0% found this document useful (0 votes)
267 views321 pages

STRAUSS - Reorientation - Leo Strauss in The 1930s (Palgrave Macmillan 2014)

The document is a compilation of scholarly works focused on Leo Strauss and his political philosophy during the 1930s, edited by Martin D. Yaffe and Richard S. Ruderman. It includes various essays that critique and interpret Strauss's thoughts, as well as appendices containing some of his writings. The book is part of the 'Recovering Political Philosophy' series published by Palgrave Macmillan.

Uploaded by

Ruben Garrote
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
267 views321 pages

STRAUSS - Reorientation - Leo Strauss in The 1930s (Palgrave Macmillan 2014)

The document is a compilation of scholarly works focused on Leo Strauss and his political philosophy during the 1930s, edited by Martin D. Yaffe and Richard S. Ruderman. It includes various essays that critique and interpret Strauss's thoughts, as well as appendices containing some of his writings. The book is part of the 'Recovering Political Philosophy' series published by Palgrave Macmillan.

Uploaded by

Ruben Garrote
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 321

REORIENTATION

R E C OV ERI N G PO LITI CAL PHILO SOPHY


SERIES EDITORS: THOMAS L. PANGLE AND TIMOTHY W. BURNS

PUBLISHED BY PALGRAVE MACMILLAN:

Lucretius as Theorist of Political Life


By John Colman

Shakespeare’s Political Wisdom


By Timothy W. Burns

Political Philosophy Cross-Examined: Perennial Challenges to the Philosophic Life


Edited by Thomas L. Pangle and J. Harvey Lomax

Eros and Socratic Political Philosophy


By David Levy

Reorientation: Leo Strauss in the 1930s


Edited by Martin D. Yaffe and Richard S. Ruderman
REORIENTATION: LEO STRAUSS
REORIENTATION
IN THE 1930s

Edited by
Martin D.Yaffe
and
Richard S. Ruderman
REORIENTATI
ON
Copyright © Martin D.
D Yaffe and Richard S.
S Ruderman,
Ruderman 2014.
2014
Softcover reprint of the hardcover 1st edition 2014 978-1-137-32438-2
All rights reserved.
First published in 2014 by
PALGRAVE MACMILLAN®
in the United States—a division of St. Martin’s Press LLC,
175 Fifth Avenue, New York, NY 10010.
Where this book is distributed in the UK, Europe and the rest of the world,
this is by Palgrave Macmillan, a division of Macmillan Publishers Limited,
registered in England, company number 785998, of Houndmills,
Basingstoke, Hampshire RG21 6XS.
Palgrave Macmillan is the global academic imprint of the above companies
and has companies and representatives throughout the world.
Palgrave® and Macmillan® are registered trademarks in the United States,
the United Kingdom, Europe and other countries.
ISBN 978-1-137-37423-3 ISBN 978-1-137-38114-9 (eBook)
DOI 10.1057/9781137381149

Library of Congress Cataloging-in-Publication Data


Reorientation : Leo Strauss in the 1930s / edited by Martin D. Yaffe and
Richard S. Ruderman.
pages cm.—(Recovering political philosophy)
Includes index.
1. Strauss, Leo—Criticism and interpretation. 2. Political science—
Philosophy. I. Yaffe, Martin D. II. Ruderman, Richard S., 1958–
JC251.S8R46 2014
181.06—dc23 2013025811
A catalogue record of the book is available from the British Library.
Design by Newgen Knowledge Works (P) Ltd., Chennai, India.
First edition: July 2014
10 9 8 7 6 5 4 3 2 1
For Connie and Anne
This page intentionally left blank
CO T TS
CONTENTS

Acknowledgments ix
Abbreviations xi

Editors’ Introduction 1
Martin D.Yaffe and Richard S. Ruderman
1. How Strauss Became Strauss 13
Heinrich Meier
2. Spinoza’s Critique of Religion: Reading the Low in the Light of the High 33
Steven Frankel
3. The Light Shed on the Crucial Development of Strauss’s
Thought by his Correspondence with Gerhard Krüger 57
Thomas L. Pangle
4. Strauss on Hermann Cohen’s “Idealizing” Appropriation
of Maimonides as a Platonist 69
Martin D.Yaffe
5. Strauss on the Religious and Intellectual Situation of the Present 79
Timothy W. Burns
6. Carl Schmitt and Strauss’s Return to Premodern Philosophy 115
Nasser Behnegar
7. Leo Strauss on the Origins of Hobbes’s Natural Science
and Its Relation to the Challenge of Divine Revelation 131
Timothy W. Burns
8. Leo Strauss on Farabi, Maimonides, et al. in the 1930s 157
Joshua Parens
9. The Problem of the Enlightenment: Strauss, Jacobi,
and the Pantheism Controversy 171
David Janssens
viii CONTENTS

10. “Through the Keyhole”: Leo Strauss’s Rediscovery of Classical


Political Philosophy in Xenophon’s Constitution of the Lacedaemonians 191
Richard S. Ruderman
11. Strauss and Schleiermacher on How to Read Plato:
An Introduction to “Exoteric Teaching” 203
Hannes Kerber

Appendices: Seven Writings by Leo Strauss 215


Preliminary Note to Appendices
A Leo Strauss: “Conspectivism” (1929) 217
Translated by Anna Schmidt and Martin D.Yaffe
B Leo Strauss: “Religious Situation of the Present” (1930) 225
Translated by Anna Schmidt and Martin D.Yaffe
C Leo Strauss: “The Intellectual Situation of the Present” (1932) 237
Translated by Anna Schmidt and Martin D.Yaffe
D Leo Strauss: “A Lost Writing of Farâbî’s” (1936) 255
Translated by Gabriel Bartlett and Martin D.Yaffe
E Leo Strauss: “On Abravanel’s Critique of Monarchy” (1937) 267
Translated by Martin D.Yaffe
Supplementary Note on Archival Materials 269
Jeffrey A. Bernstein
Editorial Note to Appendices F and G 271
Hannes Kerber
F Leo Strauss: Exoteric Teaching (1939) 275
Edited by Hannes Kerber
Supplement 1: Early Plan of “Exoteric Teaching” 287
Supplement 2: Later Plan of “Exoteric Teaching” 291
G Leo Strauss: Lecture Notes for “Persecution and the Art of Writing” (1939) 293
Edited by Hannes Kerber
Notes on Contributors 305
Index 309
AC
ACKNOWLEDGMENTS
O DG TS

E arlier versions of chapter 1 have appeared previously as “Vorwort des


Herausgebers” in Leo Strauss, Gesammelte Schriften, Band 2: Philosophie undd
Gesetz—Frühe Schriften, edited by Heinrich Meier, Stuttgart and Weimar: Metzlerr
Verlag, 1997, ix–xxxiii, and in translation as “How Strauss Became Strauss,” in
Enlightening Revolutions: Essays in Honor of Ralph Lerner, edited by Svetozar Minkov,,
363–82, Lanham, MD: Lexington Books, 2006. We are grateful to Heinrich Meierr
for his kind permission to republish.
Earlier versions of chapters 7 and 9 have appeared previously as “Leo Strauss
on the Origins of Hobbes’s Natural Science,” Review of Metaphysics 64, no. 4 (June
2011): 823–55, and “The Problem of the Enlightenment: Strauss, Jacobi, and the
Pantheism Controversy,” Review of Metaphysics 56, no. 3 (March 2003): 605–31,,
respectively. We are grateful to Review of Metaphysics and its editor, Professor Jude P..
Dougherty, for their kind permission to republish.
The German originals of appendices A, B, C, D, and E have appeared previ-
ously in Strauss, Gesammelte Schriften, Band 2, as “Konspektivismus” (pp. 365–75,,
620–21), “Religiöse Lage der Gegenwart” (pp. 377–90, 621), “Die geistige Lage
der Gegenwart” (pp. 441–62, 623), “Eine vermißte Schrift Farâbî’s” (pp. 167–77,,
614), and “Zu Abravanels Kritik der Königtums” (pp. 233–34, 615), respectively..
We are grateful to J. B. Metzlersche Verlagsbuchhandlung und C. H. Poeschel
Verlag, GmbH, for their kind permission to publish translations of these and to
Heinrich Meier for his help in arranging for their permission. Thanks especially to
the Earhart Foundation of Ann Arbor, Michigan, for its generous financial support
for that purpose.
Our series coeditor Timothy W. Burns has been unstinting in his encourage-
ment and judicious in his counsel during the routine and not-so-routine ins andd
outs, ups and downs of the preparation of this volume. And Daniel Burns, Guy
Chet, Kenneth Hart Green, Joshua Parens, and Alan Udoff have provided timely
assistance concerning Hebrew and Arabic and related matters.
This page intentionally left blank
ABBREVIATIONS

Titles of Leo Strauss’s Writings


AAPL The Argument and the Action of Plato’s “Laws.” Chicago: The University
of Chicago Press, 1975.
APT “On Abravanel’s Philosophical Tendency and Political Teaching.””
In Isaac Abravanel, edited by J. B. Trend and H. Loewe, pp. 93–109..
Cambridge, UK: Cambridge University Press, 1937.
CaM “Cohen and Maimonides.” Translated by Martin D. Yaffe and Ian
Moore. LSMC C 173–222.
CM The City and Man. Chicago: Rand McNally, 1964.
CuM “Cohen und Maimuni.” GS–2 393–436.
EMFL “Einleitung zu Morgenstunden und An die Freunde Lessings.” GS–2
528–605.
EP “Einleitung zu Phädon.” GS–2 485–504.
EPLJJ Das Erkenntnisproblem in der philosophischen Lehre Fr. H. Jacobis. GS–2
237–92.
EPM “Einleitung zu Pope ein Metaphysiker!” GS–2 476–72.
ESG “Einleitung zu Sache Gottes, oder die gerettete Vorsehung.” GS–2
514–27.
ESL “Einleitung zu Sendschreiben an den Herrn Magister Lessing in Leipzig.””
GS–2 473–75.
FP “Fârâbî’s Plato.” In Louis Ginzberg Jubilee Volume. New York: American
Academy for Jewish Research, 1945, 357–93.
FPP Faith and Political Philosophy: The Correspondence Between Leo Strauss
and Eric Voegelin, 1934–1964. Edited by Peter Emberley and Barry
Cooper. Columbia, MO: University of Missouri Press, 2004.
GA With Jacob Klein. “A Giving of Accounts.” The Collegee (Annapolis,,
MD) 22, no. 1 (April 1970): 1–5.
GA–JPCM
– M “A Giving of Accounts,” JPCM M 457–66.
GS–1 Gesammelte Schriften, Band 1: Die Religionskritik Spinozas und gehörigee
Schriften. Herausgegeben von Heinrich Meier. Stuttgart und Weimar::
J. B. Metzler Verlag, 1996; 3rd, rev. and expanded ed., 2008.
GS–2 Gesammelte Schriften, Band 2: Philosophie und Gesetz—Frühe Schriften..
Herausgegeben von Heinrich Meier. Stuttgart und Weimar: J. B..
Metzler Verlag, 1997; 2nd rev. printing, 1998; 3rd printing, 2004.
xii A B B R E V I AT I O N S

GS–3 Gesammelte l Schriften,


h f Band 3: Hobbes’ politische Wissenschaft, and zugehörige
Schriften—Briefe. Herausgegeben von Heinrich Meier. Stuttgart und
Weimar: J. B. Metzler Verlag, 2001; 2nd, rev. ed., 2008.
HBS “How to Begin to Study The Guide of the Perplexed.” Maimonides, Moses..
The Guide of the Perplexed.Translated by Shlomo Pines. Introductory Essay
by Leo Strauss. Chicago: University of Chicago Press, 1963.
HCR R Hobbes’s Critique of Religion and Related Writings. Translated and Edited by
Gabriel Bartlett and Svetozar Minkov. Chicago: University of Chicago
Press, 2011.
ISP “The Intellectual Situation of the Present.” Appendix C of the present
volume.
JPCMM Jewish Philosophy and the Crisis of Modernity. Edited by Kenneth Hart
Green. Albany, NY: State University of New York Press, 1997.
LAM Liberalism Ancient and Modern. New York: Basic Books, 1968.
LSEWW The Early Writings (1921–1932). Translated and Edited by Michael Zank..
Albany, NY: SUNY Press, 2002.
LSMC C Leo Strauss on Maimonides:The Complete Writings. Edited by Kenneth Hart..
Chicago, IL: University of Chicago Press, 2013.
LSMM M Leo Strauss on Moses Mendelssohn. Translated, Edited, and with an
Interpretive Essay by Martin D.Yaffe. Chicago, IL: University of Chicago
Press, 2012.
NRH H Natural Right and History. Chicago, IL: University of Chicago Press,,
1953.
ONI “On a New Interpretation of Plato’s Political Philosophy.” Social Research
13 (1946): 326–67.
OPSS On Plato’s “Symposium.” Edited by Seth Benardete. Chicago, IL: University
of Chicago Press, 2001.
OPSPS “The Origins of Political Science and the Problem of Socrates.””
Interpretation 23, no. 2 (1996):127–208.
OTT On Tyranny. Revised and Expanded Edition, Including the Strauss-Kojève
Correspondence. Edited by Victor Gourevitch and Michael S. Roth. New
York: Free Press, 1991.
PAW Persecution and the Art of Writing. Glencoe, IL: Free Press, 1952.
PG
G Philosophie und Gesetz: Beiträge zum Verständnis Maimunis und seinerr
Vorläufer. Berlin: Schocken Verlag, 1935.
PLA Philosophy and Law: Contributions to the Understanding of Maimonides andd
His Predecessors. Translated by Eve Adler. Albany, NY: SUNY Press, 1995.
PLB Philosophy and Law: Essays Toward the Understanding of Maimonides andd
His Predecessors. Translated by Fred Baumann. Philadelphia, PA: Jewish
Publication Society, 1987.
PPH The Political Philosophy of Hobbes: Its Basis and Its Genesis. Translated by
Elsa M. Sinclair. Reprint. Chicago: University of Chicago Press, 1952.1
PoP “The Place of the Doctrine of Providence according to Maimonides.””
Translated by Gabriel Bartlett and Svetozar Minkov. Review of Metaphysics
57 (2004):
( ) 537–49.
A B B R E V I AT I O N S xiii

PoR “Progress or Return?” JPCM M 87–136.


RCPR R The Rebirth of Classical Political Rationalism. Edited by Thomas L. Pangle..
Chicago, IL: University of Chicago Press, 1989.
RSP “Religious Situation of the Present.” Appendix B of the present volume.
SA Socrates and Aristophanes. New York: Basic Books, 1966.
SCRR Spinoza’s Critique of Religion. Translated by Elsa M. Sinclair. New York::
Schocken Books, 1965.
SR “Some Remarks on the Political Science of Farabi and Maimonides.””
Interpretation 18 (1990–91): 1–30.
SSTX “The Spirit of Sparta or the Taste of Xenophon.” Social Research 6 (1939)::
502–36.
TM Thoughts on Machiavelli. Glencoe, IL: The Free Press, 1958.
TWM “Three Waves of Modernity.” In An Introduction to Political Philosophy: Ten
Essays by Leo Strauss. Edited by Hilail Gildin. Detroit, MI: Wayne State
University Press, 1989.
WIPPP What is Political Philosophy? And Other Studies. Glencoe, IL: Free Press,,
1959.
XS Xenophon’s Socrates. Ithaca, NY: Cornell University Press, 1972.
XSD
D Xenophon’s Socratic Discourse: An Interpretation of the “Oeconomicus.” Ithaca,,
NY: Cornell University Press, 1970.

Other Titles
JMW
W The Jew in the Modern World: A Documentary History. Edited by Paul
Mendes-Flohr and Jehuda Reinharz. 2nd ed. New York: Oxford
University Press, 1995.
NHE
E Karl Löwith. Nature, History, and Existentialism. Edited by Arnold Levison..
Evanston, IL: Northwestern University Press, 1966.
MPP–1 Medieval Political Philosophy: A Sourcebook. Edited by Ralph Lerner and
Muhsin Mahdi. New York: Free Press of Glencoe, 1963.
MPP–2 Medieval Political Philosophy: A Sourcebook. 2nd ed. Edited by Joshua
Parens and Joseph C. Macfarland. Ithaca, NY: Cornell University Press,,
2011.

Other Abbreviations
Gk. Greek
Ger. German
Heb. Hebrew
{HM} Heinrich Meier (appended to footnote as its author)
Lat. Latin
Lit. literally
{LS} Leo Strauss (appended to footnote as its author)
xiv A B B R E V I AT I O N S

Note
1. Originally: Oxford: Clarendon Press, 1936. This edition will be referred to as Thee
Political Philosophy of Hobbes (1936). It differs from the 1952 reprint in that the latterr
includes a two-page “Preface to the American Edition” by Strauss, which begins as
follows: “For various reasons this study is here reissued in its original form. It is, no
doubt, in need of considerable revision. But it still seems to me that the way in which
I approached Hobbes is preferable to the available alternatives. Hobbes appeared to me
as the originator of modern political philosophy. This was an error: not Hobbes but
Machiavelli deserves this honor. But I still prefer that easily corrected error, or ratherr
its characteristic premises, to the more generally accepted views which I was forced to
oppose and which are less easily corrected” (p. xv).
EDITORS’
D TORS’ INTRODUCTION
TROD CT O

Martin D. Yaffe and Richard S. Ruderman

A s a young scholar in 1920s Germany, Leo Strauss, under Ernst Cassirer,


earned his doctorate with a dissertation on F. H. Jacobi, went on to study
with Edmund Husserl and Martin Heidegger, worked on a series of studies
on primarily Jewish themes while employed at the Academy for the Scientific
Study of Judaism in Berlin, and, as he soon afterward said about his intel-
lectual outlook during his twenties, was “so dominated and bewitched” byy
the thought of Friedrich Nietzsche “that I literally believed everything that
I understood of him.”1 What this perhaps hyperbolic remark seems to have
meant is that Strauss was convinced by Nietzsche that “God is dead”—namely
that not only the Biblical God but also the high claims of rationalism (Plato’s
Ideas, Aristotle’s teloi, the modern scientific reconstruction of human life) had
all been shown to have lost their power to convince or compel the modern (orr
nascent postmodern) soul to assent to their putative truths. Indeed, Truth itselff
(as a plausible concept) would seem to have breathed its last. And while the
charm of Plato continued to hold sway over the young Strauss—he remarked
in a reminiscence late in life that he would once have loved nothing less than
to have been a rural postmaster who spent his free time raising rabbits and
studying the Platonic dialogues 2—he could only have done so in the somewhat
impotent manner of the amused end-of-history thinker looking to fill his days
(or, as a hostile critic would later say, “rolling the classics round the tongue like
old brandy”3), for he remained convinced that a “return to the classics was not
possible.”
What were the then widely accepted reasons for this supposed impossibil-
ity? Advanced thinking of the time (at least in Germany, the most intellectu-
ally advanced country in the world) held that modern rationalism had refuted
classical thought and had itself then been refuted by the historicist insights
of Heidegger, leaving us without God, nature, or reason. By the end of the
1920s, however, Strauss was beginning to suspect that the natural situation
of man had not yet been completely obliterated or shown to be a sham, but
remained rather as a hidden possibility to be uncovered. That possibility was
far from being a simple “given” in the situation of modern men, immediately
2 M A RT I N D. YA F F E A N D R I C H A R D S . RU D E R M A N

or directly available to any who sought it. For modern man’s situation, as
Strauss’s studies of Hobbes (whom he took at the time to be the founder off
modernity) were beginning to reveal to him, was not a natural but rather a
constructed one.4 Man, according to the early modern philosophers who were
dedicated to improving his lot, was no longer to be understood in terms off
what he aspired to be. Such longings to transcend his given situation through
eros, political ambition, an after-world, or such philosophical thought as culmi-
nates in some form of metaphysics, were to be understood—and rejected—as
so many “imaginary republics and principalities” (Machiavelli, Prince, ch. 15).
Instead, he should be posited, as it were, as he would have to be in order to
achieve scientific and political certainty in his situation. If political science was
to achieve the Euclidean certainty that Hobbes sought for it, man would have
to be simplified so as to take on the countable and interchangeable nature off
numbers. And so, Hobbes rejected all claims of human inequality as the wish-
ful thinking of vainglorious thinkers who mistook their dreams for reality. Iff
men could be understood as equal—or if they could simply view themselves
as though they were equal (see Leviathan, ch. 15, end of paragraph 21)—they
could enter into a social contract that would stabilize their politics and thereby
liberate their lives for the safe pursuit of all civilized pastimes. In the absence
of any naturally ordered world, Hobbesian man would have to construct one
by means of modern science.
Now, the classical philosophers would not have denied that men had always
begun from a “made” or “created” world, the world that Plato called the
“cave.” But that world was created by poets, not scientists or philosophers. And
those poets—at least those of the caliber of Homer and Hesiod—offered cap-
tivating tableaus of human nature that stood as invitations to their audience to
reflect on and deepen their understanding of that nature. Such worlds or caves,
then, retained openings to the light, to the natural world outside of any and all
societies, such that potential philosophers who detected contradictions in the
authoritative poetic accounts could escape from them into the world of nature.
But the modern world was to be a construct, not of poets, but of philosophers
and scientists themselves. And, since the world of nature could no longer be
viewed as one of light but rather of mute darkness (random matter in motion,
as Hobbes put it), modern philosophic liberation had an altogether different
character than that of classical philosophic liberation. Modernity ultimately
came to be the first time in human history when the authority of philosophy
itself stood opposed to the search for human nature or any and all timeless
truths.
The 1930s marked Strauss’s “change of orientation,” a change that repre-
sented or consisted of his rediscovery of political philosophy and therewith the
possibility of an escape from the limitations of modern thought, culminating
in a return to classical philosophy. On the suspicion that the natural starting
point for philosophical reflection (Plato’s “cave”) had been lost to us through
the construction of an unnatural, man-made world (modern philosophy had,
in Strauss’s arresting formula, dug for us a “pit beneath the cave”), Strauss set
out in the 1930s to test four grand arguments: first, that the modern world was
E DI T OR S ’ I N T RODUC T ION 3

a constructed one that could not, strictly speaking, be considered either natural
or rational; second, that the failure of that world to become rational meant that
modern philosophy had neither refuted religion nor established its own basis
as philosophy; third, that the great medieval political philosophers were not,
as was widely thought, convinced that monotheism had fundamentally altered
the nature of man but rather that they showed a way back to true rationalism
from within changed circumstances; and fourth, that only by reopening the
great moral and theological questions as genuine questions (i.e., by admit-
ting man’s natural interest in them, irrespective of the external pressures off
social, political, and theological actors) could one recover the genuine clas-
sical approach to grounding the still possible life of philosophy. This meant
that Strauss had to dedicate himself to thoroughgoing reexaminations of the
founders of modernity (chiefly Hobbes, whose subordination of the truth to
political goals continued to have overwhelming influence, even on his chieff
critic, Carl Schmitt) and of modern, dismissive Biblical criticism (Spinoza);
of the medieval thinkers and their forgotten art of esoteric writing (Farabi,
Maimonides); of various misleading efforts of those who would recover earlierr
thought (F. H. Jacobi, Hermann Cohen, Franz Rosenzweig); and finally, off
the ancients properly understood (Strauss began with Xenophon at the end off
the 1930s). This collection constitutes the first comprehensive effort to exam-
ine Strauss’s astonishingly wide-ranging essays and books of the 1930s (some
of which having become available only recently to English-speaking readers,
including several herein) with a view to their unifying theme of recovering
classical political philosophy.
Reorientation: Leo Strauss in the 1930s seeks to explain the “change in orienta-
tion” that Strauss underwent during that eventful decade. While Strauss arrivedd
in America in 1937, and only began publishing the books that garnered attention
in the 1950s, it was in the 1930s (a decade of much personal as well as political
upheaval) that Strauss made the series of fundamental breakthroughs that enabled
him to recover, for the first time since the Middle Ages, the genuine meaning off
political philosophy and therewith the possibility of recovering classical philoso-
phy. Despite this being a period of extraordinary and wide-ranging productivity,
Strauss’s research in this period remains overlooked and largely unknown. This
has primarily been true in the English-speaking world, since so many of his
essays in this period have not been translated into English until quite recently
(we add several new translations in this volume). In fact, one major manuscript
(Hobbes’s Critique of Religion) was not published at all until the twenty-first cen-
tury. The lack of attention paid to this period is unfortunate, for several reasons.
During this period, Strauss’s manner of writing, though extremely intricate and
demanding, was relatively lacking in the allusiveness and extreme restraint that
made some of his later works sometimes difficult to access. Second, it is dur-
ing this period that the reader can follow Strauss, step by step, as he works out
his liberation from the “pit beneath the cave” (namely, the artificial situation
imposed on man by modern political philosophy and modern science that must
be overcome merely to return to the “cave,” the natural moral and political
co sc ous ess o
consciousness of man
a from
o w which
c aand
doonlyy from
o w which
c itt iss poss
possible
b e to asce
ascend
d
4 M A RT I N D. YA F F E A N D R I C H A R D S . RU D E R M A N

to genuine classical philosophy). In the 1930s, we see Strauss recovering genuine


philosophy.

* * *

The recovery Strauss undertook involved multiple, interlocking investigations.


This volume assembles in one place an examination of the various parts of Strauss’s
project so that the reader can more fully appreciate the manner in which the parts
rely on and support one another. In his opening, comprehensive chapter (“How
Strauss Became Strauss”), Heinrich Meier focuses on the meaning of “political
philosophy” and its centrality to Strauss’s project. Unlike those who treated it as
a subset of philosophy—a way of thinking about politics that presupposed the
existence and validity of philosophy itself as an undertaking—Strauss treated
political philosophy as the essential introduction to or vindication of philosophy.
This entailed addressing the pre-philosophic political opinions (including the
moral and religious opinions) of communities in their fullness, especially in their
characteristic view that philosophy was either “vicious or useless” (Republicc VI).
Political philosophy, then, had to perform the function of justifying philosophy
before the tribunal of the city. As Meier shows, Strauss came to see that political
philosophy’s central theme is not “belief and knowledge” but, rather, “law and
philosophy.” Not different ways of knowing but different ways of life are at issue
in the confrontation between reason and revelation. And this means that the
relation between reason and belief is one of conflict, not of two alternative ways
of knowing the world that can stand side by side or that can be synthesized into
some combination.
Steven Frankel looks at the dramatic change in Strauss’s understanding of the
seminal Jewish apostate Baruch Spinoza. It was, after all, Spinoza, more than any
other thinker, who initially convinced Strauss of the impossibility of a return
to classical thought. Frankel focuses on the so-called autobiographical Preface
to the 1962 English translation of Strauss’s Spinoza’s Critique of Religion. He
shows that that Preface, rather than being genuinely autobiographical, is in fact
a kind of series of “Socratic dialogues” pitting various philosophical authorities
against one another—Martin Buber against Heidegger, Cohen against Spinoza,
Rosenzweig against Maimonides, and liberalism against Zionism. Frankel helps
us to see that Strauss, rather than trying to show or illuminate the specific his-
torical influences that shaped his youthful (limited) understanding of Spinoza
and related questions, had, by 1962, become capable of seeing the ways in which
the apparently time-bound disputes of the 1920s and 1930s were in fact iterations
(strikingly new ones in many ways) of various permanent questions.
In the only chapter in the volume covering a personal friendship of Strauss,
Thomas L. Pangle examines the epistolary exchange with Gerhard Krüger.
Krüger was alone among all of Strauss’s intellectual friends and acquaintances
to join with him in taking a “philosophical interest” in theology. Krüger and
Strauss were distinctive in recognizing that the (widely understood) crisis off
modern Enlightenment thought lay not so much in its failure to provide a
rational
at o a etethics
cs with
wt w which
c to replace
ep ace Biblical
b ca et
ethics,
cs, but in its
ts pprior
o failure
a u e to
E DI T OR S ’ I N T RODUC T ION 5

adequately meet the challenge to rationalism itself from “supra-rational and


contra-rational revelation.” Only Krüger was able to share in Strauss’s eager-
ness to elevate consideration of Nietzsche’s claim that “God is dead” from the
sociological level (a mere description of the religious-mental state of modern
European man) to the philosophical level (i.e., to raise the question of whether
modern scientific and philosophical thought had indeed refuted revelation in
general or even Christian revelation specifically). The correspondence, Pangle
shows, proceeds by indicating both the inadequacy of the modern treatment off
the religious question and the initial steps Strauss was beginning to take—with
the acknowledged assistance of the great medieval political philosophers, chiefly
Farabi and Maimonides—toward the adequate and even definitive ancient,
Platonic approach to the question. For even if orthodoxy had been marginal-
ized and weakened by the successes of modernity (as Krüger and Strauss each
acknowledged), the contemporary call (by men such as Karl Barth and Franz
Rosenzweig) for a return to orthodoxy (or at any rate a kind of “‘post-modern’
neo-orthodoxy”) posed (and would continue to pose) a theoretical challenge
to philosophy. Strauss traced the inadequacy of the modern approach to its cri-
tique of miracles—a critique, he showed, that presupposed “enlightened” man
(who could accept nothing on faith) and thereby begged the question. The cor-
respondence, then, shows Strauss completing his escape from the Heideggerian
influence (at least with respect to the modern atheistic assumptions toward
revelation) with which he was partly entangled when he wrote his first book
(Spinoza’s Critique of Religion). This is marked by his evolving understanding off
the “ancient” critique of religion as being, at its deepest, not Epicurean, but
Platonic. Now, as Pangle further shows through a brief discussion of Strauss’s
later consideration (in 1959) of Krüger’s final book (Insight and Passion: The
Essence of Platonic Thinkingg (1939)), Krüger was not in fact able to follow Strauss’s
lead or take his hints as to how to recover an adequate understanding of the
Platonic approach. For he, no less than Heidegger himself, could not entertain
the possibility that Plato supplied the answer to the theological question, in
large measure because they still labored under Christian assumptions about the
altered and supposedly deepened nature of man brought about by the now-
disintegrated Christianity.
Martin D. Yaffe next examines the question of whether in Strauss’s under-
standing Hermann Cohen’s efforts at “Platonizing” Maimonides were
convincing—and whether Cohen may ultimately have been right that
Maimonides is best understood as a Platonist, but for different reasons than those
adduced by Cohen. Cohen tried to show that Maimonides presented God as a
moral-ideal for humanity, thereby (Cohen thought) siding with Plato over and
against Aristotle (who placed intellectual virtue ahead of moral virtue). But,
Yaffe explains how Strauss showed that, on this key question of the relative
superiority of the intellectual to the moral virtues, Plato was in agreement with
Aristotle—and that Maimonides’ view of God does indeed comport with Plato’s
view, properly understood. Moreover, insofar as Maimonides understood the
Torah as itself a document of political philosophy that served as a “propaedeutic
to pphilosophy,”
osop y, hee was ultimately
u t ate y indeed
deed more
o e a Platonist
ato st tthan
a an
a Aristotelian.
stote a .
6 M A RT I N D. YA F F E A N D R I C H A R D S . RU D E R M A N

In the next chapter, Timothy W. Burns considers two remarkable talks that
Strauss gave in 1930 and 1932 on the “Religious” and the “Intellectual” situa-
tion of the present. (The two talks, never before translated, are presented in ourr
Appendices B and C.) In these talks, Strauss elaborates, in the wake of his unpub-
lished book review of the sociologist Karl Mannheim (Appendix A), his under-
standing of the “present” as a curious amalgamation of temporal and permanentt
concerns. Focusing (in front of his young German-Jewish audience) on “the
Jewish problem,” Strauss discusses both the “theologico-political predicament,” and
the “theologico-political problem.” The former is a time-bound, local, if excep-
tionally pressing, problem for German Jews in the era of nascent Nazism. The
latter, however, is a permanent problem, perhaps the central permanent problem:
should man be guided by revelation or by his freely-employed reason? In order to
see his way to the genuine contours of this problem, Strauss has to begin by show-
ing the inadequacy of the “new thinking” and new theology of Rosenzweig—as
well as of the inadequate defense of reason mounted by Mannheim against the
greatest philosophical critic of reason, Heidegger. Burns shows the way in which h
Strauss powerfully demonstrates both the allure and the fundamental dishonesty
or incompleteness of “the Present.” For the Present, as interpreted by the chil-
dren of modernity and “ultra-modernity” (what today is called postmodernity), is
nothing other than a summing up and dividing of all purported and equally validd
“truths” into a sort of “mean” that is meant to satisfy all by satisfying none. “Our”
Present—that is, the purported result of our dedication to reason—is character-
ized by its smug certainty that reason teaches there is no “Truth.” Strauss, Burns
argues, teaches that the greatest obstacle to reason in “our” time comes not from
frankly antirational outlooks such as revealed religion, but from the false reason
of modernity and ultramodernity masquerading as philosophy.
Nasser Behnegar, in the subsequent chapter dealing with Strauss’s “Comments
on Carl Schmitt’s The Concept of the Political,” helps us understand what was,
according to Strauss himself, “the pivotal moment” in his life. Schmitt was perhaps
the most thoughtful critic of liberalism in Germany at the time. And more than a
few features of his critique were apt to interest Strauss. Above all, Schmitt wished
to show that liberalism was deeply hostile to the fundamentally political nature off
man, a nature that had been illegitimately (if astoundingly successfully) covered
over or assumed away by the prestidigitation of Hobbes’s “state of nature.” Now,
insofar as Hobbes had been engaged in a political critique of the political nature
of man (i.e., a critique that was meant to advance his brand of politics) it won
Schmitt’s respect as an unusually successful instance of polemics. But Schmitt, as
Behnegar shows through his intricate analysis of Strauss’s argument, labors under
a massive contradiction: he argues both that man’s nature (in the “natural orderr
of things”) is political andd that politics is always fundamentally polemical, that is,
time-bound and related to a “concrete political existence.” Behnegar’s analysis
reveals, then, the deepest basis for Strauss’s interest in Schmitt: the latter exem-
plified the tempting but ultimately flawed way of escaping from what Schmitt
perceived to be the debilitating tyranny of liberalism by adopting or remaining
under the unconscious sway of liberalism’s chief assumptions. Those assump-
t o s, at least
tions, east when
w e inverted,
ve ted, provided
p ov ded tthee oonlyy coge
cogentt assu
assumptions—other
pt o s ot e than
t a
E DI T OR S ’ I N T RODUC T ION 7

“murderous hatred of the Jews”5—that lay behind National Socialism. As Strauss


memorably puts it in his “Comments”: Schmitt’s effort at repoliticizing man,
tolerant as it was toward any and all convictions that are willing to risk war and
death, “proves to be liberalism preceded by a minus sign.”6 Liberalism, that is,
considers as true whatever tends toward peace; Schmitt’s thought considers as
true whatever tends toward war (for only in war can the noble nature of man
emerge). Now, as Strauss puts it in a later writing, “we should not behave toward
Thrasymachus as Thrasymachus behaves.” 7 We see in Behnegar’s elaboration off
Strauss’s “Comments” the first concrete step Strauss was able to take in rejectingg
liberalism’s depoliticization of man without making the apparently concomitant
Schmittian error of valorizing war or deadly decision-making. Instead, Strauss
points to the Platonic manner of rejecting that depoliticization: we need to raise
anew the question of the best regime, a political question (submerged by Hobbes
and the liberalism that arose on his basis) that without contradiction reopens the
question of the true nature of man and thereby of his satisfaction. Strauss, we see,
wishes to salvage Schmitt’s natural longing for the full coming-to-be of man’s
noblest possibilities from the derailment it suffers through remaining enmeshed
in the Hobbesian notion that human seriousness is a willed matter (which willingg
Schmitt promotes and Hobbes sought to end). And, as Behnegar finally shows, it
is through Strauss’s recovery of the sense of evil as “moral depravity” (as opposedd
to the “innocent” evil that Hobbes had understood to emanate from unen-
lightened man) that enabled Strauss, unlike Schmitt, to “restore the connection
between politics and morality.” Schmitt could not see the moral basis of his
condemnation of liberalism because he fundamentally and mistakenly accepted
liberalism’s view of morality as a private affair. If, instead, we recover the sense off
morality as indeed a force in human nature—a force not to be bracketed through
a somewhat cowardly retreat to the supposedly neutral, morality-free plane off
“theory”—we will finally once again be in a position to engage profitably in the
problem represented by the “antagonism between belief and unbelief.”
In the following chapter, Timothy W. Burns examines Strauss’s arguments
in his unfinished manuscript from 1933–34 (translated and published by Gabriel
Bartlett and Svetozar Minkov in 2011), Hobbes’s Critique of Religion, the first
extended if ultimately incomplete product that followed on the “change in ori-
entation” that he later identified as having first received public expression in
“Comments on Carl Schmitt’s The Concept of the Political.” The echo, in the title,
of Strauss’s first book (Spinoza’s Critique of Religion) should not mislead the readerr
into thinking that Strauss was rehearsing the same arguments with little changed
but the names. For not only had Strauss come to see how much more daunting
the task modern philosophy had undertaken was in refuting religion in the pro-
cess of justifying itself as a genuinely rational enterprise; he had also learned how
much more penetrating and satisfactory (because more open-minded) Hobbes’s
critique of miracles was than that of Spinoza (or Descartes). Strauss reveals, in
Burns’s reading, that Hobbes’s critique of religion (which included a critique of
classical philosophical politics, insofar as that classical approach had supplied the
“dangerous weapon” of incorporeal substances to Christian theologians) was not
t e result
the esu t o
of hiss materialist
ate a st pphilosophy—as
osop y as tthee Leviathan
eviathan appears
appea s to argue—but
a gue but
8 M A RT I N D. YA F F E A N D R I C H A R D S . RU D E R M A N

became rather the presupposition or foundation of his whole philosophy. Hobbes’s


procedure involves granting the possibility of a mysterious, all-powerful God (a
concession the more polemical Spinoza did not grant)—which possibility ren-
ders nature and hence philosophy radically questionable. To contend with—and
ultimately to dispose of—this possibility, Hobbes inaugurates a grand “civiliza-
tional” project by which man and the world will be transformed in such a way
as to render him and it impervious to miracles. As the scientific mind (initially
lacking a true foundation) becomes more pervasive, belief in and even interest in
miracles can be counted on to fade. This hope in the progress of the human mind
proves then to be the foundation of modern rationalism. And, as Burns shows
through his discussion of Strauss’s later, rather different, analysis of Hobbes’s
critique in Natural Right and History, Hobbes’s hope obscured from him the “ulti-
mate futility of all human endeavors” (a futility that Hobbes’s materialistic—and
even “idealistic”—predecessors recognized all too well).
Joshua Parens’s study of Strauss’s “Farabian turn” (the phrase is Daniel Tanguay’s)8
provides valuable insight into the thought and extraordinary influence of Farabi
on both Maimonides and Strauss. Through a careful examination of a series off
very little-known essays on Farabi, Maimonides, and Abravanel, Parens shows that
Strauss was far more open in his writings of the 1930s than in his more well-known
essays of later years with regard to Farabi’s crucial influence on Maimonides. A
proper appreciation of that influence enables us not only to interpret Maimonides
better, but even to understand the reasons for Strauss’s growing tendency to down-
play that influence. For Farabi supplies the two keys essential for understanding
Maimonides: the centrality of political science and of esoteric writing. Prior to
seeing the scope of his debt to Farabi, Strauss was still apt to read Maimonides as
turning to “supernatural resources” to supply the defect or limits of natural human
knowledge. He still, that is, saw in Maimonides another, if perhaps the greatest,
synthesizer of reason and revelation. After seeing that Maimonides shared with
Farabi the belief that there could be no synthesis of these two ultimately antagonistic
outlooks—and that esoteric writing was the historical and even necessary mannerr
in which that antagonism would have to be carefully expressed—Strauss’s under-
standing of Maimonides (and the struggle between belief and unbelief ) deepened
considerably. From Farabi, Strauss learned that the political—both political science
as means of analysis and the Law as the fundamental social, moral, and psychologi-
cal aspect of man—was “central” to Maimonides’s undertakings in the Guide. In
turning his focus from “theoretical philosophy” and “metaphysics” to the “Law”
and “particular providence,” Strauss was able to find a more, not a less, rational-
ist Maimonides. Moreover, this emphasis on political philosophy—on the effort
to resolve on the moral plane the dispute between philosophy and the city and its
Law or its gods—revealed to Strauss the Platonic core of Farabi’s and Maimonides’s
approaches. Farabi, Parens shows, was of unusual importance to Strauss, insofar as
he was the first thinker to show that it was possible to return to classical thought
from within a far-from-classical world, and how to do so.
In his chapter on “The Problem of the Enlightenment: Strauss, Jacobi, and the
Pantheism Controversy,” David Janssens provides the prehistory of the chief con-
cerns animating Strauss in the 1930s. He does so through a careful and revealing
E DI T OR S ’ I N T RODUC T ION 9

analysis of how Strauss was and was not influenced by the subject of his 1921 dis-
sertation, F. H. Jacobi. Jacobi, a younger friend and admirer of Lessing, became an
outspoken and influential critic of the Enlightenment, in both its radical (Spinozist)
and its moderate (Mendelssohnian) forms. While despising the latter position as an
unconvincing effort to stand between two stools (as well as to replace biblical rev-
elation by the subterfuge of “common sense”), Jacobi had a certain respect for the
former as one of the key consistent and rigorous engineers of the Enlightenment
outlook. That outlook, which claimed to liberate human understanding and there-
with moral and political life from traditional authorities, was recognized by Jacobi
as fundamentally destructive of all that could not be made a “clear and distinct”
object of rational, scientific thought. By applying radical doubt to the world (in
order to ground our knowledge more securely), the early Enlighteners were guilty
of reducing “Being or reality to non-Being or Nothingness” (which he termed,
in a new coinage, “nihilism”). More importantly (as such and to Strauss’s devel-
opment), Jacobi recognized the nontheoretical and in fact moral impetus behind
the modern reconstitution of science and of moral/political life. In order to grant
man autonomy, morality had to be reconceived as following the norms that we
ourselves generate (following on the model of a science that understands what it
creates or posits). But—anticipating Strauss’s rediscovery of the crucial importance
of the Law—Jacobi saw that only out of obedience to a norm, via heteronomy,
“does moral insight emerge.” Moreover, Jacobi saw that, to liberate man from het-
eronomy, the Enlightenment had to “liberate man from the authority of transcen-
dence.” While Jacobi understood this “liberation” as amounting to nothing otherr
than a revolt against God—and consequently proposed faith (in both God and in
the given world) as the antidote to the Enlightenment’s impoverishing of man—
Strauss recognized that enlisting transcendence for a moral or political purpose was
just as suspect as was condemning it for a contrary purpose. (Strauss reproduced
this line of thought throughout the 1920s in his Zionist writings, where he found
both the efforts of the political Zionists to liberate the Jews from God, and those off
the religious Zionists to enlist God, to be equally guilty of presupposing a “truth”
in order to satisfy a human, moral need.) And so, while Strauss was impressed
and influenced by many aspects of Jacobi’s critique of the Enlightenment, Janssens
powerfully demonstrates that he could not and did not share Jacobi’s turn to irra-
tionalism, authoritarianism, and faith as an alternative. Indeed, Strauss saw that
Jacobi remained enmeshed in modern presuppositions even as he sought to counter
them. Janssens completes his analysis of Jacobi and his influence on Strauss with a
consideration of Strauss’s 1930s writings on Mendelssohn. Eager to defend “natu-
ral religion,” Mendelssohn rejected all religious doctrines incompatible with the
goodness of God. In doing so, Mendelssohn also fell prey to the modern genuflec-
tion before autonomy. Thus, Janssens concludes, neither Jacobi nor Mendelssohn
made any serious effort to recover the ancient, Socratic outlook. For that, Strauss
turned to their common, infinitely more ironic and nondogmatic friend, Lessing
(whose Nathan the Wisee Strauss took to be “the outstanding poetic monument
erected in honor of medieval Jewish philosophy”).
In the following chapter, Richard S. Ruderman considers Strauss’s first pub-
lication
cat o devoted exclusively
e c us ve y to an
a ancient
a c e t author,
aut o , “Thee Spirit
Sp t of
o Sparta
Spa ta or
o the
t e
10 M A RT I N D. YA F F E A N D R I C H A R D S . RU D E R M A N

Taste of Xenophon.” This essay, written shortly after Strauss’s arrival in America
in 1937, examines Xenophon’s Constitution of the Lacedaemonians. Both the quiet,
bashful nature of Xenophon’s writings and the disrepute into which he had fallen
by Strauss’s time could lead the unwary reader into thinking that Strauss’s work
was a mere obscurantist scholarly production. Ruderman shows that Strauss’s
essay was timely without seeking to be so, since the “spirit of Sparta” (the attrac-
tion of which having been relaunched in modernity by Rousseau)—namely the
longing for total human transformation or perfection through political life—was
very much on the march in 1930s Europe. More important, Strauss demonstrated,
through a careful reading of the Constitution, that Xenophon was satirizing, not
praising, the Spartan way of life—and was doing so from the perspective of phi-
losophy. He was thereby able to show that Xenophon was able to demonstrate,
in his inimitably unassuming manner, that philosophy—precisely by taking the
claims of the city, its founders, and its morality more seriously than did the city
itself—would always and inevitably end up critiquing the city and its gods. But,
following the Xenophontic Socrates’s refusal to separate wisdom from modera-
tion, Strauss presented to his readers two of the crucial features of philosophy that
had been obscured or denied throughout the course of modernity: first, it was an
essentially theoretical enterprise that neither could nor should seek to transform m
the world; and second, it could only dismiss politics or ethics (as Heidegger, forr
instance, did) at the cost of losing the opportunity to engage in the kind of politi-
cal philosophy that alone could justify its practice as a way of life.
In the final chapter of the volume, Hannes Kerber discusses Strauss’s final writ-
ing (though unpublished until after his death) of the 1930s, “Exoteric Teaching.”
While Strauss had made a variety of observations regarding exoteric and esoteric
writing throughout the 1930s, and would continue to do so in the 1940s and
beyond (most famously in Persecution and the Art of Writing), “Exoteric Teaching” is
unique in its exclusive focus on European authors (avoiding the medieval Islamic
and Jewish authors from whom he had learned of the phenomenon in the first
place). In particular, Strauss here considers Lessing (whose writings confirmed
and deepened Strauss’s suspicions regarding esotericism more than perhaps any
others) and Schleiermacher (whose powerful and influential denial that Plato was
an esoteric writer Strauss sought to refute). Schleiermacher was not unacquainted
with the thesis of esotericism. Moreover, he usefully distinguished esoteric “sub-
ject matters” (as in Pythagoreanism) from the esoteric “manner of presentation.”
He then denied that either view applied to Plato. Kerber shows that Strauss agreed
with Schleiermacher in rejecting Platonic esotericism thus understood. For Plato
did not follow the Pythagoreans in refusing to discuss “in public” (i.e., in his
writings) certain topics and he did not divide his oeuvre into popular or “exo-
teric” and scientific or “esoteric” writings. There is, in all of Plato’s writings,
both a comprehensive account of all subjects, so to speak, and a perfect fusion orr
interweaving of esoteric and exoteric presentations. Schleiermacher, in short, had
failed to grasp the specific nature of Plato’s intended audience—the potential phi-
losopher—and thus had failed to see that the Platonic mode of education presentt
throughout the dialogues consisted of broaching every subject “exoterically” (i.e.,
through an appeal to some sort of popular or accepted opinion) and then to move,
not “continuously” (as though deepening an essentially correct view with fullerr
E DI T OR S ’ I N T RODUC T ION 11

insights), but “discontinuously” via a conversion or “turning-around” of the soul


to the “esoteric” level. And, as Kerber points out, Schleiermacher’s inability to
see Plato’s exotericism was connected to his failure to see the needd for it: Socrates
was not engaged in a mere quarrel with polytheism or popular religion (as though
striving to purify the object of the Athenians’ sound but confused religious long-
ings) but was questioning the “gods of the city of Athens,” which gods under-
wrote the political and moral basis of Athenian life. Schleiermacher’s rejection off
Platonic esotericism, then, was connected to his failure to see the radical content
of Socratic questioning. Finally, Kerber notes perhaps the chief insight or advance
made by Strauss in the course of the 1930s due to his deepened understanding off
esotericism: while he had still understood Maimonides to be nothing more than
a “believing Jew” in his start-of-the-decade Spinoza’s Critique of Religion, he came
to see, by the decade’s end, that Maimonides was a philosopher who, through his
example and his teaching, made the return to classical philosophy a plausible andd
indeed compelling possibility.
Our appendices include translations of several fascinating writings of Strauss’s
from the 1930s that have been unavailable in English till now. In “Conspectivism”
(1929), Strauss uncovers the questionable moral motive behind the contemporary
effort to replace the notion of timeless truth with something like the “spirit of the
age,” an outlook not precisely held by anyone at the time, but an assumed con-
glomeration or averaging of all partial but equally valid outlooks. One cannot
help but express Strauss’s intention here as being an effort to clarify the nature
of “our” cave and to do so in such a way as to incline us to the suspicion that
an egalitarian moral motive (and not a genuine theoretical insight) lay behind
its novel (and purportedly theoretical) denial that one could escape from the
cave. Strauss then critiques at length the efforts of sociologist Karl Mannheim to
defend, through conspectivism, rationalism (i.e., the effort to celebrate the fact
that multiple philosophies exist that cannot and do not presume to refute one
another and thereby to ground themselves).
In “Religious Situation of the Present” (1930) and “The Intellectual Situation
of the Present” (1932), both analyzed in detail in this volume by Timothy W.
Burns, Strauss undertakes not merely to summarize those respective situations,
but to bring a radical critique to bear on the very notion of their importance orr
existence. Indeed, he shows the very problematical nature of the always-pow-
erful enticement of “the present.” Not deeper entanglement with the “present,”
but a return to the fundamental, original meanings of the Bible and of Plato is
what is needed. Through reinforcing or bowing to the moral prejudices of the
time, the religious thinkers and intellectuals can be seen as standing athwart
Thinking and yelling “stop!”
We should add that the translators of the three foregoing writings have anno-
tated them somewhat sparingly, so as to call attention to Strauss’s evidently
not considering the German originals ready or worthy to be published as they
stand.
“A Lost Writing of Farâbî’s” (1936) and “On Abravanel’s Critique of Monarchy”
(1937) supply intricate supplemental background to Joshua Parens’s penetratingg
discussion of Strauss’s studies of medieval political philosophy throughout the
1930s. The former essay proceeds in a somewhat scholarly fashion to disentangle
12 M A RT I N D. YA F F E A N D R I C H A R D S . RU D E R M A N

the genuine Farabi from the rather less penetrating version of him transmit-
ted by the Jewish thinker Falaquera—the Farabi who, as Strauss concludes, was
“the towering figure who laid the ground for the later development and set
down its limits by making his task the revival of Platonic-Aristotelian philoso-
phy as philosophy as such.” The latter is a brief but arresting effort to disentangle
Abravanel’s “theocratic” critique of monarchy from Joseph ibn Caspi’s “aristo-
cratic-philosophical” one (both of which are considered more or less identical
by the essay under review). In the course of his demonstration, Strauss traces
Caspi’s argument to a somewhat overlooked passage in Maimonides’ Guide of thee
Perplexed, thereby offering a small and early example of how to read Maimonides
as an esoteric thinker.
Finally, we present the original, unfinished work “Exoteric Teaching,” and
“Lecture Notes for ‘Persecution and the Art of Writing,’” both analyzed herein by
Hannes Kerber. The former supplies Strauss’s pre-Persecution and the Art of Writingg
views on exoteric writing in general and the proper way of understanding Plato’s
exotericism in particular. And the latter supplies a plethora of evidence (much off
which was not provided to the readers of the later essay “Persecution and the Art
of Writing”) from writers both ancient and modern that they employed and had
to employ esoteric writing as well as a brief interpretation of Don Quixotee that
establishes it too as a masterpiece of esotericism.
The volume as a whole attempts to introduce the reader to the extraordinary
and rapid evolution of Strauss’s thought in the 1930s. His efforts at analyzing the
current state of thinking (and of the political and religious possibilities thought to
constitute our horizon) provide incentive to escape from the cave. And his stud-
ies of the medieval philosophers and, through them, the classics, provided him
(and us) with proof of the possibility of such an escape. We hope, in presenting
these studies, to ensure that Strauss’s efforts will no longer be able to be dismissed
as essentially Schmittian (or Jacobian) efforts at the authoritarian rejection off
democratic modernity, nor as “rationalist” efforts at restoring religion, nor as
“pain-loving antiquarianism.”9 For it was in the 1930s that Strauss made the first
efforts in 400 years to uncover the true meaning of philosophy and the mannerr
in which to vindicate it as a way of life.

Notes
1. Letter to Karl Löwith, June 23, 1935, GS–3 648; trans. George Elliot Tucker, in
Independent Journal of Philosophy 5/6 (1988): 183; quoted in JPCM
M 63.
2. GA–JPCM
– M 460.
3. Brian Barry, Political Argumentt (London: Routledge, 1965), 290.
4. See HCR.
5. SCRR 3.
6. SCRR 350.
7. CMM 74.
8. Daniel Tanguay, Leo Strauss: An Intellectual Biography, trans. Christopher Nadon
(New Haven: Yale University Press, 2007), 7.
9. CMM 1.
C APT R 1
CHAPTER

HOW STRAUSS BECAME STRAUSS

Heinrich Meier

A mong the great philosophers of the twentieth century, political philosophy


had only one advocate. For a long time it seemed to have been dispensed
with by history or to have fallen into oblivion, thus to have become a thing off
the past in every sense. That political philosophy has been recognized to have a
present and ultimately a future is the achievement of Leo Strauss. It is especially
this achievement that marks his place in the history of philosophy. And it alone
would suffice to draw the keen attention of later philosophers to his oeuvre. Forr
nothing less than their ownmost, their vital and rightly understood self-interest
is in question with the disappearance or the recovery of political philosophy. Yet
the revolution of thought that Strauss brought about single-handedly has a pre-
history. The revolution is preceded by a periagoge, or a decisive turn, in Strauss’s
thought, just as political philosophy itself is based on a periagoge in philosophy.
Whoever would like to understand this turn better, whoever is interested in how
Strauss became Strauss, can now—with the aid of the first three volumes of his
Gesammelte Schriften, which appeared between 1996 and 2001 and which come
to some 2,000 pages, nearly half of which contain previously unpublished texts
and letters—retrace step by step the path of thought that Strauss traversed in the
formative period of his philosophy.1
The second volume of Strauss’s Gesammelte Schriften comprises 35 texts that
were written between 1921 and 1937. Together with the first and third volumes,
which contain the books on Spinoza and Hobbes, as well as the accompanyingg
writings from the same period, it makes accessible—in the original languages
(German, French, and English)—Strauss’s entire early work up to his emigra-
tion to the United States of America in 1938. Well over a quarter of the second
volume consists of first publications. Part I makes available once again, afterr
more than six decades, Philosophie und Gesetz, a book that marks the beginningg
of a radically new understanding of Maimonides and his Arabic “predecessors”
in the horizon of Platonic political philosophy. The new edition of Philosophiee
und Gesetz is complemented by four important essays from 1936–37 on Farabi,
14 H EIN RICH M EIER

Maimonides, and Abravanel, which continue the new interpretative approach andd
are supplemented by previously untranscribed marginalia from Strauss’s personal
copies. Part I concludes with a brief, previously unknown comment entitled “Zu
Abravanels Kritik des Königtums.” Part II, entitled Frühe Schriften, begins with
the first publication of Strauss’s doctoral dissertation, Das Erkenntnisproblem in derr
philosophischen Lehre Fr. H. Jacobis, from 1921, and ends with a preface that Strauss
wrote in 1937 in Cambridge to an unwritten book entitled Eine Erinnerung an
Lessing. Between them lie numerous essays, lectures, articles, and reviews, which
are presented in chronological order. These include the first publication of “Der
Konspektivismus” (1929), “‘Religiöse Lage der Gegenwart’” (1930), “Cohen
und Maimuni” (1931), and “Die geistige Lage der Gegenwart” (1932).
Leo Strauss’s second book-length work, Philosophie und Gesetz. Beiträge zum
Verständnis Maimunis und seiner Vorläufer, r was published by Schocken Verlag in
Berlin in time for the festivities surrounding the eight-hundredth anniver-
sary of Maimonides’s birth on March 30, 1935. Just a few days later, Ludwigg
Feuchtwanger announced the new book in a review that began: “The concern
that access to Maimonides would be more obstructed than opened up by the vast
literature from the anniversary year is eliminated by Leo Strauss’s revolutionary
writing, Philosophie und Gesetz.”2 The volume’s appearance on the Maimonides
anniversary did not result, however, in either greater circulation or an apprecia-
ble public reception. Since it was a publication by a “Jewish author” on a “Jewish
subject” with a Jewish publishing house—a circumstance that alone made it
possible to publish Philosophie und Gesetz in the Germany of 19353—it went vir-
tually unnoticed beyond the few Jewish organs still in existence. Most libraries
in Germany were closed to it. It did not even make it into the holdings of some
university libraries. The contemporary debate passed over it almost completely
in silence.4 One can therefore say: Philosophie und Gesetz had been in print since
spring 1935 without ever really having been made public. It remained a book
for the few. In addition to the circle of specialists in the thought of the Jewish
and Islamic Middle Ages, it reached “initiates” such as Ernest Barker in England,
Étienne Gilson in France, and Carl Schmitt in Germany. And it received atten-
tion early from friends and acquaintances to whom Strauss himself sent the book:
in philosophy, for example, to Hans-Georg Gadamer, Gerhard Krüger, Karl
Löwith, Jacob Klein, Alexandre Koyré, and Alexandre Kojève; in Arabic and
Judaic studies, to Fritz Baer, Shlomo Pines,5 Paul Kraus, and Gershom Scholem.
Their names may give an indication of the subterranean influence that Philosophiee
und Gesetz exerted from the very beginning.
The Maimonides anniversary was the reason for Schocken Verlag to publish
Philosophie und Gesetz in 1935. Not so for Strauss. He had not written the book
in view of the commemorative year, nor did he even write it as a book in the first
place. He published it with an eye to his application for a professorship at the
Hebrew University in Jerusalem that he decided to pursue in 1933–34 since his
grant from the Rockefeller Foundation—which had led him first to Paris from
October 1932 to December 1933 and then on to London in order to continue his
Hobbes research—ended on September 30, 1934, and he had no hope of securingg
a pos
position
t o in eeither
t e France
a ce o
or England.
g a d. From
o autu
autumn 1933
933 on,
o , St
Strauss
auss ttried
ed to get
HOW S T R AU S S BE C A M E S T R AU S S 15

various positions in Jerusalem.6 After Julius Guttmann, 19 years Strauss’s seniorr


and his superior at the Akademie für die Wissenschaft des Judentums (Academy
for the Science of Judaism) in Berlin until 1932, had received the professorship
for Jewish medieval philosophy in 1934, Gershom Scholem advised Strauss to
publish something in Jewish studies as soon as possible in view of another vacant
professorship. At the time Strauss had two essays ready for publication that fit-
ted the bill, copies of which had already been circulating for some time and
with which Scholem was quite familiar: the study “Maimunis Lehre von derr
Prophetie und ihre Quellen,” 7 which was completed in July 1931 and was origi-
nally to appear in the Korrespondenzblattt of the Akademie für die Wissenschaft
des Judentums but did not because, for financial reasons, the Korrespondenzblattt
could not be continued from 1931 on; and an essay of roughly the same length
that was finished in September 1933 and that takes Julius Guttmann’s Philosophiee
des Judentums published in the same year as the occasion for a fundamental state-
ment and clarification of Strauss’s own position.8 On August 2, 1934, Strauss
writes to Scholem: “To follow your advice to publish something soon in Jewish
studies, I wrote to Nyberg, the editor of Monde Oriental, who had accepted my
essay on Maimonides’s prophetology and its sources for publication in the 1933
volume of the journal, asking him to print my essay finally.” Strauss continues:
“So since, as things stand, the publication might be delayed for some time still, I
have considered whether it would not be feasible to publish either my Guttmann
review or my Maimonides essay or both in Hebrew translation in a Palestinian
journal. What do you think about that? And might you know of someone who
would undertake the translation?” On December 14, 1934, Strauss is able to
tell Scholem about an entirely different publication project: “Thanks to Ernst
Simon’s and Buber’s kind mediation, there is now some hope that my essay on
Maimonides’s prophetology and its sources, as well as my notes on Guttmann’s
book, will finally be printed. Schocken Verlag’s editor—Dr. Spitzer—writes me
that he wants to recommend to Schocken that the essays be published on the
occasion of the Maimonides anniversary as soon as a suitable title has been foundd
that would justify the publication of the two at least formally quite different
essays together in one volume. I suggest adding a short essay entitled “Die geset-
zliche Begründung der Philosophie,” in which I discuss the treatment of this
theme in Ibn Rushd, Maimonides, and Levi b. Gershom, and then to entitle the
volume Philosophie und Gesetz. Ein Beitrag zur Maimuni-Forschung. The decision is
completely in Mr. Schocken’s hands in Jerusalem. I would be enormously grate-
ful to you if you would put in a good word for me with Schocken. An emphatic
recommendation by Buber and an evaluation by Schaeder have been sent to
Spitzer.”9 Philosophie und Gesetz now takes shape so rapidly and so concretely that
Strauss is able to inform Guttmann not even six weeks later, on January 22, 1935,
about the quite advanced preparations for its publication: “I have already final-
ized two essays for publication: the review of your book and a short article on the
legal justification of philosophy in Ibn Rushd, Maimonides, and Gersonides. The
essay on Maimonides’s prophetology and its sources I hope to be able to send off ff
in two or three days. And if I still have the time and leisure for it, I want to write
a introduction
an t oduct o tthatat w
will motivate
ot vate tthee pub
publication
cat o oof tthese
ese essays, w
which
c aaree so
16 H EIN RICH M EIER

diverse.” We shall soon have occasion to speak of this introduction—the first part
of the book, though the last to be written. On March 27, 1935, Strauss is able to
report to Guttmann and his other correspondents that the book has been pub-
lished. In just under three months, the plan had been realized. Strauss announces
to Gerhard Krüger in Marburg “that in the next few days you will receive two
copies of a booklet, one of which is for you and the other for Gadamer. I would
be delighted to hear what each of you thinks of it sometime.”
The history of the genesis of Philosophie und Gesetz is of interest insofar as it
allows an adequate dating of the individual parts of the volume, and the chro-
nology of the four essays making up Philosophie und Gesetz deserves, like the
chronology of other works that Strauss wrote between 1929 and 1937, some
consideration, since during these nine years a far-reaching reorientation off
Strauss’s thought occurs and his philosophical positions undergo a significant
change. Strauss himself drew attention to the fact that after the completion off
Die Religionskritik Spinozas in 1928 he reached a caesura that was of the greatest
importance for his further path of thought. At the end of the long “autobio-
graphical preface” to the English translation of his Spinoza book, he speaks of a
“change of orientation,” which took the form of a turn away from the premise,
sanctioned by a powerful prejudice, “that a return to premodern philosophy
is impossible,” a premise, as he expressly adds, on which Die Religionskritik
Spinozas was based. Strauss’s change of orientation found its first expression,
“not entirely by accident,” in the 1932 essay “Anmerkungen zu Carl Schmitt,
Der Begriff des Politischen.”10 What caused this change of orientation? How is
it to be defined more precisely? And what insights does the change disclose?
The edition of Strauss’s Gesammelte Schriften, which in the second and third
volumes presents together for the first time all the writings from the decisive
years between the conclusion of the Spinoza book and his new beginning
in America, places the careful reader in a position to answer these and simi-
lar questions himself. A few remarks concerning the development of Strauss’s
thought, in addition to information about the chronology of his oeuvre,11 may
nevertheless be appropriate.
At the center of Strauss’s change of orientation lies the emphatic turn to polit-
ical philosophy in the sense of a conscious return to a thorough examination off
the political opinions and theological convictions that precede philosophy orr
that are able to place philosophy fundamentally in question. The turn to political
philosophy is flanked by both a new hermeneutic openness and the rediscovery
of the art of exoteric-esoteric writing. The former is reflected in the maxim
that the greatest effort must be made to understand the philosophers of the past
exactly as they understood themselves. The latter leads to the insight into the
difference that exists, for political as well as philosophical reasons, between the
generally accessible teachings and the thought of those philosophers. Each off
the three aspects of this turn includes mutatis mutandis the turn away from the

“premise , sanctioned by a powerful prejudice,” that a return to premodern phi-
losophy is impossible. The most important considerations and encounters that
contributed to the reorientation sketched only briefly here12 can be named in
abb ev ated form:
abbreviated o : 1)) Thee co
conclusion
c us o reached
eac ed by Dieie Religionskritik Spinozas that
t at
HOW S T R AU S S BE C A M E S T R AU S S 17

modern rationalism’s answer to revealed religion remained insufficient—or that


the “offensive critique” failed and the “Napoleonic strategy,” with which the
modern Enlightenment hoped to defeat orthodoxy, ultimately aroused the sus-
picion that the Enlightenment was based merely on a “new belief ” rather than
on a “new knowledge.” 2) The question whether the self-destruction of reason
is the inevitable consequence of modern, in contrast to premodern, rationalism;
and whether radical historicism, in which that self-destruction is most clearlyy
and most resolutely expressed, must not be grasped as the position in which the
complex historical development of modern philosophy reaches its end. 3) The
judgment that the irruption of a tradition based on revelation into the world
of philosophy and the battle fought against this tradition in a manifoldly inter-
locking way over the centuries maneuvered philosophy into a second, historical
cave of increased mediation and artificiality that lies below the “natural” cave in
which philosophy, according to the Platonic allegory, begins its ascent; and the
resultant view that the historical obstacles that are added to the natural difficul-
ties of philosophizing in the present make necessary a concerted effort, to regain
the natural horizon in the form of a historical propaedeutic. 4) The insight into
the possibilities that Nietzsche’s critique of the principles of the biblical and the
philosophical traditions opens up for an original understanding of these prin-
ciples, as well as the insight into the opportunities that Heidegger’s destruction
of the tradition offers for an adequate confrontation with the thought of the phi-
losophers of the past, a confrontation oriented completely toward the subject at
issue in their thought, when this confrontation is carried out under the guidance
of political philosophy. 5) The conclusion that the philosopher in the present has
every reason to approach premodern philosophy in the manner not of one who
believes oneself able to understand medieval and ancient philosophers better than
they understood themselves, but rather of one who wants to give thought to the
alternatives and explore paths that lead out of the modern impasse, in short: as
one eager to learn from premodern philosophy and open to being overcome by
wonder, to being surprised by it.
A momentous surprise awaited Strauss in 1929 or 1930. While reading
Avicenna’s treatise On the Divisions of the Rational Sciences for the first time, he
came upon the statement that the treatment of prophecy and divine law is con-
tained in Plato’s Laws. Avicenna’s hint, which Strauss would use more than fourr
decades later as the motto of his last book, his commentary on the Laws (the
“most pious” and “most ironical of Plato’s works”),13 disclosed to Strauss virtu-
ally overnight a new, unhoped-for access not only to the Arabic philosophers
Farabi, Avicenna, and Averroes, and to Maimonides, but also to Plato himself.
This find at the Berlin National Library meant the birth of Strauss’s theme “phi-
losophy and law.” A subject to which Strauss had directed his attention primarily
for historical reasons turned out to be one of eminently philosophical interest.14
For Avicenna’s sentence said in other words that, long before the irruption off
revealed religions into the world of philosophy, Plato had at his disposal the
means that the philosophers of the Middle Ages needed and which the Platonic
political philosophers used to meet the historical challenge with which they
we e confronted:
were co o ted: tthat at iss to say, to u
understand
de sta d itt pphilosophically
osop ca y and
a d to make
a e the
t e
18 H EIN RICH M EIER

best of it politically for themselves and for their commonwealths. The Arabic
philosophers and Maimonides follow Plato when they grasp the divine law,
providence, and the prophet as objects of politics; they rely on the Laws when
they treat the teaching of revelation, the doctrine of particular providence, and
prophetology as parts of political science (and not at all of metaphysics);15 they
move in the horizon of the Republicc when they regard the founding of the
“perfect city” as the raison d’être of revelation.16 In this sense Strauss can speak
of our grasping in Plato the “unbelieving, philosophical grounding of faith in
revelation in its origin.”17 The endeavor to ground or to found faith in revela-
tion from the political perspective of the founder distinguishes the rationalism
of the Platonic political philosophers from modern rationalism, which is no
less interested in a natural explanation of faith in revelation, which ventures
no less to gain a philosophical understanding of faith in revelation, but which
does not carry out the confrontation from this perspective.18 And also in con-
trast to modern rationalism, premodern rationalism begins the confrontation
by focusing on the law: It carries out its grounding of faith in revelation as the
grounding of the law. Platonic rationalism begins by focusing on the law in the
original sense, on the comprehensive order of the commonwealth, an orderr
that unites religion and politics and that, as religious, political, moral law, lays
claim to the individual wholly, existentially.19 Premodern rationalism engages
this claim radically in order to understand the law radically. It justifies the law
in order to get beyond the law. If the philosophical justification of the law is the
mode of understanding the law from the ground up, then it is simultaneously
the locus in which the presuppositions of philosophy are engaged, in which the
self-knowledge of the philosopher is in question. For with the grounding of the
law as the politico-theological order or the way of life of the commonwealth
the question of the right of the philosophical life is raised most acutely. This
question moves into the center of political philosophy, and since the rationalism
of the medieval Platonists is essentially political philosophy, its central theme is
not “belief and knowledge” but “law and philosophy”: The philosophical justi-
fication of the law becomes the grounding of philosophy.20
Leo Strauss persistently pursued his new interpretation of the political phi-
losophy of the Islamic and Jewish Middle Ages in the essays and lectures from
1930–37 and rapidly developed it more richly. In the course of this development,
the confrontation with the conflict-laden polarity “philosophy and law” becomes
not only more profound but also clearer. Several careful statements and several
cautious hints in Parts II–IV of Philosophie und Gesetz disclose their full signifi-
cance only in light of the rhetorically bolder essays “Quelques remarques sur la
science politique de Maïmonide et de Fârâbî” and “On Abravanel’s Philosophical
Tendency and Political Teaching.” However, given his high regard for the medi-
eval enlightenment—which, as Strauss emphasizes, in contrast to the modern
Enlightenment, or the “Enlightenment properly speaking,” insisted on the eso-
teric character of philosophy—Strauss’s reticence in all the aforementioned writ-
ings (including the “very daring” introduction to Philosophie und Gesetz) should
not be underestimated.21 For instance, it is left up to the reader to bring togetherr
a gu e ts tthat
arguments at interlock
te oc but aaree deve
developed
oped in passages far a removed
e oved fromo o onee
HOW S T R AU S S BE C A M E S T R AU S S 19

another, or important statements able to shed light on one another but that the
author “scatters” over various publications, so that the reader thinks them himselff
and draws the necessary conclusions. Let us give three examples, which lead us
beyond Strauss’s early writings on medieval political philosophy: First, precisely
what Strauss had in view when, at the outset of the Introduction to Philosophiee
und Gesetz, he says that “the critique of the present, the critique of modern ratio-
nalism as the critique of modern sophistry, is the necessary beginning, the constant
companion, and the sure sign of the quest for truth which is possible in our age,”
becomes clear when one consults the critique of the “Epimethean” physics properr
to sophistry with recourse to Plato’s Protagoras in Hobbes’ politische Wissenschaft,
which he wrote at the same time or had already completed.22 Second, whoeverr
wants to understand why Strauss (once again in the Introduction) places the word
“radicalized” in quotation marks in saying that modern philosophy has “radical-
ized” the Platonic critique of the natural ideal of courage “in such a way that the
character of virtue in courage as such is really denied,” or whoever would like to
understand Strauss’s confrontation with Nietzsche’s and Heidegger’s “new kind
of fortitude” (in which the Introduction culminates) will profit from reading
what Strauss has to say in Hobbes’ politische Wissenschaftt about the radical critiquee off
the ideal of courage—now without quotation marks.23 Third and finally, in the
context of the same discussion there occurs a statement—expressed most clearly
in the revised English version—that clarifies the status of the expositions on the
“fundamental difference between Plato and Aristotle” that come at the end off
Philosophie und Gesetz and that may perhaps puzzle some readers.24 Of course,
there are also modifications of content—and new discoveries. Thus Strauss gains
greater clarity roughly in the latter half of 1935 on the political dimension off
the philosophical confrontation with revealed religions and on what he terms
“philosophical politics.” It is expressed, for example, in the account of Farabi’s
role as founding father of the first great endeavor to recover philosophy and in
the observation that the Arabic philosophers attached far greater significance to
war and courage than had Plato or Aristotle, as well as in the parallel observa-
tion that Hobbes—whom Strauss regarded at that time as the founding father off
modern political philosophy—gave primacy to foreign policy in a critical turn
against the ancient philosophers.25 The most important changes in Strauss’s posi-
tion concern, however, his account and—if we look beyond the years 1930–37
for a moment—understanding of Maimonides as a whole. Whereas Strauss still
characterized Maimonides in his Spinoza book as an “adherent of revealed reli-
gion” and as a “devout Jew,” ten years later he is convinced that Maimonides was
absolutely not a Jew in his belief; and whereas he had stuck essentially to the tra-
ditional reading in his interpretation of the Moreh nebuchim in 1928, after careful
study of the book he comes to the view, in 1938, that Maimonides with his Guidee
of the Perplexedd had succeeded to a far greater degree in doing what Nietzsche
had in mind with his Zarathustra.26 Before 1938 the most striking modification
that Strauss makes in his new Maimonides interpretation,27 which he had been
developing since 1930, is in “Quelques remarques.” There he explicitly distances
himself from the view that for Maimonides the dependence of the philosophers
o revelation
on eve at o iss based oon tthee insufficiency
su c e cy o of human
u a reason.
easo . Itt iss ppresumably
esu ab y in
20 H EIN RICH M EIER

view of the claim of insufficiency, which plays a prominent role in the account
in Philosophie und Gesetz,28 that Strauss says in retrospect that he moved, “so
to speak, contrary to Guttmann’s moderate rationalism, on a path via a Jewish
Thomism to radical ‘rationalism.’”29 The Abravanel essay, in which the contours
of the political philosopher Maimonides emerge particularly clearly through
a comparison and contrast with the political theologian Abravanel, doubtless
comes very close to the position of “radical rationalism.” With its uncommon
openness, it forms a transition to the essay “The Literary Character of the Guidee
for the Perplexed,” which was written in 1938 with all the more reticence and
extreme concentration.30
In many respects the introduction to Philosophie und Gesetz occupies a spe-
cial place. It is—notwithstanding the framing remarks, which make express
reference to the subsequent texts and without which it could not have been
published in 1935—an essay in its own right. Finished in only a few days, it
counts among the most brilliant essays that Strauss wrote. Its great themes— —
the need for a new understanding of the quarrel between enlightenment and
orthodoxy, the necessary failure of any attempt to bring about a synthesis off
“Jerusalem” and “Athens,” the self-destruction of reason as a consequence off
modern rationalism, the requirement that the philosophical confrontation with
faith in revelation be taken up anew—are programmatic for Strauss’s future
studies. Almost three decades later, Strauss granted such undiminished validity
to the analysis of “atheism from probity,” that new “atheism with a good, orr
also a bad conscience,” which “is distinguished from the atheism at which the
past shuddered precisely by its conscientiousness, by its morality,” that he trans-
lated the decisive passages from the introduction word for word into English
and included them in his “autobiographical preface.”31 Since the essay gives a
vigorous account of the philosophical intention that guides the investigation
of the seemingly historical subject of Philosophie und Gesetz, it is simultane-
ously, strictly speaking, an introduction, a text that leads the reader to Parts
II–IV, and thus that part that makes the volume a whole. On May 9, 1935,
Strauss writes to Kojève from Cambridge: “Just read the introduction and essay
no. 1. The introduction is very daring and will interest you already for that
reason. And then write me with your reaction. I myself regard it as the best I’ve
written.” There is no record of Kojève’s judgment. The contemporary reac-
tions we know of from Strauss’s circle of friends and acquaintances, however,
could hardly be more diverse. They range from Löwith’s conjecture that Strauss
sided with orthodoxy32 to Guttmann’s assumption that he identifies himselff
with the position of “modern existentialist philosophy.”33 Klein, whom Strauss
had described as perscrutatorem cordis mei a few years earlier in a letter to Krüger,
confesses a certain helplessness regarding the question to which the last pages off
the Introduction give rise: “but wheree is Maimonides’s enlightenment supposed
to lead us now? I understand full well that an answer is not immediately possible:
it is the situation in which we find ourselves in the first place: indeed, to antici-
pate an answer would mean no longer wanting to understand back [Zurück-
verstehen-Wollen].” He has his astonishment lead to the consideration: “Well,
o e could,
one cou d, followingg your
you accou
account,
t, aarrive
ve at tthee co
conclusion:
c us o : w
whyy not
ot o
orthodoxy,
t odo y,
HOW S T R AU S S BE C A M E S T R AU S S 21

then?! After all, you yourself say something very, very important, it seems to
me, when you do nott identify ‘probity’ with ‘love of truth.’ And so everythingg
remains open. Which I myselff have nothing against. But it is clear that people are
going to hold that against you.”34 By contrast, for Scholem the most important
question by no means remains open. “Shortly,” he writes to Walter Benjamin
on March 29, 1935, “Schocken will publish for the Maimonides anniversary a
book by Leo Strauß (whom I have tried very hard to get appointed to a profes-
sorship in Jerusalem) which—a remarkable act of bravery for a book that has to
be understood by everyone as one by a candidate for Jerusalem—begins with a
thoroughly (albeit completely madly) reasoned, undisguised confession to athe-
ism as the most important Jewish watchword! That even outdoes those first 40
pages of your Habilitationsschrift! I admire this morale and regret the apparently
consciously and deliberately provoked suicide of such a good mind. The inde-
pendence of mind to vote for the appointment of an atheist to a teaching post
for the philosophy of religion is here understandably to be expected from at
most three people.” Benjamin responds to the news on May 20: “the book by
Leo Strauß is of great interest to me. What you have told me about him fits the
pleasant impression I had always had of him.”35
The early writings presented in Part II of the second volume of the Gesammeltee
Schriften give clear confirmation of a statement Jacob Klein made half a century
later, namely that at the time of their first encounter at the University of Marburg
in 1920 and in the early years of their friendship thereafter Strauss was primarily
interested in two questions: the question of God and the question of politics. 36
The importance of these questions, which form the leitmotif of the articles and
reviews from the 1920s—beginning with the first two of Strauss’s publications,
an intervention in the debate over Zionism and a short essay on Rudolf Otto’s
book Das Heilige—justifies the presentation of all the publications prior to the
caesura around 1929 that, along with the dissertation, make up the first group
of the Frühe Schriften.37 Strauss himself most likely would not have included
these contributions to journals and newspapers in a collection; he did not even
gather them together for his personal archive; and he certainly did not considerr
a later publication of his doctoral dissertation, which he wrote as a 22-year-old
under Ernst Cassirer in Hamburg and which in 1970 he called “a disgraceful
performance.”38 Yet in view of the fact that the dissertation on Jacobi has since
become available in French translation 39 and its translation into other languages
is planned, it seemed time to publish a reliable edition of the German original— —
particularly as the question of God already plays a role in the dissertation. The
Zionist articles and statements from 1923–25 allow the starting point of the
intellectual development to take on sharper contours, a development on which
the “autobiographical preface” sheds light and whose philosophical interest lies
in the fact that it testifies to the truth of Plato’s Republicc according to which the
ascent of philosophy starts with the political opinions that are obligatory orr
binding for the individual and is carried out as the gaining of insight into the
nature or the limits of those opinions.
The second group of the Frühe Schriften comprises texts in which Strauss’s
“change
c a ge of
o orientation”
o e tat o iss prepared
p epa ed or
o iss already
a eady expressed.
e p essed. Since,
S ce, with
w t the
t e
22 H EIN RICH M EIER

exception of the Ebbinghaus review, none of them has been published before,
they shed still more light on the factors that contributed to that change of orien-
tation, factors that I sketched above in five points. Thus, at the end of the essay on
conspectivism, we encounter the question of “how the world in which science
arose looked before the irruption of biblical consciousness,” which Strauss rein-
forces with the announcement: “Only in the orientation towards this world is one
to gain the horizon in which alone one can now radically question and answer.”
And we are in a position to pursue how Strauss, beginning with the lecture
“‘Religiöse Lage der Gegenwart,’” introduces the figure of the “second cave”
and develops it step by step, a figure that he first uses in print in the Ebbinghaus
review.40 There is more: The critique of “conspectivism,” which Strauss intended
to rework into a critique of the “sophistry of our times,”41 can be read as an
early example of Strauss’s talent for irony and ridicule or for that “laughter”
that “belongs essentially to all enlightenment, be it Platonic or modern.”42 The
lecture bearing the title “‘Religiöse Lage der Gegenwart’” shows clearly the
ease with which Strauss is able to transform a prescribed theme into his own, his
ability to raise the theme to the level of the confrontation that seems appropri-
ate to him. In “Cohen und Maimuni” we encounter the first presentation of the
new Maimonides interpretation, which in large part found its way into Part IV
of Philosophie und Gesetz.43 Finally, “Die geistige Lage der Gegenwart” presents
us with Strauss’s historical self-reflection as an answer to historical consciousness
and sheds light on the role that Nietzsche plays in this self-reflection.
The third section of Frühe Schriften contains the introductions that Strauss
contributed to the Jubilee Edition of Moses Mendelssohn’s Gesammelte Schriften.
Strauss first wrote them on behalf of the Akademie für die Wissenschaft des
Judentums, where he worked from February 1925 to October 1932 and which
entrusted him with quite a number of Mendelssohn’s philosophical writings.44
The first two volumes of the jubilee edition that Strauss coedited, the Schriften
zur Philosophie und Ästhetik II and III 1, appeared in 1931 and 1932. Volume III 2,
which Strauss edited alone and on which he had made a great deal of progress
by 1932,45 could no longer be published due to the political circumstances in
the 1930s. In addition to his studies of Hobbes and medieval philosophy, Strauss
nevertheless continued his work on the edition of Sache Gottes oder die gerettetee
Vorsehung, as well as of Morgenstunden and An die Freunde Lessings. Strauss pub-
lished the introduction to Sache Gottes, the typescript of which was finished
in 1936, as a contribution to the festschrift for Krüger in 1962.46 By contrast,
he would not live to see the publication of the long essay on Morgenstunden
and An die Freunde Lessings, which he likewise completed during his stay in
Cambridge:47 volume III 2, and with it Strauss’s account of the quarrel between
Mendelssohn and Jacobi about Lessing’s Spinozism, which reads like a detective
story, appeared only in 1974, several months after Strauss’s death, in the con-
tinuation of the jubilee edition under Alexander Altmann’s direction. After the
Second World War, Strauss considered using the unpublished manuscript for an
essay on the quarrel over Lessing entitled “A Controversy on Spinoza,” which
would form the penultimate chapter of a book consisting partly of published
essays and
a d partly
pa t y of
o some
so e that
t at were
we e still
st to be written.
w tte . Thee volume,
vo u e, which
w c
HOW S T R AU S S BE C A M E S T R AU S S 23

Strauss estimated would be 350 pages in length and which he planned to pub-
lish in 1948, had the working title Philosophy and The Law: Historical Essays.
The titles of the 12 chapters, which Strauss sketched in a typed outline, read:
“I) Modern Jewish philosophy and its limitations. II) Jerusalem and Athens.
III) The two faces of Socrates. IV) How to study Jewish medieval philosophy.
V) Maimonides’s political science. VI) Maimonides’s ethics. VII) The liter-
ary character of the Guide for the Perplexed. VIII) Fârâbî’s treatise on Plato’s
philosophy. IX) The Law of Reason in the Cuzari. X) Persecution and the art
of writing. XI) A controversy on Spinoza. XII) Nathan the Wise.”48 This book
project, which was never realized, was replaced by the collection Persecution andd
the Art of Writing, which appeared in 1952.
The chapter on Nathan der Weisee was never written, like the book Strauss had
planned to write about Lessing in Cambridge in 1937. Thus on the author who
was particularly important for Strauss’s rediscovery of the distinction between
exoteric and esoteric speech,49 we have—besides the few pages that are devoted
to Lessing in the essay “Exoteric Teaching” (which was written in Decemberr
1939, remained a fragment, and was published only posthumously)50—only the
preface to Eine Erinnerung an Lessing, which concludes the second volume off
the Gesammelte Schriften. It ends with the words: “Incidentally, the author was
not unmindful of the debt of thanks that his nation owes to that great son off
the German nation, especially in this moment of farewell.” Strauss first wrote
“moment of parting” and then replaced “parting” (Trennung) with “farewell”
(Abschied). Toward the end of his life, he recalls in a letter to Alexander Altmann
the title of the book on Lessing he had once planned, Abschied von Deutschlandd
(Farewell to Germany): “In 1937 I had the intention to give an account of the
center of Lessing’s thoughts de Deo et mundo either in the conclusion of the intro-
duction [to Morgenstunden and An die Freunde Lessings] or in a separate text (underr
the title Abschied von Deutschland). The decisive points are as clear to me today as
they were back then. But ‘Making plans is often . . . ’ All that I could do was to
point my better students emphatically to Lessing and to say on appropriate occa-
sions what I owe to Lessing.”51
Translated by Marcus Brainard

Notes
1. As in the case of every philosopher, for an adequate study of Strauss’s thought it is
at least advisable—if not required—that one study his writings in the languages
in which they were written. The present essay is a revised and expanded English
version of my introduction to GS–2 ix–xxxiii.
2. Ludwig Feuchtwanger, “Philosophie und Gesetz. Bemerkungen zu zwei neuen
Arbeiten zum Verständnis Maimunis,” Jüdische Rundschau no. 29 (Berlin, April 9,
1935): 7. Feuchtwanger had been the editorial director of the publishing house
Duncker & Humblot from 1913 to 1933 in Munich and a close acquaintance off
Carl Schmitt since the First World War. At the end of 1932 he published Strauss’s
essay “Das Testament Spinozas” in the Bayerische Israelitische Gemeindezeitungg (now
in GS–1 414–22). On April 15, 1935, Feuchtwanger writes to Strauss: “I recently
reviewed
e e ed youyour Maimonides
a o des boo
book in the
t e Rundschau
Rundschau.. I enclose a copy of the paper
24 H EIN RICH M EIER

for you. I feel bad about it because I am well aware of the particularly demand-
ing requirements for being able to judge it: a knowledge of Aristotle and Arabic
studies would be the bare minimum. It would also be necessary to go into the wholee
of ‘existentialist philosophy’ and ontology. [ . . . ] In 1932 the author of The Conceptt
of the Politicall spoke to me about you with high esteem; but his praise was quite
unnecessary, because I knew your book on Sp. If I had not reviewed your book,
no one would have reviewed it in the Rundschau. Take that as the extenuatingg
circumstance. [ . . . ] I have high hopes for you as one of a very few who have somethingg
to say” (Leo Strauss Papers, Box 1, Folder 13, Department of Special Collections,
University of Chicago Library).
3. Thus in the case of his monograph Hobbes’ politische Wissenschaft, which obvi-
ously dealt with a “non-Jewish subject,” all of Strauss’s attempts in 1935 to find
a German publisher, or at least a German journal that would be willing to pub-
lish the study in several parts, came to nothing. The book ultimately appeared
in English translation in 1936 under the title The Political Philosophy of Hobbes:
Its Basis and Its Genesis at the Clarendon Press in Oxford, whereas the German
original was not published until 1965.
4. As a peripheral phenomenon, we mention the attack of the National Socialist
philosophy professor Hans Alfred Grunsky, who lumped Strauss together with
Hermann Cohen, Maimonides, and Philo and then reproached Philosophie undd
Gesetz for a “falsification of Plato” from the point of view of “Jewish prophet-
ism,” which struck him as “all the more brazen” since Strauss “knows very well
that the concept of the prophet is not Platonic and precisely because it presupposes
the notion of revelation, which is completely foreign to Plato” (Der Einbruch des
Judentums in die Philosophiee (Berlin: Junker und Dünnhaupt, 1937), 15–16).
5. Shlomo Pines reviewed Philosophie und Gesetz in Recherches Philosophiques V (Paris,
1936), 504–7. Almost three decades later he and Strauss edited the epoch-makingg
edition of Maimonides’s The Guide of the Perplexedd (Chicago: The University off
Chicago Press, 1963), to which Pines, assisted by Ralph Lerner, contributed a
painstaking translation of the Moreh nebuchim from the Arabic original, and a
historical introduction, and for which Strauss wrote his famous essay “How To
Begin To Study The Guide of the Perplexed” (xi–lvi; reprinted in LAM M 140–84).
6. Letters to Fritz Baer from September 24, 1933 and Gershom Scholem from
December 7, 1933. (Both letters, as well as all other letters to Scholem cited in the
present essay, are located under the signature “Arc. 4°1599/Corresp. Leo Strauss”
in the Department of Manuscripts & Archives as part of the collection of the
Jewish National and University Library, Jerusalem. Photocopies are located in Leo
Strauss Papers, Box 18, Folder 1, Department of Special Collections, University
of Chicago Library. Strauss’s correspondence with Scholem has been published
in GS–3 699–772. The letter referred to here may be found on 706–9. GS–3 also
contains the correspondence between Strauss and Gerhard Krüger (377–454),
Strauss and Jacob Klein (455–605), and Strauss and Karl Löwith (607–95), all in
the original languages in which the letters were written.) On December 7, 1933,
Strauss informs Gerhard Krüger: “I have just heard that I have certain prospects
of getting the professorship for Jewish medieval philosophy in Jerusalem” (GS–3
436). In a letter from December 31, 1933, to Jacob Klein, his closest friend from
the 1920s and 1930s, Strauss explains his stance as follows (he apparently knew
nothing of Julius Guttmann’s candidacy at the time): “My sincere thanks to you,
as well as to Krüger and Gadamer, for your Palestine-Buber operation. I have no
illusions about the fact that [Simon]
[ ] Rawidowicz will ppresumablyy come out on
HOW S T R AU S S BE C A M E S T R AU S S 25

top, and I am not even sad about that: Palestine means abandoning Hobbes and
thus much else besides. Nevertheless, I cannot afford to leave any stone unturned
so that we three aren’t left with nothing on October 1st. Besides, I am hopingg
that, should R. get the position in Jerusalem, the university there will do its
best to help me in England or America. Naturally, I would like most to stay in
England—indeed, everythingg speaks for this country . . . ” (GS–3 485).
7. In a letter to Krüger from July 25, 1931, Strauss writes: “I am enclosing my
essay for you. I would appreciate it if you would take a look at it soon. Pages 1–5
and 23–37 would be enough for you; the middle section is far too ‘medieval’”
(GS–3 391). On December 7, 1933, Strauss describes the background of the essay
to Scholem: “This study is meant as groundwork for a history of prophetology
in the Middle Ages for which I have collected a lot of material but which I can
begin to elaborate only after the provisional conclusion of my Hobbes study (in
two treatises on Hobbes’s critique of religion and on Hegel and Hobbes). The idea
for this larger investigation of things medieval grew out of several years of work
on Gersonides’s Milchamoth ha-shem and Aristotle- (or Averroes-) commentaries”
(GS–3 708).
8. On September 15, 1933, Strauss mentions to Klein “that I have written a long
review of Guttmann’s book, which is destined to become part of my posthumous
papers.” On September 24, he sends the essay to Scholem. On December 3, 1933,
Strauss asks Krüger: “Has he [Klein] given you my Guttmann critique?” On
October 10, 1933, Strauss tells Klein: “In the meantime, after a year’s hesitation,
I sent my Guttmann review to Guttmann. The result was surprising: he was
delighted and honored, and he wants to try to get it published” (see GS–3 470,
702, 436, and 523).
9. On the same day Strauss writes to Guttmann to the same effect (Leo Strauss
Papers, Box 1, Folder 14). The report from November 20, 1933, by the Orientalist
Hans Heinrich Schaeder (Berlin) gives the following judgment: “Dr. Strauss
combines to an exceptional degree the abilities of a responsible philosophizingg
that unwaveringly gets to the bottom of the matter and a historical ref lection that
seeks to be wholly concrete and that does not shy away from any investigation off
detail. He has an advantage over others who have oriented their study towards the
entire development of the problem of religion and philosophy, with his intimate
and sure knowledge of the state of the problem in the Age of the Enlightenment.
[ . . . ] His studies on the doctrine of prophetology in the Islamic and Jewish
Middle Ages, in which he is currently engaged and which he sketches in the two
aforementioned unpublished essays, signify, if I understand correctly, the openingg
up of an essentially new and incalculably fruitful perspective for the judgment off
the medieval discussion of the problem of revelation and reason. He finds a way
here to elucidate the doctrine of the bringer of revelation no longer with respect
to the constricting perspectives of modern Romantic philosophy of religion, but
rather based on the tradition of the ancient, in particular the Platonic, doctrine off
the state and the laws, and once again to give the ancient division of philosophy
its due in the consideration of its continuations in the Middle Ages. For all theirr
subtlety, his studies are just as powerful and courageous in the progression off
thought as their language is terse. The historical investigation does not once get
lost in antiquarian concerns, but remains ever philosophically relevant. Whoeverr
studies the history of religion and the history of thought in the philosophy off
religion today will have much to learn from these studies again and again and
will have to take a stand with respect
p to them. In the areas in which he works,
26 H EIN RICH M EIER

there are hardly any subjects of inquiry that are more pressing than his and there
is hardly anyone better prepared to do the job than he.” (A copy of the evaluation
is located in Correspondence Leo Strauss, Arc. 4°1599 of the Jewish National and
University Library Jerusalem.)
10. LAM M 257. Consider also Strauss’s letter of June 23, 1935, to Karl Löwith (GS–3
648). Strauss’s “Anmerkungen zu Carl Schmitt, Der Begriff des Politischen” has been
republished in GS–3 217–38. An English translation may be found under the title
“Notes on Carl Schmitt, The Concept of the Political” in my Carl Schmitt and Leo
Strauss: The Hidden Dialogue, trans. J. Harvey Lomax (Chicago: The University off
Chicago Press, 1995; rev. paperback ed., 2006), 89–120.
11. From the correspondence or based on the dates that Strauss himself entered
into his personal copies (P), insofar as they are available, the following chro-
nology can be reconstructed for the writings of relevance to our question:
1929 (P): “Der Konspektivismus”; until December 1930: “‘Religiöse Lage der
Gegenwart’”; until May 1931: “Cohen und Maimuni”; by July 1931: Philosophiee
und Gesetz, Part IV; circa October 1931: “Besprechung von Julius Ebbinghaus”;
end of October/beginning of November 1931: “Disposition” of a book entitled
Die politische Wissenschaft des Hobbes. Eine Einführung in das Naturrecht ; end off
1931: “Vorwort zu einem geplanten Buch über Hobbes”; by February 1932:
“Die geistige Lage der Gegenwart”; April–May 1932 (P): “Anmerkungen zu
Carl Schmitt, Der Begriff des Politischen”; end of 1932/beginning of 1933: “Einige
Anmerkungen über die politische Wissenschaft des Hobbes”; 1933–34: Diee
Religionskritik des Hobbes; by September 1933: Philosophie und Gesetz, Part II; circa
December 1934–January 1935: Philosophie und Gesetz, Part III (based on a manu-
script from 1932?); January–February 1935 (P): Philosophie und Gesetz, Part I; 1934
by May 1935: Hobbes’ politische Wissenschaft; circa second half of 1935: changes
and supplements to the text of The Political Philosophy of Hobbes that go beyond
the German original; August–October 1935 (P): “Quelques remarques sur la sci-
ence politique de Maïmonide et de Fârâbî”; November 1935 (P): “Eine vermißte
Schrift Farâbîs”; May 1936: “Notes additionelles” to “Quelques remarques sur
la science politique de Maïmonide et de Fârâbî”; August 1936 (P): “Der Ort
der Vorsehungslehre nach der Ansicht Maimunis”; April–August 1937 (P): “On
Abravanel’s Philosophical Tendency and Political Teaching.”
12. An in-depth confrontation may be found in my Die Denkbewegung von Leo Strauss.
Die Geschichte der Philosophie und die Intention des Philosophen (Stuttgart/Weimar:
J. B. Metzler, 1996), 19–43, and in Das theologisch-politische Problem. Zum Thema
von Leo Strauss (Stuttgart/Weimar: J. B. Metzler, 2003), 13–48. (An English trans-
lation of both texts is contained in my book Leo Strauss and the Theologico-Politicall
Problem (Cambridge/New York: Cambridge University Press, 2006), 53–73 andd
3–28 respectively.)
13. AAPL L 1 and 2; “On Abravanel’s Philosophical Tendency and Political Teaching,”
in GS–2 198; cf. Philosophie und Gesetz, GS–2 112, 114, “Quelques remarques
sur la science politique de Maïmonide et de Fârâbî,” GS–2 126, “Cohen und
Maimuni,” GS–2 425, as well as “The Spirit of Sparta or the Taste of Xenophon,”
Social Research 6 (1939): 530–32.
14. After concluding the Spinoza book, Strauss’s research concentrated on Hobbes
on the one hand and medieval thought on the other. On June 26, 1930, he offers
to give a talk in Krüger’s Augustine seminar on “Enlightenment in the Middle
Ages.” He adds the following explanation: “I had begun my work on a Jewish
Scholastic—Gersonides— purelyp y in the ‘interest of learning’
g [als
[ reine ‘Lernarbeit’],
],
HOW S T R AU S S BE C A M E S T R AU S S 27

but also because, of course, I have to give the people who pay me some writings
or other. I soon saw, however, that the work couldn’t be done so mindlessly, sim-
ply because the subject is so exciting” (GS–3 382).
15. Philosophie und Gesetz, GS–2 57–58, 59, 90, 111, 116; “Quelques remarques,”
GS–2 125, 126, 132–33, 136–37, 139; “Eine vermißte Schrift Farâbîs,” GS–2
170; “Der Ort der Vorsehungslehre nach der Ansicht Maimunis,” GS–2 182–84,
187, 189.
16. Philosophie und Gesetz, GS–2 60, 114–15, 123; “Quelques remarques,”
GS–2 143.
17. Philosophie und Gesetz, GS–2 64. Cf. GS–2 49, 86, 118; AAPL L 7–11.
18. Cf. TM M 288–90, 291–92; furthermore, my Denkbewegung von Leo Strauss, 26 (Leo
Strauss and the Theologico-Political Problem, 60).
19. Philosophie und Gesetz, GS–2 61; “Quelques remarques,” GS–2 126. “ . . . law is
originally nothing other than the way of life of the community. The first things
and the right way cannot become questionable or the object of a quest, or phi-
losophy cannot emerge, or nature cannot be discovered, if authority as such is not
doubted or as long as at least any general statement of any being whatsoever is
accepted on trust” (NRH H 84).
20. Philosophie und Gesetz, GS–2 48, 60, 62, 66, 67–68, 78; “Quelques remarques,”
GS–2 125, 128, 152; “On Abravanel’s Philosophical Tendency,” GS–2 198–99.
21. Philosophie und Gesetz, GS–2 46, 47, 67, 71, 74, 76, 78, 88–89, 115 (cf. the variant to
123, which is found on 612, and “Cohen und Maimuni,” GS–2 425); “Quelques
remarques,” GS–2 137, 144–45, 148, 152, 156; “Der Ort der Vorsehungslehre,”
GS–2 184, 186–87; “On Abravanel’s Philosophical Tendency,” GS–2 197–98.
22. Philosophie und Gesetz, GS–2 9–10 (my emphasis, H.M.); Hobbes’ politischee
Wissenschaft, in GS–3 162–63. On March 27, 1935, Strauss writes to Krüger: “To
avoid a possible misunderstanding: ‘Sophistry’ on the first page of my introduction
is meant quite literally (after Protagoras’s myth): on the basis of an Epimethean
physics (the exposedness of man) to surrender ultimately to what the Athenians
say” (GS–3 442).
23. Philosophie und Gesetz, GS–2 14, 25–26; Hobbes’ politische Wissenschaft, GS–3
166–69.
24. “It is not courage which is the highest virtue—self-mastery stands higher, and
higher still than self-mastery stand wisdom and justice. In itself wisdom stands
supreme, but justice stands supreme from an exoteric point of view. This explains why
Plato does not assert, as does Aristotle, the superiority of the theoretical life to
ethical virtue” (The Political Philosophy of Hobbes (1936) 147; my emphasis, H.M.).
In Hobbes’ politische Wissenschaftt this passage read originally: “It is not courage that
is the highest virtue—moderation stands higher than courage, higher still than
moderation stand insight and justice. And in fact it is insight in itself that stands
highest, though for man it is justice [that stands highest]. This explains why Plato
does not assert, as does Aristotle, the superiority of the theoretical life to ethical
virtue” (GS–3 167–68). Cf. Philosophie und Gesetz, GS–2 122–23.
25. “Quelques remarques,” GS–2 128–30, 136, 142, 150–51, 156–58 (cf. “On
Abravanel’s Philosophical Tendency,” GS–2 197, 205–6); The Political Philosophy
of Hobbes (1936) 161–63. The long paragraph on the primacy of foreign policy
is not in Hobbes’ politische Wissenschaft. It was added to the English text roughly
at the same time as Strauss wrote “Quelques remarques.” For the importance off
this change, consider what Strauss suggests in Thoughts on Machiavelli, 298–99
on the ancients and the moderns regardingg g “the art of war” and what the book
28 H EIN RICH M EIER

as a whole says about Machiavelli’s “spiritual warfare.” Cf. my Denkbewegung


von Leo Strauss, 24–28 (Leo Strauss and the Theologico-Political Problem, 59–61). (In
1952, in the foreword to the American edition of the Hobbes book [xix], Strauss
formally corrected his mistaken view that Hobbes was the founder of mod-
ern political philosophy: “Hobbes appeared to me as the originator of modern
political philosophy. This was an error: not Hobbes, but Machiavelli, deserves
this honor.”)—On October 2, 1935, just prior to completing the manuscript off
“Quelques remarques,” Strauss writes to Scholem: “What I claimed in this book
[Philosophie und Gesetz] purely on historical grounds, I have since come to find to
be totally insufficient: Cohen was much more right [in claiming that Maimonides
was a Platonist] than I ever would have dreamed possible (of course, his reasons
are and remain misguided, indeed absurd). What happened there between 900
and 1200 looks completely different than it appears in the light of the reception
of Islamic philosophy by Christianity (and this reception—perhaps even more
than the specifically modern prejudices—is the reason for the prevailing view).
If I have the time and energy, I want to write a book about the Moreh over the
course of roughly ten years. For the time being I am publishing an introduction
to the Moreh under the title: Hobbes’s political science in its development, which
should be published next year by Oxford Press” (GS–3 716). On the political
dimension of the confrontation with revealed religions into which the Arabic orr
the modern founding fathers entered in order to recover philosophy, see also my
preface to GS–3 xxiii–xxiv, and the newly discovered letter to Paul Kraus that is
cited there.
26. Cf. Die Religionskritik Spinozas, GS–2 208, 238, 254; PAW W 81, 124–26; letters
to Jacob Klein from February 16 and July 23, 1938, GS–3 549–50 and 553–54,
respectively. – In the first sentence of “How To Begin To Study The Guide of thee
Perplexed,” Strauss states that Philosophie und Gesetz was not based on an adequate
understanding of the Moreh: “I believe that it will not be amiss if I simply present
the plan of the Guidee as it has become clear to me in the course of about twenty-
five years of frequently interrupted but never abandoned study” (LAM M 140).
In the manuscript on which Strauss worked from May 19 to August 13, 1960,
he first wrote “36 years,” which would have put us in 1924, the year in which
“Cohens Analyse der Bibel-Wissenschaft Spinozas” was published. Strauss then
changed it to “25 years” and underscored it in red, which places the beginningg
of Strauss’s study of the plan of the Moreh in the period following the publication
of Philosophie und Gesetz. After having written “The Literary Character of the
Guide for the Perplexed,” Strauss never left any doubt that a philosophical work on
the order of the Moreh could not be adequately understood so long as the readerr
did not have its plan clearly in view, that is, as long as it is not also understood as
a work of art.
27. In GA the following oral statement by Strauss is recorded: “Maimonides was, to
begin with, wholly unintelligible to me. I got the first glimmer of light when I
concentrated on his prophetology and, therefore, the prophetology of the Islamic
philosophers who preceded him. One day when reading in a Latin translation
Avicenna’s treatise, On the Division of the Sciences, I came across this sentence (I
quote from memory): the standard work on prophecy and revelation is Plato’s
Laws. Then I began to begin to understand Maimonides’s prophetology and
eventually, as I believe, the whole Guide of the Perplexed. Maimonides never calls
himself a philosopher; he presents himself as an opponent of the philosophers. He
used a kind of writing which is, in the precise sense of the term, exoteric. When
HOW S T R AU S S BE C A M E S T R AU S S 29

Klein had read the manuscript of my essay on the literary character of the Guide
of the Perplexed, he said, ‘We have rediscovered exotericism.’”
28. Philosophie und Gesetz, GS–2 51–52, 55, 77, 82. As for what status the claim off
insufficiency has in Philosophie und Gesetz, that can remain open here. Considerr
on this point 58–59 and 64, as well as the question at the end of the introduction:
“Is it [Maimonides’s enlightenment] not even in some respects more ‘radical,’
more dangerous to the spirit of Judaism, than the modern Enlightenment alto-
gether?” In “Quelques remarques” Strauss begins his answer as follows: “Il reste
à savoir si Maïmonide, lui aussi, regarde la fondation de la cité parfaite comme
la raison d’être de la révélation. On a penséé que, selon lui, le but principal de la
révélation était la proclamation des vérités les plus importantes, surtout de celles
qui ne sont pas accessibles à la raison humaine” (GS–2 143, my emphasis, H.M.).
29. As is clear from two letters Strauss wrote to Guttmann, in 1934 Guttmann had
sent Strauss a detailed response to his review of Philosophie des Judentums (letters
from December 14, 1934 and January 22, 1935). Later Guttmann wrote a critique
of Philosophie und Gesetz, though he did not publish it. On October 10, 1939,
Strauss mentions to Klein: “Kraus spoke with Guttmann in Jerusalem. Guttmann
told him that he is writing an article against me, to which K. responded that it
is too late since in the meantime I have a new interpretation of Maimonides”
(GS–3 583). On June 22, 1952—Guttmann had died in 1950—Strauss says in
a letter to Scholem: “I knew from Guttmann himself about the reply he had
begun. He wrote me several years ago (it may be 8–10 years already) that he had
ceased to work on the reply because I had given up the standpoint I had held in
Philosophie und Gesetz. That is correct insofar as I publicly agreed to G.’s thesis
about the identity of reason and revelation in the M.A., but my earlier rejection
is ‘sublated’ in my current agreement: I have moved, so to speak, contrary to G.’s
moderate rationalism, on the path via a Jewish Thomism to radical ‘rationalism,’
am now therefore on the right wing (for the right is truth, the left is sinister,
r as no
one knows better than you), whereas I stood on the left wing in Philosophie undd
Gesetz: Guttmann ever in the middle. (I am now w attempting to reach a moderate
‘rationalism,’ but one that, I am afraid, would be even less acceptable to G. than
my two earlier positions). Be that as it may, I am of the opinion that G.’s critique
is still highly relevant.” When in 1972 Scholem offered Strauss the typescript off
Guttmann’s critique, which was found in the latter’s posthumous papers, Strauss
expressed his interest (November 17, 1972; GS–3 765). On January 2, 1973, he
confirms receipt of it (GS–3 765). The essay, which was written between 1940
and 1945, was published by Pines shortly after Strauss’s death: Julius Guttmann,
Philosophie der Religion oder Philosophie des Gesetzes?? in Proceedings of the Israell
Academy of Sciences and Humanities V, 6 ( Jerusalem 1974).
30. “On Abravanel’s Philosophical Tendency,” GS–2 197–98, 200–4, 205–6, 207n33,
209, 215, 224–25. “The Literary Character” was published in 1941 and then
reprinted in PAW W 38–94.
31. Compare Philosophie und Gesetz, GS–2 17, 18, 19, 20, 23, 24–26, with LAM M 254,
255, 256. The final sentence of the analysis reads: “The last word and the ultimate
justification of the Enlightenment is the atheism from probity, which overcomes
orthodoxy radically by understanding it radically, that is, free of both the polemi-
cal bitterness of the Enlightenment and the equivocal reverence of romanticism”
(Philosophie und Gesetz, GS–2 26). In his “autobiographical preface” Strauss adds
what was—though not expressly formulated—no less clearly discernible as the
underlying
y g thrust of the critique:
q “Yet this claim, however eloquently
q y raised,
30 H EIN RICH M EIER

cannot deceive one about the fact that its basis is an act of will, of belief, and that
being based on belief is fatal to any philosophy” (“Preface to Spinoza’s Critique
of Religion,” LAM M 256).
32. Cf. Löwith’s letter from April 15, 1935. Strauss responds to it on June 23, 1935:
“Incidentally: I am nott an orthodox Jew!” (GS–3 645–47, 650). On February 14,
1934, Strauss remarks to Scholem in another context: “That I am not orthodox
and under no circumstances the right man for an orthodox institute since I can-
not make any concessions, I hardly need to tell you” (GS–3 711).
33. Guttmann, Philosophie der Religion oder Philosophie des Gesetzes? 6; cf. 26–27.
34. Letter from May 6, 1935 (GS–3 538–39). The characterization perscrutator cordis
meii [scrutinizer of my heart] is in Strauss’s letter to Krüger from August 19, 1932
(GS–3 399).
35. Gershom Scholem, ed., Walter Benjamin—Gershom Scholem. Briefwechsell
1933–1940 0 (Frankfurt a. M.: Suhrkamp, 1980), 192–93, 197.
36. “Now, while Mr. Strauss and I were studying we had many, I should say, endless
conversations about many things. His primary interests were two questions: one,
the question of God; and two, the question of politics. These questions were not
mine. I studied [ . . . ] Hegel, mathematics, and physics” (GA 1).
37. All the publications I knew of in 1997 were taken into account. It may be that
still more newspaper articles will turn up in the future. For example, while I was
preparing volume 2 for publication the article “Ecclesia militans,” which was not
known in the entire literature on Strauss, was brought to my attention and thus
could be included in the first edition of volume 2. In spring 1999, I discovered
three articles that Strauss had published in Der Jüdische Student, which I edited
in the second, revised, and enlarged edition of GS–1 in 2001: “Bemerkung zu
der Weinbergschen Kritik” (1925), “‘Die Zukunft einer Illusion’” (1928), and
“Zur Ideologie des politischen Zionismus” (1929). And after 2001 I discovered
more of Strauss’s writings on political Zionism from the 1920s: “Zionismus und
Orthodoxie” (1924) and “Quellen des Zionismus” (1924) as well as “Die ‘Jüdischen
Schriften’ Hermann Cohens” which are now contained in an appendix to the
third edition of GS–1 and to the second edition of GS–2 (2013) respectively.
38. GA 2.
39. Revue de métaphysique et de moralee 99, nos. 3–4 (1994): 291–311 and 505–32. The
translation is based on a photocopy of the Munich typescript, which I gave to the
Department of Special Collections at the University of Chicago in the 1980s.
40. On October 15, 1931, Strauss writes to Krüger: “I have now discovered a fourth
man who shares our opinion of the present as a second cave: Ebbinghaus. His lec-
ture Ueber die Fortschritte der Metaphysik contains several excellent formulations; I
am going to review the booklet in the D. L. Z.” (GS–3 394–95). On the “second
cave,” see “‘Religiöse Lage der Gegenwart,’” GS–2 386–87, “Besprechung von
Julius Ebbinghaus,” GS–2 438–39, “Die geistige Lage der Gegenwart,” GS–2
451–52, 455–56, 461, Philosophie und Gesetz, GS–2 14, 45, and PAW W 155–56. Cf.
my Denkbewegung von Leo Strauss, 21–28 and 42–43 (Leo Strauss and the Theologico-
political Problem, 56–61, 72–73).
41. In a letter to Krüger dated February 27, 1931, we read: “Enclosed, then, the
piece on conspectivism, which I kindly request you to pass on to Gogarten. I
am thinking of developing it in such a way as to fill out the critique to which
I allude in the second half, in particular by showing how Mannheim remains
completely ‘helpless’ insofar as he asks about politics as a science and about uto-
p without havingg been enlightened
pia g byy Plato (Pardon
( the barbaric sentence!).
)
HOW S T R AU S S BE C A M E S T R AU S S 31

I give expression to this tendency by entitling the whole thing: Sophistry of Our
Times. I shall work in the theses that I expounded in my lecture on the religious
situation of the present (the 2d cave, etc.). My hope is that the essay, revised
along the lines I am planning now, will seem convincing to you and will amuse
you, at least that it will be far better than it is now. Thus: if Gogarten is preparedd
to recommend the essay, enlarged and improved, to a suitable publisher, whether
as a contribution to a journal, or as a separate booklet, then I shall be happy to
begin reworking it” (GS–3 384). In Leo Strauss Papers, Box 8, Folder 4, there is
a two-page, handwritten outline with the title “Sophistik der Zeit” (Sophistry
of Our Times).
42. Letter to Krüger from July 17, 1933 (GS–3 431).
43. On May 7, 1931, Strauss tells Krüger about the lecture that he had given on May
4 in the auditorium of the Hochschule für die Wissenschaft des Judentums in
Berlin: “Last Monday I gave a lecture on Cohen and Maimonides. I attempted to
show that Cohen was nevertheless right in saying that Maimonides was basicallyy a
Platonist, and nott an Aristotelian; of course, one cannot show that as directlyy as did
C. In this lecture I made public for the first time my thesis about Islamic-Jewish
scholasticism (that it understands revelation within the framework laid out by
Plato’s Republicc and Laws). Too bad that you were not there; I would have liked to
have heard your opinion. You would also have seen just how much I have profitedd
from your lecture course on Plato” (GS–3 385).
44. Letter to Cyrus Adler from November 30, 1933 (Leo Strauss Papers, Box 4,
Folder 1). When the Akademie ran into financial difficulties at the end of 1931,
all its employees had to be given notice.
45. On August 21, 1932, Strauss writes to Krüger that he wants to “conclude” the
commentary on Mendelssohn’s Morgenstunden “if it is at all possible, in the nearr
future.” On February 7, 1933, he mentions that he is working on the introduction
to Sache Gottes (GS–3 400, 426).
46. On September 12, 1962, Strauss writes to Krüger: “It was extremely good off
you to write to me about my contribution to the festschrift. I thought that it
was fitting for the purpose because of your deep interest in Leibniz. I regret that
by a grave error of the publisher the error was created that the article had been
published before: it was written in 1936 for Volume III b of the Jubilee Edition
of Mendelssohn’s works, and the volume could no longer appear because of the
situation at that time” (GS–3 453).
47. On May 31, 1937, Strauss writes to Guttmann, the editor in charge of the Jubilee
Edition: “As for the Mendelssohn introduction, I have already decided not
to consider at all the effects and consequences of the quarrel [about Lessing’s
Spinozism]. But the history of the genesis [of Morgenstunden] is so opaque and so
little explained that a certain thoroughness cannot be avoided, particularly since
a material understanding of Chapters XIII–XV of Morgenstunden depends entirely
on an understanding of its genesis. It is, so it seems to me, of considerable signifi-
cance that the concept of purified Spinozism never occurs in the whole prehistory
[of the quarrel]. Incidentally, I hope to be able to submit the introduction soon forr
your assessment.”
48. Plan of a book tentatively entitled Philosophy and The Law: Historical Essays. 5
pages. Leo Strauss Papers, Box 11, Folder 11. At the time of the outline, essays
VII, VIII, IX, and X had already been published, a typescript of essay IV existed,
which appeared posthumously, and Strauss wrote and published essays II and V
later. (Cf.
( Die Denkbewegung
g g von Leo Strauss,, Bibliography: A 18, B 29, 31, 36, 38,
32 H EIN RICH M EIER

61, 129; and Leo Strauss 1899–1973: A Bibliography


bl h by Heinrich Meier, The Leo
Strauss Center, Chicago: A 18, B 35, 37, 42, 44, 67, 137.)
49. In GA 3, Strauss reports in 1970: “In this study [of Spinoza’s Theologico-Politicall
Treatise] I was greatly assisted by Lessing, especially his theological writings, some
of them with forbidding titles. Incidentally, Lessing is also the author of the only
improvised live dialogue on a philosophic subject known to me. Lessing was
always at my elbow. This meant that I learned more from him than I knew at the
time. As I came to see later Lessing had said everything I had found out about
the distinction between exoteric and esoteric speech and its grounds.” Already in
1948 Strauss had expressed an emphatic word of thanks to Lessing at the end off
his lecture, “Reason and Revelation,” which he gave at the Hartford Theological
Seminary in Hartford, Connecticut: “In conclusion, I would like to name that
man to whom I owe, so to say, everything I have been able to discern in the
labyrinth of that grave question: Lessing. I do not mean the Lessing of a cer-
tain tradition, the Lessing celebrated by a certain type of oratory, but the true
and unknown Lessing. Lessing’s attitude was characterized by an innate disgust
against compromises in serious, i.e. theoretical, matters [ . . . ] He decided in favorr
of philosophy.” “Reason and Revelation” in Leo Strauss and the Theologico-Politicall
Problem, 178.
50. It was first published in Interpretation 14 (1986): 51–59. Unfortunately, the edi-
tor failed to inform the reader that the text presented is only the first part of a
longer work, the manuscript of which breaks off after the heading “II,” which
marks the unwritten second part. Its fragmentary character may also explain why
Strauss chose not to publish the text. However, a note among his posthumous
papers reveals the intended structure of the treatise. The plan of Part II contains
the following entries: “7. Aristotle’s ‘exoteric’ writings. 8. Cicero. 9. Xenophon.
Cyneg. 10. Plato’s Letters. 11. Plato’s Dialogues. Phaedrus, Rep., Timaeus.
12. Plato on the poets and Hesiod on the Muses. 13. Herakleitus. 14. The big
exceptions: Epicurus and Sophists. Cic. Rep. III” (Leo Strauss Papers, Box 12,
Folder 2). See now the critical edition by Hannes Kerber in this volume.
51. The letter to Alexander Altmann from May 28, 1971, is printed in Altmann’s
preface to volume III 2 of the Jubilee Edition (Stuttgart/Bad Cannstatt:
Frommann-Holzboog, 1974), viii. (Cf. now LSMM M 4–5.)
C APT R 2
CHAPTER

SPINOZA’S CRITIQUE OF RELIGION:


N
READING THE LOW IN LIGHT OF THE HIGH*

Steven Frankel

S trauss’s early studies of Spinoza, including his first book, Spinoza’s Critiquee
of Religion (1930; henceforth SCR), have been largely neglected in favor off
his later work. Such neglect is understandable: Strauss’s work on Spinoza spans
his entire career as a scholar and thinker, and includes his discovery of esoteri-
cism—which played a critical role in his most authoritative analysis of Spinoza.
In addition, Strauss privately conceded problems with the work. In a letterr
to Gerhard Krüger in 1930, Strauss conceded that he had been compelled to
remain silent in public about the presuppositions that were the point of depar-
ture for SCR.1 Later he criticized the work publicly for not taking seriously
the possibility of return to premodern philosophy (¶¶21, 42). It is somewhat
surprising then that more than 30 years later, after its publication in German,
Strauss decided to have the book translated into English. Ostensibly to explain
his decision, Strauss prefaces the translation with an autobiographical account off
the genesis and development of his early views as well as the inclusion of a later
essay on Carl Schmitt (1932). One can recognize many of the themes in SCR— R—
for example, the tension between Athens and Jerusalem, the inadequacy of the
Enlightenment’s critique of religion, and the development of Epicureanism,
and others—which would preoccupy Strauss over the course of his career. Still,
Strauss’s decision to resurrect this early work is puzzling.
The studied carelessness with which Strauss handles his autobiographical
account is in striking contrast to the attention he pays to the substance of the
preface. This becomes clear when we contrast the literary form of the essay with
the substantive argument: The literary form is an autobiographical narrative of a
young German Jew who seeks to escape the “theological-political predicament”
by either returning to the form of Jewish belief or else discovering a political
alternative, such as Zionism or liberalism (see ¶¶1–13). After pointing to the dif-
ficulties with both of these political alternatives, Strauss seeks a qualified return
34 STEVEN FRANK EL

to Jewish belief mediated by modern thought (see ¶¶14–23). When this proves
unworkable, he considers the return to “unqualified Jewish belief ” or orthodoxy.
This leads him to consider Spinoza’s critique of Judaism, and this is presumably
the context within which SCR R was written (see ¶¶24–39). Ultimately, Strauss
concludes that Spinoza’s critique undermines not only Judaism but also philoso-
phy, and he decides to return to “Jewish medieval rationalism and its classical
(Aristotelian and Platonic) foundation” (see ¶¶40–42).
This straightforward account of his intellectual development is under-
mined by the substance of his argument. Toward the center of his essay, Strauss
confounds the account of his development. 2 For example, as a young man,
Strauss reports that he was still considering the possibility of return to Jewish
thought as mediated by modern rationalism, particularly in the thought of Franz
Rosenzweig. According to his autobiographical narrative, he has not yet con-
sidered the possibility of return to premodern forms of rationalism, and in fact
would not do so until after the completion of SCR . This is the conclusion of his
autobiographical account: “I began therefore to wonder whether the destruc-
tion of reason was not the outcome of modern rationalism, as distinguished
from pre-modern rationalism” (¶42).
However, in ¶21 Strauss breaks dramatically from the autobiographical
narrative to assert the conclusion of the autobiographical account. After he criti-
cizes Rosenzweig’s new thinking, he writes: “One begins to wonder whetherr
our medieval philosophy, and the old thinking of Aristotle, of which it made
use, was not more ‘empirical’ . . . than an unqualified empiricism” (¶21). In
effect, Strauss has moved the conclusion of the autobiography to the middle
of the argument. This casts doubt on the central claim of the autobiography,
which presents Strauss as gradually discovering the inadequacy of modern rea-
son and the relevance of the ancients. Even if we read this claim as merely an
aside or parenthetical remark so that it does not disturb the overall narrative,
it is difficult to deny that Strauss is more concerned with emphasizing his con-
clusions than with giving an account or exploring the historical development
of his thought. 3 Throughout the essay, he presents several more examples off
interrupting the narrative to emphasize mature judgments that, though they
shape his account, emerge only after the biographical period portrayed in the
essay. One example, which we shall discuss in greater detail below, is Strauss’s
hermeneutical maxim of reading the low in light of the high (¶7). This principle
supersedes the autobiographical narrative in terms of shaping the content of the
essay. Strauss’s goal is not to encourage further reflection upon his autobiogra-
phy or offer an account of his intellectual development, but rather to guide our
attention toward his mature thought.

The Theologico-Political Problem


Strauss begins the preface by referring to himself as “a young Jew born and
raised in Germany who found himself in the grips of the theologico-political
predicament.”4 Unfortunately, Strauss does not explain the nature of the
t eo og co po t ca ppredicament,
theologico-political ed ca e t, o or indicate
d cate w
whatat spec
specifically
ca y about tthee
S P I N OZ A’S C R I T IQ U E O F R E L IG IO N 35

predicament constrained him. This is less of a mystery than it appears since the
preface itself is devoted to explaining how to read Spinoza’s Tractatus Theologico-
Politicus (henceforth TTP), which in term is intended as an introduction to
Strauss first book on the same topic. In order to see what is at issue, it is useful
therefore to remind ourselves of the main themes of the theologico-political
predicament as presented by Spinoza.
The TTP P begins with an account of the role of superstition in political life.
The perdurance of superstition creates a theological problem because men imag-
ine a divine source to explain their fortunes, and it also creates a political problem
because religious and political leaders are eager to manipulate them in order to
solidify their own power.5 Spinoza focuses on the particular theologico-political
problem in Christendom and asks his readers to consider why Christians preach
love and practice hatred (P.4.1–2). He goes on to suggest that Christianity has
been hijacked and vulgarized by unscrupulous men who pervert its teachings in
order to satisfy their ambitions for the wealth and power of ecclesiastical offices.
Among their most insidious methods for obscuring and twisting the word off
God is the importation of foreign or superstitious ideas into the Bible, partic-
ularly “the theories of Aristotelians and Platonists” (P.4.7).6 Spinoza’s account
assumes that reason can identify superstition because reason can in principle pro-
vide a full account of nature; nonetheless, philosophy is unable to adjudicate
disputes among superstitious, passionate individuals. To the contrary, philosophy
is quickly transformed into yet another instrument to bolster superstition. The
power of superstition also poses a political problem, because as superstitions gain
in intensity, the passions that accompany them also increase and threaten the
stability of the state.
The theologico-political predicament refers to the superstitious condition off
mankind and the subsequent manipulation of those superstitions by religious
and political leaders. Spinoza’s solution to the problem has both political and
theological elements. Theologically, he distinguishes the essential teachings off
Christianity so that they can be purged of superstition, and restored in author-
ity. The essential teaching of the Bible, and the minimum requirement forr
salvation, is the practice of charity and love toward all men. With her foun-
dations restored, Christianity can once again contribute to the peacefulness
of society. These few dogmas of Christianity do not exhaust our knowledge
of God. Some men may seek and achieve greater knowledge of God; how-
ever, such knowledge is not a theological or political requirement for salvation.
Politically, the pursuit of this knowledge is irrelevant since it is available to
only a few individuals (14.1.49–51). The state needs only to secure safety and
security for its citizens. For the most part, it leaves the citizens free to pursue
knowledge of God as they see fit—as long as that pursuit is consistent with cari-
tas. Spinoza envisions freedom, rather than reason or virtue, as the cornerstone
of political life with the hope that reason will flourish under such conditions.7
Strauss describes this regime as liberal democracy, and identifies Spinoza as its
founder.8
The general analysis of the theologico-political predicament, directed toward
a larger
a ge Christian
C st a audience,
aud e ce, was not
ot the
t e same
sa e problem
p ob e faced
aced by Strauss,
St auss, “aa young
you g
36 STEVEN FRANK EL

Jew in Germany.” The version that Strauss inherited reflected the success of
Spinoza’s political solution in Germany, as well as its particular reception in the
Jewish community. Spinoza’s analysis was attractive to German Jews for several
reasons: First, Spinoza himself was born a Jew and his argument involves a deep
knowledge of Jewish commentaries on the Hebrew Bible. Strauss writes that
many of his contemporaries in the Jewish community celebrated Spinoza “on
purely Jewish grounds” (¶¶28, 26). Second, Spinoza’s solution to the theologi-
co-political predicament provides for a society where both Jews and Christians
can live together in freedom despite differences in their private beliefs. Strauss
reports that many German Jews believed that, thanks to Spinoza, “the millennial
antagonism between Judaism and Christianity was about to disappear” (¶27).
The reason for this confidence is that Spinoza had shown Jews that their religious
law was purely political, and as a result, had become obsolete with the destruc-
tion of the Jewish state. This provided the theological argument for assimilation.
Third, Spinoza’s analysis of the Jews in Chapter three of the TTP P not only envi-
sions Jews and Christians living together but also considers the possibility that
Jews might rebuild their former state. Certainly, this argument was meant to
complement his analysis of the destruction of the Jewish state and not as a politi-
cal strategy; nonetheless, Spinoza was celebrated by some Jews as the founder off
Zionism (cf. ¶11).
Strauss was in the grips of the theologico-political predicament as described
by Spinoza. The various solutions that Spinoza suggests, namely secular liberal-
ism, religious liberalism, and Zionism, appear to be the only options available to
the young Strauss. The organization of the essay adopts the framework set out
by Spinoza, considering each possibility in turn. As we shall see, each solution
falls short and the thread connecting these failures is the audacious confidence in
the power and capacity of reason to recognize and redress superstition. Spinoza’s
solutions are political solutions to the problem of superstition since they concede
that most people will remain superstitious; nonetheless, they offer strategies forr
prudently managing superstition. By framing the issue in this way, as a matter off
reason ameliorating superstition, Spinoza suggests the political value of religion
for a stable community, while undermining the philosophical basis for revela-
tion. The result is that we are left with only unpalatable options. After turning to
Spinoza’s solutions to the theological political problem, Strauss next considers the
modern variants of Spinoza’s analysis, Heidegger’s atheism, and Rosenzweig’s
piety. Both fail to resolve the tension between politics, religion, and philosophy:
either we reject reason and embrace revelation, or else we reject revelation and,
with it, the well-being of our political communities.

Liberalism in Weimar
Strauss reports that the majority of German Jews celebrated Weimar and its
historical founder, Spinoza. Although Weimar originated with the defeat off
Germany and his humiliation in the Treaty of Versailles, she could trace her roots
to a nobler, deeper heritage in the French Revolution, and ultimately Spinoza’s
accou t of
account o the
t e theological-political
t eo og ca po t ca problem.
p ob e . From
o their
t e point
po t of
o view,
v ew, liberal
be a
S P I N OZ A’S C R I T IQ U E O F R E L IG IO N 37

democracy had deep roots in Germany so that their Jewish faith presented no
obstacle to assimilation. They interpreted Weimar in the best or highest light,
emphasizing its moderate character and its attempt to integrate the “principles
of 1789” with the “highest German tradition” (¶2). This tradition includes the
recognition of the rights of man and a well-organized government of highly
trained civil servants. Under this regime, Jews flourished and participated in the
cultural life of the Republic, and Jewish life was strong; it even developed a new
“science of Judaism” (¶7). Although the German tradition also included strong
anti-Jewish sentiments that dominated medieval society, liberalism corrected
this situation by protecting the rights of religious minorities and purging the
government of superstitious and irrational goals. No wonder Weimar appeared
to German Jews all the more precious and noble.9
In direct contrast to this view, the Zionists, with whom Strauss openly sided,
viewed Weimar as weak at its very foundations and unable to defend itself when
challenged by ruthless enemies who viewed the Jews as outsiders who threatened
German Kulturr with foreign culture and “Civilisation.”10 Strauss presents quota-
tions from Goethe, Nietzsche, and Heidegger that demonstrate the resilience off
anti-Jewish feelings in Germany (¶8). Ignoring their “precarious situation,” the
Jews of Weimar clung to the vain hope that liberalism would somehow prevail.
Strauss suggests that the election of Hindenburg in 1925 “showed everyone who
had eyes to seee that the Weimar Republic had only a short time to live,” because
the nonliberal tradition was “stronger in will” (¶3, emphasis added, S.F.). Strauss’s
phrase suggests that the weakness and vulnerability of Weimar were obvious, yet
few, particularly within the Jewish community, recognized it. The problem off
superstition comes to light in the first instance as the belief in the durability off
liberalism in Weimar.
History appears to have vindicated this Zionist critique of German Jewry.
Weimar is likely to be remembered as a cautionary tale of democracy without
strength, “of justice without a sword” (¶3). However, the Zionist vision was also
limited by the fact it is the product of the liberalism that it criticizes, rather than
of the Jewish tradition (cf. ¶11). The difficulty of making a balanced judgment on
Weimar is that it requires measuring the low elements against the higher ones.
Strauss suggests the following rule of thumb: “It is safer to understand the low in
the light of the high than the high in the light of the low. In doing the latter one
necessarily distorts the high, whereas in doing the former one does not deprive
the low of the freedom to reveal itself fully as what it is” (¶5).11 To judge Weimarr
only on its sorry record of weakness and collapse ignores its deeper and noblerr
roots and “its moderate, non-radical character” (¶2). As a principle of interpreta-
tion, taking one’s bearings from the highest rather than the lowest possibilities
seems to be a prudent rule of thumb.
But this principle is hardly self-evident. Strauss’s justification of this principle
on the grounds of safety raises the question: safer for whom? In the context off
the preface, where Strauss discusses repeatedly the “precarious” situation of the
Jews in Weimar and Europe, the primary consideration appears to be the safety
of the Jews in the Diaspora (¶¶6–8). His subsequent analysis highlights the dan-
ge s involved
gers vo ved in ignoring
g o g thet e lowest
owest political
po t ca elements
e e e ts in favor
avo of
o dreams
d ea s about
38 STEVEN FRANK EL

perpetual peace. He takes a particularly severe view of Hermann Cohen’s faith


in progress and his unwillingness to confront the less palatable side of power and
coercion in politics (cf. ¶¶32–33). Cohen was so concerned with interpreting the
high in light of the low that he underplayed the reality of the low. He condemned
Spinoza “for his Machiavellian-inspired hard-heartedness” only to fall victim to
“the opposite extreme” (¶32). But even Cohen appears to be a sober realist com-
pared to the “fantastic flights” that mesmerized those German Jews who feted
Spinoza as a savior (¶29). Strauss openly sides with the Zionists’ interpretation off
the Jewish Question, a position that confirms the wisdom of neglecting the high
in politics in order to protect oneself from the low. Nonetheless, as we shall see,
his analysis ultimately reveals the serious flaws with its neglect of the high.

Zionism as a Political Solution to the “Jewish Problem”


Spinoza inspired the Zionist movement by suggesting that there is a political
solution to the Jewish problem, which involves giving priority to politics overr
religion.12 In his critique of purely political Zionism, Strauss draws our atten-
tion to the work of Ahad Ha‘am (Asher Ginsberg, 1856–1927), “the founder off
cultural Zionism.” For Ahad Ha‘am, the theological questions had been already
settled by Darwin and science, both of which had allegedly proven that revelation
was false (¶15). This distinction helps to understand Strauss’s critique, specifically
how Zionism was largely correct in its analysis of liberalism in Germany and
yet not able to offer Jews a completely satisfying account of Judaism. In addi-
tion, this analysis of a political solution that is theologically inadequate casts the
theologico-political problem in sharper relief. The political situation of the Jews
involves questions of power and safety as well as theoretical questions about the
meaning of Judaism. To read the low in light of the high means to judge politi-
cal solutions in light of their higher, theoretical considerations. Strauss’s analysis
shows that, while it is dangerous to ignore practical political concerns, they need
to be separated from the higher theological questions. His critique of both the
Zionists and the assimilationists confirms this reading.
While Zionism has the virtue of grasping the limitations of liberalism in
Europe (cf. ¶6), it does not reject liberalism itself. Instead it attributes the fail-
ure of liberalism to the fact that the Jews were at the mercy of other nations.
Strauss quotes Herzl on the failure of assimilation: “We are a nation—the
enemy makes us a nation whether we like it or not” (¶10). The assimilationists
abandon Judaism in the vain hope that they will be welcomed into a universal
society. Political Zionism exposes the fact that liberalism, despite its promise
of a universal solution to the Jewish Question, cannot create a view of justice
strong enough to overcome discrimination. Liberalism cannot eliminate hatred
or superstition; at best, it can provide only legal—rather than social—equality.
But Strauss does not stop there. He argues that Zionism also betrays Judaism
in the belief that it can resolve the Jewish Question by treating it merely as a
problem of power, not divine punishment.
The virtue of Zionism was its sober claim that the Jewish Question could be
so ved by powe
solved power alone.
a o e. Even
ve cultural
cu tu a Zionism
o s attempts
atte pts to combine
co b e this
t s view
v ew
S P I N OZ A’S C R I T IQ U E O F R E L IG IO N 39

with a project for the revival of Judaism, but its account of the high is vague and
unsatisfying. As a result, cultural Zionism tends to slide in opposite directions,
“politics (power politics) and divine revelation” (¶12). This result follows from
Spinoza’s account of the Jewish Question. Cultural Zionism attempts to escape
the theological-political grip of Spinoza, but it cannot explain the basis of the
high; instead it either ignores the high in the case of political Zionism or reads
the high in light of the low in the case of cultural Zionism.

The Highest Goals of Political Life


Strauss exposes the failure of Zionism to resolve the Jewish Question as part off
a broader critique of liberalism. Following Spinoza, liberalism advocated the
separation of politics from religion on the grounds that such a separation would
allow politics to pursue more effectively the goals of safety and security.13 By
focusing on the low, liberalism promised to leave people the freedom to pursue
the high in safety. The problem is that this freedom also allows for superstition,
and thus liberalism does not solve the problem of discrimination. To do so, lib-
eralism would have to prevent discrimination by severely limiting freedom off
speech in both the public and the private sphere. This would, in effect, destroy
the private sphere and, with it, the raison d’être for the liberalism. The Zionists
recognized that the problem of superstition, and more particularly of anti-
Jewish beliefs, was not resolved by freedom of thought in the private sphere.
To the contrary, prejudice against Jews seemed to spread all the more widely
in liberal democracies. Liberalism cannot resolve the Jewish problem, or more
generally the problem of the imagination in political life. In Strauss’s words, the
Jewish Question is an example of an “infinite, absolute problem [which] cannot
be solved” (¶12).
The claim that there are permanent problems shifts our attention from the
particular perspective of a young Jew in Weimar toward a broader under-
standing of the limits of liberalism. The Jewish Question is one example off
neglecting the high to manage the low. By allowing superstitious men freedom
of thought, liberalism contributes willy-nilly to the decline of philosophy and
reason in political life. Strauss again breaks from the autobiographical chronol-
ogy with a judgment that shapes the narrative. As with his claim about judg-
ing the lower in light of the higher, he forces us to consider the claim more
carefully and reflect on the role of reason in political life. The Jewish problemm
is “the most manifest symbol” of the political dimension of the “human prob-
lem,” which cannot be fully resolved (¶12). He contrasts this with the problem
of the Jewish individual who has become alienated from his faith. This prob-
lem can be resolved by rejoining the Jewish community. However, the largerr
problem of the Jewish community’s relation to other, nonJewish communities
cannot be resolved. Liberalism tries to resolve it by finding a common under-
standing of justice or “a universal human society,” which is united by its vision
of justice (¶14).
Strauss presents the liberal solution as it appeared to his contemporaries,
namely
a e y that
t at liberalism
be a s was thet e best solution
so ut o to the
t e Jewish
Jew s problem
p ob e either
e t e because
40 STEVEN FRANK EL

Judaism itself had been refuted (¶14) or because liberalism simply followed from
Jewish principles of justice (¶28). Spinoza provides the basis for both claims since
his argument is built on the dual premises of criticizing the Biblical teaching
while returning to its essential principle, the practice of charity.14 The principle
of charity and toleration has the advantage of a broad consensus among various
religious faiths and even nonbelievers. For adherents of the Bible, Spinoza argues
that the truest and most accurate reading of the Bible reveals that its deepest
teaching, the one most necessary for salvation, is the practice of caritas. For all
good men—honestos as Spinoza calls them—the benefits of this religious view
are so apparent that they would embrace it just as enthusiastically whether or not
they are believers (TTP P 14.1.36).15 In addition, by subordinating all other reli-
gious beliefs to charity, liberalism allows a good deal of freedom of thought. Any
views, even atheism, are acceptable if they promote charity.16

The Question of Return


In his critique of Weimar and Zionism, Strauss had cast a doubt on whetherr
Spinoza’s strategy of accommodating all views can work effectively without
identifying a firmer theoretical basis for the morality of caritas. He concludes
that “the Jewish problem is insoluble,” by which he means that the “liberal state
cannot provide a solution to the Jewish problem” (¶13). Strauss qualifies this
claim by admitting that liberalism is better than the alternatives of commu-
nism and National Socialism, and that it does provide an “uneasy ‘solution to
the Jewish problem’” (¶13). Then, in striking contrast to his previous claims,
he declares in the next paragraph: “There is a Jewish problem that is humanly
soluble” (¶14). This problem, however, is not the wider political problem of the
community in a precarious situation, but rather of “the problem of the Western
Jewish individual.” The individual can return to traditional beliefs and avoid the
illusions of liberalism. In this sense, there is hope for solving our “deepest prob-
lem” and “most vital need” theoretically, even if this does not end discrimination
or superstition (¶14).
In moving from the political situation to the question of individual return,
Strauss has not entirely freed himself from the grips of the theologico-political
predicament. Although he has explained why a political solution cannot fully
solve the Jewish problem, he has not yet exposed the theoretical underpinnings
of liberalism. These underpinnings are first revealed when Strauss points out
that “some of his contemporaries” believed that return was impossible because
the traditional faith “had been overthrown once and for all.” But the theo-
retical legacy of liberalism is more complicated than simply rejecting revela-
tion. This is indicated by the fact that many more of Strauss’s contemporaries
believed that it was possible to grant science and history complete authority
“without abandoning one iota of the substance of the Jewish faith” (¶15). In
short, the theologico-political predicament is ambiguous: does it destroy faith
or preserve it? In the next section of the essay, Strauss examines the reasons
for this ambiguity. As we shall see, both positions are the result of a deliberate
a b gu ty about tthee status o
ambiguity of Sc
Scripture
ptu e wwithin
t Sp
Spinoza’s
o a s argument.
a gu e t. Ass St
Strauss
auss
S P I N OZ A’S C R I T IQ U E O F R E L IG IO N 41

frees himself of Spinoza’s theologico-political grip, he parts company with


more of his contemporaries.
The path Strauss follows takes him through the intellectual currents of his
time. I emphasize that he follows this path, which he describes as a “qualifiedd
return” or a return mediated by modernity, because it has already been marked
out by its founders including Martin Buber and Franz Rosenzweig as “new
thinking.” As the very title of the movement, “new thinking,” indicates, the
movement represented a conscious departure from the tradition so that the ques-
tion of return takes on a radical new meaning. The new thinking has several ele-
ments that interested Strauss and explains why he chose to pursue it rather than
traditional Judaism. For one thing, the movement assigns authority to “present
experience” rather than revealed law or tradition. Further, the new thinkingg
relies heavily on a broader critique of reason and philosophy found in the work
of Friedrich Nietzsche, a seminal thinker for Strauss (¶17). Unlike earlier efforts
to make Judaism compelling by showing that “the truth of traditional Judaism is
the religion of reason,” the new thinking rejected reason as a criterion altogetherr
(¶15). The movement followed Nietzsche in arguing that the “return to Judaism
requires today the overcoming of what one may call the perennial obstacle to the
Jewish faith: traditional philosophy” (¶16).
This critique of philosophy was based not only on its awareness of the lim-
its of reason but also on its thoroughgoing historicism. Following Nietzsche,
Rosenzweig argued that “the human soul has no unchangeable essence or limits
but is essential historical” (¶20). This claim constitutes the core of historicism’s
claim to wisdom. The historical dimension of all thought reveals previous claims
to wisdom as defective. It allows us to understand previous thinkers better than
they understood themselves. Historicism also provides a new approach to tradi-
tion: whereas the question of return was also seen as a return to the revealed law,
the new thinking claims that we have always selected from the tradition accord-
ing to the needs of our age; now we can and should do so consciously.
According to the autobiographical account of the narrative, Strauss’s turn to
the new thinking was motivated by his question of whether a “return to Jewish
orthodoxy was not both possible and necessary” (¶24). Yet Strauss does not
return to orthodoxy. Instead he launches an investigation of Spinoza’s critique
of religion to see if there are grounds philosophically that prevent a return. The
results of this investigation are less clear in terms of orthodoxy, but point instead
to the continuing relevance of ancient philosophy. Like Socrates, who claims in
the Apology that his lifelong philosophical quest began with his pious attempts to
understand and verify the oracle of Delphi, Strauss frames his discovery of phi-
losophy in terms of his pious effort to return to orthodoxy. Whereas Socrates hadd
investigated the belief in the gods of the city, Strauss investigates the superstitions
of his age, the belief in historicism under the guise of the “new thinking.”

The New Thinking


Strauss begins with Martin Buber’s foray into the new thinking in I and Thou.
Rat e than
Rather t a begin
beg with
w t the
t e tradition,
t ad t o , hee starts
sta ts with
w t “present
p ese t experience,”
e pe e ce, which
w c
42 STEVEN FRANK EL

is more immediate and less doubtful. If we are open to an encounter with the
divine, that is, if we have not dogmatically accepted the rationalism of Greek
philosophy—the path that recognizes only “what man knows by himself ”—we
may experience an absolute call that comes from outside of man and “goes
against man’s grain,” suddenly becomes more compelling (¶16). This experience
finds confirmation in the call of the prophets in the Bible. The prophets too
have experienced the speechless “call” of God and, in response, offer a human
interpretation. Prophecy, the speechless experience of the absolute, is in principle
open to everyone. Our return depends on our experience of the absolute and ourr
recognition of this same experience in the Bible.
Buber’s analysis places great weight on our interpretation of the absolute
divine experience. But which interpretation of this experience is the truest one?
Cannot the experience be tampered with? Is not the experience itself deter-
mined by our historical conditions? Buber cannot answer these questions by
using reason to judge because he assumes that reason necessarily undermines
such an experience. To show us the danger that Buber’s position has invited,
Strauss invents a dialogue between Buber and Heidegger regarding whether the
prophets sought to challenge or affirm the people’s security. Although Buberr
would many years later consider Heidegger’s critique, during the 1930s no such
dialogue was possible because “[a]t that time, Heidegger expressed his thought
about revelation by silence or deed rather than by speech” (¶17). To create a
dialogue between Heidegger and Buber, therefore, Strauss has to use later writ-
ings from each author that he could not have had access to during this period off
his development.17 The confrontation between Buber and Heidegger once again
confounds our efforts to trace Strauss’s historical development.
Strauss’s dialogue reveals that Heidegger has a “deeper understanding” off
the meaning and implications of the new thinking (¶17). Heidegger claims that
Buber’s interpretation of the absolute is merely wishful thinking. Indeed, against
Buber’s assertion that the warnings of the prophets do not “go against our grain,”
Heidegger shows that they merely confirm our wish for security, particularly
the security of our moral judgments by providing a “supra-human support forr
justice” (¶18). Rather than dismiss philosophy, he identifies a goal shared by both
ancient philosophy and biblical revelation to find or demonstrate “the security
of justice.” The desire for justice is in turn related to the desire for the eternal,
which does not come into being or pass away. But the new thinking exposes this
desire as “stemming from ‘spirit of revenge’” rather than from either revelation
or reason. Buber had dismissed reason with the confidence that revelation could
take its place as the basis for justice. But Heidegger’s atheism cuts off this possi-
bility and forces us to confront our terrifying condition of insecurity and uncer-
tainty. For him, terror and cruelty are the true signs of “intellectual probity.”18
There is no ultimate ground for its belief in a summum bonum; philosophy can no
longer pretend to be anything more than an act of will.
Strauss presents Heidegger’s critique of revelation in terms of a critique of the
high. In an unmistakable attack on Heidegger, he observes that “[n]ot every man
but every noble man is concerned with justice and righteousness and therefore
with any possible or extra-human support of justice, or with the security of
S P I N OZ A’S C R I T IQ U E O F R E L IG IO N 43

justice” (¶18). In Strauss’s presentation, Heidegger’s pursuit of justice takes the


form of a demand for “intellectual probity,” which in turn attempts to expose
every concern for justice with the desire for security.19 Intellectual probity should
not be confused with “the old love of truth.”20 The difference, Strauss, explains,
is that probity is more explicit, even dogmatic, about its atheism. As a result, it
rejects all attempts to obscure the consequences of atheism, or to pretend that
some sort of “semi-theism” is possible.21 In Heidegger’s hands, liberalism’s neu-
trality regarding the high gives way to the explicit rejection of the transcendent
and the apparent embrace of the low. Conversely, Strauss shows the common
ground shared by reason and revelation, and in fact by all noble men, namely the
concern for justice. Strauss’s return aims at restoring this common ground ratherr
than showing the victory of one alternative. Of course, this personal attack does
not mitigate the force of Heidegger’s objections. Instead, it exposes a deeperr
problem: modern philosophy, which began with a criticism of superstition in
favor of reason, now abandons reason as a guide. The result is that philosophy
participates in the superstitions that it had previously criticized and the cruelty
that it had once abhorred.22
Strauss’s path of return moves from Heidegger toward the origins of moder-
nity and Spinoza. 23 First, though, he considers a more serious attempt at a
qualified return to the Bible offered by Franz Rosenzweig. Paradoxically,
Rosenzweig begins his quest to return by denying the possibility of a return
to Biblical faith. He draws a distinction between what the authors of the Bible
meant and how we understand it today. The former is a historical concern
characteristic of what he calls the “old thinking,” while the latter is practiced
in full awareness of the historical text and our current situation, that is, how
the text affects the present situation. It is important to note that Rosenzweigg
rejects Spinoza’s principle of selecting from the text by distinguishing between
its essential and unessential parts. Such a distinction can hardly be as objective
as it claims, since it presumes the wisdom that it seeks to find in the Bible.
Rosenzweig avoids this problem by admitting that a “force” rather than a
principle guides our decision. The past, the “whole reality of Jewish life,”
offers materials for us to select and by virtue of the selection, transform into a
living force in our lives. The past becomes a set of resources for building the
future, but we are the ones who select from the past.
But as we scrutinize Rosenzweig’s selections, we cannot help but notice
that vitality is measured by consistency with modern liberal regimes. Thus,
for example, Rosenzweig does not find compelling the orthodox view of the
Torah’s law, which sees it in terms of prohibition and rejection; instead he inter-
prets the law in terms of liberation and transformation. This represents a farr
more optimistic view of political possibilities and of the fate of liberal societies.
Another example of vitality is Rosenzweig’s rejection of Biblical miracles. Here
vitality is associated with skepticism and science. Rosenzweig claims he did not
believe all the miracles reported in the Torah but, rather, claims to be open to
belief.24 The orthodox approach, in contrast, had far less confidence in the judg-
ment of each individual not only because of the limits of reason but also because
suc individuality
such d v dua ty could
cou d undermine
u de e the
t e cohesiveness
co es ve ess ofo the
t e community.
co u ty.
44 STEVEN FRANK EL

Like Buber, Rosenzweig wishes to make Judaism more consistent with expe-
rience, but in practice this means making it more consistent with liberalism.
Instead of beginning from the Torah as law, as does Maimonides, for example,
Rosenzweig begins from one’s awareness of being a member of the chosen peo-
ple, because such awareness is “the primary condition of the possibility of Jewish
consciousness” (¶21). This is at odds with the traditional view that the law as
revealed is the basis of the community. Strauss suggests that Rosenzweig con-
sciously begins with chosenness rather than law in order to reshape the tradition
so that it is more compatible with Christianity. This is in keeping with Spinoza’s
liberalism, where Judaism and Christianity are assimilated into a universal reli-
gion that stresses caritas as the sole requirement for salvation. Rosenzweig did not
hope to return to traditional Judaism, but to adapt the tradition so that it would
be at home in Weimar. Strauss exposes Rosenzweig’s effort to reshape the tradi-
tion, yet he does not criticize this reform simply as a betrayal of the tradition.
Rather, he wonders how to take one’s bearing in such a project: how can one
know that one’s proposals to change the tradition will result in its preservation orr
deepening? Has Rosenzweig understood “the old in its depth”? His attempt to
transform Judaism in order to make it compatible with liberalism, a regime that
is ambiguous with respect to a summum bonum, suggest otherwise.25 By trans-
forming Judaism to meet the needs of liberalism, Rosenzweig has lowered the
tradition rather than deepened it.
Strauss also suggests that Rosenzweig’s proffered reform has not deepened
the tradition, by contrasting him with Maimonides, who similarly attempted
to alter the tradition’s attitude toward philosophy. Curiously, Strauss approaches
the superiority of Maimonides in terms of his method for reshaping the tradi-
tion. Maimonides was more loyal to the Jewish People, or at least more careful
to preserve the appearance of his loyalty to the tradition, than was Rosenzweig.
Nothing exemplifies this better than Maimonides insistence on placing Judaism
above philosophy and politics. Strauss says that Maimonides, in sharp contrast
to Rosenzweig, wrote Jewish books, not philosophical ones, “as a Jew he gives
his assent where as a philosopher he would suspend his assent” (cf. Guide off
the Perplexedd 2.16; HBS xiv). In the preface, Strauss makes the same point in
order to contrast Maimonides with Rosenzweig, He points out that [w]hereas
the classic work of what is called Jewish medieval philosophy, the Guide of thee
Perplexed, is primarily not a philosophic book but a Jewish book, Rosenzweig’s
Star of Redemption is primarily not a Jewish book but a “system of philosophy”
(¶21). The difference between a Jewish book and a philosophic book is that
in a Jewish book, the authority of law precedes and is distinguished from the
authority of reason.26 Rosenzweig, however, has already rejected both ancient
philosophy and reason as a guide. To make matters worse from Strauss’s point off
view, Rosenzweig’s philosophy is much more consistent with Christianity than
is traditional Judaism because its starting point is not the law, but a secondary
category like chosenness. In this sense, he treats the tradition as a quarry, or set
of resources, for crafting a Judeo-Christian society. This view is important to
keep in mind because Strauss will later call attention to Rosenzweig and Cohen’s
c t que o
critique of Sp
Spinoza,
o a, wwhich
c esse
essentially
t a y cclaims
a s tthat
at hee was ddisloyal
s oya to hiss peop
peoplee
S P I N OZ A’S C R I T IQ U E O F R E L IG IO N 45

by exposing its flaws for all to see. In fact, Strauss here shows that Rosenzweig
has essentially done the same thing, with even less consciousness of his debt to
Spinoza.

The Return to Spinoza


The second half of Strauss’s preface (¶¶ 24–42) is devoted to Hermann Cohen’s
analysis of Spinoza. This is surprising, insofar as Cohen predated Rosenzweigg
and Buber. Moreover, his analysis stops short of the new thinking in that he
holds reason in high esteem, capable of guiding men and teaching them a uni-
versal moral law consistent with Jewish law. Possibly Strauss was genuinely more
impressed by Cohen than by Rosenzweig. He describes Cohen as “a Jew of rare
dedication” who “symbolized more than anyone else the union of Jewish faith
and German culture.” Nor did Cohen neglect the high. In fact, he “surpassed in
spiritual power” his colleagues in philosophy and theology. However, we must
be careful not to overstate the influence of Cohen’s critique; after all, Strauss
largely rejects Cohen’s analysis of Spinoza. His tribute to Cohen is actually a
refutation of Cohen’s approach to philosophy. Indeed, he raises the question
of whether Cohen could even be considered a philosopher, suggesting instead
that he is a rather limited, if well-meaning, scholar.27 In short, his presentation
of Cohen is from the point of view of the mature Strauss who has thoroughly
rejected him.28
Yet Rosenzweig’s judgment on Cohen plays an important role in Strauss’s
presentation. Rosenzweig admits that Cohen’s treatment of the evidence may
not be just; however, Rosenzweig argues that there is a deeper justification forr
Cohen’s position. Spinoza took “for granted the philosophic detachment or
freedom from the tradition of his own people” (¶35). Philosophy may appear to
confer freedom by detaching one from the community, but this is an illusion
on par with the erroneous view that reason provides a liberating vantage point.
Rejecting such illusions, Rosenzweig claims that this detachment is nothingg
more than the rejection of loyalty, love, and sympathy for one’s own community.
Had Spinoza cared more for his people, he would have helped them reinterpret
their tradition “in light of the highest [possibility, or failing that,] if neces-
sary, better than they understood themselves” (¶35). Thus, although the explicit
topic of this section of the essay is how to read Spinoza (or how Cohen misreadd
Spinoza), the judgment of Rosenzweig opens the radical question about chang-
ing a tradition. Cohen, who was temperamentally conservative, was unable
to grasp, “the fact that the continuous and changing tradition . . . [depends on]
revolutions and sacrileges” (¶38).
As for Cohen’s critical reading of Spinoza, Strauss does show how Cohen
helps expose several central and apparently deliberate contradictions in the
TTP P. These contradictions would reappear in Strauss’s mature interpretation off
Spinoza; however, Cohen was unable to resolve them correctly because of his
conviction that he has understood Spinoza better than Spinoza understood him-
self. This “idealizing interpretation” is a variety of historicism, which explains
a author’s
an aut o s work
wo as a product
p oduct of
o history
sto y rather
at e than
t a deliberate
de be ate choice.
c o ce. Cohen,
Co e ,
46 STEVEN FRANK EL

for example, did not consider that Spinoza’s style reflected a deliberate choice
in light of persecution and his political project. Strauss playfully and explicitly
refutes this, and suggests instead that Cohen “understood Spinoza too liter-
ally because he didn’t understand him literally enough” (¶37). The difficulty
in reading Spinoza correctly is determining whether he means what he says,
that is, whether we should read him literally. The question of historical cir-
cumstances undoubtedly plays a role in Spinoza’s presentation, but in order
to measure this influence, we must understand Spinoza’s metaphysical project.
If Spinoza’s assertions contradict his metaphysical teaching, then we are justi-
fied in considering his political circumstances to explain his argument.29 Thus,
Strauss’s section is meant to guide the reader away from his historicist prejudice
to a more thoughtful reading of Spinoza. The claim of the overall narrative is
that one can escape Spinoza’s theologico-political grip only by reading him
correctly.

The Development of Strauss’s Interpretation of Spinoza


In addressing Cohen’s historicist interpretation of Spinoza, Strauss begins by put-
ting aside the autobiographical narrative to offer an account of Spinoza’s meta-
physics (¶27). To see more clearly the defects in Cohen’s hermeneutics, Strauss
begins with a literal reading of Spinoza’s political thought. According to this
account, Spinoza must be understood as part of modern philosophy’s break with
ancient philosophy. Spinoza did not initiate this break, but was “the heir of the
modern revolt” (¶29). Modern philosophy rejected the theoretical contempla-
tion of nature in favor of a more practical and useful approach. It sought to
master rather than contemplate nature. This was part of a political project to
improve man’s comfort and thereby secure the freedom to philosophize. Spinoza
embraces the modern project, but also attempts to restore the “dignity of specu-
lation on the basis of modern philosophy or science” (¶29).
Spinoza’s metaphysical starting point is also distinctively modern: all things
proceed from the one, but the process is not one of decay or descent, neither is
it one of creation. Rather it is an unfolding of individual, particular things that
we are to understand sub specie aeternatatis.30 The “last word on the subject” is the
Ethics, where Spinoza argues that the highest form of knowledge is of particularr
things.31 In sharp contrast to the Bible, God is not therefore the highest source
of knowledge, but the process or unfolding of nature. Thus, on the question off
the high, Spinoza’s thought is ambiguous. On the one hand, he maintains the
view of ancient philosophy that philosophy is the highest activity for man; on
the other hand, his view is not grounded in nature but rather in the endless (and
directionless) unfolding of nature.
Spinoza’s political project, liberalism, reflects his ambiguity about the status off
the high. He begins, or appears to begin, from metaphysical axioms that provide
no basis or support for justice.32 In fact, the goal of politics must be understood
in light of the mechanical universe within which it operates. Rather than referr
to intellectual or moral virtue, the regime must be directed toward security and
co o t. No
comfort. Nor ca
can po
political
t ca leaders
eade s rely
eyo on so
somee abso
absolute
ute sta
standard
da d o
of just
justice
ce ((from
o
S P I N OZ A’S C R I T IQ U E O F R E L IG IO N 47

nature or God) to fix their goal. They must guide man but rather by the tough-
minded control of destructive passions and the fostering of more constructive
ones through, for example, a commercial society. Spinoza equates the passions
with the right of nature, so as to indicate that they are natural, even if they are
not directed toward any end other than the striving to persist.

Cohen’s Critique of Spinoza


The celebration of Spinoza by German Jews reflects their appreciation of lib-
eralism as a regime that offered them unprecedented opportunities. Spinoza
was the saint who inaugurated “a new religion or religiousness which was to
become a wholly new kind of society, a new kind of Church” (¶27). That theyy
neglected the Machiavellian elements of his political thought and ignored his
rejection of the Biblical God is not surprising given their apparent flourishing in
liberal democracy. To his credit, Cohen refused to ignore the less attractive ele-
ments of Spinoza’s teaching (¶29). He was particularly disturbed by the effort to
annul the original excommunication, arguing instead that Spinoza had indeedd
betrayed the Jewish people. Cohen was not concerned, according to Strauss,
with violations of ceremonial law or the denial of Mosaic authorship. Instead,
he “condemned Spinoza because of his infidelity in the simple human sense, off
his complete lack of loyalty to his own people, of his acting like an enemy of the
Jews and thus giving aid and comfort to the many enemies of the Jews, of his
behaving like a base traitor” (¶19).
Cohen’s particular charges constitute a damning indictment: Spinoza accepts
and extends the Christian critique of Judaism. He falsely accuses Judaism off
commanding hatred of the enemy. He presents a picture of Judaism as carnal and
particularistic in contrast to the universal and spiritual picture of Christianity.
His efforts to show Judaism in a poor light lead him to contradict himself by
then claiming that prophecy is universal in order to promote the universalism
of Christianity. In Spinoza’s version of Judaism, the law is particularly devalued
as having a purely political, rather than moral or spiritual, role. Having reduced
the law to a tribal code, Spinoza suggests that the law’s author, the God of Israel,
is merely a tribal God. Not only then does Spinoza betray the Jewish people,
he also blasphemes God while suggesting that the Torah is “of merely human
origin” (¶18).
In Cohen’s reading, Spinoza is so filled with malice—“he has no heart for
the people, no compassion”—that he grows obsessed with denigrating Judaism
by any means necessary (¶30). He does not hesitate, for instance, to praise
Christianity or encourage the anti-Jewish views of his Christian readers. He
ignores contradictions in his own argument if they undermine Judaism. His
betrayal of his people and philosophy is “humanly incomprehensible” and even
demonic in that it is rooted in a denial of any standard of morality. Cohen
also argues that there was a more comprehensible motive for Spinoza’s hostility
toward Judaism. Like other Marranos, whose fear of the Inquisition turned into
hatred of its causes, namely Judaism, Spinoza aimed his hatred at the victims
rather
at e than
t a tthee perpetrators.
pe pet ato s.
48 STEVEN FRANK EL

Strauss agrees with Cohen’s overall judgment that Spinoza is “amazingly


unscrupulous”; indeed, he argues that his own judgment is “in some ways even
stronger than Cohen’s” (¶31). But Cohen reads the TTP P too literally and therefore
mistakenly thinks that Spinoza genuinely prefers Christianity to Judaism because
it appears more spiritual and universal. In his view, Spinoza wishes to ennoble
Christianity by convincing Christians to purge themselves of carnal relics off
Judaism, that is, by liberating Christianity from Judaism. This is where “Cohen
fails” to follow Spinoza’s argument. 33 Spinoza’s goal according to Strauss was to
fashion a liberal society that embraced the freedom to philosophy. The argument
that he crafted was not a betrayal of the Jewish people, nor does his primary alle-
giance to philosophy mean that he is indifferent to the Jewish people.34 To the
contrary, Spinoza provides a society wherein they could be citizens. In addition,
the universal dogmas of catholic faith in TTP P Chapter 14 “are equally acceptable
to both Jews and Christians” (¶31). Such a society requires the abrogation of the
Mosaic law, but this law had become obsolete with the destruction of the state.
Strauss’s defense of Spinoza amounts to this: he “liberated Jews in the only way
he could think of ” given his philosophy (¶31).
Strauss’s critique of Spinoza is stronger than Cohen’s because Strauss does not
ignore the Machiavellian core of Spinoza’s project. Cohen was committed to a
theologico-political view that combined the Bible with Kant, who viewed God
as an “idea” that guarantees the existence of a universal moral law discoverable
by reason. The “idea” of God is the basis of morality, as well as of the optimistic
belief in the ultimate triumph of Good over Evil. 35 Kant’s moral law is realiz-
able first only in particular states, but it points toward a universal moral order,
grounded in international law and enforced by a league of nations. In Cohen’s
version of creation and providence, the politics generates the theology. In otherr
words, we need the “idea” of God to justify a certain political order. Reason
comes to theology’s aid to help determine the requirements of a universal and
just political order, and instructs theology on its presuppositions.
But all of these beliefs are exposed as superstitions, albeit salutary supersti-
tions, within the confines of Spinoza’s metaphysics. Cohen’s thought and his
interpretation of Spinoza have a clear sense of the high, and he seeks whereverr
possible to read the low in light of the high. According to Strauss, this leads to
various problems within Cohen’s own thought. For example, he tends to coverr
over the criminal violence and destruction that attend revolutions. Similarly,
he sees the purpose of punishment solely in terms of promoting the well-beingg
of the criminal. In his single-minded attention to the high, Cohen appears
to ignore the low altogether. He “ignored the harsh political verities which
Spinoza had stated so forcefully” (¶36). In this, one wonders whether he did
not help to prepare the generation of Jews who did not have “eyes to see” the
disaster that was approaching. In any case, this tendency to neglect the low ren-
ders Cohen a woefully inadequate interpreter of Spinoza because “the kind off
interpretation which Spinoza calls for is not idealizing, since his own doctrine
is not idealistic” (¶36).
Strauss focuses our attention on two related problems in Cohen’s interpre-
tat o tthat
tation at eexpose
pose tthee pproblem
ob e tthat
at eemerges
e ges w
with
t hiss idealizing
dea g pphilosophy.
osop y.
S P I N OZ A’S C R I T IQ U E O F R E L IG IO N 49

First, Cohen is unable to conceive of the relation between reason and revela-
tion as anything other than harmonious. In addition, Cohen cannot grasp
the meaning of Spinoza’s denial of any natural support for the high. Both
examples point us directly to Maimonides and imply that not only did Cohen
misinterpret Spinoza and Maimonides incorrectly but also that Maimonides’s
point of view was superior.
Cohen claims that Spinoza’s interpretation of Judaism does not recognize
the universalism of the prophets. For instance, Spinoza draws our attention to
Maimonides’s claim that a person cannot be saved unless he believes in Mosaic
revelation. This view suggests that obedience rather than reason is primary to
piety. Similarly, Maimonides claims that a non-Jew can be considered pious if he
performs the Noahide commandments as commandments of God, adding that
if a non-Jew performs them because they are rational, he is neither pious norr
wise. Its wisdom teaches that reason cannot supplant piety as a source of political
authority. Finally, Maimonides suggests that the Torah’s commandments them-
selves are not rational, but are meant to address and cure idolatry, “an irrational
practice.” Maimonides begins from the opinions of the community before deter-
mining the best way to enlighten the tradition. Irrational opinions may justify
the commandments while reason itself may not.36 Leaving aside Spinoza, Strauss
reports that the Jewish tradition, as per Joseph Caro, confirms this view. Cohen
ignores those elements of the Jewish tradition, which suggests a tension between
wisdom and piety because he overestimates the power of reason and its compat-
ibility with the law.
The other example that Strauss chooses focuses on the question of the high-
est good. Cohen points out the following contradiction in the TTP: Spinoza
claims that Moses’s law is a divine law, that is, it points man toward the highest
good or intellectual love of God; yet, he denies that the Mosaic law aims at the
highest good. Strauss denies that this is a contradiction, because Spinoza does
not mean the law is divine in the sense of aiming toward salvation or even intel-
lectual knowledge of God. Rather, Spinoza means that the law is divine because
it is believed to be divinee by Christian readers whom Spinoza addresses. There is
no divine law in the sense of a law with natural or rational ends. Cohen has not
come to grips with the fact that neither God nor reason supports the moral law.
Spinoza’s morality is always tied to a relentless egoism. Following his teacherr
Machiavelli, Spinoza conceives of nature in terms of individuals struggling to
persist in their own being. Nothing stands apart from or above this struggle that
would allow us to judge it in moral terms.
Cohen wants to understand the tradition as something that we can shape by
understanding the low or superstitious in light of reason, or the high. Despite
Strauss’s insistence on understanding the low in light of the high, he rejects
Cohen’s reading of the Jewish tradition. The reason for this is that Cohen has
not escaped the horizons set out by Spinoza. He “does not come to grips with
the fact that Spinoza’s critique is directed against the whole body” of Jewish
teachings and tradition (¶39). He believed that Spinoza had refuted orthodoxy,
especially miracles and law. Furthermore, Cohen agrees with Spinoza’s opposi-
tion to rabbinical Judaism, which put great emphasis on ceremonial law. Modern
50 STEVEN FRANK EL

Judaism has liberated itself from the rabbinical view of law. Cohen’s understand-
ing of Judaism, by virtue of which he attacks Spinoza, is grounded in the “his-
torical understanding of the Bible,” which in fact was originated by Spinoza.
Cohen recognizes that his understanding of Judaism, for example, his distinction
between mythical and historical elements of the Bible, is not consistent with “our
traditional exegesis.” In short, Cohen offers a more positive view of Judaism as
rational, but ultimately his understanding of Torah is consistent with, and rests
on Spinoza. Cohen’s attempt to understand Spinoza better than he understood
himself failed; in fact, Spinoza understood Cohen better than Cohen understood
himself.

Strauss’s Judgment on Spinoza


Strauss concludes the Preface by summarizing his argument in SCR R on the con-
sequences of Spinoza’s critique of religion. Spinoza claims to refute orthodoxy,
and its particular claims about the divine inspiration of every word of the Bible,
that Moses was the author of the Torah, the account of miracles as violations
of natural law, etc. The TTP P did succeed in refuting orthodoxy, but only iff
orthodoxy makes claims that it cannot defend. But if Orthodoxy responds by
making less ambitious claims, for example, by admitting that such claims “can-
not claim to possess the binding power peculiar to the known,” than Spinoza has
not refuted orthodoxy (¶40). Biblical religion ultimately rests on the claim that
God’s will is unfathomable. This is the premise by which all orthodox claims
are possible.
Spinoza has succeeded only in showing that there are contradictions in the
Biblical narrative but this only undermines an overly ambitious variety of ortho-
doxy, which claims paradoxically to know the ways of God. Traditional Judaism,
however, makes no such claims and thus “cannot be refuted by experience or by
recourse to the principle of contradiction” (¶40). Spinoza was forced to over-
reach by claiming to know more than he could about the whole of things, and
he had resort to mockery and ridicule in the place of argument to unseat religion.
This reveals his critique as fundamentally resting on an act of will or a decision
rather than reason. By exposing the defects of Spinoza’s critique, Strauss does
not mean to suggest that orthodoxy has triumphed over philosophy. Rather,
he wishes to show the defects inherent in the Enlightenment’s political project,
which leaves men freedom without guidance. The “self-destruction of reason” in
modernity invites another kind of return, a return to a preliberal thought. Such
a return requires an investigation into method by which “heterodox thinkers off
earlier ages wrote their books” (¶42).

Conclusion: Progress or Return?


As we have seen, Strauss presents an autobiographical account meant to
explain the development of his understanding of Spinoza in the Preface to
the translation of his first book on Spinoza. But a careful examination of the
Preface reveals the fact that Strauss is neither explaining the development of
S P I N OZ A’S C R I T IQ U E O F R E L IG IO N 51

his thought, nor simply giving an account of his life as a young man. This
becomes clear when we pay attention, for example, to the fact that Strauss’s
mature judgments guide his presentation of the autobiographical materials.
His assertion that it is safer to understand the low in light of the high is par-
ticularly important in this regard. But what does Strauss mean by the “high”?
Strauss forces the reader to puzzle out this critical question, which is the key
to understanding the essay.
Instead of identifying the meaning of the “high,” he provides an account of his
efforts to escape “the grips of the theologico-political predicament” bequeathed
to modernity by Spinoza. This escape or liberation initially takes the form off
return, particularly a return to revelation. But this attempt to return proves
exceptionally difficult because Spinoza’s critique of revelation profoundly shapes
the way subsequent thinkers understand the tradition. Significantly, Spinoza is
ambiguous about the source and meaning of “the high”; specifically, he is unclearr
whether the divine law is in some sense true or whether it is merely the product
of superstition. This ambiguity allows for a variety of readings of Spinoza, which
were pursued by Strauss’s community in Germany. The more optimistic inter-
pretation affirmed the divine foundation of morality, while others took Spinoza’s
atheism as corroborating the new virtue of “intellectual probity.” Neither view,
however, managed fully to escape Spinoza’s grip and appreciate the possibility
of return. Strauss suggests that the liberation from Spinoza is possible if we read
Spinoza carefully, that is, if we understand his political project, including his
ambiguous treatment of the high, in light of his metaphysics that denies the pos-
sibility of the high. In other words, the political project must be understood as
an attempt to obscure Spinoza’s view of nature, which is entirely indifferent to
human flourishing.
To liberate oneself from the grip of Spinoza means to understand his project
and also to recognize an alternative to his metaphysical and political views.
Strauss suggests that this is possible because Spinoza has not succeeded in refut-
ing revelation. “The genuine refutation of orthodoxy would require the prooff
that the world and human life are perfectly intelligible without the assumption
of a mysterious God. . . . [M]an has to show himself theoretically and practically
as the master of the world and the master of life” (¶40). Spinoza has not done
this; instead, in his attempt to do so he has overstated the capacity of reason
to provide a full account of the whole and thereby rendered it questionable
as a rational activity. To the clear-sighted descendants of Spinoza, philosophy
appears to be the result not of rational inquiry, but an act of will. This permits
the provisional triumph of orthodoxy, but without reason. Invariably, ortho-
doxy descends into fanaticism. Strauss envisions a “return” to a piety that is
checked by reason, and a philosophy that is mindful of its limitations. Thus
the restoration of the “high” requires the simultaneous recovery of reason andd
revelation as potential candidates for the best life. Nor is the possibility of such
a recovery a matter for despair. Since we have the important works of medieval
and ancient philosophy, we need only to excavate how such issues appeared to
thinkers before Spinoza. In teaching us to read Spinoza carefully, Strauss also
teac es us to read
teaches ead hiss ppredecessors
edecesso s ca
carefully.
e u y.
52 STEVEN FRANK EL

Since Strauss himself was not an observant Jew, it is easy to conclude that
philosophy has, in his eyes, already ultimately overcome revelation in its claims
to the high.37 In fact, Strauss appears to say as much: “Philosophy or science,
the highest activity of man, is the attempt to replace opinion about ‘all things’
by knowledge of ‘all things’; but opinion is the element of society; philosophy
or science is the attempt to dissolve the element in which society breathes, and
thus it endangers society.” The result is that philosophy “must respect the opin-
ions on which society rests” without “accepting them as true.”38 Thus, even iff
we attribute to Strauss the view that philosophy is the highest or best activity
for man, we cannot avoid the political problem of accommodating this view
to our community, which rests on a settled standard of justice. This tension
between conventional opinion and philosophy resembles the Platonic cave. Yet
this tension has been entirely obscured by modern philosophy, as exemplified in
Spinoza’s metaphysics, which claims that nature offers no support to either reason
or revelation. In this sense, believers and philosophers are in the same predica-
ment: in order to restore a sense of the high, they must first liberate themselves
from the cave beneath the cave.39

Notes
* Strauss published two editions of the autobiographical preface: the first as a preface to
the translation of SCR R and the second as part of a collection of essays in LAM. The two
versions are nearly identical except that in the latter version Strauss divides several of the
longer paragraphs into shorter ones so that where there are 42 paragraphs in the original
essay, while there are 54 paragraphs in the later edition. Another important difference is
that in the later edition, Strauss refers to himself as being in the “grips” of a theologico-
political problem. This change, from “grip” is discussed below (see note 22). This essayy
uses the paragraph numbering from the original essay in SCR. My thanks to Professors
Terence Marshall, Thomas Meyer, Richard Polt, Timothy Sean Quinn, John Ray, and
Martin D. Yaffe for their thoughtful comments (and objections) to this essay.
1. Strauss’s letter to Krüger is dated January 7, 1930 (GS–3 380–81). He blames his
“boss,” Julius Guttmann, for preventing him from writing more explicitly. Forr
an overview of the correspondence between Krüger and Strauss, see Thomas L.
Pangle, “Light Shed on the Crucial Development of Strauss’s Thought by his
Correspondence with Gerhard Krüger,” chapter 3 of the present volume.
2. One of Strauss’s central hermeneutical principles is that careful authors have
developed their own manner of writing by studying other careful writers. In
general, therefore, “we learn to write by reading. A man learns to write well by
reading well good books, by reading well most carefully books which are care-
fully written” (PAW W 144). Strauss’s commentaries on philosophical texts offerr
guidelines for reading carefully. One method that he frequently employs, forr
example, is counting the number of paragraphs to determine the central para-
graph (cf. PAW W 24–25).
3. A good example of this is David Janssens’s fine study, Between Athens and Jerusalem:
Philosophy, Prophecy, and Politics in Leo Strauss’s Early Thoughtt (Albany: SUNY
Press 2008), 8–26. Janssens begins by attempting to follow Strauss’s preface as
an account of his earlyy thought,
g but is quickly
q y forced to cite other materials and
S P I N OZ A’S C R I T IQ U E O F R E L IG IO N 53

sources to fill in the picture even to explain the meaning of Strauss’s description
of the “theological-political predicament.”
4. The original essay indicated that Strauss was in the “grip” of the theologico-
political predicament. Strauss decided to make “grip” plural in the subsequent
edition. His choice may ref lect the fact that he believed the theologico-politi-
cal predicament had multiple sources and expressions. I wish to thank Thomas
Meyer for directing me to this difference and its possible meaning.
5. For a more extended analysis of the theologico-political predicament, see my
“Politics and Rhetoric: The Intended Audience of Spinoza’s Tractatus Theologico-
Politicus,” Review of Metaphysics 52 (1998–99): 897–924.
6. See my “Review Essay of Martin D. Yaffe’s translation and commentary off
Spinoza’s Tractatus,” Interpretation 32, no. 2 (Spring 2005): 171–78.
7. Spinoza also hoped to mitigate the damage done by theology, by creating a uni-
versal faith whose only requirement was the practice of charity. In addition, he
prescribed political institutions based on power rather than theology, which were
more consistent with the natural asocial nature of man. Finally, by liberating phi-
losophy from politics and theology, he enlisted the help of science in promotingg
the well-being of the citizens.
8. Spinoza “was the philosopher who founded liberal democracy, a specifically
modern regime” (¶27).
9. Allan Arkush, “Leo Strauss and Jewish Modernity,” in Leo Strauss and Judaism:
Jerusalem and Athens Revisited, ed. David Novak (Lanham, MD: Rowman &
Littlefield, 1996), 115.
10. See Steven Smith, “How to Commemorate the 350th Anniversary of Spinoza’s
Expulsion,” Hebraic Political Studies 3, no. 2 (Spring 2008): 173.
11. Emil Fackenheim also identifies this passage as critical to understanding Strauss,
but Fackenheim thinks that Strauss has not fully confronted the low as revealed in
the Holocaust. See his “Leo Strauss and Modern Judaism,” in Emil Fackenheim,
Jewish Philosophers and Jewish Philosophy, ed. Michael Morgan (Bloomington:
University of Indiana Press, 1996), 103, 104.
12. Contrast this position with Hannah Arendt’s critique of Zionism. After the war,
Strauss continued to defend Israel on the grounds that it restored national pride
(see his letter to National Review w in JPCMM 413–14.). The preface was written in
August 1962, less than 3 months after Eichmann was hung in Jerusalem. Strauss’s
endorsement of Zionism on the grounds of restoring Jewish pride stands in sharp
contrast to Arendt’s Eichmann in Jerusalem, written during the same period.
13. See TTP, P chapter IV.
14. Strauss explains in his 1959 Spinoza seminar: Spinoza insists that “[w]e must go
back to the Scriptures. And this Scripture is the document of revealed truth, and
forget everything which the theologians claim to know. And in that moment—up
to this point he says, by implication, there is only one authority: the word off
God as delivered in the Bible. Then he turns it around. We must even examine
the authority of the Bible. . . . That the Bible itself is true and divine cannot be
assumed. It must be found out. Perhaps it is not true and divine, or is only partly
true and divine. Therefore, to repeat, the argument of this work is ambiguous
throughout. There is a Biblicist argument, based on the accepted authority off
the Bible, but only of the Bible, as the word of God, and there is another argu-
ment which questions this very premise. The real teaching of Spinoza is not the
Biblicist argument, but the other one” (29).
54 STEVEN FRANK EL

15. After quoting this passage, Strauss observes that the dogmas of universal piety
“must be of such a kind that all moral men, all decent men, regardless of whetherr
they are Confucians or Christians, or Greek pagans for example, would neverr
disagree. But this must somehow be in accordance with the Bible. That is the
difficulty. The dogmas must be acceptable to all decent men. Now I hope the
philosophers are decent men. Must they not also be dogmas which are accept-
able to the philosophers? And philosophers according to Spinoza cannot believe
in a legislating God, in a God who exercises providence. That’s the prob-
lem. . . . after having said that the Bible demands nothing but charity, meaning
love of neighbor—therefore, opinions are free. But Spinoza knows that this prin-
ciple . . . necessarily has theoretical premises. . . . If that is so, the denial of these
theoretical principles must be prevented. And if the philosophers by definition
deny these principles, the philosophers must be persecuted” (Lecture 9, p. 172).
The solution, Strauss suggests, is that philosophers must pay lip-service to the
essential dogmas, even if they do not believe in them (cf. 176–77).
16. See PAW W 193–96.
17. Richard Velkley shows how Strauss’s thought acquired a “renewed and deeperr
engagement with Heidegger in the 1950s as Strauss acquired the publications off
[Heidegger’s] later thought.” See Heidegger, Strauss, and the Premises of Philosophy
(Chicago: University of Chicago Press, 2011), 69.
18. See Nietzsche, Human, All Too Human, sec. 120 and 227 (trans. Marion Faber and
Stephen Lehmann [Lincoln: University of Nebraska Press, 1984], 87, 140–41).
19. “Heidegger’s political engagements show that the furthering of [the renewal off
the question of being] takes precedence over any considerations of the good, the
moral, and the just, as these have been understood in the philosophic tradition as
having some universal articulation, ref lecting ends (happiness, perfection, virtue)
inherent in human nature or reason” in Velkley, Heidegger, r 93.
20. See PLA 137–38 (endnote 13).
21. Ibid.
22. Strauss makes this point quite sharply in WIPP P 26–27: “The crucial issue con-
cerns the status of those permanent characteristics of humanity, such as the
distinction between the noble and the base.” He goes on to associate the neglect
of these characteristics with Heidegger: “It was the contempt for these permanen-
cies which permitted the most radical historicist in 1933 to submit to, or rather to
welcome, as a dispensation of fate, the verdict of the least wise and least moder-
ate part of his nation which it was in its least wise and least moderate mood, and
at the same time to speak of wisdom and moderation. The biggest even of 1933
would rather seem to have proved, if such a proof was necessary, that man cannot
abandon the question of the good society, and that he cannot free himself from
the responsibility of answering it by deferring to history or to any other powerr
different from his own reason.”
23. Strauss will argue that “the ultimate justification of Spinoza’s critique is the
atheism from intellectual probity, which overcomes orthodoxy radically by under-
standing it radically, i.e., without the polemical bitterness of the Enlightenment
and the equivocal reverence of romanticism” (¶39).
24. Buber and Rosenzweig did not succeed in preserving the substance of the
Jewish tradition. Their strategy, which Strauss describes in Philosophy and Law w
as “internalization,” manages to reinterpret the claims of the Jewish tradition
(e.g., creation, miracles, revelation, and prophecy) so that they would no longerr
conf lict with “intellectual probity.” For a lucid outline of Philosophy and Law, see
Eve Adler, “Translator’s Introduction,” PLA 1–20.
S P I N OZ A’S C R I T IQ U E O F R E L IG IO N 55

25. Hobbes is less ambiguous and therefore more revealing than Spinoza on the ques-
tion of the greatest good. See Leviathan, chapter 11.
26. As Hillel Fradkin points out: “there are according to Maimonides different sci-
ences of the law, or more precisely two: the legalistic science of the law or juris-
prudence and the true science of the law. The Guidee is devoted to the latter.” See
Fradkin, “A Word Fitly Spoken: The Interpretation of Maimonides in the legacy
of Leo Strauss,” in Leo Strauss and Judaism: Jerusalem and Athens Revisited, ed. David
Novak (Lanham, MD: Rowman & Littlefield, 1996), 59.
27. “Cohen commits the typical mistake of the conservative, which consists in
concealing the fact that the continuous and changing tradition . . . [depends on]
revolutions and sacrileges” (¶38). By “conservative,” Strauss does not mean
politically conservative. (Cohen’s left-wing political views, including his endorse-
ment of socialism and his hostility to capital punishment, are well known.)
Rather, Strauss means by “conservative” a strong orientation toward preservingg
the tradition that precludes questioning and challenging it.
28. As Strauss admits in 1972, it had been many decades since he had read Cohen seri-
ously: “I grew up in an environment in which Cohen was the center of attraction
for philosophically minded Jews who were devoted to Judaism; he was the masterr
whom they revered. But it is more than forty years since I last studied or even read
the Religion of Reason, and within the last twenty years I have only from time to
time read or looked into some of his other writings.” See “Introductory Essay to
Hermann Cohen,” JPCM M 267.
29. This also explains a host of apparent contradictions in Strauss’s explanation off
Spinoza. For example, Strauss says that the Ethics is the first and last word off
Spinoza, but he praises Cohen for beginning with the TTP, P which he says is more
revealing.
30. See Richard Kennington, “Analytic and Synthetic Methods in Spinoza’s Ethics,”
in On Modern Origins: Essays in Early Modern Philosophy, ed. Pamela Kraus and
Frank Hunt (Lanham, MD: Lexington Books, 2004), 205–28. For an extended
discussion of Kennington’s argument, see Joshua Parens, Maimonides and Spinoza:
Their Conflicting Views of Human Naturee (Chicago: University of Chicago Press,
2012), 193–212.
31. But Strauss praises Cohen for focusing his study on Spinoza’s TTP P rather than
the Ethics, a method that Strauss himself follows in his own writing and teach-
ing (¶39). In his seminar on Spinoza in 1959 at the University of Chicago, forr
example, Strauss promises to turn to the Ethics after reading the TTP P and the
Political Treatise, but does not do so.
32. The ambivalence runs throughout Spinoza’s critique of the Bible. Spinoza wishes
to use the Bible to establish a liberal society, which in turn, calls into question the
very foundations of that society. As Strauss wrote in his 1936 study of Hobbes:
“Exactly as a Spinoza did later, Hobbes becomes an interpreter of the Bible . . . to
make use of the authority of Scriptures for his own theory, and then . . . to shake
the authority of Scriptures themselves.” Cited in Leora Batnitzky, “Leo Strauss
and the ‘Theologico-Political Predicament,’” in The Cambridge Companion to Leo
Strauss, ed. Steven B. Smith (New York: Cambridge University Press, 2009), 47.
33. In PAW W 187, after claiming that Spinoza deliberately contradicts himself and
that we should always resolve these contradictions by accepting the “the state-
ment most opposed to what Spinoza considered the vulgar view,” Strauss says
that “[o]nly by following this rule can we understand Spinoza’s thought exactly
as he himself understood it and avoid the danger g of becomingg or remainingg the
56 STEVEN FRANK EL

dupes of his accommodations.” Here Harry Wolfson takes the place of Cohen as
an author who explains Spinoza’s contradictions with historical reasons, that is,
reasons “primarily based, not on Spinoza’s explicit statements, but on the history
of the author’s life.” Wolfson “admits that he is trying to understand Spinoza bet-
ter than he understood himself ” (PAW W 188).
34. “[T]he Treatisee is linked to its time, not because Spinoza’s serious or private
thought was determined by his ‘historical situation’ without his being aware off
it, but because he consciously and deliberately adapted, not his thought, but the
public expression of his thought, to what his time demanded or permitted. His
plea for ‘the freedom of philosophizing,’ and therefore for ‘the separation of phi-
losophy from theology,’ is linked to its time in the first place because the time
lacked that freedom and simultaneously offered reasonable prospects for its estab-
lishment” (PAW W 192).
35. See also “Introductory Essay to Hermann Cohen,” JPCM M 280: Cohen’s
“‘optimism’ was too strong.”
36. On the importance of the distinction between wisdom and piety in Maimonides,
see Raymond Weiss, Maimonides’ Ethics: The Encounter of Philosophic and Religious
Morality (Chicago: University of Chicago Press, 1991).
37. This is Werner Dannhauser’s argument in “Athens and Jerusalem or Jerusalem
and Athens” in Leo Strauss and Judaism: Jerusalem and Athens Revisited, ed. Davidd
Novak (Lanham, MD: Rowman & Littlefield, 1996), 155–71. See Hilail Gildin’s
useful rejoinder, “Déjà Jew All Over Again: Dannhauser on Leo Strauss and
Atheism,” Interpretation 25, no. 1 (Fall 1997): 125–33.
38. See “On a Forgotten Kind of Writing,” in WIPP P 221–22.
39. Philosophy may not disagree with Spinoza’s conclusions as much as his presenta-
tion of those conclusions. In his essay on Kurt Riezler, Strauss presents a disagree-
ment between Socrates and Thucydides on whether the high is supported by
nature, or whether it is vulnerable precisely because it lacks support. Thucydides
appears to deny the existence of such support, but nonetheless is not indifferent to
its existence. See “Kurt Riezler,” WIPP P 260.
C APT R 3
CHAPTER

THE LIGHT SHED ON THE CRUCIAL DEVELOPMENT


OF STRAUSS’S THOUGHT BY HIS CORRESPONDENCE
WITH GERHARD KRÜGER

Thomas L. Pangle

T he rather complex private correspondence between Leo Strauss (1899–1973)


and Gerhard Krüger (1902–72) runs from late 1929 through 1935.1 Readers
will presumably be acquainted with Strauss, but a few words are in order to
introduce Krüger—whose fulfillment of his great promise was severely hindered
by the oppression of National Socialism and then, in his early fifties, was cut
short by strokes that left Krüger mentally incapacitated.2
Like Strauss, Krüger studied at the University of Marburg, a center off
Kantian thought that, starting in 1923, experienced the electrifying teaching
of Heidegger (whom Strauss encountered when, after completing his disserta-
tion at Hamburg under Cassirer, he went to Freiburg in 1921 to study with
Husserl). Krüger completed his “Habilitation” in 1929 with a dissertation that
was published in 1931 as Philosophy and Morality in the Kantian Critique. 3 This
book’s unusual stress on Kant’s “anthropology,” or analysis of lived human
experience, indicates the impact of Heidegger. At the same time, the book’s
unusual insistence on the affinities between Kant’s doctrine, as one of obedi-
ence to higher law, and the thought of St. Augustine indicates Krüger’s shar-
ing with Strauss, though from a Christian perspective, deep interest in the
resurgence of morally and intellectually demanding theology—a reawaken-
ing led by Karl Barth, Franz Rosenzweig, Friedrich Gogarten, and Rudolph
Bultmann (with the last of whom Krüger was especially close). Above all, the
book testifies to Krüger’s growing conviction, shared with Strauss and also
Jacob Klein, that classical Greek philosophy (to which Heidegger had given
fresh access but which he regarded as historically surpassed) contained deci-
sive wisdom about the human condition that modern philosophy had covered
over and lost.
58 T H O M A S L . PA N G L E

This trajectory is articulated at the conclusion of Krüger’s Kant book, the


last sentence of which reads: “That the decisive question remains true, even iff
it does not find an answer, the example of Socrates can teach to whoever asks
as did he.” Strauss, in the retrospective 1964 Preface to the German edition
of his Hobbes book, says: “the final sentence of Krüger’s Kant book, which
corresponded completely to my view at that time and with which I wouldd
still today, with certain reservations, agree, explains why I turned entirely to
‘the true politics’ and why I did not write about Hobbes as a Hobbesian.” In
a letter to Krüger of June 1, 1931 (GS–3 387), in which Strauss reacts to his
receipt and first perusal of Krüger’s Kant book, Strauss writes: “I have already
flown through the introduction and the conclusion, which give a clear picture
of your direction: instead of understanding Plato in the light of Kant—as do
the neo-Kantians—on the contrary, to put into question Kant and us through
Plato.”
Krüger’s second book, published in 1939, is entitled Insight and Passion: The
Essence of Platonic Thinking.4 In 1959, Strauss paid this book the high compliment
of referring to it repeatedly in his seminar on Plato’s Symposium (the transcript
of which has been published by the University of Chicago Press in 2001 as
On Plato’s Symposium (OPS)). Strauss introduced Krüger as “a very philosophic
interpreter” (OPS S 39)—a high compliment, coming from Strauss. But I believe
that the subsequent references Strauss makes to Krüger’s book in the course off
the seminar indicate that Strauss found himself more and more dissatisfied with
Krüger’s interpretation, and indeed with Krüger’s whole approach to interpreting
Plato. The core inadequacies of Krüger’s approach, in Strauss’s eyes, are already
visible in the correspondence of the 1930s, in the comments Krüger makes about
Plato when Krüger takes issue with Strauss, and in Krüger’s failure to respond to
Strauss’s countercriticism. In general, Strauss found that Krüger interpreted Plato
not sufficiently on Plato’s own terms, but still too much in the light of the con-
ventional approach, rooted in the tradition of Christian Platonism (see the last
sentence of the unsent draft of a letter from Strauss of December 25, 1935, GS–3
450); this is perhaps the most important respect in which Krüger remained underr
the spell of what Strauss called that “powerful prejudice” according to which it
is impossible to return to an understanding of Plato exactly as Plato intended
himself to be understood.
Let me now focus in on the immediate philosophic context of the correspon-
dence. What at that time bound the two together in their distinctive critical
posture toward modern rationalism was not merely the shared conviction that
the thought underlying the Enlightenment had become evidently a failure,
and that hence the entire modern civilization rooted in the Enlightenment
had entered a state of protracted crisis—that conviction was shared by many
who had experienced the philosophic impact of Heidegger, and his predeces-
sor Nietzsche. What distinguished Strauss together with Krüger, even from
Strauss’s closest intellectual friend and interlocutor Jacob Klein, was the pair’s
growing confirmation of the suspicion that at the heart of modernity’s failure
and crisis was the unanswered theologicall question—the un-disposed-of chal-
lenge to rationalism from suprarational and contrarational revelation. Again in
S T R AU S S ’ S C O R R E S P O N D E N C E W I T H G E R H A R D K RÜ G E R 59

the Preface to the German edition of his Hobbes book, explaining the develop-
ment of his thought around 1930, Strauss writes:

The re-awakening of theology, which for me is marked by the names of Karl Barth
and Franz Rosenzweig, appeared to make it necessary to investigate how far the
critique of orthodox theology—Jewish and Christian—deserved to be victori-
ous. Since then the theologico-political problem has remained thee theme of my
investigations. . . . The philosophic interest in theology linked me with Gerhard
Krüger; his review5 of my Spinoza book6 expressed my intention and my result
more clearly than I myself had done.7

We learn now, when we read Strauss’s long letter to Krüger of January 7, 1930
(GS–3 378–81), that the latter’s review was so helpful in part because Strauss
had—in this letter—explained to Krüger the intention and result of the book,
while prefacing this explanation with the disclosure that part of the reason forr
the book’s deplorable obscurity was the fact that Strauss’s “boss,” and more gen-
erally the institute where he worked (The Academy for the Scientific Study off
Judaism), compelled him to remain silent in public about the true presupposi-
tions that were the point of departure for the book.
This revealing letter to Krüger of January 7, 1930 clarifies and extends what
can be gathered from published writings concerning the meaning, for Strauss,
of the “philosophic interest in theology” that Strauss reports he shared with
Krüger. I would put the key points as follows. Krüger assisted and reinforced
Strauss’s discovery that modern philosophic rationalism had failed in the
grand theological-political project that was its most profound aim. Prior to the
Enlightenment, all of post-classical Western Civilization was dominated by bib-
lical revelation as supremely normative—by biblical law, in either its Mosaic,
or Islamic, or Christian versions. The deepest motivation and meaning of the
Enlightenment was the liberation of civilization from that domination, and, more
positively, the replacement of suprarational revealed law, as the supreme civili-
zational norm, with rational or scientific supreme norms. Strauss and Krügerr
agreed that this vast cultural revolution had been partly successful—but only
partly: modern, Enlightenment rationalism managed to eject revelation from its
cultural or civilizational throne, while endowing mankind with unprecedented
material, technological power. Modern rationalism failed, however, to discoverr
any adequate moral authority or norms, rooted in reason and science, that could
take the place of the previous civilizational guidance by revelation. The con-
sequence has been an ever more obvious spiritual vacuum, apt to be filled by
cultural irrationalism of all sorts, and accompanied, in the realm of so-called
philosophic rationalism, by desperate or even fanatical recoil from all serious
rational thought about foundations. As Strauss put it in his 1951 Preface to the
American edition of his Hobbes book, describing the perspective from which
that book was written, in the early 1930s: “I had seen that the modern mind had
lost its self-confidence or its certainty of having made decisive progress beyond
premodern thought; and I saw that it was turning into nihilism, or what is in
practice the same thing, fanatical obscurantism.”
60 T H O M A S L . PA N G L E

This crisis of modern rationalism compels a reopening of the possibility that


it is necessary to return to revelation—either simply, to traditional orthodoxy;
or to what we might today call “postmodern” neo-orthodoxy. And each of these
possibilities was powerfully expressed in the 1920s: the first above all by Karl
Barth, and the second above all by the “New Thinking” proclaimed by Franz
Rosenzweig (to whose memory Strauss dedicated his Spinoza book). What pre-
vented Strauss and Krüger (or at any rate the side of Krüger to which Strauss
was drawn) from embracing either orthodoxy or neo-orthodoxy is implied in
Strauss’s characterization of theirs as a shared “philosophic interest” in theology:
Strauss and Krüger approached theology not as theologians, but as philosophic
men. They saw the revival of orthodox and neo-orthodox theology as a challenge,
which had to be met, and disposed of, if genuine (rationalist) philosophizing,
if philosophic science—if philosophy as a way of life—were to be tenable for a
serious human being.
In his letter to Krüger of January 7, 1930 Strauss writes that the question
animating his Spinoza book was: “how was it possible, that the Enlightenment
has been victorious?” Strauss then proceeds to provide an illuminating, and, so farr
as I know, otherwise unavailable, summary of the position from which he wrote
the book—a summary that affords, in his words, a “share in the insufficiently-
clear apparent core of my reflections.”
Strauss begins by explaining his dissatisfaction with the answer to his
question commonly given by the contemporary new theologians, andd
expressed by Rosenzweig. They try to respond to the question by assertingg
that the Enlightenment was victorious only “over Scholasticism, but nott overr
Revelation,” nott over “the world of the Bible.” The “sign of the inadequacy
of this answer”—the inadequacy of this attempt to confine the victory of the
Enlightenment—Strauss finds in the present status of the Enlightenment’s
critique of miracles:

the concept of the miraculous is biblical, and, as a consequence of the Enlightenment,


this concept has lost its power and its truth. (It is today an embarrassment: just
read, please, carefully Rosenzweig’s Star of Redemption pp. 119ff.;8 Rosenzweig
has recognized, that the problem of miracles is central; and how does he have to
“interpret” the enlightenment critique of miracles, in order to be able to maintain
the miraculous—and what does he have to regard as miraculous!).

Having indicated the gross inadequacy of the answer typified by Rosen-


zweig, Strauss next asks: “But now what has the Enlightenment achieved in
regard to miracles?” Strauss answers: “It has achieved only this. That it itself,
that is, that the already enlightened human being, is immune to miracles; it
has created a position that is unreachable by miracles.” But—Strauss counters:
“the miraculous is, however, according to its own meaning, only capable off
being experienced as a miracle on the foundation of faith—and thereby, the
Enlightenment offensive is thus rendered impotent.”
“At this point, at the latest,” Strauss continues, “it becomes clear that the
Enlightenment
g te e t does not ot owe its
ts victory
v cto y to assertions
asse t o s of
o the
t e scientific
sc e t c refutation
e utat o
S T R AU S S ’ S C O R R E S P O N D E N C E W I T H G E R H A R D K RÜ G E R 61

of revealed religion.” Instead, “its victory is thanks to a certain will,


ll that one may,
with a grain of salt, characterize as Epicurean.” And “this will,” Strauss declares,
“appears to me to be no foundational justification for the Enlightenment, against
revealed religion.” “The clear indication of this,” Strauss submits, “is the fact,
that it is evident to anyone who has even only an intellectual grasp of what prayerr
is, that the understanding, grounded in the Epicurean disposition, of religion,
is inadequate” (see also SCR R 207–8). Now, given this, Strauss continues, it fol-
lows that “in order for the social victory of the Enlightenment—a nonbinding
state of affairs—to become total, there must emerge anotherr will against revealed
religion.” Strauss says that he sees “such a will disclosed in Machiavelli, Bruno,
and Spinoza,” and “reaching its most extreme representation in Nietzsche, and
its completion in—Being and Time.” “I mean,” Strauss continues, “in the inter-
pretation of the calll of conscience, and in the answer given there to the ques-
tion; who then is calling?” It is “from Heidegger’s Dasein -interpretation that forr
the first time an adequate atheisticc interpretation of the Bible may be possible.”
Strauss seems to be concluding when he writes: “religion is then for the first time
overcome, when it can be given an adequate atheistic interpretation.” But his
statement of the thought that constitutes the foundation of his first book in fact
concludes with an abrupt new tack: “So: the victory of the Enlightenment, that
is, the victory of the ‘scientific view of the world’—which I only understand to
include the loss of the possibility of believing in miracles—is justifiable solely on
the ground of a resolute conviction [bestimmten Gesinnung], not on the ground off
this ‘worldview’ itself.”
This complex and ambiguous statement by Strauss of the thought underly-
ing and animating his Spinoza book makes clear the impact of Heidegger’s
Being and Time on Strauss; and at first, or reading only this far in the letter, one
gets the impression that Strauss is indicating that he wrote his Spinoza book
from a position that was fundamentally Heideggerian, as regards the decisive
issue.9 But: what Strauss says later in this same letter, more specifically about
the contents of the Spinoza book, indicates a deep uneasiness with the previ-
ously elaborated quasi-Heideggerian position. Strauss writes that in his Spinoza
book, “I sought to make clear for myself the various grounds for atheism; thatt
is the reason forr the apparently typological presentation of the first sections off
my work.” In thus clarifying the various grounds of Heidegger’s atheistic pre-
decessors, above all Spinoza, Strauss was in effect bringing to clear view the
various grounds underlying Heidegger, on which Heidegger stands. Strauss
was clarifying the historical-philosophic grounds that Heidegger is apparently
making firm through his atheistic phenomenological analysis of the call off
the conscience. But these historical-philosophic grounds Heidegger himselff
has failed to bring to light. Strauss is in effect going historically deeper than,
and thus criticizing, Heidegger. Strauss is clarifying the various versions of,
first, the “Epicurean” will to atheism; and then, secondly, the additional and
distinct modern Enlightenment will to atheism—the two of which togetherr
comprise the “resolute conviction” (bestimmten Gesinnung) that precedes, cul-
minates in, and is presupposed by, is meant to be vindicated by, Heidegger’s
p e omenology
pheno e o ogy of
o the
t e conscience.
co sc e ce.
62 T H O M A S L . PA N G L E

This letter to Krüger thus helps us to see that Strauss’s Spinoza book is in
fact also tacitly intended to be a critical interrogation of Heidegger’s unarticu-
lated and insufficiently investigated historical-philosophic grounds for atheism.
The Spinoza book brings to sight both the strengths and the troubling shakiness
of those grounds.10 And this letter to Krüger—written a year after the book
was completed, as Strauss notes—shows that Strauss is now aware that he did
not himself sufficiently appreciate, when he wrote the book, how veryy problem-
atic were the implications he was showing, or discovering. For Strauss goes on
to say, coming to the close of this letter: “That this orientation, starting from
my presuppositions, can no longer justify itself, I of course concede to you.”
Strauss had opened the letter by expressing his “heartfelt thanks” to Krüger forr
the latter’s previous letter (this important letter has unfortunately not survived),
and its critique of Strauss’s whole position in the Spinoza book: “most heartfelt
thanks for your letter,” that “will substantially contribute to transforming my
general uneasiness about my work into concrete doubts, and transformations off
my previous questions.”
As we learn from the next surviving letter to Krüger, written half a yearr
later (on June 26, 1930; GS–3 382–83), Strauss soon stumbled across the source
from which he was to discover the path out of the aporia in which he foundd
himself. Strauss writes to Krüger with a request to be invited to give a seminarr
in Krüger’s class on Augustine the coming winter, and explaining his request in
the following words:

I had begun my work on a Jewish Scholastic—Gersonides—as a work of “pure


scholarship,” and also because I must deliver some concoction to the people who
pay me. But soon I observed that the work was not so boring to carry on, simply
because the subject is exciting. It concerns the problem of that moderate (i.e. non-
atheistic) enlightenment, about which I have learned so much from your Kant
work.

Strauss proceeds to summarize what he has discovered thus far about the dis-
tinctive character of the “moderate” or nonatheistic Judeo-Arabic medieval
enlightenment—in contrast to the modern “moderate” enlightenment, typi-
fied by Krüger’s reflections on Kant, but also probably by Strauss’s own work
on Mendelssohn (and also in contrast to the medieval Christian-Scholastic
enlightenment).
To begin with, Strauss formulates what he has found the Judeo-Arabic
version shares with the eighteenth-century moderate enlightenment: “preva-
lence of belief in providence, as belief in the good God over and above the
God who calls one to account; and therefore belief in the sufficiency of reason.”
Then Strauss outlines what he has found distinguishes the Judeo-Arabic moder-
ate enlightenment from the eighteenth-century moderate enlightenment: first,
the “primacy of theory,” as opposed to the “primacy of morality (veneration
of Socrates)” in the eighteenth-century enlightenment; second, and linked to
the preceding, in the Judeo-Arabic moderate enlightenment, “natural right”
or natural law plays no role, or at least not anything like the role that it plays in
S T R AU S S ’ S C O R R E S P O N D E N C E W I T H G E R H A R D K RÜ G E R 63

Christian medieval thinking—whereas, in the eighteenth-century moderate


enlightenment, the moral law is developed as a “natural right” that demands
a new and specific kind of constitutional law. This leads immediately to the
thirdd and most important contrast: in the Judeo-Arabic moderate enlighten-
ment, the place of natural right or natural law is taken by the divinely revealed
positive law of Moses, or of Mohammed—which divinely revealed law is,
however, reconceived and reinterpreted as fully intelligible in the light off
the principles of Plato’s Republic. Moses and Mohammed are reconceived and
reinterpreted as “philosopher-lawgivers”; and the scriptural law itself is recon-
ceived and reinterpreted as the “sole sufficient and binding norm, leading life
to the happiness that consists in theorizing.” Strauss concludes that the Judeo-
Arabic middle ages “has, through the link to the ancient idea of the concrete
nomos [law] and nomothetes [lawgiver], a much greater possibility than does
the natural right-preoccupied 18th century of accepting the concrete orderr
of revelation.”
Now in the Spinoza book, Strauss had already given some attention to the
Judeo-Arabic critique of religion, and its relation to the modern Enlightenment’s
critique. He had done so in the first place through his discussion in the openingg
chapter (SCR R 46–49 = 13–17) of “Averroism”—“a tradition which has remained
active for about five hundred years.” But as the preceding words suggest, Strauss
had seen and stressed continuity rather than contrast:

Three tendencies and traditions of very different origin underlie seventeenth-


century criticism of religion. They are traditionally designated by the names
Epicurean, Averroist and Machiavellian. From an early time they were in such
close association that it becomes difficult to characterize the general movement
of criticism of religion in the seventeenth century by one name rather than by
another.

Strauss had been interested not so much in the authentic philosophizing off
Averroes himself or his predecessors as in the Western European tradition
believed to have grown out of Averroes: “in Christian Europe knowledge off
the true Averroes is more and more replaced by the legend of Averroes.” Strauss
had taken notice of “Averroism” not as a profound critique of, or even engage-
ment with, revelation but as an attempt to conceive of religion as “needed forr
the guidance of the ignorant many, for the sake of law and order.” And the sixth
chapter’s lengthy discussion of Maimonides had treated the latter as a target
of Spinoza’s critique of religion. From this perspective, Maimonides came to
sight in rather conventional garb, as a kind of Jewish scholastic who made use
of philosophy conceived as Aristotelian natural science to defend revealed cre-
ationism against rationalist natural science. “According to the inner structure off
Maimonides’s science,” Strauss writes, “the insight into the insufficiency of the
human understanding—an insight gained on the basis of Aristotelian science, in
principle prior to the introduction of the central theological presupposition— —
motivates the recourse to revelation; this insight inclines man to the acceptance
of revelation” (SCRR 158 = 141).
64 T H O M A S L . PA N G L E

In contrast, the letter of June 26, 1930 shows that what has revolutionized
Strauss’s approach to the medieval Judeo-Arabic “moderate” Enlightenment is
the stress Strauss now places on the link to “antiquity” as meaning especially
the link to Plato, and his Republic. In the Spinoza book, Strauss sees only the
Aristotelianism of Maimonides; Strauss is unaware of Maimonides’s profound
debt to Plato and above all Plato’s politicall philosophy—as thee decisive way off
addressing the theological question. In the Spinoza book, Strauss still thinks
that: Maimonides’s context of thought may be summed up as a nexus of sci-
entific reasoning” (SCR R 161 = 144). To be sure, Strauss does make “ancient”
Greek philosophic-critical theology a major theme in the Spinoza book. But
the ancient philosophic-critical theology he has in mind is Epicurean, which
he takes as the most serious (even the sole serious) ancient critical-theological
thought. Similarly, in the earlier letters to Krüger, Strauss equated “ancient”
theological criticism with “Epicurean” philosophic theology (see esp. 380:
“ancient [Epicurean] critique”). From this new letter of June 26, 1930, how-
ever, we see that in the preceding six months it has begun to dawn on Strauss
that what is most profound and valuable in the medieval Judeo-Arabic think-
ers is the guidance they give back to Plato—as above all a politicall philosopher:
that is, to a Plato, and to Platonic dialogues, that are radically differentt from
what is conventionally understood by modern scholarship, shaped by the tradi-
tion of Christian or Augustinian Platonism. What Strauss has begun to dis-
cover through Maimonides and his Muslim teachers, above all Farabi, is a Plato
whose dialogues and political philosophy have to be completely reinterpreted,
as thee philosophic key to meeting the challenge posed by revelation, seen as
“the theologico-political
- l problem” (see also the draft of Strauss’s unsent letter off
December 25, 1935, 449–50).
This becomes still more explicit in the next two surviving letters from Strauss
to Krüger. The first, dated February 27, 1931 (GS–3 383–84), refers to Strauss’s
delivery on December 21, 1930, of his lecture entitled “Religious Situation off
the Present”—in which for the first time Strauss spoke in public of the modern
historical-spiritual condition as that of humans dwelling in “a cave, beneath that
cave,” which is described in the Platonic Socrates’s famous image at the begin-
ning of Book Seven of the Republic. The second letter, dated May 7, 1931 (GS–3
384–85), refers to a public lecture Strauss has just given on Hermann Cohen and
Maimonides, which, Strauss reports, is his first public presentation of his thesis
that “the Islamic-Jewish Scholastics understood revelation within the framework
laid out in Plato’s Republicc and Laws.” In this letter Strauss speaks of “how much
I have profited from your Plato lecture.” In other words, Strauss continues to
speak as if his own incipient recovery of the genuine Platonic philosophic under-
standing of and response to revelation is in accord with, and even helped by, the
development of Krüger’s thinking.
This goes with the strong suggestion of the earlier, first letter mentioning
Plato ( June 26, 1930, GS–3 382–83), where Strauss writes as if he is mov-
ing in Krüger’s direction—that is, toward a quest for the grounds of what
Strauss there calls a “moderate” or “nonatheistic” Enlightenment. BUT: do
S T R AU S S ’ S C O R R E S P O N D E N C E W I T H G E R H A R D K RÜ G E R 65

Strauss and Krüger agree in their understanding of what this “nonatheistic


Enlightenment” means? Above all, do they agree on the meaning of the “possi-
bility” Strauss refers to as: “the philosopher’s ‘acceptance’ of the concrete order
of revelation?”11 To what extent is this expressedd closeness to Krüger giving
voice to Strauss’s self-understanding at this time frankly and straightforwardly,
OR, to what extent is Strauss writing in a manner designed to try to begin
to entice his friend to join in pursuing a path of self-discovery, along which
Strauss has already proceeded much further—far enough to see that the truths
Strauss is beginning to unearth from his study of Plato, guided by the medieval
Judeo-Arabic thinkers, will require a radical and wrenching, even agoniz-
ing, self-transformation on the part of Krüger: the profundity, and the crux,
of which self-transformation, Strauss knows from his own transformation in
the intervening months? This much is clear: in the correspondence of the two
subsequent years, it becomes more and more evident that, and why, Krügerr
cannot follow Strauss. Krüger cannot bring himself seriously to entertain the
possibility that Plato supplies the answerr to the theological question; and this is
chiefly because Krüger remains tied to the thought that Strauss identifies as the
core of historicism.
In a December 27, 1932 letter (GS–3 419–22) replying to Krüger’s expressed
reservations about Strauss’s contention that all specifically “modern” thought
dwells in a “cave beneath Plato’s cave,” Strauss writes: “the ‘substantial and his-
torical core’ of historicism is, as you correctly say, ‘the factual domination of Christ
over the post-classical world.’ What follows from that, for him, who does nott
believe, who thus denies the right, that is the divine right, of this domination?”
Strauss answers: “the immediate consequence—in Heidegger among others—is:
Christianity has brought to light facts of human life, that were unknown orr
inadequately known to ancient philosophy”; this means, “fundamentally: afterr
the disintegration of Christianity, there remains, and first becomes possible, phi-
losophy that preserves the ‘truth’ of Christianity; it is as such deeper and more
radical than ancient philosophy.” Strauss continues: “maybe this consequence
is correct—but it must be as such proven. And this is possible only on the basis
of a directt confrontation of modern with ancient philosophy. So much on the
legitimation of my project as regards Hobbes—I mean, the direct confrontation
with Plato.”
These letters to Krüger allow us to observe the point of departure from which
Strauss set out on the journey, taking up the rest of his life, confronting Socratic
with non-Socratic thought, ancient and modern, and thereby vindicating everr
more powerfully Socratic political philosophy—as the decisive response to the
most fundamental and all-encompassing question.
In subsequent letters to Krüger, Strauss tries repeatedly to give his friendd
leading or thought-provoking clues to precisely what sortt of questioning it is that
one can find to be the decisive questioning, in the Platonic–Socratic dialogues, iff
the dialogues are studied correctly—and that means in liberation from all mod-
ern scholarship, with its post-Christian and historicist prejudices. Sadly, Krügerr
seems never to have caught g on.
66 T H O M A S L . PA N G L E

Two of the passages giving such clues seem to me to be especially helpful. In


the letter of December 27, 1932 that I quoted just previously Strauss writes:

Granting the greater depth of post-Christian philosophy: but, is depth what counts?
Is the perspective of depth not itself a Christian perspective, that for its part requires
proof? Is depth identical with radicalism? Is it not perhaps the case, that “depth” is
not really radical? Depth is at home in self-examination. Self-examination presup-
poses a standard. The question as to the standard is the radical question. I find, that
the moderns on the whole have neglected this question, while they apparently or
really have demanded self-examination.

A year and a half later, on August 18, 1934 (GS–3 439–42), responding to his
reading of a later Kant essay by Krüger, Strauss returns to this point and elabo-
rates, as regards Kant, in the following most revealing words: “I do not know
whether one should proceed on the basis of the distinction between theoretical
and practical knowledge as much as do you, following Kant. As important as this
difference is, it seems to me to be secondary. Practical knowledge is knowledge off
an obligation, on the basis of an obligation.” “But,” Strauss continues, “more fun-
damental than obligation is, whatt is obligatory and what, ‘merely for us humans,’
takes on the character of the obliging. Platonic philosophy is concerned with the
character of this what—which does not in itself have the character of a law in the
precise sense.” For Kant, “the question about the law first emerges in connection
with the application of the standard to human beings.” More generally, “modern n
morality is from the beginning conceived as an applicablee morality, and, I believe,
despite the incomparable radicalization it undergoes in the Kantian morality,
this (latter) still is in this sense specifically modern.” In profound contrast, Plato’s
critical philosophy, while understanding itself to be “summoned by the law, asks,
not so much about the law, but rather about the right ordering of human life— —
and therefore about the principle of the ordering.” Strauss interjects this poignant
remark: “But I am writing as if we just yesterday had a conversation, and am not
taking account of the fact that these indications are perhaps comprehensible only
for me.” By this time, Strauss had left Germany for good.12
In 1958 there was published by Karl Alber Press a collection of essays Krügerr
had written before his brain strokes, entitled Freedom and World Administration
(Freiheit und Weltverwaltung). Krüger sent a copy to Strauss, eliciting a letter off
June 21, 1958 (GS–3 450–51) in which Strauss expresses his “heart’s joy” in hear-
ing again from Krüger through the book, which Strauss says he has read twice.
“How near we come in the questions and even in the general direction in which
we seek the answers! Your discussions of the division between the Ancients and
the Moderns has especially taught and delighted me.” But then Strauss proceeds
to expatiate on how and where he departs from Krüger:

you concede, it seems to me, the necessity of distinguishing between the human as
physical-earthly being with its inadequate perspectives, and the human as absolute
wayfarer, who is on his way to thee Truth. A corresponding distinction would also
be necessaryy as regards
g the essential tension between the ariste politeia
p [
[best regime]
g ]
S T R AU S S ’ S C O R R E S P O N D E N C E W I T H G E R H A R D K RÜ G E R 67

and the actual “natural community”—not to mention the fundamental difference


between the highest praxis, which is only theoria, and all other praxeis. . . .
Decisive, I believe, is the difference in regard to the “natural community.” Its
“naturalness” is doubtful, because in the strict sense only the ariste politeia is natu-
ral (cf. the problem of the agathos anerr [good man] in contrast to the agathos polites
[good citizen] in [Aristotle’s] Politics III). To take the extreme Platonic expression
of this state of things, the polis is the cave—there exists a necessary tension between
the polis and philosophy (for that very reason the ariste politeia needs the kalon pseu-
dos [noble lie]). The problem becomes hidden, but not solved, when one replaces
the polis by the ethnos [nation].

Notes
1. In GS–3 377–454, Heinrich Meier has made these letters available through pains-
taking editorial work, and in his introduction has brought intelligent learning to
bear in framing their context (esp. GS–3 xxviii–xxx). Unless otherwise noted,
all page references will be to this edition; italics in quotations from Strauss andd
Krüger are in the original.
2. For a fuller account of Krüger’s career, see esp. the obituary by Krüger’s lifelongg
friend Hans-Georg Gadamer in Archives de Philosophiee 47 (1984): 353–63.
3. Philosophie und Morale in der Kantischen Kritik (Tübingen: Mohr, 1931; 2nd ed.
1967).
4. Einsicht und Leidenschaft: Das Wesen des platonischen Denkens (Frankfurt:
Klostermann, 1939; 2nd ed. 1948; 3rd ed. 1961).
5. In Deutsche Literaturzeitungg 51 (December 20, 1931): 2,407–12. An English trans-
lation, by Donald J. Maletz, was published in the Independent Journal of Philosophy,
vol. 5/6 (1988): 173–75. Quotation is from the latter.
6. Die Religionskritik Spinozas als Grundlage seiner Bibelwissenschaft: Untersuchingen zu
Spinozas Theologisch-politischem Traktatt (Berlin: Akademie-Verlag, 1930; repr.,
Hildesheim: Georg Olms, 1981); republished in GS–1 1–362; English translation
by Elsa Sinclair published as Spinoza’s Critique of Religion (New York: Schocken,
1965). All references here will be to pages of the latter, abbreviated as SCR,
sometimes with equation marks in parentheses indicating page numbers of the
German original (printed in the margins of GS–1).
7. When Strauss agreed to have an English translation of his Spinoza book executed
and published, his revision profited from Krüger’s critical suggestion, at the close
of his review (175), that “the specific divisions provided by the table of contents
would very much facilitate the reading if they were still more detailed and indi-
cated in the text by more than dashes.”
8. Franz Rosenzweig, Der Stern der Erlösungg (Frankfurt am Main: J. Kauffmann
Verlag, 1921), 119–42, subtitled “Über die Möglichkeit, das Wunder zu erleben”
[On the Possibility of Experiencing Miracle]; trans. William W. Hallo (New
York: Holt, Rinehart and Winston, 1971), 93–111, or Barbara Galli (Madison,
WI: University of Wisconsin Press, 2005), 101–21.
9. This is not ruled out by what Strauss says when he goes on to express in very
personal terms the distinctivee way in which he shares in the “will” and the “convic-
tion” [Gesinnung] that he sees animating modern atheism: “I must justify myselff
[mich rechtfertigen] before the forum of the Jewish tradition”: and “truly, without
anyy philosophy-of-history
p p y y ref lection,” but “simply
p y because I hold it to be not
68 T H O M A S L . PA N G L E

defensible [nicht
h vertretbar
b r] that I abandon out of thoughtless lightness and indolent
comfort a cause, for the sake of which my ancestors took upon themselves every-
thing conceivable.” It is this Jewish moral passion, for self-justification before the
tribunal of his ancestors, that Strauss indicates he understands to be driving his
relentless quest for the truth about the most momentous question.
10. See esp. SCR R chap. 5, sec. C, 144–46 (= 126–28), “The Premises and the
Limitation of the Critique of Orthodoxy”; also 123, 179; and Ch. 7, sec. A,
193–200 (= 182–90), “Calvin’s Position as Immune to [unerreichbar für] r Spinoza’s
Critique” as well as sec. B., 200–4 (= 190–94), “The Illusion of the Critique.” See
also Krüger’s review, 175: “The general discussion about the difference between
modern and ancient thought receives here for once an ‘existential’ sharpness:
Strauss shows in concreto how much the modern ‘disposition of method, of cul-
ture’ (44; 71) is a historical antithesis, that is, an unprovable negative life-decision
opposed to that past which believed in revelation.”
11. See Strauss’s highly paradoxical formulation in Philosophy and Law’s first chap-
ter, which Heinrich Meier informs us (GS–2 xvi n11; trans. as n11 of ch. 1 in
the present volume, above) was originally completed in September 1933: “It is
in the Laws that Plato undoubtedly stands closest to the world of revealed law,
since it is there that, in accordance with a kind of interpretation anticipating the
philosophic interpretation of the revealed law among the medieval thinkers, Plato
transforms the ‘divine laws’ of Greek antiquity into truly divine laws, or recog-
nizes them as truly divine laws. In this approximation to the revelation without
the guidance of the revelation we grasp at its origin the unbelieving, philosophic
foundation of the belief in revelation.”—PLA 76.
12. Krüger’s leading student, Klaus Oehler, reports that in 1951 “Krüger asked me,
‘Do you know who Leo Strauss is?’—to which I had to reply at that time that
I did not. Then he said to me: ‘If Leo Strauss had not been compelled by the
German political situation to depart, philosophy in Germany would have taken a
different direction.’” Klaus Oehler, Blicke aus dem Philosophenturm: Eine Rückschau
(Hildesheim: Georg Olms, 2007), 185; see also 179.
C APT R 4
CHAPTER

STRAUSS ON HERMANN COHEN’S “IDEALIZING”


APPROPRIATION OF MAIMONIDES AS A PLATONIST

Martin D. Yaffe

H ermann Cohen’s “Characteristics of Maimonides’s Ethics” appeared in


1908, in a collection of essays on Moses Maimonides by various German
Jewish scholars.1 Cohen’s essay was the subject of a 1931 lecture by the young
Leo Strauss to the Academy for the Scientific Study of Judaism in Berlin, entitled
“Cohen and Maimonides.”2 At that time, as Strauss later recalled, Cohen was
“the center of attraction for philosophically minded Jews who were devoted
to Judaism . . . the master whom they revered.”3 Strauss’s lecture vindicates his
older contemporaries’ reverence for Cohen to the extent that it shows Cohen to
be an extraordinarily helpful (if ultimately inadequate) guide for understanding
Maimonides when one starts, as they do, from modern premises, that is to say,
from premises supplied by the (modern) Enlightenment. Cohen himself, Strauss
shows, reveres Maimonides as an enlightened Jew, indeed as a role model for mod-
ern enlightened Jews like himself, and in this regard interprets Maimonides as a
Platonist rather than, more conventionally,4 as an Aristotelian. Strauss endorses
Cohen’s Maimonides interpretation—up to a point. He demurs at Cohen’s ret-
rofitting onto Maimonides his own, “idealizing” way of interpreting Jewish
sources.5 Accordingly, Strauss’s lecture probes the merits and shortcomings off
Cohen’s “idealizing” appropriation of Maimonides and offers a corrective.
Strauss’s corrective emerges from three interconnected criticisms: 1) Cohen’s
reverence for Maimonides as an enlightened Jew is paradoxical. 2) Cohen’s inter-
pretation of Maimonides as a Platonist rests on the unsustainable claim that Plato
unlike Aristotle privileges ethics over philosophy; rather, both Plato and Aristotle
privilege philosophy over ethics. 3) Cohen’s interpretation likewise rests on the
further unsustainable claim that Maimonides privileges ethics over philosophy;
rather, like both Plato and Aristotle, Maimonides too privileges philosophy overr
ethics. In what follows, I consider each criticism in turn, along with Strauss’s
corrective.
70 M A RT I N D. YA F F E

Cohen’s Reverence for Maimonides as an Enlightened Jew


What Strauss finds paradoxical about Cohen’s reverence for Maimonides has
to do with Cohen’s notion of an enlightened Jew. Strauss argues as follows.
The term “enlightened Jew” is, on the surface, self-contradictory. Judaism
(the Torah) emerges historically within a mythical horizon. By its own lights,
then, the Torah would seem oblivious to science. But “enlightenment” means sci-
entific enlightenment. Hence Judaism and enlightenment would seem at bottom
incompatible. Admittedly, one might wish to construe the Torah’s distinctive
doctrines— creation, providence, etc.—as compatible in retrospect with compa-
rable scientific doctrines. Yet this would not prove that the original authors off
the Torah understood its doctrines in that way. Nor, being anachronistic, would
it be consistent with the principles of modern critical (scientifically-enlightened)
historical investigation. One would need to show in addition, or instead, that the
Torah as understood by its original authors somehow enjoins scientific enlight-
enment. As Cohen points out, this in fact is what Maimonides claims. Since
the Torah’s enjoining scientific enlightenment is not evident from a merely lit-
eral reading, however, Maimonides interprets the Torah allegorically.6 Thus, forr
example, he identifies ma‘aseh bereshith and ma‘aseh merkabah7 allegorically with
Aristotelian cosmology.8 Cohen takes Maimonides’s allegorizing as a precedent
for his own idealizing interpretation. Strauss finds this paradoxical.
The paradox shows up in that Maimonides’s allegorizing and Cohen’s ideal-
izing interpretation differ in their respective aims. Both differ as well, of course,
from literal interpretation. Maimonides’s allegorizing agrees with literal inter-
pretation in aiming to understand the Torah’s original meaning. The difference
between them turns on whether the original meaning is to be understood by
limiting oneself to a plain-sense reading of the text, as, say, modern biblical criti-
cism does,9 or whether instead it is to be understood by inferring an inner mean-
ing of which the plain text is merely the outward expression, as Maimonides’s
allegorizing does. In contrast to both modern biblical criticism and Maimonides,
Cohen claims to understand the Torah betterr than its original authors.10 Ratherr
than simply interpreting the Torah, he consciously reinterprets it. Yet doing so,
he claims, is in line with the Torah’s real intention. Here Cohen appeals partly to
modern biblical criticism and partly to Maimonides. Like modern biblical criti-
cism, Cohen rejects allegorizing as historically dubious. Like Maimonides, how-
ever, he is guided by the biblical prophets’ intrabiblical critique of the Torah’s
laws about animal sacrifices to the effect that those laws had an ulterior purpose.
According to Maimonides, the laws in question were meant in the long run to
wean worshipers from animal sacrifices altogether and to habituate them to focus
instead on what really matters to the Torah.11 Cohen takes Maimonides’s allego-
rizing reading, as it were, literally. According to Cohen, what really matters to
the prophets—and hence retroactively to the Torah—is not animal sacrifices as
such but the pure worship of God which enjoins human beings to ethical action.
That is to say, with their putative assertion that the laws about sacrifices point to
enjoining human beings to ethical action, the prophets look beyond the Torah’s
mythical horizon toward a more enlightened horizon. They thereby set in motion
an ongoing, biblically sanctioned struggle for the progressive development of
S T R AU S S ON M A I MON I DE S A S A P L AT ON I S T 71

reason for the sake of ethical action. Cohen appeals to that intrabiblical prec-
edent, so construed, while going beyond it.12 Appropriating Maimonides as
his enlightened role model, he radicalizes the prophets’ reinterpretation of the
laws about sacrifices by regrouping alll the Torah’s laws around the intention the
prophets are seen to attach to those particular laws—progressive enlightenment
for the sake of ethical action. Such, in brief, is the drift of Cohen’s idealizing
interpretation. Even so, says Strauss, it differs from Maimonides’s allegorizing in
a paradoxical way.
The paradox is this. For Maimonides, enlightenment would seem to mean
replacing the Torah’s original mythical horizon with a horizon constituted by
Aristotelian cosmology. Paradoxically, however, Cohen polemicizes against
Aristotelian philosophy as fundamentally misguided—indeed as a betrayal off
philosophy as understood by Aristotle’s teacher Plato, namely, as has been said,
philosophy for the sake of ethical action.13 To avoid the paradox, Cohen inter-
prets Maimonides at bottom as a Platonist rather than the Aristotelian he seems
to be on the surface. Before examining Cohen’s Maimonides interpretation,
then, Strauss considers Cohen’s claim about Aristotle and Plato.

Cohen’s Claim that Plato unlike Aristotle Privileges


Ethics over Philosophy
Notwithstanding the polemical character of Cohen’s complaint against Aristotle,
Strauss corroborates much of what Cohen claims differentiates Aristotle from
Plato. As regards Plato, Cohen’s claim reduces to three points. First, Plato, orr
Plato’s Socrates, is concerned only with the question of the just life.14 Second,
the question of the Good (as the highest and most urgent question, which
emerges from and guides the question of the just life) is an ongoing question
that is never fully disposed of but keeps showing up in day-to-day moral ques-
tions.15 Third, moral questions are inseparable from political life as such.16 Now,
lest the question of the Good as the highest and most urgent question be thought
to be merely self-evident and not distinctive to Plato as Cohen claims, Strauss
subjects this part of Cohen’s claim to a brief thought-experiment. Under what
conditions, Strauss asks, would the question of the Good nott be the highest and
most urgent question? Two conditions suggest themselves: first, if the just life
were nott questionable, that is to say, if we knew what the Good is; and second,
if human life were nott the most important thing. As it happens, these same
two conditions characterize Aristotle’s philosophy, as Strauss points out and
Cohen also recognizes. First, as Cohen says of Aristotle: “The moral is not a
problem for him.”17 (In corroboration, Strauss quotes Aristotle himself on the
aim of ethical inquiry: “we are inquiring not in order to know what virtue is
but in order to become good.”)18 Second, as Cohen also says of Aristotle: “His
metaphysics is directed toward the universe.”19 (Again, Strauss quotes Aristotle
in corroboration: “it is absurd to believe that politics or prudence is the most
serious [knowledge] if man is not the best thing in the cosmos.”)20
Strauss’s thought-experiment lets him differentiate Aristotle from Plato in
agreement with Cohen without being distracted by the polemical character off
Cohen’s complaint against Aristotle—so as to be able to examine the validity
72 M A RT I N D. YA F F E

of Cohen’s complaint free of polemics. Since Strauss’s focus here is on Aristotle


and Plato in the context of Maimonides, he postpones that examination till dis-
cussing Maimonides directly. Meanwhile he reformulates Aristotle’s differences
from Plato nonpolemically:21 First, whereas Platonic-Socratic philosophy is con-
cerned with the question of the just life, Aristotle restores the possibility of the
life spent in pure contemplation, or theorizing (theôrein), originally exemplifiedd
in the Pre-Socratic philosophers but “called into question by Plato’s Socrates.”
Second, whereas Platonic-Socratic philosophy is questioning about the good off
human life, hence about “the true state,” Aristotelian philosophy is “pure con-
templation of existing [things] and understanding of Being.” Finally, whereas the
aim of Platonic-Socratic philosophy is inherently political, for Aristotle “politics
moves into second place.”
Before proceeding to Maimonides, Strauss is prompted to ask: Does it not
follow from Aristotle’s privileging of pure theorizing over political inquiry that
Aristotle is enlightened in the full sense of the word? Why does Cohen nev-
ertheless consider Plato more enlightened? The answer, Strauss goes on to say,
has to do with Cohen’s extending the struggle against myth as taken up by the
biblical prophets to a struggle against pure theorizing as such. Mythical man is
said to concern himself exclusively with the cause of the world in the sense off
its origin—its whencee or wherefrom. Only (biblical) religion, says Cohen, raises
the question of the cause of the world in the sense of its purpose—its whitherr
or whereto. Aristotle, being concerned with the world’s origin only, therefore
remains within the horizon of the mythical. That is to say, even though the
highest knowledge according to Aristotle is knowledge of God, Aristotle’s God
is not the biblical God, who manifestly enjoins ethical action.22 Cohen infers that
if the biblical God is to become the object of knowledge, as for Maimonides, this
change must affect theorizing too. Thus Maimonides in Yesodei ha-Torah, when
discussing the commandment to love and fear God, asks what is the way to love
and fear of God, and answers that it is knowledge of the world, that is to say,
(Aristotelian) physics.23 Here Maimonides, by interpreting physics in connection
with the putative moral purposiveness of the biblical God, is seen to go beyondd
Aristotle’s own understanding of physics (and therefore of metaphysics or theol-
ogy). In Cohen’s expression: when knowledge of God becomes knowledge of the
biblical God, theorizing itself is “uprooted.”24

Cohen’s Claim that Maimonides like Plato Privileges


Ethics over Philosophy
Nevertheless Strauss faults Cohen for ascribing to Maimonides a Platonism
conceived, or rather misconceived, as philosophy for the sake of ethical action.
The fault shows up in Cohen’s misconstruing Maimonides’s negative theology.
Maimonides argues that assigning positive attributes to God is incompatible with
God’s unity and uniqueness.25 Even so, positive knowledge of God, as human
beings’ highest possibility, remains possible. What Maimonides denies of God
are attributes that amount to defects, or privations. Positive attributes, properly
understood, are “negations of privations.” To call positive attributes negations off
p
privations is not to sayy somethingg trivial. For example,
p to call God “powerful”
p
S T R AU S S ON M A I MON I DE S A S A P L AT ON I S T 73

is not to say merely that He is “not weak,” but that His existence suffices in
bringing forth things other than Himself.26 Here Cohen consciously interprets
Maimonides in line with his own theology—although he alerts his reader to this
fact and invites him to double-check. (Strauss will respond to Cohen’s invitation
in a moment.) In Cohen’s interpretation, rather than simply denying positive
attributes, Maimonides is replacing immanent or absolute attributes, that is,
attributes characterizing God as such, by transitive ones—attributes describing
God in His correlation with the world. By limiting our knowledge of God to
knowledge of that “correlation,” Cohen is saying that God is knowable not as the
object of metaphysics but only as a moral being, or rather as the paradigm or (in
Cohen’s word) Idea of morality. In short, Cohen says Maimonides equates “nega-
tions of privations” with what Maimonides calls “attributes of action” (which
Cohen interprets to mean moral action).
Strauss finds three difficulties here. First, the passage that Cohen claims
equates “negations of privations” with “attributes of action” instead differ-
entiates these as two ways of stating something of God.27 Second, according
to Maimonides some negations of privations do nott involve correlations—forr
example, “God lives” means no more than “God is not without life,” “not
ignorant,” “not dead,” etc.28 Third, whereas Cohen is content to say that in
the Torah God is revealed exclusively as a moral being, 29 Maimonides sees a
need to explain why the Torah limits itself to revealing God in that way, given
that God also has attributes that concern His actions vis-à-vis the extra-hu-
man world. 30 In Maimonides’s explanation, revealing God’s moral attributes is
needed specifically for the existence and guidance of human beings in political
communities. To obscure the difference between God’s moral actions and His
actions vis-à-vis the extra-human world, as Cohen does here, is to assume that
morality (human morality) is the purpose of the world. This is in fact the view
Cohen ascribes to Plato in contrast to Aristotle.31 But does Maimonides teach
otherwise than Aristotle here, Strauss wonders? Maimonides recognizes the
opinion that according to “our” view—the view of those who believe in the
creation of the world (in contrast to those who adhere to Aristotle’s doctrine
that the world is eternal)—the ultimate purpose of the world can be asked about
and that its purpose is for the human race to worship God. 32 Yet Maimonides
rejects this opinion by saying that the world has no purpose besides God’s free
will or wisdom and by referring to how small human beings are as compared
to the cosmos (understood in Aristotle’s terms as the spheres and separate intel-
ligences). Moreover, Maimonides appropriates Aristotle’s distinction between
dianoetic and ethical virtues by teaching that moral perfection is lower in rank
than intellectual perfection, since moral perfection concerns the mutual rela-
tions of human beings and is at bottom useful only for an individual human
being in relation to others, whereas intellectual perfection benefits the indi-
vidual alone and simply. 33 In short, Strauss finds Cohen’s basic assertion—that
knowledge of God as knowledge of the God of Israel “uproots” theorizing—in
need of correction.
In correcting it, Strauss notes how Maimonides rethinks Aristotle’s distinc-
tion between moral and intellectual perfection. On remarking that whereas the
moral man depends on others the wise man is self-sufficient to the degree that he
74 M A RT I N D. YA F F E

is wise, Aristotle adds that the wise man may also experience incidental benefit
to himself from those who collaborate in his theorizing.34 Maimonides inverts
Aristotle’s addendum in saying instead that those who understand for themselves
may also benefit others by helping them understand. 35 Maimonides’s inversion
invites an explanation. Strauss looks for it in Maimonides’s account of prophecy,
since while Maimonides like Aristotle teaches that the philosopher is superior to
the moral man, unlike Aristotle he teaches also that the prophet is superior to
the philosopher. “Hence,” Strauss infers, “if there is a limitation, a calling into
question, an uprooting of the Aristotelian ideal in [Maimonides], then it must
show up in his prophetology.”36 Strauss therefore turns to Maimonides’s account
of prophecy.
In Maimonides’s account, the prophet is a philosopher and more. He can sur-
mise future events by bodily visualizing them in front of him. Exercising his
vivid power of imagining as well as that of knowing, he can present philosophical
insights figuratively so as to communicate them to the unphilosophical multi-
tude.37 His imaginative power matches that of statesmen, lawgivers, diviners, andd
magicians (miracle-workers)—yet he exceeds any or all of these by being philoso-
pher, statesman, lawgiver, diviner, and magician combined. The Torah accords
the highest rank of prophecy to Moses, a lawgiving prophet,38 but does not spell
out exactly what distinguishes him from the others: knowledge? statesmanship?
divination? magic? The basic question here, says Strauss, is: What is ultimately
important about prophecy—what purpose or need does it ultimately address?
Maimonides, Strauss admits, offers only an implicit answer. Strauss sketches
it as follows,39 before pointing to its likely source in Maimonides’s Muslim pre-
decessors. Human beings, diverse and contrary as they are individually, need
socializing by a leader who can bring about agreement among them through
law. Leaders divide into lawgivers and governors; governors, unlike lawgivers,
presuppose laws already in place. Lawgiving itself can aim either at the perfec-
tion of the body only or at the perfection of the body in service to the soul. The
latter sorts of laws are called divine laws, and their promulgators prophets. Since
the perfection of the soul is the perfection proper to a human being as such, the
purpose of prophecy is the founding of a political community whose aim is the
proper perfection of human beings. That is to say, the prophet must be a philoso-
pher, statesman, diviner, and magician combined in order to be the founder off
the perfect or ideal political community.
Is this implicit answer Maimonides’s own answer? Strauss corroborates it by
citing Shem Tov ibn Falaquera,40 who in turn cites Avicenna to that effect.41 Off
the three moments of prophecy besides philosophy (sc., statesmanship, divination,
magic), Avicenna ranks statesmanship highest. Accordingly, he treats prophecy
as part of politics. Strauss quotes Avicenna as saying in that context: “The first
purpose of the lawgiver in lawgiving is the division of the city into three parts:
leaders, artisans, and guardians.”42 Avicenna’s prophetic lawgiver thus aligns the
class divisions of his city with those of Plato’s Republic. Like Maimonides’s otherr
Muslim predecessors, he understands himself in general as a student of Plato and
Aristotle, as Strauss also shows.43 Yet at bottom Avicenna et al. take their bear-
ings by Plato rather than Aristotle. Strauss returns to the difference between
Plato and Aristotle to discover why.
S T R AU S S ON M A I MON I DE S A S A P L AT ON I S T 75

As has been said, Plato and Aristotle differ over whether philosophy consists
in asking about the Good (Plato) or in pure contemplation and understanding
(Aristotle). In Plato’s view, Strauss goes on to say, asking rightly about the Goodd
needs a preparation—a detour consisting of such questions as: What is the soul?
What are its parts? What is science (or knowledge)? What is being? What is the
One? etc. These are the same questions Aristotle asks apart from questions about
the Good. But this means that contemplation and understanding as set forth
in Aristotle’s treatises are already worked out in Plato’s dialogues. If so, Plato
does not disagree with Aristotle that human happiness means pure contempla-
tion and understanding, or that these are a human being’s highest possibility.44
They disagree in how they approach this possibility. Whereas Aristotle sets
or leaves philosophers free of political obligations, Plato forces them to live
up to those obligations.45 For Plato, philosophy is not absolutely sovereign. In
Strauss’s words: “The [Platonic-Socratic] philosopher who has elevated himselff
in the beholding of the truth about the beautiful, just, and good in purity, lives
and wants to live in it, is bound back to the state by the harsh command of the
lawgiver, which considers the order of the whole and not the happiness of the
parts.”46
Now Strauss points out that what Plato requires—philosophy’s standing
under the law—is given in the age of the revealed religions. Accordingly,
philosophers who live in that age must defer to the law—by seeking legal
authorization to philosophize through deriving philosophy as a legal obligation
and by keeping under wraps (or secret) the freedom they take in interpreting
the law in that way (that is to say, exoterically). Yet because the law the philoso-
phers defer to is already given through a prophet, they no longer need to make
it the main theme of their philosophizing, as does Plato. Like Aristotle, they
can ask metaphysical questions apart from, and at more length than, the ques-
tion of the just order of human life. They still need to justify and understandd
the law, however; and when they do, it is within the horizon established by
Plato. Here Strauss recalls Avicenna’s identifying the prophetic lawgiver with
the founder of the Platonic state. “But this,” Strauss infers, “means that revela-
tion, the Law, is understood in the light of Plato.”47

Conclusion
“All honor to the God of Aristotle; but truly he is not the God of Israel,” says
Cohen pivotally in interpreting Maimonides.48 Strauss agrees with Cohen’s piv-
otal statement, though not entirely in Cohen’s sense. Interpreting Maimonides
as a Platonist, Cohen nevertheless criticizes him for appropriating Aristotle’s
theology while underestimating how Aristotle’s downgrading of ethics endan-
gers philosophy in what Cohen takes to be Plato’s meaning, namely, philosophy
for the sake of ethical action.49 At the same time, Cohen adds that Maimonides
could overlook this danger more easily since he saw the privileging of ethics
safeguarded by the Torah. Strauss notes that in calling Maimonides a Platonist,
Cohen is on to something. Even so, Strauss finds Cohen’s “idealizing” interpre-
tation misleading in its gratuitously substituting ethics for law as understood by
Maimonides and his predecessors.50
76 M A RT I N D. YA F F E

Notes
1. Hermann Cohen, “Charakteristik der Ethik Maimunis,” in Moses ben Maimon: Sein
Leben, seine Werke und sein Einfluß, ed. W. Bacher, M. Brann, and D. Simonsen
(2 vols. in 1; Leipzig: Gustav Fock, 1908; reprint, Hildesheim and New York:
Olms, 1971), I, 63–134. Unless otherwise indicated, page numbers to Cohen’s
essay refer to this volume. Cf. Cohen, Ethics of Maimonides, trans. Almut
Bruckstein (Madison, WI: University of Wisconsin Press, 2002). No attempt has
been made to harmonize my renderings of Cohenian passages with Bruckstein’s
considerably freer renderings. “Charakteristik der Ethik Maimunis” is reprinted in
Cohen, Jüdische Schriften, ed. Bruno Strauss, intro. Franz Rosenzweig (3 vols.;
Berlin: C.A. Schwetschke & Sohn, 1924), III, 221–89; also in Cohen, Kleineree
Schriften IV, 1907–1912, ed. Hartwig Wiedebach (Hildesheim and New York:
Olms, 2009), 161–269.
2. CuM 393–436; CaM 173–222.
3. Strauss, “Introductory Essay for Hermann Cohen, Religion of Reason out of thee
Sources of Judaism,” LAM M 233; originally in Cohen, Religion of Reason out of thee
Sources of Judaism, trans. Simon Kaplan, intro. Leo Strauss (New York: Ungar,
1972), xxiii.
4. See, for example, Julius Guttmann, Die Philosophie des Judentums (München:
Verlag Ernst Reinhardt, 1933), 174–205; Philosophies of Judaism, trans. David W.
Silverman (New York: Holt, Rinehart and Winston, 1964), 172–207; Guttmann
concludes his account of Maimonides by speaking matter-of-factly about
Maimonides’s “theistische Aristotelianismus [theistic Aristotelianism],” 205; trans.
206–7. See also Isaac Husik, A History of Medieval Jewish Philosophyy (New York:
Macmillan, 1916), 216–311; Husik speaks of Maimonides’s “evident purpose . . . to
harmonize Judaism and philosophy, to reconcile the Bible and Talmud with
Aristotle,” 236.
5. See Hermann Cohen, Religion der Vernunft aus den Quellen des Judentums (2nd
ed.; Frankfurt am Main: J. Kauffmann, 1929), 78, 89–90, 94, 172, 303–4, 307;
cf. Cohen, Religion of Reason, 68, 77, 81, 148, 260, 263. Cf. also Cohen, “Diee
Bedeutung des Judentums für den religiösen Fortschritt der Menschheitt [The Significance
of Judaism for the Religious Progress of Humanity],” Jüdische Schriften, I, 18–19;
reprinted as “Die Bedeutung des Judentums für den religiösen Fortschritt,” Kleineree
Schriften IV,
V 429; “The Significance of Judaism for the Progress of Religion,”
ibid., 456.
6. Cf. Strauss, Die Religikonskritik Spinozas, in GS–1 197; SCR R 148.
7. The two Hebrew expressions mean, respectively, “The Account of the Beginning”
(Genesis 1) and “The Account of the Chariot” (Ezekiel 1 and 10).
8. See Moses Maimonides, Guide of the Perplexedd Part I, Introduction (trans. Shlomo
Pines, intro. Leo Strauss [Chicago: University of Chicago Press, 1963]), 6; hence-
forth Guidee I, Introduction (Pines 6).
9. Cf. Strauss, Die Religikonskritik Spinozas, in GS–1 321–25; SCR R 258–62.
10. Here Cohen emulates his own philosophical authority Immanuel Kant, who
claims to be able to understand Plato better than he understood himself. See
Critique of Pure Reason A314 = B370.
11. See Guidee III 32 (Pines 529–31).
12. CuM 401–2; CaM 184–86. Here Strauss quotes Religion der Vernunft, 44; cf. Religion
of Reason, 38 (translation modified, M.Y.): “Against every routine approach, the
insight must prevail that progress in religious knowledge has been accomplished
S T R AU S S ON M A I MON I DE S A S A P L AT ON I S T 77

through the revision and reinterpretation of the sources, while these themselves
remain preserved in their individual layers and have been at most rearranged orr
given different emphasis.” This sentence of Cohen’s serves to answer the (rhe-
torical) question of his which Strauss quotes immediately beforehand (CuM 401;
CaM 185–86): “And it is a question whether such reshaping is not the best form
of annihilation [of the mythical].” Religion der Vernunft, 204; cf. Religion of Reason,
175. The translation of the latter sentence is Strauss’s own; see SCR R 24–25 orr
LAM M 250.
13. “Charakteristik der Ethik Maimunis,” 70–72, 82–84 (cf. Bruckstein, 14–19, 51–55).
CuM 403–405; CaM 188–90.
14. In Cohen’s words, which Strauss quotes: “Ethics as the doctrine of man becomes
the center of philosophy.” (CuM 413; CaM 201) See “Charakteristik der Ethik
Maimunis,” 63–64 (cf. Bruckstein, 2).
15. In Cohen’s words, which Strauss also quotes, these show up as “ever new ques-
tions, [with] every new solution only raising new questions” (CuM 413–14; CaM
201). See “Charakteristik der Ethik Maimunis,” 70 (cf. Bruckstein, 15).
16. In Cohen’s words (commenting on Plato’s Crito), which Strauss quotes as well:
“There is no self-consciousness that is to be acquired without regard for the state
and without being guided by the idea of the state” (CuM 414; CaM 201). Strauss
quotes Cohen, Ethik des reinen Willens (Berlin: Bruno Cassirer, 1904), 250; refer-
ences to the Ethik in the text of Strauss’s lecture are to this edition, whose pagina-
tion differs slightly from the 2nd, revised edition and its subsequent reprints (2nd
ed., Berlin: Bruno Cassirer, 1907; cf. 7th ed., Hildesheim and New York: Olms,
2008)—although the changes in the reprints have to do with the addition of an
editorial apparatus, etc., only; the corresponding page number in the revised edi-
tions is 265.
17. CuM 414; CaM 202. Strauss quotes “Charakteristik der Ethik Maimunis,”
70 (cf. Bruckstein, 15).
18. Nicomachean Ethics 1103b7f.
19. CuM 414; CaM 202. Strauss quotes “Charakteristik der Ethik Maimunis,”
72 (cf. Bruckstein, 17).
20. Nicomachean Ethics 1141a20ff.
21. CuM 414; CaM 202.
22. CuM 416; CaM 204–5.
23. Moses Maimonides, Sefer HaMada‘/ ‘ The Book of Knowledge, ed. and trans. Moses
Hayamson ( Jerusalem: Boys Town, 1963), 35b.
24. CuM 416, with 408, 421, 422; CaM 205, with 195, 211, 212. The emphasis is
Strauss’s.
25. See Guidee I 52, 56 (Pines 117–18, 131).
26. See Guidee I 53, 58 (Pines 122, 136).
27. CuM 418f.; CaM 208. See Guidee I 59 (Pines 142) with I 58 (Pines 136).
28. CuM 419; CaM 208. See Guidee I 58 (Pines 135f.).
29. See Cohen on Exodus 34:6–7: Religion der Vernunft, 109; Religion of Reason,
94. Cf. Cohen on Micah 6:8: Religion der Vernunft, 39; Religion of Reason, 33.
30. CuM 419; CaM 209. Quoting Exodus 33:19, Strauss cites Guidee I 54 (Pines 123ff.)
but also calls attention to Guidee 3.54 (Pines 635ff.).
31. “Charakteristik der Ethik Maimunis,” 71 (cf. Bruckstein, 17).
32. CuM 420; CaM 210. Strauss cites Guidee III 13 (Pines 452ff.).
33. CuM 420; CaM 210f. Strauss cites Guidee I 54 (Pines 123ff.), with Aristotle,
Nicomachean Ethics X 7, 1177a34–b1.
78 M A RT I N D. YA F F E

34. Nicomachean
h Ethics
h 1177a32ff.
35. CuM 421; CaM 212. Strauss cites Guidee II 37 (Pines 373–74), with II 11 (Pines
275). The emphasis is Strauss’s.
36. CuM 422; CaM 212. Throughout his lecture, Strauss calls Maimonides by his
traditional Hebrew acronym, RMbM (for “Rabbi Moshe ben Maimon”; vocal-
ized as “Rambam”).
37. CuM 422; CaM 212–13. Strauss cites Guidee II 38 (Pines 377–78).
38. In contrast, Abraham is prophetically inspired in the highest degree but only
instructs or teaches; and the post-Mosaic prophets only exhort their addresses
to obedience to the Law given by Moses. CuM, 423; CaM 213–14. Strauss cites
Guidee II 39 (Pines 378f.);
39. CuM 423–24; CaM 214–15. Strauss cites Guidee 2.40, III 27 (Pines 382,
511–12.).
40. CuM 424; CaM 216. Strauss cites Shem Tov ben Joseph ibn Falaquera on Guidee
II 40, in Moreh ha-Moreh, ed. Yair Shiffman ( Jerusalem: World Union for Jewish
Studies, 2001), 289–90.
41. Avicenna, Healing: Metaphysics X, trans. Michael Marmura, in MPP–1 99–111 orr
MPP–2 78–88; or Avicenna, The Metaphysics off ‘The Healing’, A Parallel English-
Arabic Text = al-Ilahīy ī āt min al-Shifā
f ’, trans. Michael Marmura (Provo, UT:
Brigham Young University Press, 2005), 364–68. For a fuller discussion, see PG G
97–108 or GS–2 98–109; PLA 111–20.
42. CuM 425; CaM 217. See MPP–1 104 or MPP–2 84; The Metaphysics of ‘Thee
Healing’, 370.
43. CuM 424–26; CaM 216–18.
44. That is to say, in Strauss’s paraphrase of Plato, Platonic philosophers live in the
belief that they have already been relocated during life to the Islands of the Blessed
(Republicc 519c). CuM 426; CaM 218.
45. Plato “compels” them “to care for others and to guard them” (Republicc 519d–520a).
CuM 426; CaM 219. The emphasis is Strauss’s.
46. CuM 426; CaM 219. The emphases are Strauss’s.
47. CuM 427; CaM 220. The emphasis is Strauss’s. Strauss traces the provenance off
the merging of philosophy, statesmanship, divination, and magic in the prophet-
ology of Maimonides and his Muslim philosophers through, for example, Cicero’s
De divinationee I.40.89, though he adds (while acknowledging that more needs to
be said) that its ultimate provenance is Platonic. See CuM 428; CaM 220. For his
somewhat fuller discussion, see PG G 117–22 or GS–2 118–23; PLA 127–33.
48. CuM 416, 421, 428; CaM 204, 211, 221. See “Charakteristik der Ethik Maimunis,”
81 (cf. Bruckstein, 50).
49. CuM 427; CaM 219. Strauss quotes “Charakteristik der Ethik Maimunis,” 87
(cf. Bruckstein, 64): “[Maimonides] underestimated the danger residing in the
depreciation of ethics in Aristotle. And from his standpoint he could overlook
this danger more easily, since he saw the value of ethics kept safe in his religion.”
The emphasis is Strauss’s. I have used the translation of these sentences found
in PLA 133.
50. CuM 428–29; CaM 221–22.
C APT R 5
CHAPTER

STRAUSS ON THE RELIGIOUS AND INTELLECTUAL


SITUATION OF THE PRESENT

Timothy W. Burns

T his chapter focuses on two talks of startling freshness given by Leo Strauss
in 1930 and 1932, translated into English for the first time by Anna Schmitt
and Martin D. Yaffe.1 Addressed to a Jewish audience, the talks were given in
the most politically unsettled years for Germany, and the richest, intellectually,
for German Jews. Among the topics they address is “[t]he Jewish problem, whose
urgency in the age of National Socialism perhaps does not need to be proved
[any]more to anyone.” Since the talks address other problems of both “moder-
nity and ultramodernity,” or what today goes by the name postmodernism, theirr
questions are still with us, in sometimes sharpened, sometimes degraded, forms.
The talks offer guidance for those who wish to liberate themselves from the crip-
pling presuppositions through which Strauss himself had—as we see here for the
first time—fought his way to clarity.
To better grasp the talks’ significance, it may be helpful before examining them
to note an often-overlooked distinction between what Strauss, in the openingg
paragraph of the autobiographical preface to Spinoza’s Critique of Religion, calls
the “theologico-political predicament,” on one hand, and the study of the “theo-
logico-political problem” that he later identifies as the theme of his life’s work,2
on the other. The former was a narrower problem for German Jews, one that is
bound up with historicism. As Strauss explains, the German thought to which
German Jews opened themselves in the nineteenth century “was understood to
be German essentially.” It was in this respect unlike the Greek thought to which
Maimonides and others had opened themselves in the period of Spanish Jewry,
which “was understood to be Greek only accidentally.” As a result, German
Jews had not only a political dependence on Germans (the Weimar Republic)
but a “spiritual dependence.” For even the most cosmopolitan of German think-
ers attempted to understand their nation’s thought through its “culture,” which
was Christian and hence excluded Jews. 3 This was the “predicament” faced by
80 T I M O T H Y W. B U R N S

German Jews who had hoped, on the basis of liberalism, that the difference
between Jews and non-Jews would become a matter of indifference, or who
thought the state, at least, should remain neutral on that difference. That hope
proved to be misguided. Strauss’s attempt, as a morally serious young man, to
solve his predicament had led him to political Zionism, on one hand, and to the
theology of Karl Barth and Franz Rosenzweig, on the other.4 Zionism had arisen
as an attempt to solve the Jewish predicament politically by establishing a liberal,
secular state for Jews. But this, even when combined with cultural Zionism—as
eventually became the case—meant in truth a rejection of the Jewish tradition,
that is, of belief in creation, miracles, and the divine law; it meant in truth a
rejection of Judaism. However necessary as a political response to the abiding
rejection of the Jew as Jew within the modern liberal state, it could provide no
answer to the spiritual or moral longings of serious Jews.
As for Rosenzweig’s theology, while Strauss respected the attempt of this
friend and colleague5 to renew Jewish theology for the modern Jew, the two talks
before us indicate the ground of his deep dissatisfaction with it. That ground is
linked to his recovery of the thought to which Maimonides and Spanish Jewryy
had opened themselves—that is, the recovery of genuine Socratic-Platonic phi-
losophy—and the uncovering of the roots of its modern alternative, initiated by
Spinoza, Hobbes, and Descartes. It is linked to the recovery of the permanent
and universal theological-political problem. Awareness of that permanent prob-
lem, and of the two major philosophic attempts at solving it, points the way to a
means for the individual Jew to solve his contemporary “predicament.” In these
talks, we see for the first time Strauss’s recovery of the thought of Maimonides
and Plato and hence of the theological-political problem.
That recovery entails likewise the rejection of or liberation from the thought
of another representative of the “new thinking.” As Strauss indicates, these talks
take place against the rise, both meteoric and serious, of Heidegger.6 They are,
to be sure, explicitly directed against the historicism of the sociologist Karl
Mannheim, who failed indeed to face the great problem that Heidegger and his
better students did face. But in their relentless attack on the historicist doctrine
that all thinking is determined by the present—by a thinker’s time—the talks
move inevitably if quietly against the radical historicism of Heidegger. They
show us the vital concerns that had been uncovered by Heidegger and his guide,
Nietzsche, and the recovery, which they inadvertently made possible, of genuine
philosophizing—of philosophizing that begins by naively asking fundamental
questions in search of eternally true answers. To this end, they provide an initial
sketch of a nonhistoricist account of the historicism that emerged out of early
modern philosophy, one that brings historicism into focus as a “cave beneath the
cave.” They provide an account of modern philosophy that begins (in Strauss’s
case) with historicist presuppositions and that ends up turning them, like a bal-
loon, inside-out.
In his initial effort to strip down to its serious intention the assigned topic off
his first talk, “Religious Situation of the Present” (1930), Strauss recasts the topic
as “The Intellectual Situation of the Present” and then “The Question of the
Present,”
ese t, and,
a d, finally,
a y, “What
W at iss the
t e right
g t life?
e? How
ow should
s ou d I live?”
ve? Thee second
seco d
S T R AU S S ON R E L IG IOU S A N D I N T E L L E C T UA L S I T UAT ION 81

talk, “The Intellectual Situation of the Present” (1932), is then, even in its title,
a reworked version of the 1930 talk. Moreover, the second talk has, as it seems
to me, the rhetorical intention of stressing, on one hand, the original and everr
reerupting intention of the moderns, that is, the return to Greek philosophy overr
and against stale scholasticism, or Greek philosophy minus the biblical tradition,
and on the other, the dead end that historicist thinking holds for the question off
the right way of life, and, on the basis of that finding, the possibility of beginning
again that was opened up by Nietzsche. The recovery of Platonic philosophizingg
that makes both of these arguments possible is found in the first talk. I will there-
fore focus my attention on the first talk, as the archee of Strauss’s new orientation,
using the second talk to clarify some of its points, and will bring to bear upon it
earlier, contemporaneous, and later writings of Strauss, especially his correspon-
dence with Karl Löwith.

* * *

In his opening remarks, Strauss distinguishes between the many contemporary


writers on religion and the few who have “thought thoroughly,” and among the
latter he further distinguishes between commentators and the very few genu-
inely authoritative thinkers. Only these very few, he asserts, are worth address-
ing. And of more interest to “us Jews” among them is “Franz Rosenzweig,” off
whose teaching Strauss promises to give an exposition, after a preliminary “clari-
fication of the topic.” In the talk that has come down to us Strauss never arrives
at “Rosenzweig’s doctrine and its difficulty.” The talk is therefore wholly given
over to the preliminary clarification.
The clarification proceeds in six steps: 1) a reduction of the assigned ques-
tion (the religious situation of the present) to its serious purport, viz., “what is
the right way of life?” 7 2) the summoning of a goddess, “the Present,” to warn
against any naive pursuit of an answer to this serious question that is simply true
or transcends the present; 3) an account of the difficulties with the attempt to find
nontranscendent but synthetic answers in the present, that is, the difficulties with
“conspectivism”; 4) Over and against the historicist understanding of the mul-
tiplicity of answers to fundamental questions, the original, Socraticc explanation;
5) an account of the additional difficulty to philosophizing that arose through
the authority of biblical revelation, as the first “cave beneath the cave”; and 6) a
sketch of the unsuccessful modern attempt to eliminate this additional difficulty,
resulting in a new authority, a new “cave beneath the cave,” that is, historicism.
The clear theme of the talk is one of wresting oneself free of the present as from
a sub-cave that is mistaken as the true or authentic human situation.
This theme helps to explain what might otherwise appear as a somewhat petu-
lant initial tearing away of the onion layers of the assigned topic in order to reach
the serious question that lies beneath it. Strauss conveys in this initial section the
impression of one who, having liberated himself from the present sub-cave, is
being asked to give a description of its walls and has thus been assigned what is,
he bluntly asserts, “not a serious topic.” The remainder of the talk is intended to
justify this assertion, which in other circumstances could appear graceless but
82 T I M O T H Y W. B U R N S

which is here intended to free morally serious Jewish youth from the “tyrannical
conviction” under which they have posed their topic.
Strauss makes clear that he knows his listeners will be taken aback by this
opening, and at the same time he plants the beginning of doubt about their posi-
tion, by conjuring up a beautifully attired, knowing, powerful goddess, “the
Present,” who warns them against the naive effort to pose the question of the
right life. This is perhaps the most poetic rhetorical device that Strauss everr
allowed himself, and echoes Socrates (e.g., the Laws in the Crito) and Nietzsche
(e.g., the Free Spirit in the Genealogy of Morals). The goddess proclaims that each
age has its truth and, proudly possessed of this knowledge, she laughs at the
naiveté of those who in the past sought a permanent answer to the question off
the right way of life. She is told by her understanding (in a brief dialogue with
it) that her virtue or duty to be honest with herself does not permit her to fight
a direct assault on the quest for truth. Her laughter at naiveté is, she declares,
“no naïve laughter” but “reflective” and benevolent: she exculpates earlier gen-
erations, who sought the truth and who, finding in fact only the truth of theirr
present, of their age, did against their will what she now prescribes to her children,
the children of the present. Enthroned high above the past, she does not sternly
forbid the search for eternal truth, but instead charmingly appeals against it to
her listeners’ sense of what is fitting or becoming of them as rational beings: It is
now rational to want only the truth of the present.
The summoned goddess’s claim to knowledge rests, then, on her reflective,
backward glance at the past, and her moral admonition rests on the claim that
man is rational; her appeal is found, Strauss declares, in the mouths of “the most
agile, most progressive, most expert, most lively children of our time.” Through
their mouths she appeals daily and everywhere to what is trendy over and against
what stodgy old goats might say. Strauss is aware that what he will be sayingg
may be mistaken as the position of a stodgy old goat. He therefore takes up her
admonition. But where, he asks, does one encounter “the present” to which she
directs one for guidance?
It is allegedly to be found in the competing groups, parties, etc., that are
engaged in a struggle to lead the present: to be led by the present means to with-
holdd devotion to any one group and instead to synthesize the various competing
groups into a coherent totality. Strauss is here speaking of Karl Mannheim’s “con-
spectivism” (or “relationalism”), of which he had written a critique the previous
year.8 He repeats now the most salient features of that critique: 1) Conspectivism’s
mere reporting of philosophic positions stands in striking contrast to the full, dif-
ficult, long, “immense” effort of a Kant, a Marx, or a Nietzsche that is required
to arrive at their respective positions; it offers Reader’s Digestt versions of them,
the very opposite of the genuine search for understanding.9 2) Conspectivism’s
origin is in a superficial, lazy reading of texts, an openness to what appears to be
novel in those texts, and a desire to write about it, to reduce conflicting theories
to “keywords” and their connections; the result is inevitably a jargon-laden,
sloppily conceived, amusingly superficial synthesis. 3) Conspectivism displays a
“sham understanding,” a “stale and cowardly mishmash,” a failure to take “seri-
ously the
t e great
g eat men
e who
w o dominate
do ate the
t e present.”
p ese t.
S T R AU S S ON R E L IG IOU S A N D I N T E L L E C T UA L S I T UAT ION 83

In making this third and final criticism of Mannheim’s conspectivism, Strauss


refers at length to the work of “a philosopher in the full sense of the word.
Completely unknown just five years ago, today his name and work are talked off
everywhere.” He refers, of course, to Heidegger:

This philosopher in his main work has, among other things, written a few pages
about idle talk, what it means and what it inf licts. That was meant as a, so to
speak, purely factual statement, not as the author’s appeal to spare him from idle
talk. What happened? A woman—the noble word “lady” obviously forbids itself—
reads this philosopher, and before she can even have the slightest idea of what the
man really means, she goes around in London and yaks and yaks. She found the
paragraph on idle talk certainly “very fine”; she has understood him in this sense;
but she did not understand him in such a way that she would finally, finally shut
her unbearable trap. (RSP)

Strauss himself had, eight years earlier, seen and been deeply impressed by
Heidegger’s “precision, and probing, and competence . . . seriousness, profun-
dity, and concentration in the interpretation of philosophic texts.”10 That he
refers here to Heidegger without naming him stands in contrast to those many
who were now doing so.11 That he refers to him only here should not blind
us to the deep impact that Heidegger’s thinking had on him. As the passage
we have quoted makes clear, Heidegger’s philosophic thought represents in a
sense thee alternative to “conspectivism,” and of any attempt by social scientists
like Mannheim to claim an authoritative position on the right way of life. At
the same time, as we will see below, Heidegger represents a deepening of the
problematic historicism that lies behind Mannheim’s conspectivism. Heidegger,
not Mannheim, is the serious target of this talk. Before spelling out how this is
so, we must first examine its second half, in which Strauss indicates that he has
found his way out of historicism. Indebted to Heidegger, Strauss will now move
in a direction that Heidegger might have taken but did not take, that is, to an
investigation that leads out of historicism. How, exactly, has Strauss freed himselff
from it? How does he defend the activity of asking naive questions, questions
about the right way of life, from the claims of historicism and its specific “radical
reflectiveness”?

* * *

Here as elsewhere, Strauss attempts to understand fully and sympathetically the


historicists and the young men who are vaguely dissatisfied with the present.12
The very attempt of Mannheim’s conspectivism, and the difficulties with that
attempt, begin to show the way. In seeking a synthesis of present positions that
would yield an intelligible whole or totality of them, conspectivism asserts the
equivalence of all positions in the present, in order to avoid the necessary blind-
ness or partiality of any one position. In its search for guidance, it wishes to do
justice to all the positions, while the partisans of those positions cannot do jus-
tice to one another’s positions. In seeking to avoid the blindness of partisanship,
84 T I M O T H Y W. B U R N S

it aims to know or see the truth. Yet this motive that has come into sight as
prompting the search—the desire to know the truth about the present—is ill
served by treating all positions as equivalent, since some of the positions may be
simply untrue and hence not worthy of being made part of the intended synthe-
sis. More importantly, assuming that a synthesis of them all is possible, one may
still be left without the truth, since it is possible that “alll present standpoints rest
on a mistaking of the fundamental facts,” or alll present standpoints are perhaps
“ideologies.” To know whether or not this is so, to know the present as it truly
is, one must “first of all be free of the present. This freedom does not fall into
our laps; we must win it for ourselves.” The motive to know the truth about the
present impels the thoughtful person to seek an ascent out of the present.
This first step away from the present still assumes that knowledge of the pres-
ent is the goal, the vitally desired knowledge. But this first step allows one to call
into question this goal. For having taken a step out of the present, one sees that
while there has always been a “present,” an era, in which human beings lived,
it was not something that serious people felt a compelling need to be concerned
with as the source of guidance. As Strauss puts it in the second talk, “[t]houghtful
people concerned themselves with the eternal.” So too with us: the present is ourr
fate, but “our fatee is nott our task.” Strauss extends this point beyond Mannheim’s
conspectivism, to encompass “today’s man” generally: “This is the principal mis-
take to which today’s man keeps succumbing: the attempt to determine the task
from the fate. This attempt is absurd if there is no God: then fate is chance, and iff
God is, then fate is providence, and we are not allowed to want to play God.”13 Iff
our particular present is the product of mere chance, then it has no serious claim
to direct our lives; if it is on the other hand the result of God’s providence, then
we must bow to that God and seek his guidance rather than being guided by a
willing of the present. Making one’s fate one’s task falls, we may say, somewhere
in-between these two, in a confused, half-conscious attempt to find a sacredness
in the blind historical development that has produced the contemporary absence
of the sacred.
Having expressed this problem with historicism more broadly—and, inciden-
tally, brought out its theological intention—Strauss applies it to Mannheim in
particular: “This error manifests itself also in the will to synthesis: even if each
standpoint may bee a synthesis in fact—itt is nonetheless never willed as a synthesis;
what has been willed is always the truth. We have to look straight ahead; we neverr
come to know what we have to do by being reflective.” The “reflectiveness”
of historicism, its bearing in mind the historical experience of the failures off
past attempts to arrive at the truth, the historical experience of finding in those
attempts only the truth synthesized from particular epochs—all of this has now
mistakenly turned this result, the failures, into its goal.
But a return to a naive questioning, on whose necessity Strauss insists, overr
and against the sophistication of historical “reflectiveness,” is no simple thing. Iff
historicism is right in its interpretation of the multiplicity or anarchy of teachings
of the past, then escape from one’s time or “the present” is indeed impossible, and
it would make little sense to attempt vainly to transcend it. In bidding his listen-
e s, as hee again
ers, aga now ow does, to examine
e a e the
t e question
quest o “concerning
co ce g the
t e right
g t life”
e
S T R AU S S ON R E L IG IOU S A N D I N T E L L E C T UA L S I T UAT ION 85

as a question susceptible of an answer that is nott bound by one’s present, Strauss


must establish the possibility that historicism’s prescribed “reflectiveness,” on the
failure of past attempts to answer this question rests in truth on a dubious, dog-
matic interpretation of the failure in question. The historicists have, as he says,
been turning this failure of past attempts to understand the truth “into a theory,”
and asserting thereby “the inevitability of failure.” One must instead, he argues,
take seriously that failure as a warning about the possible grounds of failure, in
the effort to get at the truth. Examining those possible grounds is “truncate[d],”
however, by “the dogmatic assertion that there are no eternal truths.” A dog-
matic assertion of our present threatens to short-circuit the ascending outside off
it that has now begun.
But an investigation of such grounds of failure outside the historicist expla-
nation of those grounds presupposes that the “experience” of such failure is— —
contrary to the historicists—nott peculiar to the present: past thinkers also under-
went this “experience” of multiplicity of failure but explained it in a different,
nonhistoricist manner. Is this really the case? Strauss now argues that it indeedd
is. After posing the question “what about the historical experiencee of this failure?”
he presents Socrates as the first to have posed this question. And he presents a
sketch of the allegory of the cave as Plato’s answer to this question, stressing as
he does so the painful character of the ascent out of the cave, and that the one
who ascends would be, upon redescending, “completely incomprehensible and
laughable” to his companions down below. “Thus Plato presents the difficulties
of philosophizing, the naturall difficulties. If they are so extraordinary, is it any
wonder that there are so many conflicting opinions?” This alternative Socratic
or Platonic explanation of the multiplicity or “anarchy” of opinions is the means
to transcend the reigning historicist interpretation of that multiplicity, the means
to overcoming the dogmatic fatalism of the historicists that would short-circuit
the escape out of the present.14
Yet serious historicists deny that there is a “nature” to man such that one
can speak of “natural difficulties.” The “natural world” or “nature,” so far from
being understood by a Socrates or Plato, was (according to historicists) merely
the Greeks’ project, a “construction of the human understanding,” as Strauss will
put it two pages hence. In their articulation of Western man’s historical exis-
tence, historicists point in particular to the transformation of Western man that
occurred by dint of the biblical doctrine of the revealed, creator God.15 Has the
doctrine of creation not, they ask, deepened man by disclosing among otherr
things the contingency of all that exists, or made doubtful the alleged natural or
necessary limits to all possible change?
Strauss confronts this claim in a remarkable fashion. The “natural” difficulties
of philosophizing set forth by Plato—difficulties of “man as man, as a sensual-
intellectual being”—are indeed no longer the only difficulties. The natural dif-
ficulties were reiterated by Alexander of Aphrodisias and quoted by Maimonides
in Guide of the Perplexed16—a clear case of transcendence of time and place in
awareness of difficulties—but more importantly, to Alexander’s list of three
natural difficulties Maimonides explicitly added a fourth: the unconditionally
aut o tat ve, revealed
authoritative, evea ed writings
w t gs ofo Scripture,
Sc ptu e, which
w c had ad diminished
d s ed the
t e freedom
eedo
86 T I M O T H Y W. B U R N S

to philosophize. Far from being changed or determined in his understanding by


the biblical doctrine of creation, Maimonides now comes to sight as in agree-
ment with Plato and Alexander. He found or presented the revealed writings off
the Bible nott as offering a radically new way of thinking or existing, in a hitherto
unknown awareness of the contingency of all existence, but instead as an his-
torically added impediment to philosophizing. Philosophizing was no different and
no less possible in principle than it had ever been; it was merely impeded by an
additional difficulty.
In this presentation of Maimonides’s assertion of a difficulty to philosophizingg
added since the time of Plato, we see clearly for the first time the effect of Strauss’s
rediscovery of both Plato’s importance for medieval Arabic and Jewish thinkers,
and of esoteric writing. As late as 1928, Strauss had not understood Maimonides’s
professed belief in creation and in the Mosaic Law as exoteric.17 We see here the
first evidence that he now does so. And as Strauss makes explicit in his use off
the same Maimonidean passage two years later, he takes Maimonides’s surface
statement as requiring the reader’s own reflection or puzzlement in order to grasp
its radical import. Maimonides himself speaks, after all, of the fourth, added dif-
ficulty to philosophizing as consisting merely in a habituation and schooling in
writings that indicate God’s corporeality. Strauss remarks:

Now surely there were many Greek writings in which the gods were presented
corporeally. Why did these writings not compromise Greek philosophy? It is
therefore not being accustomed to writings in general, not having grown up in a
tradition in general, but rather being accustomed to very specificc writings, having
grown up in a tradition of a very specificc character: namely, in a tradition possessing
an authority, as unconditionall as that of the tradition of revealed religions. The fact that a
tradition based on revelation entered the world of philosophy increased the natural
difficulties of philosophizing by adding the historicall difficulty.
In other words: The natural difficulties of philosophizing have their classical
depiction in Plato’s allegory of the cave. The historical difficulty may be illustrated
by saying: there now w exists another cave beneath the Platonic cave. (ISP)

Strauss’s rediscovery of the distinction between the esoteric and the exoteric, and
of the significance not only of Aristotle but also of Plato for Maimonides, now w
comes to the fore in a striking presentation (in both talks) of revealed religion
as creating, in Maimonides’s view, a “cave beneath the cave,” something out off
which to ascend to the right life, the philosophic life.
As we know from a much later talk, this rediscovery occurred in conjunc-
tion with Jacob Klein’s study of Plato, at the time that Strauss was studying
Maimonides and Hobbes, just as he was finishing his Spinoza book. Assisted by
Klein’s rediscovery of the importance of the dramatic character of the Platonic
dialogues—even of sections that appear to be philosophic treatises18—Strauss
was able now to grasp with two hands the significance of Avicenna’s statement,
in On the Division of the Rational Sciences, that “the treatment of prophecy and the
Divine law is contained in the Laws [of Plato].” (Strauss’s first public reference to
this passage appears in “Cohen und Maimuni,” May 4, 1931).19 He could now see
S T R AU S S ON R E L IG IOU S A N D I N T E L L E C T UA L S I T UAT ION 87

that the medievals considered Plato to have understood and settled the challenge
to philosophy posed by biblical prophecy or revelation. The momentousness
of this rediscovery of esotericism in the works of Maimonides and of philoso-
phers in Islamic countries, and thereby of the significance of Plato, 20 cannot be
overestimated. The multiplicity or anarchy of philosophic teachings, which had
appeared to Strauss and his contemporaries to be evidence of the insurmount-
able, unpredictable blindness of one’s time and place to certain truths and off
unpredictable openness to other truths, began now to appear to him instead,
in the works of the greatest thinkers, as rhetorical accommodations to human
weakness and to historical manifestations of that weakness. The situation created d
by historicism would eventually compel him to put more public stress on this
rediscovery of esotericism than would have been the case otherwise, 21 but the
rediscovery is already assisting Strauss in the recovery of the naive, natural ques-
tion of the right way of life.
Now the fundamental teaching of historicism, as Strauss stresses in his talk
two years later, entails a claim to final knowledge of a truth whose content may
be lost to future generations only at the price of a rebarbarization. That teachingg
is that awareness of the radical contingency of all beings, owing to the funda-
mental thrown-ness of human existence, permits us, in our present, to know w that
all thought is historical. Historicism claims for itself this knowledge, available
to no previous age, as a central truth allegedly revealed through the historical
process. Proponents of historical consciousness argue that

historical consciousness will go away if humanity unlearns what it has learned


arduously enough over the past centuries; the renunciation of historical conscious-
ness is identical with the relapse into a stage of lesser reflection.
Let us pause here for a moment. Historical consciousness is—one cannot
emphasize this strongly enough—according to its own view a stage of higher
awareness: we know moree than the earlier generations; we know more deeply,
more profoundly, than the earlier ones that everything human is historically con-
ditioned. (RSP 245–46)

The recovery of Platonic-Socratic philosophizing must, then, somehow answerr


this claim; it must show that, to the contrary, full awareness of the possibility
of the radical contingency of human existence was faced by Plato’s Socrates and
seriously answered—not just turned away from or neglected—by him.
The recovery of Platonic-Socratic philosophizing was a recovery of the sig-
nificance of dialectic, or of the Socratic “turn to speeches,” illustrated in the
allegory of the cave. We can better understand this recovery, as an answer to the
problem of contingency, by considering the opposing, historicist, or existentialist
position of Strauss’s friend and Heidegger’s student, Karl Löwith. As Löwith saw
things, Plato’s student Aristotle presupposedd that the universe is eternal; he was
therefore not at all open to the question of “existence,” of radical contingency;
he never raised the question of why something is at all, but instead took exis-
tence for granted, neglected it. “To Aristotle, existence meant an unquestionable
element within the essential structure, order, and beauty of a dependable and
88 T I M O T H Y W. B U R N S

clearly defined cosmos, which includes the existence of rational animals called
men.”22 Again, to Aristotle, “existence as such—that there is something—was
an unquestionable element within the essential structure, order, and beauty of an
always existing cosmos without beginning and end, including the existence off
rational animals called men.”23 Löwith went so far as to claim that “the Greeks”
regarded “the physical cosmos . . . as something divine.”24 He accordingly found
that the medieval doctrine of the creator God not only made the problem off
causality—of the possibility of the contingency of all beings—clearer than it
might otherwise later have been to the moderns, but that the medievals were
the first even to seee this problem, to open it up as a problem. “On the level off
Aristotle, one can inquire only into the existence of something already existing,”
whereas for Thomas Aquinas, “existence is an exceptional category, undefinable
by a ‘what’ or essence.”25 To Löwith, finally, the problem of contingency is fully
understood and faced only at the end of modernity, with Heidegger: “With the
loss of faith in this cosmos and with the loss of faith in Christianity’s creator God,
contingency in a thrown, meaningless world comes to the fore”; “If the universe
is neither eternal and divine (Aristotle), nor contingent but created (Augustine),
if man has no definite placee in the hierarchy of an eternal or created cosmos, then,
and only then, does man begin to ‘exist,’ ecstatically and historically.”26 Someone
living in our age might well desire, Löwith notes, to readopt the ancient view off
an “immutable natural order” or “divine creation,” but he can now do so onlyy
as a choice “between the one or the other ‘project’ or Weltentwurff ” and such a
choice now “would still be an existential attitude and decision.”27 And so, see-
ing Strauss arguing now for a serious return to the ancients, to “natural man,”
Löwith was dumbfounded. As he told Strauss,

[W]hatever one might say against progressive models of history, I do agree how-
ever with them inasmuch as I also find that Christianity fundamentally modified
ancient “naturalness.” With a cat or a dog “nature” does indeed always come out
again, but history is too deeply anchored in man for Rousseau or Nietzsche or your
future hero of natural being and understanding to succeed in restoring something
which already died out in late antiquity. . . . Only when you are able to convince
me that stars, heaven, sea and earth, generation, birth and death give you . . . natural
answers to your unnatural questions, will I be able to agree to your thesis.28

Strauss replied: “You confuse the Greek man-in-the-street, and as far as I am con-
cerned also the Greek poet, with the Greek philosopher. . . . Plato and Aristotle
never believed that ‘stars, heaven, sea, earth, generation, birth and death give’
them ‘natural answers to their unnatural questions’. . . . Plato ‘flees,’ as is well
known, from these ‘things’ (pragmata) into the logoi, because the pragmata give
no answer directly, but are mute riddles.”29 Dialectics is needed, we may say,
because the sensible properties that a being manifests depend for their existence
on a mind being there to perceive them, which means that something other than
the sensible properties must be there, or there is nothingg there without man. What
that something that is there is, however, is elusive, and there is nothing that guar-
antees its permanence; perhaps, as Hesiod indeed had claimed, it and everythingg
S T R AU S S ON R E L IG IOU S A N D I N T E L L E C T UA L S I T UAT ION 89

else that is came into being from nothing. And so the possibility does open up,
for the ancients no less than for medieval thinkers and for moderns influenced by
them, that being is by a creator god or gods, a possibility that is to be excluded,
insofar as it can be excluded, by the turn to the speeches, by dialectics. At stake
in dialectics, conducted through examination of moral opinions of a believer or a
potential believer, is the alleged evidence for the existence of gods who are said to
come out of, or to make beings come out of, nothing30—the moral experiences
connected to the experience of divinity. Telling in this regard is that Löwith,
on the basis of Emil Fackenheim’s research on Farabi, Algazel, Avicenna, and
Maimonides—research assisted by Strauss—finds that to the “Arab and Jewish
Aristoteleans of the tenth and twelfth centuries” existence “is a pure accident.”31
It does not for a moment occur to Löwith that Aristotle himself may have taken
this position quite seriously.
That Strauss was prepared to see in dialectics the Platonic approach to this
problem was due in part to his prior investigation of modern philosophy, which
had been underway since his doctoral thesis on Jacobi and had been enlivened
by a study of Hobbes with Julius Ebbinghaus;32 it had to this point born its full-
est fruit in what became Spinoza’s Critique of Religion (1928; published in 1930).
In preparing to write that first book, Strauss had accepted “the conditions off
the historical consciousness,” but in such a way as already to be leading in a
direction that would take him out of it. One who thinks within the conditions
of that consciousness, he had argued, has an advantage over both Spinoza andd
his chief antagonist, Calvin, neither of whom recognized the “conditionality”
of his cause, and both of whom therefore denied “every right to the opposite
position.” This shortcoming was more consequential in the case of Spinoza,
who was deprived thereby of a “true critique” of revelation; for Spinoza,
“every critical argument, any presupposition, . . . is preceded by the convic-
tion that one can judge the truths of religion from a theoretical stance outside
the domain of obedience and belief.”33 A true critique, Strauss had argued,
would be one that does not (as Spinoza’s had) deduce fear of God from profane
fear but would instead be one that knows (from experience) “what the fear off
God, and the stirrings of conscience entailed by it, actually is.” Openness to
this nonprofane fear in the religious believer as a genuine experience is some-
thing Strauss clearly thinks the “historical consciousness” had made possible
for him and even, one could say, obligatory to those within it who wished, as
he did, to assess whether the moderns’ victory over religious orthodoxy was
in every respect deserved. He elaborates helpfully on the lack of this openness
in Spinoza’s thought by contrasting Spinoza’s “critique of legalism” with that
of St. Paul:

In this context nothing is less justified than Spinoza’s appeal to Paul’s criticism
of legalism. In Paul, the deepest awareness of sin rebels against legalism, while
Spinoza’s rejection of the Law rests on the rejection of obedience as such, and rests
ultimately on the absence of any awareness of sin. Spinoza is a homo liberr and a homo
fortis, to whom radical stirrings of conscience and moments of ultimate despair
are unknown. He either did not know or deadened in himself the characteristic
90 T I M O T H Y W. B U R N S

experiences that are connected with the interest in revelation. In any case, he did
not subject them to critique.

Spinoza, in short, lacked “reflection on the conditions of the possibility of a cri-


tique of religion,” reflection that arises, to repeat, “under the conditions of the
historical consciousness.”34 Such “reflection” is what we see Strauss renouncingg
in the talks before us, in favor of naiveté or a return to our natural situation. But
it had had the great benefit of leading him to see the need for an internal cri-
tique of religion, a critique that took seriously the need to consider the believer’s
experience of sin, of a guilty conscience, if the rational life were to truly ground
itself. It had thereby prepared him to see that in dialectic, Socrates indeed takes
that experience seriously, and to criticize as late as 1928 Freud’s repetition off
Spinoza’s failure to do so.35
To put this somewhat differently, Strauss realized already—as we can see in
the phrase “the characteristic experiences [of conscience] that are connected with
the interest in revelation”—that (as he formulates it later), “the moral man as such
is the potential believer.”36 He was therefore prepared, as he might not otherwise
have been, for seeing—with the help of Avicenna—in the Platonic dialogues
precisely such an “internal critique” of faith in a creator god and of the possibility
of the contingency of all existence, of its coming to be out of nothing, that such
a god would inevitably entail, an internal critique that begins by seeing sin not as
“animal evil” but as “moral depravity.”37 He could see now, in other words, the
Platonic dialogues’ attempt to settle this matter through the “common ground”
of believer and philosopher. As he later says, the Socratic-Platonic philosopherr
assumes a “provisional acceptance of the accepted opinions” of the city, or off
“‘natural’ valuations”;38 he sees obedience to divine law or the problem of divine
law as “the common ground” between the Bible and Greek philosophy.”39 And
so the characteristic of Maimonides’s rationalism—that it begins by placing one-
self under the law, in obedience to the law—now appeared in this new, startling
light.40 Strauss now began to see that characteristic in the way that he subse-
quently described it:

We are again confronted with the question, Why philosophy? or, Why science?
This question was in the center of discussion in the beginnings of philosophy.
One may say that the Platonic dialogues serve no more obvious purpose than
precisely this one: to answer the question, Why philosophy? or, Why science? by
justifying philosophy or science before the tribunal of the city, the political com-
munity . . . [or] before the tribunal of the law.41

Strauss came to realize (as he had not, when writing the Spinoza book) that
Maimonides and Plato, no less than and perhaps more than Heidegger, had
indeed engaged in the kind of critique absent from Spinoza and the moderns
generally, and had done so through dialectic.42 With this, he directly experi-
enced ancient thinking that was both alive and nonhistoricist.43 Strauss therefore
asserts firmly now that the natural view is fully recoverable, here and now, as
S T R AU S S ON R E L IG IOU S A N D I N T E L L E C T UA L S I T UAT ION 91

much as it was for Maimonides despite (in his day) the authoritative writings
of the Bible. He presents “the hope . . . that some day humans will be able to be
natural again,” by dint of our “need to know,” breaking out in our “question
concerning what is right.” While he holds out no possibility of an ascent from
the cave as possible for the whole of humanity, he does hold out the possibility off
a recovery of a natural ignorance, that is, a recovery of the cave, from the “cave
of modernity” beneath it.44

* * *

But what, then, did Strauss find to be the substantive result of the Socratic “turn
to the speeches” that begins with the common ground of the moral experiences
of the believer or potential believer—a result altogether different from the char-
acteristically modern, historicist position? What is learned through dialectic can
be gleaned, Strauss suggests in this talk, through an awareness of the pains that
the dialectical ascent out of the cave entails, pains upon which he lays stress. Just
what those pains are, however, he does not tell his listeners—at least not directly.
From a later talk we know that, in contrast to Klein, Strauss found that the result
of dialectic was the realization of the insufficiency or problematic character off
the moral life, which becomes visible in a sufficient number of cases to the poten-
tial believer himself. As a result, “The tension between philosopher and the polis
is . . . the highest theme of political philosophy.”

I attached much greater importance than Klein did and does to the tension between
philosophy and the city, even the best city. . . . Philosophy is as such transpolitical,
transreligious, and transmoral, but the city is and ought to be moral and reli-
gious. . . . To illustrate this point, moral man, merely moral man, the kaloskagathos
in the common meaning of the term, is not simply closer to the philosopher than
a man of the dubious morality of Alcibiades. . . .
This view of philosophy was derived from my study of premodern philosophy.
It implies that modern philosophy has a radically different character. . . . In modern
times the gulf between philosophy and the city was bridged, or believed to have
been bridged. . . . If we call moralism the view that morality or moral virtue is the
highest, I am doubtful if it occurs in antiquity at all.45

A crucial part of what he learned at this time from his reading of Plato was “con-
firmed,” Strauss goes on to say, by his examination of the roots of historicism in
modern philosophy.

I was confirmed in my concentration on the tension between philosophy and the


polis, i.e., on the highest theme of political philosophy, by this consideration. What
distinguishes present-day philosophy in its highest form, in its Heideggerian form,
from classical philosophy is its historical character; it presupposes the so-called
historical consciousness. It is therefore necessary to understand the partly hidden
roots of that consciousness. . . . Politics and political philosophy is the matrix of the
historical consciousness.46
92 T I M O T H Y W. B U R N S

The results of an investigation of modern philosophy, then, by exposing the roots


of historicism in modern political philosophy, provide an important confirma-
tion of what Strauss had learned from Plato at this time.
This helps us understand why, having presented the Bible as something
Maimonides saw as producing a “cave beneath the cave,” Strauss turns, in the
conclusion of his talk, to presenting modern philosophy’s struggle against the
Bible and its doctrine of creation and providence—that is, modernity’s effort to
remove this fourth impediment to philosophy—as having brought into being a
neww cave beneath the cave, one that we now inhabit. A sketch or map of “the
struggle of the entire last 3 centuries,” he asserts, which sought to “destroy” the
biblical impediment to philosophy and thereby restore “philosophy in its natural
difficulty,” is contained in Maimonides’s brief account of the biblical impedi-
ment to philosophizing.
This modern, Enlightenment struggle characterized itself as a struggle against
“prejudices,” as opposed to the classic philosophic struggle against appearance
and opinion, but (as the later talk especially makes clear) the “prejudice” it
chiefly had in mind was biblical faith in the creator God and His miracles.47
What this modern struggle has achieved, however, Strauss argues, is “not the
freedom of questioning, only the freedom of saying No instead of the traditional
Yes. (Mortality versus immortality, chance versus providence, atheism versus
theism, passion versus reason).” And this achievement was bought at the price off
an “entanglement” in the biblical tradition of providence in such a way as to hide
that entanglement, by causing awareness of the tradition itself to dissolve. But
finally, both of “[t]he pillars on which our tradition rested, Prophets and Socrates-
Plato, are torn down since Nietzsche,” with the result that the question of the
right way of life “has . . . regained its full sharpness.”

We can no longer read Plato’s dialogues superficially, in order to notice admiringly


that old Plato already knew this and that; we can no longer polemicize against
him superficially. And the same with the Bible: we no longer think without evi-
dence that the prophets were in the right; we ask ourselves seriously whether it
was not the kings who were in the right. We really must begin entirelyy from the
beginning.
...
But we cannot answer immediately as we are; for we know that we are deeply
entangled in a tradition; we are yet much further down than Plato’s cave dwell-
ers. We must raise ourselves to the origin of the tradition, to the level of natural
ignorance. If we wanted to concern ourselves with the present situation, we would
be doing nothing other than the cave dwellers who describe the interior of their
cave. (RSP)

The outcome of the modern struggle was, then, two-fold. On one hand, it
resulted finally in the tearing down of the tradition, achieved by Nietzsche, so
that a truly fresh start is now possible, with nothing taken for granted. On the
other, it resulted in the creation of a new cave beneath the cave, a new claim to
authoritative knowledge against the philosophic life, a new “entanglement” in
t e biblical
the b b ca tradition
t ad t o of p ov de ce.48 We may
o providence. ay squa
squaree tthee two results
esu ts by say
saying,
g,
S T R AU S S ON R E L IG IOU S A N D I N T E L L E C T UA L S I T UAT ION 93

provisionally, that Nietzsche’s tearing down of the tradition achieved its result
only through continued entanglement in the tradition of providence, so that
to make a genuinely fresh start, to make the return to the ancients sought by
Nietzsche, we need to climb out of the new cave beneath the cave created by the
tradition.49 Understanding that “entanglement” will therefore help us to under-
stand both the genesis of historicism out of modern philosophy and how it con-
firmed what Strauss had learned from Plato.

* * *

Strauss first sketches three examples of this entanglement, and then offers a six-
point “Example” of it in the moderns’ addressing of “The Problem of Creation.”
The first three examples illustrate how the Enlightenment arrived at its “No” (to
immortality, providence, theism, and understanding) through its appropriation
of aspects of the biblical tradition of providence, with opponents of enlighten-
ment appropriating in turn aspects of enlightenment thought. So religious intol-
erance was fought by the Enlightenment in the name of the biblical principle off
love of neighbor, and later, the Enlightenment understood God as a construct by
“internalizing the purposes of God into mankind: self-redemption of mankind,
self-assurance of immortality (museum, etc.), assuming the role of providence.”
Late opponents of the Enlightenment, that is, religious believers, for their part
reinterpreted traditional notions in light of the Enlightenment’s successes: “E.g.,
revelation is understood as being a human product, as morals and as form, not as
law; creation is understood not as being the creation of the world, but as what
is binding in advance on human beings.” (Here Rosenzweig’s thought seem to
be intended.)50 Strauss’s objection to this mutual “entanglement,” this series off
reactions and counterreactions, is that it was carried out with no questioning off
the foundation. “Ex[ample]: [in the] concept of the ‘irrational’—rationalism [is]
understood in the narrowest sense,” that is, only as modern rationalism, as he calls
it in the opening of Philosophy and Law w (1935). His suggestion is that modern
rationalism, in its activist battle with biblical faith, took over certain activities
and presuppositions of that faith, in such a way as to make this rationalism ques-
tionable as genuine rationalism, and, whatever its initial success in advancing
the acceptance of atheism, to make it easy for believers or potential believers to
appropriate or dismiss as unimportant the findings of modern enlightenment
rationalism.
What Strauss has in mind becomes clearer in the second part of the account
of the “entanglement” of modernity in the tradition of biblical providence, the
six-point example of “The Problem of Creation.” This “example” is a telescoped
picture of the whole development of modernity, in its protracted response to
(as he says in Point #1) the biblical, radically free, creative God who works
miracles. It provides the first outline of what Strauss would later call a nonhis-
toricist account of historicism—of the development of the doctrine taken forr
granted or neglected by Heidegger. He begins (as it seems to me) with Spinoza,
who presented a new, materialist metaphysics that would correspond with the
recent
ece t discoveries
d scove es in physics
p ys cs (Point
( o t #2:
# : “miracles
ac es are
a e unworthy
u wo t y of
o God as thet e
94 T I M O T H Y W. B U R N S

perfect being”). But as Strauss had already argued in Spinoza’s Critique off Religion,
l
the effectivee part of Spinoza’s critique was not his materialist metaphysics; it was
instead the “positive” critique, that is, the claim that the mind disposed to a sci-
entific explanation of things, the mind that had progressed from ancient super-
stition and ignorance, was far better able to understand the truth about miracles
than were their alleged witnesses in the remote past. This positive critique arose
from Spinoza’s teaching: he had not only attempted to explain miracles scien-
tifically but had done so with a consciousness of the superiority of the scientific
mind over what is ancient and therefore backward or barbaric. “Reason must
become Geistt in order to be able actively to experience its more than royal free-
dom, its sovereignty which is incapable of being shaken by anything.”51 And as
Geist, reason does not take seriously the Calvinist position that a creator-God
is behind all that is and can work miracles in accord with his will—the position
that is compatible with or the theological equivalent of “skepticism.”
What Strauss was discovering at the time of these talks, through his work
on Hobbes (articulated in the manuscript he left unfinished and unpublished,
Hobbes’s Critique of Religion) is that this most effective part of the modern critique
of revelation, this secular advance and triumph of the “positive mind,” was in
fact intendedd to be such—to be the very means of defeating the challenge posedd
by revelation, in its most radical, Calvinist form. For unlike Spinoza, Strauss
was discovering, Hobbes—along with Descartes—had grantedd to revelation the
possibility that the world is indeed radically contingent, that is, that it is possibly
the work of a hidden, deceptive God. To secure the possibility of a mechanistic-
materialistic physics against this serious possibility, they had conceived a novel
science based on “a retreat into consciousness,” from which the positive mind
would prescribe its own laws to nature, so that man would become the owner
or possessor of nature. (Point #3: “Modern physics understands nature com-
pletely on its own terms: No scientific proof of God possible.” And Point #4:
“Nature a construct of the human intellect.”) But to secure a ground for this
new science, a science that had granted so much to revelation—to “avoid the
cavils of the skeptics,” as Strauss would later put it 52—a transformation of the
given world and of human beings was needed. A new moral-political teaching,
aiming at a change in orientation of everyone—a grand civilizational project— —
would be the means to the groundingg of the new science, over and against the
claims of religious believers. The new science as a retreat into consciousness
requires and anticipates this result. The civilizational project is an attempt to
replace providence, to dispel the mundane fear that was thought to lie behind
the need for gods by world transformation; it is a “rebellion” against a harsh,
meaningless nature by a science that could offer humanity the true goods off
this world withheld by religious doctrines and the manipulation of clerics. This
modern attempt represents a modification of the Epicurean orientation, such
that its concern is no longer with liberating the philosophic individual from fearr
(of the numen) but instead with securing the happiness of all, by defeating those
earthly political powers that have used the delusion of religion and its imaginary
“other-worldliness” to cheat their peoples of the genuine goods and which have
d s upted the
disrupted c v peace.53
t e civil
S T R AU S S ON R E L IG IOU S A N D I N T E L L E C T UA L S I T UAT ION 95

For in their “No” to immortality, the early moderns had denied the natural-
ness of a concern for immortality; they held that it was an artificial, biblically-
induced concern, one supported by biblical “tales.” Relatedly, they denied the
naturalness of human sociability. Both of these denials stand in marked opposi-
tion to the Platonic Socrates, who saw the desire for immortality both as natural
and as a desire that proved to express itself in the human concern for justice orr
morality. With its “no” to immortality, modern thought turned away from the
Socratic questioningg of moral opinion through dialectic, and hence from the ques-
tioning of the (sometimes buried) hope for immortality.54 The early moderns’
attempt thus evinced an “entanglement in the tradition” of providence insofarr
as it represented a “rebellion” against a niggardly nature and against an alleg-
edly artificially induced desire for immortality and an attempt to overcome it
with a new, this-worldly morality that was compatible with the new materialist-
mechanistic physics. The attempt was aware of itself as “progressive,” over the
“prejudices” of the peoples of biblical faith. This is the “historical consciousness
of the Enlightenment.”
Historicism arises as a “correction” of that consciousness.55 The first majorr
indications of a need for its correction began to occur precisely as the modern
civilizational project seemed to have reached success. “Now, around 1750 the
structure of mechanistic physics and the politics resting on it is completed: the
consciousness of its problematic comes into the foreground.”56 The early mod-
erns’ attempted “enlightenment” of human beings, with its expectation that the
kind of happiness indicated would emerge from this enlightenment, produced
two reactions, one philosophic, found in “Hume and above all Rousseau,” the
other more political, found initially in Burke but more broadly in the “historical
school” that dominated German jurisprudential and political thought through-
out the nineteenth century.
The first—the philosophic—reaction to the modern teaching had two related
parts: a return of the very skepticism about causality that the early moderns’
civilizational project had attempted to expel, and an attempt to find a workable
political substitute for the now fully discredited doctrine of Christian natural law
and for the conscience, the means of that law’s promulgation, which had been
undermined by the new science and the political doctrine of the state of nature.
The reaction comes in Rousseau:

One sees that the promise of enlightened politics (Hobbes, Encyclopedia) to create
the just order through the propagation of mechanistic physics and anthropology
cannot be kept; one sees it (one, that is, Rousseau) because one learns to see again
from Plato the problem “science-politics . . . society needs ‘religion.’”57

Rousseau’s teaching was grounded in an awareness of the looming political-moral


threat to the new science and to the civilizational project on which it rested. He
attempted to preserve that project, by articulating on one hand an idea of happi-
ness that would be compatible with the new understanding, and on the other, a
notion of the general will—something tribal, preceding any social contract, and
sustained by a civil religion—that could replace the discredited Christian natural
96 T I M O T H Y W. B U R N S

law doctrine. Perhaps on account of his partial recovery of the Socratic-Platonic


awareness of the incapacity of most human beings to live in the light of science,
Rousseau’s presentation, in the Second Discourse, of the accidental development
of man’s very humanity in history—his replacement for the biblical account off
creation and its support of the Christian doctrine of the conscience—was written
in such a way that Kant could take it to be a vindication of divine providence.58
But it was of a providence now understood to be fully scrutable as a develop-
ment, through the pursuit of low ends, toward humanity’s present, developed
peak. That peak, as Kant establishes in his critique of practical reason, is the full
development of prudence and the “good will,” a version of Rousseau’s “general
will.”59 There is thus a further “entanglement,” within the philosophic tradition,
with the biblical notion of providence, within the Enlightenment itself.
Rousseau and Kant were nonetheless both attempting to secure the victory
of science or reason, and therefore attempting to correct the political or civi-
lizational project through which reason or science was to be vindicated. The
other reaction to the limited success of the early modern civilizational project
was directed by more strictly political concerns that nonetheless were informed
by both a philosophic skepticism and a faith in divine providence. This reaction
is most visible in the “historical school,” which Strauss had been examiningg
as early as 192560 and whose primary concern was not philosophic but politi-
cal-theological. Of this second—political—reaction Strauss says, in a letter to
Löwith, “historical consciousness is a result of the discontent of modernity with
itself.”61 The way toward the development of this school of thought was paved,
Strauss later argued, by Burke, whose opposition to theory or metaphysics was
founded in a kinship with “the skeptical tendencies of his contemporaries Hume
and Rousseau” and who adhered to a faith in divine providence (as transformed
by modernity).62 Burke’s skepticism gave him the same disposition toward philo-
sophic theory or metaphysics as that of Calvin, and therefore an understanding off
the beautiful as a thing divorced from reason, and of the best political constitu-
tion as one that is by chance, or “grown.” He argued for “prescription,” that is,
he championed the doctrine that the liberties of a people belong by right to them
because they have belonged to them time out of mind, without any reference
to a higher law; they were written before (prescriptum) and therefore trump any
other appeals, including the appeal to natural law. To Burke, the justice of what
history had “accidentally” produced was vouchsafed by his belief in (scrutable)
providence.
These twin developments paved the way, however, for a further development
of the modern civilizational project in Hegel, who now took man’s actions in
history as the proper subject of theoreticall life, of a new metaphysics that was
not about the whole or eternity but the development of spirit (Geist) through
History. (Point #5: “The whole of ‘culture,’ and religion with it, is understood
by analogy with this construction [of nature], as a construction of the human
spirit.”) Hegel argued (famously) that the development of rational human con-
sciousness was at an end: History was over, its development fully surveyable and
comprehensible as the accidental development of Geist. While his philosophy
S T R AU S S ON R E L IG IOU S A N D I N T E L L E C T UA L S I T UAT ION 97

of history significantly enhanced the status of man within the whole, it had the
effect of leaving almost no one satisfied.
Over and against Hegel’s doctrine the historical school attempted to main-
tain an undetermined future for significant human action, one that would
provide an alternative to the revolutionary or bourgeois-supportive appeals to
modern “natural law.” It looked for objective but nonuniversal or particularr
standards within the history of a given people or tribe. History was to supply
peoples with the moral standards formerly sought in revelation or in nature.
But when the historical school’s attempt at discovering objective but non-
universal standards in a particular history failed, its premise, the historicity
of man, was not abandoned, but instead maintained as a hitherto unknown
dimension of human existence. At the same time Kierkegaard, also in rebel-
lion against Hegelian rationalism, articulated the possibility of “a human life
which has a significant and undetermined future,” living authentically, in the
face of death,63 that is, what became “existentialism.” The ground of modern
science—the alleged advanced consciousness—now became questionable, as
representing one particular consciousness among many and one that, more-
over, lacked depth. In Strauss’s brief description of this anti-Hegelian think-
ing, we meet up again, it seems, with the thought of Rosenzweig. (Point #6:
“Novel understanding of the original religious attitude [in contrast to mysti-
cism]: the demanding God who summons before him. The abandonment off
creation remains.”)
This six-part “example” of the entanglement of modernity in the biblical
tradition of providence thus begins to expose the hidden “roots” of the allegedd
experience of “history” as a dimension of human existence. And—to move now
to the outcome of the development that Strauss here sketches—Nietzsche, hav-
ing accepted the fundamental premise of the historical school, deconstructed both
pillars of the tradition, biblical and philosophic, calling into question both the
Christian God and the philosophic enterprise of “plebian” Socratic philosophy
against the man of megalopsychia.64 Nietzsche, Strauss would later suggest, came
close to recovering classical rationalism. Judging “theory” to be deadly to “life,”
he faced this alternative: he could “insist on the strictly esoteric character off
the theoretical analysis of life—that is, restore the Platonic notion of the noble
delusion—or else he could deny the possibility of theory proper and so conceive
of thought as essentially subservient to, or dependent on, life or fate.”65 He did
the latter. And so in his effort to build a bridge back to the natural man of antiq-
uity, instead of restoring Platonic esotericism—as Strauss himself has now begun
doing—Nietzsche promulgated his doctrine of the eternal return, to combat
modernity’s faith in the progress of humanity toward the degraded notion off
“happiness” offered through the abolition of suffering. He offered a cosmological
doctrine that called for “endurance,” over and against the

millennia-old pampering due to belief in creation and providence. The rebellion


against the indifference of the universe, against its aimlessness, which lies at the
root of modern civilization, is an essential part of this pampering.66
98 T I M O T H Y W. B U R N S

In sum, the early moderns attempted to eliminate the obstacle posed to philoso-
phizing by the Bible and thereby restore the natural situation; they attempted
to secure a new kind of natural science, a constructivist science, by transform-
ing human consciousness, through a humanly providential transformation of the
world. The attempt held human fear of a nature indifferent to human suffering to
be conquerable by dint of awareness of scientific progress in conquering nature.
Historicism arose as a makeshift political-philosophic-theological effort to cor-
rect this attempt, to provide a morally satisfying human life by grounding moral
meaning in a “sacred” process of human history. When this attempt, too, proved
to be a failure, the alleged historical dimension of life was not abandoned, but was
now considered a “discovery.” Heidegger’s “radical historicism” seeks to provide
the philosophic, ontological ground of this historical consciousness, to demon-
strate that the experience of contingency is a true or genuine experience.
We draw the following conclusion. The recovery of Socratic dialectics, which
in the best cases produces a liberation from the moral and political opinions off
one’s cave and to the philosophic life, was “confirmed” for Strauss by bringing
to light the hidden roots of historicism, which arose owing to what was a seri-
ous mistake about human beings on the part of modern political philosophers.
Over and against the historical consciousness, Strauss has recovered in Plato the
preliminary, dialectical ascent to the philosophic life. This recovery was the
prerequisitee of his understanding of the deficiencies of early modern thought that
had called into being historicism as a reaction to that thought.67 What Strauss
realized, on the basis of his recovery of Plato, and of the pains entailed in the
dialectical ascent from the cave, was what one might call the inevitability of the
failure of the modern attempt at enlightenment, and so of the rise of the histori-
cal consciousness against it. He provides here an initial sketch of the nonhistori-
cist account of historicism that would permit the true fresh, tradition-less start
that Nietzsche’s tearing down of the tradition had otherwise made possible.

* * *

As should be clear by now, this talk, which begins with a critique of Mannheim,
expands into an implicit critique of the more powerful, more serious, more radi-
cal historicism of Heidegger, the proponent of what Strauss calls, in the second
talk, “radical reflection.” To grasp better this implicit critique of Heidegger and its
ramifications, political and philosophic, a preliminary statement on the relation
of Mannheim and Heidegger is in order.
Mannheim was a follower of, among others, Max Weber (and for a time a
friend and colleague of Weber’s brother Alfred, in Sociology, at Heidelberg), and
a founder of “the sociology of knowledge.” But on the crucial issue of whetherr
reason could provide ethical or moral guidance, Mannheim broke with Weber.68
For Weber had, for his part, been led by his reading of Nietzsche to see the
depth, seriousness, and vitality of the life of religious faith. Repulsed by the
emerging shallow world ruled by science—a world of “specialists without spirit
or mind, and voluptuaries without heart”—he had declared the impossibility off
a swe g tthee quest
answering question
o o of tthee right
g t way o
of lifee ((leading
ead g to tthee famous
a ous fact/value
act/va ue
S T R AU S S ON R E L IG IOU S A N D I N T E L L E C T UA L S I T UAT ION 99

distinction). While Weber had, out of intellectual probity, held on to the life off
scientific reasoning “as a vocation,” and attempted to explain the specific path
that had led to it, he was closer to Heidegger than to Mannheim on the central
issue of the impossibility of a rational ethics. As Strauss later put it,69 Mannheim
had attempted, over and against Weber, to show that reason and argument could
reach agreement on the fundamentals—that the urgent question of the right
way of life could and must be settled not, as theologically inclined thinkers
like Rosenzweig were claiming, by a return to revelation, nor, as thinkers like
Schmitt and Jünger were claiming, through politicall authority, but by examining
the presentt as the fruitful locus of all divergent opinions.70
Given his confidence in the possibility of rational guidance of nations, it is not
surprising that Mannheim’s Ideology and Utopia included an attack on Heidegger:
“He who would wish to have the irrational where, de jure, the clarity and astrin-
gency of reason should still hold sway, is afraid of looking in the eyes of the
mystery in its proper place.” Heidegger responded to this attack in his 1931–32
winter semester lectures on Plato, in which he examined at length the Republic’s
image of the cave. We recall that in that image, he who ascends out of the cave,
sees the truth and returns, is laughed at, and told that it is not worth going up.
Heidegger turns these laughing critics of the philosopher into Karl Mannheim
and his students: The philosopher, Heidegger says,

would be told he was one-sided, he was, coming from somewhere, holding (in their
view) a one-sided attitude; and presumably, nay, certainly, they have down there a
so-called “sociology of knowledge,” with whose help he would be informed that
he was operating with so-called ideological prerequisites, which of course gravely
disturbed the community of collective opinion in the cave and therefore had to
be rejected.

Mannheim’s thought belongs, in short, to “cave prattle.” 71 It thus came about


that both Strauss and Heidegger critiqued Mannheim by means of Plato’s alle-
gory of the cave—Strauss first, and Heidegger a year later.
Yet Heidegger manifestly differs from Strauss both in thinking that the ascent
from the cave can never be transhistorical or without “reflective reflectiveness,”
and in thinking, at the same time, that the inhabitants of one’s cave can and ought
to be transformed, one and all, by the one who has ascended out of it in a reso-
lutely historicist manner. To Strauss, this is evidence that Heidegger no less than
Mannheim failed to recover genuine Socratic philosophizing, inhabiting instead
the historicist cave beneath the cave, and would as a consequence remain funda-
mentally in the thrall of “the present.” Strauss makes this substantial agreement
between Heidegger and Mannheim, and its political consequence, explicit in a
talk he gave in 1940. Mannheim’s attempted dialectical synthesis of the opinion
of the present is, Strauss states, absurd, unless one assumes a progress in conscious-
ness that god is dead, that is, unless one is expecting a new, atheistic morality.
This new atheism would be opposed to the biblical morality, to progress, to the
brotherhood of man, equality of dignity of all men, etc., and would favor cour-
age and steadfastness.72 The synthesis for which Mannheim held out a hope could
100 T I M O T H Y W. B U R N S

become actual through Heidegger’s


d thought, but in a way that Mannheim, who
was fundamentally a liberal, would have found abhorrent.
The welcoming posture toward the present and its answers to the ques-
tion of the right way of life is, to be sure, far more obvious in Mannheim
than in Heidegger. In the year before this talk was written, Heidegger and
Ernst Cassirer had engaged in their famous debate at Davos, and the signifi-
cance of Heidegger’s defeat of Cassirer lay, for Strauss, in the fact that Cassirer
had silently dropped the claim that ethics is possible, while Heidegger faced
squarely the problem of the impossibility of ethics.73 Facing this impossibil-
ity goes hand-in-hand with Heidegger’s stance against liberal pluralism—off
which Mannheim’s “conspectivism” was the latest version—as degraded, not
permitting an experience of the Call of Being but instead ensnaring the spirit
in everydayness and utility.
Heidegger himself claimed indeed that there are no ethics in his thought.
For despite the fact that we find ourselves existing among other human beings,
existence of an other is, he argues, always as an “other,” an “also being there,”
and not as anything demanding reciprocity of behavior. (In the language off
Rosenzweig or Buber, the “other” is for Heidegger not a “thou.”) To Heidegger,
being with others is a being “together,” but not a being with each other. To be
sure, the “confrontation with the other” that our contemporary postmodernist
democratic theorists tend to rely upon as the basis of an “ethics” is an outgrowth
of Heidegger’s thought, but whether it permits the determination of one’s authen-
tic being, through resoluteness to be itself in the face of its certain death—upon
which Heidegger insisted—is doubtful. It would seem instead to foreclose such
determination, which Heidegger thinks is required for the “proper” (eigentlich)
existence, through the solicitude and concern that it would provide. There is no
ethics for Heidegger, secondly, because the resoluteness against one’s anticipated
death lacks a definite aim.74 Heidegger stands for the individual’s engagement orr
decision, with the claim that the authentic, knowing subject is ever the subject
engaged with finitude, death.
Yet as Strauss had noticed, there is, despite his denials an “ethics” or moral teach-
ing in Heidegger.75 Heidegger’s “resolute” human proves to be actively engagedd
with the present (over and against the merely imagined) into which he is “thrown.”
Heidegger stresses indeed that death permits us to become aware of our individual
temporality, of our being in time, but death is also for him, as Löwith points out,
the hidden motive of all historicalness (Geschichtlichkeitt) which is the substance off
existence, that is, of being engaged resolutely in something and exposed to one’s
fate. A resolute existence is one that “renews the possibilities of the past by way off
its anticipating resolve,” that is, by way of facing death. Resolute existence sub-
jects our “today” to criticism when and as it is merely living on the past. But as it
“hands itself over to itself,” taking upon itself responsibility for its individual being,
through resolute anticipation of death, it also hands over to itself past possibilities.
It is only these combined deeds, in fact, that lead to a “there and now” (a Da) forr
oneself, a historical “situation”76 in which the resolute human is not a spectatorr
or critic, but actively engaged.77 “Being present now on behalf of ‘one’s time,’”
o tthee “resolution
or eso ut o into
to the
t e world-historical
wo d sto ca ssituation” equ ed,78 and
tuat o iss required, a d with
w t it,
t,
S T R A U S S O N R E L I G I O U S A N D I N T E L L E C T U A L S I T U A T I O N 101

eternity is given up. By “one’s time” Heidegger may not mean the vulgar version
of Mannheim, but there is no other “world-time” than the vulgar version.79 And
in 1933, the “being present now on behalf of one’s time” meant becoming a Nazi
leader of Freiburg University. At that point Strauss, who had already seen the path
to which “resoluteness” was leading Heidegger, ceased reading him for 20 years.80
Those who would attribute to Strauss, with the greatest perversity imaginable,
sympathy with Heidegger’s National Socialism mistake Strauss’s recovery of eter-
nity for a recovery of the radically historical here and now.
For just as he took for granted the tradition of atheism that had yielded the cur-
rent situation, Heidegger took for granted the superiority of the moral, engaged,
political life. His radical historicism is driven, one might say, by the desire for an
elevated, morally committed life of decision, in the face of our apparent thrown-
ness. It mustt reject philosophy as having been practiced in ignorance of this
thrownness, as having been practiced in fact as an escape from this thrownness,
which allegedly reveals itself to one who faces death authentically. Heideggerr
must claim of philosophy that its search for universal causes is no more than
the attempt to avoid our thrownness, to avoid the thought that out of nothingg
everything comes to be. He needs this doctrine in order to recapture the resolute-
ness of the Christian moral life, but atheistically. He must in fact avoid anything
approaching certainty of death, since that would result in resignation and not in the
agitation he seeks. Moral men may be drawn to living authentically and enthused
in the being-toward death prophesied by Heidegger. But such Existenzz is pos-
sible only if one in fact resists apodictic certainty of death and resignation to it,
in order to elicit experiential awareness of death so as to call forth a noble life off
some sort—in order to experience, through confrontation with the experience off
death, a “call” of Being. Heidegger needs, no less than biblical faith, a Being that
calls and that is responsible for bringing all beings into being from nothing.
The idealizing philosophy of Hegel against which Heidegger, with so many
others, rebelled, allegedly dilutes the reality of death, which alone can make
one’s being “still there” manifest. This is what one may call the absolute empiri-
cism of Heidegger; the thinking being is the being aware that he will die. Now
Strauss for his part does not dispute this, or run for cover in the “old thinking.”
On the contrary, on the basis of his recovery from the tradition, his bringingg
to light the roots hidden by the tradition, he finds that the new thinking off
Heidegger merely continues, despite its claims to a stern facing up, the hidingg
of eternity (and hence of one’s mortality) begun by Hobbes and continued in
Hegel.81 While Strauss does not in this talk venture such a critique of Heidegger,
he does stress its fundamental ground, in his account of Plato’s cave allegory:
he stresses the terrible pains that, according to Plato’s Socrates, are experienced
by one ascending from the cave. Among those pains is, of course, the genuine
awareness of mortality and, as that awareness is made through dialectic about
the just and the noble into knowledge, the accompanying awareness of the lack
of cosmic support for one’s deepest wishes—which find their responses in the
shadows on the cave wall—or, in Heidegger’s terminology, the awareness that
the Call is an echo off the cave walls, unheard after the (dialectical) ascent, that
is,
s, precisely
p ec se y as the
t e genuine
ge u e knowledge
ow edge o of deat
death iss made
ade cclear.
ea .
102 T I M O T H Y W. B U R N S

These fundamental differences between Strauss and Heidegger are reflected in


their respective approaches to ancient texts. In “The Word of Nietzsche: God is
Dead,” Heidegger makes clear that a hermeneutic is always required and at work
in reading them; he claims that one cannot understand thinkers betterr than they
understood themselves, yet one cannot but understand them differently than they
understood themselves. He defends this distinction by criticizing the “layman”
who arrogates a right to himself.

Not only must any commentary gather the substance from the text, it must imper-
ceptibly and without being too insistent, add something of its own to it, from its
substance. This supplement is what the layman, regarding what he takes to be the
content of the text, always feels as an interpolation; it is what he, with the right
he arrogates to himself, criticizes as arbitrary. A proper commentary, however,
never understands the text better than its author understood it, though it certainly
understands it differently. Only this difference in understanding must be such that
it encounters the same thing which the explicated text is meditating.82

Strauss, with a deep irony, presents himself as the layman who would understand
thinkers as they understood themselves.
Yet while radicalizing historicism, and hence further encrypting one in the
cave beneath the cave, Heidegger had opened the way, without intending it, to
an ascent out of that secondary cave. Strauss found, with the help of Klein, more
compelling than Heidegger’s teaching the examination of the roots of the “tradi-
tion” that Heidegger’s Destruktion of the tradition had exposed:

While everyone else in the young generation who had ears to hear was either com-
pletely overwhelmed by Heidegger, or else, having been almost completely over-
whelmed by him, engaged in well-intentioned but ineffective rearguard actions
against him, Klein alone saw why Heidegger is truly important: by uprooting and
not simply rejecting the tradition of philosophy, he made it possible for the first
time after many centuries—one hesitates to say how many—to see the roots of the
tradition as they are and thus perhaps to know, what so many merely believe, that
those roots are the only natural and healthy roots. . . . Klein was the first to under-
stand the possibility which Heidegger had opened without intending it: the possibil-
ity of a genuine return to classical philosophy, to the philosophy of Aristotle and
of Plato, a return with open eyes and in full clarity about the infinite difficulties
which it entails.83

As Strauss put it in 1940, “Heidegger had made it clear that Plato and Aristotle
have not been refuted, because they had not been understood by modern phi-
losophers,” yet “Heidegger’s interpretation of Aristotle . . . is no more than a
beginning.”84

* * *

We have examined what was to be the clarifying preface to a talk on “Franz


Rosenzweig’s doctrine and his problematic.” While the promised talk on
S T R A U S S O N R E L I G I O U S A N D I N T E L L E C T U A L S I T U A T I O N 103

Rosenzweig has not come down to us as such, it is not difficult to see, from
the few allusions to his teaching that we have noted, that Strauss now found
Rosenzweig, no less than Heidegger, to be entangled in a modern tradition that
he neglects, to be reliant upon an Enlightenment critique of the creator God
that he has not examined.85 And despite the differences between Rosenzweig
and Heidegger, what Strauss argues here against historicism applies to the one
as to the other. To Strauss, the historical consciousness that leads to an enthrall-
ment to the present is a tyranny, a sub-cave, presided over by a bewitching god-
dess—something from which one ascends to lead a genuine human life. What is
entailed in that life can be understood only through a naive, careful reading off
the Platonic dialogues. But it is clear that from that present Strauss does not take
flight into the mystical eternity of “blood” or nationhood of an allegedly eternal
people, as does Rosenzweig. Strauss instead finds in Plato the ground for a resig-
nation to one’s own mortality. Like Strauss, Rosenzweig claimed, to be sure, that
the starting point of all thinking about the whole is “man in the plainest sense as
he still exists,”86 yet he too engages in an analysis of Existenz that is “reflective,”
aware of its “historical existence.”
In the talk he gave two years later, moreover, Strauss notes the deleterious
effect, for an honest investigation of the right way of life, of Rosenzweig’s histor-
icism. After chronicling the rise of political and cultural Zionism out of the failed
hopes of liberal assimilation, Strauss explains in that talk how this movement
actually meant that what had been understood by the tradition as the revealed
word of God and His Law were instead being understood, in the Enlightenment
manner, as a product of the human mind. It thus cloudedd the central issue of the
right way of life––the issue of faith or reason. Rosenzweig is presented as an
example of how this clouding affects even outstanding men in the ranks of the
Jewish “consolidation” movement:

Those Jews of our time who took an active part in the consolidation process, who
in this way came to accept the Law and who have not been driven mad by the
difficulty of believing in God, conceive of the Law differently than the Jewish
tradition. I recall the outstanding man of this group, Franz Rosenzweig, who
reproaches Jewish orthodoxy for de facto having granted priority to prohibitions
over commands (e.g., with respect to $# [shabbos; i.e., Sabbath]); in contrast, he
wants to regard prohibitions only as the other side of commands. In his struggle
against the rigidity of the Law, he goes so far as to want to dissolve entirely the
distinction between minhagg [custom] and din [law]. (RSP)

What Strauss notes here concerning Rosenzweig’s alteration of the tradition, his
accommodating “development” of it, is further elaborated upon and critiqued
in the autobiographical preface to Spinoza’s Critique of Religion, where Strauss
speaks of Rosenzweig engaging in a “conscious and radical historicization off
the Torah.”87 And over and against Rosenzweig’s praise of the justice of Cohen’s
condemnation of Spinoza, on the ground that Spinoza should have remained
faithful to a “living and hence changing tradition,” “deepening” the old by
reshaping it and thereby annihilating it, Strauss remarks:
104 T I M O T H Y W. B U R N S

This is indeed the question: whether the loyal and loving reshaping or reinterpre-
tation of the inherited, or the pitiless burning of the hitherto worshiped is the best
form of annihilation of the antiquated, i.e. of the untrue or bad. . . . Is the conser-
vatism which is generally speaking the wise maxim of practice also the sacred law
of theory?88

Strauss, far from being “conservative” in his thinking, equates the hermeneutic
“new thinking” of Rosenzweig (as well as the neo-Kantianism of Cohen) with
the wise principle of conservative political practice, and rejects it as utterly inap-
propriate to theory, that is, to thinking, to an individual’s serious pursuit of the
truth.89
It is, in short, only once one has wrested oneself free of the historicizing
attempts of the “new thinking,” and begins to takes seriously what a return to
genuine religious orthodoxy would mean, that the theologico-political prob-
lem comes to sight in its natural form. The true initial situation of the Jew is
not merely as a member of a specific people, but as someone who confronts
the revealed Law. This is how Maimonides begins his guidance of a perplexed
youth—as if he were a believing Jew and even an enemy of philosophy. For as
Strauss puts it in “A Giving of Accounts,” “[P]hilosophy is subversive . . . the
virtue of the philosopher’s thought is a certain kind of mania, while the vir-
tue of the philosopher’s speech is sophrosyne.”90 True radicalness of thoughtt will
always be accompanied by an apparent conservatism. We who have grown up
within modernity, in which the difference between genuinely independent
thinking and thinking aimed at and hence limited by political ends has become
obscured, are invited by the apparently conservative Strauss to recover this radi-
cal insight.

Notes
1. See Appendixes B and C of the present volume. I wish to thank Devin Stauffer and
Stuart Witt for their helpful comments on an earlier draft of this essay.
2. The autobiographical “Preface to the English Translation” (hereafter referred
to as “Preface to SCR”) written in 1962, appears in the English translation, by
Elsa M. Sinclair, of Die Religionskritik Spinozas als Grundlage seiner Bibelwissenschaftt
Untersuchungen zu Spinozas Theologish-Politischen Traktatt (Berlin: Akademie-
Verlag, 1930), reprinted in GS–1 1–361; SCR R 1–31 (also in LAM M 224–59 orr
JPCMM 137–77). The statement on the theological-political problem appears in the
German “Preface” to Hobbes Politische Wissenschaftt (Neuwied am Rhein und Berlin:
Hermann Luchterhand Verlag, 1965), the first publication of the original German
manuscript of PPH. An English translation of the German Preface, by Donald J.
Maletz, appeared in Interpretation, A Journal of Political Philosophyy 8 (1979–80), 1–3
and is reprinted in JPCMM 453–56.
3. Preface to SCRR 1, 3.
4. These two alternatives, though distinct, were not mutually exclusive for German
Jewish youth. On Rosenzweig’s association with and inf luence upon German
Zionist youth movements, see Paul Mendes-Flohr, “Rosenzweig and the
Kameraden: A Non-Zionist Alliance,” Journal of Contemporary History 26, no. 3/4
(September 1991): 385–402.
S T R A U S S O N R E L I G I O U S A N D I N T E L L E C T U A L S I T U A T I O N 105

5. Strauss’s first book, Spinoza’s Critique off Religion,


l is dedicated to the memory
of Rosenzweig. Between 1922 and 1925 Strauss worked at the Freies Jüdisches
Lehrhaus in Frankfurt-am-Main, founded by Rosenzweig, and published articles
in Der Judee and the Jüdische Rundschau. Thanks to Michael Zank, English transla-
tions of these are available in LSEW W 63–137.
6. On the relation between Rosenzweig and Heidegger, Strauss recommends (in
note 14 to the Preface to SCR) Karl Löwith’s Gesammelte Abhandlungen: Zur Kritik
der geschichtlichen Existenz 68–92. The German title of the essay is “M. Heideggerr
und F. Rosenzweig, ein Nachtrag zu Sein und Zeit.’” An English translation off
the essay appeared as “M. Heidegger and F. Rosenzweig, or Temporality and
Eternity” in Philosophy and Phenomenological Research 3, no. 1 (September 1942),
53–77. It appears as “M. Heidegger and F. Rosenzweig: A Postscript to Being andd
Time,” in NHE E 51–78.
7. Rosenzweig and Heidegger, too, attempt to set aside the superf luous and arrive at
“the one” thing that is driving toward decisions “because the traditional contents
of modern civilization no longer prove indisputable.” NHE E 55–56.
8. “Conspectivism” (1929), an unpublished review of Mannheim’s Ideology andd
Utopia. Translated in Appendix A of this volume.
9. Strauss’s impatience with what I have called “Reader’s Digest” versions of serious
thought is already present in his criticisms of Walter Kinkel’s Herman Cohen:
Eine einfuuhrung in sein Werk (Stuttgart: Strecker und Schröder, 1924), made in
1925. Strauss decries the dissolution of the “precision and consistency of Cohen’s
thought” for the sake of “popularity.” See “The ‘Jewish Writings of Hermann
Cohen’ by Dr. Leo Strauss” ( Jüdishe Wochenzeitung fur Cassel, Hessen und Waldeck ,
Year 2, Nr. 18, May 8, 1925), 1–3. Translated into English by Michael Zank in
“More Early Writings by Leo Strauss from the Jewish Weekly for Cassel, Hessen
and Waldeck (1925–28),” Interpretation 39, no. 2 (Spring/Summer 2012), 109–37.
10. See “Existentialism,” a talk given in February 1956 to the Hillel Foundation
of the University of Chicago, in Interpretation 22, no. 3 (Spring 1995): 303–20.
“Existentialism owes its overriding significance to a single man: Heidegger.
Heidegger alone brought about such a radical change in philosophic thought as is
revolutionizing all thought in Germany, in continental Europe, and is beginning
to affect even Anglo-Saxony. I am not surprised by this effect. I remember the
impression he made on me when I heard him first as a young Ph.D. in 1922. Up
to that time I had been particularly impressed, as many of my contemporaries in
Germany were, by Max Weber, by Weber’s intransigent devotion to intellectual
honesty, by his passionate devotion to the idea of science, a devotion that was
combined with a profound uneasiness regarding the meaning of science. On my
way north from Freiburg where Heidegger then taught, I saw in Frankfurt am
main Franz Rosenzweig whose name will always be remembered when informed
people speak about Existentialism, and I told him of Heidegger. I said to him: in
comparison with Heidegger, Weber appeared to me as an orphan child in regard
to precision, and probing, and competence. I had never seen before such serious-
ness, profundity, and concentration in the interpretation of philosophic texts. I
had heard Heidegger’s interpretation of certain sections in Aristotle. Sometime
later I heard Werner Jaeger in Berlin interpret the same texts. Charity compels me
to limit the comparison to the remark that there was no comparison. Gradually
the breadth of the revolution of thought which Heidegger was preparing dawned
upon me and my generation. We saw with our own eyes that there had been
no such phenomenon
p in the world since Hegel”
g ((304).
) See also “An Unspoken
p
106 T I M O T H Y W. B U R N S

Prologue to a Public Lecture at St. John’s College in Honor of Jacob Klein, in


JPCM M 450: “Nothing affected us as profoundly in the years in which our minds
took their lasting directions as the thought of Heidegger. . . . Certain it is that no
one has questioned the premise of philosophy as radically as Heidegger.”
11. This disposition—a public modesty for the sake of serious thinking, especially
with respect to the work of Heidegger—is not peculiar to this talk. Much later,
in the course of describing the impact of Heidegger on himself and Jacob Klein,
Strauss speaks of the “all too justified abhorrence” of “publicity” or “the lime-
light.” “An Unspoken Prologue to a Public Lecture at St. John’s College in Honorr
of Jacob Klein,” in JPCM M 449. Similarly, in a talk written in 1940, Strauss speaks
of the quiet serious work being done in German Postwar thought and, without
mentioning himself, presents his own position, over and against that of Heidegger.
“The Living Issues of German Postwar Philosophy” (1940), were published as an
appendix (115–40) to Leo Strauss and the Theologico-Political Problem, by Heinrich
Meier (Cambridge: Cambridge University Press, 2006). That talk also provides
a helpful roadmap of the period’s leading thinkers as their thought appeared to
Strauss.
12. On this point, see “The Living Issues of German Postwar Philosophy,” 133. And
on the failure of the German intellectual authorities, liberal as well as Marxist,
to understand the youthful critics of the present, see “German Nihilism” [April
1941], Interpretation 26, no. 3 (Spring 1999): 362: “Those opponents [of thinkers
like Spengler, Schmitt, Jünger, Heidegger] committed frequently a grave mistake.
They believed to have refuted the No by refuting the Yes, i.e., the inconsistent iff
not silly positive assertions of the young men. But one cannot refute what one has
not thoroughly understood. And many opponents did not even try to understand
the ardent passion underlying the negation of the present world and its potentiali-
ties. As a result, the refutations confirmed the nihilists in their belief.”
13. What Strauss may have in mind here by “fate” is (among other things) the experi-
ence of “existence,” allegedly made possible by both medieval theology and its
belief in the creator God and by the alternative to that belief, articulated by Farabi
and Avicenna, that what is, is the product of chance. See Emil L. Fackenheim,
“The Possibility of the Universe in Al-Farabi, ibn Sina, and Maimonides,” in
Proceedings of the American Academy for Jewish Research 16 (1946–47), 39–70, which
is explicitly indebted to Strauss. Already at the time of this talk Strauss had been
directed by Avicenna to Plato’s Laws for the understanding of “prophecy,” and
would have encountered there the Athenian Stranger’s remarkable and troubling
claim that we are the playthings of the gods, that is, of chance (Laws 804b–c).
14. Strauss elsewhere presents historicists as echoing Calvin in seeing the multiplicity
of philosophic opinions as “disgraceful.” NRH H 18, and “On The Bible Science off
Spinoza and His Predecessors,” in LSEW W 185.
15. Heidegger, “Was ist Metaphysik?” in Wegmarken, ed. F.-W. von Herrmann
(Frankfurt: Klostermann, 1976), 119–20; “What is Metaphysics?” Pathmarks, ed.
W. MacNeill (Cambridge: Cambridge University Press, 1998), 95–96. See also
Karl Löwith, “Heidegger: Problem and Background of Existentialism,” NHE E
29, 38–44, and in the same volume, “M. Heidegger and F. Rosenzweig,” NHE E
67–68. Cf. Löwith’s letters to Strauss, July 13, 1935 and August 18, 1946. Finally,
see Strauss’s letter to Gerhard Krüger, December 27, 1932: “The ‘substantial and
historical core’ of historicism is, as you correctly say, ‘the factual domination off
Christ over the post-classical world.’ . . . The immediate consequence—above all
in Heidegger—is:
gg Christianityy has brought
g to light
g dimensions of human life,
S T R A U S S O N R E L I G I O U S A N D I N T E L L E C T U A L S I T U A T I O N 107

that were unknown or inadequately known to ancient philosophy; . . . [this means]


most fundamentally: after the disintegration of Christianity, there remains, and
first becomes possible, a philosophy that preserves the ‘truth’ of Christianity; it is
as such deeper and more radical than ancient philosophy. Maybe this consequence
is correct—but it must be as such proven. And this is possible only on the basis of a
direct confrontation of modern with ancient philosophy. That is the legitimation
of my project as regards Hobbes—I mean, the direct confrontation with Plato.”
Translated by Thomas L. Pangle in “The Light Shed on the Crucial Development
of Strauss’s Thought by his Correspondence with Gerhard Krüger,” chapter 3 in
the present volume, pp. 129–130 above.
16. The Maimonidean text to which Strauss refers and which he partially quotes is
Guidee I 31 (ed. Munk, 34b–35a).
17. See “On the Bible Science of Spinoza and His Precursors” (1926), LSEW W 173––
200, esp. 178–81. This work is a roadmap, required by Julius Guttmann, for what
would become SCR, completed in 1929 and finally published in 1930; the same
understanding of Maimonides is present in SCR. It must be stressed that Strauss’s
rediscovery of exotericism in Maimonides goes much beyond the recognition
of esotericism, driven by a fear of persecution, to be found in Spinoza’s writing:
Strauss had already recognized such esotericism as early as 1924. See, for example,
“Cohen’s Analysis of Spinoza’s Bible Science,” LSEW W 140–72, especially 148,
151, 153, 158.
18. See “Preface to Hobbes Politische Wissenschraft,” JPCM M 454; “An Unspoken
Prologue to a Public Lecture at St. John’s College in Honor of Jacob Klein,”
JPCM M 450; and GA– –JPCM
M 462–63: “We have rediscovered exotericism.”
19. CuM 424–26; CaM 216–19. I use here the translation from Avicenna that Strauss
used for the epigraph of his AAPL. A translation of the whole passage, by Michael
Marmura, reads as follows: “the treatment of kingship is contained in the book by
Plato and the book by Aristotle on the regime, and the treatment of prophecy and
the Law is contained in their two books on the laws.” MPP–1 97 or MPP–2 75.
20. On the meaning and significance of this recovery of and return to Plato, over and
against radical historicism, that is, Heidegger, see Thomas L. Pangle, “The Light
Shed on the Crucial Development of Strauss’s Thought by his Correspondence
with Gerhard Krüger,” and Christopher Bruell, “The Question of Nature and the
Thought of Leo Strauss,” in Klesis, Revue Philosophiquee 19, Special Issue, Autour dee
Leo Strauss, ed. Timothy Burns and Lucien Oulahbib ( June 2011): 92–101.
21. That situation may perhaps best be appreciated by the counterclaims against a
return to classical thought made by his friends Löwith and Krüger. On Löwith, see
below, notes 22–31. On Krüger, see note 15, and see the letters to Krüger, January
7, 1930, June 26, 1930, February 7, 1931, May 7, 1931, and Pangle, “The Light
Shed on the Crucial Development of Strauss’s Thought by his Correspondence
with Gerhard Krüger.”
22. Löwith, “Nature, History, and Existentialism,” NHE E 29.
23. Löwith, “Heidegger: Problem and Background of Existentialism,” NHE E 42.
24. Löwith, “The Fate of Progress,” NHE, 160.
25. Löwith, “Heidegger: Problem and Background of Existentialism,” NHE E 43.
26. Löwith, “Nature, History, and Existentialism,” NHE E 29.
27. Löwith, “Heidegger: Problem and Background of Existentialism,” NHE, 49.
28. Löwith to Strauss, August 18, 1946.
29. Letter to Löwith, August 20, 1946. See also PPH H 142–43: “Confronted with
this absurd conclusion [that Anaxagorean, ‘Epimethean’ physics . . . necessarily
108 T I M O T H Y W. B U R N S

leads . . . to subjection to ‘what the Athenians believe’]


bl Plato does not without
further ado oppose to materialist-mechanistic physics a spiritualist-telelologi-
cal . . . but keeps . . . to what the ‘Athenians’ say.” And see NRH H 117ff.
30. See “The Problem of Socrates” (1956), Interpretation 22, no. 3 (Spring 1995):
329.
31. Löwith, “Heidegger: Problem and Background of Existentialism,” NHE E 44. See
Emil L. Fackenheim, “The Possibility of the Universe in Al-Farabi, ibn Sina,
and Maimonides” (cited by Löwith, NHE E 44n8) and consider especially 54n16
and 68n30–69n31. Fackenheim repeatedly thanks Strauss for his assistance in this
research.
32. See “Preface to Hobbes Politische Wissenschaft” in JPCM M 453, and GA– –JPCMM 461.
33. “On the Bible Science of Spinoza and His Precursors” (1926), LSEW W 186.
34. Ibid., 185.
35. “Sigmund Freud, The Future of an Illusion” (1928), LSEW W 211.
36. APT 99–109, esp. 108; see also 103ff., esp. 103, 106, 109. FP 388; “The Law off
Reason in the Kuzari,” PAW W 140–41; Preface to SCR R 29, and see also SSTX: [P]
olitical life and philosophic life . . . are incompatible in the last analysis: political
life, if taken seriously, meant belief in the gods of the city, and philosophy is the
denial of the gods of the city” (532). “[T]he large majority of men, the philoso-
phers of the past thought, would be deprived of the very basis of their morality iff
they were to lose their beliefs” (535).
37. “Comments on Der Begriff Des Politischen by Carl Schmitt,” SCR R 344–45. The
German original was published as Die Religionskritik Spinozas als Grundlagee
seiner Bibelwissenschaft Untersuchungen zu Spinozas Theologisch-Politischem Traktatt
(Akademie-Verlag, Berlin, 1930). See also Preface to SCR R 19.
38. FP 383. PAW W 17. See also Strauss’s account of Platonic dialectic in PPH H 136–69.
See also RCPR R 181: “[T]he reasoning is in Plato’s dialogues integrated into the
human drama. The reasoning is frequently, not to say always, faulty––deliber-
ately faulty, as it should be within an imitation of human life.”
39. “Progress or Return?” RCPR R 248; cf. 246. See also Strauss’s letter to Voegelin,
February 25, 1951: “I believe still today that the theioi nomoii is the common
ground of the Bible and philosophy––humanly speaking.” FPP P 78. See also Leo
Strauss “On the Interpretation of Genesis” [1957], L’Hommee 21, no. 1 ( janvier– –
mars 1981): 6, where Strauss refers to “the agreement as well as the disagreement
between the biblical and the Greek heritage.”
40. See the very last topic listed in the outline of the 1932 talk: “ v.) Plato and thee
Nomos and Revelation.”
41. How to Begin to Study Medieval Philosophy,” 216–17. See also “On Classical
Political Philosophy,” WIPP P 92. That Strauss had already arrived at this under-
standing at the time of these talks is evident from the concluding part of CuM
419–29; CaM 209–22.
42. What we see here corresponds, in other words, to what Strauss later calls “the
self-destruction of historicism”: “The study of earlier thought, if conducted with
intelligence and assiduity, leads to a revitalization of earlier ways of thinking. The
historian who started out with the conviction that true understanding of human
thought is understanding of every teaching in terms of its particular time or as an
expression of its particular time, necessarily familiarizes himself with the view,
constantly urged upon him by his subject matter, that his initial conviction is
unsound. More than that: he is brought to realize that one cannot understand the
thoughtg of the past
p as longg as one is guided
g byy that initial conviction,” PAWW 158.
S T R A U S S O N R E L I G I O U S A N D I N T E L L E C T U A L S I T U A T I O N 109

One might add that with the Platonic attention to the problem of contingency,
the case for Heidegger’s “metaphysics of presence” to explain the alleged blind-
ness of the ancients to contingency loses its force.
43. As he later expressed this in a general principle, “By bringing the process to light,
we free ourselves from the power of the result.” “The Living Issues of German
Postwar Philosophy,” 134.
44. Strauss elaborates on the “pit beneath the cave” in “How To Study Spinoza’s
Theologico-Political Treatise,” PAW
W 155–56.
45. GA–JPCM
– M 463–64.
46. GA–JPCM
– M 464.
47. Spinoza had attributed belief in miracles to “prejudice” (SCR R 134ff.), and already
in SCR, Strauss had distinguished this struggle against prejudice from the natural
struggle of classical philosophy against “appearance and opinion” (SCR R 178–81).
48. In “The Living Issues of German Postwar Philosophy” Strauss credits Nietzsche
with introducing him to the notion of an artificial cave protecting one from the
elementary problems, a cave visible as “the pre-modern tradition (of providence)
and also of “the modern tradition” (138–39).
49. For a helpful elaboration on Strauss’s understanding of Nietzsche’s (f lawed)
attempt to return to the ancients, see his correspondence with Löwith, June 23,
1935, Löwith’s response on June 28, 1935 and July 13, 1935, and Strauss’s rejoin-
der on July 17, 1935. These were published, with English translations, in Thee
Independent Journal of Philosophyy 2 (1978): 5–12, and ibid. 4 (1983): 105–19. See
especially Strauss to Löwith, June 23, 1935: “I think that you do not take seriously
enough those intentions of Nietzsche which point beyond Nietzsche’s teaching.
You do not enter into these enough. For it is not sufficient simply to stop where
Nietzsche is no longer right; rather one must ask whether or not Nietzsche him-
self became untrue to his intention to repeat antiquity, and did so as a result of his
confinement within modern presuppositions or in polemics against these.”
50. See PLB B 9.
51. SCR R 135, 146.
52. NRH H 169–70, 172; cf. 311–12 and 319 with 20 and 29–30.
53. Die Religionskritik Spinozas, 50–51, 61, 209–10, 224; see also PLB B 17, and Preface
to SCR R 29.
54. See Christopher Bruell, “The Question of Nature and the Thought of Leo
Strauss,” Klesis: Revue Philosophiquee 19 Special Issue, Autour de Leo Strauss, ed.
Timothy Burns and Lucien Oulahbib, ( June 2011): 92–101.
55. “Living Issues of German Postwar Philosophy,” 135.
56. Strauss to Löwith, August 20, 1946.
57. Ibid. See also the first part of Chapter 6 of NRH.
58. See NRH H 316 and the accompanying note 102.
59. GA–JPCM
– M 463–64: “The high point was reached in Kant’s teaching on the
primacy of practical, i.e., moral reason, a teaching prepared to some extent by
Rousseau: the one thing needful is a good will, and of a good will all men are
equally capable.”
60. See the review of Paul de Lagarde's work in LSEW W 90. And for an extended
examination of the Historical School, see NRH, Ch. 1.
61. Strauss’s letter to Löwith, August 20, 1946. To the quoted statement, Strauss adds
“Cf. Savigny, Beruf.”f Friedrich Carl von Savigny was one of the founders of the
German Historical School of jurisprudence; it stood in opposition to appeals to
modern rationalist natural law doctrine (Vernunftsrecht)
f t) which, with the export of
110 T I M O T H Y W. B U R N S

the principles of the French Revolution by Napoleon, were being made all overr
Europe. Beruff refers to his 1814 pamphlet Vom Beruf unserer Zeit für Gesetzgebungg
und Rechtswissenschaftt (new edition, 1892). For a discussion of the Historical School
and its preparation by Burke’s doctrine of “prescription,” see Strauss’s Autumn,
1963 University of Chicago course on Vico, 10–12 of the original transcript. See
also the second part of Chapter 6 of NRH.
62. NRH H 311–12. On Burke’s embrace of Democritean-Epicurean physics as the
ground of his skepticism, see NRH H 311–12 with 169–70 and 172. On his faith
in divine providence as behind and observable in the humanly unintended orr
accidental results of history, see NRH H 316ff.
63. NRH H 320–21.
64. See the last sentence of this section, where Strauss refers to megalopsychia. Cf.
CuM 407; CaM 194.
65. NRH H 26.
66. Strauss to Löwith, June 23, 1935, emphasis added. On the moderns’s “rebellion”
to which Strauss here alludes, and the role of biblical faith in that rebellion, see
CM M 41–43. Strauss there suggests that while the modern resolve to liberate man
from nature appears, to the moderns themselves, to be less naive than Aristotle’s
understanding of nature, it is better understood as a failure of resignation, a disap-
pointed hope in the existence of a caring God and a consequent, confused sense off
a “right to rebel,” in order to do for humanity what such a God would have done.
See Timothy Burns, “Leo Strauss on the Origins of Hobbes’s Natural Science,”
chapter 7 of the present volume, note 20, pp. 142–44.
67. See “The Living Issues of German Postwar Philosophy,” 134: “[W]hen studying
the genesis of historical consciousness, we look at it with the eyes of pre-modern
philosophy—we stand on the other side of the fence.”
68. Contrast Colin Loader, The Intellectual Development of Karl Mannheim (Cambridge:
Cambridge University Press, 1985), 25–27, 96–101.
69. “The Living Issues of German Postwar Philosophy.” Mannheim is discussed on
129–130.
70. As Strauss makes clear in his 1932 review of Schmitt’s The Concept of the Political,
he considered Schmitt to have succumbed, through historicism, to a version off
“presentism.” “Comments on Der Begriff des Politischen” (1932), published in trans-
lation in SCR R 331–51. As Strauss argues, to Schmitt, “‘all mind is only present
mind’; ‘all mental concepts [of the spiritual-intellectual sphere] . . . can be under-
stood only from the concrete political existence’ and . . . ‘all political concepts,
ideas and words have a polemical meaning’” (SCR R 331). Schmitt had, to be sure,
hoped to launch a critiquee of the present, of liberalism. And he had accordingly
been critical of Mannheimian liberal syntheses: “Agreement at any price is possible
only as agreement at the price of the meaning of human life, for such agreement is
possible only when man abandons the task of raising the question regarding what
is right, and when he abandons this question, he abandons humanity” (SCR R 348).
To this extent, there is agreement between Strauss and Schmitt. But to Schmitt,
the state of nature is to be returned to over and against “civilization,” in the man-
ner that Heidegger recommends being toward death, over and against “every-
dayness”: “Affirmation of the state of nature does not mean affirmation of war,
but ‘relinquishment of the security of the status quo.’ Security is relinquished, not
because war is something ‘ideal,’ but because one must return from ‘dazzling rep-
resentation,’ from the ‘comforts and convenience of the existing status quo,’ to the
‘cultural or social void,’ to the ‘mysterious,
y unimpressive
p origin,’
g to ‘undefiled,
S T R A U S S O N R E L I G I O U S A N D I N T E L L E C T U A L S I T U A T I O N 111

uncorrupted nature,’ so that ‘by virtue of pure, unpolluted knowledge . . . the


order of human things’ may arise afresh’” (SCR R 348). So too, Schmitt thinks
that evaluating the political “would be a ‘free, untestable decision’” (SCR R 349).
His true aim, according to Strauss, is against atheistic, this-worldly, anarchistic
socialism and technology; liberalism merely gets in the way of this real aim, as
would-be mediator, whom he would gesture aside with contempt, in the name
of “the opposite spirit and faith, which it seems, does not yet have a name” (SCR R
350–51). Strauss may imply that that name is National Socialism orr some kind
of otherworldly activism. Schmitt has in any event not yet escaped liberalism,
according to Strauss; he remains entangled in a polemic against it. “This involve-
ment is no accidental failure but is the necessary consequence of the principle that
‘all concepts of the spiritual-intellectual sphere . . . can be understood only from
the concrete political existence,’ and that ‘all political concepts, ideas and words
have a polemical meaning’” (SCR R 351). Strauss finds that this principle is “itselff
altogether based on liberal presuppositions” (SCR R 351). That is, it begins with the
prepolitical right of the individual. Schmitt’s “presentism,” as it were, causes him
to be unable to get outside of liberalism.
71. Gesamtausgabee 34, 86. This, and the Mannheim statement about Heidegger are
quoted by Rüdiger Safranski in Martin Heidegger: Between Good and Evill [transla-
tion of Ein Meister aus Deutschland], trans. Ewald Osers (Cambridge, MA: Harvard
University Press, 1998), 214.
72. “The Living Issues of German Postwar Philosophy,” 129–30. Consider also
Strauss’s brief remark, in the 1929 review of Ideology and Utopia, on Mannheim’s
treatment of fascism, and the following remark in the autobiographical preface to
SCR: “It was obvious that Heidegger’s new thinking led far away from any char-
ity as well as from any humanity” (SCR R 9). Strauss argues in the Preface to SCR R
that biblical thinking would be an explicit, key component of the new synthesis:
“Heidegger wishes to expel from philosophy the last relics of Christian theology
like the notions of ‘eternal truths’ and ‘the idealized absolute subject.’ But the
understanding of man which he opposes to the Greek understanding of man as the
rational animal is, as he emphasizes, primarily the Biblical understanding of man
as created in the image of God. Accordingly, he interprets human life in the light
of ‘being towards death,’ ‘anguish,’ ‘conscience,’ and ‘guilt’” (SCR R 12). Contrast
Karl Löwith, who finds Heidegger’s fundamental concepts—death, conscience,
guilt, anxiety, fear, and corruption—to have a “hidden history” in “the Christian
tradition,” such that Heidegger was “unconsciously” prevented from presenting
certain acts (like suicide) as consistent with an independent, resolute existence.
“M. Heidegger and F. Rosenzweig,” in NHE E 67–68. See also Ch. 3 of the same
volume, “Heidegger: Problem and Background of Existentialism”: “I think it
would be very difficult to refute the so-called ‘nihilism’ of existential ontology,
on theoretical as well as morall grounds, unless one believes in man and world as a
creation of God or in the cosmos as a divine and eternal order—in other words,
unless one is not ‘modern’” (NHE E 38). “The fundamental limitation of Greek
ontology is, in regard to the problem of Being, that it understands man’s being
and Being in general in orientation to the worldd as cosmos and physis, and in regard
to the problem of Time, that it takes its orientation from what is presentt and always
presentt or eternal” (NHE E 39–40).
73. “There was a famous discussion between Heidegger and Ernst Cassirer in Davos
which revealed the lostness and emptiness of this remarkable representative off
established academic philosophy
p p y to everyone
y who had eyes.
y Cassirer had been
112 T I M O T H Y W. B U R N S

a pupil of Hermann Cohen, the founder of the Marburg neo-Kantian School.


Cohen had elaborated a system of philosophy whose center was ethics. Cassirerr
had transformed Cohen’s system into a new system of philosophy in which ethics
had completely disappeared: it had been silently dropped. He had not facedd the
problem. Heidegger did face the problem. He declared that ethics is impossible
and his whole being was permeated by the awareness that this fact opens up an
abyss.” “Existentialism,” 304. For a helpful analysis of the discussion at Davos,
see Peter Eli Gordon, Continental Divide: Heidegger, Cassirer, Davos (Cambridge:
Harvard University Press, 2010), Ch. 4.
74. As Löwith explains, “only when a decision is in the making, is the necessary
vagueness of its ‘for what’ replaced by a definite aim. To make up one’s mind
depends on the actual possibilities of the historical situations. Hence, Heidegger
refuses to be positive or even authoritative as to existential liabilities. . . . The
resolve, thus does not come to any conclusion; it is a constant attitude, formal
like the categorical imperative and through its formality open to any material
determination, provided it is radical.” “M. Heidegger and F. Rosenzweig,” NHE E
66.
75. GA–JPCM
– M 457–66. “What I could not stomach was his moral teaching, for despite
his disclaimer he had such a teaching. The key term is ‘resoluteness,’ without any
indication as to what are the proper objects of resoluteness” (GA–JPCM
– M 461).
76. Sein und Zeit, 299.
77. As Strauss puts it in “The Intellectual Situation of the Present,” “[t]he ‘free deci-
sion of the person’ that does not want to depend on any justification ‘does not
take place in a vacuum.’ It is conditioned by the history in which the person con-
cerned stands.”
78. Sein und Zeit, 385; cf. 383–84, 299–300, 391
79. Sein und Zeit, Sections 78–81. See Karl Löwith, “M. Heidegger and F.
Rosenzweig,” NHE E 73–74.
80. “There is a straight line which leads from Heidegger’s resoluteness to his siding
with the so-called Nazis in 1933. After that I ceased to take any interest in him m
for about two decades.” GA–JPCM
– M 461.
81. See NRH H 175–76, and Timothy Burns, “Leo Strauss on the Origins of Hobbes’s
Natural Science,” chapter 7 of the present volume.
82. Nietzsche’s Word: “God Is Dead,” in Holzwegee (1950), trans. Julian Young and
Kenneth Haynes as Off The Beaten Track (Cambridge: Cambridge University
Press, 2002), 160.
83. “An Unspoken Prologue to a Public Lecture at St. John’s College in Honor off
Jacob Klein,” JPCM M 450, emphasis added. See also GA– –JPCMM 457–66. “In 1925
Heidegger came to Marburg. Klein attended his classes regularly, and he was,
naturally, deeply impressed by him. But he did not become a Heideggerian.
Heidegger’s work required and included what he called Destruktion of the tradi-
tion. . . . He intended to uproot Greek philosophy, especially Aristotle, but this
presupposed the laying bare of its roots, the laying bare of it as it was in itselff
and not as it had come to appear in the light of the tradition and of modern phi-
losophy. Klein was more attracted by the Aristotle brought to light and life by
Heidegger than by Heidegger’s own philosophy. Later Klein turned to the study
of Plato, in which he got hardly any help from Heidegger. Klein convinced me
of two things. First, the one thing needed philosophically is in the first place a
return to, a recovery of, classical philosophy; second, the way in which Plato is
read, especially by professors of philosophy and by men who do philosophy, is
S T R A U S S O N R E L I G I O U S A N D I N T E L L E C T U A L S I T U A T I O N 113

wholly inadequate because it does not take into account the dramatic characterr
of the dialogues, also and especially of those of their parts which look almost like
philosophic treatises” (GA–JPCM
– M 462).
84. “The Living Issues of German Postwar Philosophy,” 135 and 137.
85. See also on this point PLB B 8–9.
86. Kleinere Schriften (Berlin: Schocken, 1937), 363. Quoted by Löwith in “M.
Heidegger and F. Rosenzweig,” NHE E 55.
87. Preface to SCR R 8–15, 22–28, esp. 14.
88. Preface to SCR R 25; cf. 27 on Cohen. Strauss is aware that in the passage off
Cohen’s Religion of Reason Out of the Sources of Judaism (Religion der Vernunft, ch. X,
§19) that he here quotes, Cohen himself rhetorically puts the question of annihi-
lating the tradition. See CuM 401; CaM 185–86.
89. See also “The Living Issues of German Postwar Philosophy,” 115: “We may go so
far as to say that, generally speaking, German philosophy implies a more or less
radical criticism of the very idea of civilization and especially of modern civiliza-
tion—a criticism disastrous in the political field but necessary in the philosophi-
cal, in the theoretical field.”
90. GA–JPCM
– M 463. Cf. SA 137 and 281–82, with CM M 229–30.
C APT R 6
CHAPTER

CARL SCHMITT AND STRAUSS’S RETURN TO


PREMODERN PHILOSOPHY

Nasser Behnegar

B y his own account Strauss’s “Anmerkungen Zu Carl Schmitt, Der Begriff des
Politischen” (hereafter referred to as “Comments”)1—the only German essay
he published in English—marks the pivotal moment of his life, the moment
when he first expressed the possibility and even the necessity of a return to pre-
modern philosophy:

The present study [his book on Spinoza, which was completed in 1928] was based
on the premise, sanctioned by powerful prejudice, that a return to pre-modern
philosophy is impossible. The change of orientation which found its first expres-
sion, not entirely by accident, in the article published at the end of this volume
[“Comments,” originally published in 1932] compelled me to engage in a number
of studies in the course of which I became ever more attentive to the manner in
which heterodox thinkers of earlier ages wrote their books.2

What is at stake in this reorientation Strauss indicates by recollecting a train off


reasoning from Spinoza to Nietzsche that terminates in the dethroning of reason
and the “victory of orthodoxy” over philosophy.3 While he could not reject the
logic of this development, Strauss had philosophical, political, and even religious
reasons for being dissatisfied with its outcome, which dissatisfaction led him to
wonder whether “the self-destruction of reason” is “the inevitable outcome off
modern rationalism as distinguished from pre-modern rationalism.”4 Thus, forr
Strauss the fate of rationalism or philosophy came to depend on the soundness
of premodern philosophy. Although Strauss does not explain here the substance
or the power of the objection against premodern philosophy, he implies that the
“Comments” disarms that objection, at least insofar as he understood it then.
Being the first “expression” or articulation5 of his change of orientation, the
“Comments” is his first meaningful step toward the vindication of philosophy
or the life of reason.
116 NASSER BEHNEGA R

But is not Strauss being somewhat forgetful here? After all, in 1931 Strauss
published a review of On the Progress of Metaphysics in which he praised Ebbinghaus
for rejecting the modern “prejudice that thee truth has not already been found in
the past.” There he speaks for the first time of his notion of modernity as a cave
beneath the natural cave, and of the necessity of the historical recovery of the
original cave “from which Socrates can lead us to light.”6 Given that Ebbinghaus
was in Strauss’s mind in 1964–65 (he refers to him in his introduction to Hobbes’
t 7 it is more likely that his disregard of that review is due
politische Wissenschaft),
to his sense of its inadequacy. There Strauss argues that modern progress has led
to “the anarchy of systems” and that the proper response to our ignorance is to
read old books with the aim of learning from them or to move from our his-
torical ignorance to the natural ignorance that was the starting point of Socratic
thinking. Yet it is possible to break with the Socratic tradition while knowingg
the natural cave and the natural ignorance: “In [a horizon beyond liberalism]
Hobbes completed the foundation of liberalism” (“Comments,” ¶35).8 A return
to premodern philosophy requires a criticism of the roots of the modern philoso-
phy as well as of its results. The Ebbinghaus review does not even examine these
roots, which come to view only in an attempt to carry out a radical criticism off
modernity.
According to Strauss, it was “not entirely by accident” that his questioning off
the modern rejection of premodern philosophy found its first expression in the
“Comments.” That there is at least some essential connection between Schmitt’s
thought and Strauss’s turn to premodern philosophy is evident from the startling
final sentences of the “Comments”: “A radical critique of liberalism is thus pos-
sible only on the basis of an adequate understanding of Hobbes. To show what
can be learned from Schmitt in order to achieve that urgent task was therefore
the main concern of our comments” (¶35). Since Strauss in the “Comments”
characterizes Hobbes as the founder of liberalism, this adequate understanding
of Hobbes is necessarily a critical one.9
What precisely did Strauss learn about Hobbes through his criticism off
Schmitt’s position? This question, which Strauss identifies as the main concern off
the “Comments,” receives scant attention in the most impressive and influential
account of this exchange. For Heinrich Meier, Schmitt is a political theologian,
a man whose political position rests on his Catholic faith; Strauss a political phi-
losopher, a man whose views (political or otherwise) rest on evidence available
to man as man; and the gulf dividing them is an “insuperable” one.10 When one
frames the encounter between Schmitt and Strauss as that between a political
theologian and a political philosopher, the theme of the critique of modernity is
apt to disappear, and so it does in Meier’s reading. Whether Meier’s understand-
ing of Schmitt is correct is beyond the scope of this chapter;11 it is, however,
necessary to point out (as Meier himself initially acknowledges) that it does not
accord with Strauss’s exposition of Schmitt’s position.12 For Strauss, Schmitt is a
critic of modernity who is also a modern man, an opponent of liberalism whose
position in part rests on liberal presuppositions, presuppositions incompatible
with those of Biblical faith and of premodern philosophy.
S T R AU S S ’ S R E T U R N T O P R E MODE R N PH I L O S OPH Y 117

The “Comments” is an extraordinary record of one serious man understanding


another. “You’ve got to read this [the ‘Comments’],” Schmitt said to his assistant.
“He saw through me and X-rayed me as nobody else has.”13 The “Comments”
not only marks the moment of Strauss’s return to premodern philosophy; it also
marks the moment in which Strauss’s prodigious interpretive powers came into
their own. Strauss follows Schmitt’s arguments in their own terms—step by
step—until he arrives at the true foundation of his position. We witness Strauss’s
attention to contradictions, both between different statements of Schmitt’s and
between a statement and what is implied in an argument; his grasp of the difficul-
ties that caused these contradictions; and his uncanny gift of discerning which
side of a contradiction expresses Schmitt’s deeper opinion. In the end Strauss
understands Schmitt better than Schmitt understood himself because he tries
to understand Schmitt exactly as he understood himself. In exposing Schmitt’s
numerous contradictions, he at the same time reveals Schmitt’s thoughtfulness;
he reveals him as a modern man trying to break through the modern prison. The
“Comments” is indeed “very critical,”14 but it is also very helpful in a way that
few criticisms are. Schmitt’s grateful reception15 testifies to this as it does to his
own intellectual seriousness, for such criticisms, as it hardly needs saying, are not
always taken as the gifts that they are.
In the interpretation that follows, I attempt to identify the roots of the contro-
versy between premodern and modern philosophy as they manifest themselves in
the “Comments.” I do this by focusing on the twists and turns of Strauss’s argu-
ment. But to makes these intelligible I reproduce Strauss’s argument in a mannerr
that necessarily suffers in comparison to the original.

Schmitt’s Intention and Strauss’s Hermeneutic Principle


The “Comments” begins with a contradiction that is not fully resolved until
the very end of the work. Schmitt’s exposition of the concept of the politi-
cal is an attempt to understand the state radically, for the state presupposes the
political. This attempt has become necessary on account of the contemporary
dissolution of the consensus regarding the purpose of the state (as evident in
the tension between the limited liberal state and the total state of democratic
or mass movements). Thus, the contemporary crisis has caused Schmitt to raise
and address a general question, the question of “the order of human things,”
which Schmitt identifies with the question of the state (¶1). Yet, Strauss observes
that this account, derived from Schmitt’s own statements, is inconsistent with
Schmitt’s “general principles of understanding,” which deny the existence off
“eternal truths,” understand all concepts of the spiritual sphere “in terms of theirr
concrete political existence,” and maintain the polemical character of “all politi-
cal concepts, presentations, and words” (¶2). Strauss thus tries to understand
Schmitt in light of his own principles of understanding. Schmitt’s attempt to
answer the question of the state rests on the premise that “the political precedes
the state,” which premise is asserted in opposition to liberalism, the movement
in which the modern goal of neutralization and depoliticization has reached
118 NASSER BEHNEGA R

“its greatest efficacy” (¶3). Since liberalism attempted to construct the rational
society by denying the political (through its individualism, and its insistence
of autonomy of the spheres of economics, ethics, art, and religion), the failure
of liberalism has persuaded Schmitt that an adequate account of the state must
first posit the political: “Schmitt’s basic thesis is completely dependent on the
polemic against liberalism; it can only be understood as a polemical thesis, ‘only
from concrete political existence’” (¶3). While the guiding question of Schmitt’s
treatise is unpolemical, universal, and transhistorical, his attempt to prepare an
answer to it is completely polemical, concrete, time-bound.
Yet Strauss implies that there is something philosophical about Schmitt’s
polemic against liberalism. Schmitt sees the failure of liberalism in its self-
contradictory character: its attempt to banish the political from the world leads
only to carrying out politics “by means of anti-political speech” (¶4). He objects
to liberalism because it obscures reality: “Liberalism has not killed the political,
but merely killed understanding of the political” (¶4). Accordingly, his inten-
tion is not to bring down liberalism by provoking the indignation of the masses;
rather, he attempts to find a genuine alternative to liberalism (¶5). Of the dif-
ficulties of this task he has a rare grasp. Liberalism may contradict reality but
liberal thought is “astonishingly consistent” and saturates the intellectual and
moral horizon of European man. Accordingly, he maintains that liberalism has
not been replaced by any other system even in thought, which assertion pro-
vokes Strauss’s extraordinary assessment: “by this awareness he [Schmitt] stands
alone among the opponents of liberalism, who usually carry in their pocket a
fully-worked out illiberal theory” (¶6). Strauss suggests, and the essay ultimatelyy
proves, that Schmitt himself is forced to use liberal elements not just for rhetori-
cal reasons but because his own thought is entangled with them. To his credit,
Schmitt was to some extent aware of this state of affairs and therefore presented
his work in a provisional manner. From this Strauss draws a principle of inter-
pretation: “the critic is in duty bound to consider more wherein Schmitt differs
from the prevailing view than wherein he follows it” (¶6). This means that when
one sees a contradiction in Schmitt’s thought it is likely (all things being equal)
that the nonliberal element is expressive of his deeper opinion.

Hobbes’s State of Nature and the Idea of Culture


Strauss here argues that Schmitt’s concept of the political (his contention that
the distinction between friends and enemies is the ultimate political distinction)
is meant to provide a radical criticism of the idea of culture as the intellectual
framework within which the state is to be understood. This criticism in turn
leads him to restore and reexamine Hobbes’s notion of the state of nature as the
true ground of the state.
But what is culture? And why does Schmitt’s polemic against liberalism lead
him to criticize it? According to Strauss, the philosophy of culture is “the ulti-
mate self-consciousness of liberalism” (¶14). Liberalism subordinates the state to
society; the former does not rule but protects the latter. The philosophy of cul-
ture maintains that all human affairs (both the state and society) are the product
S T R AU S S ’ S R E T U R N T O P R E MO DE R N PH I L O S OPH Y 119

of the sovereign creation of the human mind. It thus incorporates idealism’s


critique of earlier thought (its contention that the human mind orders our real-
ity) into the liberal understanding of both the state and society. It further divides
society into different independent domains (economics, aesthetics, morality, etc.)
with their own independent criteria. This means, for instance, that morality, the
ultimate distinction of which is between good and evil, cannot have author-
ity over economic and aesthetic affairs, which have their own ultimate distinc-
tions (profitable and unprofitable; beautiful and ugly). It thus protects different
domains of society, especially the world of art and business, from intervention
from other domains in the affairs of the state, expanding the liberal spirit off
the society. Now, Strauss observes that some of Schmitt’s comments give the
impression that he merely wants to recognize the political as an independent
domain along with the other domains. In other words, it appears that Schmitt
wants to elevate the place of the state (giving it the same autonomy enjoyed
by the different spheres of society) while preserving the liberal aspirations for
autonomy (¶8). But other comments show that this cannot be his intention: “the
distinction between friend and enemy is ‘not equivalent and analogous . . . to those
other distinctions’; the political does nott describe ‘a new domain of its own’” (¶7).
Indeed, Schmitt’s concept of the political does more than preserve the autonomy
of politics. The distinction between friend and enemy (enemy meaning public
enemy) presupposes the possibility of war, and the possibility of men physically
killing each other is the extreme possibility not only in politics but also in all
human life. In such a situation, this distinction simply becomes the authoritative
distinction. The claims of artists, businessmen, and moralists give way to that off
the statesmen when the existence of a country is threatened. While such threats
do not always exist, their ever-present possibility casts its light on the normal
situation. Schmitt’s concept of the political is meant to establish the authority off
the political (¶9).16
But the authoritativeness of politics is incompatible with the understanding off
politics as a kind of cultural phenomenon: no creation of the human mind can
exercise an authority over that mind. Strauss discerns in Schmitt a desire to criti-
cize the whole notion of culture. He notes that Schmitt expressly maintains that
in order to define the political one must begin not with the genus to which the
political belongs but with what is specific to the political. Since no adequate defi-
nition of a thing can do without its genus, Schmitt’s choice of a partial definition
of the political implies that he intends to pioneer “a path to an original answerr
to the genus question by using the phenomenon of the political to push the
most obvious answer [culture] ad absurdum” (¶7). Strauss shows what is entailed
in Schmitt’s will to deny that politics is fundamentally a cultural phenomenon
by deconstructing the idea of culture. Culture is the cultivation of nature, and
therefore presupposes the notion of nature. The liberal notion of culture as the
sovereign creation of the mind is possible “only if the nature being cultivated has
been presupposed to be the oppositee of mind, and been forgotten” (¶10). A critical
inquiry into this notion of culture demands that one should articulate the natural
state of men. While Hobbes’s civil state is consistent with the liberal notion off
cu tu e, hiss accou
culture, accountt o
of tthee state o
of nature
atu e in so
somee ways mirrors
o s Sc
Schmitt’s
tt s co
concept
cept
120 NASSER BEHNEGA R

of the political. Hobbes’s state of nature is a state of war, not in the sense of actual
fighting but of known disposition to fighting. Hobbes’s state of nature is “the
genuinely politicall status” (¶11). Accordingly, Strauss notes that Schmitt restores
“the Hobbesian concept of the state of nature to a place of honor” (¶11).17 Politics
is not fundamentally a cultural phenomenon because politics is the natural state
(order) of man. For Schmitt politics is natural because “there is a primaryy ten-
dency in human nature to form exclusive groups.”18
While Schmitt restores the seventeenth-century concept of the state of nature,
he also changes much of its content, turning it into a political state. Strauss
exposes the difference between the Hobbesian concept of the state of nature and
Schmitt’s own concept:

For Hobbes, it is the state of war of individuals; for Schmitt, it is the state of war
of groups (especially) nations. For Hobbes, in the state of nature everyone is the
enemy of everyone else; for Schmitt, all political behavior is oriented toward friend
and enemy. (¶12)

Strauss suggests that Schmitt’s notion is the truer one, for Hobbes’s account is
corrupted by his polemical intention. Hobbes conceived the state of nature as
the war of all against all because he wanted to show the necessity of abandoning
the state of nature (¶12). The two accounts are more than just different; they are
opposed to each other; insofar as Hobbes’s state of nature was designed with the
view of being negated, it is the presupposition of the liberalism against which
Schmitt is fighting. This is already evident in the political teaching that Hobbes
himself draws from the state of nature. Hobbes rejects Schmitt’s contention that
a political community has the right to “demand . . . from the members of its own
nation the readiness to die” (¶13). Hobbes conceives of the public good in a way
that leads “to the ideal of civilization, that is, to the demand for rational social
relations of humanity as a singlee ‘partnership in consumption and production’”
(¶13). By being the founder of the ideal of civilization, Hobbes is also the founderr
of liberalism: his articulation of that right to secure mere life “has fully the char-
acter of an inalienable human right, that is, of an individual’s claim that takes
precedence over the state” (¶13).
The philosophy of culture presupposes Hobbes’s state of nature. Schmitt’s cri-
tique of the philosophy of culture leads him both to restore the cognitive claim
of this forgotten basis and to criticize its content: “Schmitt returns, contrary to
liberalism, to its author, Hobbes, in order to strike at the root of liberalism in
Hobbes’ express negation of the state of nature” (¶14). While Hobbes negates
the state of nature, Schmitt affirms it. By observing this Strauss goes beyond
Schmitt’s own self-understanding. Though by the time of the second edition
of “The Concept of the Political” Schmitt may have become aware of a tension
between his position and that of Hobbes (¶14n), it is doubtful that he conceived
of Hobbes as his fundamental adversary. Perhaps he failed to see this because
he was impressed by Hobbes’s argument for obedience to the state (individuals
ought to obey the state because it protects them) and by Hobbes’s recognition
of what Strauss calls here “the natural evil of man” (¶13). Nonetheless, Strauss’s
S T R AU S S ’ S R E T U R N T O P R E MODE R N PH I L O S OPH Y 121

contention remains valid. Whether he knew it or not, Schmitt’s thought involves


a confrontation with Hobbes’s concept of the state of nature. This is the theme
of Strauss’s final section, which theme explains the otherwise surprising final
sentences of the work.

From Necessity to Morality: Schmitt’s Confrontation with Hobbes


In the final section of the “Comments,” Strauss shows step by step that Schmitt’s
moral aloofness hides an intense moral seriousness. He takes us from Schmitt’s
claim that one only needs to recognize the reality of the political, to the claim
that one only needs to recognize its necessity, to the claim that one must rec-
ognize its desirability, and finally to the claim that the political is a demand off
moral seriousness. On the one hand, he traces this movement to reflections that
require the abandonment of the fundamental premise underlying the Hobbesian
concept of the state of nature. On the other hand, he traces Schmitt’s resistance
to that movement (or his will to conceive of the political in amoral terms) to his
acceptance of that premise.
Strauss begins with Schmitt’s claim that it is immaterial whether one rec-
ognizes “the political as desirable or detestable” so long as one recognizes its
undeniable reality (¶15). He observes that Schmitt’s intent here is the opposite
of that which animates value-free social science. He does not want to leave open
but to close “all possibilities for taking an evaluative stance toward the political”
(¶15). Any attempt to evaluate the political presupposes a standpoint higher than
the political. Any such attempt leads to degradation of the political: “the politi-
cal is constituted by reference ‘to the real possibility of physical killing’ of men
by men; and ‘there is no rational purpose, no norm however correct . . . that can
justify men’s killing one another for its own sake’” (¶15). But if the political can-
not be evaluated, how can it be a guiding principle for the state? The political
could be a guiding principle if it were inescapable, if it were a basic and neces-
sary character of human life (¶18). Schmitt sometimes is satisfied with less. Since
according to him all concepts are related to their time, he occasionally writes as
if it is enough to know that the political constitutes our present reality, that today
men still form exclusive groups with the view of the possibility of war (¶16).
This response, however, is inadequate, for existing groups act with a view to the
future and “there is todayy a powerful movement striving toward the total elimi-
nation of the real possibility of war and hence the abolition of the political” (¶17;
emphasis added, N.B.). Since Schmitt’s position is especially directed against
such movement, Strauss argues that he, more than others, cannot ignore its hope
for the future. Schmitt is not always oblivious to this difficulty. Accordingly, he
argues that every attempt to abolish the political or the possibility of war has to
take a political form if it is to succeed: it has to become a “war against war.” But
such wars are likely to be especially inhumane, for the enemy is understood to be
the enemy of humanity and hence a monster. The attempt to preserve humanity
by eliminating war only leads to an increase in inhumanity. Even if the pacifists
win they will lose, because the inhumanity that such a war engenders forces men
to form exclusive groups to defend themselves from each other (¶18). Politics
122 NASSER BEHNEGA R

cannot end because the dangerous character of man cannot be eliminated: “the
thesis of man’s dangerousness is the ultimate presupposition of the position of the
political” (¶19).19
Having articulated what at first glance appears as a perfect summary off
Schmitt’s position, Strauss argues that this train of thought is “not Schmitt’s last
word, and it is certainly not the most profound thing that he has to say” (¶20).
Schmitt cannot maintain the inescapable necessity of the political, because he
cannot claim to know that man will always be evil or dangerous. Schmitt him-
self “qualifies the thesis of man’s dangerousness as a ‘supposition,’ as an ‘anthro-
pological confession off faith ’” (¶21). As we shall see, Strauss shows that given his
understanding of human evil Schmitt could not rule out the possibility of man
becoming tame in the future (¶26). This means that the attempt to end all warr
does not need to take place through a “war against war.” Schmitt’s argument for
the inescapability of the political presupposed the existence of at least one politi-
cal opposition (between pacifists and nonpacifists), but this opposition could dis-
appear because nonpacifists can gradually disappear (¶22). Man may escape his
political situation. Hence, the political is important to Schmitt neither because
he recognizes its reality nor because he recognizes its necessity, but because he
recognizes its desirability. He does not merely posit the political but says yes to
it (¶23).
Why does Schmitt affirm the political? In other words, why does he affirm
the dangerousness of man? Schmitt argues that political actions are motivated by
existential as opposed to normative considerations. But, Strauss argues, one can-
not politically affirm the political: the dangerousness of man cannot be affirmed
existentially, because one cannot wish one’s enemies to be dangerous (¶ 24). It
can be affirmed only morally. Strauss discerns two different moral sentiments
that inform Schmitt’s affirmation of the political, and which generally character-
ize the position of the Right: warlike morality and authoritarianism. One can
affirm the dangerousness of man because one honors men who can defend theirr
country; one can also affirm it because it renders authority necessary.20 Strauss
argues that Schmitt’s preference for authoritarianism over anarchism, and not
his preference for bellicose nationalism over international pacifism, is the true
motive for his affirmation of the political, partly because Schmitt raises the ques-
tion of the dangerousness of man in the context of a world state that does not
face threats from the outside (¶25). But why does Schmitt prefer authoritarian
political theories to anarchistic theories?
Strauss does not explicitly raise this question, but he does implicitly answerr
it in the course of a criticism of Schmitt’s effort to ground authoritarian politi-
cal theory in a thesis about human evil. For Schmitt “the quarrel between the
authoritarian and the anarchistic theories concerns whether man is by nature
evil or good. But ‘evil’ and ‘good,’ here are ‘nott to be taken in a specifically morall
or ethical sense’ but are to be understood as ‘dangerous’ and ‘undangerous’”
(¶26).21 Strauss clarifies with the help of Schmitt the meaning of the two ways off
understanding human evil. An evil man can be either a defective human beingg
or a blameless man whose natural powers are harmful to others. The latter was
t e sense
the se se in which
w c seventeenth-century
seve tee t ce tu y thinkers
t e s such
suc as Hobbes
obbes and
a d Spinoza
Sp o a
S T R AU S S ’ S R E T U R N T O P R E MO DE R N PH I L O S OPH Y 123

understood human evil. Schmitt regards evil in the latter sense as a possible and
even the preferable basis of his political theory, while Strauss denies this pos-
sibility. Strauss shows this by first stating the reasons that led Hobbes to develop
this notion of evil: “Hobbes had to understand evil as innocentt ‘evil’ because he
denied sin; and he had to deny sin because he did not recognize any primary
obligation of man that takes precedence over every claim qua justified claim,
because he understood man as by nature free, that is, without obligation” (¶26).
This understanding of human evil is compatible not only with liberalism but also
with anarchism: “Once one understands man’s evil as the innocent ‘evil’ of the
beast, but of a beast that can become astute through injury and thus can be edu-
cated, the limits one set for education finally become a matter of mere supposition”
(¶26). Schmitt then had attempted to ground his authoritarian political theory on
a view of man that actually supports the position of his opponents.22 Accordingg
to Strauss, Schmitt’s attempt at a radical criticism of liberalism requires him “to
return to the view of human evil as moral baseness or depravity; only in this way
can Schmitt remain in harmony with himself if indeed ‘the core of the political
idea’ is ‘the morally demanding decision’” (¶26). Strauss’s use of this passage from m
Political Theologyy makes it clear that the notion of human evil as moral depravity is
necessary not because it preserves the thesis of man’s dangerousness but because it
restores the connection between politics and morality that Schmitt had discerned
in that earlier work, a connection that apparently had become obscured by the
time he wrote The Concept of the Political. Insofar as authoritarian governments
are necessary to correct the moral depravity of man, Schmitt’s affirmation of the
political, his choice of authoritarian political theory over and against liberal and
anarchistic ones, is rooted in his concern with morality.
Indeed, Strauss makes this point explicit by noticing the moralizing characterr
of Schmitt’s “polemic against the ideal that corresponds to the negation of the
political,” an aspect of his polemic that Schmitt tries to conceal (¶27). I invite
the reader to see how Strauss in two passages from Schmitt discerns his moral
condemnation of that ideal and the method by which he conceals it (¶27). Here
I will address only the substance of his condemnation. While the ideals of lib-
erals and anarchists may be high, their goal threatens the seriousness of human
life by creating societies in which entertainment becomes the object of human
attention: “it is impossible to mention politics and the state in the same breath
as ‘entertainment’; politics and the state are the only guaranteee against the world’s
becoming a world of entertainment; therefore, what the opponents of the politi-
cal want is ultimately tantamount to the establishment of a world of entertain-
ment, a world of amusement, a world without seriousness” (¶27). But why would
serious men in effect will a world without seriousness? Strauss finds an answer to
this question in “Schmitt’s description of the modern age as the age of depoliti-
cization” (¶28). Politics and the possibility of war have survived in the modern
age, but Schmitt notices that the matters over which men quarrel have changed:
theological matters in the sixteenth century, metaphysical matters in the seven-
teenth, moral matters in the eighteenth, economic matters in the nineteenth,
and technological matters in the twentieth. He notices a law governing these
c a ges: “the
changes: t e ‘tendency’
te de cy towa
towardd neutralization,’
eut a at o , tthat
at is,
s, tthee st
striving
v g to ga
gain a
124 NASSER BEHNEGA R

ground that ‘makes possible security, clarity, agreement, and peace’” (¶28). This
law receives its original motive in the desire “to avoid the quarrel over the right
faith.” But Strauss brings out the cost of this desire: “it is always possible to reach
agreement regarding the means to an end that is already fixed, whereas there
is always quarrelling over the ends themselves: we are always quarreling with
each other and with ourselves only over the just and the good (Plato, Euthyphro
7B-D and Phaedrus 263A)” (¶28).23 The goal of modernity is a peaceful order,
an order that is desired for the sake of humanity, but this goal requires human
beings to “relinquish asking the question of what is right; and if a human beingg
relinquishes that question, he relinquishes being a human being” (¶28).
If the price of peace is the abandonment of the question of what is right, the
price of seriously raising this question is the ignition of quarrels that follow from
the problems inherent in this question. Schmitt is an authoritarian but his fun-
damental concern is neither warlike morality nor physical or spiritual security:
“The affirmation of the state of nature does not mean the affirmation of war but
‘relinquishment of the security of the status quo’” (¶29). He turns away from the
security, comfort, and ease of civilization to nature in its purity so that out off
pure knowledge of that nature “‘the order of the human things’ can arise again”
(¶29).
We have followed Strauss as he moves from the surface claim that the political
is an inescapable necessity to the deeper but hidden claim that the political is a
demand of morality. We noted that he catches Schmitt moralizing while conceal-
ing that moralizing. The question naturally emerges as to the source of Schmitt’s
confusion. If the affirmation of the political is ultimately a moral affirmation,
how are we to understand Schmitt’s polemic against the primacy of morals overr
politics? The most obvious explanation is that by morality Schmitt only means
humanitarian morality, the very morality that negates the political (¶30). Schmitt
is in a way trapped by his opponents’ conception of morality. He is not trapped
in the sense that he cannot pass a moral judgment on humanitarian morality. He
does condemn it, though he conceals this condemnation. He is trapped because
he accepts its premise, namely, that all morals are essentially private matters (see
his contention that no norm can justify men killing each other). He tries not
to defend the political in moral terms because morality as he understands it is
incompatible with the political, which has “the character of transprivate obliga-
tion” (¶31). Morality as he understands it may recognize obligatory duties, but itt
cannot recognize obligatory duties to one’s community. Accordingly, Schmitt
has to give political obligation a different basis and a different meaning: “obliga-
tion cannot be conceived as such, as duty, but can be conceived only as inescap-
able necessity” (¶31). He is thus forced to repeat the crucial aspect of Hobbes’s
teaching regarding natural right, which he himself regarded as “the characteris-
tic presupposition of the ‘individualistic-liberal society’” (¶31). Strauss thus lays
bare the difficulty that led Schmitt to assert the inescapability of the political and
to conceal his moral judgment as soon as his subject forces him to stop main-
taining this assertion (¶31). Schmitt cannot recognize the moral authority of the
community over the individual, presumably because he agrees with Hobbes that
“mana iss by nature
atu e free,
ee, that
t at is,
s, without
w t out obligation”
ob gat o (¶26).
(¶ 6). If Schmitt’s
Sc tt s true
t ue motive
ot ve
S T R AU S S ’ S R E T U R N T O P R E MODE R N PH I L O S OPH Y 125

for the affirmation of the political is concern with the preservation of the serious
life (a life that recognizes obligations to one’s community), his affirmation of the
political is a substitute for a morality that is unavailable to him because he accepts
the basic premise of Hobbes’s political philosophy—politics as a prosthetic for a
mutilated morality.
Having explained Schmitt’s tendency to deny and to conceal the moral con-
cern informing his affirmation of the political, Strauss shows the price he pays forr
“the affirmation of the political in abstraction from the moral” (¶32). To affirm
the political in this way is to affirm a state of affairs in which men orient them-
selves, toward the “dire emergency,” toward war. He who affirms the political is
silent about the cause for which men fight and thus “comports himself neutrally
toward all groupings into friends and enemies” (¶32). Schmitt is against liberal-
ism’s will to neutralization, its tendency to put off decisions facing man for the
sake of peace, but “his eagerness for any decision regardless of content” presupposes
a kind of neutrality. His decisionism avoids deciding between different answers
to the question of what is right:

He who affirms the political as such respects all who want to fight; he is just as
tolerantt as the liberals—but with opposite intention: whereas the liberal respects
and tolerates all “honest” convictions as long as they merely acknowledge the legal
order, peace, as sacrosanct, he who affirms the political as such respects and toler-
ates all “serious” convictions, that is, all decisions oriented to the real possibility of
war.r (¶32)

Schmitt’s polemic against liberalism, perhaps like every polemic, becomes entan-
gled with its opponent: “the affirmation of the political as such proves to be
liberalism with a minus sign” (¶32).24
Strauss, nonetheless, discerns a rational provisional purpose for Schmitt’s
polemic against liberalism in his work: “The affirmation of the political as
such can . . . only prepare forr the radical critique of liberalism” (¶33). He refers
to Schmitt’s observation in Political Theologyy on Donoso Cortés: “He ‘despises
the liberals, whereas he respects atheistic and anarchistic socialism as his mortal
enemy’” (¶33). Men like Cortés and Bakunin turn their gaze away from liber-
alism so that they can look at the other as the enemy. By tolerating intolerant
positions while negating liberalism, the affirmation of the political paves the way
for this attitude. In Schmitt’s deeper view liberalism needs to be negated not
because it is the enemy, but because its compromises and its penchant for putting
off decisions through endless debates obscure the real field of the battle: “The
polemic against liberalism can therefore only signify a concomitant or prepara-
tory action: it is meant to clear the field for the battle of decision between the
‘spirit of technicity,’ the ‘mass faith that inspires an anti-religious, this worldly
activism’, and the opposite spirit and faith, which, as it seems, still has no name”
(¶33). Schmitt’s real enemy is the most extreme development of the spirit off
modernity, and he treats this enemy with extraordinary respect. For instance, in
“The Age of Neutralization and Depoliticization” (the second essay in Schmitt’s
text) he argues against liberals who deny to technicity its spiritual character. In
126 NASSER BEHNEGA R

this respect, he seems more interested in seeing the real issue facing man than
in defeating an opponent. Accordingly, Strauss argues that neither the polemic
against liberalism nor the affirmation of the political is Schmitt’s ultimate con-
cern: “His last word is ‘the order of the human things’” (¶33).25
But this is not always his last word. The polemic against liberalism, which
could only legitimately serve as a preparatory measure, sometimes becomes
for him the goal of his reflections. Similarly, his decisionism (liberalism with a
minus sign) is not simply a provisional position. According to Strauss, Schmitt
is confused, and he traces this confusion to his principle of understanding. His
entanglement with the polemic against liberalism “is no accidental failure but the
necessary result of the principle that ‘all concepts of the spiritual sphere . . . are to
be understood only in terms of concrete political existence,’ and that ‘all political
concepts, presentations, and words’ have ‘a polemicall meaning’” (¶34). Whereas
earlier he identified these as two principles (¶2), he now treats them as one: if alll
concepts must be understood in terms of their concrete political existence, and iff
all political concepts are polemical, all concepts are in fact polemical. The result
is a complete politicization and historicization of all concepts. Strauss observes
that Schmitt rejected this principle of understanding “in concreto . . . by opposingg
his unpolemical concept of nature to Hobbes’s polemical concept of the state
of nature; and he fundamentally rejects this principle by expecting to gain the
order of human things from a ‘whole knowledge,’” a knowledge that can be
gained only through a return to “pure and uncorrupted nature” (¶34). Despite
this movement, he could not completely free himself from this principle, which
Strauss notes is “bound to liberal presuppositions.” The natural freedom of man
and the consequent understanding of the state of nature that can guide man only
through being opposed comes to mind here. If Strauss’s suggestion is correct,
Schmitt was entangled in a polemic against liberalism because he failed to suc-
ceed in his attempt at a radical criticism of liberalism.
Strauss unambiguously endorses Schmitt’s launching the radical criticism off
liberalism, but he shows that Schmitt has failed to complete this task: “his cri-
tique of liberalism occurs in the horizon of liberalism; his illiberal tendency is
restrained by the still unvanquished ‘systematics of liberal thought’” (¶24). To
complete what Schmitt started one needs to gain a horizon beyond liberalism
and one needs to show that Hobbes’s establishment of liberalism in such a hori-
zon was somehow flawed. While Strauss sides with Schmitt against Hobbes, he
clearly regards Hobbes as the greater thinker. Whereas Schmitt failed to com-
plete his critique of liberalism in the horizon of liberalism, Hobbes “completedd the
foundation of liberalism” in a horizon beyond liberalism (¶35; emphasis added).
Hence, it is doubtful that one can undermine Hobbes’s position by exposing its
inconsistencies. The radical criticism of Hobbes has to address the truth of the
foundational premises of his thought. By bringing out the human cost of liber-
alism, Schmitt’s work makes it possible for us to look into these premises with
eyes not blinded by modern promises. His entanglement with liberal premises
has the virtue of identifying these premises and their power. Of these, nothing
is more fundamental and powerful than the thesis that man is by nature free,
t at is,
that s, without
w t out obligation.
ob gat o . Accordingly,
cco d g y, the
t e radical
ad ca criticism
c t c s ofo liberalism
be a s or o a
S T R AU S S ’ S R E T U R N T O P R E MODE R N PH I L O S OPH Y 127

return to premodern philosophy requires above all the restoration of the notion
of human evil as moral depravity (¶26). Only through such a restoration can a
consistent alternative to liberalism or modern philosophy emerge. This restora-
tion also bears on the cause of philosophy or rationalism. Others have noted that
this conception of morality, unlike that of Hobbes and Spinoza, can serve as a
bridge between philosophy and revealed theology, potentially allowing one to
walk (as opposed to leap) from one position to the other.26 The restoration off
the older conception of morality can serve as such a bridge because theoretical
disputes turn into serious quarrels only insofar as they are connected with moral
disagreements (¶28): the antagonism “between unbelief and belief, is ultimately
not theoretical but moral.”27 If “revelation or the polemic against revelation is
what makes the acceptance of classical politics impossible for Hobbes,”28 if pre-
modern philosophy’s apparent inability to deal with the problem of revelation is
the living justification of the modern break with the past, Strauss’s discovery off
a bridge between premodern philosophy and revealed theology may well be the
decisive insight that makes possible a return to premodern philosophy. But there
is also a more elementary and related consideration. If morality is indeed a force
in human nature, this restoration bears on the right of reason to rule man, forr
the freedom that is gained through abstraction from man’s nature is no freedom
but self-imposed slavery.

Notes
1. The original essay was published in 1932 in Archiv für Sozialwissenshaften undd
Sozialpolitik 67, no. 6. It is also available in GS–3 217–38. In 1965 Strauss attachedd
a translation by E. M. Sinclair with the title of “Comments on Der Begriff des
Politischen by Carl Schmitt” to Spinoza’s Critique of Religion. He permitted this
translation to appear in Carl Schmitt, The Concept of the Political, trans. George
Schwab (New Brunswick, NJ: Rutgers University Press, 1976). In 1965 he also
attached the original German article to Hobbes’ politische Wissenschaft. It is a testi-
mony to the importance of this article that it is the only one (in any language) that
Strauss published four times.
2. SCR R 31.
3. SCR R 28–30.
4. SCR R 30–31.
5. David Janssens has correctly noticed in Strauss’s unpublished writings from 1931––
1932 evidence of a development toward this reorientation, but he does not give due
weight to the decisive questioning that manifests itself first in the “Comments.”
Accordingly, he interprets “expression” to mean only “publication” (“A Change off
Orientation: Leo Strauss’s ‘Comments’ on Carl Schmitt Revisited,” Interpretation
33, no. 1 (Winter 2005): 100).
6. LSEW W 214–15.
7. JPCM M 453.
8. There are two translations of Strauss’s review of Schmitt: Sinclair’s (referred to in
note 2) and J. Harvey Lomax’s, which is attached to Heinrich Meier, Carl Schmittt
and Leo Strauss: The Hidden Dialogue, trans. J. Harvey Lomax (Chicago: University off
Chicago Press, 1995), 91–119. I have used both translations, occasionally modifying
them when appropriate.
pp p For ease of reference,, I have used pparagraph
g p numbers.
128 NASSER BEHNEGA R

9. This critique presupposes Schmitt’s very high opinion of Hobbes. In the


“Introduction” to the German edition of his Hobbes book, Strauss writes:Carl
Schmitt, in quiet unconscious opposition to Ebbinghaus, asserted in his essay,
“The Concept of the Political” ( . . . , 1927) that Hobbes is “by far the greatest and
perhaps the only truly systematic political thinker.” Schmitt’s judgment about the
greatness and significance of Hobbes, a judgment which corresponded to my feel-
ings or taste at that time, strengthened, understandably, my interest in Hobbes”
( JPCMM 453).
In the “Comments” Strauss observes that by 1932 Schmitt had changed his
assessment of Hobbes, though not sufficiently.
10. Meier, 43.
11. The opposition between political theology and political theory is not so clearr
in Schmitt’s writings because his conception of political theology is informed
by the modern notion of secularization: “all significant concepts of the modern
theory of the state are secularized theological concepts” (Political Theology, trans.
George Schwab (Chicago: University of Chicago Press, 2005), 36). For a defense
of Meier’s reading, see Anna Schmidt, “The Problem of Carl Schmitt’s Political
Theology” Interpretation 36, no. 3 (Summer 2009): 219–52.
12. Meier in several places acknowledges the profound tension between his reading
of Schmitt and Strauss’s text. Although “original sin” is “an expression neverr
once uttered by Strauss,” in Meier’s reading of Schmitt this notion is “the deepest
foundation” of the political (53). Meier notes that “Schmitt’s affirmation of the
political, in Strauss’s interpretation, is the affirmation of the ‘state of nature,’” but
in his own view “Schmitt denies that men ever existed or could exist in a state off
nature” (39, 49). Whereas Strauss traces Schmitt’s historicism to his acceptance off
liberal presuppositions, Meier traces it to his Christianity (83). Meier even goes
so far as to assert: “Leo Strauss’s critique ‘overlooks’ what is decisive for Schmitt”
(69). The quotation marks around the verb imply that Strauss deliberately dis-
torted Schmitt’s teaching by turning it into a philosophical doctrine.
13. Ibid., xvii.
14. Ibid., 8n. The quoted words are Schmitt’s.
15. Schmitt supported the publication of the “Comments,” helped secure a fellow-
ship for Strauss, and years later attached the “Comments” to the American edition
of The Concept of the Politicall (ibid., 8, 8n, 9n).
16. He thus can equate the question of the state with that of the order of human
things.
17. Schmitt, The Concept of the Political, 65.
18. Strauss, “Letter to Schmitt, September 4, 1932,” in Meier, 124.
19. In the course of this reasoning, Strauss in effect separates the true element off
Schmitt’s notion of the state of nature from its questionable element (the danger-
ous or evil character of man). The true element is the unpolemical account off
that state which he opposes to Hobbes’s polemical account. For Schmitt, the state
of nature is a political state and is composed not of “individuals but of totalities”
and “not every totality is an enemy of every other totality” (¶16). This part of the
account only shows that the political is possible but not necessary.
20. In a letter to Schmitt, Strauss sees a connection between bellicose nationalism
and authoritarianism: men are so resistant to obedience that “they can be unified,
only in a unity against—against
t other men” (“Letter to Schmitt, September 4,
1932,” in Meier, 125).
S T R AU S S ’ S R E T U R N T O P R E MODE R N PH I L O S OPH Y 129

21. According to Schmitt, authoritarian theories can assume that man is evil “in a
specifically morall or ethical sense” or in an amoral sense so long as they assume
that man is dangerous. Schmitt even seems to prefer an amoral understanding
of evil to a moral one. For instance, he discerns a connection between the theo-
logical notion of original sin and the political notion of evil, but he does not
like theological interference: it “generally confuses political concepts because it
usually shifts the distinction into morall theology” (Schmitt, The Concept of thee
Political, 65). He seems to prefer Machiavelli, Hobbes, and Fichte to Catholic
reactionaries because the former only assumed “the reality and the possibility off
the distinction of friend and enemy” (ibid.).
22. Schmitt’s “correction” of Hobbes’s view of evil may avoid the anarchistic poten-
tial of Hobbes’s notion but at the price of complete confusion. Whereas Hobbes
wants to combat human evil, Schmitt speaks of it “with an unmistakable sym-
pathy” (¶26). By encouraging men to be evil one may buttress the need for gov-
ernment, but, Strauss argues, such admiration is a kind of aestheticism, which
Schmitt criticizes as a path to the universal economization of life (¶8, ¶26), and
it is absurd in itself because it is an admiration for a defect, for man’s need to be
ruled (¶26).
23. Strauss omits the noble, which is mentioned along with the just and the good
in the passage from Euthyphro, which omission is justified by the passage from
Phaedrus as well as by Schmitt’s criticism of aestheticism. The noble is less con-
troversial than the just and the good. Note also that for Schmitt theological and
metaphysical issues are more controversial than moral ones, while Strauss seems
to suggest the opposite.
24. While his polemic against liberalism is in truth motivated by a concern for moral-
ity, his affirmation of the political as such leads to an even greater alienation from
man’s moral concerns, for peace is more akin to justice than war (see note 22
above).
25. This abbreviated phrase of Virgil is literally how Schmitt’s book ends, but Strauss
obviously means that the concern for the right order of human life is Schmitt’s
ultimate concern.
26. SCR R 204.
27. SCR R 29.
28. HCR R 26.
C APT R 7
CHAPTER

LEO STRAUSS ON THE ORIGINS OF HOBBES’S


NATURAL SCIENCE AND ITS RELATION TO THE
CHALLENGE OF DIVINE REVELATION

Timothy W. Burns

eo Strauss’s unfinished manuscript, Hobbes’s Critique of Religion (1933–34),1


L which Gabriel Bartlett and Svetozar Minkov have now translated into
English,2 belongs to the period that Strauss identifies as the crucial one in his
intellectual development, when he underwent what he calls, in the 1965 preface
to Spinoza’s Critique of Religion, his “change of orientation.” He moved at this
time from the premise that a return to premodern philosophy was impossible
to the recognition that such a return is possible. That change had as its founda-
tion the recognition, noted in his review of Carl Schmitt’s Concept of the Political,
that the modern attempt to see human evil as bestial and hence innocent evil,
was inferior to the starting point of Socratic dialectic, wherein evil is seen as
moral depravity.3 The unpublished and unfinished Hobbes manuscript from this
period provides us with some important, if provisional, results of the ten-yearr
study of Hobbes that Strauss undertook in his effort to understand the roots off
the peculiar situation at which the project of modern rationalism had arrived.
In it, Strauss argues that Hobbes’s positivist-phenomenalist science (“metholo-
logical materialism” or “hypothetical materialism”) is a result of the somewhat
unexpected realization that, as a working hypothesis, the omnipotent God has to
be assumed to be behind nature, rendering nature unintelligible. That is, Strauss
argues that according to Hobbes, the doctrine of an omnipotent God creating the
world ex nihilo at each moment—a doctrine that Calvin, as well as the Islamic
Kalâm, had come to adopt—had to be initially granted as a possibility, if revela-
tion were to be refuted. Twenty years later, in Natural Right and History (1953),
Chapter 5, Strauss offers what appears to be a different account of the origin off
Hobbes’s natural science; there he argues that it resulted from Hobbes’s recogni-
tion of the impossibility, for knowledge of anything, of conceiving of mind as
derivative from matter and hence subject to the f lux of mechanical causation,
132 T I M O T H Y W. B U R N S

as both Hobbes and some pre-Socratics had wished to conceive of mind. This
article examines the similarities and differences between the two arguments, and
attempts to explain why and to what extent Strauss abandoned his earlier argu-
ment in favor of the one he later articulated in Natural Right and History.

I
To better grasp the intention of the earlier work, it may be helpful to first considerr
Strauss’s remarks about its immediate predecessor, Spinoza’s Critique of Religion.
Both books manifest what Strauss later (in his 1965 German Preface to Thee
Political Philosophy of Hobbes)4 describes as his “philosophic interest in theology,”
an interest prompted by the failure of modern philosophic rationalism’s attempt
to liberate the West from biblical revelation and direct it by rational norms. As
Thomas Pangle has shown in his very helpful explication of that preface and off
the Strauss letter to Gerhard Krüger of January 7, 1930,5 Strauss understood his
Spinoza’s Critique of Religion to be a book on Heidegger, in two senses. First, it
was originally conceived from a position that was fundamentally Heideggerian
as regards the theological question: Strauss himself continued, as he tells Krüger,
to have the “will” (Wille) e and “stance” (Gesinnung) of the Enlightenment,
and Strauss understood that will or stance to have reached “its completion in
Being and Time—I mean in the interpretation of the call of conscience, and in
the answer given there to the question” of who is calling, that is, the atheistic
answer. Second, Strauss undertook an explication of the historical forms of pre-
Heideggerian philosophic atheism precisely in order to clarify the various
grounds of Heidegger’s atheistic predecessors; the resulting work clarified the
various modern versions of the “Epicurean” will, the “stance” (Gesinnung) that
he found merely presupposedd in, though meant to be vindicatedd by, Heidegger’s
Dasein-interpretation, a stance Heidegger himself had failed to bring to light.6
As Strauss explains in the letter to Krüger, the Enlightenment’s attempt to defeat
revelation depended on its refutation of the possibility of miracles. Summarizing
the findings of his Spinoza book on the achievement of the Enlightenment with
respect to miracles, Strauss claims that achievement to be limited to this:

that the already enlightened human being, is immune to miracles; [the Enlightement]
has created a position that is unreachable by miracles. But a miracle is, according
to its own meaning, only capable of being experienced as a miracle on the founda-
tion of faith—and thereby, the Enlightenment offensive is thus rendered impotent.
At this point, . . . it becomes clear that the Enlightenment does not owe its victory
to assertions of the scientific refutation of revealed religion. It owes its victory to a
certain will, which one may, with a grain of salt, characterize as Epicurean. This
will seems to me to be no foundational justification for the Enlightenment, against
revealed religion. . . . [I]n order for the social victory of the Enlightenment . . . to
become total, there must emerge another will against revealed religion. Such a will
I see disclosed in Machiavelli, Bruno, and Spinoza . . . reaching its most extreme
representation in Nietzsche, and its completion in Being and Time—I mean in the
interpretation of the call of conscience, and in the answer given there to the ques-
tion: who is calling? g It is in Heidegger’s
gg Dasein -interpretation
p that for the first
S T R AU S S , HOBBE S , A N D N AT U R A L S C I E NC E 133

time an adequate atheistic interpretation of the Bible becomes possible. Religion is


then for the first time overcome, when it can be given an adequate atheistic inter-
pretation. Therefore: the victory of the Enlightenment, that is, the victory of the
“scientific view of the world”—which I only understand to include the loss of the
possibility of believing in miracles—is justifiable only on the ground of a certain
intellectual stance (Gesinnung), not on the ground of this “worldview” itself.

The victory of the Enlightenment that had come about through the activity and
spread of modern natural science was not, Strauss had concluded, a victory that
was justified; not the findings of science, but the will and stance behind those
findings, were responsible for the limited victory, and the stance was unable
to sustain a justification of itself against the alternative stance—unable, that is,
to defeat the most radical position of revelation, that of Calvin, as Strauss had
brought out in Spinoza’s Critique of Religion (and as he reminds us, in footnote
references to Spinoza’s Critique of Religion, in Hobbes’s Critique of Religion).
Second, and relatedly, Strauss had come to recognize that the modern enlight-
enment project was failing. Here is how he puts his recognition of that failure
in the 1951 Preface to the American edition of The Political Philosophy of Hobbes,7
describing the perspective from which that book was written:

I had seen that the modern mind had lost its self-confidence or its certainty of hav-
ing made decisive progress beyond pre-modern thought; and I saw that it was turn-
ing into nihilism, or what is in practice the same thing, fanatical obscurantism.

As noted above, and as both The Political Philosophy of Hobbes and Philosophy andd
Laww reveal, Strauss had in this situation already worked his way out of the “pow-
erful prejudice” against a return to premodern philosophy. But he still sought
to unearth the fullest or strongest version of the modern attempt to dispose off
the theological-political problem, and in the process to understand the peculiarr
impetus behind modern (as opposed to ancient) natural science. And he did so
while simultaneously (in other researches and writings) clarifying ancient/medi-
eval rationalism. It is possible that in his long study of Hobbes Strauss might have
discovered a more satisfactory critique of miracles, and hence of revelation, than
he had found in Spinoza, or a better understanding of the relative success and
failure of the Enlightenment.

II
The opening of Hobbes’s Critique of Religion suggests that he may have. Strauss
claims that while Spinoza’s Theologico-Political Treatisee is bolder than Leviathan,
the boldness is “bought at the price of renouncing the proper foundation off
the critique [of religion], which is found much more in the Leviathan than in
the Theologico-Political Treatise” (HCR
R 23 / GS–3 267). For while Spinoza had
remained a metaphysical or systematic materialist, Strauss was persuaded by
Tönnies’s argument that Hobbes’s natural philosophy is nott “a materialistic meta-
physics” but instead the “foundation of modern natural science” (HCR R 24 /
134 T I M O T H Y W. B U R N S

GS–3 267–68), and hence that the critique of religion is not a “byproduct” off
Hobbes’s natural philosophy, as indeed Strauss himself seems to have thought it
to be when he wrote the Spinoza book.8 Hobbes’s critique is instead, initially, an
“integral feature” of his political science, and even a “presupposition of modern
politics” (HCR R 25, 26 / GS–3 269, 270). His critique of the philosophic politics
of classical antiquity, the denial of their acceptability, itself rests on the “polemic
against revelation” (HCR R 26 / GS–3 270). For since an effective political orderr
required the elimination of “fear of powers invisible,” Hobbes had to overcome
the Christian teaching on hell, and ancient philosophic politics, far from beingg
helpful on this score, had provided theologians with the “dangerous weapon” off
the doctrine of incorporeal substances that supported the teaching on hell.
But in undertaking his critique of revelation for the sake of an effective politi-
cal order, Strauss claims, Hobbes’s critique becamee the “presupposition of Hobbes’s
philosophy in general” or “the genuine founding (Grundlegung) of . . . his whole
philosophy” (HCR R 28, 30 / GS–3 272, 274). What Strauss means by this remark-
able claim emerges late in the text. Before explaining it, he attempts to account
for the fact that Leviathan seems to show that the very opposite is the case—that
the critique of revelation “rests on [Hobbes’s] elaborated philosophic teaching,”
that is, on a materialist physics. Against this, Strauss argues that the “structure off
Leviathan” conceals the “real foundation of the critique of revelation”; the critique
that we see in Leviathan is nott one that “sets forth his actual opinion unequivo-
cally.” Instead, Hobbes proceeds “by beginning with fully or mostly orthodox-
sounding statements, in order to lead these statements afterwards, in a more orr
less veiled manner, ad absurdum” (HCR R 28, 30, 32 / GS–3 272, 274, 277). And to
understand the true foundation of the critique of revelation in Leviathan, Strauss
argues, one must “extract the truly foundational elements of the critique of rev-
elation” (HCR R 30 / GS–3 275). Roughly the first half of Strauss’s manuscript
consists of this extraction. That is, there Strauss attempts to lead the reader in the
direction that Hobbes wished to lead the attentive reader. And that direction is
toward the manifestation of a deficiency in the surface argument of Leviathan, a
deficiency of which Hobbes was aware.
For example, the first part of Strauss’s argument (to HCR R 45 / GS–3 293) details
Hobbes’s critique of immaterial substances on the basis of Scripture, and hence
Hobbes’s critique of the possibility of suffering eternally in hell. But:

It . . . is . . . not to be doubted that it was fully clear to Hobbes that the denial of spirits,
taken by itself, would not suffice to secure the absolute unity of the civil power—
that is, the absolute exclusion of a spiritual power. For since Hobbes has to carry
out his critique of spiritualism on the basis of Scripture, he has to acknowledge the
dualism of God and creation as Scripture understands it, and therefore the possibil-
ity of miracles.” (HCR 45 / GS–3 292–93, emphasis added in the first phrase)

And while writing Leviathan Hobbes “could not exclude this possibility insofarr
as he did not openly want to renounce his belief in Scripture.” The result is that
while Hobbes provides a critique of spiritualism in order to secure an absolute
unity for the civil power, “[b]ecause this critique does not exclude the possibility
S T R AU S S , HOB BE S , A N D N AT U R A L S C I E NC E 135

of miracles, it has holes in it that can be filled in only with a special investiga-
tion” (HCR R 45, 46 / GS–3 293, 294). The whole critique of Scripture on the
basis of Scripture is, Strauss eventually says, “exercised only fictitiously and only
subsequently on the basis of Scripture, [but] in truth and originally on the basis
of philosophic presuppositions thoroughly independent of Scripture.” So while
Strauss goes through that critique, it “is not,” he states, “that critique of religion
which we identified at the outset as the presupposition of both Hobbes’s political
science and his philosophy in general” (HCR R 64 / GS–3 314).
What, then, is the latter critique? Strauss first identifies its “foundation” as
“Epicureanism”—by which he means not the doctrine of Epicurus “but rather an n
interest natural to man, a uniform and elementary outlook (Gesinnung), which
merely found its classic expression in the philosophy of Epicurus” (HCR R 64, 65 /
GS–3 315–16). So at least to this point in his work, Strauss has arrived at some-
thing like the position that he had concluded with respect to Spinoza in his firstt
book: whatever Hobbes’s genuine critique of revelation may be, the Epicurean
Gesinnungg is its foundation. According to Strauss, both the “will for the critique
of religion” and the “structure of the critique” is “predetermined by this stance,”
one that is revived generally in the seventeenth century and one that entails a
mechanistic determinism.9 “Hobbes’s critique of the religious tradition presents
itself as a post-Christian modification of Epicureanism” (HCR R 67 / GS–3 317),
that is, Socianism, a latitudinarian Christianity that was the unitarian univer-
salism of his day.10 Epicurean unbelief is likewise presupposedd in the “historical
critique” of Scripture that Hobbes uses to attack the knowability of revelation,
rather than being a consequencee of that critique (HCR R 75–81 / GS–3 327–34).
To attack those who believe in the revealed character of Scripture, Hobbes was
aware, and wished his attentive readers to be aware, he had to move from such an
attack on the knowability of revelation to an attack on its possibility. And at this
point in his argument Strauss refers the reader, in a footnote,11 to those pages in
Spinoza’s Critique of Religion where he had spelled out the inadequacy of Spinoza’s
attempted critiques of orthodoxy and of the unfathomable God. That is, Strauss
signals that Hobbes, who had become aware of these deficiencies, goes beyond
Spinoza in the reach of his critique, and Strauss aims in the rest of the book to
spell out in what way it does so.
Strauss accordingly moves from Hobbes’s critique of the tradition to his cri-
tique of the possibility of revelation (HCR 81ff. / GS–3 334ff.). Here again,
Hobbes does not “express” his denial of this possibility but “leads the readerr
to it” (HCR R 82 / GS–3 334). Hobbes begins, as he had in the critique of the
tradition, by “granting” the revealed character of Scripture. He presupposes the
impossibility of prophecy on the “conviction” that “it is impossible that God
speaks,” on the rejection of “inspiration,” etc. (HCR 82, 84 / GS–3 337, 338).
But this is not decisive concerning the possibility of revelation. And now (HCR R
85 / GS–3 338), Strauss states the point most clearly, and with the ground now
cleared, it seems, he lays out the issue to which Hobbes has led us:

If God is omnipotent and incomprehensible, one can indeed prove that human
statements about God’s activityy to be absurd,, but one can never refute the claim that
136 T I M O T H Y W. B U R N S

God’s activity is carried in a manner fully incomprehensibleh ble to man, that therefore
God in particular brings forth, in a fully incomprehensible, supernatural manner,
dreams and visions, that, in contrast to the natural products of the imagination,
have as their purpose and content the divine guidance of man. In other words, as
long as the presupposition of the incomprehensible omnipotence of God, as long as
the possibility of miracles is not shaken, the impossibility of prophecy and revela-
tion has not been proven. The critique of revelation leads further, therefore, to a
critique of miracles: the critique of miracles is the center of the critique of religion.

The footnote accompanying this statement again points us to passages in Spinoza’s


Critique of Religion, this time to passages arguing that the positive or scientific
mind rests on a “will,” not on an establishable progress of consciousness, at least
in the case of the Spinozistic Enlightenment.12 This is the point, in other words,
at which Spinoza’s positive critique of revelation undermined itself. Is the same
true of Hobbes’s critique? To answer this, Strauss now turns, in Part C of the
chapter (HCR R 85ff. / GS–3 339ff.), to Hobbes’s critique of miracles.
The central argument of the believer concerning miracles is that everything
that is is miraculous. Given God’s omnipotence, no distinction between super-
natural works of God and natural occurrences is even possible; since God can and
does do all that he wants, all “natural” occurrences are as incomprehensible as
supernatural ones. In other words, there is no nature, and there is no such thingg
as “cause,” properly speaking (i.e., necessity). In Hobbes’s effort to show that
miracles are not knowable for believers, he grants this presupposition as a pos-
sibility, according to Strauss, making it the foundation of his science. To make
his case, Hobbes “must, taking over the presupposition of his opponents, argue
on the basis of belief in the omnipotence of God,” and is thereby “compelled to
negate the possibility of natural knowledge.” Now Strauss argues that this clearly
does not “correspond to Hobbes’s real view,” which is that natural knowledge is
possible. This opinion Strauss calls “the genuine presupposition of his critique off
miracles.” Hobbes, he stresses, “had no reason for regarding natural science as
impossible.” But having made the “concession to his opponents” of an omnipo-
tent God, having had therefore skepticism “at first only imposed from without,”
Hobbes was led to a position in which the skepticism became “an integral ele-
ment” of his thought (HCR R 89–90 / GS–3 343–44).
For, to say that miracles are unknowable owing to the unknowability of an
omnipotent God is a critique that “leads only to Calvin’s teaching on mira-
cles” (HCR R 90 / GS–3 344), that is, to a teaching that Strauss had explainedd
in his Spinoza’s Critique of Religion as left unrefuted by Spinoza.13 Hobbes, hav-
ing arrived at this position, will move beyond it and thus beyond the failure off
Spinoza. But, Strauss argues, “this fundamental failure of his critique of religion
[to answer Calvin’s teaching on miracles] does not come to Hobbes’s awareness.”
In what appears to be a speculation on Strauss’s part as to the path that Hobbes’s
thinking took, he tells us that Hobbes’s initial apparent success brings him “into
an apparently hopeless situation, a situation of such a kind that he can believe that
in being freed from it, that he has once and for all put revealed religion behind
him” (HCR R 91 / GS–3 345).
S T R AU S S , HOBBE S , A N D N AT U R A L S C I E NC E 137

In support of this claim, Strauss reviews the steps that he sees Hobbes as hav-
ing taken. Hobbes wished to “attack revealed religion in general,” and so placed
himself “on the ground of belief in the omnipotence of God” (ibid.), adopting
his “opponents’ presupposition” that “everything which is has its ground in the
incomprehensible works of God,” and therefore, “there is nothingg to compre-
hend.” “In other words, in order to refute his opponents, he moves . . . to the
complete abandonment of the idea of naturee as a comprehensible orderr.” But in doingg
so, “Hobbes makes questionable, at the same time, revealed religion and natu-
ral reason.” This is the “hopeless situation” in which Hobbes found himself— —
presumably long before he wrote Leviathan. “How,” Strauss asks, “can he liberate
himself from this predicament, from this situation that seems completely hope-
less?” (ibid.)
Strauss then describes retreat into consciousness as the liberation, or at least the
beginning of the liberation, from the situation (HCR R 91–92 / GS–3 345–46):

He liberates himself from this situation, and at the same time from the power that
brought him into this situation, by withdrawing to a dimension that is removed
from the grip of God (and of a God who is thus not in fact omnipotent, or rather,
who does not make full use of his omnipotence). This dimension is the world of
consciousness, that is, a world that is as much of the material that is given to him
as of principles freely created by him. God may dispose of nature as he wants; in
the extreme case, he could even annihilate it, but insofar as I only remain, my
representations of nature remain, and with them the material and basis of science.
This material takes the form of science in being developed according to principles
that we ourselves create at will, which principles are thus to a higher degree in
our power than are the representations (which remain even through the fictitious
destruction of the world): even if nature is annihilated, the possibility of science
would survive so long as I survive, and insofar as the material of science (the ideas
given to us), as well as its form (the principles of knowledge created by us) are in
our power. But the possibility of a genuine natural science has not yet thereby been
vouched for. For the causes of the natural things that are sought for by this science
are not perceptible, and therefore do not belong to the world of our ideas, nor are
they created by us in the way in which the principles of knowledge are; they are,
therefore, in no sense in our power, but are simply in God’s power. Since nature, as
created by an omnipotent God, is beyond our grasp, natural science is possible only
in this way, that, starting from the ideas given to us in accord with the principles
of knowledge that we ourselves create, we arrive at the possiblee causes of natural
things in terms of these principles, without our ever being able to know, or need-
ing to know, whether the causes that we assume are possible are the real ones.

Modern, phenomenalist-positivist science, which understands the principles off


science as a construction of the mind, and understands the causes that it posits forr
phenomena as hypothetical, emerges owing to a concession to the unknowability
of the world, a concession made in order to overcome the challenge to knowledge
posed by the possible existence of the unfathomable will of an omnipotent God.
Strauss claims that “[t]his concept of natural science is the presupposition off
Hobbes’s thesis that miracles are, in principle, not knowable to natural reason”
138 T I M O T H Y W. B U R N S

(HCRR 93 / GS–3 347). As the footnotes indicate, Strauss bases this claim on
arguments made by Hobbes before and after Leviathan.14 In a footnote (HCR R
93n239 / GS–3 347n239), he explains why the presupposition is not found in
Leviathan:

Hobbes’s explicit justification of this thesis does not allow for an immediate rec-
ognition of his real view (see above, n. 222). Cf., in this connection, the following
paragraph.

The “explicitt justification of this thesis,” which Strauss had already spelled out (on
HCR R 85–87 / GS–3 339–40), is that Scripture tells us that miracles are know-
able not to everyone but only to the elect. After stating that the hypothetical orr
(merely) possible causes of “natural events” can or do serve as the “explanation”
of them and hence, too, of miracles, Strauss explains (in the subsequent para-
graph) Hobbes’s “civilizational” project. And it is this, he now indicates, that
allows Hobbes the final road out of his difficulty, and is something that Hobbes
does indeed indicate in Leviathan (HCR R 93–94 / GS–3 347–48):

The science that enables man to explain nature enables him at the same time
to explain “miracles.” Experience shows that the less natural-scientific knowl-
edge men have at their disposal, the more they are inclined to regard processes as
miraculous. In this sense it is true that miracles are only directed to the “elect”:
the “elect” are precisely those same poor in spirit who are without any scientific
culture. Hence, one can expect that, with the progressive cultivation of natural
science, belief in miracles will lose more and more of its significance, ultimately
disappearing entirely. For natural science is still in its beginnings, and gradually
even the unwise multitude will be educated and thereby become mistrustful of
reports of miracles that come from a dim past, that is, from an age in which there
was no science. Modern science, which excludes the possibility of miracles so little
that it rather has as its actual foundation a concession to that possibility, secures
itself subsequently against that possibility by claiming, on the basis of the con-
sciousness of progress that belongs to it, and thus on the basis of a historical ref lec-
tion, that belief in miracles is relative to the prescientific stage of humanity.

When Hobbes speaks in Leviathan (Chapter 37) about miracles being knowable
only to the elect, he is indicating there, according to Strauss, that the “elect” are
those without any scientific culture, and that the more men are enlightened by
the new science about possible causes of unusual phenomena, the less the multi-
tude will believe claims of miracles, and the more the belief in them will seem
to be something that “belongs to the prescientific stage of humanity” (HCR R
94 / GS–3 349). The “securing” of science is thus achieved through an initial
granting of the possibility that all things are miraculous, and the subsequentt claim,
“on the basis of the consciousness of progress that belongs to it,” that belief in
miracles belongs to a prescientific age.
Now—with the exception of the argument concerning the granting off
the unintelligibility of the world—Strauss had already, in Spinoza’s Critique off
Religion, made the argument concerning the awareness of the positive mind’s
S T R AU S S , HOBBE S , A N D N AT U R A L S C I E NC E 139

progress as the most effective critique of religion—muchh more effective, indeed,


than Spinoza’s systematic critique—but had there shown it to be inadequate.
(See above, notes 11 and 12). What he has now discovered, it seems, is that this
awareness, and the civilizational project upon which it depends for its authen-
tication, is, for Hobbes, the sole basis of the new science, and is the sole basis
precisely because Hobbes, unlike Spinoza, has granted to the believer the pos-
sibility that the world is the unfathomable work of what he calls, in Leviathan,
the “incomprehensible” God. And having granted this possibility, Hobbes grasps
what Spinoza denies or obscures—the degree to which modern rationalism rests
on the hope in progress of the human mind.

III
The final part of Hobbes’s Critique of Religion is a comparison and contrast off
Hobbes and Descartes, whose response, in the Meditations, to the Deus deceptorr
argument appears so strikingly similar to the argument, which we summarizedd
above, that Strauss has put together from works of Hobbes other than Leviathan.
Strauss promises that the comparison of Hobbes with Descartes will allow us “to
see the genuine basis of Hobbes’s critique of religion,” which is “by no means
the new science as such” (HCR R 94 / GS–3 348). What he seems to mean is the
following. No modern scientific finding of material causes, nor any reflections
on the right method of understanding causes, moved Hobbes to doubt the pos-
sibility of miracles. Rather, Hobbes “arrives at the new science by first carrying
out a response to the claim of miracles by revealed religion.” And this means that
there was already, as the “basis” of the response to revealed religion, a “primary
skepticism about miracles” (HCR R 95 / GS–3 348–49). This skepticism is, then,
the basis of the critique of religion. And Strauss links this primary skepticism m
to the medieval critics of religion and hence to the ancients (HCR R 95 / GS–3
349: “and antiquity”; “and classical”); he presents this “primary skepticism,” orr
what he also calls “the horizon within which” arguments against miracles “first
become possible” (HCR R 95 / GS–3 349), as something earlier than modern sci-
ence, pointing out that the evidence of this skepticism, that is, the thesis that
knowledge of miracles is practically impossible, was already held by medieval
critics of religion. The comparison and contrast with Descartes brings to the fore
this “pre-modern basis of the critique of religion,” and at the same time shows
“what is indeed the presupposition preceding the foundation of modern science”
(ibid.), that is, it shows the characteristically modern presupposition that distin-
guishes Hobbes’s critique of religion from the medieval and ancient one. Both off
these tasks, then, are achieved by examining Hobbes’s published (1642) response
to Descartes’s Meditations.
I wish to note only two things about this final section of the manuscript.
First, it clarifies the source of Hobbes’s adoption of the argument that the world
may be the work of an incomprehensible God; while that argument is in agree-
mentt with what Calvin came finally to argue, Hobbes did not learn the argu-
ment from Calvin, but from Descartes. Strauss shows that Hobbes acknowledges
t e Cartesian
the Ca tes a necessity
ecess ty of
o beginning
beg g philosophy
p osop y with
w t universal
u ve sa doubt, that
t at is,
s,
140 T I M O T H Y W. B U R N S

that the “foundation of philosophy” is “the regression into consciousness, into


the world of consciousness,” begun with the “fiction of the annihilation of the
world, with the retreat to the ‘ideas’” (HCR R 97 / GS–3 351). And, like Descartes,
Hobbes makes this retreat in truth only on the basis of the possibility of a Deus
deceptor. The “founding of science by means of a retreat into consciousness” is,
Strauss clearly suggests, something that Hobbes learns from Descartes (HCR R 98
/ GS–3 352). Hobbes at the same time rejects Descartes’s resolution of the Deus
deceptor difficulty (found in Paragraphs 10 and 11 of Descartes’s Sixth Meditation),
that is, he rejects Descartes’s claim that we can know through rational theology
that God is not a deceiver; Hobbes denies that the createdness of the world follows
from the positing of the existence of God as a first cause, and denies that God’s
truthfulness can be proven; God is, Hobbes argues, incomprehensible (HCR R
98–99 / GS–3 352–53).15
This is so, Strauss stresses, even if “genuine natural science” operates on the
basis of the existence of things, inquiring into their causes, or “does away with”
the “fiction of the annihilation of the world” (HCR R 101 / GS–3 354–55). Forr
insofar as it deals with the causes of our sense perceptions—the causes of the
“phenomena of nature” that are in our mind—natural science is dealing with
things that are “not in our power and are accordingly not accessible to any abso-
lutely certain knowledge.” We can deduce these causes “only according to the
principles of mechanics created by us ourselves” (HCR R 101 / GS–3 355). That
there aree things in themselves, however—that they exist—is for Hobbes “certain,
indisputable, self-evident.” Unlike Descartes, Hobbes does not doubt that the
world exists, but merely founds science on the fiction of its nonbeing or anni-
hilation. Hobbes even presupposes, Strauss argues, that the corporeality of the
things in themselves is “self-evident” (HCR R 102 / GS–3 356). For what exists
is known to be corporeal not through science but instead merely through the
“refinement” of the prescientific view according to which bodies, as distinct
from spirits, are beings that resist our force and are tangible or palpable. Science
merely recognizes that spirits, too, if they exist, have resistance and palpability
(HCR R 102–103 / GS–3 356–57). Understanding being as resistance and pal-
pability, Hobbes, unlike Descartes, holds the existence of things in themselves
to be self-evident; we “experience the world in the resistance that it offers us,
because we feel we experience it as independent of our thoughts” (HCR R 104 /
GS–3 359). Not a full blown Cartesian retreat into consciousness, then, nor any
denial that man is a rational animal, part of and bound up with the world, but
instead an abstraction from the existence of the corporeal world, or a retreat into
an independent consciousness “only in the context of considering the faculty off
knowledge,” characterizes Hobbes’s phenomenalism (HCR R 105 / GS–3 360, first
emphasis added). Our consciousness, an accident of a living body, is itself subject
to “the irresistibly greater power of the resistant world,” of undoubtedly existing
bodies (HCR R 106 / GS–3 361).
What all of this suggests is that Hobbes was, or was “originally,” a meta-
physical materialist, as Strauss discerns through the extraction of various passages
in The Elements of Law w and Leviathan, including claims about the corporeal-
ity of God. Hobbes’s confidence in this original metaphysical materialism was
S T R AU S S , HOBBE S , A N D N AT U R A L S C I E NC E 141

shaken, however, according to Strauss, “above all by Descartes, and he was


forced into ‘phenomenalism’” (HCR R 107 / GS–3 362).16 Hobbes understands
the Deus deceptor argument only as a pointed expression of “the possibility of a
fully incomprehensible God,” and merely a “symbolization of that possibility”
(HCR R 108 / GS–3 363).17 It is, in short, only the comprehensibilityy of the world,
and not its existence, that is a genuine problem for Hobbes. Hence, in his assent
to the Deus deceptor argument, “Hobbes . . . confronts, not the possibility of a
Deus deceptor,
r but the . . . much more threatening, and much more crediblee possibil-
ity that the world is the incomprehensible work of a simply incomprehensible
God” (HCR R 109 / GS–3 365).
Second, Strauss concludes by saying that “the retreat into consciousness is not
the foundation of philosophy” for Hobbes (though it appears to be the “founding
of science”). The reason is this: the retreat, as much as it may rescue a certain kind
of science from the possible incomprehensibility of the world, “cannot . . . help me
in a positive way with an orientation in a world that is completely incomprehen-
sible” (HCR R 109 / GS–3 364). It cannot tell one whether that science should be
practiced, since the “orientation” at issue is that of philosophic/scientific reason-
ing, on one hand, and faith in an omnipotent, incomprehensible God (the God
of Calvin), on the other. How then does Hobbes secure himself against such a
God, and thereby secure philosophy? Strauss has already provided an answer: the
consciousness of progress over and against the prescientific mind. What he does
next, in what happens to be the last section of the (incomplete) work, titled “The
Basis of Hobbes’s Critique of Religion,” (HCR R 109–14 / GS–3 364–69) is show w
what guidance Hobbes found or provided toward this orientation rather than the
other. The guidance proves to be “memory of art,” understood in a new way, a
way that emerges out of understanding the world as “resistance,” or understand-
ing man as the “resistor.”18

IV
“Hobbes’s fundamental critique of religion consists in his confrontation, carriedd
out on the basis of this presupposition, with the possibility that the world is the
work of a simply incomprehensible God and is therefore not only resistant and
overpowering but also fully incomprehensible” (HCR R 110 / GS–3 365). But “[a]
s a consequence of this possibility, every orientation in the world becomes radi-
cally problematic. In what fact does Hobbes find protection against this threat,
protection against the God of revelation?” (ibid.) Hobbes orients himself, Strauss
argues, not by the “fact of nature” but by the “fact of art,” since the works of art
are comprehensible in principle.” Withdrawal in defense against the possibility
that the world is the work of an incomprehensible God is withdrawal to “the fact
of art.” And, Strauss now repeats, Hobbes’s philosophy “is a philosophy of civili-
zation: it wants to contribute to the securing and the advancement of civilization
through knowledge of the conditions of civilization” (HCR R 110–11 / GS–3 366).
What characterizes civilization is that it has “at its disposal incomparably more
and more highly developed arts” than does barbarism. So the “orientation” is
“toward civilization and therefore toward the arts,” understood not as imitation
142 T I M O T H Y W. B U R N S

of nature but as the means by which self-interested human beings19 repel or over-
come nature (HCR R 111 / GS–3 366). Hobbes’s “ultimate division of being” is,
then, “what is by nature and what is by art.” But “art” understood as “the human
capacity to bring about useful effects on the basis of reflection” (ibid.).

The center of the critique of religion is the critique of miracles. The ultimate
presupposition of the assertion of miracles is the belief that God can simply do any-
thing, that the works of God are hence simply incomprehensible. Over and against
the threat emanating from this belief to the original security of his orientation
in the world, man finds his first protection in remembering that he understands
that which he himself produces, i.e., in remembering art. When man has secured
his orientation in the world by remembering art, he is in a position to take steps
against the claim of belief that there are miracles. (HCR R 112 / GS–3 367)

V
When we turn to the discussion of Hobbes in Natural Right and History, we seem
to find Strauss making a new argument: the skepticism about the knowability off
the world, and the modern scientific solution to it, is now presented as arisingg
out of a problem that attends any “materialist-mechanistic physics,” rather than
out of an acceptance of the possibility of an unknowable God who has created
the universe and who makes impossible any notion of “cause.” But there is reason
to think that this difference between the two accounts is less significant than it
might seem.
Strauss quickly establishes that while Hobbes was indebted to the tradition off
political philosophy for the view that political philosophy or political science is
possible or necessary, and that Hobbes sided with the apparently “idealistic” tra-
dition of political philosophy that public-spiritedly sought the “simply just social
order,” Hobbes attempted to combine this “political idealism with a materialistic
and atheistic view of the whole” (NRH H 167, 170). The novel result is that his
political science can be characterized as “political atheism” (NRH H 169). It is by
way of justifying this claim that Strauss turns to his account of Hobbes’s natural
philosophy, telling us that we must keep that philosophy in mind when trying to
understand Hobbes’s political philosophy (ibid.). Hobbes’s “natural philosophy is as
atheistic as Epicurean physics,” he states; it represents an attempt to combine that
“materialist-mechanistic” physics with a “Platonic physics” that is “mathematical”
(NRH H 170). And this synthesis, this mathematical physics, is one that Strauss rep-
resents as occurring through “the abandonment of the plane on which ‘Platonism’
and ‘Epicureanism’ had carried on their secular struggle” (ibid.). What Strauss
appears to mean by this is the struggle between materialistic atheism, on one hand,
and those that would claim there are incorporeal substances (ideas), on the otherr
(idealism). He later speaks of this as the “secular conflict between materialism andd
spiritualism” (NRH H 174). “Platonism,” in other words, means the positing of an
incorporeal mind (cf. NRH H 172), which for Hobbes was “out of the question.”
Before addressing the manner in which Hobbes conceived of the new
physics, Strauss first turns to the problem to which the new physics was a
S T R AU S S , HOBBE S , A N D N AT U R A L S C I E NC E 143

solution: the problem of “skepticism,” over and against the “actualization off
wisdom.” “Skeptical philosophy” had always accompanied dogmatic philoso-
phy, he argues, “as by its shadow” (NRH H 171), and was to be eradicated by
Hobbes by “doing justice to the truth embodied in skepticism.” And this was
to be achieved by first giving “free reign to skepticism” and erecting a new
edifice on the “foundation of extreme skepticism.” In a note to this argument,
which pithily summarizes the section on Descartes from Hobbes’s Critique off
Religion, Strauss invites us to “[c]ompare Hobbes’s agreement with the thesis off
Descartes’s first Meditation.” The skepticism to which Strauss is referring here
is, then, explicitly tied to the Cartesian argument of the Deus deceptor, even
though the problem is now being called the shadow of “skeptical philosophy.”
And Strauss has dropped the entire argument showing disagreement between
Hobbes and Descartes.
He then offers the reason—in Hobbes’s mind—for the persistence of the skep-
ticism that Hobbes set out to end. That reason is that the “predominant philo-
sophic tradition” was one of “teleological physics” (NRH H 172) and “the difficulty
with which every” such physics “is beset.” Strauss does not state what that dif-
ficulty is, but it seems safe to assume that he has in mind the lack of evidence
concerning a naturally ordered whole. Still, this difficulty for a teleological phys-
ics does not itself give rise to skepticism; to the contrary, it opens the way to a
“mechanistic physics,” one that had “never been given a fair chance to show its
virtues” owing to “social pressures of various kinds” (ibid.). Skepticism arises out
of, is “engendered by,” that materialist-mechanistic physics that Hobbes wishedd
now to make victorious. Strauss explains Hobbes’s understanding of that skep-
ticism as follows: Hobbes had “learned from Plato or Aristotle that if the uni-
verse has the character ascribed to it by the Democritean-Epicurean physics, it
excludes the possibility of any physics, of any science, or, in other words, that
consistent materialism necessarily culminates in skepticism.” Skepticism arises
out of the difficulty attending all materialist-mechanistic physics, namely, that
by understanding the mind as enslaved to the flux of mechanistic causation,
materialists in fact make the world unintelligible. In Hobbes’s Critique of Religionn
the problem of the intelligibility of the world had been presented as arising from
the need to grant the premise of revelation according to which the world is the
work of a mysterious, omnipotent, willing God. Here, however, the problem off
the intelligibility of the world is presented as arising out of a mechanistic physics,
and as something that Hobbes has “learned” from Plato and Aristotle.
But does this change represent a substantial or only a rhetorical difference?
And if the former, what precisely is its significance? It may help to note that
in the Aristotle chapter of The City and Man, where we receive Strauss’s final
presentation of the Deus deceptor argument, he indicates that the problem forr
philosophy that attends a mechanistic physics is the same problem as that of a
willful, omnipotent God.

We must reckon with the possibility that the world is the work of an evil demon
bent on deceiving us about himself, the world, and ourselves by means of the
144 T I M O T H Y W. B U R N S

faculties with which he has supplied us or, which amounts to the same thing, that
the world is the work of a blind necessity which is utterly indifferent as to whether
it and its product ever becomes known.20

While this formulation raises what appears to be a new concern—that is, off
whether the fundamental cause that has produced the world cares or is indif-
ferentt to “whether it and its product ever becomes known”—it also conflates,
as “amounting to the same thing,” the problem for philosophy posed by the
Deus deceptor and by mechanistic or blind necessity, that is, the problem of the
knowability of the world.21 In other words, the important challenge posed to
philosophy by the biblical God of revelation seems to represent only a version,
if one that is both more public and may have had other effects on the mod-
erns’ solution—of the problem already recognized by Plato and Aristotle.22 Not
Strauss’s understanding of Hobbes, but instead his understanding of Descartes,
has undergone a change. Strauss now reads Descartes’s theological arguments
as exoteric, and reads the Deus deceptor argument as having for its genuine
intention the posing of the problem of the intelligibility of a world ruled by
mechanistic-materialistic causation.
Still, it was indeed, historically, biblical revelation that had posed the problem
for Hobbes and Descartes, and in Natural Right and History Strauss considerably
mutes this fact. To be sure, he is not altogether silent about it: the discussion in
Natural Right and History of the turn to the new physics is begun, after all, as an
explanation of the claim that Hobbes’s politics is “atheistic.” And its descrip-
tion of the mind’s activity in consciously constructing the world as an act off
our “arbitrary will” or as “creation in the strict sense” (NRH H 173) sounds much
closer to the activity of the biblical God than had been even the description off
that activity given in Hobbes’s Critique of Religion. Finally, the account in Naturall
Right and History all but concludes with the striking statement that “the buildingg
of the City of Man on the ruins of the City of God rests on an unfounded hope”
(NRH H 175). Yet precisely as indications or suggestions, these are considerably
less forcefully presented than was the thesis in Hobbes’s Critique of Religion. What,
if anything, beyond the changed understanding of Descartes, is the cause and
significance of this change?
An explanation begins to suggest itself in the way in which Strauss describes,
in Natural Right and History, the relation between Hobbes, on one hand, andd
Plato and Aristotle, on the other. He appears intent on framing the argument
on Hobbes’s physics in such a way as to draw our attention to what Hobbes
“learned” and “did not learn” from Plato and Aristotle.23 To consider only the
first instance of these: Strauss claims (in a perhaps deliberately clumsy sentence,
or at least one with three negatives in it) that what Hobbes “learned from Plato’s
natural philosophy was not that the universe cannot be understood if it is not
ruled by a divine intelligence.” But since Plato’s dialogues may well lead read-
ers, and indeed did lead many Platonists, to the conclusion to which Hobbes was
nott led, we might wonder why Hobbes did not learn it from Plato. And given
Hobbes’s published doubt of Aristotle’s own belief in the Aristotelian doctrine off
S T R AU S S , HOBBE S , A N D N AT U R A L S C I E NC E 145

incorporeal or separate substances,24 it could well be that Hobbes did not acceptt
the doctrine of a divine, ruling intelligence as Plato’s serious teaching. Strauss,
from whom more than anyone we might expect such a hypothesis, appears
to immediately rule it out: Hobbes failed to learn this, he implies, because
“[w]hatever may have been Hobbes’s private thoughts, his natural philosophy
is as atheistic as Epicurean physics.” Given his atheism, Hobbes considered an
incorporeal mind to be “out of the question,” as Strauss says a little later on the
same page. In other words, Hobbes was simply closed to this as a possibility—not
as a serious possibility for Plato but rather in itself.
With the stress Strauss puts on what Hobbes learned and did not learn from m
Plato and Aristotle, we are led to wonder whether there were not other aspects off
their natural philosophy that Hobbes overlooked or too readily dismissed, and to
which Strauss is inviting our attention. That there were indeed such aspects, and
that they were even behind the Socratic turn to dialectics, that is, to the found-
ing of political philosophy and hence of the “tradition” of political philosophy
that Hobbes took for granted or overlooked (NRH H 170), is made clear in Strauss’s
much more frank discussion of the matter found in the remarkable final chapter off
The Political Philosophy of Hobbes. There he states in the following way the problem
with a mechanistic physics that drove Socrates to found political philosophy as
dialectics:

Anaxagoras and others had tried to understand the things and processes in the
world by their causes, by tracing them back to other things and processes in the
world. However, this procedure affords no possibility of true understanding. [The
accompanying note 3 here refers us to Phaedo 97b ff., where Socrates says that he
no longer thinks it possible to know if “2” is by division or by addition, or how
anything is.] Against this explanation of nature by the physiologists there is not
only the objection that it is an insufficient explanation or no explanation at all;
physics of the type of the Anaxagorean, “Epimethean” physics, which as such
takes—whether expressly and intentionally or implicitly and unintentionally is of
no importance—not the ordering power of reason, but disorder and irrationality
as the principle of nature, necessarily leads to the destruction of all certain and
independent standards, to finding everything in man’s world very well as it is, and
to subjection to “what the Athenians believe.” Confronted with this absurd conclu-
sion, Plato does not without further ado oppose to materialist-mechanistic physics
a spiritualist-teleological physics [the accompanying note 4 is to Phaedo 99c–e], but
keeps to what can be understood without any far-fetched “tragic” apparatus, to
what the “Athenians” say. (PPH H 142–43)

As Strauss goes on to argue in the same place, Hobbes misunderstood this Socratic
turn to speeches.
In Chapter 4 of Natural Right and History, Strauss likewise presents Socrates
as both founding political philosophy and as having never ceased his investiga-
tion into the nature of the beings. And the passage I have just quoted from Thee
Political Philosophy of Hobbes helps us notice that in the section of Natural Right andd
History immediately preceding the account of the Socratic founding of political
146 T I M O T H Y W. B U R N S

philosophy, that is, the conclusion of Chapter 3, Strauss had likewise referred to
the Protagorean “Epimethean” doctrine:

Epimetheus, the being in whom thought follows production, represents nature in


the sense of materialism, according to which thought comes later than thought-
less bodies and their thoughtless motions. The subterraneous work of the gods is
work without light, without understanding, and has therefore fundamentally the
same meaning as the work of Epimetheus. Art is represented by Prometheus, by
Prometheus’ theft, by his rebellion against the will of the gods above. (NRH
H 117)

Protagoras, whom Plato presents as ascribing to the Epimethean doctrine, is


therefore unable to do away with the possibility that what is held about the
gods by pious human beings is true. The Epimethean, materialist doctrine of an
unintelligible world and hence of the possibility that what religious believers say
concerning revelations of gods’s actions and intentions could be true, holds pre-
cisely the difficulty to which the Socratic turn to political philosophy had been
addressed. Hobbes’s founding of the new, modern phenomenalist physics, and off
its necessary task of enlightenment or civilizing the world,25 was undertaken in
ignorance (see NRH H 172: “as matters stood”) of this fact. Strauss’s own rediscov-
ery of the original purpose of the Socratic turn rectifies this situation, and hence
opens up the possibility of grounding the philosophic or rational life in a way
that does not suffer from the difficulty to which the Hobbesian grounding (the
progress of consciousness) has been shown to suffer.
Strauss’s intention in framing the argument of the Hobbes section of Naturall
Right and History in such a way as to draw our thoughts to what Hobbes learnedd
and did not learn from Plato and Aristotle has to be understood in light of the
intention of the work as a whole. From the first chapter, that intention appears
to be at its deepest level to address the argument raised by Heidegger againstt
the very possibility of philosophy. The fullest reason for the change that we
have traced might therefore best be sought in the digression (NRH H 175–76) that
appears in the Hobbes section that we have been examining. That digression
concludes with the first reference to Heidegger’s thesis about Dasein to be made
since Chapter 1. Strauss leads up to it as follows. After the statement concerning
Hobbes’s “unfounded hope” to build the City of Man on the ruins of the City off
God, Strauss speaks of how we, looking back on this attempt, find it difficult to
know how Hobbes could have been so hopeful about the overcoming of human
misery by means of the conquest of nature. The deepest reason for Hobbes’s hope-
filled error, and hence of the whole modern atheistic project, Strauss argues, is
that the “experience, as well as the legitimate anticipation, of unheard of prog-
ress within the sphere which is subject to human control must have made him m
insensitive to ‘the eternal silence of those infinite spaces’ or to the crackings off
the moenia mundi” (NRH H 175). That is, Strauss reminds us that the premodern
materialists, the Epicureans, were atheists of an apolitical sort nott because they
had never thought of combining mathematics and physics, but rather because,
unlike Hobbes, they allowed nothing to obscure from them the ultimate futility
of all human endeavors, nor to obscure the lack of significance of human things.
And as Strauss next argues, the hopes Hobbes and his contemporaries kindledd
S T R AU S S , HOB BE S , A N D N AT U R A L S C I E NC E 147

for overcoming human misery through human “construction” have not been
extinguished by the “long series of disappointments that subsequent generations
experienced.” The persistence of these hopes has relied, Strauss argues, on repeti-
tions of Hobbes’s obscuring of the vision of the cracking of nature’s walls, and on
a corresponding false estimate of the significance of human life. The first instance
of such post-Hobbesian obscuring, Strauss argues, is “the unplanned workings off
‘History,’” which had the effect of “enhancing the status of man and of his ‘world’
by making him oblivious of the whole or of eternity” (NRH H 175–76). As this
statement implies, what Strauss means by awareness of the whole or of eternity— —
at least in part—is an (Epicurean) awareness of the cracking of the walls of the
world, that is, the ultimate destruction of what is.26 This awareness was, more-
over, as he points out a little later, also part of the Socratic or “idealistic” tradition
prior to Hobbes.27 Strauss then concludes this brief digression by referring to a
doctrine of Heidegger as another manifestation of the hopeful, post-Hobbesian
obscuring of the whole, that is, of its ultimate destruction (NRH H 176):

In its final stage the typically modern limitation expresses itself in the suggestion
that the highest principle, which, as such, has no relation to any possible cause or
causes of the whole, is the mysterious ground of “History” and, being wedded to
man and to man alone, is so far from being eternal that it is coeval with human
history.

Dasein, by this account, is an expression of the turning away from eternity andd
hence entails an artificial “enhancing the status of man and his ‘world.’” Despite
the fact that Heidegger, in Being and Time, speaks much about death and beingg
toward death, the atheism of Heidegger, Strauss had concluded, is an atheism
achieved by the turning away from the ultimate decay of all things.28 It carries
within it the forgetting of eternity that had been begun by Hobbes. Strauss’s
new account of Hobbes’s founding of modern science is designed to highlight,
therefore, the manner in which Hobbes, unlike the classical political philoso-
phers, turned away from the eventual destruction of the world, and how this
was continued in various forms right through to Heidegger, who had himselff
undertaken a certain return to the ancients. The immediately preceding chapter,
on classic natural right, shows the ancient approach to the problem that Hobbes
confronted, an approach that begins with a concern for justice as our primary
stance or orientation in the world.29 That Strauss already speaks, at the time he
was composing Hobbes’s Critique of Religion, of writing a book on natural right,
suggests that had he completed the manuscript of Hobbes’s Critique of Religion, its
conclusion would have moved in the direction of the argument that we find in
Natural Right and History.

Notes
1. See Heinrich Meier, GS–3 ix–xii.
2. Hereafter referred to as HCR. All translations are from this Bartlett-Minkov trans-
lation. Page numbers in parentheses refer to this translation, followed by the cor-
responding page numbers to the German original in GS–3.
148 T I M O T H Y W. B U R N S

3. SCR, “Preface,” 19, and “Comments on Der Begriff ff Des Politischen


l h by Carl Schmitt,”
344–45. The original of this work was published as Die Religionskritik Spinozas als
Grundlage seiner Bibelwissenschaft Untersuchungen zu Spinozas Theologisch-Politischem
Traktatt (Berlin: Akademie-Verlag, 1930).
4. An English translation of this preface is available in JPCM M 453–56.
5. Thomas L. Pangle, “On the Strauss/Krüger Correspondence,” chapter 4 of the
present volume. I have relied on the translations of the Strauss-Krüger correspon-
dence made by Pangle in his chapter.
6. Both the inf luence of Heidegger and the critique of Heidegger are manifest in the
following passage, SCR R 214: “Just as the assertion of miracles is called in ques-
tion by the positive mind, positive critique of miracles is called in question by the
mind that waits in faith or in doubt for the coming of the miracles. The weapon
which the positive mind believes it has discovered in the fact that the assertion off
miracles is relative to the pre-scientific stage of mankind, is taken away from the posi-
tive mind by the observation that this fact permits the oppositee interpretation. Is
the willl to ‘establish,’ which needs only to have become victorious for experience
of miracles to become impossible, itself something to be taken for granted? Does
not man come to his most weighty and impelling insight when he is startled outt
of the composure of observation by which facts are ‘established,’ when he finds
himself in the condition of excitation, in which alone miracles become perceptible
at all?”
One sees here the raising of Heideggerian arguments, not only in the denial
of the depth of the everydayness of facts but in the reference to the agitated zone
between life and death in which there is an intensification of experience that
permits an “event” to happen. But as Strauss here suggests, the atheism that one
sees in Spinoza, and that Strauss understands to have reached its culmination in
Heidegger’s Dasein interpretation of the call of conscience, the atheism of the
positive or scientific mind, is something that Strauss understood to rest on a
“will,” not on an establishable progress of consciousness.
7. A work that, as HCR R 75n11 makes clear, Strauss had already completed as he
continued composing HCR.
8. See SCR R 90: “He does not contend against religion, but against unmethodical
seeking after causes. The anti-religious implication is not primarily intended. His
critique of religion is not the object, but only the subsidiary result of analyzing
and defining science.”
9. Strauss goes through the features of this outlook at HCR R 65–67 / GS–3 316–17:
Fear of gods and of death prevent man’s happiness and can be banished by the
science of nature alone, which demonstrates that the fear is based on ignorance
of nature. Science comprehends all of nature as necessitated or determined, not
arbitrary or chaotic, and hence nature is not in need of gods. Nature is without
riddle or secret in principle. “Thus only corporeal substances could be acknowl-
edged as substances, and only local motions as alterations: the Epicurean out-
look demands a mechanistic-corporealistic physics” (HCR R 66–67 / GS–3 317).
This is followed by an account of Socianism as the outward, exoteric version off
Epicureanism.
10. HCR R 67ff. / GS–3 317ff. This “modification” is according to Strauss prepared by
Epicurus himself, who argues for following the myth of a beneficent God ratherr
than “to be a slave to the fate of which the physicists speak” (HCR R 67 / GS–3
318). This “slavery” seems to be what, in Chapter 5 of NRH, Strauss concentrates
on as the problem
p that led Hobbes to pphenomenalist science.
S T R AU S S , HOB BE S , A N D N AT U R A L S C I E NC E 149

While Hobbes outwardly adopts all kinds of things from the Socinians, his
“critique of the tradition . . . is based on . . . an Epicureanism that ventures into the
light only under the cover of Socinianism” (HCR R 72 / GS–3 322). Nevertheless,
Strauss indicates that Hobbes did take something of importance away from
Socianism (HCR R 70 / GS–3 320): “From Socianism, accordingly, Hobbes came
to understand the hope for immortality in the true Epicurean way of thinking as
a simple guarantee against the fear of death, and not primarily as the reproach-
ful reminder of man’s duty and guilt. The presupposition for this conception off
immortality is that the significance of God’s punitive justice, if it is not denied in
general, at any rate recedes behind his mercy.” That is, the link between justice
and the moral life and the longing for immortality was severed.
11. The footnote reads “Cf., in this connection, Spinoza’s Critique of Religion, 144f.,
204ff. [in the English translation].” SCR R 144f. is the conclusion of Chapter 5, “The
Critique of Orthodoxy.” Here Strauss states that the “positive mind” (moved by
empirical and positive science) can be said to lack an organ, to see less, than the
believer. It succeeds only by mockery of belief, not by successfully defeating the
believer in argument. At SCR R 204ff., part of the section called “The Critique off
Calvin,” Strauss presents it as impossible for the positive mind to offer a sufficient
critique of the unfathomable God.
12. The footnote directs us to SCR R 212ff. There, in Section E of “The Critique off
Calvin” chapter, Strauss states the following: “[T]he basically problematic char-
acter of [Spinoza’s critique of revealed religion] . . . becomes most plainly manifest
when brought face to face with the position upheld by Calvin.” That position,
the position of “revealed religion,” has as its “central assumption” that “God is
unfathomable will.” Strauss’s claim is that “[p]ositive critique finds itself face
to face with this central assumption in particular when it contests the reality off
miracles.” The faith of the one who makes the assumption that God is unfath-
omable will is characterized by “Trust in God, obedience to him,” and this trust
discerns “in each cosmic process (not only in the stirrings of the human heart),
the hand of God at work,” and does not distinguish between miracles and nature
(SCR R 213). Strauss includes as part of “positive critique” of miracles the argument
concerning a progressive development of consciousness, and indicates that that
argument dooms the positive critique, in the end. He initially suggests, at SCR R
212, that this attempt of positive critique, as opposed to “systematic critique,”
cannot refute the possibility of miracles. But then, at SCR R 213, he argues that in
fact positive critique has been the most effective critique:
Positive critique does not merely prove that miracles are not knowable for
the unbelieving understanding. It simultaneously detects, by virtue of the
self-consciousness of the positive mind, the relativity of the accounts of
miracles to the pre-scientific “vulgar” stage of mankind. But is the assertion
of miracles not more completely undermined by this than by any fruitless demon-
strations that miracles are not possible?? When one considers the final result
from all the efforts made in the course of seventeenth and eighteenth-
century critique of miracles, one cannot but conclude that positive cri-
tique of miracles, which at first sight appears to be so inconspicuous and
which does no more than inquire how miracles are to be recognized, is of
more enduring significance than the attempt, at first sight so attractive,
made in the metaphysics of the Enlightenment, to prove the impossibility
of miracles. Positive critique demonstrates that the positive mind, apply-
ingg precise
p observation and stringentg analysis,
y is incapable
p of pperceivingg
150 T I M O T H Y W. B U R N S

miracles. Previous to this, positive critique establishes that miracles must be


accessible to that mind if they are to be indubitably established.
But then Strauss pivots, and shows that positive critique ultimately fails in this
historicizing effort. He first raises an objection that Spinoza seems able to answer,
concerning the disposition of the positive mind as one that matches a disposition
described in Scripture (SCR R 214):
But are miracles—understood as primarily meant—“established”? Are
not miracles looked forward to, implored in prayer and supplication? In
Spinoza’s sense, one may say against this, that according to the testimony of
Scripture miracles were experienced also by those who did not await them
in trust and faith.
Strauss then points out that the apparent parallel adduced by Spinoza does not in
fact obtain: the unbelieving spectators in the contest between Jahweh and Baal,
cited by Spinoza, experienced “doubt,” but this is quite different from the dispo-
sition or moral attitude of the positivist mind (ibid.):
These unbelieving “spectators” were however not convinced by merely
seeing the miracle, but by a form of seeing that had a peculiar pre-
supposition. They see—after waiting, not in faith, but in doubt, in uncer-
tainty, to see whether the eventt announced will occur, by which the ques-
tion, “Jahweh or Baal” will be decided. Can the man who has understood
the meaning of this question even wish to “establish” anything?
What Strauss appears to be pointing out is that the positive mind does not have
uncertainty or doubtt at all; the Biblical doubter is one who has presented prayerr
and supplication to Baal, after all. The doubter waits for the “event”; his disposi-
tion is altogether different from that of the positive mind. The believer or even
the doubterr will therefore eventually turn the argument concerning the alleged d
development of the “self conscious mind” against the enlightenment (ibid.):
Just as the assertion of miracles is called in question by the positive mind,
positive critique of miracles is called in question by the mind that waits in
faith or in doubt for the coming of the miracles. The weapon which the
positive mind believes it has discovered in the fact that the assertion of mir-
acles is relative to the pre-scientific stage of mankind, is taken away from
the positive mind by the observation that this fact permits the opposite
interpretation. Is the will to “establish,” which needs only to have become
victorious for experience of miracles to become impossible, itself some-
thing to be taken for granted? Does not man come to his most weighty and
impelling insight when he is startled out of the composure of observation
by which facts are “established,” when he finds himself in the condition of
excitation, in which alone miracles become perceptible at all?
13. In response to a point made by Krüger in his review of SCR, Strauss here explains
in a footnote that what he had presented as “Calvin’s teaching on miracles” is only
the “final outcome” of Calvin’s argument, and that it is only “as to this result” that
“Hobbes agrees with Calvin.”
14. The notes to this paragraph of HCR R (¶¶ 236 and 237, with reference also ahead to
notes 238 and 283, and thence also back to note 265) quote sections of De Corporee
and De Hominee and De Civee and Elements of Law. Elements of Law w was published
in 1640, De Civee in 1642; De Corporee was begun in 1642, even if it was not pub-
lished until 1665. Leviathan was published in 1651. Questions Concerning Liberty,
Necessity, and Chancee was published in 1656. De Hominee was not published until
1657. Later, Strauss will make reference to Hobbes’s objections j to Descartes’s
S T R AU S S , HOB BE S , A N D N AT U R A L S C I E NC E 151

Meditations
d . Those objections (and Descartes’s replies), along with objections off
and replies to six other prominent philosophers, were published in the first edi-
tion of the Meditations, published in Paris in 1641. (His objections and Descartes’s
replies are the third set.)
15. Following the argument of Tönnies, Strauss initially and tentatively (see Note
263) makes the case that Hobbes is a more radical “phenomenalist” than Descartes:
Hobbes in some places argues that only the objects of our mind’s representations,
and not the existence of corporeal things, can be known with any certainty. That
is, for Hobbes, “body” appears to be only a “positing of consciousness, of think-
ing”; even body characterized by extension is a “mere name,” or something that
“appears to” (but may not in fact) exist. If this is so, then while Hobbes appears
repeatedly to equate substance and body, that is, to be a “metaphysical material-
ist,” in fact his materialism would be only “methodical,” that is, it would be a
materialism that leaves open the question of whether sensibly perceived bodies
exist in themselves.
16. Only in his confrontation with Descartes, Strauss points out, does Hobbes
prove the existence of God by reasoning back from the “ideas” (in our minds)
to their ultimate cause; in all other places he reasons from “observed corpo-
real effects to their ultimate cause,” that is, God. Strauss takes this as decisive
evidence that “materialism,” not phenomenalism, is Hobbes’s “fundamental
conviction” (362).
17. Reckoning with the possibility of a God who actually wastes his time and energy,
as it were, deceiving human beings is not something Hobbes considered worth-
while; for God (understood as body), he thought, treats men with “complete
indifference.” And as we have seen, the world’s resistance vouches for its genuine
existence rather than its simulation.
18. Strauss opens this final section by declaring that the basis of the critique of reli-
gion “is not ‘phenomenalism’; the ‘phenomenalist’ thesis hardly appears in the
writings of Hobbes devoted to a critique of religion; it is in any case, of no
significance for his critique of religion” (HCR R 109 / GS–3 364). For the “pre-
supposition that Hobbes steadily makes use of in his critique of religion is ratherr
‘materialism,’ the monism of substances (see [HCR R 64 / GS–3 314).” Strauss then
argues that the “materialism” is also nott the “basis,” since “materialism,” as he had
already argued, is “the product of the scientific elaboration of the pre-scientific
concept of body” (HCR R 109–10 / GS–3 364). And that elaboration depends on
“the articulation of being into resistant and nonresistant (into ‘bodies’ and ‘spir-
its’).” Materialism, to be sure, does away with this “articulation of being,” by
eliminating from being the nonresistant, or “spirits” (HCR R 110 / GS–3 365).
Materialism cannot completely overcome the presupposition of this articulation
of being: the articulation into the resistant world, on one hand, and “we . . . who
assert ourselves against the world by acting” on it, on the other, or into the two
classes, “man and nature.” (There is no resistance without man there to do it.) So,
we can finally say, the presupposition of the critique of religion is the articulation
of being as man and nature, which permits materialism to be established.
Yet “‘phenomenalism’ is also present in Hobbes’s original presupposition”
(ibid.). For while being is articulated into man and nature, one part of that articu-
lation, “man,” is understood as a being with images or representations of being in
himself, or an inner world into which he can retreat. This means that phenom-
enalism, just like materialism, has its “origin in” the “original presupposition”
that beingg is man and nature, or rather, that “wewe men are in the power of a resistant
152 T I M O T H Y W. B U R N S

world,
ld but
b in suchh a way that
h we can withdraw
hd from
f this
h world ld into our inner world.”
ld (ibid.)
Strauss is now calling this “Hobbes’s fundamental presupposition.”
19. Strauss not only includes consciousness of self-interestedness (“sound common
sense”) and hence suspicion of alleged prophets, as part of Hobbes’s understanding
of “remembering art,” but points out a certain agreement on this score between
Hobbes, on one hand, and Socrates and the sophists, on the other (HCR R 111–13 /
GS–3 367–68).
20. CMM 43. In the full argument (CM M 42–43), which concerns Aristotle’s defense
of his view of the possibility of a happiness that is according to nature from the
assertion of the poets that the divine is envious of man’s happiness, Strauss begins
with the suggestion that the assertion that the whole is the work of an evil demon
or god—an assertion that he calls a modified version of the poetic assertion that
the gods bear malice toward us—relies upon an understanding of the human
good that lacks cogency: we allegedly cannot know what the good is, yet we
know that the end toward which we are by nature inclined (i.e., knowledge) is
bad. Strauss “therefore” examines the modern argument more carefully, that is,
because of the argument’s lack of cogency, Strauss does not think that this first
version can possibly be what the moderns were really saying. The second ver-
sion he arrives at in the following way. He first examines “the modern criticism
of Aristotle’s principle.” To say that they criticized it because they rejected final
causes is not sufficient, Strauss argues, since the ancient materialists had rejected
final causes as well but had not denied that the good life is life according to
nature, and they had claimed that nature provides the necessary things easily. In
place of this unsatisfactory argument concerning the moderns’ rejection of final
causes, Strauss therefore offers a somewhat surprising alternative: the moderns’
rejection of Aristotle’s principle begins by pondering what Aristotle himselff sug-
gests when he declares that our nature is enslaved in many ways: nature is a harsh
step-mother, or “the true mother of man is not nature.” This is something that
Aristotle and the moderns agree upon; it is not “peculiar to modern thought.”
What is peculiar to modern thought is, instead, “the consequentt resolve to liberate
man from that enslavement by his own sustained effort.” The argument to this
point has nothing to do with the problem of knowledge of the whole, or the need
to refute revelation or miracles.
But then how did this “resolve” to liberate man arise as a “consequence” of the
enslavement-to-nature argument, when it hadn’tt arisen for Aristotle from that
same argument? Strauss’s suggestion is again somewhat surprising. The resolve
shows itself, he states, in the “demand for the ‘conquest’ of nature: nature is
understood and treated as an enemy who must be subjugated.” And this “resolve,”
Strauss now suggests, is bound up with a Christian understanding of virtue:
“Accordingly science ceases to be contemplation and becomes the humble and
charitable handmaid devoted to the relief of man’s estate.” (Not just humble but
also charitable, and a “handmaid,” as science was to theology in the Christian
Middle Ages.) Strauss here describes the moderns as if they were runaway slaves
bent on the destruction of their former masters, the Christian theologians, still
possessed of the virtues inculcated by their masters but working now for the
good of all mankind, that is, for a different master. In any event, by this new
telling, the moderns did not begin by rejecting the notion that our natural end
is good because they accepted the possibility that the world is the work of an
evil demon. To the contrary: the moderns are now said to understand science
as beingg “for the sake of power,
p i.e., for putting
p g at our disposal
p the means to
S T R AU S S , HOB BE S , A N D N AT U R A L S C I E NC E 153

achieving our natural ends.” The notion of natural ends was still present, at this
point, then, but the whole business seems to be a project with a moral component
(“ourr disposal”). We still do not quite know, however, how this project acquired
its motivation. “Those ends can no longer include knowledge for its own sake;
they are reduced to comfortable self-preservation.” Strauss next explains why
not, and in explaining it, both spells out another or alternative manner in which
Christianity/the battle with Christianity was bound up in the motives behind the
moderns’ “resolve” to liberate man:
Man as the potential conqueror of nature stands outside of nature. This
presupposes that there is no natural harmony between the human mind
and the whole. The belief in such harmony appears now as a wishful or
good-natured assumption. We must reckon with the possibility that the
world is the work of an evil demon bent on deceiving us about himself,
the world, and ourselves by means of the faculties he has supplied us with
or, which amounts to the same thing, that the world is the work of a blind
necessity which is utterly indifferent as to whether it and its product ever
becomes known.
So now the presupposition of the conquest of nature is the Deus deceptor, r pre-
sented as the equivalent of the argument of a mechanistic materialism, a world
of “blind necessity.” A certain disappointment in a world whose cause is funda-
mentally indifferent to whether we achieve the knowledge that appears to be
our natural end, a disappointment that would only be possible if there were an
original hope that the cause behind the world somehow cares about our coming to
know it, seems in this account to have been behind the resolve to “liberate” man.
And to those affected by this disappointment, the ancients’ apparent assumption
that there is a “harmony between the human mind and the whole . . . appears now
as a wishful or good-natured assumption.”
But did the ancients even hold the assumption in question concerning the “nat-
ural harmony between the human mind and the whole”? The appearance here
that Strauss thought that they did is somewhat misleading. That harmony was
something Strauss had indeed presented, on 41, to be an Aristotelian doctrine:
“There is a natural harmony between the whole and the human mind. Man would
not be capable of happiness if the whole of which he is a part were not friendly to
him.” Yet he gives two versions of this optimistic doctrine. He first suggests that
it means that nature purposefully “supplies” man with food and his other wants.
But he soon after affirms that in truth, according to Aristotle, “the nature of man
is enslaved in many ways” (and, as we have seen, this includes the fact, disclosedd
on 42, that according to Aristotle nature does nott supply our wants or is “not a
kind mother but a harsh stepmother”). Between these two statements on 41 is his
claim that according to Aristotle, the world is the best possible world; that is, the
world could not be a better one because the evils with which it abounds could
not be removed without greater evils, so that “man has no right to complain and
to rebel.” The rosy version of the world as the best possible world is that nature
provides, the city is natural, and there is a harmony between the whole and the
human mind. In the less rosy version, “only a rare few, and even these not always,
achieve happiness or the highest freedom of which man is by nature capable.” Iff
even this version implies some kind of harmony between the human mind and
the whole, it does not seem to be a “wishful or good-natured assumption.” It is,
rather, one that entails resignation, especially to mortality, a resignation that may
well be needed precisely
p y for theorizing,
g or seeingg what we can see of the world as
154 T I M O T H Y W. B U R N S

it is but as it may well not be forever. The modern “resolve” to liberate man may,
then, in fact have been driven by a failure of resignation, a confused sense of a
“right to rebel” (against whom?) even though that resolve appears to be less naive
than the ancient position. On the Platonic-Aristotelian disposition toward ourr
“enslavement” to nature, or our being “playthings of the gods,” see also below,
note 27.
21. Strauss’s conf lating of the two appears to follow the argument of the last fourr
paragraphs of Descartes’s first Meditation, where Descartes presents as alterna-
tive to creation by an omnipotent God “fate or chance or a continuous chain
of events, or by some other means.” “Fate” seems to mean mechanistic deter-
minism; and “a continuous chain of events,” especially if one guided by noth-
ing but chance, would also suggest mechanism. Already in his 1959 lectures on
Nietzsche’s Zarathustra (6 of the transcript) Strauss had stated that Descartes did
not believe in the possibility of a Deus deceptorr, but was describing the problem off
the knowability of a world caused by mechanistic-material causes.
22. Strauss’s claim that these “amount to the same thing” is puzzling in that the
unknowabilty is in each case quite different. The difficulty in the case of a mate-
rialist-mechanistic physics is that its own account of causation is self-defeating. Iff
all human thinking is understood to be caused by a chain of blind material causes,
this very understanding is caused by such a chain, in the person who makes it.
And no mind would be moved to agree with the arguments of such a person by
the principle of noncontradiction or rules of logic, but would instead be moved
strictly by material causes. The doctrine itself would be the product of a series off
blind material causes, rather than a doctrine that could be true or could convince
another. The difficulty in the case of an omnipotent creative god, on the otherr
hand, is that there are no causes at all; all that is comes to be out of nothing, and
an oak tree may very well turn instantly into a jellyfish.
But there is this relation between the two doctrines: the unintelligibil-
ity resulting from the materialist-mechanistic physics—that is, from the claim
that mechanistic determinism is the universal and sole cause of change in the
universe—must if true undermine the trust we have in our minds to discoverr
the truth. And it is this that drives Protagoras to declare that what any and all
Athenians believe, is; or more generally, that “[each] man is the measure of all
things,” so that those who believe themselves to have had a revelation of gods
overturning necessities are as correct, their statements about the world as true, as
Protagoras’s materialism.
23. The first statement appears at NRH H 169–70:
[Hobbes’s] natural philosophy is of the type classically represented by
Democritean-Epicurean physics. Yet he regarded, not Epicurus or
Democritus, but Plato, as “The best of the ancient philosophers.” What
he learned from Plato’s natural philosophy was not that the universe can-
not be understood if it is not ruled by divine intelligence. Whatever may
have been Hobbes’s private thoughts, his natural philosophy is as atheistic
as Epicurean physics. What he learned from Plato’s natural philosophy was
that mathematics is “the mother of all natural science.”
The next two appear at NRH H 172:
For he had learned from Plato or Aristotle that if the universe has the
character ascribed to it by Democritean-Epicurean physics, it excludes the
possibility of any physics, of any science, or, in other words, that consistent
materialism necessarilyy culminates in skepticism.
p . . . On the other hand,
S T R AU S S , HOB BE S , A N D N AT U R A L S C I E NC E 155

what he had learned from Plato and Aristotle made him realize somehow
that the corporeal mind, composed of very smooth and round particles
with which Epicurus remained satisfied, was an inadequate solution.
24. See Leviathan, Ch. 46, Section 18, end.
25. “This implies that the whole scheme suggested by Hobbes requires for its opera-
tion the weakening or, rather, the elimination of the fear of invisible powers. It
requires such a radical change of orientation as can be brought about only by
the disenchantment of the world, by the diffusion of scientific knowledge, or by
popular enlightenment. Hobbes’s is the first doctrine that necessarily and unmis-
takably point to a thoroughly ‘enlightened,’ i.e., a-religious or atheistic society as
the solution of the social or political problem” (NRH H 198).
26. The footnote to this sentence brings out the point still more clearly. It includes
quotations from Friedrich Engels and from J. J. Bachofen. Here is a translation off
the Engels quotation:
[N]othing exists in the eyes [of dialectical philosophy] but the unbroken
process of becoming and passing away, the endless ascent from the lower to
the higher. . . . We need not here go into the question of whether this way
of looking at things agrees completely with the current state of natural sci-
ence, which predicts a possible end of the earth’s existence but a quite certain
endd of its habitability, which therefore also confers on human history not
only an ascending but also a descending. We find ourselves in any case quite
far from the turning point.
And the Bachofen quotation:
The East pays homage to the natural standpoint, the West replaces it with
the historical. . . . One could feel oneself tempted to recognize, in this sub-
ordination of the divine idea to the human, the final stage of the falling away
from an earlier, more elevated standpoint. . . . And yet this relapse contains
the seeds for a very important progress [Fortschritt]. t For we have to regard
as such that freeing of our mind from the crippling shackles of a cosmic-
physical view of life. . . . If the anxious Etruscan believes in the finiteness of
his race, the Roman takes pleasure in the eternity of his state, which he is not
even capable of doubting.
(The emphases in both quotations are Strauss’s.)
27. See NRH H 177 “No Scipionic dream illuminated by a true vision of the whole
reminds his readers of the ultimate futility of all that men can do.” Not only in
the Epicurean tradition, then, but likewise in what Strauss has been calling the
“idealistic” tradition of political philosophy, one finds a “vision of the whole” as
one of man’s ultimate destruction. On this point, and the resigned moderation
that results from such a vision of the whole, see also “The Problem of Socrates,”
Second Lecture, in RCPR R 133, concerning the “false estimate of human things”
as a “fundamental and primary error,” and the subsequent discussion of spirited-
ness in the Fourth Lecture (RCPR R 167). Consider also the following statement
from TWM 85–86:
Man has a place within the whole: man’s power is limited; man cannot
overcome the limitations of his nature. Our nature is enslaved in many
ways (Aristotle) or we are the playthings of the gods (Plato). This limi-
tation shows itself in particular in the ineluctable power of chance. The
good life is the life according to nature, which means to stay within cer-
tain limits; virtue is essentially moderation. There is no difference in this
respect
p between classical ppolitical philosophy
p p y and classical hedonism which
156 T I M O T H Y W. B U R N S

is unpolitical: not the maximum of pleasures but the purest pleasures are
desirable; happiness depends decisively on the limitation of our desires.
(The Plato statement can be found at Laws 709a1–3; cf. 644d7–e4 and 803c4–5.)
28. Consider in this light the very helpful remarks of Christopher Bruell, “Death
in the Perspective of Philosophy,” lecture delivered at the Carl Friedrich von
Siemens Stiftung, Munich, July 17, 2003, 12–16: Heidegger calls not those who
might be capable of it but rather “one and all” from the f light away from death,
from absorption in the every day; he is a “prophet of doom,” but one who also
holds out the promise of a Daseinsganzheit, that is, of achieving a wholeness off
being or fully existing in and through the proper stance toward death, and who
also concedes no legitimacy to our natural if unfulfillable concern with immor-
tality, to the point of tracing the belief in a prospect or image of an unendingg
time and of the wish to bring time to an halt, to the outlook of “decadent Being.”
And his account of death as “individualizing” us misses what genuinely does so,
and hence holds out the promise of a “Being-together” in a fated struggle with
others as some kind of authentic being. And, most importantly, while Heidegger
distinguishes empirical certainty of death from apodictic, theoretical certainty
of it, he shows little interest in attempting to acquire such apodictic, theoretical
certainty of it.
29. See note 10, above.
C APT R 8
CHAPTER

LEO STRAUSS ON FARABI, MAIMONIDES, ET AL.


IN THE 1930S

Joshua Parens

T he end of the 1930s marked a turning point in Strauss’s thought that Daniel
Tanguay has aptly called his “Farabian Turn.”1 Although Strauss’s most
extended work on Farabi, Maimonides, et al. in the 1930s is Philosophy and Law w
(1935),2 the turn Tanguay identifies is more evident in the smaller pieces that
came out in the wake of Philosophy and Law—including
w “Some Remarks on the
Political Science of Maimonides and Farabi” (1936)3 but especially “The Place
of the Doctrine of Providence according to Maimonides” (1937).4 In these two
pieces, Strauss evinces a growing awareness of the depth of Maimonides’s debt to
Farabi. Tanguay identifies two main features of the Farabian Turn: the focus on the
political in Maimonides and a growing awareness of the centrality of esotericism..
Prior to the Turn, in Philosophy and Law, Strauss elevated two features in his inter-
pretation of Maimonides that fade into the background as early as the two pieces
written in 1936 and 1937: the importance of Avicenna’s account of prophecy andd
Maimonides’s apparent reliance on supernatural resources for knowledge of things
beyond the limit of natural human knowledge. Beginning in 1936, Strauss avoids
any suggestion that the supernatural plays such a role.5
Our initial focus in this chapter will be on how “Place of Providence” pro-
vides a missing piece in Strauss’s effort to show that political science is central in
Maimonides’s Guide. Before the Farabian Turn, Strauss was bold in declaring this
centrality. And with the beginning of the Farabian Turn in 1936 and 1937, he
remains bold in declaring its centrality and Maimonides’s debt to Farabi regarding
its centrality. Subsequently, however, and as early as his (1941) “Literary Characterr
of the Guide for the Perplexed,”6 Strauss speaks far more cautiously about the role off
political science in Maimonides’s thought. Indeed in “Literary Character,” his main
reference to political science in the Guidee is to how “there is practically complete
agreement among the students of Maimonides” that the Guidee does not concern
it (PAW
W 44). Indeed, there are moments in the Guidee (end of II 39) as well as in
158 J O S H UA PA R E N S

Maimonides’s Logicc (Chapter 14, end) that appear to exclude any concern with
political science or philosophy.7 Strauss moves from boldly underlining the impor-
tance of political science in and for the Guidee in the 1930s to downplaying direct
reference to political science in connection with Maimonides. As Tanguay notes,,
after the Farabian turn Strauss decided more and more to emulate premodern eso-
tericism.8 That emulation includes writing in an increasingly esoteric manner about
the character and objectives of the Guide. Because of Strauss’s greater frankness in
his writings in the 1930s, they provide an indispensable window into understand-
ing Strauss’s lifelong interest in Maimonides and especially his Guide. Although his
insights into Maimonides deepened even into the last decade of his life, his earliest,,
at times ambiguous, insights were often proffered with the greatest frankness.
After our initial focus on the pieces on Farabi and Maimonides, we will turn
to consider Strauss’s writings from the 1930s on Isaac Abravanel. Strauss came
to see that Maimonides had been too readily assimilated to medieval Christian
philosophy.9 Because Abravanel’s approach to political philosophy was deeply
indebted to medieval Christian thought, Strauss came to see that Abravanel’s own
thought as well as his interpretations of Maimonides followed Christian pat-
terns. We will add Strauss’s understanding of Abravanel to our prior focus on his
Farabian reading of Maimonides to paint a picture of Strauss’s emerging under-
standing of Maimonides in the 1930s.
Arguably, the most striking insight overall in Strauss’s writings on Maimonides,,
Farabi, et al. of the 1930s is also the broadest structural insight: In Philosophy andd
Law, Strauss shows that the prophetology (Guidee II 32–48) is part of Maimonides’s
political science, and in the “Place of Providence” Strauss shows that his teach-
ing on particular providence (Guidee III 8–24) is also part of his political science..
In other words, Strauss argues that the central chapters of the Guidee are part off
political science.10 A potential source of confusion is worth nipping in the bud::
Maimonides states that after Guidee III 7 (the culmination of what every reader off
the Guidee readily acknowledges is the most elusive discussion in this most elusive
of books, the biblical Account of the Chariot [Guidee III 1–7]), “after this chapter,,
yyou will not hear from me even a single word about this subject.” Strauss interprets
this as Maimonides’s establishment of the end of his discussion of divine science orr
metaphysics in the Guide. Maimonides implies thereby that Guidee III 8–24 is not
a metaphysical discussion. Now if one overinterprets Strauss, one could suppose
that this implies that everything preceding Guidee III 7 concerns metaphysics. This
would obviously undercut the claim that the prophetology is part of political sci-
ence. It is one thing to say that henceforth I will not say anything more about x,,
and quite another thing to say that all I discussed previously was about x. And it is
obviously false to suggest that everything prior to Guidee III 7 concerns metaphys-
ics or divine science. In short, it is at least conceivable that both the prophetol-
ogy and the teaching on particular providence are part of Maimonides’s political
science.
Aside from the bold insinuation that the central arguments of the Guidee are
political, what is the significance of Strauss’s claim that these two sections are part
of political science? To answer this question, we begin by considering why one
might have assumed that these two sections were a part of metaphysics or divine
science.Throughout the Guide, the exact content of divine science is in contention..
L E O S T R AU S S O N F A R A B I , M A I M O N I D E S 159

After all, the Introduction to the First Part opens with the riddle that equates the
biblical or rabbinic mysteries, the Account of the Beginning and the Account off
the Chariot, with natural science or physics and divine science or metaphysics..
There is little agreement among scholars about what this means. It could mean
that (a) the secret teachings of the Torah are these philosophic theoretical sciences
(cf. Guidee I 71), (b) the secret teachings of the Torah are some revealed equivalent
to these philosophic sciences, or (cc) the secret teachings of the Torah are a blend orr
synthesis of philosophic and revealed teachings. Perhaps the most widely prevailingg
interpretation is some version of (cc). Strauss seems to have adopted some version off
(c)
c in Philosophy and Law.
In keeping with some version of (c), c one would think that the prophetology
would be a part of the Account of the Chariot or divine science because the central
role that the Active Intellect, the lowest of the separate intellects, plays in the recep-
tion of prophecy by the prophet. To say that the prophetology is part of political
science is to discount the centrality of the theoretical psychology of prophecy.11
Strauss does not discount completely that psychology in Philosophy and Law, but he
does begin a process of shifting more and more emphasis from the psychology to
political science. Similarly, one would think that Maimonides’s teaching on particu-
lar providence would be part of his divine science. After all, who but God is respon-
sible for particular providence? At this point, we may venture the conjecture that
Strauss’s break from his earlier stress on the role of the supernatural in prophetic
action and human knowledge of the metaphysical and his turn to this unexpected
approach to particular providence in “Place of Providence” are connected. As we
will see, Strauss’s analyses of prophecy and particular providence lead Strauss step by
step back to Plato by way of Farabi, not only in Philosophy and Law, Chapter 3, and
“Place of Providence,” but perhaps especially in “Some Remarks.”
In Philosophy and Law, Strauss begins the shift in orientation away from more
theoretical inquiries toward political science in his analysis of Maimonides’s Guidee
well before he gets to Chapter 3. In the Introduction, Strauss shows that what
makes Maimonides’s medieval enlightenment distinctive is that it, unlike the mod-
ern Enlightenment, takes as its “leading idea ...the idea of Law” (PLA 39). Because
the modern Enlightenment is oriented elsewhere, namely, the autonomous indi-
vidual or the subjective consciousness, interpreters like Julius Guttmann, whom
Strauss devotes so much attention to criticizing in Chapter 1, treat “philosophy off
religion” or “religious consciousness” as the central phenomena of all Jewish think-
ers, including Maimonides (PLA 73). Since at least Spinoza, political philosophers
have made every effort to transform religion into a part of man’s private or social,,
as opposed to political, experience. Strauss discovers in Maimonides’s concern with
Law or the divine Law a different focus upon something both public and total..
The Law, unlike contemporary religion, covers every aspect of life. Consequently,,
the Law proves to be the natural counterpoint to philosophy as a total way of life
(PLA 73). The more partial or less encompassing a religion’s demands become, the
less relevant they are as a counterpoint to philosophy.
In Chapter 1, Strauss identifies in reverse order the two main steps of his argu-
ment: (Chapter 3) the philosophic foundation of the law and (Chapter 2) the legal
foundation of philosophy (PLA 60). Already in Chapter 2’s focus on the legal foun-
dation, we can see the centrality of Law—and in a light that anticipates the focus off
160 J O S H UA PA R E N S

Strauss’s subsequent studies of Platonic political philosophy. Any reader of Strauss’s


writings on not only Plato but also Aristophanes will be reminded by Chapter 2
of the fate of Socrates in Aristophanes’s Clouds and Plato’s Apology and Phaedo..
Through the lens of the city, the philosopher appears to be a renegade. The Law is
the highest expression of the city’s way of life.
The main focus of the legal foundation of philosophy (Chapter 2) is sufficiently
difficult to detect in Maimonides that Strauss devotes little attention to it (cf. PLA
89–92, approximately three pages in Adler’s translation). Figuring prominently in the
middle of it is a declaration that we will not find after Philosophy and Law: “[H]uman
intellect has a limit which it cannot cross; for this reason man is obliged for the gloryy
of the Lord, to halt at this limit and not to reject the teachings of revelation that
he cannot comprehend and demonstrate. Philosophy is free—in its own sphere. Its
sphere is nature, not super-nature” (PLA 90–91).12 In every other respect, Strauss
associates Maimonides’s position with that of Averroes.
Strauss presents Maimonides’s legal foundation of philosophy as a mean between
Averroes’s defense of philosophy before the bar of the law in his Decisive Treatisee and
Gersonides’s oddly proto-modern claims to pursue philosophy in his Wars of thee
Lord.13 Averroes, unlike Strauss’s Maimonides in Philosophy and Law w (PLA 66–67),,
defends philosophy before the law by tacitly assuming that all of the relevant dog-
mas in the philosopher’s confrontation with the law are “accessible to the unaided
reason of man” (PLA 88).Yet Averroes is notoriously circumspect about how much
of the truth the philosopher should make available to the multitude. Rather than the
public teaching of the truth, he argues for the philosopher’s bondage to the public
teaching of scripture (ibid.)—a bondage that, by the way, Strauss identifies as the
main source of intellectual freedom in the premodern enlightenment. In contrast,,
though Gersonides appears oddly proto-modern in his calls for “the freedom of publicc
communication ...of philosophic truths” (PLA 96), his understanding of the preex-
istence of the Torah undercuts any possibility of discovering truths at odds with it
(PLA 100). Here, in Philosophy and Law, we already see hints of Strauss’s teachingg
regarding esotericism. Although his argument regarding philosophy’s bondage as
the source of its freedom (PLA 88) clearly anticipates his more fully developed
arguments about esotericism in Persecution and the Art of Writing,14 Strauss’s under-
standing of esotericism in Philosophy and Law w (PLA 102–3) has mixed in it anotherr
more widely recognized sense of esotericism, the esotericism of incommunicable
mysteries (PLA 66–67 and 95)—a sense that Strauss deemphasizes in his subse-
quent writings. Once again, Strauss’s stress in Philosophy and Law w on the supernatu-
ral element and the putative limitedness of human understanding of Maimonides’s
teaching undercuts his full embrace of esotericism.
We must renew our question as we turn to Chapter 3 of Philosophy and Law: How
and why does Strauss shift the emphasis in his account of the philosophic founda-
tion of the Law (i.e., the prophetology) from psychology to Law? Strauss answered
this question in outline in his critique of Guttmann’s approach to Maimonides in
Chapter 1:

[T]he philosophic foundation of the law, in spite of outward appearances, is not a


teac g among
teaching a o g others
ot e s but iss the
t e place
p ace in the
t e system
syste of
o the
t e Islamic
s a c Aristotelians
stote a s and
a d
L E O S T R AU S S O N F A R A B I , M A I M O N I D E S 161

their Jewish pupils where the presupposition of their philosophizing comes under
discussion. Now if they are following Plato in the philosophic foundation of the law,
this means that these philosophers are Platonists not because they follow this or that
Platonic theorem, however important—in this sense they are Aristotelians rather
than Platonists—but because in the foundation of philosophizing itself, they are
guided by Plato to answer a Platonic question within a framework laid out by Plato.
Ultimately, they differ from Plato only in this, though decisively in this: for them
the founder of the ideal state is not a possible philosopher-king to be awaited in the
future, but an actual prophet who existed in the past. (PLA 75)

The prophet is understood in the light of Plato’s philosopher-king. Even though


Strauss in Philosophy and Law w insists on the supernatural superiority of the prophet
vis-à-vis the philosopher-king (PLA 66), his analysis of the prophetology leads him
eventually to the striking admission that all or nearly all of Maimonides’s account off
the prophetology is natural rather than supernatural. And the remainder that is not
natural is merely the possibility that God might withhold prophecy.15 Maimonides
so likens the revealed understanding of prophecy to the philosophic understand-
ing developed by the Islamic falāsifaā (esp. in Guidee II 32) that Strauss is compelled
to acknowledge that for Maimonides “prophecy as such is natural” (PLA 104)..
Wary of the possible implications of this, Strauss presses Maimonides’s teaching,,
noting, for example, that in Guidee 2.38 he comes close to siding with Avicenna
in insinuating that the difference between the prophet and the philosopher-kingg
is that the prophet possesses “direct” knowledge of “the upper world” whereas
the philosopher possesses only “indirect” knowledge, that is, through the study off
nature (PLA 109 and 116). This account of the difference between philosopherr
and prophet is a far cry from the earlier insistence on the supernatural insight
provided by revelation. One of the most striking features of Strauss’s account off
Maimonides’s prophetology is the way he finds again and again that apparent dif-
ferences between Maimonides and the Islamic falāsifa ā prove ultimately to be just
that, merely apparent.
In the last two sections of Philosophy and Law, Strauss focuses intensely on the
exact sense in which prophecy is political rather than psychological. He frames this
opposition by reference to the opposition between the mantic role and the politi-
cal role of the prophet (PLA 120). Starting from the political center of the Guidee
(II 40), Strauss shows that Maimonides, by way of Avicenna and Farabi, character-
izes politics in a manner that cannot but remind the reader of Aristotle’s Politics
(Guidee III 6).16 Although Strauss no longer appeals to the modern Enlightenment
here, it is not difficult to see that the understanding of politics that he finds in
Maimonides17 is one that emphasizes not only the well-being of the body but also
of the soul, that is, of the intellect (PLA 124). In other words, Maimonides’s prophe-
tology culminates in a political science that stresses the perfection of the individual
intellect—which, of course, can be achieved only by the philosopher (cf. PLA 126
with Guidee III 34). In the final section, Strauss famously argues that the Law so
conceived enables those who live under it to “aristotelize” (PLA 133). Because they
are no longer beset as were Plato and Aristotle by the task of discovering the “ideal
state,” believers in revealed laws may now turn their efforts to philosophy. It is as iff
the Jew under the Law lives in the end of days as described by Maimonides in his
162 J O S H UA PA R E N S

Mishneh
h h Torah. h One wonders though whether discovering philosophy in the Law
or among its aims means that the Law has divined the insights of the philosophers..
Even though the perfect Law has been given, its perfection may not be fully appar-
ent without proper interpretation. Perhaps due to some residual influence of the
supernatural, direct knowledge and neo-Pythagorean mantic elements that Strauss
finds in Maimonides in Philosophy and Law w (PLA 129), he is prevented from fully
plumbing the depths of the Platonic political philosophy he has begun to uncoverr
in Philosophy and Law. Indeed, a certain overemphasis on a traditional understand-
ing of theoretical philosophy obscures the full depth of Strauss’s insight even into
the 1940s.18
We turn now to the “Place of Providence” to consider the political supple-
ment to the prophetology, Maimonides’s teaching on particular providence. In
comparison with his account of prophecy, Maimonides’s teaching on particularr
providence is extremely secretive. He is untroubled by stating openly that the
true account of prophecy is equivalent to the philosophic view with the sole
exception of divine withholding. In contrast, Maimonides’s own views on partic-
ular providence are, according to Strauss, revealed elusively under the guise of his
interpretation of Job (cf. PoP 544n20 with Guide III 23 and III 17). Although all
scholars acknowledge the surprising proximity of Maimonides’s views on proph-
ā , far fewer would assent to Strauss’s claim that
ecy to that of the Islamic falāsifa
even with respect to particular providence Maimonides proves to be a “rational-
ist” (PoP 544n20). It is likely that it was the very secretiveness of Maimonides’s
position on particular providence that makes “Place of Providence” and “Some
Remarks” such fruitful ground in the development of Strauss’s views on premod-
ern esotericism.
Scholars have sometimes complained about Strauss’s characterization off
Maimonides in the “Literary Character” as a mutakallim or dialectical theologian.19
As has already been mentioned, “Literary Character” is itself far more elusive about
the character and objectives of the Guidee than his writings of the 1930s—appar-
ently the closer Strauss got to the nerve of Maimonides’s argument, the more
uneasy he became about revealing it to his audience.20 In “Literary Character””
Strauss’s point is neither that the Guidee is devoid of philosophic content nor that
Maimonides never adopts the viewpoint of the philosopher therein but that its
exotericc surface is kalām (dialectical theology). Those who characterize the Guidee
as “Jewish philosophy” blend together the exoteric with the esoteric layers of this
multilayered work—thus obscuring each of the layers and their relation to one
another. In resisting the label “Jewish philosophy,”21 however, Strauss was deepen-
ing the insights of some of his predecessors. In “Place of Providence,” he underlines
his agreement with Jacob Guttmann and Shlomo Pines in observing similarities in
the structure of the Guidee with the Mu‘tazilite kalām (PoP 542n16). Now Saadya
Gaon’s (892–942 ce) debt to this kalām is well documented.22 Yet Maimonides is
so critical of the dialectical theologians as a group (though somewhat more spar-
ingly regarding the Mu‘tazilites) that we cannot but be somewhat surprised to
see that the structure of the Guidee is highly similar to what he criticizes. Unlike
Guttmann and Pines, however, Strauss underlines a crucial Maimonidean depar-
ture. The Mu‘tazilites are famous as proponents of the unity and justice of God.
L E O S T R AU S S O N F A R A B I , M A I M O N I D E S 163

Consequently, they divide their theological treatises into a first part on God’s unity
and a second part on God’s justice. The unity section covers God and the angels;;
the justice section covers prophecy, the Law, and (particular) providence. Strauss’s
key insight is that Maimonides’s philosophic or political approach to the prophetol-
ogy severs the connection between his account of particular providence and God’s
justice. In other words, Maimonides’s account of particular providence falls within
the ambit of political science rather than the Mu‘tazilite kalām teaching on God’s
justice (PoP 542). Although Maimonides imitates the Mu‘tazilite kalām in his exo-
teric teaching, even that imitation is superficial. The surface structure of the Guidee
is similar to that of the theologians, but its meaning and intention are opposed to
theirs. As Maimonides explains in Guidee I 71, albeit somewhat indirectly, he will
pursue a defense of Judaism that avoids the pitfalls of traditional kalām: Its main
pitfall was to defend a given faith using any means fair or foul.23 Above all, the
mutakallimūn were willing to violate the “nature of existence,” which nature was
discovered by the philosophers. Indeed, according to Maimonides, kalām originated
less as defense of one revealed religion against others than as defense of revelation
against philosophy (Guidee I 71).
Although Strauss highlights the opposition between the exoteric surface of the
Guidee and its esoteric depths in “Place of Providence” (PoP 545), he does not yett
insist as he does in “Literary Character” that Maimonides is a practitioner of “enlight-
ened kalām.” Rather, in “Place of Providence” Strauss speaks more boldly perhaps
than anywhere else in his oeuvre about the subject matter of the Guide. He takes farr
more literally than most interpreters the significance of what Maimonides identifies
in the Introduction to the First Part as its subject matter, namely, the “true science
of the Law.” According to Strauss, the theme of the Guidee is the Law, and the Law w
is analyzed from within the philosophic discipline of political science (PoP 548).24
Most interpreters view the true science of the Law as the study of the theological
foundations of the Law or of the fundamental articles of faith (cf. Pereq Heleq). In
contrast, Strauss views the true science of the Law as the political-philosophic studyy
of the aims of the Law. As we saw in Philosophy and Law, the highest aim of the Law w
studied by political science or philosophy is the philosophic aspiration to intellectual
perfection. It would appear then that the theological foundations stressed so heavilyy
by other interpreters reappear in Strauss as the enlightened theological surface off
the Guide.
We need to turn back momentarily to Strauss’s main focus in “Place of Pro-
vidence”: particular providence. According to Strauss, from Guidee I 1 to III 7, at
the end of which Maimonides announces his intention to cease theoretical discus-
sions of the Account of the Chariot, Maimonides at times discusses providence
but only or for the most part general providence (PoP 539n9). The discussion
of particular providence then is limited to nontheoretical or practical philosophy
(PoP 540). A crucial consequence of this is that discussions such as of the “doctrine
of divine reward and punishment” are part of Maimonides’s “exoteric doctrine””
(PoP 545– 46)—or his “edifying doctrine[s],” as opposed to his “fundamental doc-
trine” treated in the prophetology. One must be careful then not to leap to the
conclusion that Strauss’s stress on the political character of Maimonides’s teaching
implies that there is nothing but edifying surface without any depth. The political
164 J O S H UA PA R E N S

(as well as super-political) end(s) of the Law is (are), according to Strauss,“a true and
demonstrable fundamental doctrine” (PoP 548–49).
It may seem odd that we turn back briefly now to Strauss’s earlier “Some
Remarks.” After all, Strauss underlines an important development from “Some
Remarks” (1936) to “Place of Providence” (1937).25 Yet we do so because
“Some Remarks” reveals more clearly than “Place of Providence” the inter-
relation of particular and general providence in the thought of Maimonides..
Ultimately, Strauss argues for a shocking convergence of Maimonides’s teachingg
on particular providence with Laws 10: A convergence emerges even though
Plato moves from a more philosophic teaching toward a teaching oddly anticipa-
tory of the revealed teaching, and Maimonides moves in the opposite direction..
In both cases, the teaching on particular providence proves ultimately to be the
exoteric face of a philosophic teaching on general providence. Their different
contexts, however, demand opposite rhetorical strategies. We cannot consider the
details of Strauss’s investigation of Maimonides’s possible sources for the devel-
opment of his position on particular providence here. Let it suffice to say that
he shows that Maimonides either knew of or divined by himself the teaching off
Plato’s Laws 10 with the help of intermediaries such as Alexander of Aphrodisias
(SR 22–24). Strauss sets the stage for this linking of Maimonides to Plato by
offering what may be the first glimpse of the thesis of Persecution and the Art off
Writing. Although Strauss refers in Philosophy and Law w to esotericism and even dif-
ferences between modern and medieval types of esotericism (PLA 102–3), “Some
Remarks” hints at the likelihood that the philosopher, like the prophet, is subject
to persecution. Protecting oneself from persecution seems to be on the tip off
Strauss’s tongue here. Over the course of several pages (SR 18–21), Strauss shows
the striking similarities between the prophetic call to Israel to change its ways andd
return to God and the perfection of Zion, that is, the idea of the perfect city, on
the one hand, and Socrates’s divining of a perfect city in the Republicc as well as
his philosophic call to his fellow citizens to strive for virtue, on the other. How
ironic that the one who demands justice should be persecuted. Most notable off
all, however, may be that the return to Zion in the end of days is reimagined byy
Maimonides. Contrary to Jewish tradition regarding the Messianic age, the lion
does not lie down with the lamb—in other words, nature is not controverted. On
the contrary, the key natural divide among men is not undermined; it is in a sense
reinforced or deepened: “Only [when the Messiah has come will] . . . the privileges
of the philosophers be fully recognized” (SR 20).
With Strauss’s subtle adumbration of the differences between the prophetic,,
biblical view and the philosophic view still before our eyes, we turn to his analy-
sis of the political teaching of Isaac Abravanel (1437–1508 ce), which includes
Abravanel’s transformation of Maimonides’s political science. Because Abravanel
is deeply indebted to Christian political thinkers, including most notably Thomas
Aquinas, Strauss’s “On Abravanel’s Philosophical Tendency and Political Teaching””
(1937)26 enables us to deepen our understanding of the key differences between
Maimonides and the main line of Christian political thought. Should the readerr
wish to have a more nuanced understanding of Strauss’s interpretation of different
L E O S T R AU S S O N F A R A B I , M A I M O N I D E S 165

trends in medieval Christian political thought s/he will have to wait for his “Law
of Reason in the Kuzari” (1943).27 Be that as it may, two clear and striking opposi-
tions characterize the lines between Abravanel and Maimonides in Strauss’s 1937
piece: On the one hand, Maimonides is a “thoroughgoing rationalis[t]” (APT
104); on the other, Abravanel sides with the “‘mythical’ or ‘mystical’ tendencies
of the Midrash” (128). And, on the one hand, Maimonides adopts not only a
pro-monarchical interpretation of the Bible but also a political interpretation off
it; on the other, Abravanel adopts not only an antimonarchical interpretation but
also an antipolitical one. The former opposition is evidently simpler and more
straightforward. The character of Abravanel’s mysticism is connected with what
Strauss describes as “the fact that for him political philosophy loses its central
importance” (APT 104).28 It has a “much more restricted field than it had forr
Maimonides” (APT 105)—a field reminiscent of Aristotelian political thought,,
which, of course, takes the lead in medieval Christian thought in general (cf. APT
105 with 96–99).29 As we saw in Philosophy and Law, the characteristically modern
view of philosophy of religion as concerned with subjective religious conscious-
ness is more closely linked to this restricted view of political philosophy than the
ā and Maimonides. Rationalism, then, is closely
total teaching of the Islamic falāsifa
linked to (Platonic) political philosophy broadly understood.
The former opposition between Abravanel and Maimonides regarding the
status of monarchy and politics is a bit more complicated than the opposition
regarding rationalism and mysticism—though they are intimately connected, as
is already evident from our inability to avoid referring to politicall philosophy to
characterize that rationalism. The former opposition is more complicated because
Maimonides’s positive view of monarchy is characteristic of the vast majority off
the rabbinic tradition (APT 112, 119). Yet the biblical attitude toward politics
and the founding of the city is far closer to Abravanel’s view than Maimonides’s
(APT 129).30 Ultimately, Maimonides’s positive attitude toward monarchy and
politics is more deeply rooted in his philosophic orientation than the rabbinic atti-
tude toward monarchy. He, like Farabi and other Islamic falāsifa ā , favors monarchy
because it squares with his conception of Moses and prophets more generally as
philosopher-kings. In contrast, not only does human kingship fill Abravanel with
“disgust” (APT 117)—indeed, so deeply that he can be found speaking somewhat
favorably of a “‘mixed’ constitution”—but also Abravanel prefers the rule of judge
and prophet or priest to that of philosopher-king (APT 118, 126). In other words,,
Abravanel’s animosity toward monarchy and politics is rooted in the view that
“man’s ‘natural’ state” is most analogous to Israel’s life “in the desert, where Israel
had to rely entirely for everything on miraculous providence” (APT 110–11)..
Here we arrive at the opposite pole to the convergence of Maimonides and Plato
regarding particular providence we saw in “Some Remarks.” Maimonides’s ratio-
nalism places roughly as much stress on human self-reliance, as Abravanel’s mysti-
cism places upon faith in divine support.
In the 1940s, Strauss began to write about Maimonides in such a way that his
debt to Farabi in political philosophy became more difficult, not easier, to detect..
I do not mean that Strauss did not consistently acknowledge this debt, but he did
166 J O S H UA PA R E N S

little in writing to explain that debt, especially from the side of Maimonides. In
“Literary Character,” Strauss moved away from the emphasis in his 1930s writings
on the political character of Maimonides’s teaching to an emphasis on the enlight-
ened kalām surface of the Guide. At the same time, of course, Strauss began to make
the difference between exoteric surface and esoteric depth a central theme of his
writing. For those sympathetic to Strauss’s insights, then, it is not difficult to see
that he eventually argued that the Guidee consists in an enlightened kalām surface
and a political philosophic depth. To the unsympathetic, he appears to be makingg
Maimonides into a mere dialectical theologian. That is because of the unsympa-
thetic tend to overstate the importance of the theological aspect of Maimonides’s
writings—thinking of him primarily as a “Jewish philosopher.” Strauss insisted
rightly that interpreters would never understand Maimonides properly until they
understood the surface of the Guidee for what it was, namely, enlightened kalām. It
must be underlined that Maimonides learned from Farabi how to make and use a
kalām surface.31
Fortunately, among the many useful appendices included in this volume is a
translation of “A Lost Writing of Farâbî’s” (1936).32 This often dry explanation—off
how Strauss discovered that what were thought to be parts of Farabi’s Enumeration
of the Sciences were in fact parts of the first part (the Attainment of Happiness) of a lost
Farabian trilogy, The Philosophy of Plato and Aristotlee (consisting of the Attainmentt
of Happiness, The Philosophy of Plato, and The Philosophy of Aristotle)—contains one
of Strauss’s frankest statements of the depth and significance of Maimonides’s debt
to Farabi. Strauss concludes this piece by arguing that now that we know the
outlines of this lost trilogy, we are finally in possession of the “central writings” off
Farabi—because we can add this writing to the Enumeration (or Encyclopedia), the
Virtuous City, and his Political Regime.33 In explaining the importance of this dis-
covery, Strauss cites the letter Maimonides wrote to Samuel ibn Tibbon explainingg
the unparalleled importance of Farabi’s writings and especially his Political Regime..
Strauss concludes this piece by stating that at last we are in a position to assess
properly “the Islamic and Jewish philosophy of the Middle Ages: At the beginning
of this epoch of the history of philosophy there stands not just any ‘predecessor,’’
but the towering spirit [Farabi] who laid the ground for the later development
and set down its limits by making his task the revival of the Platonic-Aristotelian
philosophy as philosophy proper.”The Farabian Turn set Strauss on the course to a
Plato who was far less metaphysical than had been previously believed. In effect, it
was through his study of Farabi and Maimonides that Strauss came to understand
that Plato’s “political philosophy broadly understood . . . [is] first philosophy.’”34
This chapter has explored the more explicit declarations of Strauss’s writings in
the 1930s about the centrality of political science or philosophy in Maimonides’s
writings to facilitate a recovery of the distinctiveness of the “Islamic and Jewish
philosophy of the Middle Ages,” especially that of Farabi and Maimonides. The
Farabian Turn in Strauss’s thought, though often elusive in the written word is there
for anyone who can see and attend to Strauss’s actions. Even while he grows qui-
eter about the centrality of political philosophy in Maimonides’s writings, Strauss’s
exploration of Maimonides’s enlightened kalām demonstrates a key, oft-forgotten
aspect of
o political
po t ca philosophy.
p osop y.
L E O S T R AU S S O N F A R A B I , M A I M O N I D E S 167

Notes
1. Daniel Tanguay, Leo Strauss: An Intellectual Biography, trans. Christopher Nadon (New
Haven:Yale University Press, 2007), 80.
2. Originally published as Philosophie und Gesetz: Beiträge zum Verständnis Maimunis
und seiner Vorläufer (Berlin: Schocken Verlag, 1935). Translated by Eve Adler (Albany::
SUNY Press, 1995); henceforth referred to as PLA.
3. Originally published as “Quelques remarques sur la science politique de Maïmonide
et Farabi,” Revue des Études Juives 100 (1936): 1–37. Translated by Robert Bartlett in
Interpretation 18, no. 1 (1990): 3–30; henceforth SR.
4. Originally published as “Der Ort der Vorsehungslehre nach der Ansicht Maimunis,””
Monatschrift für Geschichte und Wissenschaft des Judentums 81, no. 1 (1937): 93–105..
Translated by Gabriel Bartlett and Svetozar Minkov in Review of Metaphysics 57
(2004): 537–49; henceforth PoP.
5. Tanguay, Leo Strauss, cf. 55, 57, 61 with 65, 66, 67, and 97.
6. Originally published in Essays on Maimonides, 37–91, ed. Salo Baron (New York::
Columbia University Press, 1941), it was eventually included in PAW W 38–94.
7. Parens,“Strauss on Maimonides’s Secretive Political Science,” in Leo Strauss’s Defense off
the Philosophic Life: Readingg “What Is Political Philosophy?” ed. Rafael Major (Chicago::
University of Chicago Press, 2012), 116–36.
8. As Tanguay argues (Tanguay, Leo Strauss, 84), Farabi is the thinker who made Strauss
aware of a different, more political and less metaphysical understanding of Plato..
All the while that Maimonides’s debt to Farabi became more apparent to Strauss,,
however, he seems to have grown increasingly uneasy about underlining the political
character of Maimonides’s teaching.
9. See Tanguay, Leo Strauss, 56, and Joshua Parens, “Escaping the Scholastic Paradigm,””
in Encountering the Medieval in Modern Jewish Thought, ed. Aaron Hughes and James
Diamond (New York and Leiden: Brill, 2012), 207–27.
10. Why do I call these “central chapters”? (After all, the numerically central chapter in
the Guidee is II 13, the first of two crucial chapters on eternity versus creation.) First,,
these chapters surround what Maimonides seems to be setting up as the holy off
holies, III 1–7. Second, the prophetology is of the essence of the theme of the Guide,,
the true science of the (divine) Law. Third, these two sections identified by Strauss
consist of 17 chapters each. As Nasser Behnegar has noted, according to Strauss, the
number 17 stands numerologically for “nature”—in a sense, the true center of all
philosophizing (“Reading ‘What is Political Philosophy?’” in Leo Strauss’s Defense off
the Philosophic Life, 41n5).
11. In the Aristotelian tradition to which Maimonides adheres at least in this, psychology
is a theoretical science containing elements of natural science and divine science. Cf..
On the Soull 402a5–7, 403a3–b18.
12. This stress on supernaturalism and the limits of human knowledge should be comparedd
with the approach to Maimonides developed by Shlomo Pines and his students afterr
Pines published his “Limits” article. Compare my Maimonides and Spinoza (Chicago::
University of Chicago Press, 2012), Chapter 1 for evidence that Maimonides’s argu-
ments regarding the limits of knowledge are anything but straightforward, or as
Strauss states immediately before the above quotation from PLA, “beyond doubt.”
13. Averroes or Ibn Rushd (1126–1198 ce), Decisive Treatise and Epistle Dedicatory,,
trans. Charles E. Butterworth (Provo, UT: Brigham Young University Press, 2001)..
Gersonides or Levi ben Gershom (1288–1344 ce), The Wars of the Lord, trans..
Seymour Feldman (3 vols.; Philadelphia: Jewish Publication Society, 1987, 1999).
168 J O S H UA PA R E N S

14. PAW W was published in 1952. Cf. note 6, above.


15. Cf. Parens, Maimonides and Spinoza, 112–15, on withholding and whether it is natural
or supernatural.
16. And this despite Strauss’s insistence that the Islamic Aristotelians did not possess the
Politics. Cf. Shlomo Pines, “Aristotle’s Politics in Arabic Philosophy,” Israel Orientall
Studies 5 (1975): 150–60.
17. And in Avicenna’s Healing: Metaphysics 10, Division of the Rational Sciences (in MPP–2
74–88), and Farabi, The Virtuous City (in Al-Farabi on the Perfect State, ed. Richard
Walzer [Oxford: Clarendon, 1985]).
18. Parens, “Strauss on Maimonides’s Secretive Political Science,” in Leo Strauss’s Defensee
of the Philosophic Lifee (Chicago: University of Chicago Press, 2012), 116–36.
19. See W. Z. Harvey, “Why Maimonides Was Not a Mutakallim,” in Perspectives on
Maimonides, ed. Joel L. Kraemer (Oxford: Oxford University Press, 1991), 105–14,,
cited in Sarah Stroumsa, Maimonides in His World: Portrait of a Mediterranean Thinkerr
(Princeton: Princeton University Press, 2009), 37n47.
20. See Heinrich Meier, Leo Strauss and the Theologico-Political Problem (New York::
Cambridge University Press, 2006).
21. See PAW W 19.
22. See Haggai Ben Shammai, “Kalam in Medieval Jewish Philosophy,” in History off
Jewish Philosophy, ed. Daniel H. Frank and Oliver Leaman (London: Routledge,,
1997), 127–32.
23. See Farabi, Enumeration of the Sciences, Chap. 5 in MPP–2 21–23.
24. I believe that Strauss’s scholarly understanding of political science or political phi-
losophy in Maimonides developed in the direction of what he would refer to in
the Platonic context as “political philosophy broadly understood” (CM M 20). The
view that political science or political philosophy is a mere discipline, still expressedd
here in “Place of Providence,” is due in part to the continued influence of medieval
Christian views of the division of the sciences on Strauss. I say that despite Strauss’s
evidently growing awareness of the divide between Maimonides and the Islamic
ā , on the one hand, and medieval Christian thought of the Thomistic school,,
falāsifa
on the other (e.g., PoP 541, 543n18; PLA 73n25). For more on medieval Christian
views on the status of political science, see the discussion of Abravanel below.
25. See esp. PoP 542–43n17. It appears that in SR Strauss did not understand fully that
the political center of the Guidee was in the prophetology. In SR 12, he linked the
articles or principles of faith that Maimonides enumerates in Commentary on thee
Mishnah, Pereq Heleq, with Guidee III 25–50 rather than with II 39–40. He did so
by referring to III 25–50 as Maimonides’s discussion of “the divine law.” In otherr
words, in SR he had not fully grasped that the prophetology was the very heart off
Maimonides’s analysis of divine law.This might help to explain the extent of Strauss’s
focus on the psychology of prophecy in Philosophy and Law. (Then there’s the prob-
lem of when these three pieces were actually written. Though they came out in
1935, 1936, and 1937, it was not necessarily the case that SR was not written before
Philosophy and Law.) Perhaps not until PoP did Strauss come to understand fully the
extent to which the core of the true science of the (divine) Law, the prophetology,,
was part of political philosophy.
26. Henceforth APT.
27. See esp. PAW W 96–97.
28. Readers who have read widely in Strauss’s thought will recognize here a phras-
ing that Strauss will use again, nearly verbatim, to describe the position of Martin
Heidegger in “Philosophy as Rigorous Science and Political Philosophy,” SPPP P 30..
See also Parens, “Strauss on Maimonides’s Secretive Political Science,” 118n5, 130.
L E O S T R AU S S O N F A R A B I , M A I M O N I D E S 169

29. See the general introduction to MPP–2 12–17, and Parens, “Escaping the Scholastic
Paradigm.”
30. In Appendix E of this volume, we have published for the first time a translation into
English of “On Abravanel’s Critique of Monarchy” (1937); first published as “Zu
Abravanels Kritik des Königtums,” in GS–2 233–34. In that piece, Strauss took H..
Finkelscherer to task for arguing in a journal essay in Monatsschrift für Geschichte undd
Wissenschaft des Judentums (1937) that Ibn Caspi (1297–1340 ce) already maintainedd
the kind of hostility toward monarchy expressed by Abravanel. (In APT 101n1,,
Strauss merely alludes to the divide between Abravanel and Ibn Caspi.) In “On
Abravanel’s Critique of Monarchy,” Strauss notes that Ibn Caspi likens monarchy
to sacrifices. Finkelscherer did not think through the significance of this reference
to sacrifices, which in Guidee III 32 are paradigmatic of legislation that expresses the
second rather than the first intention of the legislator. They are merely conditionally
rather than absolutely good.This kind of goodness is a far cry from Abravanel’s rejec-
tion of monarchy as bad.
31. See PAW W 40n9—and the Book of Religion, referred to below in note 33, a book that
Strauss did not have the good fortune to see.
32. Originally published as “Eine vermisste Schrift Farâbîs,” in Monatsschrift für Geschichtee
und Wissenschaft des Judentums 80 (1936): 96–106; translated in appendix D of the
present volume.
33. Little did Strauss know that there was at least one more “central writing” by Farabi
to be discovered, namely, his Book of Religion. See Alfarabi’s Political Writings: “Selectedd
Aphorisms” and Other Texts, trans. Charles E. Butterworth (Ithaca: Cornell University
Press, 2001), and the critical Arabic edition by Muhsin Mahdi, Abú Naṣr al-Fârâbî,,
Kitâb al-Milla wa Nụ
N súṣ Ukhrâ (Beirut: Dar al-Mashriq, 1968).
34. CMM 20. See Strauss’s observation in “A Lost Writing” in appendix E below that
Ibn al-Qifti was wrong in thinking that the missing metaphysics at the end of Thee
Philosophy of Aristotlee was due to textual corruption.
C APT R 9
CHAPTER

THE PROBLEM OF THE ENLIGHTENMENT: STRAUSS,


JACOBI, AND THE PANTHEISM CONTROVERSY

David Janssens

Denn was die Philosophen sogar ein wenig nachsehend und parteiisch gegen Enthusiasten und Schwärmerr
macht, ist, dass sie, die Philosophen, am allermeisten dabei verlieren wurden, wenn es gar keine Enthusiasten
and Schwärmer mehr gäbe.

—Lessing, “Über eine zeitige Aufgabe””

I
In his first book, Leo Strauss provides the reader with an interesting clue to one
of the sources of his groundbreaking critical study of Spinoza’s Theologico-Politicall
Treatise. While identifying the guiding question of his undertaking, he also points
out its pedigree:

Even if all the reasoning adduced by Spinoza were compelling, nothing would have
been proven. Only this much would have been proven: that on the basis of unbeliev-
ing science, one could not but arrive at Spinoza’s results. But would this basis itself
thus be justified? It was Friedrich Heinrich Jacobi who posed this question, and by so
doing lifted the interpretation of Spinoza—or what amounts to the same thing—the
criticism of Spinoza on to its proper plane.1

This statement is both literally and figuratively singular: the only reference to Jacobi
in the whole book, unaccompanied by any mention of its source, it makes us won-
der about the importance of this author for Strauss’s endeavor. A renowned critic
of the Enlightenment, Friedrich Heinrich Jacobi (1743–1819), singled out Spinoza
as one of the main targets of his attacks.2 Similarly, in Spinoza’s Critique of Religion,,
Strauss casts doubt on the legitimacy of Spinoza’s attack against revealed religion,,
thereby also questioning the foundations of the Enlightenment. In a discerningg
review of the book, his contemporary Gerhard Krüger noted that in Spinoza’s
Critique
q off Religion
g “there is concealed a fundamental philosophic discussion of the
172 DAV I D J A N S S E N S

problem of the Enlightenment.”3 If this interpretation is sound, then from a merely


formal point of view the procedure followed by Strauss closely resembles that off
Jacobi: to address the problem of the Enlightenment by means of a critical assess-
ment of Spinoza.4
However, even if Strauss’s critique of Spinoza may be said to take its cue from
Jacobi, it is not clear whether the latter’s influence reaches beyond this initial impulse,
nor is it clear to what extent. Recently it has been suggested not only that Spinoza’s
Critique of Religion is “by its own account, ‘Jacobian’ in orientation” but also that
“the Jacobian dilemma and the critique of rationalism [remained] fundamental forr
Strauss’s perspective” throughout his career.5 Moreover, these assumptions carry an
implicit criticism, to the extent that Strauss may be said to be heir to the irratio-
nalism, conservatism, and authoritarianism attributed to the anti-Enlightenment
with which Jacobi is commonly associated.6 This chapter will attempt to show that
such assessments are in need of qualification. It will be argued that even if a certain
affinity between Strauss and Jacobi can be shown to exist, this affinity is far more
complex than it seems.
In order to bring out this complexity, a closer look will be taken at those writings
in which Strauss discusses Jacobi. To begin, there is his doctoral dissertation, which,,
although he later disparaged it as “a disgraceful performance,” nevertheless merits
closer investigation.7 A comprehensive account, moreover, must broaden the inquiry..
After the completion of Spinoza’s Critique of Religion, Strauss worked as a coeditorr
of the Jubilee Edition of Moses Mendelssohn’s collected works. As a part of this
employment, he conducted research into the so-called Pantheism Controversy.This
debate was launched by Jacobi, with Moses Mendelssohn as its principal addressee,,
and initially concerned the philosophical legacy of the thinker and writer Gotthold
Ephraim Lessing. However, it soon developed into a fullblown debate concerningg
the foundations and the legitimacy of the Enlightenment, involving such promi-
nent contemporaries as Johann Georg Hamann, Immanuel Kant, Karl Reinhold,,
and Johann Gottfried Herder.8 The results of Strauss’s research are published in the
introductory commentaries he wrote to several of Mendelssohn’s writings. Besides
a thorough treatment of the latter’s position in the controversy, these introductions
contain an astute and well-documented investigation of its general background,,
particularly of the positions of Jacobi and Lessing.9 In this respect, they enable us to
come to a more exact determination of Strauss’s own perspective. It will become
apparent that, although Jacobi does play a very important role, this role differs con-
siderably from what the current view has made it out to be.

II
Judged solely by its title and its structure, Strauss’s doctoral dissertation is typical
of the genre: an inconspicuous exhibition of academic proficiency. Yet his treat-
ment of The Problem of Knowledge in the Philosophical Doctrine of Fr. H. Jacobii reveals
some distinctive traits.10 Thus, at the beginning of the work, Strauss expressly states
his intention to treat Jacobi as a competent thinker in his own right, not as the
romantic enthusiast and proponent of the Sturm und Drangg he is often made out
to be. Moreover, he cautions the reader that he intends to deal “not so much with
T H E P RO B L E M O F T H E E N L I G H T E N M E N T 173

Jacobi himself, but rather with the problems or perspectives designated by the title
‘Jacobi.’”11 Near the end of the work, this approach is reaffirmed and put within a
larger framework, when it is asserted that “the eidos ‘Jacobi,’ so rich in consequences
for intellectual history,...corresponds to a distinct and autonomously coherent
complex of problems on the level of timeless problems.”12
For various reasons, these statements merit closer examination. To begin with,,
they are formulated in a language with unmistakable Platonic resonances, oddly
prefiguring the interpretation of Platonic Ideas as timeless problems in Strauss’s
later works.13 More important, they suggest that Strauss is more concerned with
the philosophical problem brought to the fore by Jacobi than by the particularr
solution he propagated. This does not do away with the fact that the problem only
becomes visible through the solution argued by Jacobi in his polemic against the
Enlightenment. As Strauss shows in his analysis, this polemic is deployed on both
the epistemological and on the ethical-political level although both dimensions are
ultimately rooted in a single conviction characteristic of Jacobi’s thought.
On the level of epistemology, the main target of Jacobi’s critique is the Cartesian
method of radical universal doubt at the heart of modern rationalism. As is welll
known, this method attempts to secure the reality of Being by reducing it to an
indubitable condition of possibility, from which it then sets out to reconstruct Beingg
according to the requirements of reason. In Jacobi’s view, this amounts to a system-
atic reduction of Being or reality to non-Being or Nothingness—a procedure forr
which he coined the term “nihilism.” All that remains is the pure thinking subjectt
that thereby becomes the only source of reality and the sole warrant of genuine
knowledge. Strauss rephrases this view of the Cartesian program as follows: “We can
only know what we can produce. Thus, the philosopher who wants to understandd
the world must become the creator of the world.”14
Moreover, according to Jacobi, the Cartesian procedure is deliberately selective::
it filters out those aspects of its object that withstand reduction and rational control..
In this way, it ignores or even destroys such vital elements of the object as it can
never artificially replace or reconstruct. These elements point to what Jacobi calls
“natural certainties” (natürliche Gewissheiten), which are known prior to any attempt
at rational knowledge and therefore constitute the possibility of such knowledge..
From his point of view, both the source of knowledge (human understanding) and
its object (reality) are “irrational” or, to be more exact, “superrational” (überrationall)..
They come to light in propositions that are grasped with intuitive immediacy and
therefore cannot be made the object of subsequent rational proof, such as “I am””
and “There is a world outside of me,” but also the reality of God.15 As a result, Jacobi
rejects Kant’s notion of God as a regulative idea of reason. The latter, he argues,,
reverses the original primacy of God with regard to reason, and thus is devoid off
any content and hence both theoretically and ethically useless.16
Because of its deliberate disregard of these natural certainties as the limits off
knowledge, Cartesian rationalism and the modern sciences based on it can be noth-
ing more than the organization of ignorance, Jacobi holds.17 For if the method off
rational demonstration exerts its power within a domain limited by the irratio-
nal and transcendent certainties, a strict determinism can rule only within those
bounds. Unable fully to justify the precedence of radical doubt over against natural
174 DAV I D J A N S S E N S

certainty, rationalism can never attain the truth because it is based on an initial
surrender of the truth.18 Rationalism sacrifices theory or contemplation in orderr
to radically exclude irrationality. In Strauss’s words: “Doubt is the relinquishment
of theoretical life (truth) for the sake of the theoretical evil (irrationality) which is
necessarily related to it.”19
On the level of ethics and politics, Jacobi’s argument runs parallel to his episte-
mological critique. In this case, his polemic is directed against the idea of autonomy
at the heart of the moral and political program of the Enlightenment. In his dis-
sertation, Strauss summarizes the main contention as follows:

Autonomism is the ethical form of general doubt, of the principle of modern culture,
which invokes the autonomy of religious conscience, of scientific reason, and of moral
legislation (sola fides, sola ratio, “only a good will”). In opposition, Jacobi emphasizes
that, in ethical matters, it is simply unnecessary for the acting subject to understand
the norm and to affirm it out of its own insight. It is not the case that insight precedes
and obedience follows, but precisely the reverse: only out of obedience, as a result of
following the norm, from the penetration of the norm into the center of our lives as
a consequence of obedience, does moral insight emerge.20

Just as the principle of radical doubt and the belief in proof and demonstration
express a refusal to submit to the transcendence of reality, so does the concept off
autonomy disclose a rejection of the ethical norms inherent in this reality, incited
by man’s proud desire to be the sole source of morality. Correspondingly, just as it
leads to organized ignorance and determinism on the level of knowledge, so does
rationalism lead to atheism and fatalism on the level of morality and politics, Jacobi
asserts. Contrary to its own claim, rationalism is incapable of replacing what is has
destroyed; it is unable to establish morality on purely immanent grounds.
The intrinsic relationship among epistemology, ethics, and politics postulatedd
by Jacobi explains why his opposition to the Enlightenment takes the form off
a critical discussion of Spinoza. The latter, in Jacobi’s view, exemplifies the defi-
ance of Cartesian rationalism in the face of transcendence. In his introduction
to Mendelssohn’s contributions to the Pantheism Controversy, Strauss recalls how
Jacobi locates the root of rationalism in “the tendency to prove everything and to
accept nothing as given; if one follows this tendency honestly, i.e., without com-
punction, it leads to Spinozism, i.e., to atheism and fatalism ...the origin of the
tendency to prove everything is the will of man not to be dependent on a truth
that transcends him, the will ‘not to obey the truth, but to command it,’ pride,,
vanity.”21 With unrivalled clarity, Jacobi argues, Spinoza’s thought shows that the
common root of the Enlightenment’s philosophy and politics is a rebellious and
revolutionary effort to liberate man from the authority of transcendence. As Strauss
notes subsequently, Jacobi was “still too closely tied to the theistic tradition not to
be compelled to see in atheism (and ‘Spinozism is atheism’) a result of anti-theism,,
of the revolt against God.”22
According to Jacobi, however, the motive underlying this revolt proved to be at
least as tyrannical as that of its putative opponent: Descartes and Spinoza heralded
a new metaphysical despotism of autonomous demonstrative reason, which found
its political complement in the new political despotism of Hobbes’s Leviathan. His
T H E P RO B L E M O F T H E E N L I G H T E N M E N T 175

objections notwithstanding, Jacobi respected both Spinoza and Hobbes for the
consistency and rigor of their thinking. In fact, he preferred these “classics of des-
potism” to the German Aufklärerr of his time. What he perceived as the latter’s half-
hearted rationalism and readiness to compromise with autocratic regimes provoked
his aversion to such an extent that he went so far as to defend the ideal of a liberal
state.23 Nevertheless, he remained intensely critical of rationalism because of the
lack of justification and the “nihilism” characteristic of Cartesian doubt.
Claiming at least equal justice, Jacobi’s own philosophic doctrine takes precisely
this deficit as its point of departure. His procedure is first to pursue rationalism to
its ultimate consequences, up to the point where its fatalism, atheism, and nihil-
ism become apparent, as well as its rootedness in ignorance. The knowledge of this
ignorance (Wissen des Nicht-Wissens) then becomes the basis for a salto mortale: a leap
out of rationalism and nihilism into faith or Glaube, motivated by the willingness
and the courage to take the risk of believing reality instead of doubting it. As Strauss
emphasizes, the concept of Glaubee at the heart of Jacobi’s doctrine is not primarily
religious: it comprises both “faith” and “belief ” in the Humean sense, accordingg
to which human knowledge is ultimately based on indemonstrable beliefs. In this
respect it proved to be a most powerful weapon in Jacobi’s polemic against the
Enlightenment, for it enabled him to argue that even the choice for rationalism and
demonstration rests upon a primary belief, an initial act of faith.24
In Jacobi’s view, Glaubee is not only an epistemological but also and even pri-
marily an ethical category: affirmation of the transcendence of reality is the basic
prerequisite for true virtue (Tugendd), which in its turn is the necessary condition forr
true knowledge.Without the recognition of his heteronomy and of the necessity off
loving obedience to God’s commands, man can never hope to attain true knowl-
edge. In fact, Jacobi goes so far as to equate virtue and knowledge: the Platonic
character of this identification, far from being accidental, actually points to the
foundations of his thought, Strauss emphasizes. According to Jacobi, the history off
philosophy is determined by the predominance of one of two typical theoretical
attitudes, whereby each type is rooted in a more general type of intellectual and
moral attitude. The first, which Jacobi dubs “Platonic,” is characterized by nobility,,
audacity, confidence, faith, and love and is therefore able to gain access to truth andd
virtue.25
The other type, called “non-Platonic,” displays the opposite qualities: baseness,,
apprehension, diffidence, distrust, disbelief, doubt, and pride, and accordingly the
inability to attain truth and virtue. According to Jacobi, the non-Platonic attitude
has become dominant in modern philosophy, and this decline has reached its nadirr
in the age of the Enlightenment. The latter, in spite of its earthly accomplishments,,
is animated by a Cartesian fear of the immediacy of transcendent reality and char-
acterized by the subsequent attempt to circumvent its claims. Faced with what he
perceives to be the dire consequences of this refusal, Jacobi’s doctrine of Glaubee is
an emphatic attempt to restore the Platonic attitude. Through a change in morality,,
it seeks to reaffirm the transcendence of reality with a view to reinstating what has
been lost and thus accomplishing a renewal of philosophy.26
Because of the largely analytic and descriptive nature of Strauss’s account, it is
d cu t to assess tthee impact
difficult pact tthee eidos “Jacobi” may have had on his own thinking.
176 DAV I D J A N S S E N S

Looking ahead at his ulterior writings, it is nevertheless possible to highlight a feww


aspects. The first of these concerns the critique of Cartesian methodical doubt..
Very likely, Jacobi’s challenging the legitimacy of radical doubt informs Strauss’s
question in Spinoza’s Critique of Religion, mentioned at the beginning, whether “the
basis of unbelieving science” underlying Spinoza’s philosophy is justified. For, as
Strauss argues in the same book, it is precisely with an appeal to Cartesian doubt
that Spinoza excludes both the possibility of miracles in general and of prophecy in
particular (as a miraculous collaboration of reason and imagination).27 In the same
context, Jacobi’s comment on the selective character of methodical doubt may very
well have provided the background for a basic question, raised by the later Strauss,,
concerning the Ethics: c “But is Spinoza’s account clear and distinct? ...Is its clarity
and distinctness not due to the fact that Spinoza abstracts from those elements off
the whole which are not clear and distinct and which can never be rendered clearr
and distinct?”28 In Natural Right and History, he seems to answer this question in
the affirmative when criticizing modern philosophy’s “dogmatic disregard of every-
thing that cannot become an object, that is, an object for the knowing subject, orr
the dogmatic disregard of everything that cannot be mastered by the subject.”29
Second, among the elements of transcendent reality that are not and cannot be
rendered clear and distinct under the auspices of rational demonstration, Jacobi
gives pride of place to the existence of God. Against Kant, he argues that, as Strauss
puts it in his dissertation, “The [philosophical] system must accommodate itself to
the existence and meaning (Sinn) of God; the fundamental religious phenomenon
may not be diverted (umgebogen) for the sake of the system.”30 Interestingly enough,,
we see Strauss himself making frequent use of a similar argument during the 1920s
and 1930s, in various discussions involving the main currents within contemporary
Judaism, such as political Zionism, cultural Zionism, Jewish orthodoxy, and the so-
called return movement or neo-orthodoxy. Since it would lead us too far to discuss
each of these debates in detail, a few examples must suffice.31
In his very first publication, a contribution to a Zionist debate written in 1923,,
Strauss argues against those Zionists who advocate a qualified acceptance of reli-
gious contents for purely political reasons, mainly with a view to the needs of the
human soul: “Wholly inseparable from the essence of religion is a minimum off
doctrinal content, which minimum is the existence of God, wholly independent off
human existence and human need.”32 Two years later, the same criticism is directed
against Jewish orthodoxy when Strauss reproaches its adherents for “obeying the
law for the sake of their people or for the sake of all peoples, and not—or not
primarily—for the sake of God.”33 In like manner, finally, Strauss addresses cul-
tural Zionism and the “return movement,” through representatives like Hermann
Cohen, Franz Rosenzweig, Martin Buber, and Julius Guttmann.34 In his view, the
attempts of each of these thinkers to return to the Jewish tradition remained par-
tial and qualified by reservations because of their failure to consider “the original,,
non-‘internalized’ meaning of the basic tenets of tradition.”35 In sum, Strauss criti-
cizes contemporary Jewish thought for reducing one of the fundamental tenets off
Judaism, the radical transcendence of a personal God, either to a postulate of reason
or consciousness, to a product of religious experience, to a psychological need, to a
cultural or even a natural phenomenon. An adequate philosophical understanding
T H E P RO B L E M O F T H E E N L I G H T E N M E N T 177

of the condition of modern Judaism, he argues, requires that this transcendence— —


with regard to the Jewish nation, to human consciousness, and to reason—some-
how be accounted for in its original radicalness.
What makes these criticisms particularly puzzling and perplexing is that they
are leveled by someone who presents himself not only as an adherent of “simple,,
straightforward political Zionism” but also as an admitted nonbeliever.36 As becomes
apparent from recently published material, the young Strauss was a political Zionist
in the strict sense of the term, advocating the political self-organization of Jews on
a purely secular, unbelieving basis. In his early publications, he systematically argues
that modern science and politics have destroyed the closed world of the tradition,,
that the requirements of modern politics and those of the tradition are mutually
exclusive, and that “political Zionism which wants to found itself radically must
found itself as unbelieving.”37
At the same time, however, Strauss never ceases to stress that the fundamental
claims of the tradition somehow continue to exert their power and hence cannot
be dismissed forthwith. Regarding this quandary, he admits there is no obvious
way out. Precisely for this reason, unbelieving political Zionism is compelled to
acquire full clarity regarding its own position. This, however, can only succeed on
the condition that it rejects attempts to secure a middle ground and approaches “the
question of God” with at least as much seriousness as “the question of politics.”38
As Strauss has already perceived at this initial stage, doing justice to both requires a
renewed understanding of the old quarrel between belief and unbelief and thus off
its latest installment, Spinoza’s critique of religion.
Although he is never mentioned by name in these early writings, it is hard to
disregard the impression that Strauss’s perception of this fundamental issue and off
the road to be taken is partially indebted to Jacobi.39 This is borne out most clearly
by what might be called the “theologico-political” aspect of Jacobi’s polemic. As we
have seen, Jacobi subordinates the epistemological question to the ethical-political
question and focuses on the specific motive underlying modern rationalism. A simi-
lar approach may be said to be one of the distinctive characteristics of Strauss’s studyy
of Spinoza. In an article written in preparation of the book, he points out thatt
Spinoza’s theoretical critique, according to its own view of religion as based on
obedience and faith, necessarily presupposes disobedience and unbelief. As a result,,
a critical reading of Spinoza must concentrate on “the ‘Why’ of theory as the ‘Why’’
of disobedience and unbelief. This ‘Why’ precedes every theory; it is no theoretical
insight or conviction, but a motive.”40 Subsequently, in the book Strauss traces this
motive to the Epicurean tradition and its attempt to relieve the human condition
by liberating man from fear of the gods, the cause of the greatest unrest and the
gravest crimes. On the basis of this connection, he argues, “the concern for safe-
guarding and alleviating life may be designated as the characteristic concern of the
Enlightenment.”41 As such, it animates Spinoza’s construction of a world in which
there is no place for an inscrutable God or for a revealed law teaching man what is
good and what is evil.
At the same time, this concern reveals the limits and flaws of Spinoza’s cri-
tique. As Strauss argues at various points in the book, Spinoza’s neo-Epicurean
motive
ot ve caused him to ddisregard
s ega d tthee fundamental
u da e ta ddistinction
st ct o between
betwee supe
superstitious
st t ous
178 DAV I D J A N S S E N S

or profane fear and genuine fear of God, which is the traditional prerequisite forr
true love of God and true obedience to the revealed law. This shows that Spinoza
did not even attempt to understand or to take seriously his opponent. Driven by
a Cartesian “will to immediacy,” he was incapable of understanding the “will to
mediacy,” which, as a response to the original revelation, is the basis of obedience
and loyalty to the tradition of revelation. Instead, he urged the legitimacy of his
position to the point of committing a fatal petitio principiii by presupposing the
freedom of philosophizing that he set out to argue against revealed religion. While
this precluded any offensive tactics and only allowed for a defensive position, it left
his opponent essentially intact. The defensive critique, however, could only uphold
itself and its freedom by permanently turning revealed religion into ridicule, thereby
concealing its own questionableness.42 In the final analysis, there is at “the basis off
unbelieving science” a moral motive that is no more or no less problematic than
that of revealed religion, Strauss concludes. This motive, furthermore, differs from m
its Epicurean precursor on a crucial point: its probity and its conscientiousness,,
which reveal it to be a descendant of biblical morality.43
Both the approach and the results of Strauss’s investigation seem to be closely akin
to those of Jacobi. For both, the inquest into the motive of the Enlightenment points
to a revolutionary antitheism animated by proud human reason, self-postulating and
therefore deeply problematic. In his dissertation, Strauss even goes so far as to sub-
scribe to Jacobi’s typological characterization of modernity as an age of fear, distrust,,
and pride:

In any case, it seems to us that a specific moment of modern culture is viewed here for
the first time in such a comprehensive manner. How little one has reason to regard—
and to disregard—this expression as a mere circumstance of Jacobian sentiment, is
made evident most clearly by the fundamental agreement in which it finds itself with
the results of the research of contemporary sociologists (such as Troeltsch, Sombart,
Max Weber, Scheler).44

However, Strauss immediately goes on to qualify his assent by adding that this does
not mean he also shares the strong evaluative judgment (Bewertung) g Jacobi appends
to it. At this stage, Strauss is less dismissive regarding the claims of modern rational-
ism, even if in many ways he shares Jacobi’s insight into its flaws. Thus, he makes
the critical remark that although Jacobi is fundamentally aware of the scope off
Cartesianism as “a general philosophic principle of method,” he fails to do justice
t 45 As a critical response
to “its profound practical legitimacy (tiefes sachliches Recht).”
to Cartesian doubt, the foundation of Jacobi’s doctrine of Glaubee seems to be at
least as questionable as that of its opponent. Accordingly, in Spinoza’s Critique off
Religion, though viewed with an increasingly critical eye, rationalism is still treated
with more impartiality than is meted out by Jacobi.46
In a different form, this reservation is also visible on the ethical-political level..
Commenting on Jacobi’s defense of heteronomy, Strauss notes that it is basically
the expression of the principle of traditionalism or, to be more exact, conservatism::
“The principle of tradition—which doesn’t mean the recognition of a particularr
tradition. Rather,, one should say: p of conservatism.”47 To the extent that
y principle
p
T H E P RO B L E M O F T H E E N L I G H T E N M E N T 179

the “leap of faith” implies the espousal of the principle of traditionalism, it leaves
undetermined what particular tradition is embraced. Nevertheless, Jacobi insists
that “the leap of faith” necessarily requires an espousal of the Christian tradition..
Even though he justifies this view by equating the principle of Christianity with
the absolute principle of religion as such, this cannot hide the fact that, as Strauss
puts it, “the difficult problem of the specification (Besonderung)
g of the highest moral
norm does not exist as a theoretical problem for Jacobi.”48 As a result, the leap off
faith bears the mark of decisionism, in its attempt to affirm what, according to
Jacobi’s own doctrine, is in no need of affirmation. In exposing the act of faith at
the basis of rationalism, the Jacobian option against Cartesian doubt succeeds in
restoring the balance, but it fails to do better than its opponent.
These and other considerations suggest that Jacobi’s influence on Strauss’s early
thinking, although certainly not negligible, is not as univocal as it appears to be..
At the same time, they indicate that, as a result of his investigations into Spinoza’s
critique, Strauss had more or less reached a quandary. On the one hand, his youth-
ful allegiance had become deeply problematic: modern rationalism was seen to be
based on a questionable moral motive with biblical roots. On the other hand, the
Jacobian alternative turned out to be equally flawed and unacceptable. In order to
see how Strauss found his bearings in this dilemma, we must take a further look at
the introductions to Moses Mendelssohn’s collected works. Written between 1931
and 1937, these writings are of particular interest, not only because they reveal a
profound knowledge of Mendelssohn’s thought but also because they focus on the
latter’s dispute with Jacobi, which came to be known as the Pantheism Controversy..
Above all, they suggest that Strauss had begun to find a way out of the quandary in
which he found himself. Since both the Pantheism Controversy and its aftermath
have been amply and excellently documented from a variety of perspectives by
different authors, the discussion will be limited to such aspects and features as are
salient in Strauss’s analysis.49

III
The beginning of the Pantheism Controversy is well known: in 1783, Jacobi
informed Mendelssohn, by way of a mutual acquaintance, that “in his last days,,
Lessing had been a committed Spinozist.”50 For Mendelssohn, this disclosure
amounted to nothing less than a slanderous degradation of the highest to the low-
est. At that time, the German intelligentsia revered Lessing as a champion of the
Enlightenment while it denounced Spinozism as a heretical, atheistic, and anar-
chistic doctrine. By the same token, Jacobi cast a shadow over Mendelssohn’s long-
standing friendship with Lessing.
With his declaration, Jacobi wanted to buttress his contention that the
Enlightenment and its rationalism as such ultimately led to atheism and fatal-
ism. Lessing, he claimed, had reached the same conclusion and had consistently
embraced its radical consequences. By making this publicly known, Jacobi intendedd
to force upon the Aufklärer the dilemma of either following in the footsteps off
Lessing and accepting the destructive effects of rationalism or rejecting rationalism
in favor
avo of
o hiss own
ow doctrine
doct e ofo Glaube
Glaube.. Ass a result,
esu t, Mendelssohn
e de sso was co compelled
pe ed to
180 DAV I D J A N S S E N S

defend not only the memory of his friend but also his own position as a protagonist
of the moderate Enlightenment.
As Strauss argues, Jacobi’s attack struck home because he and Mendelssohn
found themselves on common ground. Both faced the same problem: “the final
crisis of modern metaphysics of Cartesian-Leibnizian stamp.”51 More particularly,,
they both grappled with “the knowledge that the attempt of modern metaphysics
to found the concept of God particular to faith by means of unbelieving specula-
tion had failed.”52 The result of this attempt, generally known as natural theology
or natural religion, had become increasingly problematic as the radical premises off
“unbelieving speculation” had come to the surface and demanded a hearing. As
we have seen, Jacobi responded to this crisis by a wholesale repudiation of modern
metaphysics and the attempt to return to traditional faith. For Mendelssohn, this
solution was out of the question. Refusing to abandon the moderate wing of the
Enlightenment, he held on to the idea of a natural religion and to the possibility off
harmonizing religion and reason, not least because it provided the cornerstone off
his defense of Judaism as a religion of reason.53
In the course of his introductions, Strauss critically discusses several key elements
of Mendelssohn’s natural theology, showing how it became increasingly embattledd
by the atheism of radical Enlightenment, on the one hand, and by the Jacobian
return to faith, on the other. For the present inquiry, these are relevant only to the
extent that they enable Strauss to single out general characteristics and general
problems. In this perspective, the most important point in his treatment is his obser-
vation that Mendelssohn systematically privileges goodness as the primary attribute
of God. This, Strauss holds, is a central characteristic of the Enlightenment:

The whole of the Enlightenment, insofar as it implicitly or explicitly preserves a


relationship with the tradition rooted in the Bible, is characterized by the fact that it
combats the traditional doctrines and convictions by having recourse to the goodness
of God. More precisely, proper to the Enlightenment is the unequivocal priority it
accords to God’s goodness over His power, His honor and His punishing wrath; for
the Enlightenment, God is not primarily the demanding, summoning God, but rather
the benevolent God.54

The priority of goodness over the other divine attributes determines almost all
of the distinctive tenets of Mendelssohn’s natural theology, Strauss maintains. It
provides the basis for his demonstration of the immortality of the soul, of human
perfectibility and freedom, his concomitant rejection of eternal punishment and
his denial of revelation. A good and benevolent God, Mendelssohn holds, does not
need to make himself known by revelation but enables man to acquire knowledge
of his design by studying the perfect order of creation. Moreover, a benevolent God
could not have created man but with a view to happiness, so that man must be infi-
nitely perfectible. As a consequence, Mendelssohn rejects the ceaseless suffering off
eternal damnation, for it contradicts human perfectibility as well as the perfection
of creation. In addition, human perfectibility also implies that every individual pos-
sesses both an irreducible existence and certain inalienable rights that not even God
can violate. This can never lead to difficulties, Mendelssohn assures, for any conflict
between the rights of man and those of God is excluded.55
T H E P RO B L E M O F T H E E N L I G H T E N M E N T 181

According to Strauss, however, giving priority to divine goodness does not


express “a theological concern of any kind, but instead the concern for the substan-
tiality, the independence, the autonomy, and the proper right of the Ego (das Ich)::
the unconditional goodness of God is given priority because it is in accord with
the claims of the autonomous Ego.”56 In other words, Mendelssohn’s natural theol-
ogy proves to be ultimately guided by and accommodated to interests particular to
modern philosophy.57 This is rendered manifest by several observations. Thus, forr
Mendelssohn, one of the principal tasks of modern metaphysics consists in securingg
human happiness and individual progress by liberating man from the fear of death
and divine wrath. Not surprisingly, he once referred to his natural theology as to a
“rather Epicurean” theism.58
This modern character also becomes apparent in Mendelssohn’s attempts to
“correct” the doctrines concerning the immortality of the soul of two of his revered
predecessors, Plato and Leibniz. As Strauss shows, Phädon, Mendelssohn’s translation
of Plato’s Phaedo, contains many alterations and emendations to the effect that
the original teaching and its exigencies are mitigated and moderated. In a typical
manner, for example, Mendelssohn’s Socrates emphasizes the consoling effect off
the idea of immortality of the soul, whereas Plato’s Socrates does not regard this as
a valid argument but instead considers it an obstacle to philosophizing. A similarr
approach marks Mendelssohn’s Sache Gottes, oder die gerettete Vorsehung, ostensibly
an elaboration of Leibniz’s Causa Dei. Whereas Leibniz argues divine providence
by asserting that God’s justice is his goodness guided and limited by his wisdom,,
Mendelssohn reverses the order of wisdom and goodness. As a result, he must reject
eternal punishment and suffering, which Leibniz could still justify as a necessary
component of the best of all possible worlds.
However, although it was intended as a defense of the orthodox religious view
of providence, Leibniz’s concept of divine justice implied a radical break with the
tradition since it no longer allowed divine justice to be distinguished from divine
goodness and divine wisdom. In this way, Strauss argues, Leibniz prepared the
momentous transition from the old notion of law to the modern notion of right::
“[B]y dissolving the classical concept of justice which had preserved the original
meaning of justice as obedience with regard to the law, he had considerably pre-
cipitated the process that aimed at the eradication of law understood as obligation
in favor of right understood as claim.”59
Mendelssohn, a self-confessed follower of Leibniz, could not but accept this
result and adapt his natural theology in accordance with it. However, his edifice
started to topple when his faith in the power and the authority of demonstration
was decisively shaken in acrimonious disputes with critics who attacked his natu-
ral theology: “Compelled to defend his Judaism and his rationalism at the same
time, he had to present Judaism as a purely rational religion. In any case, however,,
the teaching of the Bible is not demonstrative....Saving Judaism was only possible
for him in this way, that he severely restricted the right and the significance off
demonstration.”60 This restriction found its expression in Mendelssohn’s introduc-
tion of the notion of “common sense” or “plain human understanding” (gesundes
Menschenverstand),d a specific human capacity to grasp intuitively and with full clar-
ity
ty ce
certain
ta esse
essential
t a ttruths
ut s tthat
at specu
speculative
at ve reason
easo aalone
o e ca
cannot
ot de
demonstrate.
o st ate. In
182 DAV I D J A N S S E N S

Mendelssohn’s view, since common sense alone could provide a basis for agree-
ment among men, it had to guide and to supplement reason, which he had come
to regard as insufficient.
Not surprisingly, Strauss is critical of this move. In the first place, he notes that
this new configuration of reason and common sense is merely a reiteration of the
traditional religious notion of revelation as a necessary guide for insufficient reason..
Confronted with the failure of Cartesian-Leibnizian metaphysics as a substitute forr
traditional faith, natural theology could do no more than to seek refuge on “the
neutral isle of common sense,” while the realm of speculation was invaded by the
radical atheist metaphysics of Spinozism.61 As Mendelssohn himself admitted, this
move did not differ essentially from Jacobi’s leap of faith out of speculation andd
demonstration. In both cases, the appeal to a faculty beyond speculation proved to
be the only way of saving teleology.
Second, Strauss challenges Mendelssohn’s judgment—foreshadowing the cur-
rent view—that Jacobi’s doctrine of Glaubee threatens philosophical speculation
and leads to irrational “enthusiasm” (Schwärmerey). On the contrary, he argues, it is
precisely common sense that endangers speculation: “For common sense lets the
animating conviction appear as self-evident, whereas [Jacobi’s] admission that this
conviction is merely believed, implies or may imply the knowledge of ignorance
and therewith an impulse to speculation.”62 Differently stated, Jacobi’s teachingg
preserves an unexplored latitude for philosophical speculation, which is altogetherr
excluded by Mendelssohn’s notion of common sense.
Third, Strauss argues that the notion of common sense merely compounds the
predicament it seeks to escape from. Cartesian philosophy, he explains, was moti-
vated by the view that traditional philosophy had relied too much on everyday
language. As a result, it called for a distinct and purely scientific language. This
demand, however, could not be brought into agreement with the equally important
requirement that the new philosophy enlighten humanity in general by supplanting
the old popular beliefs, for:

especially in its “language,” this philosophy was further removed from the language
of common sense than the earlier philosophy; it tended to extreme unpopularity.
However, it thus became entirely incapable of replacing the “popular system,” and
therewith of fulfilling one of its most important functions, that of “Enlightenment.”
Small wonder, then, that “enthusiasm” reared its head anew. However, small wonder,
as well, that common sense, which had allowed itself to be enlightened to the best
of its abilities by modern metaphysics, when it perceived that it could expect a new
“obscurantism” from the “subtleties” of this metaphysics, dismissed its nurse without
further ado and declared itself mature.63

It did so, however, in the illusion that it could now freely marshal clear and distinct
metaphysical truths since it regarded the latter as having been assimilated within
everyday language. Hence, although it was introduced to remedy the shortcom-
ings of Cartesian philosophy, the notion of common sense remained within the
horizon established by modern philosophy’s estrangement from everyday language..
As a result, it did not lead to a serious reconsideration of “earlier philosophy” in
relation to premodern, “nonenlightened” common sense. As Strauss points out,,
T H E P RO B L E M O F T H E E N L I G H T E N M E N T 183

Mendelssohn was convinced that premodern metaphysics had been definitely sur-
passed by modern metaphysics. He therefore persistently identified metaphysics
with modern metaphysics and thus proved incapable of understanding premodern
thought as it understood itself. One example of this failure is his distorting appro-
priation of the Platonic teaching concerning immortality.64
According to Strauss, however, this general critique applies with equal force to
Jacobi. The latter, in spite of his sweeping repudiation of modern metaphysics, also
remains decisively bound to its presuppositions and exhibits a similar blindness to
premodern thought. This becomes apparent in a central ambiguity of his critique
of Spinoza. One of Jacobi’s main objections against Spinozism is that it gives prior-
ity to action over thinking, whereby the latter is regarded as the mere continua-
tion of action (die Handlung im Fortgang).g However, he himself adopts precisely this
proposition in his polemic against the Enlightenment when he asks, rhetorically
and polemically: “Can philosophy ever be anything more than history?” and when
he asserts that “every age has its own truth, just as it has its own living philosophy,,
which describes the predominant manner of acting of the age in question in its
continuation (in ihrem Fortgange).”65 These assertions show that Jacobi’s irrational-
ism and traditionalism, according to which true knowledge can only result from
virtuous action motivated by obedience to transcendent reality, are actually rootedd
in historicism. This accounts for the decisionism characteristic of his “leap of faith,””
as well as for his attempt to bring about a renewal of philosophy through a change
in morality.
In spite of his efforts, Jacobi remained equally captive to the horizon of modern
historical thought. Strauss concludes: “Persisting in his critique of Spinoza to the
end, he would not have been able to appeal to history against the Enlightenment,,
nor to faith (Glaube) understood within the horizon of the concept of history.”66
The implications of this terse remark deserve our attention: a sustained critique off
Spinoza, it seems, would have called into question “the concept of history” and, per-
haps, opened the possibility of a nonhistorical approach to both the Enlightenment
and faith. It is hard to disregard the impression that, in this remark, Strauss is think-
ing of his own undertaking. For, as his principal writings of the 1930s show, this
is precisely the path his research has taken. Between Spinoza’s Critique of Religion
(1930) and Philosophy and Law w (1935), he has radically questioned the universal
claim underlying historical consciousness by tracing its origin to the concept off
“prejudice” introduced by Descartes and adopted by Spinoza. By the same token,,
he has begun to reopen the quarrel of the ancients and the moderns and to recoverr
the nonhistorical horizon of premodern thought, the horizon of “the old concept
of law” shared and disputed by “earlier philosophy” and revealed religion. Guided
by the medieval Enlightenment of Islamic and Jewish philosophers such as Farabi,,
Avicenna, Averroes, and Maimonides, he is led back to Plato and, ultimately, to the
enigmatic figure of Socrates.67

IV
If Jacobi’s thought was certainly influential in the development of Strauss’s early
t oug t, itt was by no
thought, o means
ea s dec
decisive.
s ve. Ge
Generally
e a y spea
speaking,
g, Jacob
Jacobi made
ade St
Strauss
auss awa
awaree
184 DAV I D J A N S S E N S

of the problem of the Enlightenment in two different ways: by means of his tren-
chant critique of the epistemological, theological, ethical, and political character-
istics of modern rationalism, but equally by the shortcomings of the solution he
proposed. On the first account, Strauss adopted several of Jacobi’s criticisms and
insights in his own study of Spinoza as well as in his discussions with contemporary
thinkers. On the second account, he expressed a number of fundamental reserva-
tions as to Jacobi’s doctrine. The most important of these pertains to Jacobi’s failure
to sustain his critique of Spinoza and his subsequent failure to question radically
the horizon of modern thought. Although this constitutes a crucial and, in a certain
respect, decisive point of divergence between Strauss and Jacobi, it is by no means
the occasion for a critical farewell, nor does it fully express Strauss’s final appre-
ciation. Throughout his analyses, he repeatedly suggests that the many ambiguities
and contradictions in Jacobi’s position may have been deliberate. In his campaign
against the Aufklärung, Jacobi constantly changed tactics and continually moved the
line of battle. Depending on the conduct of his opponent, he would alternately
take the side of radical atheists such as Spinoza and Hobbes against the prudent
dogmatism of the Aufklärer or launch a spirited defense of Christianity and tradi-
tion against the antitheistic dogmatism of radical Enlightenment. For Strauss, this
agility suggests that Jacobi may have been neither a dogmatic atheist nor a religious
enthusiast. Commenting on Mendelssohn’s bewilderment regarding the position off
his opponent, he writes:

It has been noted that he lacked Jacobi’s spiritual freedom, and that, as a result, Jacobi’s
vacillating between atheism and Christianity remained incomprehensible to him: at
times, he really did not know whether he encountered in Jacobi an atheist or a
Christian; only for a brief instant did he prove capable of rising to the insight that
Jacobi was a philosopher.68

According to Strauss, Jacobi’s great example in this particular modus operandi was
none other than Lessing himself. In various debates that had established his reputa-
tion as a writer and thinker, Lessing had alternately defended radical orthodoxy andd
radical Enlightenment without paying allegiance to either camp. He was unable
to adhere to any doctrine, and his loyalty only regarded an elusive truth that lies
between and beyond. The search for this truth enabled him to experiment with
conflicting opinions without accepting any of them as final. This philosophical
independence found its expression in his well-known love of paradox as well as in
his preference for conversation over doctrine. For Mendelssohn, who disapproved
of his friend’s “theater logic” (Theaterlogik), this vital aspect of Lessing’s thought was
inaccessible: he could not accept a paradoxical truth but dismissed every philo-
sophical dispute as “mere verbal disagreement” against the fixed background of the
certainties perceived by common sense understanding. Identifying philosophy with
ontology, he was unable to appreciate Lessing’s dialectical style of thinking, which
Jacobi had partly emulated:

If one pays attention to the How rather than to the What—and for Jacobi and Lessing
a e, the
alike, t e great
g eat manner
a e of o thinking
t g held
e d more
o e weight
we g t than
t a the
t e recognition
ecog t o ofo this
t s
T H E P RO B L E M O F T H E E N L I G H T E N M E N T 185

or that opinion—one will be inclined to reckon with the possibility that Jacobi was
the most intelligent follower Lessing found among his contemporaries....Jacobi felt
himself to be, not entirely without justification, the legitimate heir of Lessing and of
the latter’s radical, i.e., undogmatic way of thinking.69

To be sure, the praise implicit in this assessment is qualified. In the first place, it does
not detract from the pertinence of Strauss’s objections to Jacobi’s historicism and
his failure to pursue his critique of modern rationalism to the end. Furthermore,,
Strauss elsewhere suggests that even Jacobi did not fully fathom the extent off
Lessing’s irony and may have become its dupe.70 In any case, it suggests that any
influence Jacobi may have exercised over Strauss’s thinking is secondary to the
impact of Lessing and is even conditioned and mediated by the latter. Although an
adequate assessment of this impact would exceed the limits of this chapter, it is not
amiss to give two examples in support of this contention.
First, there is some evidence that Lessing played an important role in Strauss’s
rediscovery of the art of writing of ancient and early modern thinkers. This may be
inferred from the introduction to Mendelssohn’s Sache Gottes. Discussing the dif-
ference between Mendelssohn and Leibniz regarding the defense of eternal punish-
ment, he makes the following noteworthy remark:

Leibniz, however, did not believe in eternal damnation as it was understood by the
Christian tradition....The fact that he was nevertheless able to defend the ecclesiasti-
cal doctrine is ultimately rooted in the conviction that determines the content of his
defense: in the conviction of the unconditional priority of the beauty and order of
the whole over the happiness of the parts, hence also of human beings, and in the
conviction, inseparable from the former, that beatitude consists in the contemplation
of the universal order. For the ideal of contemplation carries with it the division of
mankind into the “wise” and the “many,” and therewith the recognition of a twofold
way of communicating truths, an esoteric and an exoteric.71

Although Strauss does not mention it in his commentary, he is well aware that
the source of this view is none other than Lessing. This is borne out by “Exoteric
Teaching,” a text he wrote three years later in 1939 but chose not to publish at the
time.There he refers to “Leibniz von den ewigen Strafen,” in which Lessing comments
as follows on Leibniz’s defense of the orthodox view of eternal punishment:

He did no more and no less than what all of the ancient philosophers used to do in
their exoteric speech. He observed a sort of prudence for which, it is true, our most
recent philosophers have become much too wise....I admit that Leibniz treated the
doctrine of eternal damnation very exoterically, and that esoterically he would have
expressed himself altogether differently on the subject.72

At the end of the same text, Lessing indicates that one ancient philosopher who
observed this sort of prudence was Socrates, who “believed in eternal punishment
in all seriousness, or at least believed in it to the extent that he considered it expe-
dient (zuträglich) to teach it in words that are least susceptible of arousing suspicion
and most explicit.”
p 73 That this remark did not escape Strauss is evinced by the fact
186 DAV I D J A N S S E N S

h Art off Writing.74 As he indi-


that he reproduced it in its entirety in Persecution andd the
cates in “Exoteric Teaching,” “Lessing was the last writer who revealed, while hid-
ing, the reasons compelling wise men to hide the truth: he wrote between the lines
about writing between the lines.”75 By the same token, Lessing may well have been
the first writer from whom Strauss began to learn how to read between the lines.
Second, as an assiduous student of his undogmatic way of thinking and of his art
of writing, Strauss knew that Lessing was familiar with medieval Jewish and Islamic
theology and philosophy.76 It is not unlikely that this had some bearing on his
own investigations. Thus, in 1946 he sketched the plan of a book tentatively titled
Philosophy and the Law, the final chapter of which was to be devoted to Lessing’s
Nathan the Wise. Although this celebrated play is generally regarded as a tribute to
Mendelssohn, the symbol of enlightened tolerance, Strauss hints at a strikingly dif-
ferent interpretation: “The recollection of the man Maimonides was probably one
of the motives underlying Lessing’s Nathan the Wise, the outstanding poetic monu-
ment erected in honor of Jewish medieval philosophy.”77 It is still far from clearr
whether and to what extent Lessing guided Strauss to Maimonides’s art of writingg
and perhaps also to the theater logic of Plato’s dialogues, that poetic monument
erected in honor of Socratic classical philosophy.78 Nonetheless, if there is some
foundation for the view that Strauss was a Jacobian, there is all the more reason to
explore the possibility that he was a committed Lessingian.79

Notes
1. SCR R 240.
2. Friedrich Heinrich Jacobi, Uber die Lehre des Spinoza in Briefen an den Herrn Moses
Mendelssohn (Breslau: Löwe, 1785). Abridged English translation in: F. H. Jacobi, Thee
Main Philosophical Writings and The Novel ‘Alwill, trans. George Di Giovanni (Montreal::
McGill–Queen’s University Press 1994).
3. Gerhard Krüger, “Review of Leo Strauss’ Die Religionskritik Spinozas als Grundlagee
seiner Bibelwissenschaft,” trans. George Elliot Tucker, Independent Journal of Philosophy
5/6 (1988): 173. Very likely, Krüger’s use of the word “concealed” is deliberate. As
becomes apparent from his correspondence with Strauss, the latter encountered
objections from Julius Guttmann, his superior at the Akademie fur die Wissenschaft
des Judentums. Guttmann demanded that certain passages in the book be altered orr
even omitted. Though Strauss deferred to Guttmann, he asked Krüger to criticize
the opaqueness of his work so as to guide his readers to his true intentions. See the
Strauss–Krüger correspondence, in GS–3 379, 393.
4. See Frederick C. Beiser, The Fate of Reason: German Philosophy from Kant to Fichtee
(Cambridge, MA: Harvard University Press, 1987), 448.
5. See, respectively, Susan Shell, “Taking Evil Seriously: Schmitt’s ‘Concept of the
Political’ and Strauss’s ‘True Politics,’” and John G. Gunnell, “Strauss Before
Straussianism: Reason, Revelation and Nature,” in Leo Strauss, Political Philosopherr
and Jewish Thinker, ed. Kenneth L. Deutsch and Walter Nicgorski (Lanham, MD::
Rowman & Littlefield, 1994), 183 and 171.
6. Compare Stephen Holmes, The Anatomy of Antiliberalism (Cambridge, MA: Harvard
University Press, 1996).
7. Leo Strauss (with Jacob Klein), GA–JPCM
– M 460.
8. As Strauss notes, the Pantheism Controversy marked the “formal reception off
Spinoza.” This reception was followed by a wave of “Spinoza enthusiasm” that lasted
T H E P RO B L E M O F T H E E N L I G H T E N M E N T 187

into the twentieth century, until the spell was broken by Hermann Cohen’s renewedd
excommunication. Cohen’s attack formed the occasion for Strauss’s reassessment off
Spinoza’s critique. Leo Strauss, “Preface to Spinoza’s Critique of Religion,” in JPCM M
154–58. See also Leo Strauss, “Das Testament Spinozas,” in GS–1 415–22; English
translation in LSEW W 216–23.
9. Both the dissertation and the introductions to Mendelssohn were published in GS–2..
Henceforth, all references to those writings are to this volume, as EPLJJ and EMFL,,
respectively. All translations of them in the present chapter are my own, D.J.
10. EPLJJ 237–92. The dissertation, defended on September 17, 1920, was written
under the direction of Ernst Cassirer, who conducted a large-scale research proj-
ect on the problem of knowledge in modern philosophy. See Ernst Cassirer, Das
Erkenntnisproblem in der Philosophie und Wissenschaft der neueren Zeitt (Berlin: Verlagg
Bruno Cassirer, 1906). A French translation of the dissertation is available in Revue dee
métaphysique et de moralee 99 (1994): 291–311 and 505–32.
11. EPLJJ 243. In a handwritten note appended to a summary of his dissertation, Strauss
calls his work “a non-Jacobian approach to Jacobian problems,” adding the remark
that “I have not presented ‘Jacobi as such’, but only insofar as I needed him”; EPLJJ
297.
12. EPLJJ 283.
13. In a similar vein, Strauss asserts in the introduction to the dissertation: “Certainly, a
philosophy that understands itself and refuses to surrender to relativism, must con-
ceive the truth it pursues as an independent and coherent condition, which it does
not create but rather seeks, discovers and recognizes.” EPLJJ 244. Compare WIPP P 39;;
NRH H 123–24, 150n4; CM M 119–21; RCPR R 174–76.
14. EPLJJ 249. Cf. NRH H 173–74, 174n9, 201; RCPR R 243–44; LAM M 212.
15. EPLJJ 249. Cf. Beiser, The Fate of Reason, 46.
16. EPLJJ 285. Cf. Beiser, The Fate of Reason, 81, 89–91.
17. EPLJJ 258. In this light, Beiser’s assertion that “natural science is the source of nihil-
ism” (The Fate of Reason, 85) must be supplemented: with equal justification, one
might say that, for Jacobi, nihilism is the source of the natural sciences.
18. EPLJJ 258–59.
19. EPLJJ 252.
20. EPLJJ 281.
21. EMFL 537–38; cf. LSMM M 71. Compare EPLJJ 278.
22. EMFL 549; cf. LSMM M 85.
23. EMFL 533–35; cf. LSMM M 65–68. Compare Frederick C. Beiser, Enlightenment,,
Revolution, and Romanticism: The Genesis of Modern Political Thought, 1790–1800 0
(Cambridge, MA: Harvard University Press, 1992).
24. Cf. Beiser, The Fate of Reason, 89.
25. EPLJJ 242–43, 270, 274–75, 277, 279–80, 282.
26. EPLJJ 245–47, 252.
27. Die Religionskritik Spinozas als Grundlage seiner Bibelwissenschaft, GS–1 235–47.
28. “Progress or Return? The Contemporary Crisis of Western Civilization,” in JPCM M
117. Compare “Preface to Spinoza’s Critique of Religion,” in JPCM M 253–54: “The
Ethics thus begs the decisive question-the question as to whether the clear and distinct
account is as such true and not merely a plausible hypothesis....[T]he clear and dis-
tinct account of everything which it presents remains fundamentally hypothetical.”
29. NRH H 30. Cf. David R. Lachterman,“Laying Down the Law:The Theologico-Political
Matrix of Spinoza’s Physics,” in Leo Strauss’s Thought:Toward a Critical Engagement, ed..
Alan Udoff (Boulder, CO: Lynne Rienner Publishers, 1991), 123–53.
30. EPLJJ 251–52.
188 DAV I D J A N S S E N S

31. For a more extensive account, see David Janssens, “Weimar Revisited: Judaism,,
Zionism, and Enlightenment in Leo Strauss’s Early Thought” (in Hebrew), Iyyun 50
(2001): 407–18.
32. “Antwort auf das ‘Prinzipielle Wort’ der Frankfurter,” GS–2 305 (cf. “Response to
Frankfurt’s ‘Word of Principle,’” LSEW W 69).
33. “Ecclesia Militans,” GS–2 353 (cf. “Ecclesia Militans,” LSEW W 126). See also “Biblischee
Geschichte and Wissenschaft,” GS–2 357–59 (cf. “Biblical History and Science,” LSEW W
131–33).
34. “Preface to Spinoza’s Critique of Religion,” in JPCM M 144–55.
35. PLA 136n3. Compare “Bemerkungen zu der Weinbergschen Kritik,” GS–1 429
(cf. “Comments on Weinberg’s Critique,” LSEW W 119); “Die Zukunft einer Illusion,””
GS–1 433 (cf.“Sigmund Freud, The Future of an Illusion,” LSEW W 203);“Zur Ideologie
des politischen Zionismus,” GS–1 447.
36. GA–JPCM
– M 460.
37. “Die Zukunft einer Illusion,” GS–1 433 (cf. LSEW W 204). Cf. “Zur Ideologie des poli-
tischen Zionismus,” GS–1 445: “Political Zionism is the organization of unbelief in
Judaism; it is the attempt to organize the Jewish people on the basis of unbelief.”
38. According to his closest friend Jacob Klein, these two problems were at the center off
the young Strauss’s attention. Compare Strauss, GA–JPCM – M 458.
39. Another indication is the fact that in his early writings Strauss frequently acknowl-
edges his indebtedness to Rudolf Otto’s The Holy. This seminal work initiated a
renewal of theology by restoring the transcendence of God as its primary object
and by identifying the irrational as the core of the divine. Already in his dissertation
Strauss appeals to The Holy, pointing out that Otto’s thought is substantially con-
nected to Jacobi by way of the German philosopher Jacob F. Fries. See, respectively,,
“Das Heilige,” “Biblische Geschichte und Wissenschaft,” and “Zur Auseinandersetzung mitt
der europäischen Wissenschaft,” GS–2 307–10, 357–62, and 341–50.
40. “Zur Bibelwissenschaft Spinozas and seiner Vorlaufer,” GS–1 404.
41. Die Religionskritik Spinozas, GS–1 265.
42. Die Religionskritik Spinozas, GS–1 166, 193–94, 225, 247. Cf. PLA 28–30.
43. Die Religionskritik Spinozas, GS–1 266n276. Regarding the “free spirit” that animates
modern theory, Strauss remarks: “it presupposes itself, like faith presupposes itself ”;;
GS–1 214. See also PLA 37; “Preface to Spinoza’s Critique of Religion,” JPCM M 151
and 172.
44. EPLJJ 247. Consider also the later Strauss’s well-known elaboration on Plato’s simile
of the cave, to the extent that modern thought has become trapped within a second
cave beneath the first. As he suggests, fear may have been at the origin of this event::
“People may become so frightenedd of the ascent to the light of the sun, and so desirous
of making that ascent utterly impossible to any of their descendants, that they digg
a deep pit beneath the cave in which they were born, and withdraw into that pit”;;
PAW W 155 (emphasis added).
45. EPLJJ 247–48.
46. Die Religionskritik Spinozas, GS–1 229–46.
47. EPLJJ 282n135.
48. EPLJJ 282.
49. See Chapters 2, 3, and 4 of Beiser, The Fate of Reason. In his account, Beiser names Strauss’s
introduction among the best treatments of the controversy. Compare Beiser, The Fatee
of Reason, 335n12. See also Alexander Altmann, Moses Mendelssohn: A Biographical Study
(London: Routledge and Kegan Paul, 1974) and Hermann Timm, Gott und die Freiheit::
Studien zur Religionsphilosophie der Goethezeitt (Frankfurt: Klostermann, 1974).The main
T H E P RO B L E M O F T H E E N L I G H T E N M E N T 189

documents of the controversy were edited and published in Die Hauptschriften zum
Pantheismusstreit zwischen Jacobi und Mendelssohn, ed. Heinrich Scholz (Berlin: Reutherr
and Reichard, 1916; reprint,Waltrop: Hartmut Spenner, 2004). A concise discussion off
Mendelssohn’s position in the controversy, critical of Strauss’s account, can be found
in Allan Arkush, Moses Mendelssohn and the Enlightenmentt (Albany: State University off
New York Press, 1994).
50. EMFL 531; cf. LSMM M 64. Cf. Beiser, The Fate of Reason, 61.
51. EMFL 572; cf. LSMM M 110.
52. EMFL 587; cf. LSMM M 126.
53. Cf. Moses Mendelssohn, Jerusalem, or On Religious Power and Judaism, trans. Allan
Arkush, with a commentary by Alexander Altmann (Hanover, NH: University Press
of New England, 1984).
54. EP 491; cf. LSMM M 35–36. Cf. PLA 44.
55. EMFL 583–86; cf. LSMM M 122–25.
56. EMFL 585; cf. LSMM M 125.
57. In PLA 78n28, Strauss renders this criticism more explicit by pointing out the
Hobbesian pedigree of Mendelssohn’s “surrender of the ancient natural right off
duty in favor of the modern natural right of claim.”
58. EMFL 573–74; cf. LSMM M 110f.
59. ESG 527; cf. LSMM M 160–61.
60. EMFL 578; cf. LSMM M 116–17. Challenging Strauss’s thesis, Arkush argues that
Mendelssohn “could have defended Judaism without downplaying the importance
or denying the possibility of philosophical knowledge of religious truths,” because
he never did “place such an absolute value on philosophical knowledge” in the first
place. Rather, Mendelssohn regarded the balance between reason and common sense
he tried to strike as a temporary settlement in anticipation of an ultimate demonstra-
tive proof of God’s existence. See Arkush, Moses Mendelssohn and the Enlightenment,,
88–93.
61. EMFL 581; cf. LSMM M 120.
62. EMFL 587; cf. LSMM M 127.
63. EMFL 575–76; cf. LSMM M 113–14.
64. EMFL 573; cf. LSMM M 110. At EMFL 577 (cf. LSMM M 115n304), Strauss comments
on Mendelssohn’s “pride in this progress [of metaphysics] and, at the same time, the
concomitant inability to understand the character of Aristotelian ethics, which had
been adopted by Maimonides.”
65. EMFL 588; cf. LSMM M 127. With slight alterations, I reproduce Beiser’s translation
of both quotations in The Fate of Reason, 88–89. On 88, Beiser aptly dubs Jacobi’s
doctrine an “epistemology of action.”
66. EMFL 588; cf. LSMM M 128.
67. For an account of this stage in Strauss’s development, see David Janssens, Between
Athens and Jerusalem: Philosophy, Prophecy, and Politics in Leo Strauss’s Early Thoughtt
(NY: State University of New York Press, 2008), 77–108.
68. EMFL 570; cf. LSMM M 107.
69. EMFL 542; cf. LSMM M 76–77. Compare the characterization of Mendelssohn’s posi-
tion in ESL 474 (LSMM M 15–16), with EMFL 556 (LSMM M 92–93). Cf. EPLJJ 282. Cf..
PPHH 145.
70. As Strauss observes, before admitting to Jacobi that for him “there is no other phi-
losophy than Spinoza’s,” Lessing had already qualified his commitment by saying, “Iff
I were to name myself after anyone, then I know no one better.” In a similar vein,,
to Jacobi’s avowal that “my creed is not in Spinoza,” he had rejoined ironically: “I
190 DAV I D J A N S S E N S

hope it is in no book,” that is, not even in Spinoza. See EMFL 546; cf. LSMM M 80–81..
Referring to Jacobi’s conversation with Lessing in “A Giving of Accounts,” Strauss
praises the latter as “the author of the only improvised live dialogue on a philosophic
subject known to me.” In the same context, looking back on SCR, he states: “In this
study, I was greatly assisted by Lessing,” that is, not Jacobi. See Strauss, GA––JPCMM
462.
71. ESG 522; cf. LSMM M 155.
72. Gotthold Ephraim Lessing,“Leibniz von den ewigen Strafen,” in Werke in acht Bänden
(München: Carl Hanser Verlag, 1979), 7: 180–83. Cf. Strauss, “Exoteric Teaching,” in
RCPR R 65; or appendix F in the present volume, p. 278. In his quotation, Strauss does
not reproduce the final sentence.
73. Lessing, “Leibniz von den ewigen Strafen,” Werkee 7:196.
74. PAWW 182.
75. Strauss, “Exoteric Teaching,” in RCPR R 64; or appendix F in the present volume,,
p. 276. Cf. Clemens Kauffmann, Strauss und Rawls. Das philosophische Dilemma derr
Politik (Berlin: Duncker & Humblot, 2000), 129–41.
76. In 1747, Lessing completed a comedy entitled Der junge Gelehrte. At the beginningg
of the play, the protagonist, a young scholar, is reading Maimonides’s Mishneh Torah..
See Lessing, Werkee 1:282. In the famous Fragments purposely published by Lessing,,
Hermann Reimarus praises Maimonides as “the wisest (verstandigste) of all Jews.” See
Lessing, “Von Duldung der Deisten. Fragment eines Ungenannten,” Werkee 7:322..
Compare Friedrich Niewohner, “Vernunft als innigste Ergebenheit in Gott. Lessingg
and der Islam,” Neue Zürcher Zeitung, no. 262 (2001): 83.
77. Strauss, “Plan of a Book Tentatively Titled Philosophy and the Law,” in JPCM M 470. In
the same context, Strauss announces: “While preparing the edition of Mendelssohn’s
metaphysical writings for the Jubilee-Edition of Mendelssohn’s works, I discovered
some unknown material which throws new light on that controversy.” However, it
is not clear from the text, nor does it become clear in the introductions what this
material consists of. Although the book was never published, there is fragmentary
evidence in Strauss’s Nachlass in the University of Chicago Library that he worked
on an interpretation of Nathan the Wise. Cf. Leo Strauss Papers, box 11, folder 7.
78. EMFL 535; cf. LSMM M 68f. Consider also Strauss’s comment on Mendelssohn’s art
of writing in casting Morgenstunden, as a dialogue: “A dialogue is a kind of drama; a
drama, being a product of poetry, is an ideal presentation of nature, in specific cases
an ideal presentation of real occurrences; and art is playful, whereas life is serious”;;
EMFL 590; cf. LSMM M 129.
79. In a letter to Alexander Altmann, written in 1971, Strauss acknowledges his great
debt to Lessing. Cf. Heinrich Meier, “How Strauss Became Strauss,” chapter 1 of the
present volume, p. 23. In 1937, Strauss wrote “Eine Erinnerung an Lessing,” a short
note in which he recorded that he did not find among his contemporaries “a single
man of Lessing’s spiritual freedom”; GS–2 607–8; cf. LSMM M 162.
C APT R 10
CHAPTER

“THROUGH THE KEYHOLE”: LEO STRAUSS’S


REDISCOVERY OF CLASSICAL POLITICAL
PHILOSOPHY IN XENOPHON’S CONSTITUTION
OF THE LACEDAEMONIANS

Richard S. Ruderman

T he decade of the 1930s was, for much of the civilized world, a dark andd
darkening period. Not only was the West in the grips of a Great Depression,,
but its self-confidence was at perhaps its lowest ebb. After all, European liberalism m
(and the individualism that was thought to be its most beautiful flower) was being
abandoned for Marxism and fascism; science—which was not a little implicated in
the deaths of millions in the Great War—was increasingly seen as both evil and as
a merely parochial Western view; and political science, having once been viewed as
the queen of the social sciences, was increasingly viewed as impotent, at least with
regard to generating the most needful thing of all, values. Collectivism in politics,,
irrationalism for the spirit, and a tired recounting of a tradition of political thought
that no longer could speak to the age—such was the “spirit of the age” that, alreadyy
a generation earlier, Spengler had dubbed the “Going-Under of the West.”
Toward the end of that decade, Leo Strauss, a then little-known German émigré
scholar to America, published his first article on classical political philosophy, an inter-
pretation of Xenophon’s Constitution of the Lacedaemonians.The immediate impression
of scholarly abstractedness, unaware of the gathering storms just outside his study,,
seems redoubled when we look back from the vantage point of the later Strauss, who
was anything but shy about openly invoking the “crisis of the West” and “the crisis off
political philosophy”—usually in the opening paragraphs of his writings—as grounds
for returning to the thought of classical antiquity.1 In this essay, however, he appears
to be silent about the state of the world in which he lives and writes. Moreover, in
selecting Xenophon’s Constitution of the Lacedaemonianss as his subject, Strauss seems
to be making every effort to evade, rather than to engage, the pressing issues off
political philosophy as then understood. After all, the greatest thinkers of the period,,
192 R I C H A R D S . RU D E R M A N

Nietzsche and Heidegger, had pronounced Socratic philosophy to be the decisive


wrong turn taken by mankind. And Xenophon was hardly viewed as a Socratic wor-
thy of Dekonstruktion: whatever his charms may be, the ability to grasp the intricacies
of Socrates’s most subtle teaching was not thought to be among them. Even if, in
some perverse way, that failing could be turned to Xenophon’s advantage—that is,,
if he could be shown to represent the poetic tradition of heroic overreaching that
Socrates had undermined or destroyed—Strauss’s choice of the Constitutionn fails to
capitalize on it: far from there being a commanding Cyrus in its pages, the Spartan
citizens (and even, to a considerable extent, its rulers) we meet there prove to be more
or less interchangeable and unheroic.
In contrast to the high drama associated with Nietzsche and Heidegger, then,,
Strauss’s first effort at a recovery of classical political philosophy seems modest to
a fault. And yet, toward the end of his essay, Strauss does briefly indicate what it is
about the contemporary political and philosophical situations that require overcom-
ing if, indeed, a return to the classics is to be viable. First, in the midst of explainingg
the necessity and purpose of esoteric writing, Strauss observes that esoteric writ-
ing disappeared “at a rather recent date” and that “its reappearance is simultaneous
with the reappearance of persecution” (535).2 This quiet allusion to the persecution
then rearing its ugly head in Soviet Russia and Nazi Germany is Strauss’s sole refer-
ence to the contemporary situation. Second, Strauss proposes that insofar as “the
restitution of a sound approach is bound up with the elimination of Rousseau’s
influence,” the writings of “men like Xenophon” are precisely the “antidote” we
need. Now, Strauss does not attempt to delineate the scope and meaning of the
“influence of Rousseau.” But, as the topic of the essay leads us to recall, Rousseau
was the most vocal and impassioned defender of “Sparta” in modern times. Nor was
Rousseau merely engaged in historical romanticism. Rousseau’s defense of “Sparta””
in the midst of the advance of enlightenment liberalism was closely connected to
his promotion of the “citizen” who was fully dedicated to his fatherland over the
“bourgeois individual” who combined the selfish exploitation of his fellow citi-
zens with a craven deference to popular tastes and public opinion. Insofar, then, as
Rousseau’s preference for Sparta had paved the way for both the Communist and
the Fascist efforts to subordinate the individual to the state, or the we-species, orr
the Volk, it could not have been more timely to reexamine the original meaning off
Sparta, or the allure of the “spirit of Sparta.”
What then is at issue in “The Spirit of Sparta or the Taste of Xenophon”? The
title, at a glance, would seem to consist of two ideas linked by a conjunctive “or”: we
need to recover something of the “taste of Xenophon” that ran toward “the spirit off
Sparta.” But the “spirit of Sparta” alludes not so much to the “actual Sparta of the
present or of the past” as to “the conviction that man belongs, or ought to belong,,
entirely to the city”: “the incarnation of the political spirit was Sparta” (531). The
spirit of Sparta, then, was alive and well, so to speak, in the “taste” of Stalin and off
Hitler (not to mention Marx and Heidegger).3 Indeed, insofar as modern liberal-
ism encourages a certain broad tolerance—tending to official indifference—toward
morality, contemporary critics of liberalism cannot but be attracted to a Sparta that
was, as Xenophon writes, “on the watch for those who are easy-going toward what
was bebelieved
eved to be tthee noble”
ob e ((Const. ac. 4.4).4 Furthermore, as Strauss makes clear
Const. Lac.
“ T H RO U G H T H E K E Y H O L E ” 193

in the course of his interpretation of the satirical nature of Xenophon’s encomium


of Sparta, the “true name of that taste which permeates Xenophon’s writings” is
“philosophy” (531). The “or” in the title is disjunctive: “Sparta and philosophy are
incompatible.” As a call for a return to “Spartan virtue,” Strauss’s essay fails—though
it succeeds in undermining Rousseau’s parallel call. But it succeeds in pointing to
the reasons why philosophy is the only available alternative to politics and life so
understood.
Strauss announces a preference for the inconspicuousness of Xenophon at the
start.This is not to say, however, that Strauss took Xenophon to be a lesser exemplarr
of the classical approach. As he wrote concerning his later book on Xenophon’s
Hiero in a letter to his friend Alexandre Kojève, he prefers to enter a room “through
a keyhole.”5 Xenophon, more than his more openly theoretical and seemingly doc-
trinaire fellow classics Plato and Aristotle, compels his readers to think for them-
selves. Or rather, he is happy to present them with this simple alternative: take me
at face value (and thus be charmed or even repelled by my apparent simpleminded-
ness) or make the effort to connect the dots, and cross the T’s in order to understand
the truth behind the surface, but misleading presentation. Now this, in itself, stands
opposed to Rousseau and his cult of sincerity. For Xenophon is ironic through
and through. Strauss’s admiration for such irony—and for esoteric writing more
broadly—has led more than a few contemporary critics to accuse him of harbor-
ing a hidden and fiercely political agenda (namely, promoting the secret rule of the
few). On the basis of this article, however, it is difficult to see how this could be so..
For Xenophon’s irony here is revealed to be, among other things, a political strategy
whereby he is able to mount a criticism of the Spartan regime—a regime that turns
out to be nothing other than the hidden rule of the few. Strauss shows through
his analysis that Xenophon chose to imitate in his writing the manner in which
Sparta—or rather its founder, Lycurgus—exercised and accomplished political
rule. To put it another way: Xenophon could have openly denounced the various
hypocrisies with which the Spartan regime was riddled. But this procedure wouldd
suffer from at least two serious defects. First, it would deny to readers the experi-
ence of uncovering those hypocrisies themselves. Since no regime or structure off
authority has ever openly announced that it was being hypocritical, Xenophon
wrote so that we might learn how to think through the various claims of author-
ity to uncover their hypocrisies. (The extent to which people must reconsider the
world they think they “know” is indicated by Xenophon in a comment he makes
about the general view regarding the complexity of the Spartan infantry formation::
“what most people imagine ...is the exact opposite of the truth [lit. of what is]”;;
11.5.) Second, unless one first feels the attraction of politics at its most alluring, one
cannot learn from the subsequent discovery of hypocrisies within it. When one is
taught something directly, textbook-style, there is no sense of what is missing; but
when one is attracted to a positive goal (here, the life of political virtue), one will
be like the Ephors described by Xenophon when outfitting the Spartans for war::
“anything missing is not at all likely to be overlooked” (11.2).
In order to uncover Xenophon’s teaching, Strauss pays particular attention to
the plan of his work. The Constitution consists of 15 chapters: the first ten chapters
out e tthee way o
outline of lifee tthat
at Lycurgus
ycu gus atte
attempted
pted to des
design
g foro tthee Spa
Spartans;
ta s; tthee next
e t
194 R I C H A R D S . RU D E R M A N

three turn to war; and the final two to the institution of monarchy at Sparta. From
this design alone, the reader is tempted to draw the conclusions that the Spartan
regime was designed to prepare the citizens for war; that the successful prosecution
of war is and ought to be the highest purpose of political life; and that the Spartan
king, as war-leader or general, is the ultimate ruler and even hero of Spartan life. As
Strauss shows, however, Xenophon quietly indicates problems with each of these
three theses by marring certain aspects of the plan of the work and by utilizingg
ambiguous expressions and references.
The most striking “flaw” in the book’s plan, Strauss argues, is the appearance off
Chapter 14, a chapter devoted to a harsh critique of contemporary Lacedaemonia..
For, it seems, Lacedaemonia has so fallen away from L Lycurgus’s original institu-
tions and laws that other Greeks, who used to “beg” her to lead them in the past,,
now band together to prevent Lacedaemonian hegemony (14.6). However effective
Lycurgus’s enactments were at the start—or were on paper—they have failed overr
time to accomplish their task. Let us examine with some care what Lycurgus sought
to accomplish and the manner in which he sought to do it.

Lycurgus’s Founding
As mentioned above, Lycurgus wanted a city devoted to virtue—or what he andd
perhaps most others “believed” to be the noble. Rather than permit the myriad
ways in which “all other cities” slacken in their pursuit of nobility, Lycurgus legis-
lated from cradle to grave—nay, from prior to birth to beyond even death—with
an eye toward developing the most “hardy” citizens, whose resourcefulness and
physical toughness would always be available to the city whenever it needed to be
at war. To produce the hardiest soldiers, Lycurgus promoted a unique and unre-
stricted diet for young girls with respect to food and drink. Spartan girls, that is,,
were encouraged to indulge their pleasures—their “animal natures” as Strauss says
(505)—and, as a result, their manners, especially in sexual matters, were lax. Now,,
Xenophon does not state any of this openly: he merely speaks of the strict andd
opposed general Greek practices. Moreover, while he speaks of the moral educa-
tion of Spartan men (which made them continent), he says not a word about the
moral education of the women. And he prepared not just the womb, but even the
sperm so as to produce the strongest offspring. For while he accepted a certain
moral looseness in the women in order to produce robust mothers, he required a
painful degree of self-control from the men to produce robust fathers: by making it
shameful for husbands to be seen entering or leaving their wives’s rooms, he made
sexual intercourse rarer and, he believed, more likely to produce vigorous children..
Lycurgus—and not only Lycurgus—believed that to deny ourselves pleasure is the
first step toward the noble.
Thus Xenophon introduces the Lycurgan approach toward what we today call
“gender differentiation.” The men, but not the women, are required to develop a
strong sense of shame with respect to pleasure (especially sexual pleasure) result-
ing, as Strauss notes, in the odd fact that Spartan men are “stronger” (or “better”)
at modesty than the women. Now, insofar as modesty is typically considered a
female
e a e vvirtue
tue (Strauss
(St auss draws
d aws our
ou attention
atte t o to Plato,
ato, Laws
aws 802e8–10), Xenophon
“ T H RO U G H T H E K E Y H O L E ” 195

thereby quietly raises the striking possibility that Spartan men, frequently regardedd
as the most manly, were in fact quite womanly in important ways. So complete
was Lycurgus’s desire for robust offspring that he was willing to denature both
male and female Spartans—and so eager was he to produce many such youngg
Spartans that he permitted both men and their wives a “surprisingly large freedom m
to indulge in adultery” (506). The denaturing, that is, was not completely success-
ful, and Lycurgus’s concessions to human nature led to a weakening of the family..
(When Xenophon goes on to note that sexual intercourse with boys was refrained
from “no less” than was incest, we cannot be too reassured, for the likelihood off
incest rises with the increased obscurity of familial relations.)
Strauss next turns to “Spartan education.” This education turns out to be entirely
“physical education”—there was apparently no liberal education in Sparta. Indeed,,
unlike the Persian education outlined at the beginning of the Education of Cyrus,,
the Spartan education did not even include an education in speech. One might say
that Sparta extended her teaching regarding continence even to the act of speech..
Xenophon’s apparent praise of this practice—which extends to his adopting a certain
reticence in speech himself when discussing Sparta—cannot be viewed as sincere
until we determine his attitude toward the importance of speech. Now Sparta, as
Strauss shows throughout his essay, ranked deeds (especially actions in war) as higherr
than speech. Was this Xenophon’s view? Not only was Xenophon, of course, a stu-
dent of Socrates, that master of speech, but he himself, in the course of leading the
retreat of the Ten Thousand Greeks from Persia, utilized speech often and with excep-
tional success. From the point of view of action or war itself, then, speech appears to
be insufficiently appreciated at Sparta. More than that, however, Xenophon raises the
question of whether not speaking of certain things is the best method by which to
minimize their presence or influence within the hearts and minds of the citizens.
Just as Xenophon suggests that speech (or thinking) may be superior to deedd
(or action), so too he holds that the soul is superior to the body (10.3). But while
Xenophon does mention the ways in which Lycurgus strove to improve the bod-
ies of the Spartans, he says next to nothing about how he sought to improve theirr
souls. Xenophon notes that Lycurgus felt “toil” was “an employment of the soul””
(7.4), but this suggests that keeping the soul distracted or exhausted (presumably by
keeping the body hard at work) is the safest way to handle its natural inclinations
toward “big ideas,” a certain “insolence,” and the pursuit of “pleasure” (3.2).6 (One
might note here that enlightenment liberalism’s promotion of equality, the private
sphere, and a generally instrumental view of politics is no less hostile to these erotic
longings of the soul than are its modern alternatives noted at the outset.) Strauss
notes that Xenophon’s sole reference to this undertaking consists of his “emphatic
statement” (512) that Lycurgus “compelled all [the Spartans] to practice all virtues
publicly” (10.4). He immediately goes on to note Xenophon’s silence regarding any
Spartan practice of the virtues of wisdom, justice, or moderation. And this should
have been apparent from the start: insofar as the Spartan education in virtue consists
of shaming and discipline alone, it cannot be expected to produce virtue, especially
those virtues that cannot be compelled.
The use of political authority, then, to produce virtue can, at best, produce only
“political virtue.” Such virtue is to genuine virtue as compulsion is to freedom.
196 R I C H A R D S . RU D E R M A N

This means that Sparta produces only such virtues as are valuable to the political
community, not virtues that perfect or fulfill the individuals in that community..
Rather than practice wisdom, moderation, and justice, the Spartans practice merely
continence, bashfulness, and obedience (514). Now, even where these virtues seem
similar, their differences are decisive. Continence might seem to share an “affinity””
with moderation, for example. But continence (or self-control) is less an actual vir-
tue than the foundation of virtues—and even, to be honest, of vices: thieves must
exercise self-control in order successfully to carry out their tasks. Moderation, on
the other hand, shapes the whole soul, as it were. It is practiced in the dark (when
unobserved) no less than in the light. And the Spartans, Strauss shows, cannot be
said to practice moderation, at least with respect to food, drink, and sex. The best
that Xenophon seems able to say of them in this regard is that they practice it with
respect to wealth. But here too there is a gap between Lycurgus’s legal intention
and actual Spartan practice. Lycurgus made every effort to limit the acquisition off
wealth. He made it a shameful practice and he even required that Spartan money
be so heavy as to make its private possession all but impossible. That he still hadd
to institute searches for gold and silver suggests that many Spartans could not be
compelled to give up their natural desire for wealth, but merely to seek natural
substitutes for the useless conventional money Lycurgus issued. In fact, Xenophon
quietly indicates not only the presence of wealthy men in Sparta but even theirr
ability and readiness to contribute purchased food to the common store in place off
the missing contributions of “those who are idle” (5.3). And while such generosity
might appear to compensate for their having amassed wealth against the spirit and
letter of the law (see 7.2–3), their capacity to purchase their contribution ratherr
than hunt for it means that they are able to buy their way out of hunting, which
Lycurgus wished to establish as the “noblest occupation” (4.7). This would seem to
supply a standing and devastating critique of Spartan nobility: not only does its holdd
over the citizens weaken when the necessity for it is removed, but its status as the
poor man’s substitute for wealth is all but broadcast.
The “spirit of Sparta,” it now becomes clear, consists in a suspicion of—orr
even hostility toward—self-concern and a corresponding effort to channel that
self-concern toward a more self-sacrificing nobility. That spirit, as we noted, was
alive in the 1930s and, in such movements as communitarianism, not to speak off
religious fundamentalism, is alive today. It becomes of pressing concern, then, to
determine whether Xenophon viewed Lycurgus as a failed attempt at carrying out
an essentially admirable project, or whether he viewed the project itself as somehoww
flawed.

The Problem of the Noble


Lycurgus, to repeat, wished to keep a watch out for those who were “easygo-
ing toward what is believed to be the noble” (4.4). That is to say, he wished to
inculcate what today is known as “moral seriousness.” But perhaps there is a gap
between “what is believed to be the noble” and genuine nobility. Let us recon-
sider Lycurgus’s view of nobility. So important to him was nobility that, despite
the general Laconic opposition to speech, Lycurgus designed mixed generational
“ T H RO U G H T H E K E Y H O L E ” 197

mess tables to encourage the kind of conversation that would turn on recountingg
“noble deeds performed in the city” (5.6). Even Lycurgus, then, concedes that
speech is needed to identify, praise, and even exhort to the performance of noble
deeds. But while Xenophon concludes the parallel passage that explains Lycurgus’s
efforts at developing the “legs, arms, and necks” of the citizens with the remark that
“he succeeded” in doing so, he is silent with regard to the “success” of mess-table
conversations about the noble (cf. 5.9 and 5.6). Were no such deeds performed that
could be recounted? Or is the recital of noble deeds exclusively doomed to fail?
That is, must not the “shameful” deeds (apparently buried in silence by Lycurgus)
also be recounted, not least to help distinguish what it is that nobility consists in?
Moreover, must not noble speeches (or at least reasoned speeches about the noble
deeds) also be offered if nobility is to be properly understood and perforce per-
formed? Strauss summarizes Lycurgus’s accomplishments as follows:

By educating the Spartans in bashfulness only, while withholding from them true
education—education in letters and speech, education in wisdom and moderation
and justice—in other words, by frightening them into submissiveness with the men-
ace of severe and dishonoring punishments, he compelled them to do forbidden
things in utter secrecy. (517)

This means, Strauss concludes, that the “famous Spartan sense of shame is ...simply
hypocrisy.” The decline in Sparta is not then, strictly speaking, a decline in Spartan
virtue: it is merely a decline in their willingness or ability to dissimulate or to hide
their deviations from the publicly permissible virtues.
There remains to be considered one final potential virtue of the Spartans,,
namely the most vaunted of their virtues, manliness or courage. Oddly, Xenophon
mentions this word only once in the entire treatise—and there in so “exceedingly
ambiguous” a usage (520) that most editors amend it (with no manuscript author-
ity whatsoever) to its opposite (9.5; cf. 4.2 for a word whose root is “manly”)! The
one appearance of “manliness” comes in a passage devoted to explaining the endless
shaming that must be faced by those Spartans who fail to live up to the noble stan-
dard of life in Sparta, above all to face willingly a “noble death” (9.1). Such losers are
left out when teams pick sides; must give way in the street even to their juniors; and
“must support their spinster relatives at home and must suffer the blame for [or sup-
ply the cause of] manliness” (9.5). Now, unraveling the meaning of this exceedingly
odd phrase is not easy. How would one explain the “cause of manliness”? We learn
at the start of this brief chapter that manliness is caused by a readiness to “choose a
noble death over a dishonorable life” (9.1). And that readiness is secured, as the end
of the previous chapter suggests, by Lycurgus’s having secured the Delphic Oracle’s
support for his laws. By doing this, Xenophon tells us, Lycurgus ensured that refus-
ing obedience to the laws would henceforth not only be illegal, but “impious” (8.5)..
Thus, Spartan manliness, the willingness to accept or even seek out a noble death,,
is connected to religious belief. Indeed, the most Xenophon can say in defense off
the willingness to die is that “more” people survive when soldiers are animated by it
(not all or even the most brave are guaranteed survival) and glory (as well as politi-
cal followers)) will attend those who do survive. Manliness “accuses” in the same
198 R I C H A R D S . RU D E R M A N

sense that the gods “accuse” those who fail to live up to their demands. Lycurgus
wanted fear of the gods, no less than fear of the regime, to help make men “will-
ingly” obedient.
The final theme of Strauss’s essay, accordingly, is the connection between
“political life” and “belief in the gods of the city” (532).While Lycurgus taught the
Spartans to believe that the Delphic god had given them their laws, Strauss observes
that Xenophon “distinguishes between the Spartans’ obedience to Lycurgus’ laws
and their obedience to the god” (532).7 For, Strauss continues, Xenophon raises
the question of whether and in what manner the gods can interact or commu-
nicate with humans. Through the use of a simple trope, Xenophon compares the
Spartans’ feeding of their children with, first, their feeding of “the king and those
with him” and, finally, their offering of “sacrifices to Zeus and to those with him””
(533). In suggesting that the Spartan sacrifices amounted to a “feeding” of the
gods for their own political purposes, Xenophon delivers his ultimate criticism m
of the Spartan constitution: the Spartans wish, through their sacrifices, to “seize
beforehand [prolambanein
[ ]” the gods’ goodwill just as they learned as children to
“steal [kleptein]” something to alleviate their hunger (Strauss encourages us to
compare 13.3 with 2.7). The Spartans’ understanding of the noble, it appears, is
fundamentally self-contradictory: the sacrifices they make in an effort to appearr
worthy of the gods’ aid are, in truth, nothing but an underhanded effort to seek
their own advantage.

The Meaning of Lycurgus’s Authority


As we have seen, the place of the missing speeches about the noble is taken by the
rich men’s act of buying their way out of the noble activity of hunting, that is, by
a silent or implicit critique of nobility. Now, their very act of becoming rich sug-
gests a flaw in Lycurgus’s basic teaching of obedience to authority. Accordingly,,
Xenophon next turns to an examination of how authority was constituted at
Sparta and in what it consists. We here learn that Lycurgus did not attempt to
introduce discipline until he had first secured “like-mindedness” among the best
(or strongest) men in the city (8.1). Now, like-mindedness is a highly ambigu-
ous term. For, as the case of the proverbial band of robbers reveals, a group can
be “like-minded” about base ends no less than noble ones. Obedience to rulers,,
as well as like-mindedness within the ruling class, is a virtue only insofar as those
rulers understand and properly enact wise laws. And, as we have seen, Xenophon
raises doubts about the wisdom of many of Lycurgus’s laws. But Strauss here raises
an important objection. Did not Xenophon’s Socrates teach that obedience as such,,
obedience to any laws, is identical to justice (Memorabilia 4.4.15)? And, we might
add, did not Xenophon leave ambiguous at the very start of his treatise whetherr
the Spartans owed their happiness to the quality or wisdom of Lycurgus’s laws—orr
merely to their obedience to them (1.2)? Strauss responds to this objection with a
digression explaining Xenophon’s purpose and procedure in the Memorabilia. Afterr
noting that its purpose was to reveal only Socrates’s speeches and deeds (and not his
thoughts), Strauss goes on to suggest that Socrates’s stated thesis about obedience to
(any) law was, at best, a political teaching and not a philosophical one. Moreover, he
“ T H RO U G H T H E K E Y H O L E ” 199

demonstrates, Socrates himself includes in his discussion of obedience the relevant


information that would enable his interlocutor to refute that thesis: there exist
“natural” or “unwritten” laws that function as standards by which to judge (and to
find wanting) any existing, written laws.
In what way, then, were the leading citizens “like-minded”? Xenophon explains
they were not only of one mind with one another, they were like (in their fear off
authority) the rest of the Spartans. Whom, then, did they fear? It is at this point in
the discussion that Xenophon quietly introduces the office of the Ephors (8.3) who
turn out to be the hidden rulers of Sparta. For the Ephors have more or less unlim-
ited authority—they can even fine or imprison the magistrates, meaning that they
(and they alone) are superior to the laws in Sparta. They behave, says Xenophon,,
like “tyrants” (8.4). The so-called wisdom of the Spartan laws, then, amounts to
nothing more than a recognition that those laws are so inadequate to their task (off
training citizens to prefer the city to themselves) that they must be supplemented
with a tyrannical admixture of fear and shame. And finally, we should resist the all-
too-natural conclusion that “like-mindedness” among the citizens is tantamount
to “unity” or “harmony” among them. For insofar as they are like-minded about
Lycurgus’s teachings (e.g., that “stealing is good,” that wealth enables one to escape
from some of the onerous demands of nobility, that surviving until old age brings
honor), the Spartan citizens are in fact prone to be in an endless contest with one
another. There is, we learn, rampant dissension, rivalry, and even spying on one
another. As Strauss observes, the classics taught that “one cannot assert that warr
against other cities is the aim of the life of the city without being driven to assert
that war of individual against individual is the aim of the life of the individual””
(524–25).
Xenophon concludes the section on Lycurgus’s institutions (Chapters 1–10)
with the summary praise of Lycurgus for laying an “irresistible necessity to practice
the whole of political virtue” on the citizens (10.7). This praise, we can now see, is
less enthusiastic than it may at first seem. For not only is “public virtue” (i.e., the
virtue that one practices in public and for the sake of the public; 10.4) a rather poorr
cousin to genuine virtue, but it is hard to see what virtue—other than obedience— —
can arise as a result of compulsion.8 Moreover, the irresistible necessity is apparently
not utterly irresistible: Xenophon immediately goes on to speak of the different
fates of those who comply with it and those who do not. And while those who
do not comply are made unhappy, those who do are merely given an equal share
(with the others who do). It should come as no surprise that, while all “praise” such
institutions, no city chooses to imitate them (10.8).
Strauss, in his concluding remarks on this section, focuses on the identity of the
“leading citizens” and of “Lycurgus” himself. Not only, as we have already noted,,
do the Ephors secretly rule (and rule like tyrants), but there were apparently vari-
ous “like-minded” powerful men from the city who accompanied Lycurgus on his
trip to Delphi. Who, Strauss asks, were these men? It seems, he suggests, that they
were the group that became the Ephors. And it is not clear how or to what extent
Lycurgus can be distinguished from this group. Lycurgus, Strauss proposes, “did not
exist at all”: he was a “mere name covering something much less solemn than an
almost divine lawgiver belonging to a remote and venerable past” (527). (Strauss
200 R I C H A R D S . RU D E R M A N

finds further textual proof of this at 13.10, where, according to the un-amended
good manuscripts, “the Lycurgus with regard to [pitching the tents] is the king.””
“‘Lycurgus’ is, then, a name designating authority or the men in authority”; 527.)
Finally, Strauss indicates that Xenophon leaves it to the reader to conclude that the
Ephors themselves either are or are in the sway of the wealthy Spartan few. This
would-be aristocracy of virtue turns out to be a hypocritical or hidden oligarchy.
The second section, on war (Chapters 11–13), follows. In this section, Xenophon
tries to make the case that the Spartans “have omitted least what is necessary in militaryy
things” (12.7). Of course, this silently implies that they may have omitted quite a lot off
what is necessary in the nonmilitary things. Furthermore, in noting the extraordinaryy
power of the Ephors to oversee the others (even the king), Xenophon suggests “onlyy
the Lacedaemonians possessed the art of war” (13.5). Now, despite giving a relativelyy
detailed account of how the Lacedaemonians prepare for war, Xenophon offers not
a word on any successful, heroic, or noble military deeds by any Spartans (and this
despite their recent victory in the Peloponnesian War). Readers of Thucydides will
not be so surprised by this, insofar as he relates very few distinguished Spartan actions
in that war (Brasidas, the least characteristic Spartan, being the sole exception), andd
suggests that, to no small degree, Athens defeated herself.
The final section, on the Spartan institution of kingship, is surprising for several
reasons. First, we discover that there are two kings, not one. Second, and more
important, it is here (in Chapter 14) that Xenophon unleashes his harsh criticism
of present-day Sparta, thereby ruining the effect, carefully nurtured from the begin-
ning, of an impassioned praise of Sparta. Regarding the king, we learn that they
are not quite the true power in Sparta—they remain subordinate to the Ephors..
Moreover, the Spartan constitution itself “incentivizes” the kings to go to war::
while they have “power and honor” in wartime, they have only “honor” in peace-
time (cf. 13.1 and 15.8).
Strauss ends with a consideration of why Xenophon would hide his criticism m
of Sparta—and why he would hide it somewhat ineptly. First, an open criticism
of Sparta would easily be mistaken for praise of Athens—and this, Strauss notes, is
something that Xenophon, in the aftermath of Socrates having been put to death
by Athens, would be hesitant to do. More than the particular praise of Athens that
he wishes to withhold, however, Xenophon does not wish to praise any other polit-
ical system in comparison with Sparta. For, in the final analysis, Xenophon wishes
to critique political life tout courtt from the point of view of the contemplative life..
And this means that, beneath his hidden critique or satire of Spartan “education””
lies a still more hidden defense or promotion of true, philosophic education. Now,,
as we recall from the start, this apparently antiquarian essay is, in fact, of no little
relevance for the age in which Strauss wrote it. For enlightenment liberalism as well
as its contemporary political enemies thwart the “natural” taste for “big ideas” orr
philosophy. Moreover, his later reputation notwithstanding, Strauss clearly taught
that war must be for the sake of peace and not vice versa.Would not the times then
call for an open account of Xenophon’s philosophic teaching? Strauss, of course,,
does not address or even raise this question. But he is clearly no fool. Germany was
in no mood—not even its leading philosopher was in the mood—to hear about the
primacy of the contemplative life over the life of war and deadly decision-making.
“ T H RO U G H T H E K E Y H O L E ” 201

And England and America needed at this time to hear, not of the ultimately supe-
rior value of peacetime, but of the grim necessity of fighting the given war that was
about to be foisted on them.
What Strauss does see fit to share with his readers are a few brief comments
about the character of a philosophic education. Such education is premised on the
understanding that man does not belong “entirely” to the city.9 This was, to repeat, a
somewhat controversial if not fantastic assertion to make in 1939. For it was denied
not only by the various illiberal politics of the day, but by the leading philosophi-
cal school of the day, namely historicism. Man, it taught, was wholly a creature off
his society, of his times, of the (temporary) world-view whose creature he inevi-
tably had to be. Even at the end of the twentieth century, leading American post-
modernist Richard Rorty could still assert man is “socialized all the way down.”10
Twentieth-century man, Strauss implied, needs to begin to relearn, in an almost
humble fashion, the elementary truths about human nature, insofar as his under-
standing of such truths is “directly opposite to ...what is” (11.5). And, because the
authors most conversant with those truths insisted on writing esoterically, he needs
to relearn how to read such literature. I close by simply quoting Strauss’s wonderful
tribute to Xenophon’s own peculiar type of esotericism:

[S]uch a man was he that he preferred to go through the centuries in the disguise
of a beggar rather than sell the precious secrets of Socrates’ quiet and sober wisdom
to a multitude which let him escape to immortality only after he had intoxicated it
by his artful stories of the swift and dazzling actions of an Agesilaus or a Cyrus, or a
Xenophon. (536)

Notes
1. See, for example, NRH H 7–8.
2. All page references are to SSTX.
3. Sparta had its less extreme, and thus perhaps more influential, champions as well,,
such as William Mitford, whose popular History of Greecee (1836) had a strongly pro-
Spartan and anti-Athenian bias.
4. All chapter and paragraph references are to Xenophon, “The Constitution of the
Lacedaemonians” in Scripta Minora, trans. E. C. Marchant (Cambridge, MA: Harvardd
University Press, 1968). I have occasionally altered the translation for greater adher-
ence to the literal meaning of the text.
5. Letter to Kojève, August 22, 1948, in OT T 236.
6. Note also that 8.4 suggests the Spartans understand by “soul” nothing more than the
life-giving force in the body.
7. Rousseau concedes that legislators (such as Lycurgus), who are “unable to use eitherr
force or reasoning” to form a people, must always have “recourse to the interven-
tion of heaven” in order to have the people abide by laws prior to their having been
formed by them (Social Contract, 2.7).
8. For a similar critique of “political virtue,” see Plato, Republicc 430c.
9. Cf. NRH H 5: “there is something in man that is not altogether in slavery to his
society.”
10. Richard Rorty, Contingency, Irony, and Solidarity (Cambridge, UK: Cambridge
University Press 1989), 185.
C APT R 11
CHAPTER

STRAUSS AND SCHLEIERMACHER


ON HOW TO READ PLATO:
AN INTRODUCTION TO “EXOTERIC TEACHING”

Hannes Kerber

What philosophy is seems to be inseparable


from the question of how to read Plato.
—Seth Benardete

“ E xoteric Teaching” is arguably one of the most important finds among the Leo
Strauss papers. As his first attempt at an explanation of the concept of “exo-
tericism” in general terms, the fragment enables the reader to cast a unique glance
at the genesis of Strauss’s hermeneutics. Its peculiar approach distinguishes the essay,,
which was written in December 1939, not only from its next of kin, “Persecution
and the Art of Writing,” but from most other studies, written and published mainly
during the 1940s, in which Strauss applied his thesis of exoteric writing: While the
main subjects of these articles are writers from the Jewish and Islamic Middle Ages,,
“Exoteric Teaching” takes a different route by presenting the issue exclusively from
what may loosely be called a Western perspective.1
“Exoteric Teaching” discusses the German philosopher Gotthold Ephraim
Lessing and the Protestant theologian Friedrich Daniel Ernst Schleiermacher, while
mentioning such writers as Aristotle, Leibniz, Zeller, Kant, Ferguson, Rousseau, and
Jacobi. The essay avoids naming or citing any non-Western thinker. This silence is
particularly puzzling in the case of Moses Maimonides, who played a crucial role in
Strauss’s discovery of exotericism well before he wrote “Exoteric Teaching.” Even
more perplexing is the fact that, while Maimonides is left out of the typescript, he
is mentioned in an early handwritten plan of the essay. In this plan, the title for the
penultimate section of the essay’s first part is: “Lessing—Leibniz—Hobbes (vera— —
pia dogmata)—Spinoza—RMbM—” (p. 287.)2 In the corresponding section off
the manuscript, which was written before the typescript, Strauss replaces Thomas
Hobbes with René Descartes but does mention Maimonides at the end of the
204 HANNES KERBER

sequence of exoteric writers: “Leibniz is [ ...] that link in the chain of the tradition
of exotericism which is nearest to Lessing. Leibniz, however, was not the only 17th
century thinker who was initiated. Not to mention the prudent Descartes, even
so bold a writer as Spinoza had admitted the necessity of ‘pia dogmata, hoc est,,
talia quae animum ad obedientiam movent’ as distinguished from ‘vera dogmata.’’
Despite, or because of, that admission Spinoza rejected Maimonides’s allegorical
interpretation of the Bible as ‘harmful, useless and absurd.’ Thus, he cannot be con-
sidered a genuine spokesman of the tradition.” (p. 286n117.)3 The last two sentenc-
es—with the ambiguous personal pronoun that could refer either to Maimonides
or to Spinoza4—were not transcribed by the typist and were also not reinserted
later by hand. Whether this omission was a deliberate and, hence, authoritative
decision or one of the many blunders of the typist overlooked by Strauss, cannot be
determined with certainty since “Exoteric Teaching” was never prepared for publi-
cation. Another and more detailed plan of the essay slightly changes the title for the
passage in question but does not resolve the problem: “Lessing—Leibniz—Spinoza
(—RMbM)” (p. 292). How should one interpret the fact that here Strauss has
put Maimonides’s name in parentheses? Do the parentheses indicate that Strauss
wanted to include the ambiguous suggestion about Spinoza’s critique? Or do they
show that Strauss decided to drop the reference to Maimonides altogether afterr
writing the manuscript? However these questions might be answered, the textual
difficulties highlight the fact that non-Western thinkers would have played only a
minor role, if any, in “Exoteric Teaching.”
The peculiar approach of “Exoteric Teaching” is reflected not only in the essay’s
silence about Maimonides, but first of all in its focus on Lessing and Schleiermacher..
These two thinkers seem to have been important for Strauss during the 1930s, but
are not prominently featured in his published works. Recalling his early studies off
Spinoza, Strauss remarked decades later that at this time “Lessing was always at my
elbow.” The influence of Lessing’s writings, according to Strauss’s own account,,
was subtle and initially unrecognized: “I learned more from him than I knew at
that time. As I came to see later Lessing had said everything I had found out about
the distinction between exoteric and esoteric speech and its grounds.”5 This laterr
reconsideration of Lessing, which took place probably around 1936–37,6 shows
that the young Strauss’s thought was ripe for the discovery of exotericism longg
before he began to write “Exoteric Teaching.”7 However, this essay for the first
time combines a general discussion of philosophico-political reasons for exoteric
communication with a sketch of the fate of exotericism in ancient times and in
modernity.
The essay’s most substantive part is a discussion of Schleiermacher’s way of read-
ing Plato and his refusal to interpret Plato as an exoteric writer. The fine intro-
duction, which Schleiermacher wrote for the first volume of his pathbreakingg
translation of the Platonic dialogues into German, shows that he, like many earlierr
writers but unlike many of his successors,8 was very much aware of exoteric writ-
ing.9 While doing away with what he sees as the most common misconceptions off
earlier interpretations, Schleiermacher reports in rather baroque German that past
interpreters of Plato “have formed the opinion, partly from individual statements
o Plato
of ato himself
se [a[and]
d] pa
partly
t y from
o a w widespread
desp ead ttradition,
ad t o , which
w c has as preserved
p ese ved
S T R AU S S A N D S C H L E I E R M A C H E R 205

itself since antiquity, of an esoteric and exoteric [dimension] in philosophy, as iff


his proper wisdom were either not at all contained in Plato’s writings or only in
secret allusions, which are difficult to locate.”10 Schleiermacher’s attempt to refute
the opinion that Plato wrote exoterically is complex because, according to his
account, the concept is “utterly vague” and “has cultivated itself in the most mani-
t 11 Yet, two views off
fold forms (hat sich in die mannigfaltigsten Gestalten ausgebildet).”
exotericism stand out for Schleiermacher: The pre-Platonic “Pythagorean” view,,
on the one hand, declared certain subjects to be “esoteric,” such as the political
teaching that the Pythagoreans did not wish to discuss in public.12 In the view off
those post-Platonic thinkers who mixed Socratic philosophy with sophistry, on the
other hand, the distinction between exoteric and esoteric concerned not so much
specific subjects as a manner of presentation:What could not be explained in a popularr
lecture, but could very well be explained in a lecture for an expert audience, was
called “esoteric.”13 Schleiermacher strongly denies that either of these views is cor-
rect in the case of Plato: “in whichever of the two senses one would want to apply
these concepts to the Platonic writings and philosophy in order thereby to divide
both into two parts, one will get caught up everywhere.”14 In Schleiermacher’s
eyes, the latter view—according to which there is one genre of philosophic writ-
ings that is easy to understand, that is, popular or “exoteric,” and another genre off
philosophic writings that is hard to understand, that is, scientific or “esoteric”15—is
futile with regard to Plato, because all of Plato’s writings are hard to understandd
and therefore he “could have confided his most difficult and most mysterious wis-
dom [to these writings] just as well.”16 The Pythagorean view, on the other hand,,
which implies that Plato spoke about certain issues only within the confines of the
Academy and on principle never in writing, almost refutes itself because “in the
field of philosophy nothing might be found on which a judgment could not be
encountered in [Plato’s] writings, [expressed] either directly and clearly or at least
according to the reasons (den Gründen nach).”17
Setting aside other differences for the moment, one cannot fail to notice that
Strauss, who does not explicitly mention Schleiermacher’s critique of these two
versions of exotericism in “Exoteric Teaching,” agrees with him in rejecting both
of them. Like Schleiermacher, Strauss would object to those interpreters of Plato
who follow (to use Schleiermacher’s façon de parlerr) either the “Pythagoreans” orr
the “Socratic sophists.” While Strauss and Schleiermacher are convinced that the
dialogues are the sole basis for understanding Plato’s teaching, the two groups in
question give up this material completeness of the dialogues: The Pythagoreans do not
rely on the dialogues but have recourse to an oral tradition. Orality, while being the
original form of philosophic communication and therefore a conceivable option
for the transmission of the esoteric teaching,18 has major defects: Not only does
oral communication require permanent political stability19 and perfect compre-
hension on the part of each successor, it also limits the potential audience to those
who happen to be a link in the chain of the tradition.20 In a word, oral traditio is
very unlikely to achieve the goal of preserving over time an undistorted traditum
for the intended audience.21 While the Pythagoreans tear apart the material com-
pleteness of the dialogues by assuming the existence of an oral tradition that com-
municates
u cates certain
ce ta “esoteric”
esote c subjects independently
depe de t y oof the
t ew writings,
t gs, the
t e Socratic
Soc at c
206 HANNES KERBER

sophists make the very same mistake in a different manner. According to their view,,
“exoteric” books are intended for nonspecialists outside the school, while otherr
writings are “esoteric,” that is, intended only for the students within the school.The
main task for the interpreter is therefore to distinguish between the two genres and
to disregard22 all books that belong to the former since they contain exclusively
“exoteric” matters that are philosophically well-nigh worthless.23 This rather crude
way of dividing the twofold audience by membership cards is problematic, not
only because every successful school has to include a variety of disciples24 but also
because it is to be expected that all writings become public eventually. Its political
precariousness alone renders this “exotericism” unusable. Hence, Strauss was aware
that if there is any need at all for exotericism, “no written exposition can be strictly
speaking esoteric.”25
After debunking these two views of exotericism in the introduction to his trans-
lation of Plato’s dialogues, Schleiermacher mentions a third view that he does not
ascribe to a specific group and that for him is not even worth discussing in detail::
“And those indeed who trace back the distinction between the esoteric [and the
exoteric] merely to the quarrel with polytheism and with the popular religion, in
fact overturn (aufheben) [the distinction] completely and either make it into a legal
safe-keeping (rechtliche Verwahrung)—which
g would be highly insufficient because
Plato’s principles on that topic are clear enough to read in his writings, so that
one can scarcely believe that his pupils might have needed further instruction, the
publication of which he shied away from—or into a childish performance (kindischee
Veranstaltung),
g which delights itself in saying in a loud voice behind closed doors
what in fact may also be said publicly but only in a low voice.”26 This provides the
first weak point for Strauss’s attack on Schleiermacher’s argument in “Exoteric
Teaching”—only because Schleiermacher refrains from calling a spade a spade is
he able to lend credibility to the view that Plato abstained from practicing “legal
safe-keeping,” that is, that Plato abstained from drawing a veil over his denial of the
gods of the city even though such impiety was forbidden in Athens under penalty
of death.27 The ambiguity of the expression “polytheism and the popular religion””
enables Schleiermacher, according to Strauss, to claim that Plato’s principles on the
topic are easily recognizable in his writings. If, however, “Schleiermacher had used
the less ambiguous expression ‘belief in the existence of the gods worshipped by
the city of Athens,’ he could not have said that Plato’s opposition to that belief is
clearly expressed in his writings.” (p. 280) Thus, Strauss not only calls into question
Schleiermacher’s criticism of exotericism for reasons of political caution, but he also
provides a toehold for that argument in favor of exotericism that he would laterr
elaborate most prominently in “Persecution and the Art of Writing”: If a writerr
wants to conceal an opinion from the censor for fear of persecution but neverthe-
less wants to communicate his true thoughts to the reader, he must write in a way
that achieves both ends.28
However, Strauss’s most important objection to Schleiermacher’s criticism is
a different one. In “Persecution and the Art of Writing,” Strauss makes clear that
Schleiermacher’s major point against those interpretations that distinguish between
Plato’s exoteric and esoteric teachings is his “unusually able argument [ ...] that there
iss only
o y onee Platonic teac g. 29 In hiss writings
ato c teaching.” w t gs o on Plato,
ato, Sc
Schleiermacher
e e ac e ttimee aand d
S T R AU S S A N D S C H L E I E R M A C H E R 207

again emphasizes the unity of Plato’s thought, which is reflected in the unity of his
teaching.30 Those who distinguish between exoteric and esoteric teachings, in his
view, are to be blamed because they are tearing this unity apart.31 At the same time,,
Schleiermacher does not simply brush aside the difficulties of interpreting Plato,,
which have led some scholars to deny the unity of Plato’s philosophy or presenta-
tion. According to “Exoteric Teaching,” Schleiermacher explains the impression off
multiplicity in Plato that has misled many scholars by asserting that while there is
only onee Platonic teaching, “there is, so to speak, an infinite number of degrees of thee
understanding of that teaching: it is the same teaching which the beginner understands
inadequately, and which only the perfectly trained student of Plato understands
adequately.” (p. 280)32 The seeming multiplicity that overshadows the underlyingg
unity is then due to the process of understanding. In this process, the reader starts
out with an imperfect and therefore fragmentary understanding of the teachingg
that will then improve ad infinitum gradually and continuously. While the under-
standing thus changes, the teaching that is understood remains the same. Plato
ensures, according to Schleiermacher, that this ongoing enhancement of the read-
t 33 Plato forces
er’s understanding has the character of a “self-activity (Selbsttätigkeit).”
the reader to think for himself by writing in the literary form of a dialogue, that is,,
by not teaching directly what the reader is searching for in a treatise but by “expos-
ing the [reader’s] soul to the necessity to search for [the end of the investigation]
and to guide [the soul] to the way on which it can be found.”34 For this purpose,,
Plato, in Schleiermacher’s presentation, first brings the reader to realize his state off
ignorance and then relays the actual thought to the reader indirectly in two ways..
This indirect communication is achieved either 1) by “weaving contradictions into
a riddle to which the intended thought is the only possible solution” and givingg
“in a seemingly strange and accidental manner such allusions which will be foundd
and understood only by him who searches genuinely and independently,”35 orr
2) by dressing up “the actual investigation with another [investigation], not as iff
with a veil but as if with a grown-on skin (angewachsenen Hautt), which conceals
from the inattentive [reader] (dem Unaufmerksamen) but only from him that which
actually ought to be observed or found, but which for the attentive [reader] (dem
Aufmerksamen) sharpens and chastens the sense for the internal coherence.”36 At first
sight, Schleiermacher’s remarks about Plato’s two sets of literary devices—on the
one hand, teaching through riddles, contradictions, and brief indications, and on
the other, a meaningful differentiation of fore- and background of the text—seem m
to be very close to Strauss’s own way of reading Plato. This proximity shows above
all that, unlike many other interpreters, both Schleiermacher and Strauss believe
that the literary form of the dialogue, which merges the advantages of spoken and
written communication, is Plato’s answer to Socrates’s critique of writing in the
final exchanges of the Phaedrus.37 Closer consideration reveals, however, a deep
disagreement between Schleiermacher and Strauss, which manifests itself in two
ways. First, while Schleiermacher stresses the continuous character of the process
of understanding, according to Strauss, understanding is characterized not by con-
tinuity but by discontinuity. Second, what Schleiermacher takes as an indefinite
multiplicity of degrees of understanding in Strauss’s view turns out to be a strict
dup c ty of
duplicity o the
t e Platonic
ato c teaching.
teac g.
208 HANNES KERBER

The continuity of understanding is that premise that in Strauss’s eyes has caused
Schleiermacher to misinterpret Plato: “Schleiermacher tacitly assumes that the way
from the beginning [of the process of understanding] to the end is continuous,,
whereas, according to Plato, philosophy presupposes a real conversion, i.e. a total
break with the attitude of the beginner” (p. 281).38 Against Plato, Schleiermacherr
describes the structure of understanding not as a periagōg ō ē but as continuous. While
Schleiermacher could argue that he was forced to depart from Plato’s view (since
Plato was wrong about the structure of understanding), Strauss shows the price
Schleiermacher had to pay for understanding Plato supposedly better than Plato
understood himself. Schleiermacher’s assumption of continuity made him not only
blind to “the difference between the morality of the beginner and the morality off
the philosopher [ ...] which is at the bottom of the difference between exoteric
and esoteric teaching” (p. 282),39 but also to the very form of exotericism that,,
in Strauss’s eyes, is employed by Plato. This form of exotericism, to come to the
second point of disagreement between Schleiermacher and Strauss, is compatible
with both the material completeness of the dialogues and the unity of the author’s
thought, because it comprises in one and the same text two kinds of teaching,,
each of which addresses a different audience.40 One teaching is conveyed by the
explicit statements that the author deposits in plain view on the surface of the text
while the other teaching is indicated only “between the lines.” This duplicity off
presentation is the reason why the process of coming to understand such a text is
characterized by discontinuity rather than continuity.The advance from the surface
teaching to the hidden teaching is something like a turnaround. This metaphorical
description of the hermeneutical experience is itself in need of interpretation and
may be best elucidated by using an example.41
In “Persecution and the Art of Writing,” Strauss lists several concrete disagree-
ments between modern historians and their predecessors rooted in the principles off
modern scholarship.42 One of these examples is the gulf between the older and the
more recent understandings of Thomas Hobbes’s attitude toward religion: Strauss
asserts that while many earlier “philosophers and theologians believed that Hobbes
was an atheist,” many contemporary historians “tacitly or explicitly reject that
view.”43 In a footnote to this statement, Strauss refers to the studies of five contem-
porary scholars who oppose the older opinion that Hobbes was an atheist. Among
these studies is Strauss’s own book Die Religionskritik Spinozas als Grundlage seinerr
Bibelwissenschaft, which was published more than a decade earlier, and in which
Strauss asserts that Hobbes, while coming very close to atheism, was not “an atheist
in the theoretical sense of the term.”44 With the reference to his own book in the
footnote of “Persecution and the Art of Writing” Strauss merely alludes to his ear-
lier misinterpretation and recent reassessment of Hobbes’s attitude toward religion,,
which he would later lay out in detail.45 This self-criticism is made in order to illus-
trate the typical consequence of the rejection of exotericism. The assumption that
there is only one doctrine for all practical purposes amounts to taking the author’s
words at face value. By doing so, the reader closes his eyes to the fact that authors
in the past did hide their thoughts for political, philosophical, and didactic reasons,,
or that not everybody has come into the world to bear witness unto the truth in
front
o t of eve yo e.46 Thinkingg tthrough
o everyone. oug tthee reasons
easo s tthat
at induced
duced a wwriter
te to ppresent
ese t
S T R AU S S A N D S C H L E I E R M A C H E R 209

his thought in a twofold manner for a twofold audience can be tantamount to a


complete turnaround of the understanding of that writer, as Strauss knew by first-
hand experience. While the young Strauss’s Hobbes was an agnostic, the mature
Strauss understood him as a philosopher who “expressed himself with great caution
in writings which he published with his name on their title pages” but who antici-
pated in his thought the infamous heretical work by Hermann Samuel Reimarus,,
which was published by Lessing a century later.47
Even more impressive than the revaluation of Hobbes is, however, the revolu-
tion in Strauss’s understanding of Maimonides. While Strauss referred to him in
Die Religionskritik Spinozas as a “believing Jew,”48 a series of careful studies written
during the 1930s brought to light the philosopher Maimonides.49 According to
Strauss’s own account, this new interpretation of Maimonides that was sparked by
an aperçu on Plato was crucial for the development of his own understanding off
exotericism.50 The fact that neither Maimonides nor Hobbes played a decisive role
in the essay “Exoteric Teaching,” whereas Lessing did, shows not so much that the
essay should be understood as an autobiographical supplement that informs the
ō ē, which we
reader of a further source of Strauss’s insight, but rather that the periagōg
associate with the name of Plato, can be experienced when studying philosophers
from any time or place.

Notes
1. A version of this essay has been delivered as the Joe R. Long Lecture at the University
of Texas at Austin on April 5, 2013. I wish to thank Thomas L. Pangle, Devin Stauffer,,
and Erik Dempsey for a challenging discussion of the talk as well as Tom van Malssen,,
Alexander Orwin, Karl Dahlquist, Jeremy Bell, Heinrich Meier, and Ralph Lernerr
for their friendly critique of an earlier version of the essay.
2. All references in brackets in the body of the text refer to my edition of “Exoteric
Teaching” or “Lecture Notes for ‘Persecution and the Art of Writing’” in this
volume.
3. For the most detailed critique of Maimonidean hermeneutics, cf. Benedict de Spinoza,,
Tractatus theologico-politicus in Opera quae supersunt omnia, ed. Carl H. Bruder (3 vols.;;
Leipzig: Tauchnitz, 1843–46), vol. III, 120–23 (cap. 7, §75–87); Spinoza, Theologico-
Political Treatise, trans. Martin D. Yaffe (Newburyport, MA: Focus Philosophical
Library, 2004), 97–100. On this question, see SCR R 173f. and Die Religionskritik
Spinozas, in GS–1 224–26.
4. The fact that Strauss left unclear whether it was “despite, or because of ” this admis-
sion that Spinoza rejected Maimonides’s allegorical reading of the Scripture, lends
further support to the supposition that the ambiguity of the personal pronoun was
not a slip of the pen.
5. GA 3.
6. See, above all, EMFL 528–605. Cf. ESG 522 and “Eine Erinnerung an Lessing,””
GS–2 607–8 (English translations in LSMM M 59–145, 155, 162).
7. The distinction is already present in Philosophie und Gesetz (1935). Cf., for example,,
GS–2 47, 82–83, 88–89, and 123 with 612. (English translation in PLA 59, 95–96,,
102–3 and 132.)
8. Cf. Arthur M. Melzer, “Esotericism and the Critique of Historicism,” American
Political Science Revieww 100, no. 2 (2006): 280: “it is difficult to find a single majorr
210 HANNES KERBER

philosopher writing before 1800 who did nott somewhere make open and approvingg
reference to the practice of esotericism, regarding either his own writings or (more
commonly) those of others.” See also his “On the Pedagogical Motive for Esoteric
Writing,” The Journal of Politics 69, no. 4 (2007): 1015–31.
9. Cf. Friedrich Schleiermacher, “Einleitung,” in Platons Werkee (second edition; Berlin::
G. A. Reimer/Realschulbuchhandlung, 1817), vol. I.1, 11–26. Reprinted in id., Überr
die Philosophie Platons, ed. Peter M. Steiner (Hamburg: Felix Meiner, 1996), 33–47..
The introduction was published for the first time in 1804 in the first edition off
Platons Werkee (A) but Schleiermacher’s revisions in the second edition (B) reflect his
mature understanding of the matter.
10. Friedrich Schleiermacher, Über die Philosophie Platons, 33 (B 11/A 11). All transla-
tions from German are my own, H.K.
11. Ibid., 33–34 (B 11/A 11).
12. “Nämlich bei den ersten Pythagoreern ging dieser Unterschied so unmittelbar auff
den Inhalt, daß Gegenstände als esoterische bezeichnet wurden, über welche sie sich
außerhalb der Grenzen ihrer innigsten Verbindung nicht mitteilen wollten; und es
ist zu vermuten, daß weit mehr ihr politisches System die Stelle des esoterischen
ausfüllte, als ihre eben so unvollkommenen als unverdächtigen metaphysischen Spe-
kulationen. Damals aber war auch die Philosophie mit politischen Absichten undd
die Schule mit einer praktischen Verbrüderung auf eine Art verbunden, die hernach
unter den Hellenen gar nicht wieder Statt gefunden hat.” Ibid., 34 (B 12/A 12).
13. “Später hingegen nannte man vornehmlich das esoterisch, was in dem populären
Vortrage, zu dem sich nach der Vermischung der Sophisten mit den sokratischen
Philosophen Einige herabließen, nicht konnte mitgeteilt werden, und der Unterschied
ging also unmittelbar auf den Vortrag, und nur mittelbar und um jenes willen erst auff
den Inhalt.” Ibid., 34–35 (B 12–13/A 12–13)
14. Ibid., 35 (B 13/A 13).
15. Ever since Jacob Bernays’s Die Dialoge des Aristoteles in ihrem Verhältniss zu seinen
übrigen Werken (Berlin: W. Hertz/Bessersche Buchhandlung, 1863), German scholars
have debated the question of exotericism intensively by the example of Aristotle’s
enigmatic references to exōterikoi logoi.They seem to have settled for the view that the
term primarily refers to a genre. Cf. Michael Erler,“Philosophische Literaturformen,””
in Der Neue Pauly. Enzyklopädie der Antikee (Stuttgart/Weimar: J. B. Metzler, 2000), vol..
IX, 874: “In Aristotle, but probably already in Plato, one can distinguish three areas
in which philosophical texts were used: literary works (‘dialogues’) for the public;;
‘exoteric’ exercises or public courses of instruction; and strictly academic lectures
and discussions within the school.” See Konrad Gaiser, “Exoterisch/esoterisch,” in
Historisches Wörterbuch der Philosophiee (Darmstadt: Wissenschaftliche Buchgesellschaft,,
1972), vol. II, 865–67.
16. Friedrich Schleiermacher, Über die Philosophie Platons, 35 (B 13/A 13).—Schleier-
macher’s critique of this version of exotericism is also directed against Wilhelm
Gottlieb Tennemann, one of the most important interpreters of Plato at the time..
According to Tennemann, Plato investigated subjects “about which most confused
and most erroneous notions (Vorstellungsarten) prevailed, but which had acquired
such a reputation by virtue of their age, by virtue of their connection to holy truths,,
[as well as] by virtue of the protection of priests and the state, that they were con-
sidered to be an inviolable property of humankind (ein unverletzbares Eigenthum derr
Menschheit).”
t Plato therefore “chose the dialogical form by which he could say truths
without being responsible for them.” Based on these considerations, Tennemann
asserts that it is likely that “the writings of his esoteric philosophy were written in a
S T R AU S S A N D S C H L E I E R M A C H E R 211

different form [than the dialogical form].” Wilhelm Gottlieb Tennemann, System der
Platonischen Philosophiee (Leipzig: J.A. Barth, 1792), vol. I, 128. On the esoteric writ-
ings, cf. also 114, 137, 141, 149, 162–64 and 264–66. See also Tennemann, Geschichtee
der Philosophiee (Leipzig: J.A. Barth, 1799), vol. II, 205–22.
17. Friedrich Schleiermacher, Über die Philosophie Platons, 36 (B 14/A 13–14).
18. In “Persecution and the Art of Writing,” Strauss makes clear that those ancient phi-
losophers who had become convinced of the essential difference between philoso-
phers and nonphilosophers had to choose one of two ways but then only discusses
the latter: “They must conceal their opinions from all but philosophers, eitherr by
limiting themselves to oral instruction of a carefully selected group of pupils, orr
by writing about the most important subject by means of ‘brief indication.’” PAW W
34–35. My italics, H.K.
19. Cf. PAWW 50.
20. The observation that “writings are naturally accessible to all who can read” (PAW W 35
with Phaedrus 275d9–e3) should be understood as a promise as well as a warning.
21. Cf. Plato, Phaedrus 277a2 and 276d4.—In the manuscript, Strauss crossed out a
sentence that seems to turn this line of argument against Schleiermacher himself::
“[Schleiermacher] forgets the fact that Plato has not written his dialogues for his
pupils only, but rather as a possession for all times, or that not all readers of Plato are
pupils of Plato.” See below p. 280n50. For κτῆμα ἐς αἰεί, see Thucydides, Historiaee
1.22.4.
22. According to a widespread, though apparently spurious tradition, Andronicus off
Rhodes, Aristotle’s eleventh successor as head of the peripatetic school, did not
include the “exoteric” writings in his edition of Aristotle’s works.—As I have indi-
cated in note 15, above, Aristotle’s usage of exōterikoi logoii has led many scholars to
believe that the term always refers to a genre of writings. This view, which entirelyy
ignores the crucial fact that there is no consistent antonym to “exoteric” throughout
the corpus Aristotelicum, can easily be refuted on the basis of Physics Δ 10 (217b29ff.)
23. Hegel’s critique of Tennemann’s notion of exotericism (cf., above, footnote 16) in
his Lectures on the History of Philosophy emphasizes the absurdity of a materiall division
of exoteric and esoteric teaching: “How simpleminded! This looks as if the philoso-
pher is in possession of his thoughts in the same way as of external things. But the
thoughts are something utterly different. Instead of the reverse, the philosophic idea
is in possession of the human being. When philosophers elaborate on philosophic
subjects, they have to follow [the course of] their ideas; they cannot keep them in
their pocket. Even when speaking externally (äußerlich) to some people, the idea
must be contained [in this speech], if the matter (Sache) has any content at all. It
does not take much to hand over an external item (Sache), but the communication
of an idea requires skill.The idea always remains something esoteric; hence, one does
not merely have the exoteric (das Exoterische) of the philosophers. These notions are
superficial.” Werke, ed. Eva Moldenhauer and Karl Markus Michel (Frankfurt am m
Main: Suhrkamp, 1986), vol. XIX, 21–22. See also ibid., 76–77: “One does not have
to make the distinction [of the esoteric and the exoteric] as if Plato had two such
philosophies: one for the world, for the people; the other, the internal, saved for the
confidants. The esoteric is the speculative that is written and printed and neverthe-
less remains hidden for those who have little interest in straining themselves. It is not
a secret but still hidden.” Cf. Strauss’s reference to Hegel, below, p. 291.
24. With one eye to the centrifugal forces within the philosophical school Strauss him-
self founded, Heinrich Meier writes in his “Preface to the American Edition,” in
Leo Strauss and the Theologico-Political
g Problem,, trans. Marcus Brainard (Cambridge:
212 HANNES KERBER

Cambridge University Press, 2006), xix: “For the school, no less than for the com-
monwealth, it holds true that different addressees have to be addressed differently,,
that they grasp the teaching differently and pass it on differently.” Cf. ibid., 15 with
xvii–xx.—For Strauss’s own critique of philosophical schools, see his “Restatement
on Xenophon’s Hiero,” in OT T 194–96.
25. PAW W 187. Cf. ONI 349: “according to the Seventh Letter, as well as according to the
Phaedrus, no writing composed by a serious man can be quite serious.” Cf. Plato,,
Seventh Letterr 344c1–d2 and Phaedrus 276d1–e3 and 277e5–278b4.
26. Friedrich Schleiermacher, Über die Philosophie Platons, 36 (B 14/A 14).
27. “Socrates was executed for not believing in the gods of Athens, in the gods of the
city. By considering and reconsidering this fact, we grasp the ultimate reason why
political life and philosophic life, even if compatible for almost all practical purposes,,
are incompatible in the last analysis: political life, if taken seriously, meant belief in
the gods of the city, and philosophy is the denial of the gods of the city” (SSTX
531–32); “In the time of Xenophon, impiety constituted a criminal offence. Thus
philosophy, which is essentially incompatible with acceptance of the gods of the city,,
was as such subject to persecution. Philosophers had therefore to conceal if not the
fact that they were philosophers, at least the fact that they were unbelievers” (SSTX
534).
28. For the first elaborate version of this argument, cf. SSTX 534. It should be noted that
Strauss immediately indicates the limitations of this argument: “It would, however,,
betray too low a view of the philosophic writers of the past if one assumed that they
concealed their thoughts merely for fear of persecution or of violent death” (SSTX
535, my italics, H.K.). Cf. also PAW W 17.
29. PAW W 28. My italics, H.K.
30. Cf. especially Friedrich Schleiermacher, Über die Philosophie Platons, 31–32 (B 9–10/A
9–10).
31. Cf., once again, ibid., 34–35 (B 12–13/A 12–13).
32. My italics, H.K. I here quote the wording of the manuscript and the alternative
reading of the typewritten versions because “degrees of understanding” shows more
clearly than “levels of understanding” the connection of this statement to the one
that occurs later in the essay: “The difference between the beginner and the phi-
losopher (for the perfectly trained student of Plato is no one else but the genuine
philosopher) is a difference not of degree, but of kind” (p. 281).
33. Friedrich Schleiermacher, Über die Philosophie Platons, 40 (B 19/A 19). Cf. also 39,,
41–42, 60, and 110. On the self-motion of the soul, see, for example, Plato, Phaedrus
245c5–246a1.
34. Über die Philosophie Platons, 41 (B 20/A 20). On rhetoric as psychagōgia ō , see Plato,,
Phaedrus 261a7ff. and 271c10ff. Cf. Plato, Gorgias 452e9–453a5 and Gorgias,,
Encomium of Helen, in Gorgias von Leontinoi, Reden, Fragmente und Testimonien, ed..
Thomas Buchheim (second edition; Hamburg: Felix Meiner, 2012), 2–16.
35. Über die Philosophie Platons, 41 (B 20/A 20).
36. Ibid., 41–42 (B 20/A 20). See below p. 281.
37. Cf., for example, Über die Philosophie Platons, 40 (B 18–19/A 18–19) with CM M 52..
It should be noted that in the Phaedrus (258d1–2) Plato’s Socrates does nott considerr
writing speeches as something shameful (aischron) in itself. Cf. 277d1–278b4.
38. Seth Benardete emphasizes this structure of understanding when he writes in
“Strauss on Plato,” in The Argument of the Action: Essays on Greek Poetry and Philosophy,,
ed. Ronna Burger and Michael Davis (Chicago: University of Chicago Press, 2000),,
409: “Something happens in a Platonic dialogue that in its revolutionary unexpect-
edness is the equivalent to the periagōgō ē, as Socrates calls it, of philosophy itself.”
S T R AU S S A N D S C H L E I E R M A C H E R 213

39. Strauss’s most radical discussion of this problem, which in Plato comes up as the dif-
ference between aretai politikaii and genuine virtue (e.g., Phaedo 82a10–b3), can be
found in “The Law of Reason in the Kuzari,” in PAW W 139: “It is hardly necessary
to add that it is precisely this view of the non-categoric character of the rules off
social conduct which permits the philosopher to hold that a man who has become
a philosopher, may adhere in his deeds and speeches to a religion to which he does
not adhere in his thoughts; it is this view, I say, which is underlying the exotericism of thee
philosophers” (My italics, H.K.).
40. In his review of Schleiermacher’s Platons Werke, the classical scholar August Boeckh
criticizes Schleiermacher’s rejection of exotericism and suggests, in passing, an
alternative. According to Boeckh, Plato’s true teaching, which he communicatedd
straightforwardly in the Academy, can also be found in some “dark corners” of his
writings: “Accordingly, the difference of the esoteric and the exoteric is based nei-
ther on the subjects nor on the external form of the presentation alone, but on the
higher or lower degree of the unveiled scientific explanation, in such a way, that
the exoteric, like the myth, has an externally manifest side, which the uninitiated
accept, but [it] also has an internal meaning, which is intelligible only for the initi-
ated. [ . . . ] Plato would have acted coyly in a curious fashion if he would have had
no exoteric [teaching] (nichts Esoterisches).” August Boeckh, Kritik der Uebersetzungg
des Platon von Schleiermacherr (1808), in Gesammelte kleine Schriften, ed. Ferdinand
Ascherson and Paul Eichholtz (Leipzig: B. G. Teubner, 1872), vol. 7, 7, cf. 4–9.
41. Cf. Seth Benardete, “Strauss on Plato,” 409: “There cannot be a method (methodos) off
thought in the thoughtful going after (metienai) of thought.”
42. In the central paragraph of the essay, Strauss lists seven items: 1) the distinction off
ancients and moderns, 2) Averroes’ attitude toward religion, 3) the Greek physicians’’
attitude toward religion, 4) the distinction between exoteric and esoteric teach-
ing, 5) Eusebius of Caesarea’s attitude toward religion, 6) Thomas Hobbes’s attitude
toward religion, and 7) the lucidity of the plan of Montesquieu’s De l’esprit des lois..
Cf. PAW W 27–29.
43. PAWW 28.
44. GS–1 145 and SCR R 100–1.
45. “On the Basis of Hobbes’s Political Philosophy,” in WIPP P 182–89. Cf. NRH H
199n43.—On Strauss’s and Hobbes’s approach, see footnote 77 in Heinrich
Meier’s “Die Erneuerung der Philosophie und die Herausforderung der Offenba-
rungsreligion,”in id.,Politische Philosophie und die Herausforderung der Offenbarungsreligion
(München: C.H. Beck, 2013), 86f.
46. At all times, Christian authors have used “milk,” instead of “solid food,” in order to
communicate the One Truth. Using the examples of Augustine and Thomas Aquinas,,
Frederick J. Crosson makes this point very clear, but also underlines the differences
between the philosophical tradition of exotericism and the Christian tradition off
“latent” teaching: “A central difference is that in the Christian tradition the manifest
teaching expressed in similitudes and metaphors and parables aims at communicat-
ing the truth, at bearing witness to the truth, in a form in which it is able to be
understood (at least partially) by all.There is only one doctrine, presented in different
depths of meaning to the two audiences” (“Esoteric versus Latent Teaching,” Review w
of Metaphysics 59, no. 1 (2005): 86).
47. WIPP P 189.
48. GS–1 238, 254 and SCR R 185, 199.
49. See, above all, the article “The Literary Character of the Guide of the Perplexed,””
which was written in 1938 but published for the first time in Essays on Maimonides::
An Octocentennial Volume,, ed. Salo Wittmayery Baron (New( York: Columbia Universityy
214 HANNES KERBER

Press, 1941), 37–91 (= PAW W 38–94). For the role Maimonides played in Strauss’s
understanding of exotericism, cf. also “Der Ort der Vorsehungslehre nach der Ansicht
Maimunis,” in GS–2 183–87 and “Quelques remarques sur la science politique de
Maïmonide et de Fârâbî,” in GS–2 137–38, 144–45, 148, 152–56; see GS–2 134n28
with Strauss’s marginal note (GS–2 160), and SR 11–12, 15–16, 18, 21–24.
50. Cf. GA 3: “Maimonides was, to begin with, wholly unintelligible to me. I got the
first glimmer of light when I concentrated on his prophetology and, therefore, the
prophetology of the Islamic philosophers who preceded him. One day when readingg
in a Latin translation Avicenna’s treatise On the Division of the Sciences, I came across
this sentence (I quote from memory): the standard work on prophecy and revelation
is Plato’s Laws. Then I began to understand Maimonides’s prophetology and eventu-
ally, as I believe, the whole Guide of the Perplexed. Maimonides never calls himselff
a philosopher; he presents himself as an opponent of the philosophers. He used a
kind of writing which is in the precise sense of the term, exoteric.” See Avicenna::
On the Divisions of the Rational Sciences, in Medieval Political Philosophy: A Sourcebook,,
ed. Joshua Parens and Joseph C. Macfarland (second edition; Ithaca, NY: Cornell
University Press, 2011), 75.
APP D C S SSEVEN WRITINGS
APPENDICES: R T GS BBY LEO
O STRAUSS
STRA SS

PRELIMINARY NOTE TO APPENDICES

O ur appendices include seven writings by Strauss himself. Five may be found in


their German originals in GS–2 and are now translated into English for the
first time. Four of these were left unpublished: a review of Karl Mannheim’s Ideologiee
und Utopiee (our appendix A);1 lecture notes for two talks to young Zionist groups
(our appendices B and C);2 and a brief reply to an article by Herbert Finkelschererr
on the medieval Jewish biblical exegete Isaac Abravanel’s conception of politics andd
society, as regards Abravanel’s critique of monarchy (appendix E).3 The fifth, on a
lost writing by the medieval recoverer of Platonic and Aristotelian philosophy Abu
Naṣr al-Fârâbî, originally appeared in Monatsschrift für Geschichte und Wissenschaft des
Judentums 80 (1936): 96–106 (our appendix D).4 The sixth is a newly edited and
annotated version of Strauss’s manuscript in English on “Exoteric Teaching”5 with
supplementary materials gathered and introduced by Hannes Kerber (our appen-
dix F). And the seventh is Strauss’s notes for a lecture in English on “Persecution
and the Art of Writing,” also newly edited and annotated by Hannes Kerber (ourr
appendix G).
Unless otherwise indicated, all footnotes in appendices A to E are the translators’..
Footnotes followed by {LS} are Strauss’s own.6 Footnotes followed by {HM} are

1“Der Konspektivismus,” GS–2 365–76, 620–21.


2“Religiöse Lage der Gegenwart,” GS–2 377–92, 621; “Die geistige Lage der Gegenwart,” GS–2
441–64, 623.
3“Zu Abravanels Kritik des Königtums,” GS–2 233–34, 615.
4“Eine vermißte Schrift Farâbîs,” GS–2 167–77, 614.
5A previous version, edited by Kenneth Hart Green, has appeared in Interpretation 14 (1986): 51–59;
reprinted in RCPR R 63–71.
6These are found mostly in “Eine vermißte Schrift Fârâbîs” and “Zu Abravanels Kritik des Königtums.”
In our translations of these (in appendices D and E), Strauss’s original footnote numbers have been freely
altered to accommodate the addition of translators’ (and editor’s) footnotes.
216 APPENDICES

Heinrich Meier’s as editor of GS–2. All footnotes designated by Arabic numerals in


appendices F and G are Hannes Kerber’s. Those designated by Roman numerals in
appendix F are Strauss’s. About the editorial apparatus to appendix F in particular,,
see Hannes Kerber’s Editorial Note.
Interpolations in square brackets are the translators’. Interpolations in parenthe-
ses inside titles or quotations from other authors are Strauss’s. All emphases, whetherr
inside or outside quotations, are Strauss’s.
Page numbers that have been inserted into the translated texts of appendices
A–E, inside curly brackets and in boldface, are to GS–2.
APP D A
APPENDIX

LEO STRAUSS: “CONSPECTIVISM” (1929)1

Translated by Anna Schmidt and Martin D.Yaffe

C onspectivism, the greatest power within present-day philosophy, is not a “direction


[of thought],”2 as materialism, positivism, idealism, etc., are “directions [of thought].””
It is, rather, a method or a style. In the previous century, the effort arose to dissolve phi-
losophy into the history of philosophy. This effort is continued by conspectivism and
transformed into a new effort. Both efforts presuppose that a naive, head-on attack on
the problems has been overcome.This overcoming has its basis in the consciousness “that
man ...after a development so long, full of sacrifice and heroic, has reached the highest
stage of consciousness.”3 At this high stage, the attempt to solve the problems is out of the
question. But while in the previous century one was still so naively reflective as to deny
the problems altogether, conspectivism is full of the reflectively reflective knowledge that
there are problems—hard, perhaps insoluble problems; it sees its very task in awakeningg
and sharpening the sense of the problematic; but it does not itself solve the problems;;

1[Leo Strauss, “Konspektivismus,” GS–2 365–75, 620–21.] Unpublished. Typescript of 13 pages with
autograph entries and corrections in ink and pencil. Page 1 dated by Strauss in handwriting: “1929.””
Leo Strauss Papers, Box 8, Folder 3, Department of Special Collections, University of Chicago Library
{HM}.
Strauss’s essay is a book review (unpublished) of Karl Mannheim, Ideologie und Utopiee (Bonn: F. Cohen,,
1929). All page numbers in parentheses in Strauss’s text are to this volume (see note 4). Emphases in
Strauss’s quotations from Mannheim are Strauss’s own.
The English translation, Ideology and Utopia, by Louis Wirth and Edward Shils (New York: Harcourt,,
Brace and World, 1936) is inexact and hence useless for helping readers understand Strauss’s criticism m
of Mannheim.
Strauss himself coins the term Konspektivismus with a polemical intention. In “Religious Situation
of the Present” (appendix B), which appropriates words and phrases and even a full sentence from m
“Konspektivismus,” Strauss associates this coinage with three synonyms: the German Zusammenschau (“syn-
opsis,” “overview”), the Greek σύνοψις, and the Latin conspectio (GS–2 2 382, with page 428, below).
2Strauss writes “Richtung” (“direction”). The German word “Denkrichtung” (“direction of thought” orr
“school of thought”) seems to be implied.
3[Handwritten note in the upper margin of the typescript:] All citations derive from Mannheim’s
Ideologie und Utopie. {HM}
218 APPENDIX A

it does not even attempt to solve them; but it keeps the option open to solve them
in the future, perhaps in the near future, the next time, so to speak; in short, it replaces
the solution of the problems and the denial of the problems by the wrestling with the
problems.
The progress from naive reflectiveness to reflective reflectiveness has the result
that philosophy creates for itself a new subject matter. Naively reflective philosophy
dissolved into the history of philosophy; it dismembered the philosophies of the
past; reflectively reflective philosophy occupies itself exclusively with the {366}
philosophy of the present. Now the return to earlier standpoints is at times still
indispensable even today; but the admirable division of labor that corresponds to
the high stage now reached allows the thinker of the present to entrust to the
historians the providing of access to the past. Let us take the example that a con-
spectivist spirit4 finds itself prompted to deal with the problem of utopia; it learns
that Thomas Münzer is of very great significance for the history of utopia; the
conspectivist spirit will then take up the pertinent literature, especially Holl’s essay
on Luther and the visionaries,5 and obtain from it an exhaustive knowledge off
the facts of the case. This procedure is unobjectionable. For even if the limited-
ness of the historian compels us to be greatly suspicious about his value judgments,,
the historian’s objectivity allows the user to gain a reliable overview of the facts
from the documents drawn on by the historian. Meanwhile, as already indicated,,
the conspectivist thinker is only occasionally dependent on the historian; usually
he occupies himself with present-day phenomena that need not be imparted by a
third party. We can now attempt a first definition of conspectivism: conspectivism
does not deal directly with the problems, as naive philosophy does; nor with the
history of philosophy, as does naively reflective philosophy; but exclusively with the
philosophy of the present.
The moment conspectivism constitutes itself, new horizons open up that were
completely unknown to earlier generations. We point here only to the bottom-
less problematic that lies in conspectivism itself. We said that it concerned itselff
only with present-day philosophy. What happens if alll present-day philosophers
are conspectivist thinkers? This possibility does not bear contemplating; but that its
realization is imminent is not to be doubted. If we see correctly, then the encoun-
ter and dialogue of conspectivist thinkers will become the theme and method off
philosophy. But, as has been said, for now we are not there yet. For now, there are
still a number of more naive spirits who deal with the problems directly. That is
why for now the conspectivist thinker still has the possibility of concerning himselff
with naive philosophers, of moving back and forth among these philosophers. This
movement is called dialectics. Dialectics is the preliminary stage of the encounterr
and dialogue, thus the preliminary stage to the stage at which the {367} conspec-
tivist thinkers will be completely among themselves. Having reached this stage, the

4The German is Geist, which can mean either “spirit,” “mind,” or “intellect.” As an adjective, geistigg can
mean “intellectual” or “referring to the intellect,” as in Strauss’s lecture title “The Intellectual [geistige
[ ]
Situation of the Present” (appendix C).
5Karl Holl, “Luther und die Schwärmer,” Gesammelte Aufsätze zur Kirchengeschichtee (3 vols.; Tübingen::
J. C. B. Mohr, 1927), 420ff.
APPENDIX A 219

spirit will have reached its being in-and-for-itself; the truth of the spirit in itself will
unveil itself in the conspectivist spirit. In the Socratic dialogue taking place amongg
Graf Keyserlingk, Peter Wust, Arthur Liebert, Margarete Susmann et al., philoso-
phy’s initial situation is being recreated at a higher stage.
The victory parade of conspectivism is not to be slowed by reactionary spirits
repeatedly shouting slogans like “To the things themselves!” “Back to the sources!””
etc. Imbued with the consciousness that everything effective at present is of equal
value, conspectivism incorporates thoroughness dialectically into its own position;;
it transforms the naive thoroughness into a thoroughness of a higher order. The
fact that even the conspectivist thinkers raise the demand for philological precision
and methodological exactness shows clearly how comfortably thoroughness can be
integrated into conspectivist thinking.
Since these remarks have the purpose of introducing the reader to conspectiv-
ism, this may be the right moment for some brief information about the most natu-
ral mode of access to the conspectivist writings. The novice should not be deterredd
by all the talk about wrestling in these writings; he must keep in mind that wres-
tling can be a beautiful, indeed a graceful, gesture. He best begins by reading the
literary and entertainment supplements included in the widely distributed demo-
cratic newspapers. Here he becomes acquainted effortlessly with the first concepts;;
he thus spares himself the time-consuming detour via the naive problems and via
the history of philosophy; he learns how positions that took a decade or longerr
to establish are overturned or even dispensed with within a few minutes by a few
clever moves, using dashes, question marks, and exclamation points; in this man-
ner he understands from the start the powerful progress in technical thinking that
the conspectivist methods have brought; he learns to apply these methods himselff
without difficulty. Trained in this way, he advances to the reading of conspectivist
periodicals, pamphlets, and books; these writings he easily recognizes by titles such
as: “Currents of Present-Day Thought”; “New Ways of (or to) Philosophy”; “Spirit
and World of Dialectics”; “The Resurrection of Metaphysics”; “Thinkers of the
Time”; etc. Once he has educated himself with the help of these classic works, he
can then turn to the preconspectivist writings and gradually apply himself to {368}
processing them dialectically by writing essays, pamphlets, and books.
If a naif has, with strenuous labor, gained a concept for himself, has thought it
through to the end “with unsparing ruthlessness even toward himself,” has put his
thoughts on paper, and, finally, published them for whatever reasons, then anyone is
free to read his book. How someone reads it depends on what kind of human beingg
he is. If he is narrowminded and has a “categorial apparatus” at hand, then he will
cast the book aside: as “metaphysical,” if he is a positivist; as “psychological,” if he is a
Neo-Kantian; as “un-existential,” if he is an adherent of the religious-metaphysical
renewal movement. But if he is open to everything new, if he is hence predestined
for conspectivism, then he notices that the book contains a word of the future. In
this case—and this case alone is of public interest—the following will then happen::
the attentive reader incorporates the new word into his vocabulary; he reads otherr
books, of which one or the other may be as significant as the first book taken as
our example; in these books he also encounters new words that he incorporates;;
his philosophy becomes more and more universal; effortlessly he overcomes the
220 APPENDIX A

onesidedness of the various naifs. Of course he reflects on the various keywords


and catchwords; he notices connections; he sits down and writes; the result off
such a wrestling is a synthesis, that is, an essay or a pamphlet or a book in which
the various words are combined dialectically. The dialectical combination of the
keywords and catchwords of present-day philosophy—this and nothing else is con-
spectivism. That is why conspectivism is no standpoint, but rather a method; more
exactly, it is a literary genre. Those books are conspectivist in which validity, value,,
Gestalt, stream of life, dialectics, the existing thinker, the conditions of production,,
Weltanschauung, structure, ontology, etc., etc., move in a bacchantic whirl.6
In order to write such books, one needs a peculiar style. Characteristic of the
conspectivist style are, for instance, sentences containing a “without”; one would d
write, say: “without underestimating the great significance of Hegelianism, one will
have enough of a critical conscience to concede that this philosophy does not fullyy
satisfy the demands of an existential worldview.” Also, adjectives ending in -haftt are
necessary in conspectivist prose; while the naif uses the {369} adjectives endingg
in -haftt only in words like fabelhaftt or grauenhaft,7 etc., it is the concern of the con-
spectivist author to develop “gestalthaft
“ ,” “bildhaft,” “raumhaft,”8 etc., for everyday use..
These indications may suffice for now. Let me just add that the conspectivist style
has exercised a fruitful effect on the language of the educated. If a merchant, judge,,
or physician because of his attitude somehow finds himself incapable of opting for a
certain worldview, or if he somehow finds a metaphysical grounding in absolute val-
ues to be required, then he owes the possibility for this to the conspectivist authors.
Only now, “ex post,” has Karl Mannheim developed the program of conspectiv-
ism long dominating the practice of philosophic authorship.9 What is better called
conspectivism, he calls “dynamic synthesis.” One would do Mannheim an injustice
if one were to describe him without qualification as a conspectivist thinker. In his
book Ideology and Utopia (here being reviewed) one finds a whole number of naive
remarks stemming from direct contact with the problems that still await conspec-
tivist reworking. Mannheim’s book is to be counted among the conspectivist lit-
erature not so much because it is itself conspectivist in form, but because, as already
said, it develops the program of conspectivism for the first time. That is why we
have to take a different position toward it than toward the properly conspectivist
literature; we shall therefore review it not on the basis of conspectivism, but on the
basis of naive thinking and in naive seriousness.
Mannheim’s book, which beckons the reader with its interesting subjects—the
three treatises of which it consists are titled “Ideology and Utopia,” “Is Politics
as a Science Possible? (The Problem of Theory and Practice),” “The Utopian
Consciousness”—is daunting at first for its wealth of technical expressions, arousingg

6Strauss’s German sentence reads: Konspektiv sind jene Bücher, in denen die Geltung, der Wert, die Gestalt,
der Lebensstrom, die Dialektik, der existierende Denker, die Produktionsverhältnisse, die Weltanschauung, diee
Struktur, die Ontologie usw. usw. in bacchantischem Taumel sich bewegen.
7Fabelhaftt means “fabulous”; grauenhaftt means “dreadful.”
8These terms mean something like “Gestalt-like,” “image-like,” and “space-like.”
9The German is Schriftstellerei, which can have the pejorative meaning of “scribbling” (in the sense off
“hack
hack writing”).
writing ).
APPENDIX A 221

admiration and fear. It just teems with “structural,”“categorial apparatus,”“outlook,””


“situation,” “synthesis,” “deliberation,” “ontology,” etc.This expenditure of technical
terms would be tolerable, perhaps even necessary, if it were in the service of con-
ceptual precision. But an expression such as “categorial apparatus” is precisely nott
possible strictly speaking. Why Mannheim embellishes the word “decision,” which
has long been recognized and accepted as a technical term {370}, with the epithet
“ontological,” why he says “lived ontic” instead of “life,” remains unintelligible so
long as one has not yet noticed how little precision there is in Mannheim’s writingg
in general. He speaks of a sociological—that is, social-scientificc—link (137) when he
should speak about a social link at most. He speaks of the solution of a synthesis
(122) when he means the solution of the task to produce a synthesis. Ellipses off
this sort do not become more tolerable by being counterbalanced with pleonasms
such as “in the truly ultimate end” and “unsparing relentlessness.”When Mannheim
says, “It could not ...be our task this timee to solve problems once and for all,” one
would like to call out to him: “but next time, please.” But in doing so one has to be
prepared that what will matter to him next time is being right; for he declares pro-
grammatically that he does “not” want “to conceal” the “contradictions, for beingg
right does not matter for now” (5). A sentence such as the following highlights
his syntax: “For magical times one would of course not be so intellectualist as to
assume that on account of epistemological considerations the magical ‘system off
order’ has disappeared ...” (57). “on a completely new manner” (124)10 is also not quite
impeccable grammar. “Chairos” (201) is presumably formed from Kairos by analogy
with Chaos–Kaos; the fact that the word is emphasized in print, and therefore could
not easily be overlooked on a proof-page, speaks decisively against the hypothesis
that we are dealing with a printer’s error.The enumeration of the formal shortcom-
ings of Mannheim’s book could be continued indefinitely; the examples cited will
suffice to justify the assertion that reflects one’s first impression: that this book has
not been written with the requisite care.Yet this is, if you will, the author’s business
only. It is the reader’s task to keep to what is worthy of attention, something that
can be found in the book nonetheless. Worthy of special attention are thee thoughts
of Mannheim’s, which may be summarized as follows.
The fact that there are various, opposing philosophical and political parties inca-
pable of convincing one another was also known previously, under quite different
conditions than today. But previously this fact was interpreted differently than it is
today. Previously one drew from it, for instance, thee consequence that prior genera-
tions had chosen a wrong approach, that it was therefore necessary to take a new path
to the {371} truth, the truth valid for all human beings and times. One put a new w
system in place of the existing systems, which might have fundamentally differed as
to method and form from the previous systems but which was for that reason no
less—a system. This possibility no longer exists today. The lasting insight of the 19th
century is the insight into the historical and social conditionality of all systems: there
is no free-floating thought, all thought is bound to its historical and social place. But

10Strauss’s point is that this phrase is grammatically incorrect. Mannheim writes “auf einer ganz neuar-
tigen Weise,” which is in the dative, whereas the preposition auff in this idiom should take the accusative::
“auf
auf eine ganz neuartige Weise.
Weise.”
222 APPENDIX A

does not knowing thereby lose its meaning? No—it merely fundamentally changes
its meaning. It gives up chasing the chimera of timeless truths; it understands that its
meaning lies in understanding the present, present-day life, the social situation from m
which it stems. The place of metaphysics is taken over by the “sociological diagnosis
of the time,” the “analysis of the situation,” the “report on the situation.”This science
grows out of the understanding of ourselves and our world that is given with our life
itself; it unfolds when we question the particular and narrowminded viewpoint that
we hold initially with the other, equally partial and narrowminded viewpoints thatt
are effective in the same social and historical space. In carrying out this confronta-
tion we are on the way to the only possible totality, to the total understanding of ourr
situation. At every moment, we must guard ourselves against the previously gainedd
insight’s positing itself as absolute, against our fleeing into a system that reassures
us by blocking the horizon from us. The inclination to such absolutizing is admit-
tedly given with human nature, with our thinking and acting. “But this is preciselyy
the function of historical research ...in our epoch, to keep rescinding these inevi-
table ...self-hypostasizings and to keep relativizing the self-deification in a constant
countermovement, thereby forcing us to be open to the addition” (40).
At this point the question becomes urgent to the reader: how indeed is the addi-
tion supposed to happen? Mannheim answers: through the other present-dayy view-
points. But who says that an adequate understanding of the present situation is indeed
achieved through the “dynamic synthesis” of viewpoints, however successful, which
exist at present? Can the possibility be ruled out from the start that all these inter-
pretations may be blind to the same fundamental facts; that one thus never even
encounters these fundamental facts if one orients oneself from the beginning only
by these viewpoints? {372} Mannheim presupposes further that the various present-
day viewpoints are equivalentt (40: “now there are too many equivalent positions, even
intellectually equally powerful ones, which mutually relativize themselves ...”). His
proof of this equivalence is that from each of these positions one sees facts that one
does not see, or at least does not see in that way, from the other positions. But are all
facts equally important? Are all aspects equally radical? What determines importance
and radicality? The totality! Now Mannheim assumes that only that viewpoint is
total that as a “synthesis” does justice to all the others. But can it be ruled out from
the start that the total viewpoint might be supremely “unjust”? In order to know w
which facts must be at the center of a total viewpoint, one must know which facts
aree central; but one does not come to know this by pitting the viewpoints dominant
at present against one another. Here Mannheim’s premature judgment of Ranke’s
“obliviousness”11 (63) comes back to haunt him; Ranke said: “All the heresies of the
world will not teach you what Christianity is—one can come to know it only by
reading the Gospel.” This sentence is neither naive nor ominous, but simply true. Iff
one understands that thought is conditioned by the situation, it does not follow that
one cannot come to see the situation originally, free of the dominant viewpoints.This
freedom does not fall into anyone’s lap; it must be won by understanding the tradition
as such in which we are caught up. Admittedly, this tradition cannot be seen clearly

11Ger.: Ahnungslosigkeit. In Strauss’s next sentence, “oblivious” is ahnungslos and “ominous” is


ahnungsvoll.
ahnungsvoll.
APPENDIX A 223

if—as Mannheim does throughout—one orients oneself only by the more recent
centuries. When Mannheim takes premodern developments into consideration, then
only “traditionalism” in contrast to modern “rationalism,” the “medieval-Christian
objective unity of the world” in contrast to the “Enlightenment’s absolutized unity off
the subject,” or at the most the “magical system” and “the prophets”: in Mannheim’s
book, which poses the question of the meaning of science, specifically of the pos-
sibility of politics as a science, the foundations of our scientific tradition which lie in
Greek antiquity are forgotten! Of all people it is Mannheim, who desires and hopes
that the history of word meanings “will be researched at the level of methodological
exactness possible at present” (38), who is guilty of this omission. {373}
But let us disregard the lack of “methodical exactness,” of “philological pre-
cision” (1), this failure to answer demands that Mannheim himself makes; let us
further suppose that in fact every analysis of a situation that is possible in that
situation is “somehow” a “synthesis” of the extreme positions effective in this situ-
ation: is it permissible therefore to make this fatee of all research into the principlee
of research? Mannheim speaks of the danger that lies in the “false contemplation
of the researching stance” vis-à-vis political practice. Well—the same danger exists
vis-à-vis scientific practice. Mannheim, who at many points is brushing the outerr
limits of liberalism—by incorporating illiberal elements into the liberalism that in
fact has a hold on him by means of a “dynamic synthesis”—does not in truth over-
step these limits. (Particularly interesting in this respect is Mannheim’s interpreta-
tion of fascism, which we cannot enter into here.) We say: liberalism has a hold on
Mannheim, and we are justified inasmuch as it is the essence of liberalism to elevate
insights gained in a contemplative attitude to principles of practice.
The “analysis of situation” sought by Mannheim does not want to be “value
free.” It is aware that the will to know what is, this will that forbids every escape
and every lie, contains a “value judgment”; that given with our life already, from
which this situation analysis arises, are valuing, taking a stance, praising and blaming;;
that it is possible only on the basis of a “decision” guiding the eye, illuminating the
horizon. This “decision” stands at the beginning; not, however, as an axiom from m
which anything one likes can be inferred, but as an implicit drive to question, which
becomes explicit, understood, tested, and doubted in carrying out the confronta-
tion. Thus, the analysis of situation that is always grounded in a “decision,” succeeds
in the unveiling of ideologies by showing for what they are “outdated and outlived d
norms and forms of thought, but also worldviews” that do not clarify the present
situation but conceal it (51). A “decision” underlies the “sociological diagnosis off
the time” that seeks a “dynamic synthesis,” and in particular underlies the “politics
as science in the form of a political sociology” (143) also demanded by Mannheim,,
namely, the decision for a politics of the center, whose support is to be the “socially
free-floating intelligence.” The {374} stratum of intellectuals that is—according to
Mannheim’s thesis—the support of the “spirit” is the “predestined advocate of the
intellectual interests of the whole.” This stratum is not a class; it owes its unity nott
to its economic situation but to “education.” The intellectuals, who as individu-
als are conditioned by class and always remain within certain limits, have in theirr
education “a homogeneous medium” in which the heterogeneous class tendencies
can confront each other; the intellectual struggle made possible by education is a
224 APPENDIX A

“downsized image” of the class struggles.The result of this confrontation is expectedd


to be a “total orientation” about the social situation, as it cannot be reached from the
extreme standpoints of classes and parties; but not a “politics peculiar to intellectu-
als,” which would hardly be possible in the time of mass movements. The stratum m
of intellectuals has the possibility and the “mission” of anticipating intellectually andd
thus beginning a “dynamic mediation” between the classes opposed to each other,,
between the “ruthless representatives of yesterday’s principles” and the “one-sidedd
emphasis on the day to come” (121–34). We leave undecided whether this type off
influence on politics is possible at all in the present situation; we only point out an
allusion that Mannheim makes in the context of his treatise on politics as a science,12
an allusion whose momentousness is deplorably at odds with how little it is devel-
oped. We recall the significance of the distinction between an ethics of intention
and an ethics of responsibility for Max Weber.13 Unless we are mistaken, the insightt
into this distinction was the deepest and strongest motive for his conviction that it
is impossible to decide scientifically between opposing moral principles; for here
two irreducible, unjustifiable, unconditional possibilities of human behavior seemedd
to face each other, compelling each human being to face an either/or. There is no
doubt: Max Weber faced this choice. Are we still facing it today? If we compre-
hend correctly the remark that Mannheim makes at the conclusion of his secondd
treatise,14 then he is of the opinion that the ethics of intention is no eternal possibil-
ity, but onee stage in the history of humanity, destined to be overcome by the ethics off
responsibility that Max Weber professed. But if that is the case, is not the meaning off
“decision” fundamentally changed? Do not then the questions that Mannheim asks
himself have to be asked entirely differently? This question leads back to the {375}
more fundamental question of how the world in which science emerged looked
before the incursion of the biblical consciousness. Only by orienting ourselves byy
this world can we gain the horizon in which alone we can radically question andd
answer henceforth. In orienting ourselves by this world we would also see that,,
under Mannheim’s implicit presuppositions, one would have to inquire not into the
possibility and necessity of utopia, but into the possibility and necessity of planning..
But Mannheim utterly lacks precisely this orientation.

12I.e., in “Is Politics as a Science Possible? (The Problem of Theory and Practice).”
13See Max Weber, “Politics as a Vocation,” in From Max Weber: Essays in Sociology, trans. H. H. Gerth
and C. Wright Mills (New York: Oxford University Press, 1946), 120: “We must be clear about the fact
that all ethically oriented conduct may be guided by one of two fundamentally differing and irreconcil-
ably opposed maxims: conduct can be oriented to an ‘ethics of intention’ or to an ‘ethics of responsibil-
ity’ [‘gesinnungsethisch
‘ ’ oderr ‘verantwortungsethisch’ orientiert].
t This is not to say that an ethics of intention
[Gesinnungsethik] is identical with irresponsibility, or that an ethics of responsibility [Verantwortungsethik]
is identical with unprincipled opportunism. Naturally, nobody says that. However, there is a gaping
contrast between conduct that follows the maxim of an ethics of intention—that is, in religious terms,,
‘The Christian does rightly and leaves the results with the Lord’—and conduct that follows the maxim
of an ethics of responsibility, in which case one has to give an account of the foreseeable results of one’s
action.” Translation modified. See Weber, Politik als Beruff in Gesammelte Politische Schriften, ed. Johannes
Winckelmann (Tübingen: J.C.B. Mohr, 1971), 551–52. Cf. also NRH H 69ff.
14I.e., of “Is Politics as A Science Possible? (The Problem of Theory and Practice).”
APP D B
APPENDIX

LEO STRAUSS: “RELIGIOUS SITUATION OF THE


PRESENT” (1930)1

Translated by Anna Schmidt and Martin D.Yaffe

I would be facing an insoluble task if I had to report to you all that is being written
about religion today; for infinitely much is being written: each day a new book,,
a new pamphlet, a new essay comes out concerning our question. The task would
be simplified—and the simplification in our case amounts not so much to the
temptation to give in to laziness as to an imperative of reason—if I limited myselff
to reporting what was being thoughtt today, being thought thoroughly. For while
there are many who write, there are few who think, who think thoroughly. I do
not mean to claim that a necessary opposition exists between thinking and writing,,
although that claim might not be all that indefensible. I am happy to admit that
there are a number of men who write after they have thought, have thought thor-
oughly. We need to concern ourselves with these men only.
Meanwhile a further restriction recommends itself. Most of even the thorough
authors are apostles of a master. And here we keep to the proverb: “One does not
go to the apprentice, but to the blacksmith.” It is indisputable that the kind of lit-
erature that advances the understanding of a great and deep mind, commentary, has
a value that should not be disparaged. But the commentator is not the author. And
in concerning ourselves with the religious situation of the present we do not want
to become confused by the multitude of commentators, but to stick to the very few
authors, to the auctores of the situation. {378}
We are interested in the religious situation of the present as Jews. For that reason
we take a stronger interest in those authors who, being Jews, determine the present

1[“Religiöse Lage der Gegenwart,” in GS–2 377–91, 621.] Unpublished. Manuscript with 13 handwrit-
ten pages and inscribed cover, in ink with penciled supplements and corrections. On the cover Strauss
has noted under the title: “Paper to be read on 21 December 1930 in the Kadimah Federation Camp
in Brieslang near Berlin.” Leo Strauss Papers, Box 8, Folder 4, Department of Special Collections,,
University of Chicago Library. {HM}
Kadimah was a national student Zionist organization founded in Vienna in 1882 (and disbanded by
the
h Nazis
N i ini 1938).
1938) IIts H
Hebrew
b name means “F“Forward.”
d”
226 APPENDIX B

situation. Of the Jewish writers of the present who are of interest in our context,,
the most important one by far is without any doubt Franz Rosenzweig.2 I will
therefore confine myself to a presentation of Franz Rosenzweig’s doctrine. In the
short time that I have, I would only cause confusion by giving a survey of the vari-
ous theories. The thorough treatment of a work that is indicative of and peculiar to
the present situation is much more suitable for our purpose.
However, I cannot even turn to my task limited in this limited way, I cannot
begin my thus limited task in the right manner, so long as I have not clarified the
topic itself. In this clarification it will turn out, however, that the topic is not a serious
topic. That is why I divide my presentation into 2 parts:

1. Clarification of the topic;


2. Rosenzweig’s doctrine and its difficulty.

The title of my presentation consists of four words—rather, since we can safely


disregard the article “the,” of three: “Religious,” “Situation,” “Present.” We want to
look at these three words in sequence in their context as determined by the topic.
1. Religious Situation of the Present. Obviously this does not mean in one par-
ticular respect: in the way that we can consider the situation of the present as
conditioned by the Versailles Treaty, by the crisis of the world economy, by the
Balfour Declaration, etc.; nor that the religious situation of the present is of the
same order as the political, social, economic, technical, scientific, artistic situation..
Instead what is meant is: the situation of the present in the most importantt respect. It
is, for instance, unthinkable that the topic of your study group, the validity of norms,,
values, and laws, would nott belong to our topic; rather, it essentially belongs to it..
For that reason, for example, the philosophic situation should not be distinguishedd
and excluded from the religious situation; but certainly the only reason why you
did nott say “philosophicc situation of the present” was because otherwise there would
have been the danger of letting loose a flood of incomprehensible technical terms..
Avoiding this will be my most pressing concern. But the matter prohibits disregard-
ing philosophy, {379} prohibits distinguishing between religion and philosophy in
looking at the present religious situation. I therefore replace “religious situation off
the present” with “intellectual3 situation of the present.”
2. Intellectual Situation of the Present. The expression is intelligible but
inappropriate:4 the intellect is not a thing that is situated, or that could have a
situation. The intellect is actual5 in seeing and searching, in believing, wishing and
hoping, in demanding and giving an account, in responsibility, in questioning and
answering. Now, questioning has priority over answering. God does not question,,

2Franz Rosenzweig (1886–1929), German-Jewish theologian, originator of the “new thinking” and
leading advocate of the “movement of return” to Judaism; see Strauss, Preface to SCR, in JPCM M 146–48
with 151–53, 453, 460.
3Ger.: geistige. (The emphasis is Strauss’s.)
4Ger.: uneigentlich. Or: inauthentic. Likewise five sentences later in this paragraph. In that same sen-
tence, “proper” is eigentlich, whose range of meanings includes “real,” “actual,” “true,” “appropriate,” andd
“authentic.”
5Ger.: wirklich
wirklich.. Or: real.
APPENDIX B 227

although he does answer. Questioning is more characteristic of the human intel-


lect than is answering. There is no answer without questioning,6 although there is
questioning without answer. It may be that questioning is not the sufficient defi-
nition of the intellect—in any case, questioning is proper to the intellect and not,,
like the situation, inappropriate. We therefore now say: the intellectual questioningg
of the present. This expression has a pleonastic effect: although we speak of “the
social question,” we do not speak of social questioning, etc. We therefore say: thee
questioning of the present. The questioning asks something; it asks a question;7 we
grasp the questioning of others—in this case, of the present—from their question..
We therefore say: thee question of the present.
3. The Question of the Present. Let us imagine, in a fanciful manner, a Kadimah
camp assembly in the twelfth century, in the century of R[abbi] M[oses] b[en]
M[aimon],8 and that you had asked a student of the RMbM to speak to you so
that he might help you, by means of what he has learned from his teacher, to free
yourselves from your confusion and perplexity. What would he have spoken to
you about? Creation, providence, the unity of reason and revelation. Hence, about
substantivee questions. In another age, one would probably have spoken about otherr
questions, but always about substantive questions. Nobody would have minded
whether or not they were questions of the present. At the time, they weree questions
of the presentt that were being dealt with, but they were not being dealt with as ques-
tions of the present. When we question, seriously question, we then by that very factt
ask questions of the present. And if we pose thee question that we are certain is thee
question, then we are asking thee question of the present.We will therefore cross out
“of the present” and say: Thee question.There can, however, be no doubt about what
thee question is that is and must be the most important one for us: it is the question,,
what is the rightt life? how shouldd I {380} live? what matters?9 what is needful?10
Thus, our modern topic of the “religious situation of the present” boils down to the
old, eternal question, thee primordial question.
There is no doubt that this is the question for the sake of which you have posedd
the question concerning the religious situation of the present. For in order to learn
the latest from the realm of intellect and wit, you did not need to travel from noisyy
Berlin to quiet Brieselang. But how does it come about, then, that the simple ques-
tion poses itself not only to you but also to the present as such, as the question off
the religious or intellectual situation of the present? This automatic alteration of the
question takes place on account of an implicit or explicit, but in any case tyrannical,,
conviction: the conviction that the answer to the question could be gained solely orr
essentially from the knowledge and understanding of the present as it exists in the
present. This conviction and its causes we have to consider first.

6Or: questions.
7Ger.: Das Fragen fragt etwas, es fragt eine Frage. Strauss’s sentence has a Heideggerian formulation.
8“RMbM” is the traditional Hebrew acronym for Rabbi Moshe ben Maimon (i.e., Moses
Maimonides).We have retained it here, along with Meier’s bracketed interpolations.The acronym recurs
in Strauss’s next clause and later on.
9Ger.: worauf kommt es an?
10Ger.: was tut not?
228 APPENDIX B

If we pose the question concerning the right life unselfconsciously and naively,,
convinced that we can answer it if we make an honest effort and do not let our-
selves be put off by any detour, then the present,11 attired in the most splendid
robes, confronts us with the raised eyebrows of a haughtily knowing, exalted per-
sonage and calls out to us:

Stop! You unsuspecting ones! Do you not know that the inexhaustible earth brings
forth new generations year after year, which, barely having reached maturity, are all
destined to charge with all the fire of youth directly at the truth, at thee truth? This
has now been happening for thousands of years. For thousands of years the attempt
has been made, and time and again it has failed. At one time, later generations did not
let themselves be confused by the failure of earlier ones; full of delusion they said to
themselves, if they failed—perhaps they approached the issue the wrong way; let’s just
begin from the beginning; let’s begin completely from the beginning. And they began
from the beginning, and they also failed. The unhappy ones did not know—what I,
the Present, the powerful goddess, know—that they hadd to fail. They had to fail since
they were seeking thee truth. For there is not the one eternall truth, but each age has its
truth, and you, you 20-year-olds, you can reasonably seek only your truth, the truth of
your age, my—the present’s—truth. Being in full possession of this knowledge, which
is {381} my greatest pride, I am allowed to smile at the past—at its naivety. I do not
hide that my smile conceals a little envy: at the élan of youth, which, in the superior-
ity that my knowledge gives me, I cannot permit myself: the long, magnificent robes
that strike your eye would hinder me very much in an assault on the truth, which can
only be dared in combat gear; I cannot go on the assault: I am stuck at the base; I do
not hide that I am sometimes ashamed before the frontline soldiers; but then my good
sense tells me: “You have no reason to be ashamed; the base is yourr virtue, yourr duty;
you would be dishonest, you would be betraying yourself, if you, such an exalted,
superior, refined personage, were to take on the dirt and hardship of the front lines. I
have it much harder, since I, too, would rather go on the assault and cannot and must
not.”—So, while my smile is not a naive smile at their naïveté, and is indeed a smile
that is not without grief and shame, it is, at the same time, also not a poisonous smile:
My smile is benevolent: I exculpate, I justify: I exculpate the earlier generations since
they did against their will what I prescribe to my children. To be sure, they did seek
thee truth, but they found the truth without time;12 they failed—measured by their
standard; measured by my standard, they reached the goal. So now, enthroned high
above the entire past, I call out to you: It is befitting for thinking beings to know what
they are doing and what they can reasonably want: therefore, know and be imbued
with it once and for all, that you can find only your truth, the truth of the present, and
therefore can reasonably seek only it.

In this way the present speaks a lot to us, not through the mouth of stubborn
goats in Scotland13 but through the mouth of the most agile, most progressive, most
expert, most lively children of our time. So let us hear more closely what they
are saying to us. We cannot seek thee answer to our question, but only the answerr
for us, for the present. But where and how to find this answer? Surely not in the

11Or: the Present. Likewise in the following.


12Ger.: ohne Zeit.
13This expression seems proverbial, with no literary antecedent.
APPENDIX B 229

study. No—only through coming to know the powers of the present! Where do we
encounter these powers? In the struggle of parties, groups, trends, currents, etc. But
are we supposed to hear what the truth of the present is from the cacophonous
noise of the public? No—these conflicting trends do not harmonize on their own;;
their harmony {382} must first be produced by us. In what manner, though? The
thoughtful person cannot devote himself completely14 to onee trend; he sees all too
clearly that in each of these trends are truth and untruth; hence he must try to do
justice to all, to the truth in all. In short—what he needs is a “synthesis.”
Now then: 1) since everything human is historical, there is not thee question, but
always only the question of the present. In order to answer this question, or even
just to pose it, we must know the situation of the present. 2) The situation of the
present—that is, the factual, effective answers of the present. 3) The answer is given
by a synopsis, σύνοψις, conspectio—conspectivism.15
1. Critique of Conspectivism.
α.The Incredible Difficulty of a Synthesis. Conspectivism is possible only because
of the complete absence of a concrete notion of the emergence of a “position.””
Every position that can at all be taken seriously is the work of an immense effort
of an individual. When Kant—who already had achievements that by themselves
would have made him immortal (Kant–Laplace theory), who was no inexperienced
young man who yet had to acquire the necessary knowledge of facts—had accom-
plished the breakthrough to his position, he needed 11 years for the Critique of Puree
Reason—not to write it, but just to think it. Let us assume that something similarr
goes for Marx or for Nietzsche. These men came to completely different results
with their immense efforts.What an exponential effort it would take to find a posi-
tion from which both positions were unified! How much deeper would someone
have to descend in order to find the common point from which....One need only
imagine these difficulties in order to comprehend that the people who today talk off
synthesis simply do not mean16 anything by it. But they must mean something by it!
As it seems to me, conspectivism comes about in the following manner.
β. The Genesis of Conspectivism from the reader turned writer. The origina-
tors of positions have laid down the results of their {383} immense effort in pub-
lished books. Everyone can buy these books or have them given to him as a gift
or borrow them and then read them. Now, there are two types of readers. Some
are narrowminded; they have a fixed and ready opinion; they read only in order to
confirm their opinion: should the book nott be of their opinion, they have enough
arguments ready-at-hand to dismiss the book. For, what aren’t there arguments for;;
certain fundamental insights of Kant’s, which today any jackass has or believes he
has, were “refuted” with sovereign superiority by jackasses among Kant’s contem-
poraries.This type of reader is harmless and innocuous. More harmful is the secondd
type. To this type belong people who are stimulated by the books, who are open to
everything new; these people are easily excited; they adopt one book’s conclusions

14Ger.: mit Haut und Haar.


15Ger.: Konspektivismus. Compare the following with Strauss’s unpublished review by this title, trans-
lated in appendix A, above.
16Lit.: think. Likewise in the next sentence.
230 APPENDIX B

and then again another’s. Since they are precisely nott narrow, they cannot resist the
conflicting theories.The theories can be formulated in certain keywords; these key-
words can easily be adopted. One reads and reflects while reading; it occurs to one
how things are related; one sits down and writes.The result of this very entertainingg
activity is a synthesis, that is, a book or a pamphlet or an essay.That is the essence off
the conspectivist spirit:17 the conspectivist spirit is the lazy reader turned writer.
γ. Sham Understanding. In truth, one understands nothing conspectively, even iff
one is very bright. I want to give an example. In our time, somewhere there lives a
philosopher in the full sense of the word.18 Completely unknown just five years ago,,
today his name and work are talked of everywhere. This philosopher in his main
work has, among other things, written a few pages about idle talk, what it means
and what it inflicts.19 That was meant as a, so to speak, purely factual statement, not
as the author’s appeal to spare him from idle talk. What happened? A woman20—
the noble word “lady” obviously forbids itself—reads this philosopher, and before
she can even have the slightest idea of what the man really means, she goes around
in London and yaks and yaks. She found the paragraph on idle talk certainly “very
fine”; she has understood him in this sense; but she did not understand him in such
a way that she would finally, finally shut her unbearable trap.
Therefore: if one takes seriously the great men who dominate the present, then
one will not wish to attempt a synthesis, which amounts to muddling and dilut-
ing what was important to them. It {384} is preferable to despair in light of theirr
contradiction than to give in to a stale and cowardly mishmash.
2. The Situation of the Present can be grasped21 in the totality of the positions effectivee
in the present. Why alll these positions? Because they are equivalent. Why are they
equivalent? Because each one sees facts that the others do not see, or see indistinctly..
But obviously it appears not to be important to see everythingg equally distinctly— —
but to see what is important distinctly and what is not important indistinctly. We
must therefore already know beforehand what is important. One answers: what is
important is the total situation of the present: what matters is the totality. The reason
why individual positions cannot convince one another, cannot do justice to one
another, is that they are not total.22 But is “justice” what matters simply? Is not
“injustice” vis-à-vis what is not true truly just? Cannot the total view be supremelyy
unjust? In truth, alll views are total.
But assuming that a synthesis were possible and that all the positions effective in
the present were equivalent: would it therefore be necessarily so that the synthesis
actually conceived thee situation of the present? For is it necessary that alll view-
points, that the truee viewpoint be contained in the present viewpoints in such a
way that it results from their synthesis? Is it not possible that alll present standpoints

17German: Geistes. Except where otherwise noted, Geistt and its cognates are “intellect” and its
cognates.
18The reference is to Martin Heidegger. See the following note.
19Heidegger, Sein und Zeitt §35 (10th ed.; 167–70)
20Ger.: Weib. In the next clause, “lady” is Frau.
21Ger.: ist fassbar. Later in this subsection, “view” is Auffassung, and “viewpoint” is Gesichtspunkt.
22Ger.: untotal
untotal..
APPENDIX B 231

rest on a mistaking of the fundamental facts? Are not perhaps all these standpoints
“ideologies”? This is in no way settled. If we want to come to know the present
just as it is, free from the dominant views, which we must first examine, then we
must first of all be free of the present. This freedom does not fall into our laps; we
must win it for ourselves.
3. Necessity of the Naivety of the Questioning. But is it true at all that we have to
come to know the situation of the present in the first place? From mankind’s always
having a present, it nevertheless does not follow that one need be concerned with
it: our fatee is nott our task. This is the principal mistake to which today’s man keeps
succumbing: the attempt to determine the task from the fate.This attempt is absurd
if there is no God: then fate is chance, and if God is, then fate is providence, and we
are not allowed to want to play God. This error manifests itself also in the will to
synthesis: even if each standpoint may bee a synthesis in fact—it t is nonetheless neverr
willed as a synthesis; what has been willed is always the truth. {385} We have to look
ahead; we neverr come to know what we have to do by being reflective.
We want to do justice to the matter, then.23 We turn to the matter,24 that is, the
question concerning the right life, with the will to answering it. But in order not
to suffer shipwreck as thousands have suffered shipwreck beforee us, we do want to
hear the warningg of the present, the call: watch out. We will not listen to the present
if it turns this failure into a theory, if it asserts the inevitability of failure. In order to
be able to get beyond the present, we must take the warning of the present seri-
ously, we must be in a position to interpret more closely the experience on which
the present insists. We therefore do not ask about the present, but about the warningg
of the present. But in this warning we let ourselves encounter the present. We thus
admit: failure will certainly not have been accidental, due to individual inadequacy,,
to the stupidity of earlier generations; it will have its serious reasons. In order to
understand these serious reasons, we must take seriously the question about them,,
we must not truncate this question by the dogmatic assertion that there are no
eternal truths.
By the way, what about the historical experiencee of this failure?
The question was posed for the first time by Socrates. Whether and in what
sense he himself gave an answer is obscure. In any case, his student Plato answered it::
in the Republic.25 In order to illustrate the difficulty of true understanding, Plato in
this work compares the situation of human beings to the situation of cave dwellers::
a cave with a long entry stretching upwards; the human beings from childhood on
are bound inside the cave by chains around their thighs and necks; they thus always
remain in the same place, and they are prevented by their neck chains from turning
their heads around; from above a firelight shines from a distance; above, between the
fire and the prisoners runs a walkway, alongside which runs a wall; along this wall
human beings carry all sorts of artifacts, statues, etc.; it appears that the prisoners
there can see only the shadows of those artifacts, which are cast by the fire’s light

23Ger.: Wir wollen also sachlich sein.


24Ger.: Sache.
25See, for the following, Republicc 514a–517a.
514a 517a.
232 APPENDIX B

onto the cave’s wall facing them; to them, therefore, the shadows would be the true
things. Now, if one of them were unchained and put into a position to gaze freely
up toward the light, which however could happen only under great pain {386}, he
would, being blinded by the glow, be incapable of recognizing the things whose
shadows he had seen before; he would be at a complete loss if he were told that
now he was seeing the things whose shadows he had been seeing until now; above
all, the very sight of the light would pain him so much that he would turn away
and want again to retreat into the dark of the cave; and it would require a longg
habituation and effort, indeed the use of force, for him to be capable of seeing the
true things, of living in the light of the truth. Brought back into the cave, he wouldd
retain the memory of his life in the light, but would be completely incomprehen-
sible and laughable to his companions precisely because of this. Thus Plato presents
the difficulties of philosophizing, the naturall difficulties. If they are so extraordinary,,
is it any wonder that there are so many conflicting opinions? Bearing in mind the
Platonic parable, we will not be misled by the anarchy of opinions, but will have to
exert ourselves as much as possible to get out of the cave.
We said: Plato presents the naturall difficulties of philosophizing. That is, those
difficulties natural to man as man, as a sensitive-intellectual being, the difficulties
that according to the Platonic view are given by his sensitivity.26 We say “natural””
because there are difficulties that are not “natural” but become effective only underr
certain presuppositions. RMbM in Moreh Nebuchim27 (I 31) expands the enumera-
tion of reasons given by a Greek philosopher for the differences of opinion in
philosophy, and therefore for the difficulty of philosophy simply, by a fourth reason;;
about this he says, literally:

In our timee there exists a fourth reason, which he [sc., Alexander of Aphrodisias] did
not mention, since it did not exist among them; namely, habituation and schooling; for
human beings by nature love what they are habituated to and incline to it;...thus
it happens to man regarding the opinions with which he has grown up: he loves
them and holds on to them and keeps away from deviant opinions. For this reason
as well, then, man is prevented from coming to know the truth. Thus it happens to
the multitude regarding God’s corporeality ...on account of habituation to writings in
which they firmly believe and to which they are habituated, whose wording seems to
indicate God’s corporeality.

Let us sum up: by the fact that a tradition resting on revelation has entered the
world of philosophy, the difficulty of philosophizing is fundamentally augmented
{387}, the freedom of philosophizing fundamentally limited.
In RMbM’s remark, the struggle of the entire last 3 centuries, the struggle off
the Enlightenment, is in a sense sketched, outlined: in order to render possible phi-
losophy in its natural difficulty, the artificial difficulty of philosophizing has to be
eliminated;28 there has to be a struggle against prejudices. In this, modern philosophy
is fundamentally different from Greek philosophy: the latter struggles only against

26Ger.: Sinnlichkeit.
27That is, Guide of the Perplexed. Strauss uses the traditional Hebrew title.
28More or less lit.: removed from the world.
APPENDIX B 233

appearance and opinion; modern philosophy’s struggle begins prior to that against
prejudices. The Enlightenment thus wants to recover Greek philosophy. What does
it achieve? It achieves: the freedom of answeringg but not the freedom of question-
ing, only the freedom of saying No instead of the traditional Yes. (Mortality vs..
immortality, chance vs. providence, atheism vs. theism, passion vs. reason.29) But this
liberation from the Yes of the tradition takes place by means of a commensurately
deeper entanglement in the tradition. For instance, the Enlightenment conducts
its struggle against the tradition in the name of tolerance, ultimately in the name
of love of neighbor; thus religion is now being based entirely on love of neighbor,,
in such a way, however, that with the doubt of love of neighbor (as understoodd
by the Enlightenment)30 religion as such becomes altogether doubtful. Or to take
an example from a later stage of the Enlightenment: when the Enlightenment
becomes openly atheistic and believes it sees through “God” as being a construct off
the human heart, the only way it does so is by internalizing the purposes of God
into mankind: self-redemption of mankind, self-assurance of immortality (museum,,
etc.), assuming the role of providence. And if opponents of Enlightenment arise at
every stage of it, then these opponents, for their part, take over the successes of the
Enlightenment and reconstruct their position in accord with these. (E.g., revelation
is understood as being a human product, as morals and as form, not as law; creation
is understood not as being the creation of the world, but as what is binding in
advance on human beings.)31 In general: since the Enlightenment, each generation
has generally reacted to the preceding generation without questioning the founda-
tion. For example, the concept of the “irrational”—rationalism understood in the
narrowest sense.32 {388}

Example:The Problem of Creation


1. God has created the world in complete freedom, out of love; he rules overr
it in complete freedom—justly and wisely, but in such a way that we are
not authorized and not able to know the ways of his justice and wisdom. In
his freedom he can perform miracles. Miracles are not in themselves “more
divine” than the usual course of the world; but there is no reason for saying
that God could not and would not perform miracles.
2. The struggle against miracles in the name of metaphysics: miracles are unwor-
thy of God as the perfect being.
3. Modern physics understands nature completely on its own terms: No scien-
tific proof of God possible. God related to the humanity of human beings.
4. Nature a construct of human intellect.33
5. By analogy with this construct, the whole of “culture,” and with it religion,,
is understood as a construct of the human mind.34
29Ger.: Verstand.
30Ger.: aufgeklärt verstanden.
31Ger.: als Vorher-verbindlich über den Menschen verfügt sein.
32[Note in margin:] Theory vs. Intuition
Mannheim’s Ideology and Utopia {HM}
33Ger.: Verstand.
34Ger.: Geistes.
234 APPENDIX B

6. Novel understanding of the original religious attitude (in contrast to mysti-


cism): the demanding God who summons before him.35 The abandonment
of creation remains.

The entanglement in the tradition is further intensified by a theory that legiti-


mizes this entanglement.While the Enlightenment itself was wholly convinced that
history was accidental, that the victorious party was not in the right just because
it happened to be victorious, in the nineteenth century the belief that world his-
tory is the world’s court of judgment becomes dominant. (A belief that can be
justified in the natural sciences, where in fact an unequivocal progress, a building
on the accomplishments of earlier generations, is possible but that is, at bottom,,
impossible.36
Thus: the question πως βιωτέον is hard to answer today not merely on account
of the naturall difficulties, not merely on account of the dominance of a tradition off
which we are aware as being a tradition, but on account of our total entanglement
in the tradition, which goes so far that we cannot express ourselves purely and freely,,
that every attempt to express and determine what we have seen and experienced is
impossible at first. But what then can we do now?)
The struggle against the tradition that leads to the complete entanglement in the
tradition, also dissolves the tradition. In every subsequent {389} generation there
is in factt ever less tradition. The struggle against the tradition was ever again forced
to appeal to the opponents of the tradition, to those against whom the tradition had
arisen and prevailed.Those elements that were supplanted by the tradition reappearr
(the pagans, Epicurus, vindication of heretics of all types, the sophists, idolatry)..
These elements became understood.The end of this struggle is the complete rejection off
the tradition: not just of its answers, also not just of its questions, but of its possibili-
ties: The pillars on which our tradition rested, prophets and Socrates-Plato, are torn
down since Nietzsche. Nietzsche’s siding with kings against prophets, with sophists
against Socrates—Jesus not only no God, also no charlatan, also no genius, but a
moron. Θεωρεῖν37 andd “good–evil” rejected—Nietzsche the lastt Enlightener.
The tradition has been shaken at its roots by Nietzsche. It has altogether forfeited
its self-evidence. We stand in the world completely without authority, completely
without orientation. Only now has the question πως βιωτέον regained its full
sharpness. We can again pose it. We have the possibility of posing it in full seri-
ousness. We can no longer read Plato’s dialogues superficially, in order to notice
admiringly that old Plato already knew this and that; we can no longer polemicize
against him superficially. And the same with the Bible: we no longer think without
evidence that the prophets were in the right; we ask ourselves seriously whetherr
it was not the kings who were in the right. We really must begin entirely from the
beginning.
We can begin entirely from the beginning: we lack any polemical passions against
the tradition (we have, after all, nothing from where we could be polemical); and at

35Ger.: der Fordernde, vor sich fordernde Gott.


36This sentence and the entire next paragraph are enclosed in parentheses in Strauss’s ms.
37Gk.: Contemplating.
APPENDIX B 235

the same time, the tradition has become completely estranged from us, completely
questionable.
But we cannot answer immediately as we are; for we know that we are deeply
entangled in a tradition; we are yet much further down than Plato’s cave dwellers..
We must raise ourselves to the origin of the tradition, to the level of natural ignorance..
If we wanted to concern ourselves with the present situation, we would be doing
nothing other than the cave dwellers who describe the interior of their cave.
We have the possibility, then, of understanding the origins of our tradition freely::
if we make the greatest effort; that is, understanding freely what has always been
handed {390} down as more or less self-evident. But what is “self-evident” is really38
always not understood.This lack of understanding is the final reason why the struggle
against the tradition has become possible and necessary. The final result; the factuall
ignorancee of the origins (e.g., μεγαλοψυχία–nobility).39
The question concerning the religious situation of the present is no serious
topic. The serious thing that is meant by this question is the question concerning
the right life. The answer to this question requires not only no special attention to
the present situation, but in fact the determined40 return to our historical origins,,
the uncompromising scrutiny of the supposed “achievements” of history.

38Ger.: im Grunde.
39Ger.: Vornehmheit. The Greek megalopsychia means, lit., “greatness of soul.”
40Ger.: entschlossenen.
APP D C
APPENDIX

LEO STRAUSS: “THE INTELLECTUAL SITUATION


OF THE PRESENT” (1932)1

Translated by Anna Schmidt and Martin D.Yaffe

Introduction: captatio benevolentiae


The topic is disreputable. A proper scholar2 adjusts his ambition to speaking only
on topics he has expert knowledge of, that is, on topics whose material he knows
and has adequately worked through intellectually. I do not have the reassuringg
awareness:3 here no one can damage or hurt me, here I am knowledgeable, as
regards this evening’s topic. I am not a specialist in the present: I have neitherr
climbed all its heights nor descended into all its nooks. The awareness that it is
impossiblee to be a specialist in the present probably prompts people who care about
their scholarly prestige to stay away from this topic.
If I nevertheless put up for discussion the intellectual situation of the present,,
it is because I feel emboldened by the following consideration. Certainly it is very
fine and comfortable to be able to step forth as if clad in iron armor. But the armorr
of scholarship always clads a being scarcely made of iron—namely, a human beingg
who questions. Now, there are questions that are hard to ask and even harder to
answer, so that one would really prefer to deny they exist. But they force themselves
onto us: we ask them and we answer them—even if with the worst conscience..
Now, the precaution of not doing things in public that one does secretly with a
bad conscience is certainly very prudent; but perhaps too prudent; is it not almost

1[Leo Strauss, “Die geistige Lage der Gegenwart,” GS–2 441–64, 623.] Unpublished. Bound manuscript
with 12 written pages (three of them loose) and inscribed cover, in ink, with additions and corrections
in pencil. Two pages (two sides) with the draft of an alternative introduction under the title Preliminaryy
Remark and a further sheet (two pages) attached with the plan of the lecture, all in pencil. On the coverr
and on the first page of the manuscript, Strauss has noted: 6.II.1932. Leo Strauss Papers, Box 8, Folder 6,,
Department of Special Collections, University of Chicago Library. {HM}
2Ger.: Wissenschaftler. Or: scientist (depending on the field of study).
3Ger.: Bewusstsein. Or: consciousness (henceforth depending on the context).
238 APPENDIX C

d 4 Therefore, if we do not want to be cowards, we should admit that


already cowardice?
now and again we nevertheless think about the intellectual situation of the present
and hence under certain conditions express these thoughts among ourselves {442}..
They are not entirely absurd, after all—perhaps they will even be of some use forr
our respectable scholarly work as well. That being said, on to the subject-matter.
1. What does the intellectual situation of the present matter to us Jews?? It matters to us
insofar as the present situation of Judaism matters to us.
Both the dissolution process and the consolidation process determined in a
European manner.5
That seems paradoxical: Does not the consolidation, in contradistinction to
Jewish self-renunciation, have the character of Jewish self-reflection, of the return
and retreat to Judaism? Does it not have the restitutio in integrum as its aim? Is not the
integrity of Judaism the guiding thought of the Jewish movement?
In order to recognize the European determination of the consolidation process,,
one need only look at the 3 steps of this process in a row.

1. Political Zionism. People as a natural group of human beings, which is held


together by a common enemy; national association as an imperative of honor..
Difference from the tradition: Pinsker’s6 motive. Trust in oneself versus trust in
God (not: power vs. spirit).
2. Cultural Zionism. Jewish tradition turns into the stuff of a European behavior::
Jewish humanism. Ahad Ha‘am7 and Hegel. Difference from the tradition: not
revelation but development.
3. Return to the Law. Rosenzweig8 and Elective Affinities.9 Difference: the Law w
as such. In principle: European reservations against the tradition (Cohen, The
Social Ideal)10—or renunciation of European prejudices. In either case: the
situation of the present matters to us.

4Ger.: Duckmäuserei. Or: hypocrisy.


5Ger.: europäisch bedingt.
6Leo Pinsker (1821–91). See his Autoemancipation (1882) in The Zionist Idea: A Historical Analysis and
d
Reader, ed. Arthur Hertzberg (New York: Harper & Row, 1959), 181–98. Cf. LSEW W 104.
7Ahad Ha‘am, pen name of Asher Hirsch Ginsberg (1856–1927), the “founder of cultural Zionism” ”
(Strauss, Preface to SCR, in JPCM M 144–45 with 341, 355n41). See Ahad Ha‘am’s address to the Firstt
Zionist Congress (1897), in JWM M 541–43. Cf. LSEW W 81, 119, 131, 203–4.
8See note 3 of appendix B.
9Ger.: die Wahlverwandtschaften. That is, the title of Goethe’s novel. Rosenzweig understands the term
m
to mean “the maze of feeling” in contrast to “the maze of action” (the emphasis is Rosenzweig’s); seee
Franz Rosenzweig: His Life and Thought, ed. Nahum N. Glatzer (2nd ed.; New York: Schocken Books,,
1963), 13. Cf. Strauss, Preface to SCR, in JPCM M 153.
10Hermann Cohen (1842–1919):“in the middle of World War I, the greatest representative of German n
Jewry and spokesman for it, the most powerful figure among the German professors of philosophy of
his time, stated his view on Jerusalem and Athens in a lecture entitled ‘The Social Ideal in Plato and thee
Prophets’” (Strauss, JA, in JPCMM 398–99); see Cohen, “Das soziale Ideal bei Platon und den Propheten,””
Hermann Cohens Jüdische Schriften (3 vols.; Berlin: C. A. Schwetschke & Sohn, 1924), I, 306–30..
Cf. LSEW W 76, 107, 109–10, 112–14, 134, 139–73, 216, etc.; Strauss, Preface to SCR, in JPCM M 154,,
158–59, and “Introductory Essay to Hermann Cohen, Religion of Reason Out of the Sources of Judaism,” in n
JPCM M 267–82; and chapter
p 4 of the ppresent volume.
APPENDIX C 239

What does the intellectual situation of the present matter to us Jews? It matters
to us insofar as the present situation of Judaism matters to us; for the present situa-
tion of Judaism is at the same time determined by Judaism’s past, by Jewish historyy
andd by the present world situation.The determination of the Jewish situation at any
given time by two factors—by Judaism and by the world—has existed at least for as
long as we have lived in the Galut.11 Determination by the second factor has con-
siderably intensified, so much that it has changedd fundamentally since the beginningg
of the age of assimilation and the dissolution of the Jewish tradition dating from this
age. Now, in the last decades, the {443} dissolution process has been countered by
the attempt at a consolidation of the setting of Jewish life,12 by Zionism in particular..
But no one who knows this attempt from seeing for himself and does not judge it
by its surface can fail to see that it is by dint of European thoughts and demands that
the consolidation of Judaism is being attempted in our time. For example, it may be
true—though in the end it is not as true as most people believe—that Jewish social-
ism has its origin in the demands and promises of the Prophets: this socialism would
not have been able to receive its present form without the authoritative influence
of European ideas. The consolidation of the setting of Jewish life is no less determined in a
European manner13 than is the dissolution of Jewish tradition preceding it.That seems para-
doxical. Is not this consolidation precisely the consequence of Jewish self-reflection
in contrast to Jewish self-renunciation in the age of assimilation?
Now this determination can be evaluated in various ways. One can say: we have
learnedd many things from Europe in the last 150 years, much that is dubious, but
nevertheless also a few things of undoubted value; that is, we have learned a few
things that we could not have learned from the Jewish tradition; we thereby have
certain reservations vis-à-vis the Jewish tradition, European reservations; what ensues
accordingly is this demand: that the consolidation of the setting of Jewish life be
carried out in a manner that takes into account these reservations that cannot be
renounced; and for just that reason we must be concerned with Europe.
One thingg about this view14 cannot be disputed: it is upright; it does not make the
task comfortable for itself; it does not smuggle in foreign elements under the coverr
of the Jewish flag, that is, by the use of biblical and talmudic passages torn out off
context. But it causes a certain unease, since it is alarmingly reminiscent of the view
of Reform15 that wanted to renew Judaism by making it up-to-date and that thereby
only rendered Judaism hollow and sentimental. Over and against all attempts at
Reform, the argument of S. R. Hirsch16 will always remain victorious—that it
would be contradictory to measure the eternal by standards of time.

11Heb.: Exile.
12Ger.: jüdischen Lebenszusammenhangs.
13Ger.: europäisch bedingt.
14[Note in the margin:] Example Cohen’s “The Social Ideal in Plato and the Prophets” {HM}
15That is, the “reformation of Judaism sought by enlightened Jews [comprising] not simply a dimi-
nution of the ritual burdens of the Jew but also an elimination or, at least, a blurring of the ethnic and
national features of traditional Judaism” (JWM
( M 156).
16Samson Raphael Hirsch (1808–88), German rabbi and intellectual founder of neo-Orthodoxy. See
his “Religion Allied to Progress” (1854) and “A Sermon on the Science of Judaism” (1855) in JMW
197–202, 234–35. Cf. CaM 184n58–58, with APT n58 as supplemented in LSMC C 603n58.
240 APPENDIX C

Let us make these facts a bit clearer to ourselves. The Jewish {444} problem,,
whose urgency in the age of National Socialism scarcely needs proof, forced out
of Herzl17 the idea of the Jewish national state, which then, simply for reasons off
realistic politics, was altered into the idea of the Jewish national home in Palestine..
Herzl imagined this state as being exactly like a European state. But whereas in
European states the national state idea was indissolubly connected with the national
culture—I recall the significance of the French Revolution for the original French
nationalism, and of France’s Catholic tradition for today’s French nationalism— —
merely political Zionism lacked that sort of native soil: Herzl’s Palestine was in
Ahad Ha‘am’s opinion nothing else than a Jewish Liberia. Thus Zionism was led
from being merely political Zionism to being cultural Zionism, the demand forr
national culture, and that means: the care and development of the Jewish tradition
came to be accepted. Now, no one could overlook that things are different with
Jewish culture than with the cultures of other peoples. Jewish culture is identi-
cal with learning and fulfilling the Law. Thus many cultural Zionists were led to
acceptance of the Law and submission to the Law, and so Zionism was getting everr
closer to Jewish tradition. There arose the possibility that European reservations
vis-à-vis the Jewish tradition were no longer even possible and necessary: the integ-
rity off Judaism seemed to become possible again.
Now there is something awkward about acceptance of the Law by cultural
Zionism, since according to the view of Jewish tradition Jewish Law is a Law given
by God. Were the Law to prescribe only actions and prohibitions, it could in the
end be fulfilled also by unbelievers; but it also and especially prescribes prayers::
how should he pray who does not believe in God? The atheistic Zionist is hence
confronted with the question, why does he not believe in God? Since the unbelieff
of a Jew in our time is nothing else than the general unbelief, the atheistic Zionist,,
in any case, sees himself compelled to concern himself with the intellectual situa-
tion of the present.
But not only he. Those Jews of our time who took an active part in the con-
solidation process, who in this way came to {445} accept the Law and who have
not been driven mad by the difficulty of believing in God, conceive of the Law
differently than the Jewish tradition. I recall the outstanding man of this group,,
Franz Rosenzweig, who reproaches Jewish orthodoxy for de facto having grantedd
priority to prohibitions over commands (e.g., with respect to $#);18 in contrast,,
he wants to regard prohibitions only as the other side of commands. In his struggle
against the rigidity of the Law, he goes so far as to want to dissolve entirely the
distinction between minhagg and din.19
The question concerning the intellectual situation of the present is ambiguous::
it can be understood in a way that is matter-of-fact or in a way that is vain; it can

17Theodor Herzl (1860–1904), father of political Zionism, founder of the World Zionist Organization,,
and author of Der Judenstaatt (The Jews’ State; 1896). See his “A Solution of the Jewish Question,” in JMW
W
533–38. Cf. LSEWW 81, 83–87, 102, 119, 128, 203, with the editor’s note at 82n2.
18Heb.: shabbos. That is, “Sabbath,” transliterated according to the Ashkenazic (Germanic)
pronunciation.
19Heb.: custom ...law. See, for example, Franz Rosenzweig, ed. Glatzer, 233–47; cf. Strauss, Preface to
SCR,, in JPCM
SCR M 153.
APPENDIX C 241

serve the seriousness of self-reflection or the frivolousness of self-satisfaction andd


self-assurance. We want to come to a rather drastic distinction between the serious
and the vain question concerning the intellectual situation of the present.We begin
priorr to this distinction: we pretend as if the question were nott ambiguous.
Let us suppose then that this question is straightforward—it is however not natu-
ral. In order to see this, one need only consider the following: at any time there
existed an “intellectual situation of the present”; but no one concerned himselff
with it; no one ever knew thatt there was an “intellectual situation of the present.””
Thoughtful people concerned themselves with the eternal; and the temporal—the
temporal that was important from [the point of view of] the eternal and for the sake
of the eternal was never the present but rather the past, the old: what was founded
in the past and has lasted for a long time has stood the test of time; and what has
stood the test of time has the opinion going for it that it is true. In contrast, the pres-
ent, the new, is held to be suspect; it is only for a brief while, it is brief, short-lived, a
momentary ripple not worth one’s attention. In recalling this earlier way of think-
ing one understands that interest in the intellectual situation of the present is not
naturall to man but is tied to certain historicall conditions, that it really is something off
the present.Thus we get our first answer to the question about the intellectual situation
of the present: this situation is characterized {446} by the question about it. If, then, we
wish to understand the intellectual situation of the present, we must ask: why does
the whole world today ask about the intellectual situation of the present?
This question cannot be settled by the statement: well, we live today, and there-
fore what happens today is of greater importance for us than what happened in the
past. This answer is justified regarding all practical tasks, small and large. If I want
to buy myself a hat, then what matters to me are the present hat designs and the
present hat prices only; if someone wants to get married, then what matters to him
are the provisions of the German Civil Code only and he does not need to take an
interest in Roman law or the Code of Hammurabi; what interests the statesman is
only the present constellation of powers, etc., etc. But if one asks about the principles
of action, about thee right and good, then it is conceivable that here it would be
extremely dubious to let oneself become involved with the present: namely, assum-
ing that the present were an age of decline. Why then do we inquire today about
the intellectual situation of the present as a matter of principle20—and not for instance
about the intellectual situation of the biblical time or of the Greek golden age?
One need only ask the question in this form in order to already have the answer::
were these ages for us still the classic ages, the authoritative ages, then they would be
at the center of our interest. But precisely this is characteristic of our age: that the
old traditions—the tradition of the Bible and the tradition of Greek philosophy— —
have lost their power. The 17th and 18th centuries, the Age of Enlightenment, has
won freedom from all tradition. The fact that a particular doctrine or institution can
invoke, be it the Bible, be it Plato or Aristotle, no longer compels us.
The Enlightenment won the freedom from traditions by fighting the traditions
in the name of the principles of the tradition: it fought against Aristotelian-scholastic

20Ger.: in prinzipieller Absicht.


Absicht.
242 APPENDIX C

science in the name of the Greek principle of science; it fought against the intoler-
ance of Judaism and Christianity in the name of the biblical principle of love off
one’s neighbor. The 19th century radicalized this fight of the Enlightenment: it
challenged the principles of the tradition; it called into question science as such andd
love of neighbor as such; it tore down the pillars on which the European world
is built. This decisive act, this completion {447} of the Enlightenment, is tied and will
always remain tied to the name of Friedrich Nietzsche.
Not only the traditions but the principles of the tradition were called into
question by Nietzsche. The freedom of thinking that had been won by the
Enlightenment, the freedom to think the opposite of what the Middle Ages hadd
thought has become infinitely exceeded; we are completely free. But free for what?
Have the principles of the tradition been replaced by other binding principles? Not
at all. We therefore do not know w at all whatt we are free for, what we live for, what
the right and good are according to which we can be at peace with ourselves. Ourr
freedom is the freedom of radical ignorance. The intellectual situation of the present is
characterized by our knowing nothing anymore, by our knowing nothing.
Out of the recognition and admission of this ignorance grows the necessity off
questioning, questioning about the right and good. And here the following paradox
presents itself: while the present is as compelled to question as any age, it is less capablee
of questioning than any age. We mustt question without being capablee of questioning..
From this embarrassment arises the question concerning the intellectual situation
of the present.
If someone says today that the first and most urgent question is the question
concerning the right life and that this question must be answered by reason, that is,,
scientifically, then the present confronts him with the verdict: this question cannot
be answered by scientific means; the answer to this question can be only a valuee
judgment, and value judgments cannot be justified and cannot be refuted by scien-
tific means; they rest on will or on belief, on the free decision of the person. Since
that is so, a universally binding knowledge about the right life is not possible—there
exists at bottom a “polytheism of values.” The de facto anarchy is thus claimedd to be
necessary and thereby consolidated.
I do not have the possibility here of proving with the requisite thoroughness
the untenability of this utterly unphilosophical view, that is, a view that cuts off ff
the real question. This proof would be identical with a radical critique of the life’s
work of Max Weber.21 Here I would prefer only to recall two points. 1) The con-
cept of “value judgment” presupposes that there are judgments that are not value
judgments—hence properly scientific judgments; it is in this sense that value-free
science mattered to Weber. Now it is not {448} that difficult to show that Weber’s
science, which he intended to be value-free, is wholly conditioned by his value
judgments; these are the ultimate presuppositions of his scientific research. 2) Now
Weber has in no way shirked the clarification of these presuppositions; on the con-
trary: his whole scientific life’s work aims precisely at this clarification, that is, the
understanding of its presuppositions from their history; thereby he provided himselff
with an altogether different basis than a merely personal decision.

21Cf. “Conspectivism”
Conspectivism (appendix A), note 40.
APPENDIX C 243

In principle: the “free decision of the person” that does not want to depend on
any justification “does not take place in a vacuum.” It is conditioned by the history
in which the person concerned stands. The knowledge of man’s being conditioned
by the history that is his history is called historical consciousness. The historical con-
sciousness that develops into historical sciencee is the closest counterauthority against
the ruling anarchy: as science, historical science makes possible universal validity.
But as it happens, historical consciousness is just that factor that thwarts the
question concerning the right life. For if man is essentially historical, then there is
not thee right life; but each age, each historical situation has its “right life,” its ideal off
life. Thereforee we cannot ask about thee right life, but only about ourr right life, about
the ideal of life that is up-to-date,22 the right life of the present. But how to know
what ideal of life is the one suitable to the present? That is possible only if the situ-
ation of the present is known.
We began from the fact that the intellectual life of the present is characterized by
the question concerning the intellectual situation of the present. We asked about the
reason23 for this question. The reason for this question is the necessity to ask about
the right life. Under the presuppositions of historical consciousness the question
concerning the right life compels us to ask the question concerning the intellectual situa-
tion of the present.
Because man is essentially historical, there are no eternal principles, no eternal
ideal of life. One can thus not ask about thee ideal of life, but only, at best, about the
present ideal of life. In order to determine the present ideal of life, one must know
the present situation.The present situation is known in the life of the present, which
we grasp above all in the intellectual productions of the present. {449}
Now, these productions themselves, however, have the character—explicitly orr
implicitly—of answers to the question concerning thee—or rather concerning the
present—ideal of life. And these answers are all in contradiction with one another..
The situation of the present—it consists in the contradictions of the present. That is
why one can only extract a unified ideal of life from the situation of the present iff
these contradictions can be resolved in the form of a higher unity: the present ideal
of life would be the synthesis of these actual answers that are effective today. But
such a synthesis is impossible. What is the higher synthesis supposed to be in which
Marx’s and Nietzsche’s ideal of life—the conviction that exploitation is bad in itselff
and the conviction that the abolition24 of exploitation is the abolition of life— —
could be unified? Or how is a synthesis of capitalism and communism imaginable?
Whoever proposes a synthesis here is confusing synthesis with compromise.
But even granted that a synthesis of the answers effective at present were pos-
sible, it is a question whether this synthesis would disclose for us the reall situation
of the present. For if all of today’s positions are mistaken about a basic character of thee
present situation, then this fundamental defect cannot be disposed of by the synthesis
of these positions: on the contrary, a synthesis would only reinforce this defect.

22Ger.: zeitgemäß.
23Ger.: Grund.
24Ger.: Aufhebung.
244 APPENDIX C

Allegedly, the situation of the present is knowable from the present positions..
How does one recognize that a position is a present one? Surely not by its being
represented in a writing published in 1932. Even now books still appear that are
written from the standpoint of Thomas Aquinas, from a standpoint, that is, that no
one will so easily describe as being a present one. In order to recognize a standpoint
as a present one, one must already have a guiding idea of the present. And this guid-
ing idea can be gained only from the knowledge of the entire historical process
out of which the present comes. In any case, the situation of the present cannot be
known from the present.
What is present may be said only on the basis of a knowledge of the entire
course of history. Now, this course is open to extremely different interpretations..
Which of these interpretations is the right one? There seem to be as many possibili-
ties as there are present positions. If the anarchy of present positions is not {450}
overcome,25 then the question of which interpretation of the entire course of his-
tory is the right one cannot be answered. And since we can determine the situation
of the present only if we can answer this question, the situation of the present is nott
determinable, not knowable.
Allegedly, the situation of the present is knowable from the entirety of positions
effective in the present. Why from alll these positions? Because they are equivalent..
Why are they equivalent? Because each one sees facts that are not seen or seen
only vaguely by the others. But obviously it is not seeing everything equally clearly
that is important—but seeing clearly what is important and vaguely what is not
important. For if one position looks at the world from the frog’s perspective and
another looks at it from the bird’s perspective, there is surely no doubt which off
these positions takes priority. I must therefore already know beforehand which facts
are important. But this presupposes that I know what is important. But if I know
this, then I know which life is the right one, and I do not need to ask about the
situation of the present at all.
Now as it happens, the reason why we ask about the situation of the present
is the fact that we do not know what is right. And it turns out that the question
concerning the situation of the present cannot be answered. Thus the question
concerning what is right cannot be answered at all. Thus the radical ignorance
remains and must remain. We are thus condemned to live without orientation; that
is, we cannot live at all. Let us attempt to determine this inability of ours to live
more exactly.
The intellectual situation of the present is determined by the historical con-
sciousness. This means that eternal, unconditional principles of living are not
recognized: all that is left are conditional, precisely historically conditioned prin-
ciples. Now, in our world, fundamentally different historical principles are effec-
tive, but these can now no longer be summed up in one universally binding,,
eternal order: the polytheism of values, anarchy, rules. Now the fact cannot be
completely forgotten that in earlier times eternal, unconditional principles know-
able to reason itselff, and hence an order, were held to be possible and necessary..

25Ger.: aufgehoben. Cf. the previous note.


APPENDIX C 245

This belief is now considered naive: we know based on a radical reflection fl that
rational knowledge about the right is not possible. Being more radically reflective,,
we are superiorr to the past. {451} Do we not thus have knowledge as an uncon-
ditional standard? No; for we merely say: iff knowledge is the ideal of life, then
the more radical knowledge is preferable to the more naive knowledge; but thatt
knowledge is the ideal of life is itself historically conditioned. In fact, conditioned
by the European tradition founded by the Greeks. In the non-European worlds
there are completely different ideals of life, and it is impossible for us to hold
these worlds to be inferior to us on account of their naivety. Hence the result: his-
torical consciousness leads to the awareness of superiority over the European past
and the awareness of the complete equality of non-European ideals; and in many
cases: contempt for the European past and prostration before everything exotic..
Now, it is natural to man to treasure and cultivate what is his own, what is handed
down to him by his forebears, whereas he confronts what is foreign proudly, suspi-
ciously, cautiously, at most with respect and admiration. Measured by this natural
stance, the stance dominant in Europe today appears to be antinatural, perverse..
Our inability to live, which manifests itself in our inability to question, is ourr
unnaturalness, the unnaturalness of our world.
Nonetheless we, too, are still in a certain way natural beings. And even if, in this
respect, we had to despair in the face of ourselves, the fact that even today children
are generated naturally and born naturally could reassure us. And even if these
children become corrupted soon enough by the dominant unnaturalness, there yet
remains the hope, so long as there are human beings on the earth, that some day
human beings will be able to be natural again.
We, too, are still natural beings. That we are still natural shows itself in the fact
that we, confronted with the ignorance of what is right, escape into the question
concerning what is right—escape from the unnaturalness of our situation.The needd
to know, and therefore the questioning, is the best guarantee that we are still natural
beings, humans—but that we are not capable off questioning is the clear symptom
of our being threatened in our humanity in a way that humans have never been
threatened.
Under the presupposition of historical consciousness, the question concerningg
the right life compels us to ask the question concerning the intellectual situation
of the present. Since this question cannot be answered, then the question concern-
ing {452} the right life seems no longer answerable. Should it be answerable, this
would be possible only by calling historical consciousness into question. But is this not
a fantastic undertaking? How w may historical consciousness be called into question?
By recognizing basically this: historical consciousness is itself historically condi-
tioned, therefore itself destined to give way to another consciousness. There is a
world, that is, a real, historical world beyond historical consciousness. That this
possibility exists in principlee no one will dispute. But, it will be said, this world is
the barbarism that awaits us no matter what; historical consciousness will go away
if humanity unlearns what it has learned arduously enough over the past centuries;;
the renunciation of historical consciousness is identical with the relapse into a stage
of lesser reflection.
246 APPENDIX C

Let us pause here for a moment. Historical consciousness is—one cannot empha-
size this strongly enough—according to its own view a stage of higher awareness::
we know moree than the earlier generations; we know more deeply, more profoundly,,
than the earlier ones that everything human is historically conditioned. But as it
happens historical consciousness is the reason why, although we are compelled to
question, we are incapable of questioning. Thus we are more incapable of question-
ing than the earlier generations—since we know more, since we know too much..
But we are compelled to question since at bottom we know nothing. Being funda-
mentally ignorant we cannot come to knowledge since we know too much. Since we believee
we know too much. We will not be able to remove our radical ignorance until this
belief that we know is abolished.
Historical consciousness includes in itself the conviction that we stand at a
higher stage of reflection than earlier human beings: we regard ourselves as havingg
progressed. Now there are without question many men especially today who are off
the opinion that our age is an age of decline. But precisely such men are mostly
of the opinion that the character of decline of our time has its reason in our beingg
too conscious, that in our time knowledge plays a role not allotted to it in healthy
times; precisely such men usually believe that there are opportunities for comingg
to knowledge as there have never been before (Spengler);26 they are of the opinion
that precisely becausee today it is twilight, the owl of Minerva could begin its flight..
That we have progressed is quite the dominant opinion. {453}
But how do things really stand concerning our progressiveness? Our progres-
siveness could only be the result of the modern development. Let us therefore
question the history of this development.
The modern centuries are dominated by the pathos of progress in knowledge
and through knowledge. At the beginning of the modern development stands the
fight against scholastic science. This science was stagnating; it essentially did noth-
ing other than transmit and explain Aristotle. Regarding physics, within which the
fight primarily took place, the founders of modern philosophy and physics were
astonished at the fact that the Scholastic philosophers investigated not nature but— —
Aristotle. (Cremonini.)27 This was possible only because Scholasticism presupposedd
explicitly or implicitly that science was essentially completed: one did not see a
possibility of advancing; one did nott have the intention of advancing; one did not havee
confidencee in advancing. Modern philosophy begins with the completely opposite
intention: plus ultra. One can illustrate this opposition in the following manner..
Science emerged with the Greeks in an age of civic flourishing, as a concern of free
citizens; it had its center in the city of Athens, of whose citizens Thucydides writes
that they are always ready to hope, always anxious to discover something new—as
opposed to the Lacedaemonians, who regard science with mistrust, who are nott
hopeful, little confident in their ability, holding on to the old. Scholastic science was
(at least in Christendom) pursued by monks. The newer philosophy is once again

26Oswald Spengler (1880–1936), Der Untergang des Abendlands [The Decline of the Westt] (1918);;
cf. PoR 100.
27Cesare Cremonini (1550–1631).
(1550 1631). Prolific expositor of Aristotle; accused of Averroism.
APPENDIX C 247

the concern of free citizens, who, as once were the Athenians, were ready to hope
and keen to do something new again. Full of self-confidence did one thus confront
classical philosophy; it was not long until a dispute began about the superiority off
the moderns over the ancients. In any case, the unimaginable revolutions in the nat-
ural sciences proved that a progress beyond classical science was possible and real. It
is a question, however, whether this progress was a fundamental progress, a progress
in the foundation. If one turns with this question to the modern philosophers, then
all will answer that also in philosophy the moderns got further than the ancients..
And this getting further is indeed quite evident; for assuming equal effort and equal
seriousness, must not {454} science also progress with the progressing of time?
Certainly so, if one assumes in addition the same starting-point.
In contrast to this predominant awareness of progress and progressiveness in the
more recent centuries, however, stands the fact that in these centuries the convic-
tion of the authoritative significance of the Greeks keeps breaking through. From this
point of view, the fight against the Middle Ages appears as an attempt to recoverr
Greek freedom, Greek science. At the beginning of modern philosophy stands the
Renaissance, the renaissance of antiquity. In fact, the fight against the Scholastics is
in considerable part conducted in thee manner of opposing genuinee Greek science— —
whether it be Aristotle himself, or Plato, or Democritus and Epicurus—to the cor-
ruptt Greek science of Scholasticism.
Now, one can say: this counter-movement
r is always a misunderstanding or only
a device or a subsequent corrective of the reall movement, which is a movement off
progress. This remark is surely justified within certain limits—namely, insofar as one
keeps to the explicit consciousness of the newer centuries. But if one looks at what
went on in reality, then one gets a different impression.
Even the fiercest opponents of the Greeks believed themselves able to put into
effect the progress they had in mind only after they had laid the foundation for it
by a return, namely, by a return to nature. Rousseau’s call to return to nature, which
has become part of our collective memory, is only onee example of that and not
even the best one. The reversionary character of modern philosophy shows itselff
much more fundamentally in the fact that is decisive for the whole span of the 17th
and 18th centuries: in the fight against prejudices that fills these centuries. The word
“prejudice” is indeed the Enlightenment’s polemical keyword—it is met with so to
speak on every page of every writing of the Enlightenment. One must freee oneselff
from prejudices, and this freeing is accomplished by retreatingg to a plane, or even a
point, from which one can finally free progress of prejudice once and for all.
Today’s reader of a writing from the Age of Enlightenment {455} in which
prejudices are fought so fiercely will often have to smile when he realizes just
how strong were the prejudices of the supposedly prejudice-free gentlemen of the
Enlightenment. One could even say: the century of the Enlightenment was the
century of prejudices. We today are therefore very cautious in the use of the word
“prejudice.” Historical consciousness has corrected us in that it is not possible for man
to live prejudice-free: every age has its prejudices; and the fight against prejudices
as such always only means the fight against the prejudices of others—for with one’s
own prejudices there is the awkwardness of not being able to know them as preju-
d ces. If then
dices. t e only
o y historical
sto ca consciousness
co sc ous ess hasas set us straight
st a g t about the
t e universal
u ve sa
248 APPENDIX C

significance of the category of “prejudice,” the overcoming28 off hhistoricall consciousness


would bring with it the overcoming of the universal significance of “prejudice.”
With respect to questioning historical consciousness it was said: if everythingg
human is historical, then historical consciousness is also historical; that is, destined
at some time to be no more. This historical conditionality of historical conscious-
ness can be demonstrated concretely by the category of “prejudice.”
The wordd “prejudice” is older than modern philosophy—but only in the Age
of Enlightenment does it achieve authoritativee significance. What Greek philoso-
phy fought against was opinion or appearance, not prejudice. (However, νόμωι
φύσει.)29 How did “prejudice” get this authoritative significance in the Age of the
Enlightenment?
About this we receive a remarkable piece of information from a famous scho-
lastic of the 12th century. This philosopher cites in an important context an enu-
meration by Alexander of Aphrodisias of the reasons for the differences of opinion
in philosophy, for the difficulties of philosophizing. 3 reasons are dealt with there,,
which all express the naturall difficulties of philosophizing, that is, such difficulties as
exist always, at alll times where human beings philosophize. Now, this scholar adds
to this enumeration the following words:

In our time there exists a fourth reason, which he (sc., Alexander of Aphrodisias)
did not mention, since it did not exist among them; namely, habitt and schooling; for
human beings by nature love what they are habituated to and incline to it ...this hap-
pens to man regarding the opinions in {456} which he has grown up: he loves them
and holds on to them and stays away from differing opinions. For this reason as well,
therefore, man is prevented from knowing the truth. This happens to the multitude
regarding God’s corporeality ...due to habituation in the writings in which one firmly
believes and to which one is habituated, whose literal meaning seems to indicate
God’s corporeality.30

Now surely there were many Greek writings in which the gods were presentedd
corporeally.Why did these writings not compromise Greek philosophy? It is there-
fore not being accustomed to writings in general, not having grown up in a tradi-
tion in general, but rather being accustomed to very specificc writings, having grown
up in a tradition of a very specificc character: namely, in a tradition possessing an
authority as unconditionall as that of the tradition of revealed religions. The fact that a
tradition based on revelation entered the world of philosophy increased the naturall
difficulties of philosophizing by adding the historicall difficulty.
In other words: The natural difficulties of philosophizing have their classical
depiction in Plato’s allegory of the cave. The historical difficulty may be illustrated
by saying: there now w exists another cave beneath the Platonic cave.
The Enlightenment’s whole fight against prejudices is in a sense sketched and
thus anticipated in the cited statement from the 12th century.The statement sheds a

28Ger.: Aufhebung.
29Gk.: by convention, by nature.
30Maimonides, Guide of the Perplexed
d I 31.
APPENDIX C 249

new light on this fight: the Enlightenment’s fight against prejudices did not have the
absolute meaning that the Enlightenment itself attached to it; and not because man
always has and must have prejudices, but precisely the reverse, because prejudices
in the strict sense of the word are only the “prejudices” of the revealed religions..
This implies that the fight against prejudices has reached its end only when revealedd
religion has been called into question in its foundation and in its consequences.
If then the Enlightenment’s fight against prejudices is only the fight against thee
historical difficulty of philosophizing, then the true goal of this fight is only: the
recovery of philosophizing in its natural difficulty, of natural philosophizing, that is,,
of Greek philosophy. {457}
[Draft of an Alternative Introduction]

Preliminary Remark
We are all convinced that there is an unambiguous distinction between good and
bad; therefore also between virtue and vice. Hence we distinguish between virtuous
and vicious actions with the confidence of sleepwalkers. We are so sure of this dis-
tinction that we are even able to distinguish between virtuous and vicious topics. It
is virtuous, say, to examine the questions of the sources of the collected Hippocratic
writings or the connection between Hellenistic philosophy and the formation off
Islamic sects; it is stilll virtuous if one grasps how Plato presents his teacher Socrates,,
or if one reconstructs a whole philosophical system from meager relics. To speak
about vicious topics in public is not unobjectionable. Let it only be said that amongg
the vicious topics there is nothing more vicious than the one we want to talk about
tonight: the intellectual situation of the present. The proof of this is that a true
scholar31 will never occupy himself with this question in his capacity as scholar; the
scholar knows that making a conjecture or finding evidence of filiation is a much
greater blessing than occupying himself with things that may be more interesting,,
but that can only be talked about vaguely. If scholars are the embodiment of virtue,,
then writers32 are the embodiment of vice. What we are attempting tonight is thus
writer-like. And that is so by necessity. Imagine, if you please, that someone wanted
to speak about the intellectual situation in the 14th century. In order to do this in
the right manner, he would have to have studied for years; he would have to have
become a specialist in the 14th century. Since without a doubt infinitely moree is
being written today than in the 14th century, how many years would someone have
to have studied the intellectual situation of the present in order to be capable off
treating it in a scholarly way. I have to confess for my part that I am far from being
a specialist in the present. I am not embarrassed to say that I have never read norr
heard a word by either Graf Keyserlingk or Margarete Susmann.33 Not only will
my claims often lack a sufficient material foundation—the claims will also often
appear to you as in themselves confused {458} or otherwise defective. I am here-
with asking for your indulgence.

31See note 3.
32Ger.: Literaten.
33See “Conspectivism”
Conspectivism (appendix A), p. 219.
250 APPENDIX C

But why talk about such a matter at all, a matter that appears not to allow forr
scholarly treatment? Well—even the most virtuous man occasionally has the need
to take a break from his virtue. Or, to say it less frivolously: it is good for even the
most respectable scholar to put the books aside for once and do some thinking like
a simple man of the people. Whatever he comes up with then does not need to be
right, it does not need to be more than a reasonable conjecture; but it may be useful
for his respectable work nonetheless.
Presupposing historical consciousness, the question concerning what is right
compels the question concerning the situation of the present. Is this path really
necessary? If all human thinking is in itself historical, then it has been arranged that
we—when we, that is, men of this present world asking about what is right as such,,
believe we have found what is right as such—have thereby eo ipso found the answerr
that corresponds to our world, the presentt ideal of life. We cannot escape the fatee off
historicity—but we need not be concerned about that in our thought.

Fate as Principle
Historical consciousness must no longer be the principle. In what sense must his-
torical consciousness be called into question? Insofar as it leads to the question
concerning the situation of the present. It leads to this question only insofar as it is
made the principlee of questioning, insofar as it wants to be more than knowledge off
the conditions and the fates of questioning.
Now, this path is in no way necessary. If everything human is in itself historically
conditioned, then, without our needing to concern ourselves about it, it has been
arranged that in searching for thee ideal of life we are bound to find only the one
that corresponds to our world, to the present, the presentt ideal of life. If everythingg
human is itself historical, then for just that reason we do not need to be concerned
with the historicity of our question. It is not the knowledge of historicity as such
that leads to the explicit question concerning the present ideal of life and therefore
concerning the situation of the present, but incorporatingg the knowledge of historic-
ity, [i.e.,] historical consciousness, into the {459} question, making it the element,,
the presupposition, the principlee of the question—when this knowledge in fact per-
tains only to the conditions and fates of questioning. But if the primary question
of the human being who does not live in a binding given order is the question
concerning a binding reasonable order, it is this question alone that must primar-
ily occupy him and not the fate and the condition of this question. And only if it
turns out that he cannot answer this primary question of his without consideringg
the conditions and fates—then and only then does his historicity matter to him.That
he needs this detour, however, must be demonstrated; it is not at all self-evident. Off
course, we cannot avoid this detour today. Why that is the case, I will attempt to
show, by34

34Here Strauss’s draft of an alternative introduction breaks off in mid-sentence.


APPENDIX C 251

[Plan of the Lecture in Draft]

1. The disreputability of the topic—the scholar adjusts his ambition to speakingg


only on topics with respect to which, as far as knowledge of the material and
its intellectual penetration are concerned, he can step forth as if clad in iron
armor. At least I for my part do not at all have the reassuring awareness—here
no one can damage or hurt me, here I am a first-class expert—concerning
the topic I am going to speak about tonight. Nevertheless it appears to me
permissible to speak about it. For the iron armor of scholarship in each case
clads a being scarcely made of iron, a human being who doubts. Not to voice
these doubts because they themselves are unclear, as are the speculations that
[come] from concerning oneself with these doubts35
2. Why does the situation of the present matter to us Jews? → Impossibility off
understanding “Law.” Reservations and prejudices.
3. a) Question ambiguous
b) Question not naturall but historical.
c) Situation of the present characterized by the question concerning it.Why?
Why not concerning situation in biblical time or Greek golden age?
d) Destruction of the power of the tradition by the Enlightenment: {460}
Enlightenment fights traditions in the name of the principles of the
tradition.
e) Nietzsche calls the principles into question: we are totally free. But forr
what?
f) No binding principles any more. Radical ignorance, necessity off
questioning.
g) Being compelled to question as any age, we are less capable of questioningg
than any age.
i) “Value judgments” not scientifically justifiable (according to Weber).— —
Free decision of the person, anarchy—does not take place in a vacuum,,
however, but is conditioned by history. Historical consciousness—seems to
guarantee the overcoming36 of anarchy.
k) Meanwhile historical consciousness makes the question concerning the
right life impossible—at best the presentt ideal of life → question concern-
ing the intellectual situation of the present. Therefore: situation of the pres-
ent characterized by historical consciousness.
l) This question not capable of being answered. Hence no orientation, and
consequently no life: we are incapable of living.
m) Perversion and unnaturalness. But nevertheless we are naturall beings—in
that we question. Natural beings threatened in our naturalness by an unnat-
ural world.

35[The text breaks off here in the middle of the line. LS has left blank a space of two lines to the next
period and has noted in the left margin:] Duckmäuserei in dem Nichtbehandeln des Themas [cowardice/
hypocrisy in the non-treatment of the topic] {HM}
36Ger.: Überwindung.
37Ger.: aufgehoben
aufgehoben..
252 APPENDIX C

n) Our unnaturalness shows itself to us in [the form of] historical consciousness


insofar as this is understood in the way that it leads to the question concerningg
the situation of the present. Therefore: division of historical consciousness.
o) Possibility of this division in principle: but not relapse into barbarism? Thus::
historical consciousness has progressedd when compared to naivee conscious-
ness. That is: Being fundamentally ignorant we cannot come to knowledge since wee
know too much, believee we know too much. If, then, we are therefore to come
to knowledge, then this belief in knowing must be overcome.37 Historical con-
sciousness must be overcome insofar as it means: it itself constitutes as such a superiorr
manner of knowing.
p) How do things really stand concerning our progressiveness? While the pro-
gressive tendency is primary, modern philosophy is always characterized by a
countermovement—and not only as a condition independent of it, but as its
condition proper.—State of nature. Liberation from prejudices.The meaning38 off
modern philosophy must be understood in light of this fundamentall intention
{461} of this philosophy.
q) “Prejudice” and historical consciousness.
i) Historical consciousness can be interpreted by means of the category of
“prejudice”.
ii) Prejudice a historical category.
RMbM39 citation
r) Natural and historical difficulties of philosophizing—the meaning of the
fight against prejudices is liberation from the historical difficulties. First and
second cave.
s) Modern philosophy’s self-interpretation as progress hides this meaning com-
pletely. It leads to a continuous erosion of the tradition, in accordance with
the intention of the reversionary tendency, while it leads at the same time
to the consolidation of a new tradition—the modern tradition. A process off
destruction that passes itself off as a process of construction.
t) Finallyy one stands at the foundation of the tradition—one stands at the start-
ing-point of the questioning from which the tradition has arisen—without
knowing it, or in any case, without drawing the consequences from it.
Nietzsche’s significance: Socratic question is not posed seriously, but is cut
off by a dictate. Inner lack of clarity: Callicles40 andd intellect41 (order off
humanity).
u) In any case, Nietzsche has enabled us to understand the Socratic question
again, to recognize it as ourr question. The Platonic dialogues are no longerr
self-evidentt for us—no longer self-evidently all right, no longer self-evidently
wrong, surpassed, out of date, but we read them as if we were conductingg
them ourselves if we were capable of doing that.

38Ger.: Sinn. Likewise in subsections r) and s).


39That is, Maimonides. See note 8 of appendix B.
40Cf. Plato, Gorgias 481b–527e (esp. 482c–486d).
41Ger.: Geist.
APPENDIX C 253

But we are not capable of doing that, since all the concepts that we aree
equipped with derive from the modern tradition. This is whatt we have to
know—this is why we have to concern ourselves with the intellectual situ-
ation of the present. The question betrays the awareness of the fact that the
question concerning what is right cannott be answered withoutt being clearr
about our incapacity to question—but this question is fundamentally mis-
guided if it is supposed to replace the real question.
The question concerning the intellectual situation of the present should
serve {462} to awaken in us the willingness to come out of the cave off
modernity—it is absurd if asked for its own sake: it would then amount to
our describing the interior decor of the 2nd cave.
Historical consciousness has the function of leading us back to the natural
questions. It is a self-misunderstanding of historical consciousness if it pre-
tends to be a higher type of knowing.
v) Plato and the Nomos and Revelation
APP D D
APPENDIX

LEO STRAUSS: “A LOST WRITING OF FARÂBÎ’S” (1936)1

Translated by Gabriel Bartlett and Martin D.Yaffe

In an article that appeared in the last volume of JQR, I. Efros tries to show that
“the second of the three parts of (Falaquera’s) Reshit hokmah, entitled ‫החלק השני‬
‫במספר החכמות‬,2 is a literal translation of the whole of Farâbî’s important work
known as the ‘Encyclopedia’3 or by its Arabic title as Iḥṣ I â al-‘Ulûm” (JQR, N.S.,,
Vol. 25, p. 227).4 This assertion is in need of considerable qualification, which Efros
manifestly neglected to undertake only because the editions of Iḥṣ I â al-‘ulûm were
not yet available to him.5

1[“Eine vermißte Schrift Farâbîs,” GS–2 167–77, 614.] First published in Monatsschrift für Geschichte
und Wissenschaft des Judentums (Breslau, 1936), 80th annum, volume 1 (January), 96–106. The essay came
into being from a handwritten submission in November 1935. {HM}
The originally published edition uses “Farâbî” throughout, rather than the more accurate “Fârâbî.”
In a personal communication to the translators, Professor Heinrich Meier reports that Strauss did not
correct the published spelling in the personal copy of the article in which he noted numerous remarks.
For this reason, the translators have likewise retained “Farâbî” as Strauss’s spelling throughout the article.
2Heb.: “The Second Part in the Enumeration of the Sciences.”
3I.e., what is now known as his The Enumeration of the Sciences. For a brief description, see Charles
Butterworth’s introduction to the selection translated in Alfarabi, The Political Writings: “Selected
Aphorisms” and Other Texts, trans. Butterworth (Ithaca, NY: Cornell University Press, 2001), 72f., or
to that same selection as reprinted in MPP–2 18f. Strauss uses the then-received title Encyclopedia (or
Encyclopedia of the Sciences) throughout.
4I.e., Israel Efros, “Palaquera’s Reshit Hokmah and Alfarabi’s Iḥṣ
I â’ al-‘Ulum,” Jewish Quarterly Review
w n.s.
25 (London, 1934–35): 227–35.
5For the editions available to Strauss but not to Efros, see note 21.
256 AP P E N D I X D

Reshit Chokhmah II falls into nine chapters (more exactly, “parts”), Iḥṣ
I â al-‘ulûm
into an Introduction and five chapters. The correspondence is

Reshit Chokhmah II IIḥṣâ al-‘ulûm


Chap. 1 Introduction
Chap. 3 Chap. 1
Chap. 5 Chap. 2
Chap. 6 Chap. 3
Chap. 7–8 Chap. 4
Chap. 9 Chap. 5

As this list leads one to suspect and the examination of the texts themselves con-
firms, the second and fourth chapters of R. Ch. II are not borrowed from Farâbî’s
Encyclopedia. The same goes for not inconsiderable parts of all the remaining chap-
ters. Not borrowed from Farâbî’s Encyclopedia are:

in the 1st chapter: the 5th and 6th “uses” of the book (ed. David,6 21.2–19);
in the 3rd chapter: the explanation of ‫שם‬, ‫ פעל‬and ‫( מדבק‬24.8–27.20);7 Falaquera
himself says, in referring to this interpolation: ‫וצריך שנבאר זה אע״פ שאין זה מכוונת זה‬
‫( הספר‬24, 18–19);8
in the 5th chapter: the last part (39.25–41.10). Falaquera himself says, in referring
to this interpolation: ‫( וכבר הארכתי וגו׳‬41.8);9
in the 6th chapter: the explanation of analysis and synthesis (43.14–28); {168}
in the 7th chapter: the last part (51.24–53.9). Falaquera himself says, in referring to
this interpolation:‫ ואני כדי‬.‫ואלו הענפים לא זכרום מקצת הפילוסופים בזכרם מספר החכמות‬
‫( שיהיה הספר שלם זכרתים‬53.8–9);10
in the 8th chapter: the last part (54.19–55.30). The interpolation begins with the
words: ‫;ויש מי שחלק זו החכמה וגו׳‬11
in the 9th chapter: the last part of the paragraph on political science (58.19–59.5).

The parts of chapters 7, 8, and 9 not taken from Farâbî’s Encyclopedia are a (more
or less literal) translation of the corresponding sections in Ibn Sîna’s Encyclopedia
(Iqsâm al-‘ulûm).12 I indicate the passages from Ibn Sîna’s writing according to the
Latin translation by Alpagus (in: Avicennae Compendium de anima etc., ab Andrea
Alpago ...ex arabico in latinum versa,Venetiis 1546):

6Moritz David, ed., Schemtov ben Josef ibn Falaqueras Propadeutik der Wissenschaften Reschith Chokmah
(Berlin: M. Poppelauer, 1902).
7Heb.: noun, verb ...preposition.
8Heb.: “And it is necessary that we explain this, although this is not part of the intention of this book.”

9Heb.: “And I have already spoken at length, etc.”
10Heb.: “And these are the branches. Some of the philosophers did not mention them when they y
mentioned the number of the sciences. But I have mentioned them so that the number would be com-
plete.” (Reading ‫ המספר‬for ‫הספר‬. Otherwise: “so that the book [sic] c would be complete.”)
11Heb.: “And there is someone who has divided this science, etc.”
12Ibn Sîna (Avicenna), Fî aqsâm al-‘ulûm al-‘qliyyah (Epistle on the Divisions of the Rational Sciences)..
Cf. Muhsin Mahdi’s introduction to a selection from it in MPP–1 95f., or to that same selection in n
MPP 2 74f.
MPP–2
AP P E N D I X D 257

R. Ch. II, chap. 7 (51.24–53.9) Avicenna, ed. Alpagus, fol. 141–42


chap. 8 (54.19–55.30) fol. 143–44
chap. 9 (58.19–59.5) fol. 140b.13

In order to identify the source of the part of the third chapter that is not taken
from Farâbî’s Encyclopedia, one must consider that this section (24.8–27.20) treats
the same subject as the greater part of the thirteenth chapter of Maimonides’s Millott
ha-higgayon,14 where, incidentally, a sentence is cited from Farâbî.The last section off
the fifth chapter (39.25–41.10) is borrowed from Farâbî’s writing on the purposes15
of Plato and Aristotle (see below, n. 39).
The most important supplement of Falaquera is the fourth chapter, which treats
the genesis of the sciences.16 It cannot yet be proved at present that this chapter is a
more or less literal translation of a section from a writing of Farâbî’s. But it is indu-
bitable that the same thoughts developed there go back to Farâbî. Let, for example,,
the conclusion of the chapter (30.28 ff.) be compared with the concluding part off
I â al-‘ulûm (or with Reshit Chokhmah 59), or the immediately preceding passage
Iḥṣ
(30.25–28)17 with Farâbî’s k. taḥṣîl al-sa‘âda,18 39–42 (or with Reshit Chokhmah
70.17–19). Reminiscent of Farâbî {169} is, also and above all, the discussion occur-
ring passim in this chapter pointing to the political function of science in that the

13For the last passage mentioned, cf. Strauss, Philosophie und Gesetz (Berlin: Schocken, 1935), 111 [=
GS–2 112; PLA 122–23]. {LS}
14Heb.: Logical Terms. See Israel Efros, Maimonides’ Treatise on Logicc (New York: Academy for Jewish
Research, 1938; reprint, Literary Licensing, LLC, 2011).
15The German Tendenz (Tendenzen in the plural) recurs frequently in Strauss’s article and is rendered
as either “purpose” or “tendency”: when it refers to Fârâbî’s writing on Plato and Aristotle whose
existence Strauss is inferring from Falaquera et al., it is “purpose(s)”; when it refers to the distinctive
characteristics of Falaquera’s own writing, it is “tendency.”
16Handwritten marginal note in Strauss’s personal copy: cf. Farâbî. De ortu scientiarum (ed. Baeumker),,
Revue néoscholastique de philosophiee 41 (1938): 84 ff. Cf. Farmer, Al-Farabi’s Arabic-Latin writings on music,,
1934. {HM}
17This passage is, incidentally, the best key to the understanding of Maimonides’s prophetology..
Falaquera says there that the art of lawgiving is the art of representing figuratively, by means of the
imagination, speculative concepts that are difficult for the multitude, and the means of bringing about
those political activities that serve the attainment of happiness, and the amphibolous speech concern-
ing those speculative and practical matters that are accessible to the multitude only in an amphibolous
manner.This suggests, among other things, that the prophets’ dependence (asserted by Maimonides as by
the Falâsifa) on the perfection of the imagination is to be understood only on the basis of the political,,
lawgiving function of prophecy. {LS}
Handwritten marginal note in Strauss’s personal copy: cf. Arist[otle] Metaph[ysics] 1074a38 and
Alex[ander of] Aphr[odisias] as well as Averroës ad loc. {HM}
Handwritten marginal note on “amphibolous speech” in Strauss’s personal copy: perhaps translation
of ‫( اﻗﻨﺎع‬cf. Reshit Chokhmah 70.19) = “persuasion” {HM}
18The k. taḥṣîl al-sa‘âda or kitâb al-sa‘âda (kitâb = book) is translated into English by Muhsin Mahdi
as The Attainment of Happiness, the first part of the trilogy whose existence Strauss is inferring in the
present essay and which also includes The Philosophy of Plato and The Philosophy of Aristotle. See Alfarabi’s
Philosophy of Plato and Aristotle, trans. Mahdi (Ithaca, NY: Cornell University Press, 1969; 2nd ed., 2001)..
Particularly as regards its second and third parts, Strauss refers to the trilogy in question usually as the
“book” (or “writing”) “on the purposes of Plato and Aristotle,” and occasionally as “The Two Philosophies.””
Hiss authorities
aut o t es heree e are
a e Ibn
b al-Qifti
a Q t (see notes
otes 32–33)
3 33) and
a d Averroes
ve oes (see notes
otes 441–43),
43), respectively.
espect ve y.
258 AP P E N D I X D

sciences emerge in the h nation,, or rather in the nations. The same consideration also
justifies the (provisional as well as hypothetical) attribution to Farâbî of the second
chapter, which treats the genesis of language in the nations.
Falaquera’s book is a decidedly Jewish book, whereas the model is not to the
same degree an Islamic book.Thus there are no citations in Farâbî from the Qur’an
or other Islamic sources corresponding to the Bible citations in R. Ch. II (54.4–5
and 54.15–17). The same difference shows itself perhaps most clearly in that, to the
“uses” of an encyclopedia of the sciences enumerated by Farâbî, Falaquera adds the
following two “uses” while remarking explicitly that both these “uses” would be off
greater importance than the previous ones (sc., borrowed from Farâbî):

1. a Hebrew encyclopedia of the sciences is necessary so that the loss of the


“wisdom of our Sages,” owing to the Galut, can be made up for again;19
2. “from this book it will become clear whether we are authorized by ourr
Torah to learn these sciences (sc., the ones about to be discussed) or not, orr
whether they contradict anything that is mentioned in our Torah....20 This
was the first intention in composing this book” (21.2–19).

Falaquera’s determining interest, to prove the agreement between the doctrines


of philosophy and the doctrines of the Law, is in no manner characteristic of Farâbî,,
as his {170} Encyclopedia of the Sciences shows in particular: according to Farâbî, the
religious sciences (fiqh
( and kalâm) are no more than branches of political science..
In this connection, let it be noted that Farâbî’s explications of the kalâm are signifi-
cantly shortened by Falaquera (cf. 59–60).
As interesting as this is in some respects, the fact that in the secondd part of Reshitt
Chokhmah the greatest part of Farâbî’s Encyclopedia as well as significant parts of Ibn
Sîna’s Encyclopedia are preserved is of no special importance; for the originals have
been edited and moreover are accessible in Latin translations that are more usable
than Falaquera’s Hebrew translation because they are more complete.21 It is other-
wise as regards the thirdd part of the Reshit Chokhmah; for this part, which contains
the translation of Farâbî’s book on the purposes of Plato and Aristotle, must replace
the original, the greatest part of which is regarded, at least for the time being, as

19Cf., in this connection, Maimonides, Guide of the Perplexed


d I 71, beginning. {LS}
20As regards this order of the questions—a. legal permissibility of philosophizing, b. contradiction or
agreement between philosophy and law—cf. my discussion of Ibn Rushd’s Faṣl al-maqâll in Philosophie
und Gesetz, 71 [= of Averroes’s Decisive Treatise, in GS–2 70f.; PLA 84]. {LS}
21Iḥṣ
I â al-‘ulûm was edited by [Uthmân] Amin, Cairo 1350 [A.H.], and A. G. Palencia, Madrid 1931
(cf. P. Kraus’s review of this edition in Islam XXII 82–85) [= Paul Kraus, “Kleine Mitteilungen und
Anzeigen: Angel Gonzáles Palencia, Alfârâbî Catálogo de las ciencias,” Der Islam 22 (1934): 82–85]. The
Latin translation by Gerard of Cremona is reprinted in the Palencia edition.––Iqsâm al-‘ulûm was edited
in the collection of Tis‘ rasâ’il, Istanbul 1298 [A.H.]. The Latin translation by Alpagus can be found in
the collection of Alpagus cited above. {LS}
AP P E N D I X D 259

lost.22 Steinschneider
h d has h already
l d suspectedd that
h the
h secondd andd third
h d parts off Reshit
Chokhmah III might be taken from the aforementioned work of Farâbî’s.23 Now,,
after Farâbî’s k. taḥṣîl al-sa‘âda has been made available by the Hyderabad edition off
1345,24 it can be proved that the entire third part of Reshit Chokhmah is a transla-
tion (albeit significantly abridged)25 of Farâbî’s book on the purposes of Plato and
Aristotle.
The first part of R. Ch. III is a (incomplete) translation of Farâbî’s k. taḥṣîl al-
sa‘ada. As proof, I first cite the beginnings of both works and then list the mutually
corresponding passages.

Farâbî: “Book on the Attainment of Happiness. The human things that must be real-
ized among the nations and the inhabitants of the cities, by means of which earthly
happiness in the initial life and the highest happiness in that life can be realized,
(fall into) four kinds: the speculative virtues, the cogitative virtues, the moral vir-
tues {171}, and the practical arts. The speculative virtues are those sciences whose
ultimate intention is solely this, that the beings and what they encompass become
intelligible (νοητά).” 26

Falaquera:27 ‫ ואומר כי הפילוסופים זכרו‬:‫החלק הראשון בביאור הדברים ההכרחיים בהשיג ההצלחה‬
‫שהדברים האנושיים אשר בהגיעם באומות ובאנשי המדינות תגיע להם עמהם הצלחת זה העולם‬
‫בחיים הראשונים וההצלחה האחרונה בחיים האחרונים ארבעה סוגים והם השלמיות העיוניים )והשלמיות‬
‫ והשלמיות העיוניים הם החכמות‬.‫( והמלאכה המעשית‬read: ‫המחשביים( והשלמיות המעשיים )היציריים‬
.‫אשר הכוונה האחרונה מהם שיגיעו עמהם הנמצאים מושכלים על אמיתתם בלבד‬

22See now, however, Alfarabius De Platonis Philosophia (Falsafat Aflâṭun), ed. Franz Rosenthal and
Richard Walzer (London, 1943) and Al-Fârâbî’s “The Philosophy of Aristotle” (Falsafat Arisṭûṭâlîs), ed.
Muhsin Mahdi (Beirut, 1961), with Mahdi’s editorial remarks in his Alfarabi’s Philosophy of Plato and
Aristotle, 151–52. Cf. also note 51.
23Alfarabi, St. Petersburg, 1869, 176–178. [= Moritz Steinschneider, Al-Fârâbî (Alpharabius), des ara-
bischen Philosophen Leben und Schriften (St. Petersburg, 1869).] {LS}
24See note 21.
25See notes 28 and 29.
26Cf. Alfarabi’s Philosophy of Plato and Aristotle, 13. In Strauss’s German text, “intelligible” is the plural
noun Verstandenen: “intelligibles.” We have rendered it as a printer’s error for verstandenen.
27Heb.: “The first part in the explanation of the things necessary in the attainment of happiness: He
says that the philosophers have mentioned that the human things by the attainment of which the nations
and the people of the cities gain both happiness in the prior life of this world and final happiness in the
afterlife are of four kinds: the theoretical virtues, (the deliberative virtues) the productive virtues, and the
practical art. And the theoretical virtues are the sciences whose ultimate intention is that existing things
thereby become intelligible by their truth alone.”
260 AP P E N D I X D

Falaquera Farâbî Falaquera Farâbî


62.2–4 (Falaquera’s supplement) 67.128–68.26 26.16–29.15
62.4–5 2.6–7 68.27–32 31.3–10
62.5–9 (Falaquera’s supplement) 69.1–12 31.17–32.17
62.9–28 2.7–3.19 69.13–15 (Falaquera’s supplement)
62.28–2929 4.2 69.16–26 36.12–37.5
62.31–63.10 4.16–5.10 69.26–27 37.16
63.10–14 (Falaquera’s supplement) 69.27–28 38.5–6
63.15–31 5.11–6.19 69.28–70.8 38.9–39.8
63.32–64.1 8.10–13 70.8–10 (Falaquera’s supplement)
64.1–12 10.7–11.3 70.10–14 39.8–12
64.12–20 11.6–18 70.14–23 42.5–16
64.20–65.23 12.14–14.9 70.23–29 42.19–43.8
65.23–66.8 15.3–16.5 70.29–71.10 44.13–45.11
66.8–9 16.15 71.11–13 (Falaquera’s supplement)
66.10–30 20.8–22.8 71.14–72.4 45.12–47.5

The Taḥṣîll concludes with the words:30

Philosophy, the description of which this (sc., what was just explicated) is, has reached
us from the Greeks as stemming from Plato and Aristotle. Neither of these {172} two
has given us philosophy without at the same time giving the ways to it and the ways
to its revival in case it has become tarnished or annihilated.We proceed in the follow-
ing to present the philosophy of Plato and the ranks of order of his philosophy: We
begin with the first part of the philosophy of Plato and thereupon follow the ranks
of his philosophy, one after the other, until we have exhausted them altogether. We
proceed likewise with the philosophy that Aristotle has given us: to wit, we begin to
present his philosophy starting from its first part. From this it will become clear that
the purpose they were both pursuing with what they have given is one and the same,
and that they have both endeavored to give one and the same philosophy.31

The Taḥṣîll is therefore the introduction to a work that was devoted to the
presentation of the philosophy of Plato and Aristotle. The same goes for the
first part of R. Ch. III: the second part of R. Ch. III treats the philosophy off
Plato, the third part the philosophy of Aristotle. It might even follow already
from the conclusion of the Taḥṣîl, but above all from the report of Ibn al-Qifti

28Falaquera justifies the omission with the words: ‫“[ וכבר דברנו בשלמיות היציריים בחלק הראשון‬and we have
already spoken of the productive virtues in the first part”] (66.31–67.1). {LS}
29Falaquera justifies the omission with the words: ‫“[ וכבר זכרתי בחלק השני וגו׳‬and I have already men-
tioned in the second part, etc.”] (62.29–30). {LS}
30Cf. the conclusion of R. Ch. III 1: ‫הפילוסופיא שזה תארה הגיע מהיונים מאפלטון ומאריסטוטליס וכל‬
‫ ואני ראיתי לזכור ספרים בפרט‬.‫( כשתאבד‬read: ‫אחד מהם נתן הפילוסופיא והדרכים לקנותה ולהדשם )ולחדשה‬
.‫“[ ואע״פ שנזכר כל זה במה שקדם בכלל‬And the philosophy of this description has come from the Greeks— —
from Plato and from Aristotle. And each of them has given philosophy and the ways to acquire it and to
renew it when it is lost. And I have seen fit to mention their books individually, even though all this has
been mentioned in general in what has preceded.”] {LS}
31Cf. Alfarabi’s
Alfarabi s Philosophy of Plato and Aristotle,
Aristotle, 49–50.
49 50.
AP P E N D I X D 261

d Lippert, 278)32 about


(ed. b Farâbî’s
âbî book
b k “OnOn the Purposes (of the Philosophy) of
Plato and Aristotle,” that this book, just like R. Ch. III, was formally divided into
three parts: into an introduction, in which “the secrets and the fruits of the sci-
ences” were investigated and “the gradual ascent” from science to science was
discussed; “thereupon he (Farâbî) begins with the philosophy of Plato by com-
municating the purpose that he (Plato) pursues with it, and his (Plato’s) writ-
ings about it; thereupon he has the philosophy of Aristotle follow this.”33 This
description corresponds completely with R. Ch. III. At the conclusion of the
edition of the Taḥṣîl, the heading of the following part is given once more; it is::
“The Philosophy of Plato, its Parts, and the Ranks of Order of its Parts, from
First to Last.”34 The heading of the second part of R. Ch. III, ‫בפילוסופית אפלטון וסדר‬
‫חלקיה מראש ועד סוף‬,35 is manifestly the translation of this.
As regards in particular Farâbî’s presentation of the philosophy of Aristotle in the
work so named, according to the report of Ibn al-Qifti it was structured as follows::
Farâbî

prefaces it (The Philosophy of Aristotle) with a significant introduction in which he


{173} makes his (Aristotle’s) gradual ascent to his philosophy known; thereupon he
begins to describe his (Aristotle’s) purposes with respect to his logical and physi-
cal writings, one book after the other, until his discussion about these things, in the
manuscript we find, arrives at the beginning of theology (metaphysics) and the proof
for this (sc., for the necessity of metaphysics), resting as it does on physics.

The third part of R. Ch. III is structured exactly so: from an introduction that
depicts Aristotle’s path to his philosophy (78.6–80.9), there follows the presenta-
tion of logic (80.10–81.13) and physics (81.14–91.1), and finally the proof for the
necessity of metaphysics, which rests on physics, and the beginning of metaphysics
(91.1 to the end): metaphysics itself is absent in R. Ch. III also. This absence, inci-
dentally, is not to be attributed to a corruption of the manuscripts, as Ibn al-Qifti
seems to think, but it corresponds to Farâbî’s plan:36 whereas he characterizes his
presentation of the philosophy of Plato with the words “We begin with the first
part of the philosophy of Plato and follow the ranks of his philosophy, one after the
other, until we have exhausted them altogether,” he says with respect to his pre-
sentation of the philosophy of Aristotle only “We begin to present his philosophy

32Ibn al-Qifti, Ta’rîkh al-ḥukamâ’, ed. Julius Lippert (Leipzig: Dieterich’sche Verlagsbuchhandlung,,
1903).
33The editor notes in connection with this: “That is all that we have found out about this book.” {LS}
The emphases in Strauss’s quotation from Ibn al-Qifti in the text are Strauss’s.
34Cf. Alfarabi’s Philosophy of Plato and Aristotle, 53.
35Heb.: “On the Philosophy of Plato and the Order of its Parts from Beginning to End.”
36Handwritten marginal notes in Strauss’s personal copy:
cf. Cuzarii V 1, Thomas ad EN N 1211; Efros’ RMbM M about Ab[raham] b[ar] Hiyya; Terminology
s.v. ‫“[ חכמה‬Wisdom”]
cf. Moreh Introd.: Politics > Metaphysics > Physics.
See below, n. 54. {HM}
For “Efros’
Efros RMbM M”:: see note 4.
262 AP P E N D I X D

fi parts on.”37 A complete presentation of the philosophy of Aristotle,


ffrom its first
therefore, was not at all intended by Farâbî.38 And that is why the third part of R.
Ch. III is a (by and large) complete translation of the third part of Farâbî’s book on
the purposes of Plato and Aristotle. Even Falaquera’s sole deviation from the struc-
ture attested to by Ibn al-Qifti of Farâbî’s work speaks for this conclusion: Falaquera
does not enumerate the individual logical writings of Aristotle. But that they had
been enumerated in his model, that this model therefore corresponds completely
to Ibn al-Qifti’s statements about the book in question by Farâbî, is proved by the
fact that Falaquera, true to his habit, {174} explicitly justifies the omission of this
enumeration:39 ‫( ארסטו׳ ונזכור‬sc. ‫וכבר זכרתי בחלק השני הספרים השמונה שחבר בה )בחכמת ההגיון‬
‫( אותה בכאן בקצרה‬80.11–12).40
In Ibn Rushd’s Long Commentary on the Metaphysics, a writing of Farâbî is cited,,
“The Two Philosophies,”41 which is identical with the writing translated by Falaquera..
The statement of Farâbî cited by Ibn Rushd from this writing can be identified in
R. Ch. III.

Ibn Rushd:42 “Una autem istarum opinionum est, quod agens creat formam, et ponit eam in
materia....Quidam dicunt quod illud agens invenitur duobus modis, aut abstractum a materia,
aut non. Illud autem, quod est non abstractum apud eos, est sicut ignis, qui facit ignem, et homo
generat hominem. Abstractum vero est illud quod generat animalia et plantas, quae fiunt non a
simili. Et haec est sententia Themistii, et forte Alpharabii, secundum quod apparet ex suis verbis
in duabus philosophiis: quamvis dubitet in ponendo hoc agens in animalibis generatis a patre
et matre.”43

37Cf. Alfarabi’s Philosophy of Plato and Aristotle, 50.


38Cf. also the similar break in the presentation after pointing from physics to metaphysics in Taḥṣîl
12–14. Let it be pointed out at least in passing that, among other things, the conclusion of the Aristotle
presentation in R. Ch. III 3 and the corresponding section in Taḥṣîll (or in R. Ch. III 1, 64–65) agree in
part literally. {LS}
39I.e., in R. Ch. 39.25–41. Falaquera has therefore taken up the relevant section of the book on the
purposes of Plato and Aristotle into his adaptation of Fârâbî’s Encyclopedia. That R. Ch. 39.25–41.10 is
borrowed from the book on the purposes shows up also in that this passage is composed in the same
narrative style that characterizes the book on the purposes: note the constantly recurring expression
here, as there: ‫“[ ואחר כך עיין‬And afterwards he examined”] (or ‫“[ חקר‬he investigated”] or the like). {LS}
40Heb.: “And I have already mentioned in the second part the eight books that Aristotle composed in
it [sc., in the science of logic], and we will recall it here in brief.”
41Handwritten marginal note on “The Two Philosophies” in Strauss’s personal copy: (sc., the exoteric
and the esoteric philosophy!) {HM}
42ad Metaph. XII (comm. 18, fol. 143, col. 2, l. 27–39). The passage is, incidentally, of importance also
for demarcating Farâbî’s teaching from Ibn Sîna’s. {LS}
43Lat.: “One of these opinions, however, is that the agent creates the form and places it in mat-
ter....Certain people say that this agent is found in two modes, either remote from matter, or not. For
them, however, the [agent] that is not remote is like fire that makes fire, and man generates man. In truth,
the remote [agent] is the one that generates animals and plants, which are not made from [something]
similar. And this is Themistius’s view, and perhaps Fârâbî’s according to what appears from his words in
The Two Philosophies: in any case, he would hesitate to place this agent in animals generated by a father
and mother.”
AP P E N D I X D 263

Falaquera: ‫… ורבים מהנפשיים יתנו אל החמרים אשר יפגשום ואשר יכינם הטבע להם נפש כי האדם הוא‬
‫אדם היה לפניו אדם והאדם מאדם וכמו כן בבעלי חיים )…( מה שאינו בבעלי חיים ומהצמחים מי שאינו מצמח‬
‫והגופים המחצביים אינם נהוים מהדומים להם במין ועל כן צריך לחקור אותו אלו ויותר מזה לחקור מי‬
‫שנתן האנושית על דרך כלל והדומה לזה משאר בעלי חיים וצורת מין ומין … ועל כן צריך לחקור מי‬
‫שנתן צורת אותו המין ועל דרך כלל מי שנתן צורות המינים אם הם הגופים השחקיים או השכל הפועל או‬
(90.22–31) 44.‫יהיה השכל הפועל נותן הצורה והגופים השחקיים נתנו תנועות החמרים‬

By this agreement, that R. Ch. III is a translation of Farâbî’s writing on the pur-
poses of Plato and Aristotle should be proved completely. {175}
This writing can therefore be reconstructed to some extent if a rule is first
secured for the use of those parts of it that are preserved only in Falaquera’s transla-
tion.45 This rule can be obtained by observing Falaquera’s tendency and technique
in his adaptation of k. taḥṣîl al-sa‘âda and Iḥṣ
I â al-‘ulûm. Falaquera in general trans-
lates very literally. Often, however, he leaves out significant parts of the original,,
occasionally by making known his deviation from the original. The supplements
are easy to recognize as such in general, without the comparison with the original
being necessary (or even only possible).46 Above all, if one considers the fact that
most of Falaquera’s supplements stem from the tendency characteristically distin-
guishing him from Farâbî, to prove the agreement between the doctrines of phi-
losophy and those of the Law. This tendency was pointed out above in examples
from his adaptation of the IḥṣI â al-‘ulûm. It shows itself also in his adaptation of the
Taḥṣîl.47 The fact that in the Taḥṣîll is found an explanation of “Imâm” (43.9–17) that
Falaquera left untranslated merely appears to contradict this; for with this expla-
nation Farâbî is pursuing the by no means believing, but philosophical intention
of leading away from the Islamic givens and toward the Platonic doctrine of the

44Heb.: “And many of the animate [beings] give to the materials that they encounter, [provided] that
nature has prepared them, a soul, as a man has been [originated] from another man before him, who
has been [originated] from a man, and in like manner with the animals....Some among the animals,,
and plants that are not [originated] from a plant, and the mineral bodies, are not [originated] from what
is similar to them in species, and therefore it is necessary to investigate these things, and moreover to
investigate who gave the human [soul] in general, and similarly [the souls] of the rest of the animals
and the form of each and every species....And therefore it is necessary to investigate who has given the
form of this particular species and in general who has given the forms of the species, whether it is the
heavenly bodies or the Active Intellect, or whether it has been the Active Intellect giving the form and
the heavenly bodies have given the movements of the materials.”
45That R.Ch. III 1 might be of use for establishing the text of the Taḥṣîll is to be noted in passing. {LS}
4672.21–25: ‫“[ ומצאנו וגו׳‬and we have found, etc.”]; 75.22–26: ‫“[ ויראה לי וגו׳‬and it has seemed to me,,
etc.”]; 77.3–11: ‫“[ ראינו וגו׳ ונשוב למה שהיינו בו‬we have seen, etc., and let us return to where we were”]. The
last example is especially important because, through this remark of Falaquera’s, it is ascertained that the
preceding report about Plato’s Republicc stems from his model. {LS}
47Cf. in particular the supplements at 68.13–15 and 71.11–13. {LS}
Handwritten marginal note in Strauss’s personal copy: 65.32: ‫—האלוה יתהּ‬cf., in contrast, taḥṣîll 15, par..
2, lines 1–2
1 2 {HM}
264 AP P E N D I X D

philosopher-king:
hl h k “The
h meaning off Philosopher
hl h andd First Leader d andd King andd
Lawgiver and Imâm is one and the same.”48
Since, therefore, Farâbî’s book on the purposes of Plato and Aristotle is able to
be reconstructed, and since his Encyclopedia of the Sciences, his Political Regime,49 andd
his book on the political regimes50 have been edited,51 his central writings are thus
preserved and available. The interpretation of his doctrine is therefore possible..
A sufficient proof that it is necessary, however, is the statement of Maimonides’s
about the “second teacher.”52 Maimonides writes to the translator of his Moreh
nebuchim:53 “Do not concern yourself with any books of logic other than those the
wise Abû Naṣr al-Farâbî has composed. For {176} everything he has composed in
general, and his The Principles of Beings54 in particular—everything is pure flour.””
And he immediately adds that the books of Ibn Sîna, with all their merits, cannot
compare with those of Farâbî. It is time that the conclusions from this authoritative

48Handwritten marginal note in Strauss’s personal copy: cf. Razi on Socrates as his Imâm (Kraus in
Orientalia 1935) [= Paul Kraus, “Raziana I,” Orientalia N.S. 4 (1935), 300–34]. {HM}
49Lit.: Perfect State. Strauss uses the German title Musterstaat, in reference to the Arabic text edited
by Friedrich Dieterici, Alfarabi’s Abhandlung der Musterstaatt (Leiden, 1895). See the following note and
notes 53 and 54.
50On the apparent redundancy here, cf. Muhsin Mahdi’s introduction to a selection from Alfarabi’s
The Political Regimee in MPP–1 31: “[The Political Regime] is known by two titles: the Principles of Beings
(or the Six Principles) and the Political Regime. The first title seems to have been extracted from the
opening passage of the work, which gives the impression that it is a treatise on the principles of the
natural world and their respective ranks of order: 1) the First Cause, 2) the Second Causes, 3) the Active
Intellect, 4) the soul, 5) form, and 6) matter. The entire first part of the work consists of an account of
these six principles and of how they constitute the bodies and their accidents. Only when one proceeds
to the second part (the human and political part ...) does one perceive that this account is an introduc-
tion to, and a preparation for, an account of political life and a classification of political regimes. Alfarabi
wrote a parallel book, the Principles of the Opinions of the Citizens of the Virtuous City, which discusses the
same themes in similar terms. As the titles indicate, however, the Political Regimee is concerned more with
regimes or constitutions, whereas the Virtuous City is concerned more with the opinions of the citizens
in these regimes.” See also note 54.
51As regards the edited texts of the Encyclopedia [sc., Enumeration] of the Sciences and the Political Regime,
see notes 3 and 49; also, Muhsin S. Mahdi, Alfarabi and the Foundation of Islamic Political Philosophy
(Chicago: University of Chicago Press, 2001), 241, 243. As regards the edited text of “the book on the
political regimes” (sc., the Virtuous City), see idem, 244.
52On this term, cf. Mahdi’s Introduction to Alfarabi’s Philosophy of Plato and Aristotle, 4f.: “Alfarabi’s sci-
entific or philosophic works proper—his commentaries, especially his large commentaries, on individual
works by Plato and Aristotle ...established his reputation as the greatest philosophical authority next to
Aristotle (Alfarabi was known as the ‘Second Master’)....Many of these works seem to be lost; the ones
that have survived remain for the most part hardly ever studied; and the few that have been edited deal
with specialized subjects whose relevance to the general character of Alfarabi’s thought and of Islamic
philosophy is not easy to establish.”
53I.e., Guide of the Perplexed. The translator is Samuel ibn Tibbon. Cf. Joel L. Kraemer, Maimonides
(New York: Doubleday, 2008), 438–43.
54The authentic title of this book is The Political Regimes. It consists of two parts, the first of which
treats the hierarchy of the cosmos, the second of which treats the hierarchy of the city. Structured in the
same manner is the book on the perfect state, which in the manuscripts of the British Museum and the
Bodleian Library is described simply as a “political book.” {LS}
AP P E N D I X D 265

explanation about the true proportions be drawn for the understanding of the
Islamic and Jewish philosophy of the Middle Ages: At the beginning of this epoch
of the history of philosophy there stands not just any “predecessor,” but the tower-
ing spirit who laid the ground for the later development and set down its limits
by making his task the revival of Platonic-Aristotelian philosophy as philosophy
proper.
APP D E
APPENDIX

LEO STRAUSS: “ON ABRAVANEL’S CRITIQUE OF


MONARCHY” (1937)1

Translated byy Martin D. Yaffe

. Finkelscherer, in an essay that appeared in the last annum of this journal,2


H has made the assertions that one of the most important antimonarchist argu-
ments of Abravanel, and “a quite similar attitude” toward the Judean monarchy, are
to be found “already in Joseph ibn Caspi.”That the attitude of both authors is quite
similar must be contested: Finkelscherer himself immediately notes the “difference
in tone between Caspi’s aristocratic-philosophical ideal and Abravanel’s theocratic
one” (81st annum, p. 506). On the other hand, the assertion that Abravanel’s argu-
mentation agrees with Caspi’s requires more serious consideration.
For Abravanel in any case, the most important part of his antimonarchical argu-
mentation, i.e., the one that is decisive for the whole question, is his explanation off
Deut. 17:14f., according to which this passage expresses no command to appoint
a king but merely a permission to do so. More precisely: he compares the provi-
sion regarding the appointment of the king with the provision regarding the ‫אשת‬
‫יפת־תאר‬.3 Accordingly, the sense of the crucial biblical passage is: if you should have
and express the wish, against the divine will, to appoint a king over yourselves,,
then you should not appoint as king over yourselves him whom you wish, but

1[Leo Strauss, “Zu Abravanels Kritik des Königtums,” GS–2 233–34, 615.] Unpublished. Typescript
with entries and corrections in Strauss’s hand. 2 pages, in the possession of Jenny Strauss Clay. Strauss’s
annotations refer to the essay by Herbert Finkelscherer, “Quellen und Motive der Staats- und
Gesellschaftsauffassung des Don Isaak Abravanel,” in Monatsschrift ffür Geschichte und Wissenschaft des
Judentums (Breslau, 1937), 81st annum, 496–508, and had evidently been intended as a brief review for
publication in this journal. Whether Strauss in fact sent the contribution to the editorial office is not
known. {HM}
2See the previous note.
3Heb.: “beautiful-looking woman” (Deut. 21:11).
268 AP P E N D I X E

him whom God will choose. Caspi says quite similarly (on Deut. 17:14): ‫וצום שלא‬
‫ישימוהו רק אשר יבחר י״י על פי נביא‬.4 Is this statement of Caspi’s necessarily saying, how-
ever, that the appointment of the king is not commanded but only permitted? He
is certainly not in any way comparing the provision regarding the appointment
of the king with the provision regarding the beautiful-looking woman, but with
the provisions regarding sacrifice (on I Sam. 8:6). He is thus pointing, as Last5 as
well as {234} Finkelscherer rightly remarks, to Maimonides’s explanation of the
sacrificial legislation. The explanation given in Moreh6 III 32, which both Last and
Finkelscherer have in mind, says: sacrifices do not correspond to the first intention
of the divine lawgiver, they are therefore not simply good, but they were ordained
only with respect to the ignorance of the people at that time, they are therefore
only conditionally good. This explanation by no means implies, however, that sac-
rifices, being ordained as only conditionally good, are not commanded but only
permitted. And therefore Caspi is in no way asserting as Abravanel does, undoubt-
edly in the wake of Ibn Ezra, that the appointment of the king is not commanded
but only permitted.
There is only one way to vindicate Finkelscherer’s assertion: one would have
to show that Caspi in fact considered the entire sacrificial legislation to be non-
obligatory. In other words: one would have to show that his words ‫למנות מלך מצוה ג״כ‬
‫כענין הקרבן‬7 contain a silent allusion not to Moreh III 32 but to Moreh III 46. For in
the latter chapter (III 102a–b, Munk)8 Maimonides says in plain words: “If we do
not perform this mode of worship at all, I mean the sacrifices, then we commit no
sin whatever.” Caspi, the thoroughgoing connoisseur of the Moreh, may very well
have been thinking of this somewhat more isolated remark, and not of the much
more famous expositions of Moreh III 32, when he compared the law of the kingg
with the sacrificial laws. But that he was actually thinking of that remark would
need proof. Only if this proof were supplied would one be obliged to assert defini-
tively that Caspi really has anticipated the decisive antimonarchical argument off
Abravanel.

4Heb.: “And he commanded them not to appoint him, except one whom the Lord will choose by
the say-so of a prophet.” Quoted by Finkelscherer, 506.
5Finkelscherer, 506n58, quotes I. H. Last’s edition of Caspi’s commentary on the Prophets, Adney y
Keseff (2 vols.; London, 1911–12), vol. I, 15, on I Sam. 8:6: ‫שיהיה לעמו מלך לשפטם‬
‫“( אין ראוי לפי התורה ש‬It is not
proper according to the Torah for our people to have a king to judge them”); Last’s editorial footnote add
loc. notes Caspi’s allusion to Maimonides's Moreh.At 506n57, Finkelscherer cites Last’s edition of Caspi’s
commentary on the Torah Matzref la-Keseff (Krakow, 1906) on Deut. 17:14, i.e., as the source for Caspi’s
view expressed in the sentence translated in note 4. And at 506n60, Finkelscherer again cites Caspi’s
Adney Keseff on I Sam. 8:6, as the source for the sentence translated in note 7, which Finkelscherer, 506,,
quotes with Caspi’s words ‫“( כענין הקרבן‬like the matter of sacrifice”) emphasized.
6I.e., Guidee [of the Perplexed].
d Strauss uses the traditional Hebrew title.
7Heb.: “And to appoint a king is a commandment, just like the matter of sacrifice.” See note 5.
8Maaïmonide, Le Guide des égar é rés, ed. and trans. S[alomon] Munk (3 vols.; Paris, 1856–66; nouv. éd.,,
Paris, 1970).
APP D E
APPENDIX

SUPPLEMENTARY NOTE ON ARCHIVAL MATERIALS

Jeffrey A. Bernstein

I n 1937, Leo Strauss wrote two essays on the thought of Don Isaac Abravanel
(1437–1508 ce)—“On Abravanel’s Philosophical Tendency and Political
Teaching” and “Zu Abravanels Kritik des Königtums” (translated as “On Abravanel’s
Critique of Monarchy” in appendix E of the present volume).1 These two essays
present a lens through which one sees Strauss’s engagement with later medieval
Jewish thinkers as it displays a newfound emphasis (characteristic of his writings
in the 1930s) on political philosophy and medieval rationalism. Further, as Joshua
Parens shows in chapter 8 of our volume, Strauss’s engagement with Abravanel
also discloses his first significant treatment of “the main line of Christian politi-
cal thought” as distinguished from the Jewish and Islamic variety (the exemplary
instance of which is the thought of Maimonides).
Around one month before the present volume went to press, I came across
a 43-page set of undated handwritten notes (in Box 20, Folder 14 of the Leo
Strauss Archive, Special Collections, Regenstein Library, University of Chicago)
ficance of which bears directly on both essays from 1937.2 Five pages off
the signifi
the manuscript, divided into three distinct sections, deal with Abravanel’s com-
mentary on I Samuel (Chapter 1,Verses 4–35). The first section carries the heading
“Abrabanels Kommentar zum Buche Samuël” while the next two sections simply
state “Abrabanel ad Samuël”. The remaining 38 pages deal with Abravanel’s com-
mentary on I Kings (Chapters 1 and 17–20). Given the degenerated state of some
of the pages, it is diffi
fficult to provide defi
finitive sections for the entire group; that
said, the overwhelming majority of pages can be sectionalized according to the des-
ignations that occur as book number-scriptural book-chapter-subscript verse—forr

1See GS-2 195–227 and 233–34, respectively.


2I would like to thank Nathan Tarcov and the Special Collections Department at the Regenstein
Library for allowing me to conduct research there. I would also like to thank Martin D. Yaffe, Richard
S. Ruderman, and Timothy W. Burns for inviting me to briefly present my findings in the present
context.
270 A P P E N D I X E : S U P P L E M E N TA RY N O T E

example, 1Reg174 (“Reg” being the abbreviation for “Regnum”—the conven-


tional Latin designation for the scriptural book).The manuscript as a whole,3 there-
fore, contains materials covered in both Abravanel essays.
One of the distinctive qualities about “On Abravanel’s Philosophical Tendency””
is its reference to many classical Jewish commentaries on the Hebrew Bible. These
occur, in that essay, largely in the notes and in one paragraph.4 “On Abravanel’s
Critique of Monarchy” (probably as a result of the scholarly focus and brevity off
Strauss’s treatment) makes no mention of these commentaries. From reading only
those essays, one might question how substantive a role these commentaries play
in Strauss’s treatment of Abravanel. The handwritten notes provide the followingg
information: At least with respect to I Samuel and I Kings (which texts figure
directly in “On Abravanel’s Philosophical Tendency” and, by implication, in “On
Abravanel’s Critique of Monarchy”), Strauss was working extremely closely with
the rabbinic commentaries. The notes largely assume a form similar to that of the
Miqra’ot Gedolot.5 He provides a verse, notes certain signifi
ficant Hebrew terms, pro-
vides Abravanel’s commentary on said verse, and then provides commentary on
the same verse from the Midrash Tanhuma, the Targum by Onkelos (ca. 35–120 ce),,
Solomon ben Isaac (“Rashi,” 1040–1104 ce), David Kimchi/Qimchi (“Radak/
Radaq,” 1160–1235 ce), and Levi ben Gershom/Gersonides (“Ralbag,” 1288–1304
ce). Strauss employs the conventional (German) shorthand for these commentators
and texts, hence, “Tanchuma,” “Raschi,” “RDK”/“RDQ,” and “RLBG.”
Strauss’s notes do not provide similar commentary from Maimonides; neitherr
do they provide analogous commentary from Christian sources (some of whom
Abravanel would have been familiar with), or philosophical analogues and/or diver-
gences.6 The question as to the occasion for which these notes were composed thus
remains open.7 Regardless of whether the notes were taken in specifi fic preparation
for the Abravanel essays or taken earlier and simply used in those essays, they off
fferr
an instructive glimpse into Strauss’s workroom and show readers the scrupulous
attention to detail that marked Strauss’s philosophical practice before, during, and
after the reorientation that he underwent in the 1930s.

3Listed on the Special Collections website as “Arabic Notebook ‘Biblical Book of Kings.’”
4¶26 in Isaac Abravanel: Six Lectures, ed. J. B. Trend and H. Lowe (Cambridge: Cambridge University
Press, 1937), 123; ¶28 in Leo Strauss On Maimonides: The Complete Writings, ed. Kenneth Hart Green
(Chicago: University Of Chicago Press, 2013), 607. Readers should note that Green’s edition follows
the GS.
5That is, the “large format Bible,” which provides verse-by-verse commentary supplied by the Targumim
(Aramaic translations of the Hebrew Bible) and Rishonim (medieval Jewish scriptural commentators).
6This is significant only insofar as “On Abravanel’s Philosophical Tendency” deals with Abravanel’s
divergence from the medieval Jewish adoption of Platonic political philosophy in favor of a more explic-
itly Aristotelian standpoint characteristic of medieval Christian thought.
7Thomas Meyer (Wake Forest University) suggests that the handwriting in the notes dates from
Strauss’s early Berlin years (1926–27). If this is correct, it opens up research possibilities into the possible
differences between his treatment of classical Jewish sources in 1) the 1920s, 2) his path through “Jewish
Thomism” (GS–3 765) in Philosophie und Gesetz, and 3) the Abravanel essays.
EDITORIAL
D TOR A NOTE
OT TO APPENDICES
APP D C S F AND
A DG

Hannes Kerber

trauss left among his papers three versions of the fragment1 “Exoteric
S Teaching”—a four-page manuscript (M) written on both recto and verso
in blue ink,2 a 19-page typescript (TS),3 and its carbon copy (CC).4 The two
typewritten versions are nearly identical5 but bear many different handwritten
corrections in pencil, red crayon, and black ink by Strauss and at least one otherr
person.
An edition based on TS was published with some editorial revisions by Kenneth
Hart Green in 1986.6 M and CC, however, throw new light on the genesis of the
essay. Further, a comparison of the typewritten versions with M strongly suggests
that the typist was not Strauss himself. Nor can TS be considered authoritative since
Strauss did not prepare the essay for publication. The deviations from M, most off
which occur in the last part of the essay, can therefore not be regarded with cer-
tainty as Strauss’s deliberate corrections but may be among the many errors made
by the typist that were missed by Strauss when he looked over TS and CC. (Forr
example, it is not clear whether Strauss wanted to leave out the second part of foot-
note XXXVI in the typewritten version or whether he simply did not notice the
typist’s failure to transcribe it.)

1In M,TS, and CC, the first paragraph is numbered “I.” However, this numeral was later crossed out in
both TS and CC. Likewise, the last line in M is the numeral “II,” which was not copied by the typist.
2Leo Strauss Papers (Box 17, Folder 2), Special Collections Research Center, University of Chicago
Library. M is written on paper of the Union College, Schenectady, NY. In the top right-hand corner off
the first page of M, Strauss initially wrote in blue ink “Dec., 1, 1939—“. Later he crossed this out and
wrote, in pencil, “Dec., 1939: I.”
3Leo Strauss Papers (Box 23, Folder 8).
4Leo Strauss Papers (Box 14, Folder 12). In the top right-hand corner of the first page of CC,,
“Carbon not corrected” was written in a hand other than Strauss’s.
5The last two pages of TS were replaced by two pages written on a different typewriter. These pages
take into account most of the corrections that appear in CC. Also, the text of footnote VI cannot be
found in CC while the superscripted number is in the main text.
6Interpretation 14, no. 1 (1986): 51–59. A more heavily edited version was included in RCPRR 63–71.
272 EDITORIAL NOTE TO APPENDICES F AND G

This edition takes TS, which Strauss corrected more carefully than CC, as the
basis from which it notes deviations from M and CC. It notes as well any hand-
written corrections in the body of the text in the three versions. In the text off
Strauss’s own footnotes, however, I have noted only obviously significant devia-
tions and handwritten corrections in order to preserve readability. Typographical
or reading errors that were later corrected, minor variations in comma placement,,
orthographical inconsistencies in German or French quotations, and discrepancies
due to the fact that Strauss used British spelling while the typist nearly always usedd
American spelling, have been corrected silently. Words underlined, by hand or by
typewriter, have been italicized.
Editorial revisions are kept to a minimum. In a few cases, mainly in Strauss’s
footnotes, I have corrected errors made by the typist and overlooked by Strauss on
the basis of the manuscript.7 Only in a handful of cases did I find emendations off
Strauss’s text indispensable.8
All footnotes with Roman numerals are Strauss’s; those with Arabic numerals
as well as all additions in square brackets are my own. The great number of devia-
tions and handwritten corrections in all three of the versions compelled resorting
to abbreviations in the apparatus. Therefore, I have adopted a simple system: 1) The
deviations from TS have been indicated by a reference to the source, followed by
the deviation. For example, the note “M: has” to the word “had” in the body of the
text is to be read as “the manuscript reads ‘has’ (whereas the typescript and the car-
bon copy read ‘had’).” 2) The handwritten corrections have been indicated with a
reference to the source, followed by the original word, an arrow and the correction..
For example, the note “M: popular → public” to the word “public” in the body
of the text is to be read as “in the manuscript, ‘popular’ was replaced with ‘public’’
(which was adopted by the typescript).” All corrections in M are Strauss’s but it is
not always clear whether the corrections in TS and CC were made by Strauss or by
someone else. Therefore, I have only identified those handwritten corrections that
can be attributed with certainty to Strauss.
The two plans for “Exoteric Teaching” (Supplements 1 and 2, below)9 reveal
that Strauss initially intended to write a much longer essay. They map out additions
to the existing text as well as details of the second part of the essay, which Strauss
never carried out. Even though the plans cannot be dated precisely, it seems fairly

7For example, TS has in footnote I “120ff.” while M has “120f.” In similar cases, I have restored the
reading of M in fn II: edd. → eds. / fn III: Lessing’ → Lessing’s / fn VII: and “The ordinary distinc-
tion between offensive and defensive wars is quite empty.” (Loc. cit., §§ 60 and 276) → (Philosophischee
Sittenlehre, § 60) and “The ordinary distinction between offensive and defensive wars is quite empty.””
(Loc. cit., § 276) / fn IX: 34ff. → 34f. / fn XIV: 153ff. → 153f. / fn XXXI: 44ff. → 44f. / fn XXXIII::
bonds → bands / proportionately → proportionally.
8Cf. the editorial comments in fn 280n53, XXIII, XXIV, XXVI, and fn 283n82. In accordance with
the original text, a few minor mistakes in the quotations in fn IV, XXXIII and XXXVII were corrected..
Cf. Adam Ferguson, An Essay on the History of Civil Society, edited by Fania Oz-Salzberger (Cambridge,,
UK: Cambridge University Press, 1995), 24f., 29 and 177.
9Leo Strauss Papers (Box 17, Folder 2). The first paragraph of an earlier draft of the essay, also enti-
tled “Exoteric Teaching” can be found on the back of the later plan. An even earlier version, entitled
“Exoteric
Exoteric Literature,”
Literature, and a third plan, entitled “Exoteric
Exoteric teaching,”
teaching, can be found in the same folder.
EDITORIAL NOTE TO APPENDICES F AND G 273

clear that the second one was written at a later time since it is both more detailed
and closer to the part of “Exoteric Teaching” that was finished.
A third supplementary document (appendix G) consists of a closely related set off
notes. On December 6, 1939, while working on “Exoteric Teaching,” Strauss gave a
lecture titled “Persecution and the Art of Writing” at Union College, Schenectady,,
NY.10 Among his papers, Strauss left a five-page manuscript (of which four pages
have survived) with elaborate notes for a lecture of the same title.11 These notes
seem to be an aide-mémoiree and some of the writing has been crossed out.12 Like
the two plans, the lecture notes are of an essentially private character and should
therefore be read in the light of Strauss’s published works. In the hope of makingg
this task easier for the reader, I have annotated these documents extensively but by
no means exhaustively, while I have kept the editorial notes for “Exoteric Teaching””
to a minimum.

* * *

The bulk of this work was made possible by a generous grant provided by the
German National Merit Foundation for the academic year 2010/2011. I am very
grateful to Robert B. Pippin for inviting me to spend that year with the Committee
on Social Thought at the University of Chicago. Nathan Tarcov, Leo Strauss’s lit-
erary executor, has kindly given his permission for the publication and has sup-
ported me constantly. I am indebted to Wiebke Meier, Svetozar Minkov, and Devin
Stauffer for help in deciphering some of Strauss’s hieroglyphs, to Stuart D. Warner,,
William Wood, and especially to Jeremy Bell for friendly critique, as well as to
Martin D.Yaffe and Richard S. Ruderman for giving me the opportunity to pub-
lish such a lengthy critical edition.To Heinrich Meier and Ralph Lerner, who have
encouraged me from the very beginning with their insights and advice, I am deeply
grateful.

10Cf. Leo Strauss to Jacob Klein, November 28, 1939, in GS–3 586. As Heinrich Meier has kindly
informed me, Strauss noted in the margin of his own copy of the article “Persecution and the Art off
Writing” that he gave lectures on the topic in October and December 1939 as well as in February,,
March, and April 1940.
11Leo Strauss Papers (Box 17, Folder 2).
12Deletions by LS are indicated by chevrons (“< ...>”) and additions by the editor in square brackets
(“[ ...]”).
APP D F
APPENDIX

LEO STRAUSS: EXOTERIC TEACHING (1939)

Edited by Hannes Kerber

Le partage du brave homme est d’expliquer librement ses pensées. Celui qui n’ose regarder fixément les deux
pôles de la vie humaine, la religion et le gouvernement, n’est qu’un lâche.

—Voltaire.1

he distinction between exoteric (or public2) and esoteric (or secret) teaching is
T not at present considered to be of any significance for the understanding of the
thought of the past: the leading encyclopedia of classical antiquity does not contain
any article, however brief, on exotericc or esoteric. Since a considerable number off
ancient writers had3 not a little to say about the distinction in question, the silence
of the leading encyclopedia cannot possibly be due to the silence of the sources; it
must be due to the influence of modern philosophy on classical scholarship; it is
that influence which prevents scholars from attaching significance to numerous4,
if not necessarily correct, statements of ancient writers. For while it is for classi-
cal scholars to decide whether and where5 the distinction between exoteric and
esoteric teaching occurs in the sources, it is for philosophers to decide whetherr
that distinction is significant in itself. And modern philosophy is not favorable to
an affirmative answer to this philosophic question. The classical scholar Zeller may
have believed to have cogent reasons for rejecting the view that Aristotle “design-
edly chose for (his scientific publications) a style obscure and unintelligible to the
lay mind”; but it must be doubted whether these reasons would have appeared to

1“It is the lot of the brave [or decent] man to explain his thoughts freely. He who does not dare to
look directly at the two poles of human life, religion and government, is only a coward.” Quoted with
some alterations from Voltaire, L’A, B, C, ou Dialogues entre A, B et C (dixième entretien, sur la religion),,
in Dialogues et Anecdotes Philosophiques, edited by Raymond Naves (Paris: Garnier, 1939), 304.
2M: popular → public
3M: has
4M: important → numerous
5TS/CC: when → where
276 APPENDIX F

him equally cogent, if he had not been assured by the philosopher Zeller that the
rejected view “attributes to the philosopher a very childish sort of mystification,,
wholly destitute of any reasonable motive.”I
As late as the last third of the 18th century, the view that all the ancient phi-
losophers had distinguished between their exoteric and their esoteric teaching was
still maintained, and its essential implications were fully understood at least by one
man. Gotthold Ephraim Lessing united in himself in a unique way the so divergent
qualities of the philosopher and of the scholar. He discussed the question of exo-
tericism clearly and fully in three 6little6 writings of his: in “Leibniz von den ewigen
Strafen” (1773), in “Des Andreas Wissowatius’ Einwürfe wider die Dreieinigkeit””
(1773) and in “Ernst und Falk” (1777 and 1780).II He discussed it as clearly and as
fully as could be done by someone who still accepted exotericism not merely as a
strange fact of the past, but rather as an intelligible necessity for all times and, there-
fore, as a principle guiding his own literary activity.III In short, Lessing was the last
writer who revealed, while hiding them, the reasons compelling wise men to hide
the truth: he wrote between the lines about the art of writing between the lines.
In “Ernst und Falk,” a character called7 Falk, who expresses himself somewhat
evasively and sometimes even enigmatically, tries to show that every political con-
stitution, and even the best political constitution8, is necessarily imperfect: the nec-
essary imperfection of all political life makes necessary the existence of what he
calls free-masonry, and he does not hesitate to assert that free-masonry, which is
necessary, was always in existence and will always be. Falk himself is a free-mason, iff
a heretical9 free-mason, and in order to be a free-mason, a man must know truths
which ought better to be concealed.IV What10 is then the concealed reason of his
view that all political life is necessarily11 imperfect?V The intention of the good

I Aristotle and the Earlier Peripatetics (translated by Costelloe and Muirhead), London 1897, I 120f.
6Added by LS in M between the lines or in the margins.
IISee Lessing, Werke, eds. Petersen and von Olshausen,VI 21–60 (“Ernst und Falk”) and XXI 138–189
(the two other treatises mentioned above). Compare also Lessing’s “Über eine zeitige Aufgabe” (XXIV
146–153).
IIILessing’s exotericism was recognized to a certain extent by Gottfried Fittbogen, Die Religion
Lessings, Leipzig 1923, 60ff. and 79ff. Fittbogen does not however see the most important implications
of his valuable remarks, since his interpretation of Lessing is based on a Kantian or post-Kantian view
of the meaning of philosophy.
7TS: character, called → character called [correction not in CC]
8M: and even the absolutely best political constitution [This part of the sentence was not transcribed
by the typist. LS reinserted it in TS and CC. However, he dropped the word “absolutely.”]
9M: heretic → heretical
IV“Falk. Weißt du, Freund, daß du schon ein halber Freimäurer bist? ...denn du erkennst ja schon
Wahrheiten, die man besser verschweigt. Ernst. Aber sagen könnte. Falk. Der Weise kann nicht sagen, was
er besser verschweigt.” Second Dialogue, loc. cit., p. 31. [“Falk: Do you know, friend, that you are already
half a free-mason? ...because you already realize truths which are better to be concealed. Ernst: But
which couldd be said. Falk: The wise man cannott say what he would do better to conceal.”]
10TS: Which → What [correction not in CC]
11M: necessarily is
VIn the 3rd dialogue (p. 40), it is explicitly stated that only such shortcomings of even the best political
constitution have been explicitly mentioned as are evident even to the most shortsighted eye.This implies
that there are other shortcomings g of political
p life as such which are not evident to “shortsighted
g eyes.”
y
APPENDIX F 277

works of the free-masons is to make good works superfluous,VI and free-masonry


came into beingVII when12 someone who originally had planned a scientific soci-
ety which should make the speculative truths useful for practical and political life
conceived13 of a “society which should raise itself from the practice of civil life to
speculation.”VIII The concealed reasons14 of the imperfection of political life as such
are the facts15 that all practical or political life is essentially inferior to contemplative
life, or that all works, and therefore also all good works, are “superfluous” as far as
the level of theoretical life, which is self-sufficient, is reached, and that the require-
ments of the lower are bound from time to time to conflict with, and to supersede
in practice, the requirements of the higher. Consideration of that conflict is the ulti-
mate reason why the “free-masons” (i.e. the wise or the men of contemplation16)
must conceal certain fundamental truths. It may be added that Lessing points out in
“Ernst und Falk” that the variety of religions is due to the variety of political con-
stitutionsIX: the religious problem (i.e. the problem of historical, positive religion) is
considered by him as part and parcel of the political problem.17
In “Leibniz von den ewigen Strafen” and in “Wissowatius,” Lessing applies
these views to an explanation of Leibniz’ attitude toward religion. The explicit
purpose of these two little treatises is to discuss “the motives and reasons” which
had induced Leibniz to defend certain orthodox beliefs (the belief in eternal
damnation and the belief in trinity).X While defending Leibniz’ defense of the
belief in eternal damnation, Lessing states that Leibniz’ peculiar way of assenting
to received opinions is identical18 with “what all the ancient philosophers used
to do in their exoteric speech.”XI By making that statement, he not only asserts
that all the ancient philosophers made use of two manners of teaching, of an
exoteric manner and an esoteric manner; he also bids us19 trace back18 all essential
features of Leibniz’ exotericism to the exotericism of the ancients. What, then,,

VI1st dialogue (at the end) and 3rd dialogue (p. 39).
VIIThe contradiction between the statement made at the beginning that free-masonry is always in
existence, and the statement made toward the end that free-masonry came into being at the begin-
ning of the eighteenth century enables us to see that free-masonry is an ambiguous term. [In M the
sentence is concluded: “ ...18th century shall enable us to see that ‘free-masonry’ is an ambiguous term,,
and that the secret meaning of that term indicates what ought to be called in unmetaphoric language—
philosophy.” In TS/CC the conclusion reads differently (“ ...ambiguous term, and that the secret mean-
ing of the term is ‘philosophy.’”) and was crossed out by LS.]
12TS: being, when → being when [correction not in CC]
13TS: life, conceived → life conceived [correction not in CC]
VIII5th dialogue (toward the end).
14M: reason → reasons
15M: is the fact → are the facts
16M: speculation → contemplation
IX2nd dialogue (p. 34f.).
17M: (i.e. the problem of positive religion) is a part of the political problem → (i.e. the problem off
historical, positive religion) is considered by him as part and parcel of the political problem
XWerke, XXI 143 and 181.
18Italics added in TS (correction not in CC).
XILoc. cit., 147.
19TS: bids us to → bids us [correction not in CC]
20TS/CC: Which are, then, → What, then, are
278 APPENDIX F

are20 the essential features of Leibniz’ exotericism? Or, in other words, what21 are
the motives and reasons which guided Leibniz in his defense of the orthodox orr
received opinion?XII Lessing’s first answer to this question is that Leibniz’ peculiarr
way of assenting to received opinions is identical with “what all the ancient phi-
losophers used to do in their exoteric speech. He observed a sort of prudence forr
which, it is true, our most recent philosophers have become much too wise.”XIII
The distinction between exoteric and esoteric speech has then so little to do with
“mysticism” of any sort that it is an outcome of prudence. Somewhat later on
Lessing indicates the difference between the esoteric reason enabling22 Leibniz
6to6 defend23 the orthodox doctrine of eternal damnation, and the exoteric reason

expressed in24 his defense25 of that doctrine.XIV That exoteric reason, he asserts, is
based on the mere possibility of eternally increasing wickedness of moral beings26.
And then he goes on to say: “It is true, humanity shudders at this conception
although it concerns the mere possibility. I should27 not however for that reason
raise the question: why frighten with a mere possibility? For I should28 have to
expect this counterquestion: why not frighten with it, since it can only be frightful
to him who has never been earnest about the betterment of himself.” This implies
that a philosopher who makes an exoteric statement, asserts, not a fact, but what
Lessing chooses to call “a mere possibility”: he does not, strictly speaking, believe
in the truth of that statement (e.g. of the statement that there is29 such a thingg
as eternally increasing wickedness of human beings which would justify eter-
nally increasing punishments).This is indicated by Lessing in the following remark
introducing a quotation from the final part of Plato’s Gorgias: “Socrates himselff
believed in such eternal punishments quite seriously, he believed in them at least
to the extent30 that he considered it expedient31 to teach such punishments in
terms which do not in any way arouse suspicion and which are most explicit.”XV
Before proceeding any further, I must summarize Lessing’s view of exoteric
teaching. To avoid the danger of arbitrary interpretation, I shall omit all elements
of that view which are not noticed32 at a first glance even by the most superficial

21TS/CC: which → what


XIICf. loc. cit., 146.
XIIILoc. cit., 147. Cf. Plato, Theaetetus 180c7–d5, with Protagoras 316c5–317c5 and 343b4–5.
22M: underlying → enabling
23M: defence of → defend
24M: underlying → expressed in
25M: explicit defence → defence
XIVLoc. cit., 153f.
26M: a moral being → moral beings
27M: shall
28M: shall → should
29In M “is” is italicized.
30M: at least so far
31M: expedient (zuträglich)
XVLoc. cit., 160. Cf. also the remarks about “believing” on pp. 184, 187, and 189. [In M this footnote
begins: “Or: ‘in terms which are least open to suspicion and most explicit’ (‘mit den unverdächtigsten
und ausdrücklichsten Worten’).”—For a different translation of the same passage, cf. PAW
W 183.]
32M: evident → noticed
APPENDIX F 279

reader of Lessing, although the obvious33 part of his view, if taken by itself, is some-
what enigmatic. 1) Lessing asserts that all the ancient philosophers and LeibnizXVI
made use of exoteric presentation of the truth, as distinguished from its esoteric
presentation. 2) The exoteric presentation of the truth makes34 use of statements
which are considered by the philosopher himself statements, not of facts, but off
mere possibilities. 3) Exoteric statements (i.e. such statements as would not 6and
could not6 occur within the esoteric teaching) are made by the philosopher forr
reasons of prudence or expediency. 4) Some35 exoteric statements are addressed to
morally inferior people who ought to be frightened by such statements. 5) There
are certain truths which must36 be concealed. 6) Even the best political constitution
is bound to be imperfect. 7) Theoretical life is superior to practical or political life..
The impression created by this summary, that there is a close connection between
exotericism and a peculiar attitude toward political and practical life, is not mislead-
ing:“free-masonry,” which 6as such6 knows of secret truths, owes its existence to the
necessary imperfection of all practical or political life.
Some readers might be inclined to dismiss Lessing’s whole teaching at once,,
since it seems to be based on the obviously erroneous, or37 merely traditional,XVII
assumption that alll the ancient philosophers have38 made use of exoteric speeches..
To warn such readers, one must point out that the incriminated sentence permits
of a wholly unobjectionable interpretation: Lessing implicitly denies that writers on
philosophical39 topics who reject exotericism, deserve the name of philosophers.XVIII
For he knew the passages in Plato in which it is indicated40 that it was41 the sophists
who refused to conceal the truth.
After Lessing, who died in the year in which Kant published his Critique off
Pure Reason, the question of exotericism seems to have42 been lost sight of almost
completely, at least among scholars and philosophers as distinguished from novelists..
When Schleiermacher introduced that style of Platonic studies, in which classical
scholarship is still engaged, and which is based on the identification of the natural
order of Platonic dialogues with the sequence of their elaboration, he still had to
discuss in detail the view that there are two kinds of Platonic teaching, an exoteric

33M: evident or obvious → obvious


XVIIn a private conversation, published only after his death, Lessing said to F. H. Jacobi about Leibniz::
“Es ist bei dem größten Scharfsinn oft sehr schwer, seine eigentliche Meinung zu entdecken.” Werke,,
XXIV 173. [“With the greatest ingenuity it is often very difficult to discover his real opinion.”]
34M: may make → makes
35M: At least some → Some
36M: ought better → must
37M: and → or
XVIICompare Clemens Alexandrinus, Stromata,V [ch. IX] 58 (365 Stählin).
38M: had → have
39M: philosophic
XVIIICf. for a similar example of Lessing’s way of expressing himself his Briefe antiquarischen Inhalts VII
(Werke, XVII 97ff.).
40M: made clear → indicated
41M: were
42M: has → seems to have
280 APPENDIX F

kind and an esoteric one. In doing this, he makes five or six extremely43 important
and true remarks about Plato’s literary devices,XIX remarks the subtlety of which
has, to my knowledge, never been surpassed or even rivaled since. Yet he failed to
see the crucial question. He asserts that there is only one Platonic teaching18—the
teaching presented in the dialogues—although there is, so to speak, an infinite
number of levels44 of the understanding of that teaching45: it is the same teach-
ing which the beginner understands inadequately, and which only the perfectly
trained student of Plato understands adequately46. But is then the teaching which
the beginner actually understands47 identical with the teaching18 which the perfectly
trained student actually understands? The distinction between Plato’s exoteric and
esoteric teaching had sometimes been48 traced back to Plato’s opposition to “poly-
theism and popular religion” and to the necessity 6in which he found himselff6 off
hiding that opposition; Schleiermacher believes he has49 refuted this view by assert-
ing that “Plato’s principles on that topic are clear enough to read in his writings,,
so that one can scarcely believe that his pupils might have needed still more infor-
mation about them.”XX Yet, “polytheism and popular religion” is an ambiguous
expression:50 if Schleiermacher had used the less ambiguous expression51 “belieff
in the existence of the gods worshipped by the city of Athens,” he could not have
said that Plato’s opposition to that belief is clearly expressed in his writings.52 As a
matter of fact, in his introduction to his translation of Plato’s Apology of Socrates, he
considers it a weak point of that writing that Plato has not made more energetic
use of the argument taken from Socrates’s service to Apollo, for refuting the charge
that Socrates did not believe in “the53 old gods.”XXI If Plato’s Socrates believed54
in “the old gods,” is not Plato himself likely to have believed in them as well? Andd
how can one then say that Plato’s opposition to “polytheism and popular religion””

43M: very → extremely


XIXF. Schleiermacher, Platons Werke, I 1, Berlin 1804, 20 (3. Auflage, Berlin 1855, 16). [The references
to the third edition in footnotes XIX, XX, and XXI were added by LS between the lines in CC, but not
in TS. In footnote XXII, LS made no reference to the third edition.]
44M: degrees [In TS and CC, LS inserted “degrees” between the lines without crossing out “levels.”]
45TS/CC: understanding that teaching → the understanding of that teaching
46M: perfectly → adequately
47TS: understands, → understands [correction not in CC]
48M: be
49M: to have
XXLoc. cit., 14 (3rd ed. Berlin 1855, 12).
50In M, one and a half sentences which originally preceded this sentence were crossed out: “But he
forgets the fact that Plato has not written his dialogues for his pupils only, but rather as a possession for
all times, or that not all readers of Plato are pupils of Plato.Yet this refutation is based on.”
51Omitted in TS, but later inserted by LS. (Correction not in CC.)
52In M a sentence originally following this sentence was crossed out: “And is not that belief some sort
of ‘polytheism and popular religion’?”
53In TS and CC, but not in M, the opening quotation marks were placed in front of “a weak point ...”
because the typist wrongly took the opening quotation marks in the sentence that was crossed out (cf.
n52) to indicate the beginning of the quotation in this sentence.
XXIPlatons Werke, I 2, Berlin 1805, 185 (3. Aufl., Berlin 1855, 128).
54TS/CC: has believed → believed
APPENDIX F 281

as such is clearly expressed in his writings? Schleiermacher’s strongest argument


against the distinction of two teachings of Plato appears to be his assertion that
Plato’s 6real6 investigations are hidden, not absolutely, but only from the inatten-
tive readers, or that attention55 is the only prerequisite for a full understanding off
his real investigations as distinguished from those investigations which are merely
the “skin” of the former.XXII But did any man in his senses ever assert that Plato
wished to hide his secret teaching from all readers or from all men? Did any man
whose judgment can claim to carry any weight in this matter ever understand by
Plato’s esoteric56 teaching anything other than that teaching of his dialogues which
escapes the inattentive readers only? The only possible difference of opinion con-
cerns exclusively the meaning of the distinction between inattentive and attentive
readers: does a continuous way57 lead from the extremely inattentive reader to the
extremely attentive reader, or is the way between the two extremes interrupted
by a chasm? Schleiermacher tacitly assumes that the way from the beginning to
the end is continuous, whereas, according to Plato, philosophy presupposes a real
conversion58,XXIII i.e.18 a total break with the attitude of the beginner: the beginnerr
is a man who has not yet for one moment left the cave, and who has even never59
turned his eyes away from the shadows of man-made things60 toward the exit off
the cave, whereas the philosopher is the man who has left the cave and who (iff
he is not compelled to do otherwise) lives61 outside of the cave, on “the island off
the blessed.”62 The difference between the beginner and the philosopher (for the
perfectly trained student of Plato is no one else but the genuine philosopher) is a
difference not of degree, but of kind. Now, it is well-known that, according to Plato,,
virtue is knowledge or science; therefore, the beginner is inferior to the perfectly
trained student of Plato not only intellectually, but also morally. That is to say, the

55M: the attention → attention


XXII
“Das geheime ...(ist) nur beziehungsweise so ...” I 1, 12.—“ ...die eigentliche Untersuchungg
wird mit einer anderen, nicht wie mit einem Schleier, sondern wie mit einer angewachsenen Haut
überkleidet, welche dem Unaufmerksamen, aber auch nur diesem, dasjenige verdeckt, was eigentlich
soll beobachtet oder gefunden werden, dem Aufmerksamen aber nur noch den Sinn für den inneren
Zusammenhang schärft und läutert.” Loc. cit., 20. (The italics are mine.) [“The secretive ...(is) only rela-
tively so ...”; “ ...the actual investigation is covered with another, not as if with a veil but as if with a
grown-on skin, which conceals from the inattentive [reader] but only from him that which actually
ought to be observed or found, but which for the attentive [reader] sharpens and chastens the sense forr
the internal coherence.”]
56M: secret or esoteric → esoteric
57In TS, “path” was written (not by LS) between the lines, and a question mark in the margin.
58M: “conversion” → conversion
XXIIIRepublicc 518c–e and 521e [recte: 521c]. Cf. also Phaedo 69a–c. [In TS and CC, LS added “and
619c–d” to the references to the Republic. However, in M he inserted the reference in footnote XXIV. V.
Cf. FP 361n11 and CM M 27n34.]
59M: not even → even never
60In M, Strauss added “of men and man-made things” in the margin. The typist dropped “men and.”
61M: and prefers the life → and who lives → and who (if he is not compelled to do otherwise) lives
62M: cave, the life on “the island of the blessed” to the life in the cave → cave, on “the island of the
blessed”
282 APPENDIX F

morality63 of the beginners has a basis essentially different from the basis on which
the morality of the philosopher rests: their virtue is not64 genuine virtue, but vulgarr
or political virtue only,65 a virtue based not on insight66, but on customs or laws.XXIV
We may say, the morality of the beginners is the morality of the “auxiliaries” of the
Republic, but not yet the morality of the “guardians.” Now, the “auxiliaries,”67 the
best among whom are the beginners, must believe68 “noble lies,”XXV i.e.69 statements
which, while being useful for the political community, are nevertheless lies. Andd
there is a difference not of degree but of kind70 between truth and lie or untruth..
And what holds true of the difference between truth and lies71 holds equally true
of the difference between esoteric and exoteric teaching; for Plato’s exoteric teach-
ing is identical with his “noble lies.” This connection of considerations, which is
more or less familiar to every reader of Plato, if not duly emphasized by all students
of Plato, is not even mentioned72 by Schleiermacher in his refutation of the 6view
that there is a6 distinction between Plato’s exoteric and esoteric teaching. Norr
does he,73 in that context, 6as much as allude6 to Lessing’s dialogues (“Ernst und
Falk” and Lessing’s conversation with F. H. Jacobi) which probably come closer74
to the spirit75 of the Platonic dialogues and their technique than any other mod-
ern work in the German language76. Therefore Schleiermacher’s refutation77 of the
view in question is not convincing. A comparison of his Philosophic Ethics with the
Nicomachean Ethics would bring to light the reasonXXVI why he failed to pay any
attention to the difference between the morality of the beginner and the morality
of the philosopher, i.e.18 to the difference which is at the bottom of the difference
between exoteric and esoteric teaching.
I return to Lessing. How was Lessing led to notice,XXVII and to understand, the
information about the fact 6that6 “all the ancient philosophers” had distinguished
63M: basis of morality → morality
64M: is not, and cannot be, → is not
65Originally, footnote XXIV was placed after “only.” LS made the correction in M in pencil.
66M: philosophy → insight
XXIV
Republicc 430c3–5, 619c–d and Phaedo 82a10–b8. [LS inserted in M, but not in TS/CC, “Rep..
619c–d.” Cf. the comment to footnote XXIII.]
67M: “auxiliaries”
68TS/CC: believe in → believe
XXVRepublicc 414b4ff. Cf. Laws 663d6ff.
69TS/CC: i.e. in → i.e. [italics added in TS, correction not in CC]
70TS/CC: is no difference of degree, but of kind, → is a difference not of degree but of kind
71M: lie,
72M: as much as alluded to → even mentioned
73M: he pay any attention, → he,
74TS/CC: come probably nearer → probably come closer
75M: spirit of the technique → spirit
76TS/CC: in the German language does → in the German language [The five words were added
between the lines in M.]
77M: refutation is not convincing → refutation
XXVI That reason can be discovered by an analysis of the following statements, e.g.: “Knowledge off
the essence of reason is ethics” (Philosophische Sittenlehre, § 60) and “The ordinary distinction between
offensive and defensive wars is quite empty.” (Loc. cit., § 276).
XXVII Cf. the remarks of the young Lessing on the relevant passage in Gellius (XX 5) in the tenth
Literaturbrieff (Werke
Werke,, IV 38).
APPENDIX F 283

between their exoteric and their esoteric teaching? If I am not mistaken, he redis-
covered the bearing of that distinction by his own exertion after having78 under-
gone his conversion79, i.e.18 after having had80 the experience of what philosophy
is and what81 sacrifices it requires. For it is that experience which leads in a straight
way to the distinction between the two groups of men, the philosophic men and
the unphilosophic men, and therewith to the distinction between the two ways off
presenting the truth. In a famous letter to a friend,XXVIII he expresses his fear that
“by throwing away certain prejudices, I have thrown away a little too much that
I shall have to fetch back82.”XXIX That passage has sometimes been understood to
indicate that Lessing was about to return from the intransigent rationalism of his
earlier period toward a more positive view of the Bible and the Biblical tradition..
There is ample evidence to show that this interpretation is wrong.XXX The con-
text of the passage makes it clear that the things which Lessing had “thrown away””
before and which, he feels, he ought to “fetch back” were truths which he descriedd
“from afar” in a book by Ferguson, as he believed on the basis of what he had seen
in the table of contents of that book. He also descried “from afar” in Ferguson’s
book “truths in the continual contradiction of which we happen to live83 and we
have to go on living continually in the interest of our quietude.” There may very
well be a connection between the two kinds of truth84: the truths which Lessing85
had thrown away formerly86 may have been truths contradictory to the truths87
generally accepted by the philosophy of enlightenment88 and also accepted by

78CC: having had


79M: “conversion” → conversion
80TS: i.e. after having made → i.e. after having had [correction not in CC]
81TS/CC: which → what
XXVIII
To Moses Mendelssohn, of January 9, 1771.
82LS inserted between the lines in M, without crossing out “to fetch back,” the alternative translation
“get back again” (which was adopted by TS/CC). In the margin of the manuscript he put an exclama-
tion mark. My decision to retain the first translation is supported by LS’s use of this translation a few
w
lines later.
XXIX Another statement about the crisis which Lessing underwent when he was about forty, occurs in
the Briefe antiquarischen Inhalts, LIV (Werke, XVII 250).
XXXSee e.g. von Olshausen in his introduction to Werke, XXIV 41ff.—Compare also Jacobi’s letter to
Hamann of December 30th, 1784:“Als (Lessings) Erziehung des Menschengeschlechts ...von einigen fürr
eine nicht unchristliche Schrift, beinahe für eine Palinodie angesehen wurde, stieg sein Ärger über die
Albernheit des Volkes bis zum Ergrimmen.” (F. H. Jacobi, Werke, I 398). [“When (Lessing’s) Education off
the Human Racee ...was considered by some as a not un-Christian writing, [but] almost as a palinode, his
anger about the absurdity of the people grew to the point of fury.”]
83brM: actually live → happen to live [In M, a few sentences of this paragraph (“The context of the
passage ...he ought to ‘fetch back’ are ‘truth in the continual contradiction of which we have to live’;;
such truths he descried ‘from afar’ in a book by Ferguson as he believed on the basis of what he hadd
seen in the table of content of that work.”) were crossed out by LS. He inserted a red cross in crayon in
the margin. The passage that is reproduced in TS and which differs slightly from M is to be found on a
brown slip of paper (brM), which also has a red cross in crayon in the margin.]
84brM: truths
85brM: L.
86TS: formerly, → formerly [correction not in CC]
87brM: those → the truths
88brM: his enlightened contemporaries → the philosophy of enlightenment
284 APPENDIX F

Lessing throughout his life.89 At any rate,90 two years later he openly rebuked the
more recent philosophers who had evaded the contradiction between wisdom andd
prudence by becoming much too wise to submit to the rule of prudence which
had been observed by Leibniz and all the ancient philosophers. External evidence
is91 in favor of the view that the book referred to by Lessing92 is Ferguson’s Essay on
the History of Civil Society.XXXIThe “truths in the continual contradiction of which
we have to live,” which had been discussed by Ferguson and which are indicated to
a certain extent in the table of contents of his Essay,XXXII concerned the ambigu-
ous character of civilization, i.e.18 the theme of the two famous early writings off
Rousseau, which Lessing, as93 he perhaps felt, had not considered in his youth
carefully enough94.XXXIII Lessing expressed his view of the ambiguous character off
civilization 6some years later6 in these more precise terms: even the absolutely best
civil constitution is necessarily imperfect. It seems then to have been95 the political
problem96 which gave Lessing’s thought a decisive turn away from the philosophy
of enlightenment indeed, yet not toward romanticism of any sort—toward what
is called a deeper, historical view of government and religion97—, but toward an
older type of philosophy. How near he apparently came to certain romantic views
on his way from the philosophy of enlightenment to that older type of philosophy98
we may learn from what F. H. Jacobi tells us in an essay of his which is devotedd
to the explanation of a political remark made by Lessing. According to Jacobi,,
Lessingg once said99 that the arguments
g against
g p 100 despotism are either no
Papal

89LS inserted “and accepted also by Lessing throughout his life” in the margin of brM. However,
“also” was dropped by the typist.
90brM: At any rate, he censured → At any rate,
91M: decides → is
92M: in question → referred to by L.
XXXI Cf. von Olshausen, loc. cit., 44f., who however rejects this conclusion on the basis of “internal
reasons.”
XXXII Cf. e.g. the following headings of sections: “Of the separation of arts and professions” [part IV,
sect. 1] and “Of the corruption incident to polished nations” [part VI, sect. 3].
93Inserted in TS/CC.
94M: dismissed in his youth somewhat too rashly → not considered in his youth carefully enough
XXXIII The influence of Ferguson’s mitigated Rousseauism on Lessing can be seen from a comparison
of the following quotations with what Lessing says in “Ernst und Falk” on the obvious reasons of the
necessary imperfection of all civil societies. Ferguson says in Part I, section 3 and 4: “The mighty engine
which we suppose to have formed society, only tends to set its members at variance, or to continue their
intercourse after the bands of affection are broken.” “The titles of fellow-citizen and countryman, unop-
posed to those of alien and foreigner, to which they refer, would fall into disuse, and lose their meaning.”
“ ...it is vain to expect that we can give to the multitude of a people a sense of union among themselves,
without admitting hostility to those who oppose them.” See also Part IV, section 2: “ ...if the lot of a
slave among the ancients was really more wretched than that of the indigent labourer and the mechanic
among the moderns, it may be doubted, whether the superior orders, who are in possession of consider-
ation and honours, do not proportionally fail in the dignity which befits their condition.”
95M: At any rate, it was → It seems then to have been
96M: question → problem
97M: society and religion → government and religion
98TS: philosophy, → philosophy [correction not in CC]
99M: had remarked → once said
100M: papal
APPENDIX F 285

arguments at all, or else they are two or three times as valid against the despotism
of princes.XXXIV Could Lessing have held101 the view that ecclesiastical despotism
is two or three times better than secular despotism?102 Jacobi elsewhere says in his
own name103 but certainly in the spirit of Lessing, that that despotism which is
based “exclusively”104 on superstition, is less bad than secular despotism.XXXV Now,,
secular despotism could easily be allied105 with the philosophy of enlightenment,,
and therewith with the rejection of exotericism strictly speaking, as is shown above
all by the teaching of the classic of enlightened despotism: the teaching of Hobbes..
But “despotism based exclusively on superstition,” i.e.18 not at all on force, cannot
be maintained if the non-superstitious minority does not voluntarily refrain from m
6openly6 exposing and refuting the “superstitious” beliefs106. Lessing had then not

to wait for the experience of Robespierre’s despotism to realize the relative truth off
what the romantics asserted against the principles of J.-J.107 Rousseau who seems to
have108 believed in a political solution of the problem of civilization: Lessing real-
ized that 6relative6 truth one generation earlier109, and he rejected it in favor of the
way leading to absolute truth, or of philosophy. The experience which he had110
in that moment enabled him to understand the meaning of Leibniz’111 “prudence””
in a manner infinitely more adequate than the enlightened Leibnizians among his
contemporaries did and could do. Leibniz is then that link in the chain of the tradi-
tion of exotericism which is nearest to Lessing. Leibniz, however, was not the only
17th century thinker who was initiated. Not to mention the prudent Descartes,112
even so bold a writer as Spinoza had admitted the necessity of “pia dogmata, hoc
est, talia quae animum ad obedientiam movent”113 as distinguished from “vera
XXXIV Jacobi, Werke, II 334 (“Etwas das Lessing gesagt hat”). Jacobi quotes in that article Ferguson’s
Essay extensively. [Cf. Jacobi, “Something that Lessing Said,” translated by Dale E. Snow, in What is
Enlightenment? Eighteenth-Century Answers and Twentieth-Century Questions, edited by James Schmidt
(Berkeley: University of California Press, 1991), 198.]
101M: Lessing held then → Should Lessing have held
102M: despotism. → despotism?
103M: name,
104M: exclusively (“einzig und allein”)
XXXV Jacobi, Werke, III 469. Cf. Lessing’s “Gespräch über die Soldaten und Mönche” (Werke, XXIV
159).
105M: reconciled → allied
106M: and attacking the increasing “superstitions” → and attacking “superstitions” → and refuting g
“superstitious” beliefs
107TS/CC: J.J. → J.-J.
108TS: who had → who seems to have
109M: before → earlier
110TS/CC: made → had
111M: Leibniz’s
112Here a footnote was added in M by LS in pencil but not transcribed by the typist:“The early Cartesians
distinguished the ‘exoteric’ Discours de la méthodee from the ‘acroamatic’ Meditationes. Cf. É. Gilson’s com-
mentary on the Discourss (Paris 1930, p. 79). Cf. e.g. Discours de la méthode, sixième partie, in princ.: writing,,
being an action, is subject to religious and political authority, but thought is not.” See Discourse on Method,,
trans. Richard Kennington (Newburyport, MA: Focus Philosophical Library, 2007), 48.
113Tractatus theologico-politicus, cap. 14, § 20 (Bruder). The quotation is taken from a longer sentence::
“Sequitur denique, fidem non tam requirere vera, quam pia dogmata, hoc est, talia, quae animum ad
obedientiam movent; tametsi inter ea plurima sint, quae nec umbram veritatis habent, dummodo tamen
286 APPENDIX F

dogmata.”XXXVI But Lessing did not have114 to rely on any modern or115 medieval
representatives of the116 tradition:117 he was familiar with its sources. It was precisely
his intransigent classicism—his considered view that close study of the classics is the
only way in which a diligent and thinking man can become a philosopherrXXXVII—
which had led him, first, to notice the exotericism of some ancient118 philosophers,
and later on to understand the exotericism of all the ancient philosophers.

is, qui eadem amplectitur, eadem falsa esse ignoret; alias rebellis necessario esset.” (“It follows, finally,
that faith does not require true dogmas so much as pious ones, that is, such as move the spirit toward
obedience—even though among them there may be very many that do not have even a shadow of
truth, yet so long as he who embraces them is ignorant of their being false. Otherwise he would neces-
sarily be rebellious.” Spinoza, Theologico-Political Treatise, trans. Martin D.Yaffe (Newburyport, MA: Focus
Philosophical Library, 2004), 164.) Cf. also SCR R 171; PAW W 180; GS–2 199.
XXXVI Tractatus theologico-politicus, cap. 14, § 20 (Bruder). [In M the footnote continues: “Cf. cap. 15
towards the end. See also Tract. de int. emend. § 17 and 14 and Hobbes, Leviathan, ch. 12 and 46.”—The
reference to the Tractatus de Intellectus Emendationee was inserted in M between the lines.—Cf. PAW
35n17.
114M: Lessing had not
115M: and → or
116M: that
117In M this sentence is preceded by two sentences: “Despite, or because of, that admission Spinoza
rejected Maimonides’ allegorical interpretation of the Bible as ‘harmful, useless and absurd’. Thus, he
cannot be considered a genuine spokesman of the tradition.” Both sentences and the footnote to the
Spinoza quotation (“Tractatus theologico-politicus, cap. 7, § 87 (Bruder).”) were not transcribed by the typ-
ist. In M, LS made two little marks in black ink, one before and one after the last two sentences.
XXXVII He writes in the 71st Literaturbrieff (Werke, IV 197), after having quoted a statement of Leibniz
in praise of criticism and study of the classics: “Gewiß, die Kritik auf dieser Seite betrachtet, und das
Studium der Alten bis zu dieser Bekanntschaft (with Plato, Aristotle, Archimedes and Apollonius) get-
rieben, ist keine Pedanterei, sondern vielmehr das Mittel, wodurch Leibniz der geworden ist, der er war,
und der einzige Weg, durch welchen sich ein fleißiger und denkender Mann ihm nähern kann.” (The
italics are mine.) Ten years later (1769) he says in the Briefe antiquarischen Inhalts XLV (Werke, XVII
218): “Wir sehen mehr als die Alten, und doch dürften vielleicht unsere Augen schlechter sein als die
Augen der Alten; die Alten sahen weniger als wir, aber ihre Augen ...möchten leicht schärfer gewesen
sein als unsere.—Ich fürchte, daß die ganze Vergleichung der Alten und Neuern hierauf hinauslaufen
dürfte.” [“Certainly, the criticism considered from this side, and the study of the ancients pushed to this
familiarity (with Plato, Aristotle, Archimedes and Appolonius) is not pedantry, but in fact the means
whereby Leibniz became who he was and the only way through which a diligent and thinking man can
approach him.” “We see more than the ancients, and yet perhaps our eyes might be worse than the eyes
of the ancients; the ancients saw less [or, fewer things] than we, but their eyes ...may easily have been
sharper than ours.—I am afraid, that the entire comparison of ancients and moderns might come down
to this.”—Cf. “Notes on Philosophy and Revelation,” in Heinrich Meier, Leo Strauss and the Theologico-
Political Problem, translated by M. Brainard (Cambridge, UK: Cambridge University Press, 2006), 178f.]
118M: classical → ancient
APP D F
APPENDIX

SUPPLEMENT 1: EARLY PLAN OF


“EXOTERIC TEACHING”

A
1) To-day the distinction between exoteric and esoteric teaching is wholly
opposed—this opposition is due to the fact that modern philosophy has
destroyed the possibility of understanding—and that class. scholarship has
made a tremendous progress.1
2) At the end of the 18th century, that distinction was still understood: Lessing.
3) Schleiermacher’s criticism: he does not see any more the moral problem m
involved. [Schleiermacher] to whom we are indebted for the deepest insights
into the element of Plato’s writings
3a) Lessing—Leibniz—Hobbes2 (vera—pia dogmata)—Spinoza—RMbM3—
4) Post-Ciceronian authors.
5) Cicero—but he himself is an exoteric writer.4
6) Plato— a) Letters Ep. II, 314a–c.5 Ep.VII, 341a–e, 344d.6
b) Phaedrus, Rep (drama and writings); Timaeus
7) Xenophon Cynegeticus.7

1Cf. the following statements in PAW:W “We are prevented from considering this possibility [i.e., the
possibility of communication of crucial issues between the lines], and still more from considering the
questions connected with it, by some habits produced by, or related to, a comparatively recent progress
in historical research” (PAW
W 26). “Modern historical research [ ...] has counteracted or even destroyed
an earlier tendency to read between the lines of the great writers” (PAW
W 31f.).
2LS first wrote “Spinoza—Hobbes” but crossed out “Spinoza.”
3That is, Rabbi Moshe ben Maimon, the traditional acronym for Maimonides.
4Cf. PAW W 34n16 and 185n85. See “Der Ort der Vorsehungslehre nach der Ansicht Maimunis,” in
GS–2 188n29; PoP 547n29.
5Cf. OPS S 29.
6Cf. PAWW 35n17 and 187n90.
7Cf. PAWW 29n11: “[C]ertain contemporaries of the ‘rhetor’ Xenophon believed that ‘what is beauti-
fully and methodically written, is not beautifully and methodically written’ (Cynegeticus, 13.6).” See also
LS to Jacob Klein, August 7, 1939, in: GS–3 576 and SSTX 502.
288 APPENDIX F: SUPPLEMENT 1

8) Plato’s remarks on Homer and especially on Hesiod.8


9) Hesiod on Muses....9
10) Herakleitus
11) The big exceptions: Epicurus and Sophists.10
cf. Cic. Rep. III 16, 26,11 N.D. I 41.12
cf. Usener.13
a) Epicurus
b) Sophists.

8Cf. LS to Jacob Klein, October 10, 1939, in GS–3 582: “To cut the matter short, what Plato says in
the Theaetetus on the poets of the past, namely that they had disguised philosophy as poetry, can really be
demonstratedd in the case of Hesiod (who occurs in the Republicc somewhere in the middle of an enumera-
tion). I am convinced that it is not different in the case of Homer. One day read the Shield of Achilles! And
the self-identification with Odysseus in the Odyssey and the strange fact that Thersites says the truth.”
9Cf. LS to Jacob Klein, October 10, 1939, in GS–3 581f., esp. 582: “The key to the book are—the
Muses, who are explicitly referred to as the main issue. The Muses have a twofold genealogy: 1) exo-
terically they descend from Zeus and Mnemosyne; 2) esoterically they are offspring of Ocean.You will
immediately guess how this is connected on the basis of the beginning of the Odysseyy as well as the
remarks in the Theaetetus and in the Metaphysics on the origin of Thales’ dictum.” Cf. LAM M 36f.
10Cf., however, PAW W 36.
11See LAM M 90 with 136n21. Cf. Cicero, Republicc 3.16.26: “Ad haec illa dici solent primum ab iis, qui
minime sunt in disserendo mali, qui in ea causa eo plus auctoritatis habent, quia, cum de viro bono queritur,
quem apertum et simplicem volumus esse, non sunt in disputando vafri, non veteratores, non malitiosi; negant enim
sapientem idcirco virum bonum esse, quod eum sua sponte ac per se bonitas et iustitia delectet, sed quod
vacua metu, cura, sollicitudine, periculo vita bonorum virorum sit, contra autem improbis semper aliqui
scrupus in animis haereat, semper iis ante oculos iudicia et supplicia versentur; nullum autem emolu-
mentum esse, nullum iniustitia partum praemium tantum, semper ut timeas, semper ut adesse, semper
ut impendere aliquam poenam putes, damna....” (My italics, H.K.). (“To such arguments as these the
following are usually the replies first given by those who are not unskilful in disputation, and whose
discussions of this subject have all the greater weight because, in the search for the good man, whom we
require to be open and frank, they do not themselves use crafty and rascally tricks of argument—these
men say first of all that a wise man is not good because goodness and justice of or in themselves give
him pleasure, but because the life of a good man is free from fear, anxiety, worry, and danger, while on
the other hand the minds of the wicked are always troubled by one thing or another, and trial and pun-
ishment always stand before their eyes. They add, on the other hand, that no advantage or reward won
by injustice is great enough to offset constant fear, or the ever-present thought that some punishment is
near, or is threatening,.... losses ....” Cicero, De Re Publica. De Legibus, translated by Clinton Walker Keyes
(Cambridge, MA: Harvard University Press, 1961), 205–07.)
12Cf. Cicero, De natura deorum, 1.41.115: “‘At etiam de sanctitate, de pietate adversus deos libros
scripsit Epicurus.’ At quo modo in his loquitur? Ut T. Coruncanium aut P. Scaevolam pontifices maxi-
mos te audire dicas, non eum qui sustulerit omnem funditus religionem nec manibus ut Xerxes sed
rationibus deorum inmortalium templa et aras everterit”. (“‘Yes, but Epicurus actually wrote books
about holiness and piety.’ But what is the language of these books? Such that you think you are listen-
ing to a Coruncanius or a Scaevola, high priests, not to the man who destroyed the very foundations off
religion, and overthrew—not by main force like Xerxes, but by argument—the temples and the altars
of the immortal gods.” Cicero, De Natura Deorum. Academica, translated by H. Rackham (Cambridge,,
MA: Harvard University Press, 1961), 111.) Cf. also 1.44.122f.: “‘At etiam liber est Epicuri de sanctitate.’’
Ludimur ab homine non tam faceto quam ad scribendi licentiam libero.” (“‘Why, but Epicurus (you
tell me) actually wrote a treatise on holiness.’ Epicurus is making fun of us, though he is not so much a
humorist as a loose and careless writer.” Cicero, De Natura Deorum, 119.)
13Hermann Usener published a collection of fragments called Epicurea in 1887. For the Cicero quota-
tions from the previous footnote, cf. Epicurea (Stuttgart: Teubner, 1966), 100.
APPENDIX F: SUPPLEMENT 1 289

B. Explanation: Gellius XX 5:14 res civiles—subtiliores



(natura)

Exoteric philosophy is a Weltanschauung σῴζων political and moral life:


makes man the center of the universe.15
Esoteric philosophy sees man in his insignificance.16
Disproportion of things political and things speculative—cf. Ar. on laws and
τέχναι.17
Quieta movere and Quieta non movere18
Philosophy as essentially unrevolutionary and as interested only in truth.
Lessing had not to wait for the French Revolution in order to separate himself
from the philosophy of Enlightenment.
Lie in the soul—lie in speech.19

14See Aulus Gellius, The Attic Nights, 20.5: “Commentationum suarum artiumque quas discipulis
tradebat Aristoteles philosophus, regis Alexandri magister, duas species habuisse dicitur. Alia erant, quae
nominabat ἐξωτερικά, alia, quae appellabat ἀκροατικά. Ἐξωτερικά dicebantur, quae ad rhetoricas
meditationes facultatemque argutiarum civilumque rerum notitiam conducebant, ἀκροατικά autem m
vocabantur, in quibus philosophia remotior subtiliorquee agitabatur quaeque ad naturaee contemplationes
disceptationesve dialecticas pertinebant” (My italics, H.K.). (“The philosopher Aristotle, the teacher off
king Alexander, is said to have had two forms of the lectures and instructions which he delivered to his
pupils. One of these was the kind called ἐξωτερικά, or ‘exoteric,’ the other ἀκροατικά, or ‘acroatic.’’
Those were called ‘exoteric’ which he gave training in rhetorical exercises, logical subtlety, and acquain-
tance with politics; those were called ‘acroatic’ in which more profound and recondite philosophy was
discussed, which related to the contemplation of nature or dialectic discussions.” Aulus Gellius, The Atticc
Nights, translated by John C. Rolfe (Cambridge, MA: Harvard University Press, 1961), vol. 3, 431–33.)
15Cf. LAMM 93; NRH H 155 and 248.
16Cf. GS–1 244f.; SCR R 190.
17Cf. CMM 21f.: “[Aristotle] is much less sure than Hippodamus of the virtues of innovation. It seems
that Hippodamus had not given thought to the difference between innovation in the arts and innova-
tion in law, or to the possible tension between the need for political stability and what one might call
technological change.”
18LS is alluding to the legal maxim “stare decisis, et non quieta movere” (“to stand by decisions
and not to move quietude”) that calls for the adherence to precedents and warns against changes. In
Germany, “quieta non movere” became well known after Bismarck mentioned the proverb in a speech
in 1891: “There is an old, good political proverb: Quieta non movere, that means, do not disturb what rests
quietly; and this is truly conservative: not to support a legislation which upsets something for which no
need for change exists.” Cf. Otto von Bismarck, Werke in Auswahl, edited by Rudolf Buchner and Georgg
Engel (Darmstadt: Wissenschaftliche Buchgesellschaft, 1983), vol. 8 (B), 73.
19Cf. Plato, Republicc 382a–e and 535d–e. See PAW W 35 and WIPP P 136.
APP D F
APPENDIX

SUPPLEMENT 2: LATER PLAN OF


“EXOTERIC TEACHING”

Plan.
I
1. Philosophy and class. scholarship; Zeller.
2. Husserl: Philos. als strenge Wiss. und Philos. als Weltanschauung.1
3. Lessing’[s] explanation of exotericism.
4. Schleiermacher’[s] criticism of exotericism. Hegel’s criticism of exoter-
icism.2
5. The basis of Lessing’s rediscovery of exotericism: the political problem.

1The second point of the list was inserted in the margin. The numbers of the first part of the plan
were changed accordingly.—Edmund Husserl’s “Philosophie als strenge Wissenschaft” was first pub-
lished in Logos 1, no. 3 (1911): 289–341. In his late essay “Philosophy as Rigorous Science and Political
Philosophy,” LS writes: “Let us see whether a place for political philosophy is left in Husserl’s philosophy..
What I am going to say is based on a re-reading, after many years of neglect, of Husserl’s programmatic
essay ‘Philosophy as Rigorous Science.’ The essay was first published in 1911, and Husserl’s thought
underwent many important changes afterward. Yet it is his most important utterance on the ques-
tion with which we are concerned” (SPPP P 34). In the same essay, LS also deals with Husserl’s view off
“Weltanschauung” (SPPP P 36f.).
2The second part of this point was inserted in the margin.—Cf. G. W. F. Hegel, Vorlesungen über diee
Geschichte der Philosophie II,
I in Werke, edited by Eva Moldenhauer and Karl Markus Michel (Frankfurt am
Main: Suhrkamp, 1986), vol. 19, 21f.: “Eine andere Schwierigkeit soll die sein: man unterscheidet exo-
terische und esoterische Philosophie. Tennemann sagt ([Geschichte der Philosophie,] Bd. II, S. 220): ‘Platon
bediente sich desselben Rechts, welches jedem Denker zusteht, von seinen Entdeckungen nur so viel,,
als er für gut fand, und nur denen mitzuteilen, welchen er Empfänglichkeit zutraute. Auch Aristoteles
hatte eine esoterische und exoterische Philosophie, nur mit dem Unterschiede, daß bei diesem derr
Unterschied bloß formal, beim Plato hingegen auch zugleich materiall war.’ Wie einfältig! Das sieht aus,,
als sei der Philosoph im Besitz seiner Gedanken wie der äußerlichen Dinge. Die Gedanken sind aberr
ganz etwas anderes. Die philosophische Idee besitzt umgekehrt den Menschen. Wenn Philosophen sich
über philosophische Gegenstände explizieren, so müssen sie sich nach ihren Ideen richten; sie können
sie nicht in der Tasche behalten. Spricht man auch mit einigen äußerlich, so ist die Idee immer darin
enthalten, wenn die Sache nur Inhalt hat. Zur Mitteilung, Übergabe einer äußerlichen Sache gehört
nicht viel, aber zur Mitteilung der Idee gehört Geschicklichkeit. Sie bleibt immer etwas Esoterisches;;
man hat also nicht bloß das Exoterische der Philosophen. Das sind oberflächliche Vorstellungen.””
(“Another difficulty is said to be the following: a distinction is made between exoteric and esoteric
292 APPENDIX F: SUPPLEMENT 2

6. Lessing—Leibniz—Spinoza (—RMbM)
7. Lessing’[s] intransigent classicism.

II3

7. Aristotle’s “exoteric” writings.4


8. Cicero.
9. Xenoph. Cyneg.
10. Plato’s Letters
11. Plato’s dialogues. Phaedrus Rep Timaeus.
12. Plato on the poets
and Hesiod on Muses.
13. Herakleitus
14. The big exceptions: Epicurus and Sophists. Cic. Rep. III

The questions: Why do they hide? and How can we decipher their truths will
be discussed in5 the continuation of this article. The historian cannot do more than
to show that the ancient philosophers did hide their thoughts, that their works
are—mixtures of truth and lies. The question of why they did it, must be answeredd
by a philosopher.

philosophy. Tennemann ([Geschichte der Philosophie,] vol. II, 220) says: ‘Plato exercised the same right that
every thinker has to communicate only so much of his discoveries as he thought good, and only to those
whom he credited with capacity to receive it. Aristotle, too, had an esoteric and an exoteric philosophy,,
but with the difference, that in his case the distinction was merely formal, while in the case of Plato it
was at the same time material.’ How simpleminded! This looks as if the philosopher is in possession of his
thoughts in the same way as of external things. But the thoughts are something utterly different. Insteadd
of the reverse, the philosophic idea is in possession of the human being. When philosophers elaborate
on philosophic subjects, they have to follow [the course of] their ideas; they cannot keep them in theirr
pocket. Even when speaking externally [äußerlich] to some people, the idea must be contained [in this
speech], if the matter [Sache] has any content at all. It does not take much to hand over an external item,,
but the communication of ideas requires skill. The idea always remains something esoteric; hence, one
does not merely have the exoteric [das Exoterische] of the philosophers. These notions are superficial.”)
3The second part of this plan has been published previously by Heinrich Meier in the introduction to
Die Denkbewegung von Leo Strauss. Die Geschichte der Philosophie und die Intention des Philosophen (Stuttgart/
Weimar: J. B. Metzler, 1996), 15n4.
4Cf. PAWW 28: “After the great theologian Schleiermacher asserted, with an unusually able argument,,
the view that there is only one Platonic teaching, the question of the esotericism of the ancient philoso-
phers was narrowed down, for all practical purposes, to the meaning of Aristotle’s ‘exoteric speeches’; and in
this regard one of the greatest humanists of the present day asserts that the attribution of a secret teachingg
to Aristotle is ‘obviously a late invention originating in the spirit of Neo-Pythagoreanism.’” (My italics,,
H.K.) Aristotle refers at least eight times in his works to exōterikoi logoii (cf., for example, Nicomachean
Ethics, 1102a26 and 1140a3, Politics, 1278b31 and 1323a22 as well as Metaphysics, 1076a28).
5LS first wrote “in a separate” but crossed it out.
APP D G
APPENDIX

LEO STRAUSS: LECTURE NOTES FOR


“PERSECUTION AND THE ART OF WRITING” (1939)

Edited by Hannes Kerber

[1 recto] Persecution and the Art of Writing.


The purpose of this lecture1 is to draw your attention to a certain approach to
earlier literature—to an approach which, to my mind, has not been sufficiently
considered. I do not say that the approach which I am going to suggest, is neces-
sarily correct, but I believe that it is worth considering. As my starting-point, I shall
choose certain phenomena which are accessible to everyone’s observation, at least
to-day, if not at all times.
Before I start, I want to point out one example of the questions which 2originally2
led me to consider the approach in question.You all know the Don Quixote3—you
know the story and the characters—you remember how Cervantes interrupts his
account of Don Quixote’s fight with the Biscaianer4 because, he says, he does
not know the continuation—how Cervantes, as he recounts, was looking aroundd
everywhere for the continuation until he discovered by chance an Arabic MS. on
the exchange in Toledo, how he got it translated into Castilian5—thus, the largerr
part of that immortal work claims to be translated from the Arabic, it claims to be
written, not by Cervantes, but by Sid Hamed, a Muslim.6—Why does Cervantes

1Cf. p. 273n10..
2Inserted by LS in pencil between the lines or in the margin.
3According to German custom, LS prefaces the title of Don Quixotee with the definite article, in orderr
to distinguish for the listeners his mentioning the book from his mentioning the hero.
4That is, the Biscayan.
5Cf. Miguel de Cervantes, Don Quixote, chapter IX. Cf., for example, NRH H 62; CMM 158; OPS S
169–70. See also “The Origins of Political Science and the Problem of Socrates,” in Interpretation 23,,
no. 2 (1996): 152.
6Cf. LS to Jacob Klein, August 18, 1939, in GS–3 580f.: “Since we are talking about exotericism— —
epaggelomai [I announce] that I have understood Don Quixote. The key is this: the book is the work off
two authors, of Cervantes and of Sid Hamed, i.e., a Christian and a Muslim. Now take away the artificial
294 APPENDIX G

tell this obvious lie? Is this just a joke for joke’s sake? Everyone admits that the Don
Quixotee is a deep book, Don Quixote is not just a fool, he represents something, cer-
tainly a folly, but a greatt folly, an eternall folly. What is then the reason that he makes
such a strange joke concerning the authorship of the book, that he attributes the
authorship to Sid Hamed? What is the relation of that strange joke to the eternal
folly represented by Don Quixote?
Cervantes satirizes the books of chivalry. This is his professed intention..
Accordingly the books of chivalry which drove Don Quixote mad, are burned by
the priest and barber of Don Quixote’s hometown, i.e. by the authority spiritual
and temporal.7 But, before they are burned, priest and barber discuss the merits andd
demerits of those books, and they find that they are innocent, and that quite a few off
them are even good literature. Why does then Cervantes satirize the whole genre?
Because of the idiotic imitations which abounded one or two generations before
Cervantes wrote the Don Quixote? Are we to believe that a man of Cervantes’ rank
shall waste his time with satirizing an ephemeral fashion? No—for all good books
are, and are meantt to be, possessions for all times.8
We might be inclined to say: well, difficulties or inconsistencies of that kindd
occur in practically all great books—aliquando dormitat bonus Homerus9—May
be. But may it not also be that we are somewhat naivee as regards those books and
their authors? That we underestimate the clarity of thought, the power of expres-
sion, the imagination, and, above all, the willingness and love for work (φιλοπονία)10
of these men? If they were inconsistent and sometimes11 insipid—may they not
have wishedd to be inconsistent and insipid? may they not have wishedd to give us
some riddles to solve? may it not be that the deficiencies of their works—all those

split of the one author, then you see that the author is Christian as well as Muslim, i.e., neither of the
two. The author is therefore a philosopher, and Don Quixote represents the founder of a religion and
Sancho Panza the believer. In fact, Don Quixote is the synthesis of Christianity (sorrowful countenance)
and Islam (holy war); he is superior to his predecessors in that he is furthermore educated and polite..
Dulcinea is Mary. The allusions to the Reformation, for example, are abundant. Consider also the role off
books in Don Quixote: Christianity and Islam are based on books.The deeds of Don Quixote are miracles..
Read the book on occasion again, and you will see that this is the case.”
7Cf. APT in GS–2 222: “Of Christian origin is, above all, Abravanel’s general conception of the
government of the Jewish nation. According to him, that government consists of two kinds of govern-
ments, of a government human and of a government spiritual or divine. This distinction is simply the
Christian distinction between the authority spiritual and the authority temporal.” Cf. GS–2 225. See
also NRH H 253–54.
8For the source of the expression, see Thucydides, History of the Peloponnesian War, 1.22.4 (“And,,
indeed, [the History] has been composed, not as a prize-essay to be heard for the moment, but as a pos-
session for all time.” Thucydides, History of the Peloponnesian War, translated by Charles Forster Smith
(Cambridge, MA: Harvard University Press, 1962), vol. 1, 41. Cf., for example, PAW W 160; CM M 142–43,,
157, 159, 228.
9“Sometimes [even] good Homer nods.”—For the source of this proverbial expression, see Horace,,
The Art of Poetry, 358–60 (“ ...et idem / indignor quandoque bonus dormitat Homerus, / verum operi
longo fas est obrepere somnum.”). Cf. PAW W 26.
10Cf., for example, Plato, Republic, 535d and Alcibiades I,
I 122c; Xenophon, Oeconomicus, 21.6. Cf. XSD
79 with 200.
11LS first wrote “somewhat”
somewhat but later replaced it by “sometimes.”
sometimes.
APPENDIX G 295

deficiencies exploited by 19th century higher criticism12—are intentionall and delib-


d lb
erate? 2At this point, we are helped somewhat by an experience we are unfortunate
enough to make to-day.2
[1 verso]
<1. Persecution → writing between the lines.
2. Persecution in the past → writing between the lines in the past.
II 3. Of the danger of reading between the lines: what is the difference between
legitimate reading between the lines and arbitrary guess work?13
Only if reading between the lines: more exactt and more exactingg than ordinary
reading, can it claim any consideration.14
Reading between the lines as regards the books in question, is necessarily
more exact than our ordinary reading: for a teaching transmitted between the
lines, is addressed to very carefull readers only.
Only the greatest care in reading can discover that teaching.15
But which are the cases in which we are entitled or rather compelled to
read between the lines?
In all cases in which ordinary reading is not sufficient to lead to an ade-
quate understanding. E.g. if we find insipid passages which a high school
[boy] would be ashamed of having written, in books of first class writers.16
III 4. Two types of persecution-literature.17
Generalization of “persecution” › social ostracism.18
How far do earlier writers, as a matter of principle, conceal their most sig-
nificant opinions?
Common to both types: conceal with regard to social conformity, which
is either merely enforced (modern type) or even desired?>

[3 recto]19 <is opposed to the orthodox view in its entirety, although he pays lip-
service to it on every page and in every sentence. If we read that author again, but
more vigilantly, and less innocently, we can be certain that we find many more
traces of his independence than the ones which struck us first.

12Cf. PAW
W 30–31.
13Cf. PAW
W 27, 30, 32; WIPP
P 224, 231; OT
T 27; ONI 351–52.
14Cf. PAW
W 30.
15Cf. PAW
W 144.
16Cf. PAW
W 30; TMM 36; WIPP P 223. The passage about the high-school boy may be an allusion to a
statement made by G. W. F. Hegel in the introduction (§3) of the Philosophy of Right. (See Grundlinien
der Philosophie des Rechts, in Werke, edited by Eva Moldenhauer and Karl Markus Michel (Frankfurt am
Main: Suhrkamp, 1986), vol. 7, 39–40.
17Cf. PAWW 33–34.
18Cf. PAWW 32–33, WIPP P 170.
19Unfortunately, the second page of the manuscript seems to have been lost.
296 APPENDIX G

l t evidence17either by the
In some cases, we are fortunate enough to possess explicit
20
authors or by intelligent philosophers proving that the author hides his real views,,
and indicates them only between the lines.>21
Lessing to MM: “You are more fortunate than other honest people who can
destroy the most loathsome structure of non-sense (sc. the orthodoxy in question)
only by pretending that they want to give it a new foundation.”22
Hobbes: points out at several occasions that he uttered certain “novel” views
during the Commonwealth only, i.e. at a time when the Elizabethan laws against
heresy were no longer valid.
and to Aubrey on Spinoza’s Tr. theol-pol.: he had not daredd to write so boldly.23
We have to base our interpretation of Hobbes preferably on the works published
under the Commonwealth, and, in case we find two sets of statements, one nearerr
to orthodoxy and another, contradictory, more remote from orthodoxy—we have
to consider the latter to be his true opinions.24
<“Of Liberty and Necessity” (London 1654, p. 35f.): “I must confess, if we con-
sider the greatest part of Mankind, not as they should be, but as they are, ...I must,,
I say, confess that the dispute of this question will rather hurt than help their piety;;
and therefore if his Lordship had not desired this answer, I should not have written
it, nor do I write it but in hopes your Lordship and his, will keep it private.”>25
4. Of the whole literature which teaches the truth concerning the crucial
questions exclusively between the lines, there are two types.26 The difference off
these two types corresponds to the difference of attitudes men may have towards
persecution.
a) The view most familiar to us, is that persecution is accidental, that it is an out-
come of a bad construction of the body politic; according to that view, persecution
oughtt to be replaced, and can be replaced, by freedom of speech; nay, persecution willl
be replaced by freedom of speech.
That view presupposes that the truth about the most important things can be
made accessible to the general public, i.e. that popular science is possible (Hobbes::
Paulatim eruditur vulgus).27 Belief in progress. A man who holds this view of per-
secution, writes and publishes his books in order to fightt persecution, in order to

20Inserted by LS between the lines. He first put “authors or by intelligent and benevolent contempo-
raries” but then crossed “and benevolent contemporaries” out and wrote “philosophers.”
21Cf. PAWW 32.
22Lessing to Moses Mendelssohn, January 9, 1771, in Werke, edited by Helmuth Kiesel with Georg
Braungart, Klaus Fischer and Ute Wahl (Frankfurt am Main: Deutscher Klassiker Verlag, 1988), vol. 11.2,
146.
23See Die Religionskritik des Hobbes in GS–3 277n20, HCR R 32n20, WIPPP 274. Cf. WIPPP 171 and
PAWW 183.
24Cf. PAWW 185–86.
25See The English Works of Thomas Hobbes, edited by William Molesworth (London: John Bohn, 1840),,
vol. IV, 256–57. Cf. PAWW 34n15.
26Cf. PAWW 33–34.
27“Gradually the vulgar become educated.” See Hobbes, De Homine, 14.13, in Opera latina, edited by
William Molesworth (London: John Bohn, 1839), vol. II, 128. Cf. PAWW 34n15; NRH H 200; ONI 360; Die
Religionskritik des Hobbes,
Hobbes, in GS–3
GS 3 348n243; HCR R 94n243.
APPENDIX G 297

contribute to the establishment of freedom of speech. And he hides h d his view merelyl
for fear of persecution, for fear of violent death or prison or exile.28
b) According to another view, persecution is essential, or necessary, and willl not
be superseded, and oughtt not to be superseded.What wee call persecution, the adher-
ent of that view would call: uniformity of thought as regards the fundamentals,,
among the citizens; such a uniformity, he holds, is a prerequisite of any 2healthy2
political life. And such a uniformity29 ought not to be endangered by public utter-
ance of divergent views, [3 verso] <however true>. <He holds that the require-
ments of political and social life are different from, and in a sense opposed to, the
requirements of philosophy or science:
the principle of political or social life is: quieta non movere30
the principle of philosophy and science, of theoretical life is: quieta movere
Arts and sciences oughtt to progress—but laws and customs ought to remain as stable
as possible.31 This view combines then intellectual radicalism with political and
social conservatism. An author of that kind hides his heterodox views not merely
for fear, but as a matter of duty towards the Commonwealth. Therefore, his tech-
nique of hiding is much more elaborate than that of an author who is interested
in political or social change. Therefore, his real views are much more difficult32 to
decipher, and, thus, his books are much more intriguing and interesting.33
2
Generalization of our topic: society and individual, thinking individual—society
l and
thought—How far do earlier authors, as a matter of principle, conceal their opinions?
How far is earlier technique of writing differentt from present-day technique?2
Generally speaking, a) is modern, and b) is ancient and medieval. Yet, we find
quite a few examples of the second type up to the 18th century.
I wish to speak mainly of the second type. For it is by far the more interesting
and important. Not merely historically, but also for us: that type produced the very
highest kind of literature in existence—a kind of literature which has provided
men, and willl provide men as long as they read at all, with the best and most solidd
kind of education.34 By being silent to all but extremely careful and vigilant readers,,
they compel us to be as careful and as vigilant, as flexible and as resourceful as we
possibly can. And thus, they educate us.>
If people hide their opinions, they will not say that they hide them, or at least they
will not say it too loud—or else they would defeat their own purpose. Therefore,,
explicit evidence in support of the view that an author hides his opinions, is rela-
tively rare. There is however a number of statements to this effect in existence.35

28Cf. SSTX 535: “It would, however, betray too low a view of the philosophic writers of the past if
one assumed that they concealed their thoughts merely for fear of persecution or of violent death.”
29LS first wrote “prerequisite” but later replaced it by “uniformity.”
30Cf. above, p. 289n18.
31Cf., for example, CMM 21–22.
32LS first wrote “hidden” but later replaced it by “difficult.”
33Cf. PAWW 34.
34Cf. PAWW 37.
35Cf. PAWW 32.
298 APPENDIX G

I have mentioned Lessingg already. Lessing has written two treatises on the the-
ology and philosophy of Leibniz,36 which show that Leibniz had two kinds off
teaching, a public and a private teaching. Lessing’s interpretation of that procedure
surpasses in depth everything I know of, written in the modern period. Anotherr
writing of Lessing’s, Dialogues on freemasonry,37 sets it beyond doubt that the methodd
of Leibniz he analyzed in the 2 treaties mentioned, was used by himself as well: it
was his settled principle not to state in his publications explicitly, what he really
thought of the then crucial question. A few years after his death, a private conversa-
tion of his was published,38 which gave people an idea, if a superficial one, that the
ordinary reader of Lessing’s writings, i.e. he who did not read between the lines, didd
not know Lessing’s view concerning the most important questions at all.39
Montesquieu is another author of that kind. In a recent discussion of his Spiritt
of [the] Laws, complaint is made of the total lack of order of that work, and of the
surprising amount of irrelevance to be met with in it.40 An extremely intelligent
contemporary of Montesquieu, d’Alembert, gives us some information about the
apparent deficiencies of Montesquieu’s work. “We say of the obscurity which one
may permit oneself in a book of that kind, the same what we said of the lack off
order. What would be obscure for ordinary readers, is not obscure for those whom
that author had in mind. Besides, voluntary obscurity is not really obscurity. M. de
Montesquieu had to present sometimes important truths, the absolute and direct
statement of which might have offended without bringing any benefits; therefore
he had the prudencee to envelop them; and by this innocent artifice, he hid them from
them [4 recto] to whom they would be harmful,41 without making them inacces-
sible to the wise.”42 Another friend of M. speaks of the “wonderful, if hidden order””
of the Spirit of [the] Laws.43 That is to say: the first task of the interpreter has to be::
to find out the reasons why M. discusses, say, this topic in this 2strange2 place.

36LS is alluding to “Leibniz von den ewigen Strafen” and “Des Andreas Wissowatius Einwürfe wider
die Dreieinigkeit,” in Lessing, Werke, vol. 7, 472–501 and 548–81. Cf. PAW W 182.
37LS is referring to Ernst und Falk. Cf. Lessing, Werke, vol. 10, 11–66. An English translation by
C. Maschler can be found in Interpretation 14, no. 1 (1986): 14–48.
38In 1785, Friedrich Heinrich Jacobi published Über die Lehre des Spinoza in Briefen an den Herrn
Moses Mendelssohn (On Spinoza’s Teaching, in Letters to Mr. Moses Mendelssohn). In this book, he reports a
conversation with Lessing in July 1780. LS gives a detailed account of the controversy between Jacobi
and Mendelssohn that followed the publication of the book (the “Spinozismusstreit”) in his introduc-
tion to Mendelssohn’s Morgenstunden and An die Freunde Lessings. The introduction, written in 1937,,
was first published in Moses Mendelssohn, Gesammelte Schriften Jubiläumsausgabe, vol. 3.2, edited by
Leo Strauss (Stuttgart-Bad Cannstatt: Friedrich Frommann Verlag, 1974), xi–xcv; reprinted in GS–2
528–605; English translation in LSMM M 59–145.
39In the margin, LS wrote “Rousseau p.m. 126, n. 2 and cf. 2nd Discours p. 40–41 with Contrat sociall IV
8.” The abbreviation “p.m.” is short for “penes me” (in my possession).
40LS is probably referring to George H. Sabine’s A History of Political Theory. Cf. PAW
W 28–29.
41LS first wrote “dangerous” but later replaced it with “harmful.”
42This passage from Jean-Baptiste le Rond d’Alembert’s Éloge de Montesquieu is quoted in PAW W
29n11.
43Cf. PAWW 29n11 with Stefano Bertolini, Analyse raisonnée de l’Esprit des Lois: “Voilà l’économie de
cet ouvrage magnifique. A la peinture que je viens de tracer, quelque foible qu’elle soit, il est aisé de voirr
que dans ce livre de l’Esprit des Lois règnent la précision, la justesse, un ordre merveilleux; ordre peut-
être caché aux yeux de ceux qui ne sauroient marcher que de conséquence en conséquence, toujours
APPENDIX G 299

Spinoza: one of the rules of life he set up for himself: ad captum vulgi loqui (to
adapt his language to the language of the vulgar).44 Tradition has it that the inscrip-
tion of his signet is: “Caute.”45 It would be a mistake to think that Spinoza’s Ethics
is nott written in the language of the vulgar. “Evasive”.
Descartes makes this entry in his diary: “Up to now, I have been a spectator off
this theatre of the world; but now being about to ascend the stage of that theatre,,
I put on a mask, just as the comedians do who do not wish that their feeling off
shame would46 become visible.” (“Ut comoedi, moniti ne in fronte appareat pudor,,
personam induunt, sic ego, hoc mundi theatrum conscensurus, in quo hactenus
spectator exstiti, larvatus prodeo.” Oeuvres X 213).47
Accordingly, he demands that some months shall be devoted to the perusal off
the 1st Meditation (VII 130).48 Writingg is an action, and as such subject to the
political and ecclesiastical authorities; but thoughtt recognizes no authority but reason
(Discours VI in princ.).49 The reall views of Desc. are not to be found in his writings
but between the lines of his writings.

guidés par des définitions, des divisions, des avant-propos, des distinctions, mais qui paroît dans tout son
jour aux esprits attentifs, capables de suppléer d’eux-mêmes les conséquences qui naissent des principes,
et assez habiles pour rapprocher et joindre dans la chaîne des vérités établies celles qui s’ensuivent, qui,,
aux yeux des connoisseurs, ne sont, pour ainsi dire, couvertes que d’un voile transparent.” (“Here is the
layout of this magnificent work. In the picture I have just sketched, however inadequate it may be, it is
easy to see the precision, accuracy, and wonderful order that reign in this book, The Spirit of Laws. It is an
order hidden perhaps from the eyes of those who can only proceed from consequence to consequence,,
always guided by definitions, divisions, forewords, and distinctions, but which appears fully illuminated
to attentive minds, who are capable by themselves of supplying those consequences born of principles,,
and who are skillful enough to bring forth and connect to the chain of established truths those truths
that follow therefrom, which in the eyes of experts, are, so to speak, covered only by a transparent veil.”)
The Analysee was reprinted in the Œuvres complètes de Montesquieu, edited by Édouard Laboulaye (Paris::
Granier, 1876), vol. III, here 60.
44In his Tractatus de intellectus emendationee (sect. 17), Spinoza declares as one of his regulae vivendi: “Ad
captum vulgi loqui, et illa omnia operari, quae nihil impedimenti adferunt, quominus nostrum scopum
attingamus.” (Opera, edited by Carl Gebhardt (Heidelberg: Carl Winters Universitätsbuchhandlung,,
1925), vol. II, 9.) In the article “How to Study Spinoza’s Theologico-Political Treatise,” LS translates the sen-
tence in the following way: “To speak with a view to the capacity of the vulgar and to practice all those
things which cannot hinder us from reaching our goal (sc. the highest good).” (PAW W 177, cf. 177–97.)
45Cf. PAWW 180.
46LS first wrote “might” but later replaced it with “would.”
47LS quotes the first few lines of the Cognitiones privataee in the edition by Charles Adam and Paul
Tannery (Paris: Léopold Cerf, 1897–1913).
48LS is referring to the Responsio ad secundas objectiones in the same edition.
49Cf. Descartes, Discours de la Méthode, edited by Étienne Gilson. (Paris: Librairie Philosophique J..
Vrin, 1939), 60 (beginning of part 6): “Or, il y a maintenant trois ans que j’étais parvenu à la fin du traité
qui contient toutes ces choses, et que je commençais à le revoir, afin de le mettre entre les mains d’un
imprimeur, lorsque j’appris que des personnes, à qui je défère et dont l’autorité ne peut guère moins
sur mes actions que ma propre raison sur mes pensées, avaient désapprouvé une opinion de physique,,
publiée un peu auparavant par quelque autre, de laquelle je ne veux pas dire que je fusse, mais bien que
je n’y avais rien remarqué, avant leur censure, que je pusse imaginer être préjudiciable ni à la religion
ni à l’État, ni, par conséquent, qui m’eût empêché de l’écrire, si la raison me l’eût persuadée, et que
cela me fit craindre qu’il ne s’en trouvât tout de même quelqu’une entre les miennes, en laquelle je me
fusse mépris, nonobstant le grand soin que j’ai toujours eu de n’en point recevoir de nouvelles en ma
créance, dont je n’eusse des démonstrations très certaines, et de n’en point écrire qui pussent tourner
300 APPENDIX G

Bacon. “I sometimes alter the uses and definitions (of the traditional terms),,
according to the moderatee proceeding in civil government; where although there be
some alteration, yet that holdeth which Tacitus wisely noteth, Eadem magistratuum
vocabula.” (Ann. I 3).50 (Advanc. p. 92).51
Note his interest in ciphers.
The arcana imperiii literature in the 16th century (Lipsius52 etc.).
“Government ...is a part of knowledge secret and retired, in both these respects in
which things are deemed secret; for some things are secret because they are hard
to know, and some because they are not fit to utterr ...even unto the general rules
and discourses of policy and government is due a reverent and reserved handling.””
(205f.).53 No wonder that he did not finish his utopia, New Atlantis, and that he
omitted practically everything politicall from it.
The writers I have mentioned, are not the inventors of such techniques. They
make use of a tradition, of the traditional distinction between exotericc teaching and
esotericc teaching. An esoteric54 teaching is not, as some present day scholars seem to
think, a mysticall teaching: it is the scientificc teaching. Exoteric = popular. Esoteric =
scientific and thereforee secret.

au désavantage de personne.” (“It is now three years since I completed the treatise that contains all
these things, and began to review it before putting it in the hands of the printer, when I learned that
certain persons to whom I defer, and whose authority over my actions can scarcely be less than that
of my own reason over my thoughts, had disapproved of certain opinions in physics, published shortly
before by someone else. I do not wish to say that I agreed with it, but since I had noticed nothing in
it before their censure that I could imagine to be prejudicial either to religion or to the state, or con-
sequently that would have prevented me from writing it if reason had so persuaded me, this made me
fear that there might nevertheless be found among my thoughts some one that was mistaken, despite
the great care I have always taken not to receive new ones among my beliefs of which I did not have
very certain demonstrations, and to write nothing that could turn to the disadvantage of anyone.”
Descartes, Discourse on Method, trans. Richard Kennington (Newburyport, MA: Focus Philosophical
Library, 2007), 48)—Cf. PAW W 182–83.
50Cf.Tacitus, The Annals, 1.3:“Domi res tranquillae, eadem magistratuum vocabula; iuniores post Actiacam
victoriam, etiam senes plerique inter bella civium nati: quotus quisque reliquus, qui rem publicam vidis-
set?” (My italics, H.K.) (“At home all was calm. The officials carried the old names; the younger men
had been born after the victory of Actium; most even of the elder generation, during the civil wars; few
indeed were left who had seen the Republic.” Tacitus, The Histories.The Annals, translated by Clifford H..
Moore/John Jackson (Cambridge, MA: Harvard University Press, 1962), vol. 2, 249.)
51Cf. Francis Bacon, The Advancement of Learning, edited by Michael Kiernan (Oxford: Clarendon
Press, 2000), 81: “But to me on the other side that do desire as much as lyeth in my Penne, to ground a
sociable entercourse between Antiquitie and Proficience, it seemeth best, to keepe way with Antiquitie
vsque ad aras; And therefore to retaine the ancient tearmes, though I sometimes alter the vses and defi-
nitions, according to the Moderate proceeding in Ciuill gouernment; where although there bee some
alteration, yet that holdeth which Tacitus wisely noteth, Eadem Magistratuum vocabula.” Cf. PAW W 183.
52For Justus Lipsius (1547–1606), cf. CM M 144; Hobbes’ politische Wissenschaft, in GS–3 100–4.
53Cf. Francis Bacon, The Advancement of Learning, 179: “Concerning gouernment, it is a part of knowl-
edge, secret and retyred in both these respects, in which things are deemed secret: for some things are
secret, because they are hard to know, and some because they are not fit to vtter: wee see all gouernments
are obscure and inuisible. [ ...] Neuerthelesse euen vnto the generall rules and discourses of pollicie, and
gouernment, there is due a reuerent and reserued handling.” Cf. PAW W 57n63.
54LS first wrote “exoteric”
exoteric but later replaced it with “esoteric”.
esoteric .
APPENDIX G 301

Tradition has it that Aristotle le wrote two kinds of books: exoteric and esoteric
books. But the content of the esoteric books was originally not destined for publi-
cation at all: they are still called acroamatic, oral.55 Scientific teaching was oral teach-
ing, because written teaching cannot remain secret.The truth cannot and ought not
to be published—i.e. the truth about the highest things—what can be published, are
things which aree public in themselves, ἔνδοξα, moral and political things.
[4 verso]The story of the correspondence between Alexander the Great and
Aristotle: Alexander complained to Aristotle that Ar. had published his oral teach-
ing. Ar. answer: those books are published and not published; for they are intelligible
only to those who have heard my lectures.56
This tradition may be spurious. But even spurious traditions are significant.
The attitude presupposed by that tradition, is certainly much older than the
Christian period. We find it clearly expressed in the 2nd and 7th of the Platonicc
letters. I believe, they are genuine57—but even if not, the “forger” knew more off
Plato than we know.

55Cf. Aulus Gellius, The Attic Nights, 20.5. Cf. above, p. 289n14.
56Cf. Aulus Gellius, The Attic Nights, 20.5: “Eos libros generis ‘acroatici’ cum in vulgus ab eo editos
rex Alexander cognovisset atque ea tempestate armis exercitam omnem prope Asiam teneret regemque
ipsum Darium proeliis et victoriis urgeret, in illis tamen tantis negotiis litteras ad Aristotelem misit, non
eum recte fecisse, quod disciplinas acroaticas, quibus ab eo ipse eruditus foret, libris foras editis involgas-
set: ‘Nam qua,’ inquit, ‘alia re praestare ceteris poterimus, si ea quae ex te accepimus omnium prosus fient
communia? Quippe ego doctrina anteire malim quam copiis atque opulentiis.’ Rescripsit ei Aristoteles
ad hanc sententiam: ‘Acroaticos libros, quos editos quereris et non proinde ut arcana absconditos, neque
editos scito esse neque non editos, quoniam his solis cognobiles erunt, qui nos audiverunt.’” (“When
King Alexander knew that he [sc. Aristotle] had published those books of the ‘acroatic’ set, although at
that time the king was keeping almost all of Asia in state of panic by his deeds of arms, and was pressing
King Darius himself hard by attacks and victories, yet in the midst of such urgent affairs he sent a letter
to Aristotle, saying that the philosopher had not done right in publishing the books and so revealing to
the public the acroatic training, in which he himself [sc. Alexander] had been instructed. ‘For in what
other way,’ said he, ‘can I excel the rest, if that instruction which I have received from you becomes the
common property of all the world? For I would rather be first in learning than in wealth and power.’’
Aristotle replied to him to this purport: ‘Know that the acroatic books, which you complain have been
made public and not hidden as if they contained secrets, have neither been made public nor hidden,,
since they can be understood only by those who have heard my lectures.’” Aulus Gellius, The Attic Nights,,
translated by John C. Rolfe (Cambridge, MA: Harvard University Press, 1961), vol. 3, 433–35.) See also
Plutarch, Life of Alexander, 7.
57Cf. LS to Jacob Klein, November 28, 1939, in GS–3 586: “By now, I am firmly convinced that alll
Platonic letters (even the first) are genuine: they are the Platonic counterpart [Gegenstück] to Xenophon’s
Anabasis: they are supposed to show that the author has not been corrupted by Socrates: while the
author constantly veils himself in the dialogues, it is the purpose of the letters as well as of the Anabasis
to show that the one who veils himself is absolutely harmless, absolutely normal. He unveils himself as
normal by writing the first three and the last letter to a tyrant (Dionysius); moreover: the letters which
are directed at philosophers deal exclusively with πολιτικά, philosophy is discussed only in letters to
πολιτικοι, and in such a way that close reading fully destroys the fiction on which the whole matter is
based: the seventh letter is precisely in the middle!”
58Cf. OPS S 29.
302 APPENDIX G

Ep. II (314a–c).58 Ep.VII (332d):59


a) the teaching cannot [be] “said” at all.
b) it can be “said” in writing or in speech, but well or badly.
c) saying it badly = stating it in writing fully to the public.

Plato’s dialogues. The Phaedrus on the danger inherent in all writing. A writ-
ing does not know to whom it ought to talk and to whom it ought to be silentt
(276a6–7). This is so important becausee the truth is not fit for everybody. Inferiority
of all writing to oral instruction. “It is hard to find the father of all things, but to
speak of him unto all men is impossible.” Timaeus.60
But Plato did write books about the most important topics: on nature, ideas, idea
of the good, soul etc. How can we reconcile his refusal to write about such top-
ics with his actual practice? Only by assuming that he did write and did nott write
about them at the same time. Just as Aristotle is said to have said of his esoteric
works: they are published and they are nott published. Books do speak and be [recte::
are?] silent according to the capacities of the reader. Plato didd write about the truth::
but he did it enigmatically.
All Platonic writings are dialogues. Dialogues are a kind of dramas (dramas in prose
and without women and nearer to comedy than to tragedy).61 What is the charac-
teristic feature of 2the2 drama according to Plato? Drama is that kind of poetry in
which the author hides himself.62 By writing dialogues, Plato gives us to understand
that he hides himself, i.e. his thought. Plato never said a word on his teaching—only
his characters do. But his main character, Socrates, does not63 speak when the highest
topic, the κόσμος, or the being, is discussed: Timaeus or the Eleatic stranger.
Plato’s school.The Academic philosophers, the successors of Plato, say: In order to
discover the truth, one must dispute pro and con as regards everything. The adver-
sary: “I should like to see whatt they have discovered.” The Academic: “We are not
used to show[ing] it.” Adversary: “But what in the world are these mysteries? or why
do you conceall your opinion, as if it were something disgraceful?” Academic: “That
those who hear, will be swayed by reason, rather than authority.” (Lucullus 60).64

59LS first wrote “314c–e” but later replaced it by “332d”.—On the Seventh Letter, cf. ONI 348–51
with LS to Karl Löwith, August 15, 1946, in GS–3 663.
60Plato, Timaeus, 28c3–5. Cf. PAW W 35n17 and FP 375n44.
61See CM M 61: “Socrates left us no example of weeping, but, on the other side, he left us example
of laughing. The relation of weeping and laughing is similar to that of tragedy and comedy. We may
therefore say that the Socratic conversation and hence the Platonic dialogue is slightly more akin to
comedy than to tragedy.” Cf. OPS S 279. Contrast LS to Alexandre Kojève, April 22, 1957, in OT T 275:
“All the Dialogues are tragicomedies. (The tragedian is awake while the comedian is sleeping at the end
of the Symp.)”
62Plato, Republic, 393c. Cf. OTT 32.
63LS first wrote “never speaks” but later replaced it by “does not.”
64Cf. Cicero, Academica, 2.59f.: “Mihi porro non tam certum est esse aliquid quod comprendi possit
(de quo iam nimium etiam diu disputo) quam sapientem nihil opinari, id est numquam adsentiri rei vel
falsae vel incognitae. Restat illud quod dicunt veri inveniundi causa contra omnia dici oportere et pro
omnibus. Volo igitur videre quid invenerint. ‘Non solemus,’ inquit, ‘ostendere.’ ‘Quae sunt tandem ista
mysteria, aut cur celatis quasi turpe aliquid sententiam vestram?’ ‘Ut qui audient,’ inquit, ‘ratione potius
quam auctoritate ducantur.’” (“For my part, moreover, certain as I am that something exists that can be
grasped (the point I have been arguing even too long already), I am still more certain that the wise man
APPENDIX G 303

It is Cicero who relates this little dialogue. Cicero himself was an academic..
Consequently, he says of himself: “we have preferably followed that kind of philoso-
phy (sc. dialogic philosophy) which, as we believed, Socrates has used, (and we did
this) in order to hide our own opinion, to free others from error, and to investigate in
each discussion what is most likely to be true.” (Tusc. V 11).65 There is a connection
between hidingg and arriving at a result which is only likely to be true, which is
only a likely tale:66 the truee tale is hidden; it is revealed perhaps in a dream (Somnium
Scipionis).67
[5 recto] Hiding one’s thought is irreconcilable with a perfectly clear and lucid
plan. A lucid plan does not leave room for hiding-places—as a consequence, an
exoteric book will not have a very lucid plan.68

Cf. Lessing, Leibniz von den ewigen Strafen: “Ich will mich so kurz fassen wie
nur möglich und meine Gedanken wenn nicht ordnen so doch zählen.”69

Cf. the twofold discussion of poetry in Rep.70—


ib.: the interruption after the enumeration of the bad constitutions—

never holds an opinion, that is, never assents to a thing that is either false or unknown. There remains
their statement that for the discovery of the truth it is necessary to argue against all things and for all
things. Well then, I should like to see what they have discovered. ‘Oh,’ he says, ‘it is not our practice
to give an exposition.’ ‘What pray are these holy secrets of yours, or why does your school conceal its
doctrine like something disgraceful?’ ‘In order,’ says he, ‘that our hearers may be guided by reason rather
than by authority.’” Cicero, De Natura Deorum. Academica, translated by H. Rackham (Cambridge, MA:
Harvard University Press, 1961), 543.)
65Cf. Cicero, Tusculan Disputations, 5.4.10–11: “Socrates autem primus philosophiam devocavit e caelo
et in urbibus collocavit et in domus etiam introduxit et coëgit de vita et moribus rebusque bonis et malis
quaerere: cuius multiplex ratio disputandi rerumque varietas et ingenii magnitudo, Platonis memoria et
litteris consecrata, plura genera effecit dissentientium philosophorum, e quibus nos id potissimum consecuti
sumus, quo Socratem usum arbitrabamur, ut nostram ipsi sententiam tegeremus, errore alios levaremus et in omni
disputatione quid esset simillimum veri quaereremus.” (My italics, H.K.) (“Socrates on the other hand was the
first to call philosophy down from the heavens and set her in the cities of men and bring her also into
their homes and compel her to ask questions about life and morality and things good and evil: and his
many-sided method of discussion and the varied nature of its subjects and the greatness of his genius,
which has been immortalized in Plato’s literary masterpieces, have produced many warring philosophic
sects of which I have chosen particularly to follow that one which I think agreeable to the practice of
Socrates, in trying to conceal my own private opinion, to relieve others from deception and in every
discussion to look for the most probable solution.” Cicero, Tusculan Disputations, translated by J. E. King
(Cambridge, MA: Harvard University Press, 1966), 435). Cf., for example, NRH H 154–55.
66Cf. Plato, Timaeus, 29d. See, for example, PAW, W 35 and SPPP, 166.
67Cf. Cicero, De re publica, 6.9ff.
68Cf. SSTX 523–24.
69Lessing, Leibniz von den ewigen Strafen (Werke, vol. 7, 481): “Ich will, was ich zu sagen habe, so kurz
zu fassen suchen, als möglich; und meine Gedanken wo nicht ordnen, doch zählen.” (“I want to try to
express what I have to say as briefly as possible; and where I won’t put my thoughts in order, I’ll at least
number them.”) Cf. LS to Hans-Georg Gadamer, February 26, 1961, in: “Correspondence concern-
ing Wahrheit und Methode,” The Independent Journal of Philosophy/Unabhängige Zeitschrift für Philosophiee 2
(1978): 6.
70CMM 133: “Socrates suddenly returns to the subject of poetry, a subject which had already been
discussed at great length when the education of the warriors was being considered. We must try to
understand this apparently unmotivated return.” Cf. CM M 134–37.
304 APPENDIX G

the twofold discussion of education in Laws.71 Repetition.72


the insertion of theology into penal law in the Laws73 and into the discussion of
noble lies in the Rep.74
the plan of Xen. Memor.75

Moreover, hiding one’s thought is not reconcilable with absolutely lucid expres-
sions: if everything is absolutely clearly expressed, there is no room for hiding places
within the sentences.
A man who hides his thought will then accept the following maxim: “What
is written beautifully and in order, is nott written beautifully and in order.” (Xen.
Cyn[egeticus]. 13, 6).76 This maxim occurs in a treatise on hunting with dogs,
which is a rather good hiding place.
Hiding one’s thoughts about the crucial things, when speaking or writing about
those things, means making misstatements about those things—or: to liee about those
things.

71Cf., for example, AAPLL 23, 27 and 104–5.


72Cf. OPS S 237:“General rule: there is never a repetition in Plato which is an identical repetition; there
is always a change, though sometimes seemingly trivial.” See PAW W 16 and 62–64.
73According to LS’s interpretation, books 9–12 of the Laws are “chiefly devoted to the penal law” ”
(AAPL L 64, cf. 126). With regard to the Athenian’s natural religion in book 10, LS writes: “The Athenian
is thus compelled or enabled to discuss what Adeimantos calls theology (Republicc 379a5–6) within the
context of the penal law, whereas Socrates discussed it within the context of pre-philosophic, nay, the
most rudimentary education.” (AAPL L 140) Cf., for example, LS to Jacob Klein, February 16, 1939, in
GS–3 567 as well as “Plato,” in History of Political Philosophyy (third edition), 85–86.
74Cf. CMM 98, 102–03.
75LS’s interpretation of the Memorabilia can be found in XS S 1–126.
76Cf. PAWW 29, SSTX 502, and LS to Jacob Klein, August 7, 1939, in GS–3 575–76.
CONTRIBUTORS

Gabriel Bartlett is Instructor in Philosophy at St. Xavier University, Chicago. He


is cotranslator of Leo Strauss, Hobbes’s Critique of Religion and Related Writings, and off
Strauss, “The Place of the Doctrine of Providence According to Maimonides.”
Nasser Behnegar is Associate Professor of Political Science at Boston College. He
received his MA in economics and PhD from the Committee on Social Thought
of the University of Chicago. He is the author of Leo Strauss, Max Weber, and thee
Scientific Study of Politics and is currently working on a book-length study of the
liberalism of John Locke.
Jeffrey A. Bernstein is Associate Professor of Philosophy at the College of the
Holy Cross. He works in the areas of Spinoza, German philosophy, and Jewish
thought. He is currently at work on a book dealing with Leo Strauss at the borders
of Judaism, philosophy, and history.
Timothy W. Burns is Professor of Political Science at Baylor University. He
is author of Shakespeare’s Political Wisdom; coauthor (with Thomas L. Pangle) off
Introduction To Political Philosophy; editor of After History? Francis Fukuyama and his
Critics and of Recovering Reason: Essays in Honor of Thomas L. Pangle; translator off
Marcellinus's “Life of Thucydides”; and author of articles on Homer, Thucydides,,
Aristophanes, Strauss, Fukuyama, Putnam, Chesterton, and modern liberal repub-
lican theory.
Steven Frankel has taught at the American University of Paris, where he received
the Board of Trustees Award for Distinguished Teaching in 2001, and now teaches
at Xavier University in Cincinnati. His scholarly work focuses on the relation-
ship between philosophy and religion; it has appeared in over a dozen journals,,
including Archiv für Geschichte der Philosophie, International Philosophical Quarterly,,
Interpretation, Journal of Jewish Thought and Philosophy, Review of Metaphysics, Review off
Politics, and Teaching Philosophy. He is also the author of French Culture and Politics::
A Guide for Students.
David Janssens is Assistant Professor of Philosophy at Tilburg University, The
Netherlands. He is the author of Between Athens and Jerusalem: Philosophy, Prophecy,,
and Politics in Leo Strauss’s Early Thoughtt and of numerous articles on ancient law,,
philosophy and poetry, and the work of Leo Strauss. His current research focuses
on the quarrel between philosophy and poetry.
306 CONTRIBUTORS

Hannes Kerber is Research Assistant at the Carl Friedrich von Siemens


Foundation in Munich. He is the author of a forthcoming study of Leo Strauss’s
hermeneutics and is currently preparing a dissertation at the University of Munich
on Gotthold Ephraim Lessing’s critique of the Bible.
Heinrich Meier is Director of the Carl Friedrich von Siemens Foundation in
Munich, Professor of Philosophy at the University of Munich, and permanent
Visiting Professor at the Committee on Social Thought, the University of Chicago..
The author of eight books, including Carl Schmitt and Leo Strauss: The Hidden
Dialogue, The Lesson of Carl Schmitt: Four Chapters on the Distinction between Politicall
Theology and Political Philosophy, and Leo Strauss and the Theologico-Political Problem, he
is also the editor of Leo Strauss’s Gesammelte Schriften. In 2005 he was awarded the
Leibniz Medal of the Berlin-Brandenburg Academy of Sciences. His studies on the
theologico-political problem, on political theology, and in political philosophy have
been translated into English, French, Italian, Japanese, Polish, and Spanish; further-
more, all of them are already or will soon be available in Chinese editions.
Thomas L. Pangle is the Joe R. Long Chair in Democratic Studies in the
Department of Government at the University of Texas, Austin, where he directs the
JJefferson Center for the Study of Core Texts and Ideas. He has published widely in
the field of political philosophy, his 70 articles and book chapters and his numerous
reviews addressing such topics as Socratic rationalism, the competing philosophic
and scriptural understandings of wisdom and of moral and civic life; classical and
modern republicanism; biblical politics; the Enlightenment as a theologico-political
project; the Federalist-anti-Federalist debate, American constitutionalism, the moral
basis of international relations; and postmodernity. He sits on advisory boards off
centers and institutes for the study of constitutionalism, of public affairs, of democ-
racy, of liberal education, and of Western civilization. He is the general editor off
Cornell University Press’s Agora editions, is on the editorial boards of three schol-
arly journals, including Political Research Quarterly. Pangle has published 15 books,,
the most recent of which is Aristotle’s Teaching in thee Politics. He is coauthor (with
Timothy W. Burns) of the forthcoming Political Philosophy: An Introduction.
Joshua Parens is Professor of Philosophy and Dean of the Braniff Graduate School
of Liberal Arts at the University of Dallas. He is the author of Maimonides andd
Spinoza: Their Conflicting Views of Human Nature, and of two books on Alfarabi— —
Metaphysics as Rhetoricc and An Islamic Philosophy of Virtuous Religions. With Joseph
Macfarland, he has coedited the second edition of Medieval Political Philosophy: A
Sourcebook, whose first edition was edited by Ralph Lerner and Muhsin Mahdi. He
has also published various articles on Alfarabi, Maimonides, and Spinoza.
Richard S. Ruderman is Associate Professor and Chair of the Political Science
Department at University of North Texas. He has published essays on “Aristotle
and the Recovery of Political Judgment,” “Homer’s Odyssey and the Possibility off
Enlightenment,” “Democracy and the Problem of Statesmanship,” “Locke on the
Parental Control of Education,” “‘Let Freedom Ring’:The Abolitionism of William
m
Lloyd Garrison and Frederick Douglass,” and “A Slingshot Recoils: The Critique
of Philosophy in Halevi’s Kuzari.”
CONTRIBUTORS 307

Martin D. Yaffe is Professor of Philosophy and Religion Studies at University off


North Texas. He is the author of Leo Strauss on Moses Mendelssohn and of Shylock
and the Jewish Question; editor off Judaism and Environmental Ethics: A Reader; coedi-
tor of The Companionship of Books: Essays in Honor of Laurence Berns and of Emill
L. Fackenheim: Philosopher, Theologian, Jew; translator of Spinoza’s Theologico-Politicall
Treatise; and cotranslator of Thomas Aquinas’s Literal Exposition on the Book of Job.
INDEX

Abravanel, Don Isaac, 8, 11–12, 14, 18, Arkush,


k h Alan,
l 53n9, 188n49, 188n53,
20, 26n11, 26n13, 27nn20–1, 27n25, 189n60
29n30, 158, 164–5, 168n24, 169n30, Ascherson, Ferdinand, and Paul Eichholtz,
215n3, 215n6, 216n3, 216n6, 267–70, 213n40
294n7 atheism; atheist; atheistic, 5, 20–1, 29n31,
Adam, Charles, and Paul Tannery, 299n47 36, 40, 42–3, 51, 54n23, 56n37, 61–2,
Adler, Cyrus, 31n44 64–5, 67n9, 92–3, 99, 101, 111, 125,
Adler, Eve, 54n24, 160, 167n2 132–3, 142, 144–7, 148n6, 154n23,
Ahad Ha‘am (Asher Hirsch Ginsberg), 38, 155n25, 174–5, 179–80, 182, 184, 208,
238n7, 240 233, 240
Alcibiades, 91. See also Plato, Alcibiades I Aubrey, John, 296
Alexander of Aphrodisias, 85–6, 164, Augustine, St., 26n14, 57, 62, 64, 88,
232, 248 213n46
Alexander the Great, 289, 301n56 Aulus Gellius, 282nxxvii, 289, 301nn55–6
Alfarabi. Seee Farabi Averroes (Ibn Rushd); Averroism, 15, 17,
Algazel (Ghazali), 89 25n7, 63, 160, 167n13, 183, 213n42,
Altmann, Alexander, 22, 23, 32n51, 188n49, 246, 257nn17–18, 258n20
189n53, 190n79 Avicenna (Ibn Sina), 17, 28n27, 74–5,
Amin, Uthmân, 258n21 78n41, 86, 89, 90, 106n13, 107n19,
Anaxagoras, 145 157, 161, 168n17, 183, 214n50, 256–7
ancient/classical/premodern philosophy, 2–4,
11, 16, 33, 41, 42, 44, 46, 51, 65, 91, Bacher, W., M. Brann and D. Somonsen,
102, 107n15, 109n47, 110n67, 112n83, 76n1
115–17, 127, 131, 133, 186, 247 Bachofen, J. J., 155n26
Andronicus of Rhodes, 211n22 Bacon, Francis, 300
Aquinas. Seee Thomas Aquinas Baer, Fritz, 14, 24n6
Arendt, Hannah, 53n12 Barker, Ernest, 14
Aristotle; Aristotelians, 1, 5, 12, 19, 23n2, Baron, Salo Wittmayer, 167n6, 213n49
25n7, 27n24, 31n43, 32n50, 34–5, Barry, Brian, 12n3
63–4, 67, 69–75, 76n4, 77n33, 78n49, Barth, Karl, 5, 57, 59–60, 80
86–9, 102, 105n10, 107n19, 110n66, Bartlett, Gabriel, 7, 131, 147n2, 167n4,
112n83, 143–6, 152n20, 154n23, 255–65
155n27, 160–1, 165–6, 167n11, Bartlett, Robert, 167n3
168n16, 169n34, 189n64, 193, 203, Behnegar, Nasser, 6–7, 115–29, 167n10
210n15, 211n22, 215, 241, 246n27, Beiser, Frederick C., 186n4, 187nn15–17,
247, 257n15, 257n18, 258–65, 270n6, 187nn23–4, 188n49, 189n50, 189n65
275, 276n1, 286nxxvii, 289n14, Bertolini, Stefano, 298n43
289n17, 291n2, 292, 301, 302 Ben Shammai, Haggai, 168n22
310 INDEX

Benardete, Seth, 203, 212n38, 213n41 dd’Alembert,


l b Jean-Baptiste lle Rond,
d 298
Benjamin, Walter, 21, 30n35 Dannhauser, Werner, 56n37
Bernays, Jacob, 210n15 David, Moritz, 256n6
Bertolini, Stefano, 298n43 Democritus; Democritean, 110n62, 143,
Bismarck, Otto von, 289n18 154n23, 247
Boeckh, August, 213n40 Descartes, René; Cartesian, 7, 80, 94,
Bruder, Carl H., 209n3, 285n113, 139–41, 143–4, 150n14, 151nn15–16,
286nxxxvi, 286n117 154n21, 173–6, 178–80, 182–3, 203–4,
Bruell, Christopher, 107n20, 109n54, 285, 299–300
156n28 Deutsch, Kenneth L., and Walter Nicgorski,
Buber, Martin, 4, 15, 24, 41–2, 44–5, 54, 186n5
100, 176 Dieterici, Friedrich, 264n49
Buchheim, Thomas, 212n34 Dionysius of Syracuse, 301n57
Buchner Rudolf, and Georg Engel, 289n18 divine law; revealed law, 17–18, 41, 49, 51,
Burger, Ronna, and Michael Davis, 212n38 59, 63, 68n11, 74, 80, 86, 90, 104, 159,
Burke, Edmund, 95–6, 109n61, 110n62 161, 167n10, 168n25, 177–8, 199, 268
Burns, Timothy W., 6–8, 79–113, 131–56; divine right, 65. See also right
and Lucien Oulahbib, 107n20,
109n54 Ebbinghaus, Julius, 22, 26n11, 30n40, 89,
Butterworth, Charles, 167n13, 169n33, 116, 128n9
255n3 education, 10, 123, 194–5, 197, 200–1, 223,
283n82, 297, 303n70, 304n73
Callicles, 252. See also Plato, Gorgias Efros, Israel, 255, 257n14, 261n36
Calvin, Jean, 68n10, 89, 94, 96, 106n14, 131, Eichmann, Adolf, 53n12
133, 136, 139, 141, 149nn11–13 Engels, Friedrich, 155n26
Caro, Joseph, 49 Enlightenment; enlightenment, 4, 8–9,
Caspi, Joseph ibn, 12, 169n30, 267–8 17–18, 20, 22, 25n9, 26n14, 29n28,
Cassirer, Ernst, 1, 21, 57, 77, 100, 111–12, 29n31, 33, 50, 54n23, 58–65, 69–71,
187n10 92, 92–3, 95–6, 98, 103, 132–3, 136,
Cervantes, Miguel de, 293–4 146, 149–50, 155n25, 171–84, 187n23,
Christ; Christianity, 5, 7, 28, 35–6, 44, 47–9, 188n31, 192, 195, 200, 223, 232–4,
54, 57–9, 62–6, 79, 88, 95–7, 101, 241–2, 247–9, 251, 283–5, 289
106–7, 111n73, 128, 134–5, 152–3, Epicurus; Epicurean, 5, 32n50, 33, 61, 63–4,
158, 164–5, 168, 179, 184–5, 213, 94, 110n62, 132, 135, 142–3, 145–7,
222–3, 224, 242, 246, 269, 270, 283, 149nn9–10, 154n23, 155n27, 177–8,
293–4, 301 181, 234, 247, 288, 292
Cicero, 32n50, 78n47, 287, 288nn11–13, Epimetheus; Epimethean, 19, 27n22,
292, 302n64, 303 107n29, 145–6
Clemens Alexandrinus, 279nxvii Erler, Michael, 210n15
Cohen, Hermann, 3, 4, 5, 14, 22, 24n4, esotericism, esoteric writing, 3, 8, 10–12,
26n11, 26n13, 27n21, 27n25, 28n26, 16, 18, 23, 32n49, 33, 86–7, 97,
30n37, 31n43, 38, 44–50, 55nn27–9, 107n17, 157–8, 160, 162–4, 166, 185,
55n31, 55n33, 56n35, 64, 69–75, 192–3, 201, 204–8, 209n8, 210n12,
76nn1–3, 76n5, 76n10, 76n12, 77n13– 210nn15–16, 211n23, 213n40, 213n42,,
17, 77n19, 77nn21–2, 77n24, 77nn27– 262n41, 275–83, 287, 288n9, 289,
33, 78nn35–40, 78nn42–50, 86, 103–4, 291n2, 300–2
105n9, 107n17, 111n73, 113n88, 176, Eusebius of Caesaria, 213n42
186n8, 238–9
Cremonini, Cesare, 246 Faber, Marion, and Stephen Lehmann,
Crosson, Frederick J.,J 213n46 54n18
INDEX 311

Fackenheim, Emil L., 53n11, 89, 106n13, Grunsky,


k Hans Alfred,
lf d 24n4
108n31 Gunnell, John G., 186n5
Falaquera, Shem Tov ben Joseph ibn, 12, 74, Guttmann, Julius, 15–16, 20, 24n6,
78n40, 255–60, 262–3 25nn8–9, 29n29, 30n33, 31n47,
Farabi (Alfarabi; Farâbî; Fârâbî), 3, 5, 8, 52n1, 76n4, 107n17, 159–60, 162,
12–13, 17, 19, 64, 89, 106n13, 108n31, 176, 186n3
108n38, 157–66, 167n3, 167n8,
168n17, 168n23, 169nn32–3, 183, 215, Hallo, William W., 67n8
255–65 Hamann, Johann Georg, 172, 283nxxx
Farmer, Henry George, 257n16 Hayamson, Moses, 77n23
Ferguson, Adam, 203, 272n8, 283–5 Hegel, G. W. F., 25n7, 30n36, 96–7, 101,
Feuchtwanger, Ludwig, 14, 23n2 105n10, 211n23, 220, 238, 291,
Fichte, Johann Gottlieb, 129n21 295n16
Finkelscherer, Herbert, 169n30, 215, Heidegger, Martin, 1, 4–5, 6, 10, 17, 19,
267–8 36–7, 42–3, 54n17, 54n19, 54n22,
Fittbogen, Gottfried, 276niii 57–8, 61–2, 65, 80, 83, 87–8, 90–1,
Fradkin, Hillel, 55n26 93, 98–103, 105nn6–7, 105n10,
Frankel, Steven, 4, 33–56 106nn11–12, 106n15, 107n20,
Freud, Sigmund, 90, 108n35, 188n35 107n23, 107n25, 107n27, 108n31,
Fries, Jacob F., 188n39 108n42, 110n70, 111nn71–3, 112n74,
112nn79–80, 112n83, 113n86, 132,
Gadamer, Hans-Georg, 14, 16, 24n6, 67n2, 146–7, 148n6, 156n28, 168n28, 192,
303n69 227n7, 230nn18–19
Gaiser, Konrad, 210n15 Herakleitus, 32n50, 288, 292
Galli, Barbara, 67n8 Herder, Johann Gottfried, 172
Gebhardt, Carl, 299n44 Hertzberg, Arthur, 238n6
Gerard of Cremona, 258n21 Herzl, Theodor, 38, 240
Gershom, Levi ben (Gersonides), 15, 25n7, Hesiod, 2, 32n50, 88, 288, 292
26n14, 62, 160, 167n13, 270 Hindenburg, Paul von, 37
Gildin, Hilail, 56n37 Hippodamus, 289n17
Gilson, Étienne, 14, 285n112, 299n49 Hirsch, Samson Raphael, 239
Glatzer, Nahum N., 238n9, 240n19 Hobbes, Thomas, 2–3, 6–8, 13–14, 19,
God, 1, 5, 8–9, 21, 30n36, 35, 42, 46–51, 22, 24n3, 24n6, 25n7, 26n11, 26n14,
53n14, 54n15, 62, 70, 72–3, 75, 79–80, 27nn22–5, 55n25, 55n32, 58–9, 65,
84–6, 88–9, 92–4, 97, 102–3, 106n13, 80, 86, 89, 94–5, 101, 104n2, 106n15,
110n66, 111n72, 112n82, 131, 134–7, 107n18, 108n32, 110n66, 112n81, 116,
139–44, 146, 148n10, 149nn11–12, 118–27, 128n9, 128n19, 129nn21–2,
151nn16–17, 154n21, 159, 161–4, 131–56, 174–5, 184, 189, 203, 208–9,
173–8, 180–1, 188n39, 189n60, 226, 213, 285–7, 296, 300
231–4, 238, 240, 248, 268 Holmes, Stephen, 186n6
god; gods, 8, 10–11, 41, 81–2, 89–90, 99, Homer, 2, 288, 294
106n13, 108n36, 146, 148n9, 152n20, Husik, Isaac, 76n4
154n22, 155n27, 165, 177, 198, 206, Husserl, Edmund, 1, 57, 291
212n27, 248, 280, 288n12
Goethe, Johann Wolfgang von, 37, 188n14, Ibn al-Qifti, 169n34, 257n18,
238n9 260–2
Gogarten, Friedrich, 30n41, 57 Ibn Rushd. Seee Averroes
Gordon, Peter Eli, 111n73 Ibn Sina. Seee Avicenna
Gorgias, 212n34. See also Plato, Gorgias Ibn Tibbon, Samuel, 166, 264n53
Green, Kenneth Hart, 215n5, 270n4, 271 irrationalism, 9, 59, 172, 183
312 INDEX

Islam, 10, 14, 18, 25n9, 27n25, 28n27, Kojève, Alexandre,


l d 14, 20, 193, 201n5,
31n43, 59, 64, 87, 131, 160–2, 165–6, 302n61
168n16, 168n24, 183, 186, 190n76, Koyré, Alexandre, 14
203, 214n50, 249, 258, 264nn51–2, Kraemer, Joel L., 168n19, 264n53
265, 269, 293n6 Kraus, Pamela, and Frank Hunt, 55n30
Kraus, Paul, 14, 28n25, 29n29, 258n21,
Jacobi, Friedrich Heinrich, 1, 3, 8–9, 12, 264n48
14, 21–2, 89, 171–86, 186n2, 187n11, Krüger, Gerhard, 14, 16, 24n6, 25n7, 30n34,
187n17, 188n39, 188n49, 189n65, 30nn40–1, 31nn42–3, 31nn45–6,
189n70, 203, 279nxvi, 282, 283nxxx, 52n1, 57–68, 106n15, 186n3
284–5, 298n38
Jaeger, Werner, 105n10 Laboulaye, Édouard, 298n43
Janssens, David, 8–9, 52n3, 127n5, 171–90 Lachterman, David R., 187n29
Jesus, 234 Lagarde, Paul de, 109n60
Jews; Judaism, 1, 4, 6–7, 9–12, 14–15, Last, I. H., 268
18–21, 24nn3–4, 24n6, 25n9, 26n14, law; lawgiver (see also divine law; moral law;
29nn28–9, 30n32, 31n43, 33, 34, natural law), 4, 8–9, 17–18, 23, 25,
36–41, 43–5, 47–50, 52, 53n9, 27n19, 31n48, 36, 41, 43–51, 54n24,
53nn11–12, 54n24, 55n28, 56n37, 59, 55n26, 57, 59, 62–3, 66, 68n11, 70–1,
62–4, 67n9, 69–70, 76nn4–5, 79–82, 74–5, 78n38, 80, 86, 89–90, 93–7,
86, 89, 103–4, 104n4, 105n9, 113, 159, 103–4, 107n19, 108n36, 109n61,
161–4, 166, 168n22, 176–7, 180–1, 123–4, 133, 140, 157–65, 167n10,
183, 186, 188n31, 188n37, 189n60, 176–8, 181, 183, 186, 187n28, 190n77,
190n76, 203, 209, 215, 225–6, 238–40, 194, 196–9, 201n7, 226, 233, 238,
242, 251, 258, 265, 269–70, 294n7 240–1, 251, 257n17, 258, 263–4, 268,
282, 289, 296–7, 304
Kant, Immanuel, 48, 57–8, 62, 66, 67n3, law of reason (see also natural law), 108n36,
76n10, 82, 96, 104, 109n59, 111n73, 165, 213n39
172–3, 176, 203, 219, 229, 276niii, 279 laws (see also Plato), 17, 28n27, 31n43, 64,
Kaplan, Simon, 76n3 82, 86, 106n13
Kauffmann, Clemens, 190n75 Leibniz, Gottfried Wilhelm, 31n46, 180–2,
Kennington, Richard, 55n30, 285n112, 185, 190, 203–4, 276–9, 284–5,
299n49 285n111, 286nxxxvii, 287, 292, 296,
Kerber, Hannes, 10–12, 32n50, 203–16, 298, 303
271–3, 275–89, 291–304 Lerner, Ralph, 24n5
Keyserlingk, Hermann Alexander Graf von, Lessing, Gotthold Ephraim, 9–10, 14, 22–3,
219, 249 31n47, 32n49, 171, 172, 179, 184–6,
Kierkegaard, Søren, 97 189n70, 190nn72–3, 190n76, 190n79,
Kiernan, Michael, 300n51 203–4, 209, 209n6, 272n7, 276–9,
Kiesel, Helmuth, 296n22 282–7, 289, 291–2, 296, 298, 303
Kimchi, David (Radak), 270 Levi ben Gershom. Seee Gersonides
Kinkel, Walter, 105n9 Liberalism, 4, 6–7, 33, 36–40, 43–4, 46–7,
Klein, Jacob, 14, 20–1, 24n6, 25n8, 80, 106n12, 110n70, 111, 116–20, 123,
28nn26–7, 29n29, 57–8, 86, 91, 102, 125–7, 129n24, 186n6, 191–2, 195,
105n10, 106n11, 107n18, 112n83, 200, 223
186n7, 188n38, 273n10, 287n7, Liebert, Arthur, 219
288nn8–9, 293n6, 301n57, 304n73, Lipsius, Justus, 300
304n76 Loader, Colin, 110n68
INDEX 313

Löwith, Karl, 14, 20, 24n6, 26n14, 30n32, Moldenhauer,


ld h Eva, andd Karll Markus
k
81, 87–9, 96, 100, 105n6, 106n15, Michel, 211n23, 291n2, 295n16
107nn21–9, 108n31, 109n49, 109n61, Molesworth, William, 296n25, 296n27
110n66, 111n72, 112n74, 112n79, Montesquieu, Charles-Louis de Secondat,
113n86, 302n59 Baron de, 213n42, 298–9
Lycurgus, 193–201 moral law, 18, 45, 48–9, 63
Moses; Mosaic, 47–50, 59, 63, 74, 78n38,
Machiavelli, Niccolò; Machiavellian, 2, 86, 165
27n25, 38, 47–9, 61, 63, 129n21, 132 Munk, Salomon, 107n16, 268
Mahdi, Muhsin S., 169n33, 256n12,
257n18, 259n22, 264nn50–2 Nadon, Christopher, 12n8, 167n1
Maimonides, Moses (Moses ben Maimon; natural right; right of nature, 47, 62–3, 124,
RMbM; Rambam; Maimuni), 3–5, 8, 147, 189n57
11–15, 17–23, 23n2, 24nn4–5, 27n25, natural science; physics, 63, 98, 131, 133,
28n27, 29nn28–9, 31n43, 44, 49, 136–8, 140, 154n23, 155n26, 159,
55n26, 55n30, 56n36, 63–4, 69–80, 167n11, 187n17, 234, 247
85–7, 89–92, 104, 106n13, 107n17, Naves, Raymond, 275n1
108n31, 157–69, 183, 186, 189n64, New Thinking, 6, 34, 41–2, 45, 60, 80, 101,
190n73, 190n76, 203–4, 209n4, 104, 111n72, 226n2
213n49, 214n50, 227n8, 248n30, Nietzsche, Friedrich, 1, 5, 17, 19, 22, 37, 41,
252n39, 257, 257n14, 257n17, 258n19, 54n18, 58, 61, 80–2, 88, 92–3, 97–8,
264, 268–70, 286n117, 287n3 102, 109nn48–9, 112n82, 115, 132,
Major, Rafael, 167n7 154n21, 192, 229, 234, 242–3, 251–2
Mannheim, Karl, 6, 11, 30n41, 80, 82–4, Niewohner, Friedrich, 190n76
98–101, 105n8, 110nn68–70, Novak, David, 53n9, 55n26, 56n37
111nn71–2, 215, 217–24, 233n32 Nyberg, Henryk Samuel, 15
Marx, Karl, 82, 106n12, 191–2, 229, 243
Maschler, Chanina, 298n37 Oehler, Klaus, 68n12
Meier, Heinrich, 4, 13–32, 67n1, 68n11, Onkelos, 270
106n11, 116, 127n8, 128nn10–14, orthodoxy, 5, 17, 20, 29n31, 34, 40, 41,
128n18, 128n20, 147n1, 168n20, 49–51, 54n23, 60, 68n10, 89, 103–4,
190n79, 211n24, 213n45, 216, 227n8, 115, 135, 149n11, 176, 169, 171, 173,
255n1, 273n10, 286nxxxvii, 292n3 176, 184, 240, 296; neo-orthodoxy, 5,
Melzer, Arthur M., 209n8 60, 176, 239n16
Mendelssohn, Moses, 9, 22, 31nn45–7, 62, 172, Otto, Rudolf, 21, 188n39
174, 179–90, 283nxxviii, 296n22, 298n38
Mendes-Flohr, Paul, 104n4 Palencia, Angel Gonzáles, 258n21
Meyer, Thomas, 53n4, 270n7 Pangle, Thomas L., 4–5, 52n1, 57–68,
Minkov, Svetozar, 7, 131, 147n2, 167n4 106n15, 107nn20–1, 132, 148n5
miracles, 5, 7–8, 43, 49–50, 54n24, 60–1, 67n8, Parens, Joshua, 8, 11, 55, 157–69, 269; and
74, 80, 92–4, 109n47, 132–9, 142, 148n6, Joseph C. Macfarland, 214n50
149n12, 150n13, 152, 176, 233, 294n6 Paul, St., 89
Mitford, William, 201n3 Petersen, Julius, and Waldemar von
modern philosophy, 2, 3, 7, 17, 19, 43, 46, Olshausen, 276nii. See also von
52, 57, 80, 89, 91–3, 112n83, 117, 127, Olshausen, Waldemar
175–6, 181–2, 187n10, 232–3, 246–8, Pines, Shlomo, 14, 24n5, 29n29, 76n8,
252, 275, 287 76n11, 77nn25–8, 77n30, 77nn32–3,
Mohammed, 63 78n35, 78nn37–9, 162, 167n12, 168n16
314 INDEX

Pinsker, Leo, 238 78n47, 92, 152n19, 156n28, 158–65,


Plato; Platonists, 1–2, 5, 7–8, 10–13, 167n10, 168n25, 214n50, 223, 234,
17–19, 21–3, 24n4, 25n9, 27nn24–5, 239, 257n17, 268n4
28n27, 30n41, 31n43, 32n50, 34–5, Protagoras. See Plato, Protagoras
52, 58, 63–7, 68n11, 69–78, 80–1, Pythagoras; Pythagorean, 10, 162, 205,
85–93, 95–9, 101–3, 106n13, 106n15, 210n12, 292n4
107nn19–20, 107n29, 108n38, 108n40,
108n42, 112n83, 124, 142–6, 153n20, Rackham H., 288n12, 302n66
154n23, 155n27, 159–62, 164–6, Radak (Rabbi David Kimchi), 270
167n8, 168n24, 173, 175, 181, 183, Rambam. Seee Maimonides, Moses
186, 188n44, 193–4, 201n8, 203–15, Rashi (Rabbi Solomon ben Isaac), 270
231–2, 234–5, 238n10, 239n14, 241, rationalism, modern and ancient, 1, 3, 5,
247–9, 252–3, 257–65, 270n6, 278–82, 8, 11, 17–20, 29n29, 34, 42, 58–60,
286–9, 291–2, 294, 301–4; Alcibiades I, 90, 93, 97, 115, 127, 131–3, 139, 165,
294n10; Apology of Socrates, 160, 280; 172–5, 177–9, 181, 184–5, 223, 233,
Crito, 82; Epistles (Letters), 212n25, 287, 269, 283
302; Euthyphro, 124, 129n23; Gorgias, Rawidowicz, Simon, 24
212n34, 252n40, 278; Laws, 17–18, Razi, Muhammad ibn Zakariya, 264n48
28, 31n43, 64, 68n11, 86, 106n13, Reimarus, Hermann Samuel, 190n76, 209
155n27, 164, 194, 214n50, 282nxxv, Reinhold, Karl, 172
298, 298n43, 304n73; Phaedo, 145, 160, revealed law, 41, 59, 63, 68n11, 104, 161,
181, 213n39, 281nxxiii, 282nxxiv; 177–8. See also divine law; revelation
Phaedrus, 32n50, 124, 129n23, 207, revelation, 4–6, 8–9, 17–20, 24n4, 25n9,
211nn20–1, 212n25, 212nn33–4, 28n27, 29n29, 31n43, 36–8, 40,
212n37, 287, 292, 302; Protagoras, 42–3, 49, 51–2, 54n24, 58–60, 63–5,
19, 27n22, 146, 154n22, 278nxiii; 68nn10–11, 75, 81, 86–7, 89–90, 93–4,
Republic, 4, 18, 21, 31n43, 32n50, 97, 99, 108n40, 127, 131–6, 143–4,
63–4, 74, 78nn44–5, 99, 164, 201n8, 146, 152, 154n22, 160–1, 163, 178,
231, 263n46, 281nxxiii, 282, 287, 180, 182, 214n50, 227, 232–3, 238,
288n8, 289n19, 292, 294n10, 304n73; 248, 253. See also divine law; revealed
Symposium, 58, 302n61; Theaetetus, law
278nxiii, 288nn8–9; Timaeus, 32n50, Riezler, Kurt, 56n39
287, 292, 302, 303n66 right; rights, 37, 180. See also divine right;
politics; political, 2, 4, 6–7, 10, 18–19, 21, natural right
30n36, 30n41, 36, 38–9, 44, 46, 48, Rolfe, John C., 289n14, 301n56
53n7, 58, 71–2, 74, 91, 95, 118–21, Rosenthal, Franz, and Richard Walzer,
123–5, 127, 134, 144, 161, 165, 174, 259n22
177, 191, 193, 195, 201, 215, 220, Rosenzweig, Franz, 3–6, 34, 36, 41, 43–5,
223–4, 240, 261n36, 289n14; political- 54, 57, 59–60, 67n8, 76n1, 80–1, 93,
theological, 96 97, 99–100, 102–4, 104n4, 105nn5–7,
Plutarch, 301n56 105n10, 106n15, 111n72, 112n74,
political philosophy, 2–5, 8–11, 13, 16–19, 112n79, 113n86, 176, 226, 238, 240
28n25, 64–5, 91–2, 125, 142, 145–6, Rousseau, Jean-Jacques, 10, 88, 95–6,
155n27, 158, 160, 162, 165–6, 109n59, 192–3, 201n7, 203, 247,
168nn24–5, 191–2, 269, 284–5, 298n39
270n6, 291n1 Ruderman, Richard S., 1–12, 191–201
probity, 20–1, 29n31, 42–3, 51, 54nn23–4,
99, 178 Safranski, Rüdiger, 111n71
prophet; prophetology, 15, 18, 24n4, 25n7, Savigny, Friedrich Carl von, 109n61
25n9, 28n27, 42, 49, 70–5, 78n38, Schaeder, Hans Heinrich, 15, 25n9
INDEX 315

Scheler, Max, 178 Steinschneider,


h d Moritz, 259
Schleiermacher, Friedrich, 10–11, 203–14, Stroumsa, Sarah, 168n19
279–82, 287, 291–2 Susmann, Margarete, 219, 249
Schmidt, Anna, 128n11, 217–65
Schmidt, James, 285nxxxiv Tacitus, 300
Schmitt, Carl, 3, 6–7, 12, 14, 16, 23n2, Tanguay, Daniel, 8, 12n8, 157–8, 167n1,
26nn10–11, 33, 79, 99, 106n12, 167n5, 167nn8–9
108n37, 110n70, 115–29, 131, 148n3, Tennemann, Wilhelm Gottlieb, 210n16,
186n5 211n23, 291n2
Schocken, Salman, 15; Schocken Verlag, Thales, 288n9
14–15, 21, 67n6, 113n86, 167n2, Themistius, 262n43
238n9, 257n13 theologico-political predicament;
Scholem, Gershom, 14–15, 21, 24n6, theologico-political problem, 6, 26n12,,
25nn7–8, 27n25, 29n29, 27n18, 27n25, 32n49, 34–6, 38, 40–1,
30n32, 30n35 46, 48, 51, 52n3, 53nn4–5, 55n32, 56,
Scholz, Heinrich, 188n49 61, 80, 100, 104n2, 106n11, 109n44,
Shell, Susan, 186n5 133, 168n20, 171, 177, 187n29,
Silverman, David W., 76n4 209n3, 211n24, 285n113, 286nxxxvi,
Simon, Ernst, 15, 24 286n117, 299n44. See also Meier;
Smith, Charles Forster, 294n8 political-theological; Spinoza
Smith, Steven B., 53n10, 55n32 theology; theological; theologians, 3–7,
Socrates, 10–11, 23, 41, 56n39, 58, 62, 16, 20, 32n49, 35–6, 38, 45, 48, 53n7,
64, 71–2, 82, 85, 87, 90, 92, 95, 101, 53n14, 56n34, 57, 58–60, 63–5, 70, 73,
108n30, 116, 145, 152n19, 155n27, 75, 80, 84, 99, 106n13, 111n72, 116,
160, 164, 181, 183, 185, 192, 195, 198– 123, 127, 128n11, 129n21, 129n23,
201, 207, 212n27, 212nn37–8, 231, 132, 134, 140, 144, 152, 162–3, 166,
234, 249, 264n48, 278, 280, 293n5, 180–2, 184, 186, 188n39, 203, 208,
301n57, 302–3m, 303n65, 303n70, 226n2, 261, 292n4, 298, 304
304n73 Thomas Aquinas; Thomism, 20, 29, 88, 164,
Sombart, Werner, 178 168, 213n46, 244, 261n36, 270n7
Sophists; sophistry; sophistication, 19, 22, Thrasymachus, 7. See also Plato, Republic
27, 31–2, 84, 152, 205–6, 210, 234, Thucydides, 56n39, 200, 211n21, 246,
279, 288, 292 294n8
Spengler, Oswald, 106n12, 191, 246 Timm, Hermann, 188n49
Spinoza, Baruch (Benedict), 3–5, 7–9, Tönnies, Ferdinand, 133, 151n15
11, 13, 16, 19, 22–3, 23n2, 26n14, transcendence, 2, 9, 43, 85, 173–7, 183
28n26, 29n31, 32n49, 33–56, 59–64, Trend, J. B., and H. Lowe, 270n4
67nn6–7, 68n10, 76n6, 76n9, 79–80, Troeltsch, Ernst, 178
86, 89–90, 93–4, 103, 104n2, 105n5,
106n14, 107n17, 108n33, 108n37, Udoff, Alan, 187n29
109n44, 109n47, 109n53, 115, 122, Usener, Hermann, 288
127n1, 131–6, 138–9, 148n3, 148n6,
149nn11–12, 159, 167n12, 168n15, Velkley, Richard, 54n17, 54n19
171–2, 174–9, 183–4, 186nn2–3, Vico, Giambattista, 109n61
186n8, 187nn27–9, 188n34, Virgil, 129n25
188nn40–3, 188n46, 189n70, 203–4, Voegelin, Eric, 108n39
208–9, 209nn3–4, 285, 285n113, Voltaire, 275
286nxxxvi, 286n117, 287, 292, 296, von Olshausen, Waldemar, 283nxxx,
298n38, 299 284nxxxi. See also Petersen, Julius, and
Spitzer,
p Moritz, 15 Waldemar von Olshausen
316 INDEX

Walzer, Richard, 168n17. See also


l Xenophon,
h 3, 10, 26n13, 32n50, 191–201,
Rosenthal, Franz, and 211n24, 212n27, 287, 294n10, 301n57,
Richard Walzer 304n75
Weber, Alfred, 98
Weber, Max, 98–9, 105n10, 178, 224, Yaffe, Martin D., 1–12, 53n6, 69–78, 79,
242, 251 209n3, 217–68, 285n113
Weimar, 36–40, 44, 79, Young, Julian, and Kenneth Haynes, 112n82
188n31
Weiss, Raymond, 56n36 Zank, Michael, 105n5, 105n9
Wirth, Louis, and Edward Shils, Zeller, Eduard, 203, 275–6, 291
217n1 Zionism, 4, 21, 30n37, 33, 36–40, 53n12,
Wolfson, Harry A., 55n33 80, 103, 176–7, 188n31, 188n35,
Wust,, Peter,, 219 188n37,, 238–40

You might also like