Germany The Long Road West Volume 1 17891933 9780199265978 - Compress
Germany The Long Road West Volume 1 17891933 9780199265978 - Compress
2
Heinrich August
Winkler
Translated by
ALEXANDER J. SAGER
3
Great Clarendon Street, Oxford OX2 6DP
Oxford University Press is a department of the University of
Oxford.
It furthers the University’s objective of excellence in research,
scholarship,
and education by publishing worldwide in
With offices in
4
Database right Oxford University Press (maker)
You must not circulate this book in any other binding or cover
and you must impose the same condition on any acquirer
5
ISBN 0–19–926597–6 978–0–19–926597–8
1 3 5 7 9 10 8 6 4 2
6
Preface
7
Contents
Abbreviations
Introduction
1. Legacy of a Millennium
Looking Ahead
Notes
Index
8
Abbreviations
9
HPB Historische-politische Blätter für das katholische
Deutschland
HZ Historische Zeitschrift
10
NZ National-Zeitung (Berlin)
VZ Volkszeitung (Berlin)
11
Introduction
12
which are still with us today.
This is a political history, but not of the traditional kind. Affairs
of high diplomacy are treated mostly in passing, battles virtually
ignored. Events play an important role, but less for their own sake
than for the significance ascribed to them by contemporaries and
by those who came later. I focus special attention on
interpretations of history, how they influenced people, and how
they informed political decision-making. Such interpretations were
and still are controversial, the objects of discourses. Accordingly,
my study is also a discourse history.
To draw is to omit, as the painter Max Liebermann once said. I
will omit many things and concentrate on what seems important to
me in the light of the central question. It goes without saying that a
different central question would elicit a different set of problems
and a different evaluation of facts and opinions.
Historical narratives require a vanishing point. Vanishing points
change in the course of time. The years 1933 and 1945 are the
vanishing points towards which histories of modern Germany have
been written after the Second World War. There is now a new
vanishing point—the year 1990. We will not reach it until the
second volume, which deals with the period between the ‘Third
Reich’ and the reunification of Germany. Nonetheless, the
vanishing point of ‘1990’ is already at work in the first volume,
which takes us to the downfall of the first German democracy,
known as the Weimar Republic, and to the threshold of the ‘Third
Reich’.
How it happened that Hitler came to power is still the most
13
important question of nineteenth- and twentieth-century German
history, if not of all German history. Ever since 1990, however, it
has been joined by a new question: why did the German question
find its answer in reunification? Or, in other words, why is there no
longer a German question after 1990, and why only after this
particular year?
In selecting the year 1990 as our latest vanishing point, we will
be investigating many interpretations German history has
undergone between 1945 and 1990. Since there is now a German
nation state once again (albeit not a ‘classical’ nation state, but
rather a ‘post-classical’ one, firmly integrated into Europe), German
history can no longer be understood as the refutation of a German
nation state, or indeed of the nation state as such. The first German
national state, which came into being in 1871, belongs not only to
the prehistory of 1933, but also of 1990. It bears within itself both
the causes of its failure in the ‘German catastrophe’ of the years
1933–45 and, at the same time, much that went into the founding
of the second German nation state. Here I will name only the key
words: rule of law, constitutional state, federal state, social welfare
state, general suffrage, and parliamentary culture. And another
point, only seldom mentioned: the Two-Plus-Four Treaty of 1990
was a confirmation of Bismarck’s ‘Little Germany solution’, at least
to the extent that the latter rejected the ‘Greater Germany solution’
of the German problem, the solution with Austria.
Towards the end of the first volume it will become clear that, by
the eve of Hitler’s accession to power, the German people were not
only weary of the democracy of 1918–19, but also dissatisfied with
the ‘Little German’ national state of 1871. Educated Germans were
14
fascinated with the idea of an empire that included Austria and
controlled central Europe, a polity that sought to be different and
more than the typical nation state. The origins of the myth of the
‘Reich’ lie deep in the German past. The first chapter of the first
volume of this German history from the end of the old empire to
the reunification begins with an enquiry into the medieval
antecedents of this myth. In the second volume, I will discuss the
question of what replaced the mythology of the empire after it
vanished along with the German Reich in 1945. Was it a particular
‘post-national’ idea of Europe? Was it, in other words, the idea of a
new German mission—the supersession of the nation and the
nation state, as an example to all of Europe?
In the prologue to his book Die Geschichten der romanischen und
germanischen Völker von 1494 bis 1535*, Leopold von Ranke wrote
that the historian should ‘merely tell how it really was’. After
Hitler, this kind of history probably can no longer be written. For
us, the question should be: Why did it happen the way it did? The
readers I have in mind for this and the second volume are not only
other historians, but all those who would like to know the answer
to this question.
I have recourse to historical sources as much as possible (and
that is not the only respect in which Ranke is not yet obsolete). I
consider narrative not the opposite of explanation, but rather its
commensurate form. My notes at the end of the volume contain, in
addition to source citations for the quotes, references to selected
secondary literature. I include more of these for the central
questions of this study, but nowhere is my goal the utopian one of
providing an exhaustive list. Footnotes, added for this translation,
15
explain specialized terms and give translations for German poems,
titles, and quotations.
16
1
Legacy of a Millennium
17
Reich seemed to many people no more than an empty shell, the
Austro-Prussian antinomy was the reason.
The Reich had always been a myth. Medieval writers expended
a great deal of effort attempting to prove that the Roman imperium
had never actually ceased to exist. To be sure, it had been divided
in 395 into eastern and western halves; yet even after 476, when
the western empire had collapsed in the tumults of the tribal
migrations, the eastern empire still endured with its capital
Constantinople, formerly Greek Byzantium. However, the claim of
the eastern basileus to be the Roman Emperor was recognized ever
less in the west, especially after the accession of a woman, the
empress-widow Irene, to the throne at the end of the eighth
century. Then, in 800, Pope Leo III placed the imperial crown on
the head of Charlemagne, the king of the Franks and Lombards and
protector of the ecclesiastic state created by his father, Pippin. The
Romans rejoiced. Thereafter the imperial mantle passed from the
Greeks to the Franks (or, as it later came to be said, to the
Germans).
This was, at any rate, how the proponents of the medieval
theory of the translatio imperii, the transfer or ‘translation’ of
Roman imperial power, viewed things. Indeed, they had every
reason to emphasize the continuity of the empire. Following early
Christian tradition, which drew on the prophet Daniel, they
considered the Roman empire the fourth and last of the great world
empires, preceded by those of Babylon, Medea-Persia, and
Macedonia. This succession also implied a shift in geopolitical
focus from east to west, from the orient to the occident, with the
imperium Romanum (from the fourth century AD onwards referred
18
to as the imperium Christianum) the westernmost of them all. It was
generally believed that the Antichrist would not arise as long as the
Roman empire endured. According to the Book of Revelation, in its
medieval interpretation based on St Jerome, the Antichrist was a
tyrant, false prophet, Jew, and chief of heretics. The New
Testament prophecy considered his rule to represent the end of
world history. Thereafter Christ would return, destroy the self-
proclaimed ‘god’, and initiate the Heavenly Kingdom. The Roman
empire was to play the role of the catechon, the power holding back
the enemy of Christ and thus delaying the end of the world. It was
not until the era of modern theological scholarship that the Pauline
authorship of 2 Thessalonians, the source text (chapter 2) of the
catechon idea, was called into question. It is now generally
considered a fictitious ascription.2
The theory of imperial continuity played an important role in
the coronation of the Saxon king Otto the Great by Pope John XII
in 962. This event represented—or was made to represent—not the
foundation of a new empire, but the return of the imperium
Romanum. To be sure, the Treaty of Verdun had divided the
Frankish empire in 843, and there had been no emperor in the west
for nearly four decades, ever since the murder of the last emperor
from the imperial Frankish nobility in 924. However, the East
Franks had had a decisive voice in the royal election of Otto’s
father, Heinrich I, in 919. Looking back on Otto’s coronation in 962
from his vantage point in the mid-twelfth century, Bishop Otto of
Freising, the uncle and counsellor of the Hohenstaufen emperor
Frederick Barbarossa, spoke of a ‘retranslation’: after the Franks
and the Lombards, the Empire of the Romans was now being
19
‘transferred back’ (retranslatum est) to the Germans (ad teutonicos)
or, as others saw it, to the Franks, who had before ‘let it slip,’ so to
speak.3
In Otto’s day there was, as yet, no mention of ‘die Deutschen’.
Latinized terms such as teutonici and teutones only became common
around the turn of the millennium. By this time, of course, they no
longer referred to the old Teutonic tribes, but to the
contemporaneous ‘Germans’: people who spoke the same language
(deutsch) despite various tribal ancestry. Thus it makes little sense
to speak of ‘the German nation’ much before 1000. The second half
of the eleventh century witnessed the politicization of the concept
deutsch. During the struggle over the investiture of imperial
bishoprics, the first large-scale conflict between secular and
ecclesiastical powers, Pope Gregory VII referred to his opponent,
Emperor Henry IV of the Salian Frankish dynasty, as rex
Teutonicorum, ‘king of the Germans’. He wanted to make clear that
a German king who had not been vetted and crowned by the pope
could be, at most, a ruler over his own people, but not the Roman
emperor. The term was intended as a humiliation, and as such it
was fully in keeping withGregory’s attitude at Canossa, the fortress
on the northern slopes of the Apennines in front of which, in
January 1077, the pope compelled the emperor to wait three days
in the habiliments of a penitent before lifting the papal interdiction
against him. North of the Alps, however, the term regnum
Teutonicorum(‘kingdom of the Germans’) soon came to be used in a
more positive manner, signalling the developing sense of
community and growing self-awareness of the Germans.4
20
Nonetheless, the German kings were not going to rest content
with a mere kingdom. The concept regnum Teutonicorum referred
only to the German part of their sovereign territories. It did not
apply to Burgundy, which had belonged to the Reich since 1034,
nor to the Italian lands under imperial control. The German kings
needed the imperial title in order to rule the Reich in its entirety
effectively. The terms imperium and imperator did not necessarily
imply a claim to suzerainty over polities that did not belong to the
Reich. However, the medieval emperors certainly did insist upon a
special dignitas and ceremonial pre-eminence over other occidental
kings. As long as they restricted themselves to this, they were not
challenged, not even in France and England. As protector of the
Christian church, the emperor merited a higher rank than other
sovereigns. But only in this capacity.5
During the reign of the Hohenstaufen dynasty, western
observers gained the impression that the German emperors were
indeed out for more that the prestige of the position of first among
equals. In 1160, at a church synod dominated by the imperial
episcopate, and thus in no way representative of the church as a
whole, Frederick I acknowledged as pope (or rather antipope) a
candidate whom, in the prior deliberations of the College of
Cardinals, only the minority fraction loyal to Frederick had
supported. One of the most famous churchmen of the age, John of
Salisbury, the bishop of Chartres, rose in protest: ‘Who is he that
subjugates the Universal Church to a particular church? Who has
appointed the Germans the judge of nations? [Quis Teutonicos
constituit iudices nationum?] Who has granted to such coarse and
violent folk the power to install a prince above the heads of
21
humanity?’6
The English critique from Chartres was a response to what we
might call Hohenstaufen ‘imperial ideology’, to use a modern
expression. In 1157, the chancery of Frederick I began to employ
the term Sacrum Imperium, ‘Holy Empire’. Hohenstaufen political
propagandists referred to the rulers of other kingdoms
condescendingly as ‘little kings’, reguli. Certain poems by the
Archpoet, a writer belonging to the circle of the imperial chancellor
Reinald von Dassel, and the Spiel vom Antichrist*, written in 1160 in
the monastery at Tegernsee, went so far as to thematize a German
‘world empire’. The anonymous author of the latter text considered
this pretension justified by the special vocation of the Germans
within the course of sacred history: as the nucleus of the people of
God, the Germans would be the last to resist the Antichrist, the
enemy of the fatherland.7
Such ideas played virtually no role in the practical politics of
Frederick I. Still, there is little doubt that his high-handed
brinkmanship vis-À-vis the papacy, indeed his entire Italian policy,
was unrealistic and catastrophic. And considering the achievements
and plans of his son, Emperor Henry VI (1190–7), one is perfectly
justified in talking about Hohenstaufen ‘world policy’. This prince,
having obtained a claim to rule Sicily through marriage, enforced it
with military power. He compelled the English king, Richard
Lionheart, whom he had taken prisoner after the latter’s return
from the Third Crusade, to receive England as an imperial fief. He
secured suzerainty over Armenia, Tunis, and Tripoli, obtained a
Hohenstaufen claim to the succession in Byzantium, and probably
22
contemplated the conquest of the whole eastern empire. He did not
succeed in subjecting France to imperial domination, but it is quite
possible that he would have extended his plans of conquest to the
west, once he had secured the east. In the event, his early death
consigns these matters to the realm of speculation. Had he lived
longer, he might also have realized yet another of his ambitious
goals: to establish the Reich as the hereditary princedom of the
Hohenstaufens. This question, too, must remain open.8
The short reign of Henry VI marks the turning point in the
history of the medieval Reich. Henry had tried to subjugate all of
Europe; now, after a seventeen-year interval plagued by competing
claims to the throne and civil war, it was left to the other European
powers to decide the issue of Henry’s succession by his son,
Frederick II. The dice fell in 1214 on the field of Bouvines in a
battle between the forces of English and French chivalry. The
military defeat of the English spelled the final political defeat of
their German ally, the Guelf emperor Otto IV, son of Frederick
Barbarossa’s inveterate enemy, Henry the Lion.
Frederick II of Hohenstaufen was crowned German king in 1215
at Aachen. The imperial coronation took place five years later in
Rome. But Frederick was much more a Sicilian than a German
prince. For Germany, the most important event of his rule was his
renunciation of royal authority in the German lands and his cession
of the rights to levy tolls and mint coins to the ecclesiastic and
secular princes. These statutes were codified in the Confoederatio
cum principibus ecclesiasticis (1220) and the Statutum in favorem
principum (1232). The latter document was directed primarily
against the cities and the urban bourgeoisie, who claimed
23
independence from feudal princes under the motto Stadtluft macht
frei, ‘city air liberates’.
The territorial princes had already gained considerably in power
during the struggle over lay investiture, during which they had
temporarily sided with the pope against the German king. They
now emerged as the true winners from the crisis of the high
medieval Reich. The documents of 1220 and 1232 consolidated the
development of Germany along the lines of the territorial state.
This development had actually begun in the previous century, not
so much through the transfer of royal privileges, but as a
consequence of the local princes’ efforts to settle and develop their
land and concentrate their power. This happened not only in the
old German territories of the west, south, and north, but also in the
regions east of the Elbe that had been ‘Germanized’ in the wake of
conquest, missionary activity among the Slavic peoples, and
colonization. In terms of general constitutional development, the
Battle of Bouvines, which had made Frederick II’s rule possible,
was merely one turning point among several.9
The battle had a greater impact on France and England. In
France, the defeat of the English and their ally Otto strengthened
the domestic position of the victorious French king. The hitherto
powerful vassals of the crown suffered a loss of influence vis-À-vis
a strongly centralizing monarchy. On the other side of the English
Channel the situation was much different. Magna Carta of 1215
forced the weakened English monarch to concede extensive rights
and privileges both to the nobility and to the bourgeoisie.
Furthermore, he had to submit monarchical authority to a certain
degree of review and control by a committee elected from among
24
the barons. These concessions laid the basis for the development of
England into a constitutional state.10
The later thirteenth century witnessed the zenith of medieval
imperial ideology, even though by this time the power of the
German emperors had long since begun to wane. In his influential
and widely read treatise Memoriale der prerogativa Romani imperii
(1289), Alexander von Roes, a canon lawyer from Cologne,
outlined the structure of what he considered a sensible and
necessary societal order in the following way: the Romans, being
the older people, ought to receive the office of the papacy
(sacerdotium) as their own; the imperial office (imperium) rightly
falls to the Germans or Franks (Germani vel Franci) as the younger
people; to the French or Gauls, noted for their mental acumen,
accrues the study of the sciences (studium).11 The author intended
this scheme of a division of labour among the nations as a
defensive manoeuvre against French attempts to establish a claim
to the emperorship. However, the idea found no resonance in
France. In a document for Philip the Fair from 1296, an anonymous
French lawyer repeated a claim that had already been made the
previous century on behalf of the French monarch: namely that in
his kingdom, he is emperor. ‘And because there was a king in
France before there was an emperor, the king may be counted the
worthier.’ This statement can almost be interpreted as a direct
response to the project of Alexander von Roes.12
In one area, however, there was agreement between the secular
rulers of the west, at least in theory: they all rejected the ‘papal
revolution’, to use the expression of Eugen Rosenstock-Huessey,
25
one of the last German universal historians of the twentieth
century, in his 1931 book about the revolutions in Europe. The
‘papal revolution’ found its manifesto in Pope Gregory VII’s 1075
bull Dictatus Papae. Gregory’s claim that the pope could depose the
emperor reversed only the practice of the emperors. His assertion
that only the pope could remove or relocate bishops, however, was
just as much a declaration of war against the kings of France and
England as against the emperor. If the pope had had his will in the
matter, the political system of all three lands would have collapsed,
since bishops not only exercised spiritual offices, but were also, in
personal union, the highest administrative officials of the crown. As
it happened, the curia achieved at best partial success. From the
early twelfth century onward (first in France, then in England, then
in Germany with the Concordat of Worms in 1122), bishops were
elected to their offices according to canon law, but in the presence
of the secular ruler, allowing the latter to continue to exert his
influence.
The Investiture Contest was only one stage in the struggle
between spiritual and secular authority. In 1302 Pope Boniface VIII
reasserted the curia’s stance that the papacy was above all other
rulers. The bull Unam Sanctam, directed at the French king Philip
the Fair, claimed that two swords lay in the hand of the pope, a
spiritual sword and a secular sword. Both were thus under the
power of the church, the only difference being that the spiritual
sword was wielded by the church, the secular sword for the
church.13
The developing nation states of France and England answered
the papal challenge by nationalizing their churches to a great
26
extent, beginning with a rigorous restriction of papal taxes from
church property. The Roman-German emperors could not go the
national route without calling into question their own universal
aspirations and provoking the German princes, many of whom
were striving to become ‘pope’ in their own territories by
consolidating their regional churches.14 The reaction of the
emperor’s ‘party’ to the secular pretensions of the church (and to
the instrumentalization of the church by France during the Avignon
papacy from 1309 to 1377) was initially ideological in nature. Two
literary advocates of Emperor Ludwig the Bavarian (1314–47), the
Italian political intellectual Marsilius of Padua and the English
Franciscan William of Ockham, argued that the transmission of the
Roman imperial mantle ‘from the Greeks to the Germans’ in the
year 800 derived from the will of the Roman people. Thus, they
placed a democratic doctrine against the curial interpretation of
the translatio imperii by means of the papal office, which had been
articulated ‘ex cathedra’ in 1202 by Innocent III in the bull
Venerabilem. Yet the reality of the Sacrum Imperium contradicted
the idea of the sovereignty of the people so radically that the
construct had little impact.15
The secularization of the church also provoked a response from
the German mystics, beginning with Eckhart (c.1250–1327). Unlike
the efforts of Ludwig’s political theorists, the ‘inward turn’ of the
mystics had a great impact on the development of the church in
Germany. In his 1929 book on the Sacrum Imperium, Alois Dempf
interpreted the struggle to deepen and enliven religious devotion in
Germany as the counterpart to the ‘political reformation’ in France
27
and England. Transforming ‘piety without priests into a widespread
pietistic movement’, German mysticism prepared the way for a
development of world-historical consequence: the Reformation. The
young Luther knew in what traditions he stood.16
The estrangement from Rome, implicit in mysticism but
restricted to the religious sphere, intensified to an early form of
German national consciousness in the course of the fifteenth
century. Both the emperor and the imperial estates could agree in
their rejection of the papacy’s financial demands. Gravamina
nationis Germanicae, the title of the document into which their
complaints were formally gathered (beginning c.1440), gives
expression to the consciousness of their commonality. The name
Römisches Reich Deutscher Nation was used for the first time in 1486
in an imperial law; the complete title Heiliges Römisches Reich
Deutscher Nation first appears in the recess of the Imperial Diet of
Cologne from 1512. The addendum ‘of the German Nation’ was not
at first intended to establish equivalence between the Roman
empire and Germany, but rather as a restriction indicating the
‘German territories’ of the Reich in contrast to the territories of the
welsch— that is, Italian—nation.
Yet even in the original spirit of the title, the German lands were
considered the core of the empire, and by means of this usage the
term ‘nation’ acquired a new significance. Whereas before it had
served as a practical means of distinguishing and organizing
national-ethnic groups at universities, church councils, and among
the foreign merchants in west European trading centres, in the
fifteenth century it broadened to a general way of looking at the
world. The defining factor in the German concept of ‘nation’ was
28
the commonality of language (Gezunge). It probably could not have
been otherwise, since, as we recall, the ‘German nation’ had no
political-administrative reality of any kind at this time. In France
and England, in contrast, the formation of the nation proceeded
from the monarchy, which lent a statist orientation to the idea of
the ‘nation’ that was not possible in Germany.
The emperor made use of the word ‘nation’ whenever he desired
the support of the imperial estates (Reichsstände) and their leaders,
the imperial electors, in matters of common interest. The seven
electors, however (the archbishops of Trier, Cologne, and Mainz,
the king of Bohemia, the count of the Rhenish Palatinate, the duke
of Saxony, and the count of Brandenburg) had reason to believe
that not everything the elected ‘Roman king’ considered necessary
was in the interest of the Reich and nation. The dynastic interests
of the Habsburgs, who had stood at the apex of the Reich since
1438, were by no means automatically the same as the interests of
the empire or the German nation.17
Conversely, the agenda of the electors—to the extent that they
could agree, which happened rarely enough—was also not
necessarily in harmony with the common good of the Reich. While
they were recognized as the co-bearers of imperial authority by the
Golden Bull, the foundational law of the Reich from 1356, the seven
electors did not alone constitute the ‘nation’. It also included the
princes and the other imperial estates, who had far less influence
on imperial policy and legislation, not to mention the cities, which
were most burdened by imperial taxation but had no voice in the
Imperial Diet (Reichstag) in the fifteenth century. Even at the
beginning of the 1400s, it was clear to many that the empire
29
needed to be thoroughly reorganized. The steps taken under
Emperors Frederick III (1440–93) and Maximilian I (1493–1519),
however, hardly merited the name of ‘imperial reform’. In
Germany, the real institutional development of the state took place
not at the level of the Reich, but in the territories, whose princes
increasingly took advantage of Roman law and employed the
services of well-educated officials trained in its application. The
princes of the larger territories were especially energetic in this
kind of regional reform and consolidation, and it was they who, in
their corresponding lack of interest in centralizing measures, stood
in the way of effective imperial reform.18
It was this rather dismal reality of the Reich that prompted the
German humanists, both before and after 1500, to call for the
restoration of the old imperial glory. The source of their hope and
inspiration was the remembrance of a distant past, the age of
Germania Magna, in which the nations of Germanic origin were still
undivided. Invoking Tacitus, whose Germania had been
rediscovered in 1455, the humanists drew an idealizing picture of
German virtue that contrasted advantageously with its distorted
counterpart: the Romans, who had long since lapsed into
decadence and debauchery. Republican Rome, not the Rome of the
Caesars, could teach the Germans the love of the fatherland. After
all, the Germans had inherited the succession to the Roman empire.
The greatness and dignity of the Reich derived from this legacy,
which was earned and legitimate. This belief did not, however,
prevent the humanists from appealing to Pope Innocent III’s bull
Venerabilem in order to prove that the empire had been transferred
from the Greeks to the Germans in the person of Charlemagne in
30
800. The aspirations arising from this view of things assumed, in
the writings of several authors, Hohenstaufen-like dimensions.19 In
his Narrenschiff*, Sebastian Brant, a member of the upper Rhenish
humanist circle, appeals to God to make the Roman empire so large
31
that came to pass too early for us to know about.20
32
from Europe the last great cultural harvest that Europe could have
gathered—the Renaissance.’ Cesare Borgia as pope: for Nietzsche,
this would have meant ‘victory’, the end of Christianity. Luther, on
the other hand, ‘this monk, with all of the failed priest’s instincts
for vengeance, rebelled in Rome against the Renaissance... And
Luther restored the Church. He attacked it... the Renaissance—an
event with no meaning, a grandiose futility.’23
The German Reformation was both: the liberation from
ecclesiastical coercion increasingly perceived as foreign rule and
the establishment of a new, interiorized, state-supporting regime of
coercion. Liberating and repressing at the same time, the
Reformation, as Marx noted, could only partially supersede the
Middle Ages. Engels was fundamentally in error when he called it
the ‘revolution no. 1 of the bourgeoisie’.24 In terms of social
history, especially in Switzerland and upper and middle Germany,
the Reformation represented the uprising of the ‘common man’ in
the country and city. The Peasants’ War of 1524–5 was the
culmination of this movement.25
In its political consequences, the Reformation is best described
as what Rosenstock-Huessey called a ‘princes’ revolution’:
The grand hierarchy of the Church Visible has lost its pathos. The
human soul is no longer where the clergy seeks it. The educational
efforts of the church can thus confidently be left to the bishops of
each locality and territory, and this bishop is the worldly authority.
Luther’s imperial elector replaces the supreme bishop ... Probably
in no other country in the world did two such different fields of
vision overlap as they did here. Above, prince and statesman fight
33
for their right and freedom of authority. Below, burghers and
peasants live and learn the pure doctrine as well as obedience to
authority within the narrow circle of their bondsman’s
understanding... the ‘unpolitical’ nature of the average German is
already implicit in the voluntary division of labour between Luther
and his sovereign.26
34
pastoral communities and reforming schools and universities.
In all their efforts, Luther and the Lutherans needed the support
of the secular authorities, who had been ordained by God and who
wielded the sword and rod to punish the wicked and protect the
devout. Many princes had not merely a religious but also a material
interest in promoting the new faith. By means of the Reformation,
they gained access to church property, thereby increasing their
government revenue and reinforcing their lordship. For Luther, the
efforts the states and cities directed toward providing a lawful
order for the new faith represented a labour of love. In making
Luther’s cause their own, secular rulers could, but were not
obligated to, adopt his rationale.27
The development of the Evangelical church system in the
territories began in 1527 with the visitations of Luther’s territorial
sovereign, Elector John of Saxony, to church communities and
schools. The other princes who professed the new faith soon
followed the example of the Saxon elector. The result was a
coercive ecclesiastical system in which, in the words of the
Evangelical theologian and philosopher of religion Ernst Troeltsch,
the human element, hitherto of secondary importance, became the
main focus:
35
legal consequences for spiritual punishments and measures. In
theory, the community was ruled by Christ and Holy Scripture; in
practice, it was ruled by the territorial sovereigns and the
theologians.28
36
argued that the Lutheran movement articulated in terms of dogma
a number of differences between the eastern church and the
western church that were only latent in the antagonism between
the two great ecclesiastical systems.
37
emperors and kings in a struggle that lasted centuries. Wherever
this separation was maintained, or could establish itself again, an
environment conducive to ideas of freedom was fostered. The
Anglican state church, introduced in England by Henry VIII in
1534, was from the beginning closely linked to the estates; in the
second half of the seventeenth century it was parliamentarized and
finally, in the nineteenth century, liberalized. In Germany, on the
other hand, the summepiscopate had, until 1918, an authoritarian
and governmental or—to put it in somewhat exaggerated terms—
caesaropapist orientation. Politically speaking, German
Lutheranism represented a step backward.
‘Lutheranism purchased spiritual liberation at the price of
earthly subjection.’ Borkenau’s verdict summarizes the
contradictory legacy of Martin Luther’s Reformation. Both sides,
the cultural and the political, must be seen in context.
38
brutal kind of megalomania.31
39
him.32
The Reformation forms one of the deepest epochal divisions in
German history. In the preface to his Deutsche Geschichte im Zeitalter
der Reformation* (1839), Leopold von Ranke spoke of ‘the most
important event of our fatherland’. The Reformation did not simply
divide the German nation; in certain respects it structured it anew.
The emperor himself, belonging as he did to one of the competing
religious parties, was less able to represent the whole of the Reich.
Not even the curia of the electors could claim to stand for the
whole, since it, too, was split into confessional factions. The same
was true of the Imperial Diet. The religious parties themselves,
however, were supra-territorial, indeed ‘national’ associations.
From the point of view of the followers of the new faith, the
German nation was the entity that came closest to embodying the
ideal entirety of the Protestant imperial estates, most—but not all
—of which had joined together in the Thuringian town of
Schmalkalden in 1531, forming a defensive alliance against the
ecclesiastical-political aspirations of Charles V. The common bond
was not political, but rather cultural: the Evangelical faith in the
sense of the ‘Augsburg Confession’, as it had been worked out in
doctrinally binding terms by Philip Melanchthon in consultation
with Luther (and not always to the satisfaction of the latter) at the
Imperial Diet of Augsburg in 1530. This was the first step toward
the ‘confessionalization’ of the new faith.
One of the most important factors for the spirit of solidarity and
community among German Protestants was Luther’s Bible
translation. This text created the supra-regional German standard
40
language, which in turn became the most important ‘national’
medium of communication and, as such, the prerequisite for the
possibility that even two centuries later, when there was still no
unified German state, educated Germans could consider themselves
members of a German cultural nation. The Evangelical segment
played such a crucial role in the formation of German national
identity that one must speak of a Protestant cultural hegemony. For
all that, however, Luther’s Volk remained surprisingly mute, as
Eugen Rosenstock-Huessey remarked. ‘The nation he awoke to life
became a nation of princes, professors, and pastors, and remained
so for a long time, all the way to the professors’ parliament at the
Paulskirche in 1848. This role of the German universities in
shaping the German nation develops in the fifteenth century.’33
Catholic Germany found little succour in the support of the
emperor. The wars between Charles V and Francis I of France, as
well as the Turkish danger, forced the emperor and the Catholic
estates repeatedly to compromise with the supporters of the new
faith, delaying the ultimate confrontation. This happened for the
first time at the Imperial Diet held at Speyer in 1526, which left it
for each imperial estate to decide whether or not to follow the
Edict of Worms, the ostracism of Luther and the condemnation of
his teachings in 1521. The second instance was the ‘Nuremberg
Standstill’ of 1532, which allowed Protestants the free exercise of
their religion for the time being. Only after the year of Luther’s
death in 1546, when Charles was no longer hampered by
involvement in foreign wars and had brought one of the most
important Protestant princes, Duke Moritz of Saxony, over to his
side (by promising him the electoral dignity, which was in the
41
hands of his cousin from another branch of the Saxon house, John
Frederick), could the emperor dare to strike against the
Schmalkaldic League.
However, the emperor did not manage to convert his military
triumph in the Schmalkaldic War (1546–7) into a political success,
since Moritz of Saxony, now an elector, allied himself to the
princely opposition, made common cause with the French king
Henry II(to whom he conceded the imperial curacy over Metz,
Toul, and Verdun), and took up arms against Charles V. The Treaty
of Passau, which the emperor found himself forced to conclude in
1552, again granted the Protestants freedom of worship until the
convention of a new Imperial Diet. This assembly, held in 1555 in
Augsburg, marks both the end of the age of the Reformation and
the conclusion of the struggle over imperial reform. From this
moment forward, all parties acknowledged the principle cuius regio,
eius religio (‘He who rules the land determines the religion’),
formulated later by a jurist. The Augsburg Confession (not the
‘reformed’ confession of Calvin and of Ulrich Zwingli, the popular
priest of Zurich) was now legally recognized in the Reich. The
Religious Peace of Augsburg did not consider the individual
believer; only the prince had the right to decide between the old
and new faiths. Dissidents had only the right to leave the country.
Imperial cities with mixed confessions were to be guided by the
principle of parity, and in ecclesiastically ruled territories a
controversial ‘clerical reservation’ was to apply: a bishop or abbot
of the Reich who converted from Catholicism to the Lutheran faith
was supposed to relinquish his office immediately; the cathedral or
cathedral chapter had the right to elect a Catholic successor. But
42
the Religious Peace of Augsburg did not compel this to be done.
The recess of the Augsburg Diet spelled the failure of two
competing solutions for a universal resolution to the crisis: on the
one hand, the restoration of Reich and church through the emperor
and in terms of the old faith, and, on the other hand, the
restoration of a unified church in terms of the new faith. The
possibility of a Reich ruled by the estates (Stände), a project
pursued in the early sixteenth century by a circle of reformers
under the patronage of Berthold of Henneberg, the elector of Mainz
and arch-chancellor of the Reich, also ultimately came to nothing.
In 1555, as Heinz Schilling wrote, the ‘estates and the crown finally
agreed that, in Germany, the princes and their territories would be
the bearers of a new conception of the state and that the Reich
would remain a pre-governmental, political union.’34
The Holy Roman empire was thus preserved. In fact, its
institutions grew even stronger. A great civil war, a war among the
princes, which as things stood would perforce have become a pan-
European war, was once again avoided. The compromise of 1555
sanctioned the right of the German territorial states to religious
particularity, though it did not yet permit the final consequence of
the increase in territorial liberty, namely full political sovereignty.
Yet even in 1555 it was clear that the government of princes would
have a greater chance than the emperor of claiming the loyalties of
the subjects in the event of conflict. Even if the Reich could offer
an ultimate organizational support and repeatedly managed,
especially in the face of external threats, to call forth waves of
‘imperial patriotism’, nonetheless a much stronger sense of
allegiance and solidarity developed at the territorial level. The idea
43
of a German nation remained an alternative empire, an empire of
faith and spirit, and one that required no emperor.
The Religious Peace of Augsburg was one of the main reasons
Protestant Germany was able to shield itself for over half a century
from the effects of the Catholic Counter-Reformation, which began
at the Council of Trent (1545–63). Countries in which the Counter-
Reformation triumphed were economically and intellectually so
lastingly set back that the effects persist to the present day. Spain,
at one time the great Catholic power due to its colonial possessions
in Latin America, succumbed to England, its most dangerous
enemy, in the second half of the sixteenth century. The defeat was
not merely military, that is, the destruction of the Armada in 1588.
Within a few decades, the Calvinist-spirited maritime trading
power of England had also economically eclipsed the countries of
the Iberian peninsula, Spain and Portugal. France, on the other
hand, successfully resisted the Counter-Reformation, remaining
true to its national ecclesiastical tradition after the bloody
Huguenot Wars of 1562–98. The conversion of King Henry IV (of
Navarre) from the Calvinist to the Catholic faith in 1593 was not
the prelude to a comprehensive re-Catholization of France; rather,
it ushered in the policy of religious tolerance that found its classic
expression in the Edict of Nantes of 1598.
Wherever it was rigorously implemented, the Counter-
Reformation almost completely destroyed any structures of modern
capitalism that had existed in the regions that remained Catholic.
In comparison, the Calvinist spirit of terrestrial asceticism and
merit promoted a dynamic spirit of enterprise in those lands where
it prevailed. An impulse of this kind was lacking in Lutheran
44
regions, which persisted in the traditional corporative order and its
correspondingly conservative economic practices. The principles of
traditional economic life were based not upon notions of individual
risk, enterprise, and continually increasing profit margins, but upon
the satisfaction of habitual, class-oriented needs and a just price. In
this respect, Lutherans differed less from the Catholics than from
the Calvinists.35
In questions of internal political order in Germany, the
confessional differences were rather minor. The governments of the
Calvinistic territories were as authoritarian as those of the Lutheran
and Catholic lands. However, the fact that the Religious Peace of
Augsburg did not recognize the Calvinists as a confession meant
that German Calvinism as a whole differed markedly in its political
profile from that of English or Dutch Calvinism. It was thus no
coincidence that the impulse to alter the confessional status quo for
the benefit of the Evangelicals in The Confessional Divide the first
decades after 1555 proceeded not from a Lutheran, but from a
Calvinist princedom, the Palatine electorate. The Palatinate thus
advanced to the position of adversary against the most energetic of
the Catholic imperial estates, Bavaria, where the Counter-
Reformation held sway.
The Calvinists began to gain ground when they were joined by a
few smaller territories and one larger one, Hesse-Kassel, in the
course of the so-called ‘Second Reformation’. The balance of power
in Protestant Germany shifted even more significantly in 1613,
when the elector of Brandenburg, John Sigismund, converted from
the Lutheran to the Reformed confession. This was an event with
long-term consequences to which we shall devote more attention
45
below. The Catholic imperial estates rallied around the duke of
Bavaria, the Protestant estates around the elector of the Palatinate.
A Protestant defensive alliance, the ‘Union’, was founded in 1608,
in the next year the counter-alliance of the ‘Catholic League’. The
opposition of the confessions could turn into a new military conflict
at any time. When Emperor Matthias prepared to rescind the
religious freedom his brother and predecessor, Rudolf II, had
granted to the primarily Protestant estates of Bohemia and
Moravia, that time had come. The Thirty Years War began in
Prague in May of 1618.36
This war was never simply a religious war, whether in its first
phases, the ‘Bohemian-Palatine War’ of 1618–23 and the ‘Danish-
Lower Saxon War’ of 1625–9, or later during the ‘Swedish War’ of
1630–5 or the ‘Swedish-French War’ of 1635–48. The marked
participation of foreign powers aroused a imperial-patriotic
reaction of protest in the empire, a reaction directed at first
primarily against Habsburg-ruled Spain, the military ally of the
Habsburg emperor, Ferdinand II, and the Catholic League. A
pamphlet of 1620, invoking the old doctrine of the four world
empires, charged the Spanish with seeking to erect a fifth
monarchy, one that would cover the earth. This was an offence
against the divine order, according to which the Holy Roman
Empire of the German Nation was the fourth and last of the world
empires. Imperial patriotism reached its high point in 1635 at the
time of the Peace of Prague, which the emperor concluded first
with electoral Saxony, then with most of the other Protestant
estates, primarily in northern Germany. Yet the attempt to establish
a ‘German peace’ came to nothing. Already that same year,
46
Catholic France intervened in the conflict on the side of Lutheran
Sweden, that is, on the confessionally ‘wrong’ side. Power politics
won a decisive victory over the religious controversy.37
The Thirty Years War lived on in the collective memory of the
Germans for a long time. It was considered the national
catastrophe, a standing that was not contested until the period of
the two world wars of the twentieth century, especially the second.
It was a catastrophe primarily in its demographic, economic, social,
and moral consequences. Large areas of Germany would not
recover from its three decades of murder and pillage until the
following century; some took even longer, others never recovered
at all. The peasants were impoverished; in the east of Germany,
many of them fell into subjection under the manorial lords. The
destruction of countless cities ended the rise of the bourgeoisie for
a long time to come. From a sociological point of view, the victors
of the war were the landed aristocracy; the part of the nobility
close to the government; state-supported merchants, entrepreneurs,
and bankers; the military, and the governmental officials—all
pillars of the emerging absolutist system. The horrors of war, mass
mortality, and deprivation prompted the survivors to retreat into
inwardness and brought about a renewal of lay piety that prepared
the way for the Pietist movement of the late seventeenth and the
eighteenth centuries in Evangelical Germany.
If one can speak of any positive effect of the Thirty Years War, it
would be an insight into the absolute necessity of religious
tolerance—a tolerance that could only be enforced by a strong state
willing to secularize itself within certain limits and thereby to
become neutral in religious matters. Not least among the causes of
47
princely absolutism was an absolutist attitude in matters of faith:
the subjects paid for their greater freedom in the inner realm with
an increase in political subjection to the secular authorities. The
latter found the most dependable support for their power in the
profound, traumatic fear that must be considered the lasting effect
of the Thirty Years War: the fear of chaos and the collapse of
societal order, of foreign bands of soldiers, of civil and fratricidal
war—the fear of the Apocalypse.38
The Peace of Westphalia of 1648 (concluded in Münster
between the Reich and France, in Osnabrück between the Reich
and Sweden) restored the Religious Peace of Augsburg of 1555,
extending it to those of the Reformed confession, which was now
granted equal rights as a Protestant observance. The geographic
boundaries between the confessions, as well as their membership,
were fixed according to the state of things in the year 1624. From
this date forward, subjects were no longer required to convert to a
new confession in the event that their ruling sovereign decided to
do so. The northern part of the Netherlands and the Swiss
confederates left the Reich for good. France took over the sovereign
jurisdiction of Alsace from Austria, definitively weakening the
position of the Habsburgs in Germany. The treaty confirmed
Bavaria’s entitlement to the Palatine electorate (acquired in 1623)
and the Upper Palatinate. A new, eighth electoral title was created
for the Rhenish Palatinate. The peace treaty recognized the
imperial estates’ right to co-determination in all affairs of the
Reich, full territorial sovereignty in secular and religious matters,
and the right to enter into alliances with foreign powers—restricted
only by a clause forbidding such alliances to be directed against the
48
emperor or the Reich, a proviso difficult to enforce. In order to
prevent the formation of an adversarial quorum, the Evangelical
and Catholic estates formed separate deliberative bodies (the
Corpus evangelicorum and Corpus catholicorum) at the Imperial Diet,
which in 1663 was transformed into a permanent congress of
delegates convening in Regensburg. Decisions could be made only
when this itio in partes reached an agreement.39
On the international stage, France and Sweden emerged as
victors from the Thirty Years War. Both countries guaranteed the
peace treaty, which was declared the basic law of the Reich. Both
were able to extend their territory at the expense of the Reich;
Sweden, in fact, which acquired West Pomerania and Rügen among
The Austro-Prussian Antagonism other territories, advanced to the
status of an imperial estate (Reichsstand). Inside Germany, the
winners were the imperial estates. In consequence of the Peace of
Westphalia, they were able to take the decisive step to full
sovereignty. After 1648, the Holy Roman Empire was no longer a
factor in European power politics. Because it helped to stabilize the
status quo, its preservation lay in the interest both of the European
powers and the smaller imperial estates. As an unwieldy, archaic,
and outmoded structure, however, the Reich bore no comparison
with the states of France or England, Spain or Sweden. It was the
‘irregular and monster-like body’ (irregluare aliquod corpus et
monstro simile) described in 1667 by Samuel Pufendorf in his
famous work about the constitution of the German empire.40
49
We turn now to the third of the basic factors shaping German
history, the opposition between Austria and Prussia. That Austria
and Prussia, the two most important states to emerge from the
medieval German colonization of middle and eastern Europe, were
able to advance to the status of European great powers was due to
the fact that both of them included significant territories not
belonging to the Reich. The Austrian Habsburgs, who had assumed
the emperorship for the first time in 1273 and held it from 1438
onwards without interruption, acquired the Netherlands and the
Free Margravate of Burgundy through marriage in the second half
of the fifteenth century. The lordship over Spain and its border
territories, along with Naples and the American colonies, were
added at the beginning of the sixteenth century. The dynastic link
with Spain was preserved after the abdication of Charles V in 1556,
when the title of Emperor was transferred to his brother, Ferdinand
I, the lordship over Spain, the Free Margravate of Burgundy, and
the Netherlands to his son, King Philip II.
In the long view, the acquisition of the Bohemian and
Hungarian crowns in 1526 was even more important. To be sure,
under Ferdinand I most of Hungary was lost to the Turks, who had
continued to advance through the Balkan peninsula after their
conquest of Constantinople (and thus the Byzantine Empire) in
1453. Nonetheless, the Turkish threat was also ideologically and
materially profitable for the Austrians. After the first Turkish siege
of Vienna in 1529, Austria could claim to be defending the Reich—
indeed, the whole of the occident—against the Ottomans and
Islam. After the second siege of Vienna in 1683, the tide turned:
Hungary was taken from the Turks, falling in 1699 to the House of
50
Habsburg along with Transylvania and large parts of Slovenia and
Croatia. Austria had become a great power.
It was able to maintain this status even after the outbreak of the
War of the Spanish Succession in 1700, occasioned by the
extinction of the Spanish line of the Habsburgs. Although Louis XIV
of France managed to break open the Habsburg encirclement and
place his grandson, Philip of Anjou, on the Spanish throne, the
Peace of Utrecht (1713) precluded the unification of Spain with
France. In the same year, Emperor Charles VI determined his own
succession in the Pragmatic Sanction: his oldest daughter, Maria
Theresa, was to inherit the Habsburg possessions undivided. After
long and difficult negotiations, the emperor’s resolution was
eventually recognized by the Habsburg hereditary lands, Hungary,
and finally by the other great powers of Europe. In October 1740,
Maria Theresa assumed the throne—almost at the same time as her
great rival, King Frederick II, came to power in Prussia.
‘Prussia’ (Preussen or Pruzzen) was originally the name of a
Baltic tribe that lived in the territory of later East Prussia during
the early and high Middle Ages. German colonization of this area
began under the auspices of the Teutonic Order, whose help the
Polish duke Conrad of Masovia solicited in 1225 against the
heathen Prussians. After its military decline in the fifteenth
century, the Teutonic Order was able to maintain only a part of its
territory as a Polish fief. Polish sovereignty was preserved when
Albert of Brandenburg, Grand Master of the Teutonic Order and
follower of Luther, negotiated with Poland to transform the
Teutonic state into the Grand Duchy of Prussia, which occurred in
1525. The duchy was bequeathed in 1618 to Brandenburg, where
51
the Hohenzollerns had been ruling as margraves and electors since
1415. In 1660 the Great Elector, Frederick William (1640–88),
successfully asserted his right to the sovereignty of the Duchy of
Prussia, which nonetheless remained outside the Holy Roman
Empire. On 18 January 1701, Frederick III, son of the Great
Elector, had himself crowned Frederick I, ‘King in Prussia’, in
Königsberg with the assent of Emperor Leopold I. This did not yet
make Brandenburg-Prussia into a great power, but it was a
significant step along the way to this goal.
Unlike those of the Habsburgs, the subjects of the Hohenzollerns
were almost exclusively German-speaking (the Masurians, who
spoke a Polish dialect, were Evangelical, and for this reason alone
did not consider themselves Poles). The other major difference
between Austria and Prussia was confessional: the Habsburgs and
most of their subjects were Catholic, the Hohenzollerns Protestant.
The rulers of Brandenburg, however, represented a special case. In
1613 Elector John Sigismund converted from the Lutheran to the
Reformed confession. Since his subjects (with the exception of the
Calvinists and the Catholics on the lower Rhine) remained
Lutheran, the Hohenzollerns were compelled to adopt a tolerant
attitude in questions of religious observance.
But that was not the only effect of the elector’s conversion. In
his Geschichte der preussischen Politik*, Johann Gustav Droysen
noted in 1870 that while John Sigismund’s conversion to Calvinism
was certainly sincere, his new confession was ‘not only of religious
significance’.41 Otto Hintze went even further in his 1931 essay
‘Kalvinismus und Staatsräson in Brandenburg zu Beginn des 17.
52
Jahrhunderts’.† Taking up ideas from Max Weber’s famous study of
the links between Calvinism and the ‘spirit of capitalism’, Hintze
enquired about the ‘elective affinity’ between Calvinism and
Staatsräson, ‘reasons of state’. His answer was that Calvinism was
‘the midwife that brought the Staatsräson of Brandenburg politics
into the world’. It served as a ‘bridge, over which the Staatsräson of
western Europe was introduced into Brandenburg’, the long-term
effects of which was ‘a basic political orientation towards the west’,
primarily towards the Netherlands and France.
The result was paradoxical. According to Hintze, Calvinism
served
53
motivations of his predecessors had been secularized, transformed,
as it were, into a worldly form, without being merely derivative of,
or completely explainable in terms of, the rationalism of the
Enlightenment.42
54
criticism on points of detail. Nonetheless, they place the
paradoxical phenomenon of Prussia in the greater historical context
in which we must view it. With regard to Brandenburg’s
pretensions to the succession of the extinct ducal house of Jülich
and of the Duchy of Prussia, Hintze himself spoke of the ‘fateful
question’ of whether Brandenburg should orient itself toward the
west or toward the east. ‘In the east, it was better served by
Lutheranism, in the west by Calvinism.’ The conversion of 1613
can be interpreted as a decision for the west, but only in terms of
the greater European political context and the organization of the
government. The sovereign’s Calvinist turn facilitated the
integration of the new, mostly Reformed estates and subjects in
Cleves, Mark, and Ravensburg, the territories of the Duchy of
Jülich granted to Brandenburg by the Treaty of Xanten in 1614. In
the Margravate of Brandenburg and the Duchy of Prussia, in
comparison, the loyalty of the estates and subjects could only be
assured at the price of abandoning any denominational coercion.
An attempt to introduce the Reformed confession into these regions
would have failed, in Hintze’s words, ‘by dint of the socio-
economic structure of these East Elbian territories with their
manorial estates. Lutheranism was well-suited to this structure,
Calvinism not in the slightest’.44
Thus the east–west dichotomy within Prussia that shaped the
nineteenth century—the disjuncture between the economically
advanced, already partly industrialized, ‘bourgeois’ west and the
manorial, ‘feudal’ east—was already emerging at the beginning of
the seventeenth century. With regard to German-speaking countries
as a whole, the manorial system (Gutsherrschaft) was a distinctly
55
East Elbian phenomenon, one that had been developing since the
late Middle Ages. A formerly free peasantry was not only
compelled to do socage for the aristocratic lord, as in the other
parts of feudal Germany; they were also reduced to a particular
form of serfdom, inherited servitude (Erbuntertänigkeit). This was
only possible because the manorial lords, the Junkers, were able to
appropriate wide-ranging legal competencies from the territorial
sovereign, giving them powers of jurisdiction and law enforcement
well beyond those of the manorial police. The result was a
transformation of the manor into a self-enclosed political entity, its
lord into a sovereign authority in his own right. For Lutherans, it
was self-evident that this authority had no less a claim to their
obedience than the territorial lord. The proximity of throne and
altar at the apex of the state found its parallel at a lower level in
the relationship between the manorial lord and ‘his’ parson.45
The intensification of manorial authority went hand in hand
with the extension of manorial boundaries. After the Junkers
succeeded in wresting the grain trade from the cities, permanently
weakening the East Elbian bourgeoisie, they also sought to increase
their cultivable lands at the expense of the peasantry. Even in the
sixteenth century, the eviction of peasants and appropriation of
their lands, a phenomenon known as Bauernlegen, was widespread
in eastern Germany; it increased in the wake of the Thirty Years
War. East of the Elbe, practically the only peasants left after the
great conflict were those reduced to a permanent and hereditary
bondage to the soil. This serfdom made the agrarian society of East
Elbia more ‘eastern’ in character, that is, more like eastern Europe.
This process was the The Austro-Prussian Antagonism converse of the
56
westernization of Brandenburg-Prussia in the seventeenth and
eighteenth centuries. In the latter case, the transformation involved
completely different aspects of society: the governmental
administration and the judicial system were modernized according
to west European models. Trade, artisan industry, and science were
fostered and promoted, and religious tolerance was secured.46
The social power of the Junkers did not always translate into
political power. The Great Elector, creator of the standing army,
had fought fiercely against the estates and thus against the political
aspirations of the manorial lords. His grandson, the ‘Soldier-King’
Frederick William I (1713–40), transferred the holdings of the
nobility into the possession of the state whenever he could. In 1733
he divided the country into levying districts or cantons firmly
delimited from each other, and built the army upon this structure.
The Junkers formed the officer corps; the soldiers were peasants. In
compensation for their duties to the state, the peasantry was
granted a certain amount of protection from the Junkers. The
manorial lords were no longer permitted to take away the lands of
‘their’ peasants.
Frederick II (1740–86) adopted a radically different policy with
regard to the nobility. Whereas his ancestors, beginning with the
Great Elector, had curtailed the power of the manorial lords, under
Frederick a process of re-feudalization took place. The Junkers
were granted sweeping privileges: the middle classes were no
longer allowed to purchase noble estates, and the government
granted credit to economically struggling manors.
These material benefits were only one part of the historical
57
compromise with the nobility that supported the rule of Frederick
II. Politically, too, the Junkers gained influence on all levels. Their
increase in power has been aptly described by Otto Büsch:
58
geographical division of the Prussian territories and their resulting
military vulnerability. The spirit of discipline and subordination,
the very condition of Prussia’s rise, was the answer to a challenge.
This answer was one-sided and required correction. By dint of its
origins, Prussia was forced to a greater extent than other powers to
become a soldier-state. Yet it could only assert itself successfully in
the long term by overcoming the hardness and harshness that
resulted from the internalization of external coercion.
The internalization of ‘Prussiandom’ was, to a great degree, the
achievement of Pietism. The roots of this devotional movement
reached back to the period of the Thirty Years War. With their
appeal for a renewal of the church from within, the Pietists were
reacting against the orthodox petrifaction within Lutheranism,
indeed against every kind of fixation on the external aspects of
dogma. In Brandenburg-Prussia, the separation from official
Lutheranism formed a bridge between the Pietists and the Calvinist
authorities. This relationship was further supported by the Prussian
Pietists’ great interest in the reform of the schools and higher
institutions of learning, efforts that found their classic expression in
the Franckeschen Stiftungen, a series of educational foundations
established in Halle by August Hermann Francke, and in the
founding of a Reform-oriented university in the same city in 1694.
Never were the ‘quiet people of the land’ so near to the throne as
they were under the Soldier-King. A Pietistically informed ethos of
the state outlasted the reign of Frederick William I, a territorial
patriotism that raised the subjects’ devotion and love towards the
sovereign to the status of a religious duty.48
Religion was also the ultimate source of the economic prosperity
59
Brandenburg-Prussia experienced in the eighteenth century thanks
to the Huguenots. Immediately after Louis IV rescinded the Edict of
Nantes in 1685, the Great Elector promulgated the Edict of
Potsdam, which invited the persecuted French Calvinists to
Brandenburg. Approximately 20,000 followed this call. More than a
quarter of them settled in Berlin, where French immigrants formed
almost a fifth of the population in 1700. In artisan crafts, trade,
and manufactures, the Huguenots introduced into Prussia the
dynamic economic ethos that so often went hand in hand with the
Calvinist religion. Yet Calvinists also excelled in academic
professions. Indeed, wherever they became active, they contributed
decisively to the very thing the naturally poor country of
Brandenburg, the ‘sandbox of the Holy Roman Empire’, needed
above all else: modernization.49
Systematic government support of trade, crafts, and industry;
extension of roads and highways; cultivation and settlement of
moorlands and wastelands; fostering of the sciences, and religious
tolerance—these were the achievements by virtue of which
eighteenth-century Brandenburg earned the reputation of marching
in the avant-garde of progress. These things were only possible
because the energetic rulers of the house of Hohenzollern had an
army of disciplined officials at their service. Frederick William I
had created the institutional framework, a rationally organized
apparatus of governmental offices with the General-Ober-Finanz-
Kriegs- und Domänendirektorium (commonly known as the General
Directory) at the top. The central administration, organized partly
along departmental, partly along geographic lines, performed with
such effectiveness that by the middle of the eighteenth century the
60
absolutist Hohenzollern state came to be considered the model of
efficient government throughout all of Europe.50
The same could hardly be said of Austria at that time. The
Reich’s most powerful estate contrasted sharply with Brandenburg-
Prussia in a number of areas. The religious compromise set down in
the Peace of Westphalia was not recognized in the Habsburg crown
lands. In the course of a forceful re-Catholization, initiated long
before 1648, far more that 100,000 Protestants fled from
Habsburg-ruled territories, an intellectual and economic loss from
which Austria never recovered. The Hohenzollerns did not merely
promulgate religious pluralism, they also practised it; the
Habsburgs, in contrast, imposed confessional homogeneity. Part
and parcel of the Catholic absolutism of Austria was a rigorous
system of censorship, in comparison to which Brandenburg-Prussia
seemed like a haven of intellectual freedom. Although under Maria
Theresa (1740–80) the administration of censorship was removed
from the hands of the church, it was also bureaucratically
systematized in the process. The spirit of the Enlightenment, which
found a home in the Prussian universities, found in Austria only
barricades raised against its entry.
But Prussia and Austria also shared structural similarities.
Austria, too, can be called a military state. As under the
Hohenzollerns, the army became the most important agent in the
process of state centralization. And like the Hohenzollerns, if
somewhat later, the Habsburgs suppressed the influence of the
estates. This was necessary in order to avoid permanently falling
behind other states, especially Brandenburg-Prussia. Even under
Charles VI, the Habsburg territorial mass represented more an
61
alliance of corporative states (Ständestaaten) with a monarchical tip
than a modern state entity. When Maria Theresa and her ministers
undertook a comprehensive reform of the government
administration in the early 1740s, they consciously oriented
themselves on the model of Prussia. The result was an absolutist
state with highly concentrated apparatus of power, less centralized
than Prussia but no less bureaucratic.51
The antagonism between the Catholic Habsburgs and the
Evangelical Hohenzollerns had long since ceased to be a religious
conflict when war began between the two countries in 1740, a few
months after Frederick II came to power and a few weeks after
Maria Theresa assumed the throne. The first war, over Silesia, was
begun by Frederick II, for both reasons of personal ambition and
reasons of state. Exploiting the weakness of Vienna after the death
of Charles VI, Frederick sought by the conquest of Austrian Silesia
to expand the territory, fortify the economy, and elevate the
political status of Prussia to that of a European great power. Silesia
did become Prussian, and Prussia in turn a great power. By the time
these results were sanctioned by the Peace of Hubertusberg in
1763, however, Frederick’s state had already experienced two
wars, the latter of which—the Seven Years War—might just as
easily have ended with Prussia’s downfall.
With respect to method, Frederick’s style of power politics
hardly differed from that of other sovereigns like Louis XIV of
France or Charles XII of Sweden. The same can even be said of the
most reprehensible of his deeds, his cooperation in the first
dismembering of Poland in 1772. What set the Prussian king apart
was the glaring discrepancy between his resources and his risks. In
62
1740 Brandenburg-Prussia was still a geographically fragmented
and fragile entity. Again and again, Frederick was willing to risk
everything, and was ultimately saved only by pure chance, an
incident that went down in history as the ‘miracle of the House of
Brandenburg’—the death of Frederick’s most dangerous enemy,
Elisabeth, Empress of Russia, in January 1762. The ultimate
success of his radical brinkmanship turned Frederick into a legend.
His example, however, also promised disaster.52
Frederick could not have earned the nickname ‘the Great’ had
he been nothing more than a successful military leader. In the
Europe of his day, he was rightly considered the representative—
indeed the very embodiment—of a new type of government,
enlightened absolutism. His programme of rationalism imposed
from the top down differed fundamentally from the typical narrow
self-interest of other absolutist rulers. Even during Frederick’s
lifetime, the phrase travailler pour le roi de Prusse (‘to work for the
King of Prussia’) meant to do something for its own sake. However,
the thing done had to be reasonable, and that meant in the interest
of the state. And the ruler-philosopher of Sanssouci, who
considered himself the first servant of the state, seemed to
guarantee that this was indeed the case.
Frederick opened his country to western ideas and prompted a
kind of cultural revolution. At the same time, he restored privileges
to the aristocracy, whom he could not otherwise lastingly integrate
into his military machine. In so doing, he again pushed Prussian
society a stretch towards the east, in the direction of an autocracy
supporting itself upon nobility and serfdom. Prussia under
Frederick I was more bourgeois in character than Russia, but in
63
comparison with France, which was such an intellectual model for
Frederick, it seemed almost a country without a middle class. A
comparison within the Reich leads to similar findings: Frederick’s
Prussia was far less bourgeois than most other states, above all in
the western and southern parts of Germany, although it was not as
feudal in character as the two duchies in Mecklenburg.
According to a saying widely attributed to Mirabeau, but
actually from an eighteenth-century German military historian,
Georg Heinrich von Behrenhorst, ‘The Prussian monarchy is not a
country with an army, but an army with a country, in which it is
only quartered, so to speak.’ Although it is true that all absolutist
states were also military states, Prussia was especially militarized.
In mid-eighteenth-century Austria, for example, one inhabitant in
sixty was a soldier; in Prussia, the ratio was one out of every
thirteen. In the hierarchy of public interests and needs, those of the
military stood just as high under the reign of Frederick II as under
his father, the Soldier-King.53
The primacy of the military, as well as the entitlement of the
nobility to which it was intimately linked, prevented the full
realization of the rule of law within the government. Nonetheless,
Prussia under Frederick continually strove in this direction. Great
progress was made in the standardization of the legal system. The
administration of justice through governmental offices was
restricted, and a new code of procedure, the Codex Fridericianus,
streamlined legal proceedings and made them more transparent.
The crowning achievement was the Allgemeines Landrecht für die
Preussischen Staaten (General Law for the Prussian States or
‘Prussian Code’), which was completed in 1791, five years after
64
Frederick’s death. As Reinhart Koselleck observes, the Prussian
Code was Janus-faced:
65
auspices of the state; and he granted to non-Catholics full rights of
citizenship as well as the right to exercise their religion in private.
However, Joseph faced stiff resistance from the Hungarians
when, in his continuing effort to unify and standardize the
administration of Habsburg lands, he tried to replace Latin with
German as the official language of government and business in
Hungary. His attempt to incorporate the church completely into the
service of the state, which involved the dissolution of many
monasteries, provoked internal unrest. Shortly before his death,
Joseph was compelled to revoke a number of his own measures.
Other of his reforms were rescinded by his brother and successor,
Leopold II (1790–2).56
Although certain aspects of ‘Josephinism’ survived, all in all the
‘revolution from above’ failed in Austria. Together with the clergy,
with whom it sympathized, the Catholic population rejected
enlightened absolutism. Here was a major difference vis-À-vis
Prussia. There, the Calvinism of the upper classes had cleared the
way for the development of enlightened absolutist rule. Frederick’s
reforms provoked no Lutheran reaction from below. In contrast to
Joseph, who remained a committed Catholic, the free-thinking
Frederick did not even attempt to challenge the religious feelings of
his subjects. In contrast to ‘Old Fritz’, who was extremely popular
due to his military exploits, Joseph never found his way into the
hearts of his people. And the sentiments inside each ruler’s country
were also mirrored abroad. This brings us to the question of how
the Prussian–Austrian antinomy affected the way the Germans saw
themselves.
66
COSMOPOLITANS AND PATRIOTS
67
Prussia as a ‘nation’ in his work Vom Tode für das Vaterland.†
According to Abbt, the answer to the question ‘What is one’s native
country?’ was not necessarily answered by the simple fact of one’s
birthplace. Rather, citizenship was a matter of free decision. ‘But
when birth or my free resolution unify me with a country, to whose
holy laws I place myself in subjection, laws that take from me no
more of my freedom than is necessary for the good of the whole
country—at this moment do I call this country my Fatherland.’58
For Austria, on the other hand, the word ‘nation’ could not be
used; the term connoted a commonality of language that did not
exist in the multi-ethnic Habsburg empire. The rule of the
Habsburgs had long since extended beyond the borders of German-
speaking Europe. It included territories in which Hungarian, Czech,
Slovak, Slovenian, Croatian, Serbian, Romanian, Italian, French,
and Dutch were spoken, as well as Polish, Ukrainian, and Yiddish
after 1772. This was a great difference vis-À-vis Prussia, which was
always considered a German state, the annexation of Polish
territory from 1772 to 1795 notwithstanding. To a far greater
extent than in Prussia, the active bearers of Austrian state
patriotism were Enlightenment-influenced civil servants. Prussian
state patriotism could claim a far broader societal basis, reaching
all the way down to the hereditary peasantry. For this reason, it
had an appeal that extended much further beyond its national
borders than was the case with Habsburg Austria.59
In the second half of the eighteenth century, however, the
concepts Vaterland and ‘nation’ did not only apply to the territorial
state in Prussia. Goethe and the conservative writer Justus Möser
68
used the word Vaterland for their respective native cities of
Frankfurt and Osnabrück. The Berlin publisher, writer, and
Enlightenment philosopher Friedrich Nikolai spoke of a ‘Bavarian
nation’ as well as ‘Bavarian patriots’ and even ‘patriots of
Nuremberg’. In 1773 the poet Christoph Martin Wieland, the
Pietist-educated son of a pastor from the village environs of the
free imperial city of Biberach, came to the conclusion: ‘The German
nation is actually not one nation, but an aggregation of many
nations.’60
With this observation, Wieland drew the balance of a process of
development that had begun with the Reformation and been
invigorated by the Peace of Westphalia. Looking back on the period
from 1000 to 1500, we might speak of a gradual formation of
German national identity. A consciousness of community,
supported primarily by commonality of language, slowly emerged
north of the Alps. Two phases are conspicuous, the first around the
turn of the millennium, the second in the decades around 1500
with the end of the Middle Ages and the beginning of the Early
Modern period. After the Reformation, however, the opposition
between the religious confessions brought discord and division to
the fore. From that point onward, the strongest feelings of supra-
territorial community were shared by Germans of the same
denomination. To be sure, the Holy Roman Empire continued to
exist, providing an ultimate institutional support for the German
nation. And imperially oriented publicists of the seventeenth and
eighteenth centuries never grew weary of emphasizing that the
Roman empire was ruled by the Germans. Nonetheless, after the
Peace of Westphalia, the importance of the Reich for the nation
69
decreased so considerably that, as early as 1667, a sober scholar of
constitutional law like Pufendorf could pronounce a final
judgement: ‘For the rest, it is obvious that Germany not only has no
advantage from her title “Roman empire”, but in fact has had great
harm and unpleasantness from it.’61
Pufendorf’s assessment anticipated the critique of the
eighteenth-century klein-deutsch historians like Heinrich von Sybel,
who spoke of the great wealth Germany wasted on its Italian
campaigns, undertaken only to gain the imperial crown. Like
Luther earlier, Sybel rejected the theory of a transmission of
imperial authority from the Greeks to the Germans, insisting that
the Roman empire had long since fallen into ruin—as a
consequence of the West Gothic invasion—by the time of
Charlemagne’s coronation. Therefore, the present-day Reich could
not be equated with the Roman empire. Only one short step led
from this Protestant view of history to Wieland’s rhetorical
question of 1793: ‘But the German patriots, who love the whole
German Empire as their Fatherland, love it above all else... —
where are they?’62
The tendency to apply the terms Vaterland and ‘nation’ only to
an individual state was more characteristic of Protestant than
Catholic authors. Nowhere was it more pronounced than in Prussia
in the second half of the eighteenth century, during which
Frederick’s kingdom rose up to become the most decided opponent
of the Habsburgs and thus of the entire obsolete majesty of the
Reich and emperor. The acclaim with which Frederick’s victories
were hailed outside of Prussia attested to the Reich’s loss of
70
prestige. This was no surprise, given the powerlessness it
demonstrated time after time. The French encroachments in Alsace
and occupation of Strasbourg in 1681 had provoked neither
reaction nor defence on the part of the emperor and the Reich. The
fact that the ‘Imperial Army’ fought on the side of the French
during the Seven Years War (and earned no military glory
whatsoever) only further increased Prussia’s prestige; for the Reich,
the debacle produced only derision and mockery.
For all that, however, imperial patriotism was still very much
alive in the eighteenth century. Especially the smaller and middle-
sized imperial estates retained the conviction that it was necessary
to reform the imperial constitution, all the while remaining faithful
to emperor and Reich. This view was expressed with particular
urgency by the jurist and Pietist Johann Jakob Moser, one of the
foremost authorities on imperial law. From his Württemberg
homeland, Moser waged a war against absolutism in the name of
the ‘old law’ of the territorial estates, an activity that earned him a
five-year prison term (an illegal sentence) on the Hohentwiel in
1759. Moser’s proposals for the restoration of the Reich formed
only a small part of his enormous scholarly output. They amounted
to bringing Prussia into close alliance with Austria, whereby the
former would be elevated so far above the other imperial estates
that it would act as a kind of co-regent of the Reich.
The wars between 1740 and 1763 reveal how far from reality
such fantasies were. Not that the imperial patriots allowed
themselves to be deterred by such realizations. One of them,
Moser’s son, the publicist Friedrich Carl von Moser, wrote a very
popular pamphlet entitled Von dem deutschen Nationalgeiste* (1765),
71
in which he contradicted the ‘false and pernicious doctrine of a
double Fatherland, a Catholic and an Evangelical Germany’, which
was deeply ingrained in the minds of contemporaries. ‘When a
Bohemian count sincerely believes that the German fatherland is to
be found in Vienna, a Brandenburg nobleman can only press his
hat to his forehead and desire to hear nothing further about any
German fatherland.’ Moser did not, however, see fit to propose any
practical suggestions as to how such ‘separatist ways of thinking’
might be overcome. He contented himself with the appeal ‘to move
the lords and heads of our fatherland to want what they should
want’, that is, to place the general good of the Reich above the
selfish interests of the imperial estates, especially the more
powerful among them.63
In addition to imperial patriotism, another kind of pan-German
sentiment persisted: a consciousness of a common culture,
expressed primarily through the cultivation of the German
language. In the half-century between 1720 and 1770, the ‘German
societies’ (Deutsche Gesellschaften) inspired by Johann Christoph
Gottsched spread over all of Germany, most notably in Evangelical
university towns such as Leipzig, where the movement first began.
Against the French-inspired culture of the aristocratic courts, the
German societies set a bourgeois German culture dedicated to the
study of grammar and lexicology, the art of speaking and
translation, and the memories of great figures and events in
German history (above all Luther and the Reformation). These
societies also placed great emphasis on the celebration of public
events (the birthday of the territorial sovereign, the anniversary of
the city founding, etc.) with due pomp and circumstance. In one of
72
his academic addresses, Gottsched described the ideal member of
the German societies as the ‘very model of an upright burgher,
sincere patriot, and, not least, honest and conscientious speaker’.
For Gottsched, there was only one ideal of form, and it was
French. In the next generation, the criticism of aristocratic court
culture prompted a more radical critique of the French model. In
1770 the Hainbund of Göttingen sought to oppose French
immorality with German virtues, which were thoroughly bourgeois
in character and generally informed with Christian piety. ‘The
German’, as Wolfgang Hardtwig synoptically puts it, ‘is upright,
noble, good, cultivates “good strict morals”, and avoids pomp.’ This
was hardly more than the literary stylization of national
stereotypes that can be traced back beyond the baroque language
societies and the humanist movement all the way to the high
Middle Ages. In these stereotypes, German self-praise could not
easily be distinguished from self-pity and a sense of inferiority. Like
the imperial patriots, the literary patriots lamented the fact that,
throughout the centuries, Germany had fallen more and more
behind France. This was indeed the case. After 1648 the hegemony
had passed to the neighbour in the west—not only politically, but
also culturally.64
A few French intellectuals, first and foremost the historian Jean
Baptiste Du Bos and the philosopher Voltaire, went so far as to
speak of a migration of cultural leadership from one people to
another. Here they had recourse to the ancient idea of translatio
artium, which, like analogous doctrine of the translatio imperii,
consisted of four golden ages. In the new interpretation, the Greece
of Philip of Macedon and Alexander the Great was followed first by
73
the Rome of Caesar and Augustus, then by the Italian Renaissance
under Popes Julius II and Leo X. The last age, the ‘great century’
that, according to Voltaire, came closest to perfection, was the age
of France under Louis XIV. The cult of Greece in German classicism
was, in part, an answer to this French challenge. As Conrad
Wiedemann notes, the one-sided emphasis on classical Hellas
implied not only a rejection of classical Rome, but also classical
France.65
If Germany wished to catch up to and surpass France, it was
clear that it could not do so in the political arena. The French
already possessed a nation state, one with a venerable history.
Some of the greatest German thinkers doubted whether the
Germans would ever become a nation. In his Hamburgische
Dramaturgie* (1763), Lessing poured scorn on the Germans, who
were ‘still the sworn imitators of all that is foreign, and above all
still the subservient admirers of the never sufficiently admired
French’. He explicitly rejected ‘the good-hearted gesture of giving
the Germans a national theatre, since we Germans are not even a
nation yet!’ For Goethe and Schiller, this should remain the case
even in the future.66 As late as 1796 they warned in the Xenien:
74
or Weltbürgertum: an imagined, invented community, not one
achieved through personal observation and concrete experience. Of
those who advised Germans against going the route of the national
state, at least politically, virtually nobody posed the question of
whether this sense of a direct, unmediated relationship to the
world did not imply a kind of national presumptuousness. But what
Goethe and Schiller regarded as an alternative, other thinkers of
the classical period considered a false dichotomy. For Herder,
differences among peoples were a natural fact, indeed a ‘barricade
against the presumptuous joining together of peoples, a dam
against foreign inundation’. At the same time, however, reasonable
Nature also sought the progressive ennoblement of humankind and
its consummation in true humanity. As Herder wrote in the third
part of his Ideen zur Philosophie der Geschichte der Menschheit,†
which appeared in 1787, ‘Humanity is the purpose of man. Thus
Nature and God have put the fate of our kind into our own hands.’
‘The course of history reveals that the destructive demons of
humankind have truly diminished with the burgeoning of true
humanity, and this has happened according to the inner natural
laws of an enlightened reason and art of government.’67
75
be rooted out and replaced by patriotism and cosmopolitanism.’68
Whether citizens of the world, literary or imperial patriots:
German poets and thinkers spoke only for a well-educated
minority. And yet, though often divided among themselves, the
members of this audience represented nothing less than the public
sphere of their day. In comparison, the patriotism of the separate
territorial states enjoyed a much broader social basis. In
Evangelical Germany, especially in Prussia, the Pietist movement
was one of the most influential factors in the formation of
territorial identity. Like the enlightened patriotism of a later
period, Pietism demonstrated a highly practical interest in the
promotion of local community projects. In any case, there was no
necessary conflict between the allegiance to an individual
territorial state and a partiality to some idea of Germany as a
whole. This was just as true of the Volk as it was of the
representatives of ‘Germany of the mind’. The Reich, meanwhile,
had come to be regarded in large sections of the population with
rather more disdain than reverence. ‘The dear old Holy Roman
empire, how does it stay together?’ Frosch’s question in Auerbach’s
cellar might have been posed by any sober-minded imperial
subject.69
Given the lack of a common German state, the most important
bond uniting Germans was still that of language, and it was with
good reason that Herder wrote that the simple people preserved the
mother tongue much better than the nobility and the educated
classes. To the latter he directed an appeal that has nothing to do
with ‘linguistic nationalism’:
76
Whoever disdains the language of his nation dishonors its most
noble public, becoming the most dangerous murderer of its spirit,
its inner and outer prestige, its creations and inventions, its fine
morality and industry. Whoever elevates the language of a people,
forming it to the most powerful expression of every feeling, every
clear and noble thought, helps to extend that most worthy and
most beautiful public, or else unite it and establish it more
firmly.70
77
transformations the Industrial Revolution was bringing to England,
the intensifying political crises in France and America. The
widespread consciousness of a new secular era was no respecter of
national borders. All aspects of tradition were summoned before
the judgement seat of Reason. The Sattelzeit,* the period of
transition from the estates and fixed corporative structures of old
Europe to a new form of society characterized by a desire for
freedom and equality, had begun.72
78
2
79
(H.A.W.)]; yet only now has humanity come to understand that
spiritual reality should be ruled by Thought.1
80
of the most astute and influential publicists of the day and an early
supporter of the revolution, condemned the deposing of the French
king as an act disrupting the proper balance between the
legislative, judicial, and executive powers.
Wieland put to his audience, as well as to himself, a question he
thought could only be answered in the course of time:
Will the new order arising from this chaos—when finally Deus et
meliora Natura [‘God and better Nature’ (H.A.W.)] gain the upper
hand—heal the numberless wounds inflicted by the democratic
cacodaemon [from kakÓs, the Greek word for ‘bad’, ‘evil’ (H.A.W.)]
of this nation drunk on freedom, heal them quickly and completely
enough to be judged a redress for so much evil?
81
of a monarchic government degenerated into the most intolerable
aristocracy’ were, in Wieland’s eyes, ‘the most praiseworthy of all
enterprises’; yet he could never call it praiseworthy ‘to set up a
monstrous, endlessly convoluted, awkward and unstable democracy
instead of a monarchy restricted to its true limits (after the
example of the English constitution) through sufficiently secured
rights of the people’. The French solution was reprehensible to
Wieland, who was convinced that another path could have been
taken in that country, too. ‘Without a doubt a reformation—both
necessary and unavoidable—should have been undertaken with the
nobility, just as with the court and the clergy.’
By January 1793, the month in which the National Convention
condemned Louis XVI to death with a majority of one vote and sent
him to the guillotine, all Wieland’s remaining doubts had been
removed.
I, too, see just as well as anybody that things are not as they should
be, and do not work as they ought to work, whether in Germany or
in the rest of Europe; and I am very convinced that the evils about
which we have cause to complain can only be addressed by a
thorough reformation of the legislative power and the constitution
as it stands today. Yet I maintain that neither the new theory of the
French demagogues nor insurrections and the overthrow of the
existing order of things can bring this about, or should even be be
attempted. What has happened in France cannot and should not
serve as a model for us, but rather as a warning to princes.3
82
Wieland did. At first they hailed the revolution as an act of
liberation, then turned away to the extent that radical political
forces and with them the urban lower classes gained influence and
the country threatened to sink into a bloody civil war. Herder, also
an impassioned defender of the watershed events of 1789, wrote
three years later that he knew nothing more repugnant ‘than a
populace agitated to madness, and the rule of a mad populace’.
What had France accomplished with its revolution, seeing that ‘it
languishes in the most fearful disorder of things’? How could one
hope for a better education from a revolution
The only thing that the Germans could do was learn from the
French experience. ‘We can witness the French revolution as a
shipwreck on the open sea observed from a safe shore, unless we,
too, are pulled by our evil genius into the sea against our will.’4
Only a very small number of German publicists had a certain
understanding for the Jacobins. Even the majority of the ‘German
Jacobins’ (the term is of questionable legitimacy) clearly rejected
the idea of emulating revolutionary France. The travelling scholar
and writer Johann Georg Forster, member of the Club of German
83
Friends of Liberty in French-occupied Mainz in 1792, wrote in
1793 that the ‘the physical, moral, and political conditions’ in
Germany were such that the only sensible path to change was a
process of ‘slow, stepwise perfection and maturity’. Germany
should ‘learn from the errors and sufferings of its neighbours and
perhaps receive gradually bestowed from above the freedom that
others have had to seize violently and all at once from below.’ The
radical writer Georg Friedrich Rebmann admitted in 1796 that he
never ‘seriously considered a German revolution after the model of
the French. That would be absolutely impossible in Protestant
lands, and in our Catholic territories nearly so.’
The idea that the reformation of the church had obviated the
need for a political revolution was a widespread article of faith in
Protestant Germany of the late Enlightenment. According to this
view, the religious renewal introduced by Luther held a general
promise of freedom, a promise to be redeemed gradually. Thus,
historically speaking, Germany was further along than France. It
could reform itself politically, since it had already been reformed
ecclesiastically. In the opinion of late Enlightenment thinkers,
Germany had to proceed along this path if it wished to avoid the
fate of France. After all, a German revolution was by no means an
impossibility. ‘It will and must happen,’ Rebmann warned in 1796,
‘if a reformation is not brought about first.’5
This insight was not foreign to a few of those who exercised
political power in Germany. In 1799 the Prussian minister Karl
Gustav von Struensee remarked to a Frenchman: ‘The revolution
that you have made from the bottom up will take place in Prussia
slowly from the top down ... In a few years, there no longer will be
84
any privileged class in Prussia.’ Those were more than mere words.
In its own domains, where the Prussian state had a free hand, it
pushed through a significant reform between 1777 and 1805, the
liberation of the peasants from the obligatory manual and draught
services (Hand- und Spanndienste) they owed to the manorial lords.
This was a prelude to the general ‘peasant liberation’ of 1808. The
sponsors of this revolution from above were the educated civil
servants from both the bourgeoisie and the nobility. This group
played the part of the ‘general estate’ that the ‘third estate’, the
bourgeoisie, ascribed to itself in France. An enlightened
bureaucracy legitimated (not only in Prussia, but with particular
effectiveness there) the claim of the enlightened ruler to unify the
executive and the legislative powers in his person. Where justice
and reason combined so harmoniously, there could be no need of
an elected body of legislators.6
Even before the French Revolution, German theorists of the state
had interpreted Montesquieu, the classical formulator of the
doctrine of the separation of powers, in this sense. After the
experience of mass violence and revolutionary terror, the idea of a
‘free monarchy’ found even greater resonance among the educated
German public. But the greatest living philosopher saw things
differently. In his Zum ewigen Frieden* (1795), Immanuel Kant
defined despotism as the governmental principle of ‘the
autonomous execution by the state of laws it has itself decreed’.
Accordingly, despotism is ‘the public will, to the extent that it is
treated by the regent as his personal will’. Against despotism Kant
placed republicanism, the form of government in which executive
85
power was separated from the legislative. For Kant, such a
republicanism was perfectly compatible with a monarchy at the
head of the state. It was not, however, compatible with a
democracy in the sense of unmediated popular sovereignty, that is,
one of the badly organized ‘democracies without a representational
system’. Thus, direct rule of the people was perforce a form of
despotism.
Kant maintained his publicly announced sympathy for the ideas
of the French Revolution through the period of the Terror. ‘Such a
phenomenon in human history can no longer be forgotten,’ he wrote
in the Streit der Fakultäten† in 1798, ‘for it has awakened within
human nature a predisposition and an ability for improvement,
such as no politician could have devised from the course of things
hitherto.’ By appealing in Zum ewigen Frieden for a ‘representative
system’ and, in the Rechtslehre of the Metaphysik der Sitten† (1797),
explicitly invoking a ‘representative system of the people’, the
Königsberg philosopher took a decisive step beyond enlightened
absolutism. The same was true of his advocacy of a constitution ‘in
which Law is autonomous and depends upon no particular
individual’. Yet because Kant insisted upon legal reforms and
sought to avoid a violent revolution, the actual addressee of his
ideas was the same as that of the other German late Enlightenment
thinkers: the enlightened absolutist state.7
‘Not the German reaction, but rather German progress was the
thing that set Germany back behind the west.’ With this trenchant
paradox, the historian Rudolf Stadelmann attempted in 1948 to
express the variety of reasons a successful revolution never took
86
place in Germany.
87
France. Moreover, the English had acquired their freedoms and
rights of political participation not only in a peaceful manner, but
also through revolutionary means. Finally, the friendly reception
Burke’s critique of the French Revolution found in Germany was
not least due to the fact that the author’s reference to resistance
and revolution as the ‘extreme medicine of the constitution’ was
typically overlooked.9
The violence of its self-emancipation was not the only
phenomenon that alienated educated Germans from revolutionary
France. The rallying cry of the ‘single and indivisible nation’ also
met with irritated dismissal in Germany. Long before 1789, indeed
even before the appearance of Montesquieu’s De l’espirit des lois* in
1748, German authors had interpreted the dualism between the
emperor and the imperial estates as a specifically German version
of the limitation of monarchical power. France was an absolutist
and comparatively centralized state. The Reich was neither the one
nor the other; in fact, it was not a state at all.
Wieland was among those who attempted to make a virtue out
of the necessity of German particularism and political-territorial
multiplicity. As he wrote at the beginning of the 1780s, the
Germans, by dint of their constitution, would never ‘think and act
as a single people, never possess what can be called in the moral
sense ‘ “National-Uniform” ’. Yet all the disadvantages of the
German political situation were
88
people will be better guarded against general foreign, domestic,
political, and ecclesiastical enslavement and subjugation, than we
Germans.
89
of every other, has above it the imperial laws, or the emperor and
the Reich, to the extent only that these latter are empowered to
administer and execute those laws; and that, from its own elected
head of state down to the mayor, master, city council, and
community of the imperial city of Zell on the Hammersbach, there
is no regent in Germany whose greater or lesser authority is not
restricted by laws, traditions, and in many other ways, and from all
sides; and against whom, if he permits himself any illegal act upon
the property, honour, or personal liberty of the least of his subjects,
the imperial constitution will not provide protection and remedy to
the aggrieved party.10
90
instrument of a mobilization of the masses.
In light of these innovations, it is advisable to reserve the term
‘nationalism’ (or at least ‘modern nationalism’) for the period
during and after the turning point in world history defined by the
French Revolution. One of the more radical consequences provoked
by the dethronement of Christian universalism, entering a new
phase in 1789, was the development of nationalist sentiment into a
substitute religion, in which loyalty to the nation was given
priority over every other commitment. And indeed, the nation
became in the post-revolutionary era what the church had been
before: a binding ethical-moral system giving meaning to the lives
of individuals and justifying supra-personal authority. This new
character of the nation separated modern nationalism from
traditional patriotism, which was an emotional tie, nearly always
religious in quality, limited in scope to an individual’s home
country, its sovereign, and the sovereign’s ‘house’, the dynasty.
Patriotism also traditionally lacked the reciprocity between the
claims of the individual upon the nation and those of the nation
upon the individual, a relation without which modern nationalism
could not have become a mass movement.
Yet, as we shall see, nowhere in Europe did old-style patriotism
simply die out after 1789. It was able to modernize, absorbing
certain elements of modern nationalism. There was also continuity
in the other direction: the new republican nationalism of the
French, in invoking such ideas as human liberty, equality, and
fraternity, had precursors and models in the community spirit that
had developed among the citizens of the northern Italian
republican city states or confederate Geneva, the home of Jean-
91
Jaques Rousseau, creator of the theories of the contrat social and
the volonté général. Nonetheless, nationalism as the unifying bond
of a large nation, recognizing nothing higher than itself on earth—
this was something that had not existed in Europe before the
French Revolution.
Only the United States of America could lay claim to a small
chronological edge when it came to determining the birthright to
the new ideology of human community. By no means, however, did
the North American subjects of the British break with their
traditional religious understanding of the just order when they
rebelled against the crown. On the contrary, their revolt and
declaration of independence were informed and sustained by
religious ideals. The American Revolution was a conservative
revolution, something that cannot in any way be said of the events
in France. American nationalism, in contrast to the French, was
both modern and traditional.11
Whoever looks for ‘modern’ forms of national feeling in
Germany during the latter half of the eighteenth century will have
the best chance of finding them in Frederick II’s Prussia. In his
above-mentioned work Vom Tode für das Vaterland (1761), written
during the Seven Years War, Thomas Abbt, the ‘Prussian by choice’
from Ulm, explained the duty of the individual subject to sacrifice
his life for his country in case of necessity. The love of the
Vaterland sets up the subject’s ‘nation as an eternal model for other
nations’. It assures everlasting fame and the gratitude of posterity
to those who give up their lives for their country. The author even
attempted to prove that ‘the love of the Fatherland (provided one
does not have the support of faith in a revealed religion) is the best
92
means to overcome fear of death.’ Why, asks Abbt, should we not
say that love of one’s king is stronger than death? ‘The love for the
Fatherland convinces us that no pleasure may figure in the face of
the pleasure to have served one’s country, and that such a death
adds more to the sum of our pleasures than we ever would receive
in the course of a longer life.’12
The love of the Vaterland as a correlation to the belief in God—
here Abbt, who had commenced his studies with theology before
switching to enlightened philosophy, the fine arts, and
mathematics, comes very close to nationalism as a political
religion. Yet however eager he was to demonstrate that a selfless
patriotic love was possible not only in a republic, but also in an
enlightened monarchy, Abbt’s patriotism exhausted itself in the
claims of the Vaterland upon the individual. He had nothing to say
concerning the voice of the citizen in the politics of his country. In
recompense for their duty of self-sacrifice, the Prussian state held
out to its subjects only the prospect of honour, which was to be
considered the highest reward the nation could bestow. Rights of
political participation, such as the French citoyen possessed—such
things were not to be deduced from the Prussian subject’s duty to
serve his country to the utmost. He had to content himself with the
fact that the laws of the enlightened state did not restrict his
liberties even more than they did.
Only in comparison to imperial patriotism could Prussian state
patriotism be called ‘modern’. The Reich, for its part, could not lay
claim to any subjects willing to give up their lives out of patriotic
devotion. Its only real support lay with a few smaller and middle-
sized estates, flanked by jurists and publicists. When the war waged
93
by Prussia and Austria against revolutionary France after April
1792 was officially designated by the Reich as ‘general war of the
Empire’ in March 1793, its popularity was in no way increased.
Even the most virulent German critics of the Jacobins generally
rejected military intervention on behalf of the older powers.
Still, when Prussia concluded the Treaty of Basel with France in
April 1795 (the treaty that prompted Kant to write Zum ewigen
Frieden), the indignation in the rest of Germany was great. The
anger was not due to Prussia’s withdrawal from the War of the First
Coalition (1792–7), but rather to its consent in this ‘Separate
Treaty’ to the transfer of the territories on the left bank of the
Rhine to the French Republic. This was a betrayal of the Reich and
was denounced as such by Austria and the imperial patriots. Then,
in the Treaty of Campo Formio in October 1797, Austria itself
agreed to the cession of most of the territory in question, including
Mainz, as well as to the renunciation of Belgium, a gesture France
honoured by turning over the French-occupied Republic of Venice
to Austria, to whom it also held out the prospect of acquiring the
Archbishopric of Salzburg and parts of Bavaria. Thus Austria, too,
abandoned the integrity of the Reich and sank morally more or less
to the level of Prussia.
When the arrangements of Campo Formio were confirmed by
the Treaty of Lunéville nearly four years later (February 1801),
Napoleon Bonaparte had already long been reigning as First Consul
in France. After their defeat in the War of the Second Coalition
(1799–1802), the Austrians, as well as the other imperial estates,
were compelled to accept the fact that the First Man of France had
become the judge of Germany. The Final Recess of the Imperial
94
Deputation of 1803 introduced into Germany radical territorial
reforms, the purpose of which was to compensate the estates
adversely affected by the loss of territories on the left bank of the
Rhine. In the course of the ‘secularization’, nearly all the
ecclesiastical princes were stripped of their landed property. These
losses drastically curtailed the power of the Catholic Church and
German Catholicism. Through the ‘mediatization’, numerous small
princes and counts, as well as all but six imperial cities, were
forced to renounce their status as self-governing subjects of the
Reich (their Reichsunmittelbarkeit). Prussia, Bavaria, Württemberg,
and Baden were able to increase their domains substantially.
Altogether, the number of imperial estates was reduced by 112
from about 300. About 3 million people found themselves under
new rulership.
In the year after the Final Recess, Napoleon subjected the Reich
to yet another provocation. In May 1804, by means of a senate
resolution, confirmed by a plebiscite the following November, he
initiated the transformation of France into a hereditary empire. On
2 December 1804 he crowned himself Emperor of the French in
Notre Dame in Paris, after being anointed by Pope Pius VII. Vienna,
expecting the dissolution of the Reich in the wake of the senate’s
resolution, responded in August 1804 by proclaiming an Empire of
Austria for all the lands of the Habsburg monarchy, an act that
once again violated the imperial constitution. In August of the
following year, Austria joined together with Russia, Britain, and
Sweden for the Third Coalition against France. Prussia and
Prussian-led northern Germany remained neutral, as they had done
during the War of the Second Coalition. The southern German
95
states of Bavaria, Baden, and Württemberg allied themselves with
France. Four months later, Austria and its allies were defeated. The
Battle of the Three Emperors at Austerlitz on 2 December 1805 was
followed by the Treaty of Pressburg on 26 December. The
Habsburg empire had to renounce its ‘anterior Austrian’ lands to
Baden and Württemberg, Vorarlberg and Tyrol to Bavaria, and
Venice to the Kingdom of Italy, ruled by Napoleon himself in
personal union. The duchies of Bavaria and Württemberg were
elevated to the status of monarchies.
The victory over Austria allowed Napoleon to extend his
influence over Germany. In July 1806 in Paris, sixteen German
princes, including the new kings of Bavaria and Württemberg and
the Margrave of Baden (who now became a grand duke), formally
withdrew from the Reich and formed the Rhenish Confederation,
the Rheinbund. This organization became part of Napoleon’s federal
system, enjoying his protection and obliged to provide him with
military support in future Continental wars. Thus the Reich was
already in shambles by the time Emperor Francis II, submitting to
Napoleon’s ultimatum, gave up the imperial crown and released all
the estates from their duties on 6 August 1806. Although he
retained the title of Emperor of Austria (as Francis I), the Holy
Roman Empire ceased to exist on that day.13
Two eulogies had already been composed before the event.
‘German Reich and German Nation are two different things,’ wrote
Schiller in his fragmentary text Deutsche Grösse,* written probably
in 1801 after the Treaty of Lunéville.
96
princes. Set apart from all that is political, the German has
established for himself his own value, and even if the Empire were
to collapse, German worth would remain uncontested. It is a moral
greatness, one that lives in the culture and character of the nation,
a nation independent of its political fate ... While the political
Reich falters, the spiritual Reich has become ever stronger and
more perfect a structure.14
97
unification would be Prussia, which as ‘an independent, sovereign,
powerful state’ was no longer in a position to ‘enter into association
with other estates on equal terms’. Rather, he expected the imperial
cities and territorial estates to collaborate with the emperor and his
hereditary lands, which, in contrast to Prussia, themselves formed a
state based on the principle of representation and in which the
populace had rights. Thus only from the emperor may ‘we expect
support for that which the world presently understands by the term
“German liberty”’.15
Hegel’s hopes were unfounded, and they were shared by only a
few. Emperor Francis II, on the throne since 1792, had never been
interested in the Reich. Little mourning or consternation was seen
in Germany upon his renunciation of the crown. The dissolution of
the Holy Roman Empire made an impression like that of an official
obituary written for someone who had taken a very long time to
die. The Reich had been a mere shadow of its former self for
decades, ever since—at the very latest—Prussia emerged from the
Seven Years War as a great power.
This state of affairs did not, however, prevent the imperial
patriots from clinging to the idea of the Reich and believing in its
restoration. The Reich had many posthumous devotees, especially
among the ‘mediatized princes’—the counts, lords, and imperial
barons, who in 1803 had lost their self-governing status under the
old Reich—as well as among the middle-sized and smaller imperial
estates. Many publicists of the Rhenish Confederation looked to the
federative unification of a ‘third Germany’ (drittes Deutschland), that
is, a Germany without Austria and Prussia, but nonetheless
consciously taking up the legacy of the old Reich. This was also the
98
line adopted by Napoleon’s ‘Prince Primate’ at the head of the
Rhenish Confederation, Karl Theodor von Dalberg, Elector and
Archbishop of Mainz and previous Arch-Chancellor of the Reich,
elevated to the title of Grand Duke of Frankfurt in 1810. Ironically,
however, the more important princes of the new confederation,
above all the kings of Bavaria and Württemberg, stubbornly
resisted granting a new central authority any more rights than the
old Reich had possessed. Since the imperial patriotism of the
Rhenish Confederation could not alter these realities, it remained
mere doctrine and political wishful thinking.16
In terms of societal development, the territories of the Rhenish
Confederation experienced a powerful burst of modernization
during the short time the alliance lasted. Napoleon pressed the
member states to adopt ‘his’ civil legislation, the Napoleonic Code,
and was at least partly successful in his efforts. The fact that,
outside of Baden, the Code was only partially implemented was
due, in the words of Elisabeth Fehrenbach, to the ‘confrontation
between a revolutionary legal code and a pre-revolutionary legal
and societal order’.
In contrast to France, the states of the Rhenish Confederation
possessed no bourgeoisie that would have supported agrarian
reforms out of their own interests. Furthermore, there were few
noble landowners who stood to benefit from the transformation of
peasant feudal dues into money payments, and few discontented
peasants who might have collaborated with the ‘Third Estate’
against the feudal order. In Germany west of the Elbe, peasants
were directly subject to manorial legal jurisdiction far more often
than in France. The vast majority of them also exercised personal
99
control of their production, were often part owners of their lands,
and were generally less encumbered by feudal burdens than their
counterparts in France under the last Bourbon kings. The German
nobility, for its part, was no ‘court nobility’, as in France since the
time of Louis XIV. While the wealthy French aristocrats controlled
their estates only indirectly through tenant administrators, the
noble landowners in western Germany leased nearly all of their
holdings directly to the peasants. Accordingly, as Eberhard We is
has remarked, the west German nobility was less alienated from
their peasants and soil than the French nobility.
The social development experienced by the states of the Rhenish
Confederation was due not least to the necessity of integrating new
territories whose populations often had a different confession than
the country to which they were annexed. This was especially the
case in Bavaria, which acquired Evangelical areas, and
Württemberg, which acquired Catholic regions. In Baden, too, the
annexations from the anterior Austrian provinces strongly
increased the percentage of Catholics in the population. These
large-scale geographical redivisions led to a sort of delayed
revolution from above, pushed through, as in Prussia and Austria,
by an enlightened bureaucracy. Events took a comparatively mild
course in Bavaria and Baden under the ministers Montgelas and
Reitzenstein. In Württemberg, however, King Frederick
autocratically disregarded the ‘good old laws’ of a country that
was, until that point, co-ruled by the estates. This challenge was
answered ten years later, after Napoleon’s final downfall, when
Frederick attempted to impose a constitution, leading to Germany’s
first constitutional conflict.17
100
Far more radical than in the Rhenish Confederation were the
transformations affecting the regions on the left bank of the Rhine
annexed to Napoleonic France. Here, the Napoleonic Code was
imposed in its totality (and remained in force until the introduction
of the Civil Code on 1 January 1900). To this was added the
adoption of the French legal procedure and court system, as well as
French governmental administration, including the division into
départements. The areas of Germany ruled by members of the
Bonaparte family (the Grand Duchy of Berg, established in 1806 on
the lower Rhine under Napoleon’s brother-in-law Murat, and the
Kingdom of Westphalia under Napoleon’s youngest brother Jerome,
formed in 1807 after the defeat of Prussia) also experienced
profound changes. Ironically, however, these showcases of direct
Napoleonic rule in Germany were also prime examples of a
dilemma. The measures Napoleon had to take in order to
consolidate his rule fundamentally contradicted the political-
societal goals he proclaimed. On the one hand, if they wished to
gain the lasting loyalty of the bourgeoisie and the peasants, the
‘Napoleonists’ were compelled to destroy the traditional feudal,
manorial, or seigneurial structures after the model of revolutionary
France. On the other hand, however, Napoleon needed the
expropriated property and estates to give to his victorious
commanders and provide a material basis for the new ‘imperial
nobility’. The second of the two goals won out over the first, thus
stripping considerable appeal from the ‘ideas of 1789’.18
A half-century later, in his Geschichte der sozialen Bewegung in
Frankreich* (1850), the Hegelian Lorenz von Stein looked upon
101
Napoleon as a great bridge-builder. By establishing the Empire,
restoring the nobility, and negotiating the Concordat with the
Roman Church, Napoleon aspired to build bridges into the past,
between revolutionary France and a not yet revolutionized Europe.
In this view, Napoleon embodies a new era in the conflict between
revolution and traditional society.
In a first phase, which Stein dates from 1789 until the end of the
War of the Second Coalition in 1802 with the Anglo-French Treaty
of Amiens, Europe’s goal was ‘simple expulsion’ of the foreign body
of revolution.
With the entrance of Napoleon, this first stage of the battle is over.
The French movement has been victorious against the attack from
the rest of the organism; the new France has become a recognized
power ... The state of things set up by the Treaty of Amiens
recognized a feudal Europe and a civil France, a part significantly
different from its whole, and thus acknowledged an absolute
contradiction as the basis of a European peace ... In consequence,
the state of war was not brought to an end with that treaty, and the
real question was not answered. The struggle had to begin again,
now with a new character. Instead of acting purely on the
defensive, France began to intervene positively in the organism of
states. Up to this moment it had been the element that set the
others in motion; it now became the constitutive element for the
European powers.’19
102
that is, a progressive agenda purified of Jacobin terror—to achieve
success in all of Europe along a course of evolutionary development
obviating the necessity of a radical break with the past. The fact
that in Germany, too, Napoleon’s relatives were more interested in
power than in social change contributed much toward the
disappointment of these high expectations.
Yet even if Napoleon had acted rigorously in the political and
social reform of Germany, he would still not, in the eyes of most
Germans, have been freed from the stigma of being the ruler of a
foreign universal monarchy. Deutschland in seiner tiefsten
Erniedrigung* was the title of an anonymous anti-Napoleonic
pamphlet disseminated by the Nuremberg bookseller Johann
Philipp Palm in 1806. The price for the patriotic deed was high. At
Napoleon’s command, Palm was condemned to death by a French
military tribunal and shot on 26 August 1806 in Braunau on the
Inn. The cause of German nationalism had its first martyr.20
103
in Dresden (published in Leipzig) in 1806 by Friedrich Gentz, the
translator of Burke and one of Napoleon’s staunchest opponents.
‘Europe has fallen because of Germany, and through Germany it
must arise again,’ Gentz writes.
104
to style themselves the guardians of ‘German liberty’, were among
the ‘destroyers of our liberty’. The liberty he had in mind, however,
was different from that of 1789. It was not to be a democratic
liberty, but a liberty of the individual estates, the precise nature of
which he did not describe. The antagonism to the French usurper
was of fundamental significance in Gentz’s national outlook, but he
did not take the next logical step: his text was not an appel au
peuple, summoning the German people to battle against Napoleon.
Instead, he turned to the educated classes. Gentz was ‘modern’ to
the extent that he viewed the German nation no longer as merely a
cultural unity, but rather as a political organization, one that had
the potential to form a state. Nonetheless, he was not a
representative of modern nationalism. To inspire a mass movement
could not have been further from his intentions. He remained a
‘European’, one who thought in terms of a European balance of
power. No longer imperial-patriotic or ‘culturally national’, not yet
nationalistic in the sense of a secular religion—Gentz’s 1806 text
documented a transition.21
The year 1806 marks a caesura in German history for yet
another reason. In this year, Prussia renounced the policy of
neutrality it had maintained since 1795, declaring war on France
on 9 October after entering an alliance with Russia. Five days later
it suffered its first severe defeat. The double battle of Jena-
Auerstädt cleared Napoleon’s path to Berlin, which he entered on
27 October. There followed the victory over the Russians at
Friedland in 1807 and the French occupation of Königsberg. In the
Treaty of Tilsit, concluded in July 1807, Prussia lost its possessions
west of the Elbe as well as most of the territories it had gained
105
from the three partitions of Poland after 1772. Of Frederick the
Great’s Prussia only a fragment now remained. The defeat was, to
be sure, a catastrophe, yet the experience was clearly necessary in
order to release the forces within the Prussian state that were
capable of and resolved upon reform.
The moral reserve army that took up the work of inner reform
after 1806 consisted largely of non-native ‘Prussians by choice’.
Baron Karl Stein, leading minister in 1807 and 1808, was the son
of an imperial-knightly house in Nassau; KarlAugust von
Hardenburg, the politically more significant, long-time chancellor
of Prussia, was a Hanoverian; Gerhard Scharnhorst, the reformer of
the military, came from a family of peasants and soldiers in Lower
Saxony; Field Marshal Neidhardt von Gneisenau was born in Saxon
Schildau. The Prussian reformers intended, and effected, the
fundamental modernization of state and society. They abolished
the inherited servitude of the peasantry and established the right—
thereafter anchored in law—of communal self-governance for the
urban middle classes (that is, for those who met a certain minimal
income); they placed Jews on a fundamentally equal legal footing
with the other Prussian subjects (without, however, giving them
access to civil and military positions within the government); they
introduced universal conscription and freedom of trade (thus
abolishing compulsory guild membership); they instituted a state
ministry divided into departments and a reformed system of
education throughout the country, the crowning achievement of
which was the founding of the University of Berlin in 1810—an
institute dedicated to the new humanistic ideal of a general
education, which abandoned the old model of learning by rote
106
hitherto dominant in Prussia, as in the rest of Europe.
Did the Prussian reforms really add up to that ‘revolution from
above’ announced in 1799 by Struensee? This question can only be
answered when one considers what the reforms did not achieve.
The ‘peasant liberation’ was a major watershed in German social
history, but its outcome was just the opposite of Struensee’s
expectation that there would be no privileged class in Prussia
within a few years. The manorial system emerged from the
agrarian reforms, which the Junkers contested bitterly, even
stronger than before. The property concessions, through which the
peasants owing draught labour services could buy their way out of
their duties to the lord, extended the holdings of the big
landowners at the expense of peasant land. Agrarian capitalism was
introduced into the Prussian manorial system; the free exchange of
goods permitted wealthy bourgeois to purchase landed estates.
After hard-fought battles, however, the nobility was able to retain
most of its privileges. Patrimonial jurisdiction remained in force
until 1848, manorial exemption from property taxes until 1861,
and manorial police authority until 1872. The thoroughly
authoritarian Gesindeordnung, the system of rules governing
relations between masters and servants, was not abolished until
1918; the manorial district (Gutsbezirk) remained as a unit of state
administration until 1927, and even the manorial lord’s authority
as patron of the church outlasted the Wilhelmine empire.
The Edict of Peasant Liberation from 9 October 1807 granted
the peasants freedom in a legal sense. The poorest among them,
however, the Kossäten, previously exempt from obligatory draught
labour, were unaffected by the edict and compelled to continue
107
rendering their services and payments. Even the small farmers who
were able to pay off their manual and draught obligations by
relinquishing land continued to belong to the manorial association.
Unlike in post-revolutionary France, where an independent class of
farmers with small parcels of land was recruited from among the
agricultural workers and landless peasants, in Prussia this class
became, to a great extent in any case, the ‘industrial reserve army’
of Marx: a reservoir of labour that made the industrial revolution
possible in Germany. Prussia’s peasant liberation thus contributed
both to the preservation of a pre-modern ruling class, the East
Elbian Junkers, and to the modernization of German society. Of
course, this modernization was quite different from the one
envisioned by the reformers of 1808.
A national system of representation was also one of the goals of
the Prussian reformers in Stein’s and Hardenberg’s circles. This was
not to be a system of direct representation through general
elections, but rather an advisory council in which indirectly elected
agents of the propertied and educated classes should have a voice.
‘Political consultation as political participation on a short leash, but
no active co-determination and sharing of power’—such was, in the
judgement of the historian Hans-Ulrich Wehler, the vision of the
reformers. Yet a national representation did not come to fruition
even in this diluted, non-parliamentary form. The plan encountered
massive resistance, first from a conservative faction of the nobility
around Friedrich August Ludwig von der Marwitz, then even from
the enlightened bureaucrats themselves. Two congresses of
notables (the first convened in 1811 by King Frederick William III,
the second an indirectly elected group, composed almost
108
exclusively of nobles and empowered to handle very few issues,
convened the following year) were enough to convince reform-
minded bureaucrats that a representative body along estate lines
would do more to hamper than to help the renewal of Prussia.
Consequently, the promise of a constitution, which Frederick
William had made in the Edict of Finance of 27 October 1810, was
never redeemed. It was not to be the last such pledge to suffer that
fate.
If we count the revolution of industry among the consequences
of the Prussian reforms, the concept of ‘revolution from above’ is
nonetheless justified. In any case, Prussia made a great leap
forward in the years after 1807. Even if Bavaria, Baden, and
Württemberg developed into constitutional states much sooner
than Prussia, the latter acquired the leadership in the
modernization of government and society for a long time to come.
With the exception of the Kingdom of Westphalia, no other German
state pushed the emancipation of the Jews as far as Prussia, where
it was secured in law in 1812. Freedom of trade, proclaimed
through edicts in 1810 and 1811, was introduced in Austria only in
1859, in the other larger German states not until the 1860s.
Freedom of commerce, the economic counterpart of freedom of
trade, was cultivated earlier and more actively in Prussia than in
any other German state with the exception of Baden. Greater co-
determination in Prussian society would have meant less social
change: such was the view of the enlightened bureaucracy, which
claimed to be the only authentic general estate. In his
Rechtsphilosophie,* Hegel, the Prussian by choice from Stuttgart,
teaching at the University of Berlin from 1818 onwards, provided
109
the theoretical basis for a Prussian self-image that had long become
political practice. The state as the ‘reality of a moral idea’—
whoever had the privilege of serving Prussia in a high position
could recognize himself in this precept.22
If Hegel laid the foundations for a new kind of Prussian-German
ideology of the state, it was Johann Gottlieb Fichte who founded
German nationalism proper. Fichte, too, was no Prussian native; he
came from the upper Lusatian region of Saxony, where he was the
son of a weaver. In 1793 Fichte stepped into the public eye with a
defence of the French Revolution. Somewhat later he gained
notoriety with a series of theses identifying God with a moral world
order, an effort that also earned him accusations of atheism and
cost him his professorship in the city of Jena in Saxon-Weimar. In
1800 he published Der geschlossene Handelsstaat,* which he
dedicated to Struensee. This text makes the case for an isolationist
polity founded upon public property and comprehensive guild
membership; a state that, in order to secure its economic
independence, is justified in expanding to its ‘natural borders’—or
limiting itself to these, as the case may be. In the 1806 work
Grundzüge des gegenwärtigen Zeitalters,† Fichte professed an
adherence to a republican state based on the principles of reason—
a place in which, on the one hand, civil liberty is guaranteed to all,
and on the other hand every citizen is obligated to subordinate
himself unreservedly to the power of the government.
The greatest sensation was caused by the Reden an die deutsche
Nation,† which Fichte delivered in Berlin during the winter of
1807–8 and published as a book in 1808. These lectures were an
110
early document of German nationalism— modern nationalism, with
its characteristic secularization, the systematic transposition of
religious into national loyalties. As we have seen, Gentz’s Fragmente
of 1806 did not go this far. Like Gentz, Fichte addressed the
‘educated estates of Germany’ in particular, but unlike Gentz he did
so with the express intention of reaching, through them, the Volk,
from whom ‘all human progress made in the German nation has
thus far proceeded’.
Fichte elevated the Germans to the status of ‘primal’ or ‘first
people’ (Urvolk), the German language to the position of ‘primal
language’ (Ursprache). In order to substantiate these claims, he
asserted that the Germans, unlike the Romanized peoples in
western and southern Europe, had remained in the original
homelands of their ancestors, successfully defending their language
against incursion from Latin and themselves against the world
domination of the Romans. In the German cities of the Middle Ages
Fichte saw ‘excellent civil’—indeed ‘republican’—constitutions at
work. In Martin Luther he viewed the German man who had
opened the eyes ‘of all the people in the world’ to the ‘damnable
deception’ of the Roman papacy. Consequently, the Reformation
was for Fichte ‘the German people’s last great—and in a sense
perfected—deed for the world’. More than any other people, the
German Urvolk was a receptacle of ‘divine’ revelations, such as
were to be ‘eternal’. As ‘bearer and pledge of earthly eternity’, Volk
and Vaterland were justified in demanding absolute devotion from
the Germans, even the sacrifice of their lives. ‘For this very reason,
the love of the Fatherland must rule the state itself, as absolutely
the highest, last, and independent authority; and it must do this
111
primarily by limiting the state in the means selected for its most
important goal, domestic tranquillity.’
Fichte was able to invoke many phenomena that had been
regarded for centuries as German achievements and central aspects
of German character. The German humanists of the fifteenth and
sixteenth centuries, as well as many an eighteenth-century poet
(like Klopstock), had celebrated the battles the Germanic tribes had
fought to preserve their liberty from decadent Rome. Luther had
been acclaimed the spiritual liberator of Germany from Roman
foreign domination ever since the Reformation; towards the end of
the eighteenth century he was also regarded as the harbinger of the
Enlightenment. The prejudice against abundant use of foreign
words (Ausländerei), especially French, could be traced back to the
Middle Ages, when the commonality of language was the most
important characteristic of German self-awareness. Fichte was also
not the first to ascribe a particular ‘cosmopolitanism’
(Weltbürgerlichkeit) to the Germans. Goethe and Schiller had both
done so, each in their own way. What Fichte accomplished was to
politicize and reinterpret in terms of the state what the Weimar
classicists considered to be a German cultural vocation. The
‘cosmopolitanism’ informing Fichte’s idea of the German nation
meant nothing less than the spiritual-intellectual world domination
of the Germans.
‘Re-creation of the human race ...from earthly and sensual
creatures to pure and noble spirits’: such was the task Fichte
assigned to the Germans as the morally highest people. In the voice
of the ancestors who had fallen in the ‘holy battle for freedom of
religion and faith’, Fichte declared to the Germans of his time: ‘So
112
that this spirit may gain the freedom to develop itself and grow up
to an independent existence—for this reason our blood has been
spilt. It is for you to give meaning and justification to the sacrifice
by elevating this spirit to the world domination for which it has
been appointed.23
Fichte repeatedly invoked the age of the religious wars, and for
good reason. In the Reformation he saw the analogy of the present.
Luther was not only the ‘model for all future ages’, but also Fichte’s
personal model. Just as Luther had fought against the clerical
world domination of the Roman church, so Fichte fought against
the rising secular world monarchy of Napoleon. The German
Protestants who willingly shed their blood for eternal blessedness
in heaven also achieved, in the opinion of the philosopher, a
profoundly secular aim: ‘through their sacrifice, more of Heaven on
this side of the grave, a more courageous and joyful gaze up from
the earth, and a freer stirring of the spirit came into all life that
followed.’
If one followed Fichte, the existence or non-existence of a
fatherland in heaven was almost immaterial. The terrestrial
Vaterland, too, promised life after death in the form of patriotic
commemoration. The prospect of earthly immortality was, in fact,
the only thing that justified the obligation of the subject to place
his life at the disposal of the Vaterland. Patriotic love gave the state
a higher purpose, higher than the ‘typical one of preserving
domestic tranquillity, property, personal liberty, the lives and well-
being of all’. Only when this purpose informed the life of the state
could the government avail itself of its ‘true sovereign right’, ‘like
God to hazard lower life for the sake of higher life’.
113
Here, the love of the Vaterland stepped into the place occupied
by the Christian faith, the community of the nation replacing the
community of the church. God was still necessary; without the
appeal to him, the new gospel would have reached only a minority
of the educated, and the Volk not at all. The ‘God’ Fichte claimed
for the service of the German nation was, however, an entirely
secular divinity, similar to the ‘highest being’ of the Jacobins. Like
the French revolutionaries, the republican Fichte regarded love of
the Vaterland as the true religion. To the German nation alone did
he grant the right to demand the highest sacrifice, not to the
Prussian nation, as Thomas Abbt had done four and a half decades
earlier in his Vom Tode für das Vaterland. Since the German nation
did not yet even exist as a state, particular effort was required to
make it the object of sentiments of loyalty. Indeed, the sense of
belonging to a German nation was experienced by growing number
of educated Germans, who maintained contact with each other
across the borders of the individual states and who read the same
poets and thinkers in the German language. To the imagination of
the simple Volk, however, the territorial state with its sovereign at
the top was much closer than Fichte’s German nation, which had
lost all semblance of concrete reality with the downfall of the Holy
Roman Empire.
The lack of a real German state and the humiliation at the hands
of Napoleon conspired to create the national inferiority complex
out of which Fichte’s compensatory nationalism was designed to
lead the Germans. The Reden an die deutsche Nation do not enter
into the question of how, exactly, the German state was to be
organized. Fichte certainly did not have the restoration of the old
114
Reich in mind, nor the union of Germany through one of its
princes, for example the Prussian king. Five years later, however,
his Entwurf zu einer politischen Schrift im Frühlinge 1813* went a
decisive step further. He justified excluding Austria as candidate for
leadership on the grounds that its emperor would use German
power only for the interests of his own country. Prussia, however,
was different:
115
and independence and the ‘revival of the venerable Reich’. The text
even referred to a constitution for the new Germany, a work ‘out of
the deep-rooted and distinctive spirit of the German people’, over
which the ‘Emperor of all the Russias’ promised to hold his
sheltering hand. Approximately two weeks after the Kalish
manifesto, on 17 March 1813, King Frederick William delivered his
appeal ‘An Mein Volk’, which became the immediate impetus for
Fichte’s Entwurf.24
Lorenz von Stein ascribed nearly epoch-making significance to
the Kalish proclamation.
The great act, by means of which the German states renounced the
old society and recognized the new, was the Proclamation of Kalish.
This proclamation has almost the same meaning for Germany as
the Déclaration des droits [the Declaration of Human Rights
(H.A.W.)] has for France. Napoleon was defeated in name of the
constitution. Through this hope, the German people withdrew from
the cause of Napoleon. The battle against Napoleon became the
battle of a developing civil society against despotism.
116
Were declarations of intent the same as deeds, the Hegelian
Stein would have been correct in his historical characterization of
the events at Kalish. However, as Hegel puts it in his
Phänomenologie des Geistes,* ‘Only the deed itself is the truth of the
intention.’26 The promises made in the Kalish proclamation were
never carried out, whether with respect to the constitution or in
terms of German unification. The claim that the late date
represents the most remarkable thing about the proclamation can
be taken with a grain of salt. When Austria took up the battle
against Napoleon in 1809, Prussia did not come to its aid, thus
contributing significantly to the victory of the French emperor.
Only after the Grande Armée was forced to withdraw in defeat from
Russia in the autumn of 1812 did the Prussian reformers manage to
assert their long-postponed agenda of arming the people and
liberating Germany.
Besides Fichte, who died in 1814, two other classical theorists of
German nationalism belong among the spiritual pioneers of the
military action: the ‘Father of Gymnastics’ (Turnvater), Friedrich
Ludwig Jahn, son of an Evangelical minister from the vicinity of
Wittenberge in Brandenburg, and Ernst Moritz Arndt, son of a
peasant from Rügen (which belonged, as did all of Pomerania, to
Sweden until 1815). Jahn, who in 1800 anonymously published a
short text entitled ‘Über die Beförderung des Patriotismus im
Preussischen Reiche’,† turned a short time afterwards to the
promotion of ‘German folk culture’ (deutsches Volkstum), a concept
he was the first to use. By this time, Jahn was willing to recognize
only two peoples who had taken up the ‘holy idea of humanity’
117
within themselves: the Greeks of classical Hellas and the Germans.
In his 1810 book Deutsches Volkstum,* Jahn, like Fichte, assigned
the Germans the task of humanity’s salvation. ‘Hard to learn, still
more difficult to exercise is the holy office of the benefactor of the
world—yet it is a sensual delight of virtue, a divine human thing to
bless the earth as savior and to plant into its peoples the seeds of
human becoming.’
Models for the present age were ‘Hermann’, that is, Arminius,
the leader of the Cheruscans, whom Jahn called a ‘savior of the
people’ (Volksheiland); the Saxon king Henry I, and Martin Luther.
Frederick I, on the other hand, whom Jahn called ‘Frederick the
Only’ (Friedrich der Einzige), was to be a model in a strictly limited
fashion. The great Prussian king certainly understood very well
how to create a state, but not how to create a people. Without a
people, a state can only be a ‘soulless work of art’, just as without a
state a people is nothing more than a ‘disembodied, airy spectre,
like the world-shy gypsies or the Jews’.
Nonetheless, when it came to the unification and leadership of
Germany, Prussia alone came under consideration. In contrast to
the multinational empire of the Austrians (to whom Jahn sought to
make clear that their historical mission lay in the domination of the
entire south-east European Danube region all the way to the Black
Sea), Prussia, this ‘youngest, fastest-growing shoot from the old
root of the Reich’, was predominantly German in ancestry and
population. The Father of Gymnastics discovered in the Prussian
state, as he professed in the introduction to his book, ‘a force
driving toward perfection’ that promised great things for the
future. ‘Thus I imagined, in and through Prussia, the timely
118
rejuvenation of the venerable old German Reich, and in the Reich a
great people who would walk in humanity the noble path to
immortality in the history of the world.’ If Prussia chose the path
predestined for it, a golden age would dawn, and not only for
Germany: ‘If Germany, at one with itself as the German
community, can develop its immense powers, powers it has never
yet used, then one day it may be the founder of perpetual peace in
Europe and the guardian angel of humanity.’27
In Jahn’s view, the inner constitution of unified Germany was to
be liberal. The imperial congress of the estates, an elected body of
representatives selected from all of society, must be a
Sprechgemeinde (‘speaking community’, Jahn’s folkish
Germanziation of ‘parliament’), not a ‘deaf-mute institution of yes-
men and sycophants, or a collection of conformists, brought
together to make the best of a bad thing. No people can be ruled
more easily and securely than one that has a firmly founded
popular [volkstümlich] constitution.’
In order to promote the love of their own Volk, Jahn
recommended to the Germans a wide range of activities and events:
Wandern in the German countryside, folk festivals, the dedication
of popular monuments and memorials, a general German
traditional costume, a publishing industry oriented towards the
people, and above all the cultivation of the mother tongue. Yet
even with the Father of Gymnastics, it was only a short step from
the appreciation of one’s own Volk to a contempt for other peoples
and cultures. Jahn did not believe that Germans should learn
foreign languages too early in life, if at all. For those who only
cultivate them as a hobby or out of love of prattle, foreign tongues
119
contain hidden poison. This was especially true of French, which
language, in Jahn’s view, was enthralling Germany’s men, seducing
its youth, and dishonouring its women. Readers who agreed with
Jahn in this evaluation would also have agreed with his advice:
‘Germans, know again with manly pride the value of your own
noble living language; draw from its never-failing wellsprings,
uncover the old sources and leave Lutetia’s [the name of Paris in
classical antiquity (H.A.W.)] cesspool alone.’28
Jahn’s sacralization of Germandom and demonization of
everything French, extreme as it was, was surpassed in the writings
of Ernst Moritz Arndt. Originally an Evangelical theologian, Arndt
expressed the pseudo-religious character of his German nationalism
in classic fashion at the beginning of 1807:
To be one people, to have one feeling for one cause, to run together
with the bloody sword of vengeance—this is the religion of our
time. Through this belief you must be united in harmony and
strong, conquering hell and the devil ... This is the highest religion:
to conquer and die for justice and truth, to conquer or die for the
holy cause of humanity, which is being destroyed by tyranny in
vice and disgrace. This is the highest religion: to love the
Fatherland more than lords and princes, fathers and mothers,
women and children. This is the highest religion: to leave behind
for our grandchildren an honest name, a free country, and a sense
of self-worth. This is the highest religion: to protect with our most
precious blood what our fathers have gained with their most
precious, freest blood. This holy cross of the salvation of the world,
this eternal religion of community and glory, preached by Christ
120
himself—make it your banner, and after vengeance and liberation
bring the joyful sacrifice under green oaks to the altar of the
Fatherland for the God who protects you.
And a monster has been born and a bloodstained horror has arisen.
And his name is Napoleon Bonaparte, a name of woe ... Arise, you
peoples! Slay him, for he is accursed of me; destroy him, for he is a
destroyer of freedom and right ... And you must recognize in
concord and peacefulness that you have one God, the old, faithful
God, and that you have one Fatherland, old, faithful Germany.
We are now looking at just such a people’s war, one for all
Germans. Only in a general rebellion against the enemy, only
121
through a fraternal and faithful unification of all German powers
can Europe and the Fatherland be rescued and the hideous might
that threatens the freedom and happiness of the world be torn
down.
I want a hatred of the French. I don’t want it simply for this war,
but for a long time. I want it forever ... Let this hatred burn as the
religion of the German people, as a holy madness in all hearts, and
let it preserve us forever in our faithfulness, uprightness, and
courage. Let it make Germany into an unpleasant land for the
French in the future, as England is an unpleasant land for them.29
122
Vosges, Jura, and Ardennes. Nothing French shall ever be desired
by the Germans, to whom it would only bring harm.’ This demand
was repeated in ‘Der Rhein, Deutschlands Strom, aber nicht
Deutschlands Grenze’++++¶ of the same year. Language
represents the only valid natural boundary. ‘Therefore, language is
what makes the right boundary between peoples ... Whoever dwells
together speaking the same language, belongs together, as God and
Nature have intended it.’
Arndt interpreted the victory of Prussian arms in the battle
against Napoleon as a revelation of the divine will.
There are now, and have been for years, a thousand signs showing
that God has great things in store for humanity and for the German
people ... Thus the Prussian people and army have revealed their
true colours ... thus have God and God’s might and an enthusiasm,
which we cannot understand, appeared also among us. God is with
the Prussians, God is with the Germans, God has come among us,
God has wrought the great deeds through which the path of
freedom has been opened—God, and not we.
123
as the Prussian ‘western marches’. Just as Prussia has always
defended Germany on the east, it must from now on do the same
on the west against France. Germany needs a ‘leader and purifier’.
‘This great and good spirit, whose highest position in Germany
nobody can and will contest, I wish to name here for all to see and
mark: its name is Prussia.’ Austria has burdened itself, to its own
misfortune, with foreign peoples, and is not even a match for them.
‘Prussia, on the other hand, stands with all its advantages and
aspirations firmly rooted and enclosed in Germany; with Germany
it must henceforth stand or fall.’30
Like Fichte and Jahn, Arndt was a vehement critic of the
German princes, especially those of middling and lower rank. In
1808 he was still invoking the ideals of the French Revolution,
whence he believed the ‘third great era of Christianity’ (after the
early Christian church and the Reformation) would proceed. In
1813 he supported a two-house system for the future German
parliament, one representing ‘the people as a whole or the
commons’. By the next year he was professing a commitment to
democracy, though not in the French sense. To be sure, the
concepts ‘democrat, democratic, and democracy’ had become in the
last twenty years a kind of ‘rat-poison for the hearts and ears of all
good folk’, but in the words themselves lay nothing repugnant. ‘The
people are as holy as the mob is unholy. A democrat is one who
wants government for the people and by the people. Someone who
wants mob rule is called an “ochlocrat”.’[from ochlos, Greek for
‘mass’, ‘mob’ (H.A.W.)] The three estates, the nobility, peasantry,
and bourgeoisie, must have advisory and co-determinative political
power in all issues and requirements of the country; the princes are
124
to exercise executive authority ‘within boundaries established by
the general laws of Germany’.31
For Arndt, Jahn, and Fichte, there existed not only no
contradiction between unity and liberty; the appeals for the unity
of Germany and for human rights anchored in law were the two
sides of the same coin. If unity and liberty had been realized as
these authors envisioned, the existing order of things in Germany
and Europe would have been revolutionized. The intellectual
pioneers of German unification under Prussia rejected traditional
notions of the European balance of power. In 1808 Fichte labelled
the ‘artificially maintained balance of power’ a ‘thoroughly foreign
creation’ that should ‘never have taken root in the soul of a
German’. Arndt spoke of the ‘balance of power’ in 1815 as the
‘bloody puppet theatre’ on whose stage the ‘German idiot’ gets
jerked around as the ‘buffoon of Europe’. The German nationalists
of the early nineteenth century were radical reformers in matters
both domestic and international. They stood on the left in the
political spectrum of their day, closer to the ideas of 1789 than
their anti-French agitation made it seem.32
Nonetheless, their differences with the French Revolution were
many and fundamental. Since there was, as yet, no German
national state, early German nationalism was not able to orient
itself according to a political order of its own that it subjectively
felt to be ideal. As an alternative, it invoked supposedly objective
entities like Volk, Sprache, and Kultur, as if these phenomena
preconditioned any political will or choice. It did not occur to
Arndt to actually ask the inhabitants of Alsace and Lorraine if they
125
wished to be Germans. In order to incorporate them into Germany,
it was enough that they spoke German and had once belonged to
the Reich. The Swiss were to be forced by a commercial war to turn
to Germany and to feel themselves to be German. The Dutch and
Flemish were to be granted limited autonomy, as long as they were
prepared to adhere closely to Germany. To be sure, Arndt no longer
considered Scandinavia part of Greater Germany, as he did in his
youth. Nonetheless, his writings from 1814 to 1815 indisputably
bear witness to an idea that goes back to the humanists of the
fifteenth century: the concept of Germania magna, which included
all Germanic nations and whose natural leader was Germany.
Even the idea of racial purity was not foreign to Arndt. He took
pains to assert that he did not simply hate ‘the Jews as Jews’ or
consider them a naturally bad people, nor did he question the
equal civil rights of Jews born in Germany, even if he did strongly
urge them finally to convert to Christianity. Foreign Jews, on the
other hand, especially those from Poland and eastern Europe,
should never be received into Germany, which was threatened
more than any other country by a Jewish inundation. The
admission of foreign Jews is ‘a disaster and plague of our people’;
the ‘peoples’ frequent mingling with foreign substances’ spells ruin
and destruction of character. ‘The Jews as Jews do not fit into this
world and into these nations, and therefore I do not wish their
numbers to be unduly increased in Germany. I also do not desire it
because they are a totally foreign people and I would like to
preserve the Germanic race [Stamm] free of any foreign elements.’
Concerning the traditional otherness of the Jews, Fichte
expressed views even more radical than those of Arndt. In his 1793
126
book about the French Revolution, the philosopher wrote that
Jewry is ‘a powerful, hostile state, at permanent war with all the
others’, a kind of ‘state within the state’. The obstacles preventing
the Jews from achieving ‘love of justice, humanity, and truth’
seemed to Fichte insurmountable, namely their morality of double
standards and their belief in a misanthropic God. The Jews ought
not to be denied human rights,
127
fantasies of power that only the Almighty could bring about.33
The self-overestimation characteristic of Fichte, Jahn, and Arndt
are reminiscent of the Hohenstaufen aspirations to world empire
during the Middle Ages. At that time, of course, the Holy Roman
Empire was a powerful force in international politics. By the time
the classical texts of early German nationalism were written, it was
nothing more than a memory. Nonetheless, there were German
intellectuals for whom the universal aspirations of the old Reich
had been transformed into the pretensions of the Germans to be
models of humanity. The Sacrum Imperium had no chance of being
resurrected. Still, even a new Reich, in order to bear comparison
with the past, would need a kind of sacralization. This was the task
of the early nineteenth-century German nationalists. They did not
‘invent’ a tradition, but rather reconstructed traditions already in
existence. Above all, they transformed into secular language and
argumentation older religious beliefs about a German mission in
the world. From this recasting emerged the phenomenon we call
German nationalism.34
This nationalism was ‘modern’ because it aimed at a
fundamental renovation of society and strove to break with the
past. The revolutionary aspect of early German nationalism is
especially clear in comparison with representatives of political
romanticism like Novalis, Friedrich Schlegel, and Adam Müller.
While the two movements have much in common in their
idealization of Germanness, the differences are more significant.
The political romantics praised the virtues of the nobility and of a
society structured according to traditional Stände. They felt an
128
affinity with the Christian universalism of the Middle Ages,
developing a strong inclination towards Catholicism, to which
several of them, like Müller and Schlegel, converted. The early
nationalists wanted to abolish the privileges of the nobility. They
were anti-universalists and ‘cultural Protestants’ (in a wider sense
than contained in the term as it was used in the Wilhelmine era).
Prussia as antagonist of Austria, as contender for the crown of a
new German Reich—such thoughts were sacrilege to the political
romantics. The early nationalists made them into a political
agenda.35
The political leaders in Prussia during the early nineteenth
century were themselves still unable to adopt such a programme
without undermining the basis of their own rule. The founding
fathers of German nationalism were certainly not Kleindeutsche,
that is, proponents of a ‘Little Germany’ who wished to exclude
Austria (there is room for doubt only in Jahn’s case). Yet to
advocate, as Arndt did, ‘Germany as a whole’, a German state ‘as
far as the German tongue is heard’, a Germany ‘from the North Sea
to the Carpathians, from the Baltic to the Alps and the Vistula’—
this was equivalent to breaking up the Habsburg monarchy. An
anti-dynastic revolution of this kind could only be endorsed by
‘ideologues’, a term Napoleon used to brand intellectuals who
engaged in politics.
The Prussian king and his ministers were neither ideologues nor
revolutionaries. The thoughts concerning Germany’s future that
Baron Stein committed to paper between 1807 and 1815 were
contradictory, but closer in spirit to political romanticism than to
German nationalism. It was never Stein’s intention to demote
129
Austria to second place behind Prussia. He desired a good
understanding between the two German great powers and a
curtailment of the influence wielded by the other German princes.
‘The firm, constant, uninterrupted accord and friendship of Austria
and Prussia’ as the keystone of a German ‘union of states’, a
voluntary and coequal affiliation of sovereign princes—such was
also the vision expounded by Wilhelm von Humboldt in his
‘Denkschrift zur deutschen Verfassung’* of 1813. Hardenberg
imagined a German federation led by Austria, Prussia, and Bavaria.
A German national state was not among the goals of Prussia’s
ruling reformers.36
Thus the early German nationalists were lacking in firm support
from the state. The support they found from within society as a
whole, on the other hand, can be fairly clearly identified. Fichte,
Arndt, and Jahn appealed above all to educated young people—
often still at university—of the Evangelical confession. This
audience was far more numerous in northern Germany than in the
Catholic and multi-confessional south and west, which had strong
reservations concerning Prussia and a certain amount of sympathy
for Napoleon. By no means, however, were all Evangelicals
necessarily apostles of German nationalism. The patriotism
preached in Prussian churches during the wars of liberation did not
generally refer to ‘Germany’, but only to Prussia. Love of the
Prussian, not German Vaterland was the sentiment that brought
women together at that time in societies for the public welfare and
in numerous associations for the care of the poor, sick, and
wounded, the nuclei of the women’s charity organizations in the
130
Evangelical church. Prussian soldiers went to war ‘with God for
King and Fatherland’. The Vaterland was embodied in King
Frederick William III (who had invented the soldiers’ rallying cry
himself) and Queen Louise (who died in 1810), as well as in the
Great Elector and Frederick the Great—not in ‘Hermann the
Cheruscan’ or in the symbolic female figure of ‘Germania’, who
began at that time her career as a symbol of the unity of the
German people.
Such was, at any rate, the case in the countryside and the
smaller cities, where the influence of Pietism had been strongest. In
Berlin and the other large (especially university) cities, the appeal
to ‘religious patriotism’ (a concept that first appeared in June 1816
in an edict of the Royal Consistory of the Province of Brandenburg
concerning the ‘religious solemnities for the remembrance of
soldiers fallen in battle’) was not always sufficient. Friedrich
Schleiermacher, pastor at Holy Trinity Church in Berlin and
professor at the capital’s newly founded university, brought
Prussian and German patriotism together. Nonetheless his
patriotism, though religiously informed, was no political religion
after the manner of Fichte, Jahn, and Arndt. Only in a limited way
can Schleiermacher be numbered among the early German
nationalists, despite his support for a unification of Germany
through Prussia.37
Between the downfall of the old Reich and the Congress of
Vienna, an organized polity called ‘Germany’ did not exist. But
neither was Germany merely an imagined community. It existed as
a network of language, culture, and communication sharing a
common memory of the Holy Roman Empire. In the years before
131
1815, only an intellectually powerful and eloquent minority of
educated Germans looked to the founding of a German national
state. The German cultural nation was a product of education, and
education was required in order for Germans to take the further
mental step from cultural to actual nation. The educated
bourgeoisie sponsored and embodied the force that united all
Germans—Kultur. The dynasties and the landed aristocracy stood
for the phenomenon that kept them apart—particularism. If the
Volk could be convinced that its own best interests lay in subduing
and transcending the petty princedoms, as well as every other form
of particularism, then the spokesmen of the educated middle
classes had a good chance of exercising far more influence in a
unified Germany than they did at the present.
This gave rise to a paradox. As described above, early German
nationalism can be seen as an expression of the emancipatory
aspirations of the bourgeoisie, making it akin to French
nationalism. Simultaneously, however, German nationalism had to
set itself in opposition to France, since it owed its most powerful
impetus, if not its very existence, to the battle against foreign
domination by Napoleon. The victors of the battles from 1813 to
1815 were not, however, the German nationalists, but the two
German great powers, Austria and Prussia, whose prestige the
defeat of Napoleon increased considerably. Whoever sought both
national unity and civil liberty at the same time continued, in
consequence, to have a much more difficult time than the French
revolutionaries of 1789. In France, the goal was the liberation of
the ‘third estate’ in an already existing national state. In Germany,
the national state had first to be created, a state the early German
132
nationalists could only imagine as liberal.
The challenge was very great. A firm political belief system was
necessary to maintain a commitment to both unity and liberty. On
the other hand, this belief system was a source of strength and
support for those devoted to the cause. Herein lay in Germany, as
elsewhere, nationalism’s great chance: it was able to fill the
emotional vacuum left in the wake of the Enlightenment, when the
old religious beliefs and commitments were summoned before the
judgement seat of Reason. The resulting spiritual crisis was felt
primarily by the educated classes. From their ranks came—not only
in Germany—the prophets and the first apostles of the new creed of
nationalism.38
133
‘Battle of the Nations’ at Leipzig in mid-October 1813, was forced
to relinquish its northern half to Prussia. That country also gained
West Pomerania (hitherto controlled by Sweden), as well as the far
more important territories of Westphalia and the northern
Rhineland, bringing the Hohenzollern state into the possession of
the most advanced German centres of industry and making it more
‘western’ than before. Bavaria, which had withdrawn from the
Rhenish Confederation before the Battle of the Nations, joining the
alliance against Napoleon, increased its territories in Franconia,
and received back the left Rhenish part of the Palatinate. The right
Rhenish area was given to Baden. Hanover, now a kingdom,
returned to the personal union with England that had existed since
1714.
The restoration of the Holy Roman Empire was not on the
agenda at the Congress of Vienna. For the German great powers
and the larger German states, the Mittelstaaten, reasons of state
dictated that no superior imperial authority could be tolerated. In
public opinion, the old Reich had virtually no remaining support.
Nonetheless, the representatives of the German states did not
consider it advisable to simply ignore the many calls for a new and
more effective form of German unity. The German Confederation
(Deutscher Bund) was the result of the efforts to bring the forty-one
German states together that had survived the Napoleonic era or
were resurrected after 1815. The German Confederation was not a
federal state, but a federation of states: a union of sovereign
nations with a single common institution, the Federal Diet
(Bundestag), a congress of legates meeting permanently in Frankfurt
am Main.
134
The German Confederation also had three non-German
members, all monarchs: the King of England as King of Hanover,
the King of the Netherlands as the Grand Duke of Luxembourg, and
the King of Denmark as the Duke of Holstein and Lauenburg. The
two most important members belonged to the Confederation only
with the parts of their territory that had formerly belonged to the
Reich: Austria (which held the presidency) with Bohemia and
Moravia, all of Styria, Carniola with Trieste, and Italian Tyrol
around Trent, but without Hungary and most of the Italian
possessions; and Prussia, the real winner of 1815, without East
Prussia, West Prussia, and the Grand Duchy of Posen (created in
1815). The recursion to the old imperial borders had significant
consequences. Since their domains extended beyond those of the
German Confederation, Austria and Prussia were permitted to
maintain larger armed forces than the Bundeskriegsverfassung of
1821, the statutes of the Confederation applying to military
matters, otherwise allowed. They availed themselves of this
privilege. As European great powers, both Prussia and Austria had
to emphasize the fact that they were more than mere members of
the German Confederation.
In principle, the Confederation could defend itself from outside
attack. However, it was ‘structurally offence-incapable’, to use a
newer term. This lay in the interest of all European powers
participating in the Congress of Vienna, especially the five great
powers, England, Russia, France, Austria, and Prussia. As
signatories of the Final Act of the Congress of Vienna (9 June
1815), these countries also approved the Federal Act (Bundesakte),
given final form the previous day. Although this charter granted
135
the non-German great powers no formal right of intervention in the
affairs of the Confederation or any of its member states, it did
entitle Europe to expect that the status quo fundamentally be
maintained in Germany. The German Confederation as the pledge
of a European balance of power—on this matter there was
consensus among all the great powers in 1815, including both the
German ones.
Domestic military intervention by the Confederation was
permitted (granted the necessary majorities in the Federal Diet) if a
member state acted in a manner inimical to the Confederation or if
its leadership saw itself threatened by rebellion or overthrow.
Furthermore, according to Article 13 of the Federal Act, the
participating states were required to introduce ‘constitutions of the
territorial estates’ (landständische Verfassungen). When and how
they chose to do so was their affair, and the actual meaning of the
term landständische Verfassung was uncertain. It could mean a
traditional corporative constitution, in which each Stand spoke and
made decisions only for itself. Equally, it could describe a modern
constitution representing the whole of the population. Of the first
German constitutions, the second type was represented by those of
Bavaria and Baden in 1818, that of Württemberg in 1819, and that
of Hesse-Damstadt in 1821, while the constitutions of several small
states in middle Germany, such as Saxony-Weimar (from the year
1816), corresponded to the first type. The plan of Chancellor
Metternich of Austria, who wanted to limit the member states to
traditional corporative constitutions, failed. Nonetheless, the Final
Act of 1820 saw to it that the rights of the representatives of the
Landstände were restricted in favour of the ‘monarchic principle’,
136
that is, the concentration of political power in the highest office of
the state. This measure contributed in great measure toward
blocking any further progress within the constitutional movement
until 1830. The two German great powers, for their part, would
submit to no constraint (except perhaps by their own people) when
it came to introducing a constitution.39
All those who had striven for a liberal, unified Germany during
the time of the Napoleonic Wars were bitterly disappointed by the
results of the Congress of Vienna. In April 1815 Joseph Görres, at
one time a Coblenz ‘Jacobin’, later one of the most passionate and
effective of Napoleon’s literary enemies and an admirer of Prussia,
wrote in his newspaper the Rheinische Merkur that the post-war
German order then emerging was a ‘miserable, misshapen,
deformed constitution, defective at birth ... four-headed like an
Indian idol, without strength, without unity and cohesion’, ‘the
mockery of future centuries’, ‘plaything of all the neighboring
peoples’. In this verdict, the Catholic publicist could be assured of
the agreement of those who went even further than he, hoping for
a German emperor from the House of Hohenzollern. The outrage
was especially great among the students, many of whom had
enlisted and fought voluntarily against Napoleon in the wars of
liberation. The most famous volunteer corps, that of Major von
Lützow, was the first to use the colours black, red, and gold, which
the newly founded ‘Fraternity of Jena’, at Jahn’s suggestion,
established as their symbol of union in 1815. The fraternities
(Burschenschaften), spreading rapidly over all of Germany, were at
first national—that is, all-German—and mostly democratic in spirit,
qualities through which they consciously distinguished themselves
137
from the regional student associations, the Landsmannschaften.
The first large public demonstration of the fraternities was the
Festival of Wartburg on 18 October 1817, which was held to
commemorate the 300th anniversary of Martin Luther’s
Reformation as well as the fourth anniversary of the Battle of the
Nations at Leipzig. The official ceremony proceeded without
incident; patriotic and liberal speeches elicited the applause of the
participants, who numbered about 500. Then, in the evening, the
gymnasts provoked a sensation, burning ‘un-German’ books
including the Code Napoléon, Samuel Ascher’s sarcastic polemic
Germanomania, and the southern German particularist journal
Allemania, as well as several hated symbols of absolutism: a
Hessian military pigtail, a corset of the Prussian uhlans, and the
stick of an Austrian corporal. Jahn’s followers, whether gymnasts
or fraternity members, also gave vent to their anti-Jewish and anti-
French feeling. In order to feel secure in their ‘Germanness’, they
apparently needed to differentiate themselves as sharply as possible
from everything they felt to be ‘un-German’.
The conservative powers, led by Metternich, were alarmed.
Revolution again seemed about to rear its head, this time in
Germany. Two years after the Festival of Wartburg, the
governments struck. The occasion was the murder in March 1819
of August von Kotzebue, Russian counsellor of state and German
comedic playwright, by Karl Ludwig Sand, a theology student at
Jena and member of a fraternity, as well as the failed attempt by
another fraternity member on the life of a councillor of Nassau,
Karl von Ibell, at the beginning of July. In the Teplitz Accord of 1
August 1819, Austria and Prussia agreed to harsh measures for the
138
control of universities, press, and parliaments. Later that same
month, the most prominent members of the Confederation adopted
the Karlsbad Decrees, which were approved in the Federal Diet in
September. They contained the legal titles for the dismissal of
professors disseminating ideas critical of the status quo, the
‘suppression of demagoguery’ throughout Germany, a ban on the
fraternities, and the censorship of newspapers, journals, and
printed matter less than twenty pages in length. Among the victims
of the ‘suppression of demagoguery’ was Jahn, who was imprisoned
from 1819 to 1825, as well as Arndt, who lost his professorship in
Bonn in 1820 and was only rehabilitated after 1840 when
Frederick William IV came to the Prussian throne. Teplitz and
Karlsbad signal the real beginning of the system of ‘restoration’ in
Germany. The German Confederation now assumed the role that
has impressed itself upon historical memory, that of an instrument
of suppression aimed at all liberal and national aspirations.40
The years 1815–19 have great historical significance for the
German unification movement. Between the end of the old Reich
and the overthrow of Napoleon, German nationalism was mainly
confined to educated circles. Between the Congress of Vienna and
the promulgation of the Karlsbad Decrees, it began to spread
among the Volk and become publicly organized (in contrast to the
clandestine organizations of the years after 1808). While the
members of the fraternities, as well as the students as a whole,
came predominantly from the educated middle classes, many
trades-men—masters as well as journeymen—became active among
the gymnasts. The same was true of the men’s choral societies
founded in the 1820s. By 1819, when the Federal Diet declared war
139
on it, the national idea was strong enough to defy the system of
repression, even though it did not yet enjoy the support of the
masses.
However, the meaning of the concept ‘national’ was subject to
considerable change in the atmosphere of Restoration politics,
especially in their Prussian version. The prestige of Prussia had
increased so greatly during the wars of liberation that even
Germans outside the Hohenzollern state, especially members of the
newly founded German Societies, began to entreat Prussia to take
up the work of German national unification. The manner in which
Prussia developed and represented itself after 1815, however, was
not calculated to win the hearts and minds of liberal forces in other
German states. On 22 May 1815 King Frederick William III once
again promised his subjects a constitution, and once again the
promise came to nothing. This time not only backward-looking
nobles opposed the move, as in the years before, but also the
bureaucratic elite itself, who wished to push forward the economic
modernization of Prussia without being hampered by conflict with
corporative and regional interests. In the 1819 Teplitz Accord,
Prussia resolved not to introduce general popular representation,
but only a central committee of representatives from provincial
congresses of the territorial estates. The year after the death of
Chancellor Hardenberg, 1823, saw the enacting of a law
introducing provincial estates (Provinzialstände), which granted the
nobles and large landowners a secure predominance over the urban
bourgeoisie. Thus the Prussians saw their hopes of national
representation and a written constitution disappointed yet again.
The struggle for national representation in Prussia between 1815
140
and 1823 was described by Friedrich Meineke as the first phase in
the battle between the ‘ common state’ or ‘state as community’
(Gemeinschaftsstaat) and the ‘authoritarian and militaristic
principle’. For the bourgeoisie, the outcome of this first test of
strength was as negative as were to be those of the second, the
revolution of 1848–9, and the third, the Prussian constitutional
conflict of 1862–6. It is true that, through its bureaucratic reform
from above, Prussia, which after 1815 extended from Memel to
Saarbrücken (though not in one continuous land mass), was able to
make great and lasting progress in its economic development and
inner unification. In the years after the Congress of Vienna, Prussia
became the leading industrial power on the Continent. At the same
time, however, the express denial of the right of its citizens to have
a voice in its politics made Prussia seem retrograde, lagging behind
the southern German constitutional states of Baden, Württemberg,
and Bavaria— which, in their turn, made far slower economic
progress than did Prussia.
As long as it had neither constitution nor parliament, the
Hohenzollern state could not be the adversary of Austria, which as
a multinational polity was the born enemy of every national
movement. Since both German great powers had other agendas
than that of German unification, the national idea lacked a clear
state-oriented political profile during the Restoration era. In other
words, whoever invoked the idea of German unity in the decade
after Karlsbad scarcely thought in terms of a national state, and
certainly not one led by Prussia. The goal of the nationally minded
was now more modest: they sought to consolidate the Germans’
consciousness of belonging to one nation despite the existence of
141
many German states. The liberal idea, narrowly linked with the
national, found even less of a home in Restoration Prussia than
before; liberals now gravitated towards the southern German states.
While these, too, were pillars of the Restoration system, at least
they remained ‘constitutional’. They kept their constitutions,
which, with the exception of Württemberg’s of 1819, were all
imposed (oktroyiert) by the ruling prince in a one-sided manner.
The southern German constitutions all possessed a two-chamber
system: the first chamber a kind of upper house or house of lords,
the second chamber representing the populace as a whole. In this
and other ways they were modelled upon the French Charte of
1814.41
In Germany, however, this paradigmatic structure of a
constitutional monarchy was understood differently than in France.
For most French liberals (the concept ‘liberal’, originating in Spain,
spread throughout Europe after 1812), the strict separation of royal
authority and parliament—executive and legislative, in other words
—was a transitional stage on the way to a parliamentary system in
which the legislative body (or, to be more exact, its second
chamber) was to be the politically decisive organ of government.
This is also the way Benjamin Constant, one of the representative
thinkers of early political liberalism much read in Germany, saw
things. In comparison, the German liberals of the early Restoration
era generally considered the dualism of parliament and
administration a permanent and indispensable fixture of
government. To think of oneself as ‘liberal’ in Germany meant to
support the strengthening of basic individual rights, above all
freedom of the press and of association, and to advocate the
142
exclusive right of the parliament to approve taxes and ratify laws.
Early German liberalism was simultaneously national and
international in scope. The goal of close cooperation between
German friends of liberty did not exclude solidarity with the
emancipatory efforts of other peoples. When the Greeks began to
rebel against Turkish rule at the beginning of the 1820s, they found
active and enthusiastic support among liberals in western and
middle Europe. In Germany, the philhellenic movement became a
symposium for liberal and democratic opposition against the
politics of the Restoration—an opposition all but muted in its
parliamentary form since the promulgation of the Karlsbad
Decrees. Moreover, within a short period of time, the friends of
Greece built a network of communications that traversed the
boundaries of the particular states and showed signs of becoming a
truly national organization of liberalism. The early nineteenth-
century Greek shepherds and small farmers had virtually nothing in
common with the heroes of classical Hellas. Nonetheless, the
German philhellenes consciously invoked the myth of Greece in
order to mobilize public opinion for a cause that was not primarily
Greek, but German in aim: the creation of a liberal society in all of
Germany.42
A completely different kind of mobilization took place in
Germany in the summer of 1819. Beginning in Würzburg, anti-
Jewish rioting, the so-called ‘HepHep Riots’, took place in many
cities, including Frankfurt am Main, Hamburg, and Heidelberg, as
well as in a number of villages. Jewish houses were attacked and
destroyed, and Jews were mocked wherever they went with the
call ‘hep! hep!’ and subject to physical aggression if they defended
143
themselves. The initiators of the riots came mainly from the ranks
of the small tradesmen and merchants. The primary motive was the
desire to eliminate Jewish competition. Indeed, the legal and
economic position of the German Jews had much improved since
the French Revolution. In a sense, the 1819 riots can be interpreted
as a protest against the beginning Jewish emancipation. The
dominant sentiment was no longer the Christian prejudice against
the supposed ‘murderers of God’, but rather a resentment against
the Jews as the beneficiaries of the process of modernization. While
already chiefly secular in nature, however, the anti-Jewishness of
the Restoration era was, in general, not yet ‘racist’ in expression. In
a word, it was no longer religious anti-Judaism and not yet
‘modern’ anti-Semitism. The new orientation of the prejudice did
not, in any case, mean that the old image of the Jews no longer
applied. The tradition lived on in conjunction with the latter-day
resentments.
For the conservative opponents of liberalism, the anti-Jewish
hatred of artisans, merchants, and peasants was perfectly
expedient. At the beginning of 1811, the Brandenburg landowner
Friedrich August Ludwig von der Marwitz conjured up the danger
of ‘our venerable old Brandenburg-Prussia’ becoming ‘a new-
fangled Jew state’ in the wake of Jewish legal emancipation. For
the anti-liberal propagandists, the emancipation of the Jews,
freedom of trade, and the free movement of labour were practically
all of a piece, and the figure of the Jew served as symbol for all the
forces that threatened custom and tradition.
But liberals, too, were in no way unanimously in support of the
civil equality of the Jews. Many of them considered the Jews
144
inveterate enemies of progress, a view that even enlightened
authors like Voltaire had held. For many liberals, it made sense to
postpone emancipation until the Jews could be re-educated and
elevated to the cultural level of the present age. In 1833 Carl von
Rotteck spoke for the majority of liberals and their constituents in
Baden when he urged delaying the Jewish civil emancipation until
a time
when the Jews cease being Jews in the strict and fixed sense of the
word, for the Jewish religion is a religion that contains and
promotes a fundamental enmity, or at least strong reservations,
against all other peoples, whereas it is the nature of the Christian
religion to strive to make brothers out of all nations of the earth.43
145
were held in high regard throughout all of Europe, and their
vitality was nowhere so great as in Germany. In France and Britain,
too, Romanticism was a powerful force challenging the
Enlightenment, but it was far more successful in Germany. Inspired
by Herder, the Romantics had discovered the ‘spirit of the people’
(Volksgeist) everywhere in Europe, but nowhere did they shape this
spirit so lastingly as in Germany. The ‘other’ and the ‘foreign’, from
which the German Volksgeist sought to differentiate itself, had
many faces, but none so sharply defined as that of the Jew.
It is true that hatred of the Jews was a pan-European
phenomenon. In 1792 revolutionary France became the first
country to introduce Jewish civil equality— a measure for which a
German author, Christian Wilhelm Dohm, was among the first to
make a case in 1781. Restoration France, too, witnessed broad-
based opposition to the social advancement of the Jews, an
advancement that continued there, as in Germany, despite all
resistance. The main difference between the two countries lay in
the fact that a countervailing force against the prejudices of the
anti-Semites was present west of the Rhine that was lacking to the
east: the proud memory of self-empowered liberation in the form of
a revolution.45
The memories of 1789 were still very much alive when, on 27
July 1830, petty bourgeois and workers, led by journalists and
lawyers, rose up to overthrow the restored Bourbon monarchy. The
immediate cause of the revolt was a series of ordinances by King
Charles X abolishing freedom of the press and curtailing voting
rights. The old regime was brought down after a three-day battle in
the streets of Paris. From the July Revolution emerged the July
146
Monarchy, the ‘civic kingship’ of Louis Philippe, formerly duke of
Orléans. The Restoration era had come to an end in France, a
circumstance that had consequences far beyond the French borders.
Once again, a French revolution had become a turning point in
European history.
147
3
148
then it could be predicted that the ranks of the proletariat and the
sufferings of the working class would increase as industrialization
took hold of the Continent. That much could be observed in
Britain, the motherland of the Industrial Revolution.
For the time being, however, the political consequences of the
new French revolution claimed the lion’s share of attention. The
events in Paris in July 1830 divided Europe into a liberal west and
a conservative east. To the one camp belonged Britain and France,
to the other Russia, Austria, and Prussia, the three states that had
inaugurated the ‘Holy Alliance’ at the Congress of Vienna in 1815
for the defence of Christianity and the old order. The signal that
went out from Paris alarmed the ruling classes, especially in the
absolutist monarchies, and gave heart to their subjects, to the
extent that their political consciousness had already been
awakened. The first in the series of national revolutions that
followed the French took place in the autumn of 1830 in Belgium,
under Dutch rule since 1815. The result was a new state, founded
with the assent of the European great powers: the Kingdom of
Belgium, which straightaway became the very model of the liberal
European state.
Completely different was the outcome of the second revolution
in the series, the Polish Revolution, which began in November
1830. In the wake of bloody battles, ‘Congress Poland’, the main
part of the country, after 1815 again allied with Russia, lost what
independence it had been granted by Tzar Alexander I. The 1815
constitution was abolished, and Congress Poland sank to the status
of a Russian province. In the winter of 1831–2 began the ‘Great
Emigration’: between four and six thousand Poles crossed through
149
Germany into France. There, as in other countries in Europe, the
Polish emigrants became very active in the cause of liberalism and
democracy. In the early 1830s the call to support the Polish
patriots had a galvanizing effect similar to the enthusiasm
generated by the Greek war of independence a decade earlier.2
In Germany the July Revolution had an effect similar to the
breaking of a spell. A wave of political and social unrest shook
many cities, among them Hamburg, Cologne, Elberfeld, Aachen,
Munich, Berlin, and Vienna. In the Mittelstaaten Brunswick and
Saxony, two of the politically most backward countries in
Germany, the protest against the status quo quickly escalated into
open revolution. In both cases, the sovereign (in Brunswick the
duke, in Saxony the king) were compelled to give up the throne to
another member of the ruling family. The unrest also spread to
Hanover and the Electorate of Hesse, which underwent a change of
government in consequence. All four states saw the introduction of
representative constitutions (leading in the Electorate of Hesse to a
constitutional conflict lasting several years). In southern Germany,
the July Revolution ushered in the era of ‘chamber liberalism’
(Kammerliberalismus), the stronghold of which was Baden, where
the liberals took control of the newly elected second chamber at
the beginning of 1831.3
If we look for a pithy expression that sums up the liberal agenda
of the Vormärz, the ‘Pre-March’ period between the French July
Revolution of 1830 and the March 1848 revolutions in Germany, it
would be the formula ‘liberty and unity’. Liberty was the cause of
the Baden liberals when they made a stand against the censorship
150
system of Karlsbad, forcing the government in December 1831 to
pass progressive legislation concerning the freedom of the press—
legislation that contradicted the statutes of the German
Confederation and was revoked several months later (July 1832)
by the grand duke at the behest of the Federal Diet. German unity
was the aim of a petition lodged in the state parliament of Baden in
October 1831 by the author of the press law, the Freiburg teacher
of constitutional law Carl Theodor Welcker, together with Carl von
Rotteck, editor of the well-known Staatslexikon. In this ‘motion’ the
liberal representative exhorted the government in the name of the
German people and German liberalism to support the cause of
national parliamentary representation. In order to reach the goal of
an ‘organic development of the German Confederation to the best
possible support of German national unity and German civil
liberty’, an elected People’s House ought to take its place at the
side of the delegate congress, the Federal Diet.
Another famous liberal declaration of German national unity
was made that same year. In his Briefwechsel zweier Deutschen,* Paul
Pfizer, jurist and deputy in the second chamber of the Württemberg
government, made an appeal for the ‘spiritual unity’ of the German
nation through the pen of one of the fictive correspondents,
‘Friedrich’. In the second edition of the book, published the
following year, the same character endorsed the peaceful
development of the German Confederation to a federally organized
constitutional state. The other correspondent, ‘Wilhelm’, advocated
a Prussian protectorate over Germany. While Catholic Austria had
become ever more alienated from Germany in the course of the
previous three centuries, a more open-minded development of its
151
own system was to be expected from Protestant Prussia. Yet
‘Wilhelm’ (whom many contemporaries considered to be the actual
mouthpiece of the author, for good reason) went even further. The
Germans must learn from the French, he wrote in the second
edition of the Briefwechsel, ‘that national independence must
precede even civil liberty and is more sacred than the latter’4
Other south-west German liberals were convinced that both
paths were doomed to failure. Neither in a limited reform of the
Confederation’s constitution nor in a Prussian hegemony lay
Germany’s salvation, but rather in the German people themselves.
Among the most eloquent advocates of this school were the
political writers Philipp Jacob Siebenpfeiffer from Lahr in Baden
and Johann Georg August Wirth from Franconian Hof. In January
1832 these two men founded the German Patriotic Press
Association (Deutscher Press- und Vaterlandsverein) in Palatine
Zweibrücken.
It was no accident that this initiative came from the Bavarian
Palatinate. The particularist mentality was much weaker there than
in the rest of the German south-west. Very few inhabitants of the
Palatinate had developed emotional ties to the Wittelsbach dynasty
after its rule was re-established in 1815, whereas many preserved
vivid memories of the civil freedoms of the ‘days of the French’.
When King Ludwig I dissolved the very self-assured Bavarian
parliament, convened the year before, in December 1831, the
protest was loudest in the left Rhenish Palatinate. The founding of
the Press Association—a party-like union of like-minded spirits that
rapidly spread beyond the Palatinate—was the first organizational
fruits of the uprising. The second was the largest demonstration of
152
German patriots and friends of liberty that had ever yet been held:
the Hambach Festival, convoked by Siebenpfeiffer, Wirth, and
others at the end of May 1832.
By the time the 20,000–30,000 participants (among them
students, tradesmen, and vintners, predominantly from south-west
Germany and above all from the Palatinate itself) gathered in the
ruins of Hambach Castle, the Press Association, as well as
Siebenpfeiffer’s and Wirth’s two newspapers, had already been
banned, the former by the Bavarian government, the latter by the
Federal Diet at Frankfurt. The All-German Festival
(Allerdeutschenfest), however, could not be stopped in time. On 27
May 1832, under banners of black, red, and gold (the official
colours of the German unification movement from that day
forward), the speakers declared to a jubilant audience their belief
in a free and democratic Germany, to be established as a unified
republic against all the resistance of the princes.
In Siebenpfeiffer’s words, the new Germany was to be a country
where
the German woman, serving maid of the ruling man no longer, but
now the free companion of the free citizen, shall nurse our sons and
daughters as babbling infants upon the milk of liberty and nourish
the meaning of true citizenship in the seeds of edifying language;
where the German maiden will recognize as most worthy the youth
who burns with the purest zeal for the Fatherland ... where noble
Germania shall stand on the pedestal of liberty and right, in one
hand the torch of enlightenment shedding the light of civilization
to the farthest corner of the earth, in the other the scales of
153
judgement, granting the sought-after law of liberty to warring
peoples—the very peoples who gave us the law of brute force and
kicked us with scornful contempt.
154
other countries when they used the term.
The moderate liberals of Germany, who backed reforms within
the existing German states, were repelled by the republican
rhetoric of the Hambach Festival. Here were the first signs of the
parting of the ways between ‘democrats’ and ‘liberals’ in the
narrower sense. And the liberal opposition to the national
aspirations of the All-German Festival was not restricted to the
question of the form of government. For Rotteck, whom the Baden
government had forbidden to travel to Hambach (as it did all state
employees), the push for German unity constituted a real threat to
the goal of liberty. At a festival of Baden liberals in Badenweiler a
few weeks after Hambach, Rotteck clearly stated his priorities:
I desire unity only with liberty, and I would prefer liberty without
unity to unity without liberty. I want no unity under the pinions of
the Prussian or Austrian eagle, nor under a strengthened Federal
Diet, as long as it retains the present form of organization. I do not
even desire unity under a general German republic, because the
way leading thither would be terrible, and the success or fruit of its
achievment seems highly uncertain in character ... In effect, I desire
no German unity sharply defined in its external form. As history is
my witness, a federation of nations is better able to safeguard
liberty than the undivided mass of a large empire.5
155
severe interpretation of the Karlsbad Decrees of 1819, the so-called
‘Six Articles’. When, in April 1833, a group of fraternity members
failed in a putsch-like attack against the Diet in Frankfurt (called
the Frankfurter Wachensturm, ‘the Storming of the Frankfurt
Bastions’), the latter deployed federal troops to secure the city. The
sweeping investigations, trials, and sentences that followed were
far more draconian than those to which some few of the main
participants at Hambach had been subjected. Despite these
circumstances, however, the ‘general subversion party of Europe’
Metternich saw at work in both Frankfurt and Hambach did not yet
exist in Germany, except in the wishful thinking of a few radicals
or in the nightmares of the ruling classes.
It was not until 1834, the year after the events as Frankfurt, that
the secret society ‘Young Germany’ (Junges Deutschland) was
founded in Switzerland by German emigrants, mostly academics
and journeymen in the craft trades. After a short time, the
organization joined ‘Young Europe’, the revolutionary international
founded during the same period, and in the same country, by
Guiseppe Mazzini. Young Germany shared its name, its radical
critique of political repression in Gemany, and its endorsement of a
German republic with a literary circle founded in 1830 around the
writers Ludwig Börne, Heinrich Heine, and Karl Gutzkow. In
contrast to these men of letters, however, the members of the
secret political society actively set about preparing for revolution.
In the years following, a diffuse network of contacts, affiliated
groups, and points of support was developed between the Black
Forest and the Odenwald. Nonetheless, only a tiny minority of
Germans were engaged in revolutionary activities of this sort. An
156
overthrow of the existing order was not to be expected from this
quarter for the time being.
Nor was there a great deal at stake politically when, at the end
of 1837, liberal Germany showed its solidarity with the ‘Göttingen
Seven’, a group of university professors, including the Grimm
brothers and the historians Dahlmann and Gervinius, who had been
relieved of their posts and exiled from the country by the new king,
Ernst August (the first ‘German’ Guelf to sit on the throne of
Hanover after the end of the personal union with England), after
they had publicly declared the continuing validity of their oath of
allegiance to the 1833 state constitution abolished by the king.
Although the Göttingen Seven were far from being revolutionaries,
their resistance to Ernst August’s coup made them into heroes
among all liberals in Germany. And more than that: in the fourth
volume of his Deutsche Geschichte im Neunzehnten Jahrhundert,*
Heinrich von Treitschke wrote in 1889 that the exile of the Seven
‘established the political power of the German professorate’, a
public acclaim that aided many university faculty when, a decade
later, they sought admittance to the Frankfurt parliament as
representatives of the people.6
The system of political repression, such as it was exhaustively
codified yet again in Vienna in the secret ‘Sixty Articles’ during
1834–5, bore the stamp of two men, the Austrian chancellor,
Metternich, and the Prussian foreign minister, Ancillon. With the
exception of Hesse-Darmstadt, the constitutional states of Germany
participated only reluctantly in the strengthening of the system of
surveillance and persecution. Although the two German great
157
powers saw eye to eye when it came to the suppression of liberal
and national forces, nonetheless there was a fundamental
difference between Austria and Prussia. While the policies of the
former were always utterly conservative in aim, those of the latter
could be decidedly innovative. Prussia liberalized its economic life
and endorsed industrialization. And it cannot be doubted that the
great pace of economic and social transformation was partly due to
the absence of a Prussian parliament in which the adversaries of
change might have balanced, braked, or blocked the power of the
enlightened bureaucracy.
The path taken by Prussia thus contrasted not only with that of
Austria, but in other respects also with that of the southern German
constitutional states, especially Baden, the paradigmatic liberal
country. The main concern of the Prussian innovators was to make
their country competitive with the most advanced industrial nation
of the era, Britain. The guiding vision of the Baden liberals was a
society of small urban and rural property owners, like the post-
feudal, yet still pre-industrial society that emerged in France after
the Revolution of 1789. The Baden approach prioritized political
participation over societal modernization. The Prussian approach
did the exact opposite. The Baden liberals paid attention to public
opinion and were especially solicitous of the interests of the broad
bourgeois and peasant strata whose trustees they considered
themselves to be. This was the reason they exercised such restraint
when it came to demands for freedom of trade and the free
movement of labour. This was also the reason—or, more
accurately, among the reasons—many of them, Rotteck for
example, opposed a swift legal emancipation of the Jews. The
158
reform-minded bureaucrats in Prussia had no such concerns and
reservations. For them, societal transformation was paramount. For
the south-west German liberals, the most important concern was
individual rights—or, at any rate, the rights important to their
constituents. The advantages and disadvantages of each path were
equally plain to see.7
The most ambitious part of Prussia’s modernization programme of
the 1820s and 1830s was the merging of northern and southern
Germany into a single customs union. This goal was all but realized
on 1 January 1834. The newly founded German Customs Union
(Deutscher Zollverein) included eighteen states with a territory of
425,000 square kilometers and over 23 million inhabitants, 15
million of them Prussian. Most of Germany outside of Austria
belonged to this Prussian-dominated organization, including
Bavaria, Württemberg, Saxony, Hesse-Darmstadt, and Thuringia.
More states joined during the first eight years (the term of the
Union’s original contract), among them Baden, Hesse-Nassau, the
Free City of Frankfurt, and Brunswick. Hanover followed in 1854,
the two Mecklenburgs in 1868. The Hanseatic cities of Hamburg
and Bremen only joined after the foundation of the new Reich, in
1888.
It was more or less self-evident that the Habsburg monarchy
would not participate in the German Customs Union. Economic
development was on widely disparate levels in the various Austrian
domains, precluding for the Vienna government even the moderate
policy of free trade promoted by Prussia and the Union
membership. In establishing the German Customs Union, Prussia
secured its position as the leading economic power in non-Austrian
159
Germany once and for all. This in no way implied that the German
question was now resolved in the direction of a klein-deutsch
national state under the aegis of Prussia. Yet the economic interests
of the industrial bourgeoisie (especially in the Rhineland and
Westphalia) were now linked with Prussia more closely than ever.
Prussia stood for the development and expansion of the national
market and thus for improved commercial potential. Liberals in
Prussia’s Rhenish provinces had a strong material interest in a
progressive unification of Germany, an interest much less
frequently encountered among the liberals of the largely still pre-
industrial German south.8
PRE-MARCH GERMANY
160
Grimm, to the Academy of Sciences in Berlin. Despite these positive
signs, however, the younger Frederick William demonstrated
within a few months of the coronation ceremonies and obeisances
in Königsberg that his understanding of the concept of liberty had
nothing to do with a constitution, but rather was grounded in the
old order of the Stände. He had no intention of giving in to
demands—especially loud in East Prussia—for a representative
constitution. ‘I feel I am king solely by the grace of God, and with
His help I will continue to feel the same until the end,’ he wrote on
26 December 1840 to the liberal Theodor von Schön, chief
administrator of East and West Prussia. ‘I leave without envy all
the glory and trickery to the so-called “constitutional” princes, who
have become a pure fiction for the people, an abstract concept, by
dint of a piece of paper. A paternal governance is the way of true
German princes.’9
The Rhine Crisis, the other great event of 1840, had its origins
in the Near East. Through its support for Muhammad Ali, the
belligerent viceroy of Egypt fighting against Turkish sovereignty,
France had made enemies out of England and Russia. The latter
found allies in Austria and Prussia. The London Treaty of July
1840, in which the four powers committed themselves to
preserving the Ottoman empire, was deeply humiliating to France,
which demanded immediate retribution. As compensation,
Germany was called upon to surrender its territories on the left
bank of the Rhine—a goal that could only be achieved, if at all,
through war.
The French demand for Rhineland territory was greeted in
Germany by a national outcry.
161
Sie sollen ihn nicht haben,
Den freien deutschen Rhein,
Ob sie wie gier’ge Raben
Sich heiser danach schrei’n.*
162
nationalist sentiment. Never before had pan-German nationalism
taken possession of the masses to such an extent as during that
year. The anti-French animus of the wars of liberation,
considerably faded since 1815 (though it had never entirely
disappeared), was suddenly alive again, this time for the long term
and without the north–south difference of the Napoleonic era. The
Rhine Crisis gave rise to a sense of external threat through France
in all political camps, among the lower classes no less than their
rulers. This feeling outlasted the event itself, causing many liberals
to reconsider seriously whether unity might not, after all, be more
important than liberty. The leader of the liberals in Hesse-
Darmstadt, the jurist Heinrich von Gagern, son of a free imperial
baron, had come to this conclusion in 1832. At the beginning of
1843, he observed that many liberals in Baden, after learning the
hard lessons of political defeat, had begun to rethink their
priorities.
163
combined strength of a constitutional ‘third Germany,’ the ‘Trias’,
could, if necessary, advance the goals of liberty and unity even in
the face of opposition from Austria and Prussia. The German–
French confrontation of 1840 taught moderate liberals that the
German question was above all a question of power, one that could
only be resolved in cooperation with the indisputably German great
power, Prussia.
The Saxon Karl Biedermann, publicist and professor of history in
Leipzig, set the new tone in 1842 when he initiated the founding of
a new ‘National Party’ with the goal ‘to establish the unity, power,
and indivisibility of the German nation on a lasting basis’. The
liberal party, too, must realize that, ‘in order for a free nation to be
built, there must first be a nation, and that this nation will never
come into existence if all we ever do is fight over the kind of
constitution it must have.’ In 1844 the Brunswick liberal Karl
Steinacker argued for the expansion of the Prussianled Customs
Union to a German national state. He contested the danger of a
‘universal Prussian monarchy in Germany’: even if Prussia has
ambitious plans, these can ‘only ever be fulfilled if they have the
sympathy and full understanding of the German people’. During the
1840s, Paul Pfizer from Württemberg definitively joined the camp
of those who pushed for the unification of non-Austrian Germany
under Prussian leadership. By 1842 he believed, like Jahn long
before, that Austria, to whom he recommended a close ‘federal and
internationally recognized’ relationship with Germany, should
extend its influence in the Danube region, a task in which
immigrants from Germany could assist. In the rest of Germany,
Pfizer anticipated and endorsed the increasing importance of the
164
national principle ‘the closer we approach the real ascendancy of
the law of reason. The national state is the normal state in the
sense of law [der rechtliche Normalstaat], the reasonable state.’
Like his compatriot Pfizer, the economist Friedrich List, who
came from Reutlingen, advocated German and Austrian dominance
of central Europe, in which the countries of the lower Danube and
the Black Sea region, as well as the whole of Turkey, were defined
as German hinterland. Yet even List could only conceive the
national unification of Germany in terms of the extension of the
Customs Union of 1834. In 1845, one year before he took his own
life in Kufstein, he expressed his belief in the German nation in
classic fashion, suggestively recasting a passage of the Bible:
165
of Alsace and Lorraine was raised again, this time more loudly than
during the wars of liberation. In the course of industrialization, the
educated and wealthy bourgeoisie increasingly favoured a powerful
Germany capable of dominating the Continent. Opinions were
divided on the issue of protectionism, that is, whether a future
German state should shield itself from the competition of
industrially more powerful Britain by means of protective tariffs (as
List and most of the Württemberg and many of the Rhenish-
Westphalian industrialists thought), or whether German interests
would be better served by the Prussian practice of moderate free
trade. Moderate liberals were united, however, in their belief that
German particularism had no future in an age that saw the
development of steamship travel and an increasingly dense rail
network stretching across the country. From the liberal point of
view, the dictates of progress called for national unification. Yet
without the active participation of Prussia, the economically most
powerful state, the project would inevitably fail. What is more, in
order to fulfil its German mission, Prussia had first to enter the
ranks of the constitutional states, something it showed no
indication of doing under Frederick William IV, to the increasing
chagrin and disquiet of the impatient liberals.11
The Prussian king’s frequent assertions of national and Christian
sentiment failed to compensate for his neglect in matters of
constitutional policy. In September 1842, he arranged a celebration
to mark the commencement of work to complete the great
medieval cathedral of Cologne as a declaration of German
greatness and a symbolic reconciliation of the confessions under
the auspices of a common cultural heritage. Cologne was an
166
opportune location for a demonstration of this kind. In the last
decade of Frederick William III’s reign, the city had witnessed a
fierce battle between the Prussian state and the Catholic Church
over the law concerning the upbringing of children in
confessionally mixed marriages, a battle culminating in the arrest
and incarceration of Archbishop von Droste-Vischering. When
Frederick William IV, in order to settle the conflict, adopted a very
accommodating policy vis-À-vis the church, liberal suspicions were
necessarily aroused. The notion of the ‘Christian state’, in which
the denominations would be reconciled (as promoted, with the
applause of the king and his conservative camarilla, by the
constitutional scholar and Lutheran convert Friedrich Julius Stahl,
a son of Jewish parents), came straight from the imagination of the
political romantics. It was absolutely incompatible with
enlightened ideas concerning the relationship between state and
society.
The view of Germany cultivated by the Prussian king and his
advisers also diverged sharply from that of the liberals. For
Frederick William IV, Germany was first and foremost the
community of its princes, led by the Austrian emperor. The
romantics saw things similarly. The champions of the liberal
unification movement of the 1840s strongly disagreed. They
wanted to create a modern nation-state of civic character, which
meant curtailing the power of the princes and particular states.
They did concur with the romantics on the importance of the
German language; liberals, too, believed that language was the best
way to determine whether a territory and its inhabitants belonged
to Germany and the Germans. ‘A Volk is the very embodiment of a
167
people speaking the same language,’ Jacob Grimm declared at the
first congress of Germanists in Frankfurt am Main in 1846. This
was a formula upon which both liberals and conservatives could
agree. Moreover, periods of threat from abroad, like the Rhine
Crisis of 1840, were conducive to at least one shared political
conviction—that the danger must be warded off. But when the
external threat disappeared, the differences grew again and the
visions of Germany remained incompatible.12
The conflict between liberals and conservatives also emerged
inside the oppositional movement of the Vormärz itself. During the
1840s, the antinomy between moderate liberals and radical
democrats grew sharper, and a socialist movement composed of
intellectuals, journeymen in the craft trades, and workers began to
form to the left of the bourgeois democrats. Liberals in the
narrower sense had no desire to overturn the existing order
through revolution; they sought to alter it through reform, working
together with the princes and governments as much as possible.
Moderate liberals had no intention of mobilizing the masses; they
wanted to keep the representation of the Volk in the hands of the
educated and propertied bourgeoisie. Thus they espoused a socially
gradated voting system, oriented on tax rates determined through
the census. The democrats advocated popular sovereignty and—
more or less openly—a German republic; they wanted universal
suffrage for men, general elections, and the right of every citizen in
good standing with the law to exercise the office of juror.
One of the sources nourishing the democratic movement of the
late Vormärz period was religious in nature. After 1815, under the
reign of Frederick William III, the Lutheran and Reformed
168
Churches had merged together into the Old Prussian Union
(Altpreußische Union). The resulting ossification of Protestantism
into a new official state church aroused dissent, on the ‘right’ from
old-style Lutherans and a new Pietist revival movement (with some
of its most important centres among the aristocratic estates in
Pomerania), on the ‘left’ from the Illuminati or Friends of the Light
(Lichtfreunde), a movement that started in 1841 in Saxony,
thereafter spreading to all of Prussia. As disciples of ‘Theological
Rationalism’, the Illuminati (as well as the Free Congregations
emerging from their ranks after 1846) opposed not only church
orthodoxy, but also the doctrine of the Christian state promoted by
Frederick William IV. This brought them into contact with the
dissident intellectual circle around the Young Hegelian Arnold
Ruge and his Hallische Jahrbücher (founded 1838), a journal
dedicated primarily to the critique of religion and the church.
The Catholic counterpart to the Protestant Illuminati was
German Catholicism (Deutschkatholizismus), a broad movement
based primarily in the petty bourgeoisie. In the beginning, the
strength of the movement lay entirely in its protest against the
exposition of the ‘Holy Coat’ in Trier in 1844. This mass
pilgrimage, condemned by the critics as a regression into the
medieval cult of relics, was organized by the church in conscious
protest against liberalism, democracy, and socialism. The success of
the ecclesiastical offensive was considerable: some 500,000 people
took part in the pilgrimage to Trier. But in many places (Silesia and
predominantly Evangelical Saxony, for example), the
deutschkatholisch opposition took on the same character of a partly
religious, partly political mass movement seen in the Illuminati.
169
The Deutschkatholiken considered themselves the avant-garde of a
unified German national church independent of Rome, and like the
Illuminati, whose membership was also mainly petty bourgeois,
they inclined politically toward the left wing of liberalism, toward
the democrats.13
170
that in 1840, immediately after the accession of Frederick William
IV to the throne. In consequence, he lost his position as
Oberpräsident. The Jewish doctor Johann Jacoby, also from
Königsberg, reiterated the same demands in 1841 in his
anonymously published polemic Vier Fragen, beantwortet von einem
Ostpreussen,† which the Federal Diet immediately banned at
Prussia’s request. But the censorship, as well as Jacoby’s acquittal
of the charges of attempted high treason and lese-majesty, only
increased the effect of the pamphlet. The call for the redemption of
the royal promise for a constitution, made in 1815, could no longer
be silenced.
The western part of the monarchy, the Rhine province,
possessed something that barely existed in the rest of Germany: a
modern economic bourgeoisie, the natural sponsor of a liberalism
that differed markedly from the liberalism of both East Prussia and
south-western Germany. Hermann von Beckerath, Ludolf
Camphausen, David Hansemann, Friedrich Harkort, and Gustav
Mevissen—the leading Rhenish liberals were bankers, merchants,
or industrialists. They were also members of the Rhenish or
Westphalian provincial parliament at various times. If nothing else,
their economic interests alone dictated strong support for a
Prussian parliament, a Prussian constitution, and Prussian initiative
in the unification of Germany. For these men, the liberalization of
the industrially most developed German state and the unification of
Germany were the two sides of the same coin of progress. Prussia
had to become a constitutional state so that its king could take his
place at the head of a united Germany; Germany had to become a
national state because, as such, it could promote the interests of the
171
German economy worldwide far more effectively that any
particular state, including Prussia.
Rhenish liberalism was more ‘bourgeois’ or bürgerlich than the
East Prussian variety. At the same time, it was less
bildungsbürgerlich, less oriented towards culture and education, and
thus less ‘idealistic’ than the liberalism of south-west Germany. It
shared with the somewhat aristocratic East Prussian liberalism a
proximity to the government: according to its leading
representatives, neither the parliamentarization of Prussian nor the
unification of Germany could be effected against the will of the
king. This attitude still remained after Frederick William IV
convened a representative assembly of the eight provincial
parliaments as United Committees (Vereinigte Ausschüsse) in Berlin
in October 1842, but then accorded this body only an advisory
function, far less than what liberals of all stripes had demanded.
Prussian liberals further to the left saw things differently. To the
extent that they had placed any hopes in the new king, they soon
learned their lesson. Up to 1838 Arnold Ruge, the publisher of the
Young Hegelian Hallische Jahrbücher, was still declaring that the
‘Protestant principle’ or ‘principle of Reformation’ was the
‘principle of Prussia’, one that made any revolution superfluous, if
not impossible. By 1841, however, he had found it advisable to flee
from the Prussian censors into neighbouring Saxony, where he
continued to publish his journal under the title Deutsche Jahrbücher.
At the beginning of 1843, Ruge published his essay ‘Selbstkritik des
Liberalismus’,* in which he proposed that liberalism disband and
become part of the democratic movement. The Saxon government
promptly banned the Jahrbücher, and Ruge left Germany for Paris,
172
the refuge of so many German democrats and radicals.14
Another even more radical Young Hegelian took the same step
the same year for similar reasons: Karl Marx from Trier, the son of
a lawyer who had converted from Judaism to Protestantism. Marx
was a doctor of philosophy and part-time editor (1842–3) of the
Rheinische Zeitung, a radical-liberal newspaper founded by Gustav
Mevissen and banned by the Prussian government in spring 1843.
In November of that year, Marx moved to Paris, where he
published the Deutsch-Französische Jahrbücher together with Ruge
before falling out with him shortly thereafter. In the only double
edition of this journal ever to appear (February 1844), Marx made
an open declaration of his final break with the philosophy of Hegel.
But the essay ‘Kritik der Hegelschen Rechtsphilosophie. Einleitung’
represents far more than a settling of accounts with the
consummate representative of German Idealism. In this text, Marx
announced the most radical of all revolutions: the proletarian
revolution, which could only be a German revolution.
The basis of this bold prophecy was a thesis concerning the
‘anachronistic nature’ of political, economic, and social conditions
in Germany, which Marx compared to the situation in France on
the eve of the 1789 revolution: ‘Negating the situation in Germany
in 1843 barely puts me in the year 1789 according to the French
calendar, much less at the focal point of the present age.’ For Marx,
Germany’s backwardness was so radical that only a radical
revolution could avail against it, a revolution so fundamental that
the emancipation of the Germans would be tantamount to the
emancipation of humankind in general. ‘The head of this revolution
is philosophy, its heart the proletariat.’ The role of philosophy was
173
predetermined by German history. ‘For Germany’s revolutionary
past is theoretical; it is the Reformation. Back then it was the monk,
in whose mind the revolution began—now it is the philosopher.’
This philosopher seeking simultaneously to replace and transcend
Luther had a name: Karl Marx.
The ideas Marx put into writing in Paris in 1843–4 suggest the
medieval doctrine of the translatio imperii transformed into
contemporary terms, a translatio revolutionis, so to speak. Just as in
the Christian interpretation the Roman empire was transferred in
800 from the east to the west, from the Greeks to the Franks or
Germans, so now the Revolution migrated from the west to the
east, from the French to the Germans. To be sure, its character was
changed in the process. What the French achieved in 1789 was the
classic bourgeois revolution; but society had moved forward in the
meantime, so that when the ‘Gallic rooster’ crowed again, he would
herald a different revolution—the revolution of the proletariat.
Once again, France would be the starting point. The decisive battle,
however, could be fought in only one country, Germany. Since
German society was so retrograde, any bourgeois revolution there
could ‘only be the direct prelude to a proletarian revolution’.
Although this insight first appears explicitly in the Communist
Manifesto of 1847–8, Marx and his friend Friedrich Engels simply
repeat here the main thrust of the ‘Einleitung zur Kritik der
Hegelschen Rechtsphilosophie’.
This correlation between the ‘proletarian’ and ‘bourgeois’
revolutions, the cornerstone of Marx’s theory of revolution, was
audacious. What had happened in France in 1789 was that an
obsolete ruling class, the feudal nobility, was stripped of power by
174
a rising class, the bourgeoisie, which was at least partially justified
in its claim to represent the entirety of non-privileged society and
to be qualified in every respect for the exercise of political power.
It was questionable whether the proletariat would ever find itself in
a parallel situation with respect to the bourgeoisie. For Marx,
however, such a development was historically inevitable. No less
bold was the inferential link he posited between the German
Reformation and the coming revolution of the German proletariat.
No German author since Fichte had identified so completely with
Luther. In demanding for himself the leadership role, Marx was
acting, as he saw it, in the name of, if not under actual commission
from, the revolutionary intelligentsia. Like the early German
nationalists, the early German socialists, those possessed of the
‘correct consciousness’ in Marx’s sense, wanted to be regarded in
society as the avant-garde. The most daring corollary of all,
however, was the third: the mental leap from Germany to the rest
of the world. In this, Marx took Fichte’s thinking to its logical
conclusion. The Germans as the people who would save humanity
by means of their revolution—one had to feel very intimate with
the World Spirit in order to assume the world would welcome such
a fate.15
The industrial proletariat in whom Marx placed such hopes was,
in Vormärz Germany, still too weak to be a revolutionary subject.
The social misery of the early factory workers was more conducive
to passive acceptance than to active resistance. Of the minority of
workers and journeymen who found the will to protest, many had
come into contact with early socialist ideas while they were
abroad, especially in France, Britain, and Switzerland. Many of
175
them were drawn to the utopian-religious communism of the tailor-
journeyman Wilhelm Weitling. Somewhat later, in 1848, the
Workers’ Brotherhood (Arbeiterverbrüderung) of the printer-
journeyman Stephan Born (one of the first to use the term ‘social
democracy’) attracted many others. In comparison, the Communist
League founded in 1847 in London with active assistance from
Marx and Engels drew very few numbers, and was destined to play
no very considerable role in the Revolution of 1848–9.
The ‘social question’ of the Vormärz concerned not only the
early industrial proletariat, but also the rural lower classes who
could find employment no longer in agriculture but not yet in the
factories. The revolt of the Silesian weavers in 1844 made the
decline of the ‘proto-industrial’ cottage industries unmistakably
clear. In the years 1845–7, an ‘old-style’ crisis in agriculture
combined with a ‘new-style’ industrial crisis. A potato famine and
large-scale crop failure provoked a general elevation of prices for
essential foodstuffs in 1845–6, leading to bloody riots by starving
mobs throughout most of Europe. At the same time a cyclical crisis
of overproduction, beginning in England, shook large parts of the
Continent, with the result that large numbers of banks collapsed
and many factories were forced to close. For the workers, the years
1846–7 brought extremely low wages, high prices, and altogether
one of the worst periods of early industrial pauperism. In the
meantime, industrial development had entered a new phase. In
1845 the ‘Industrial Revolution’ began, the ‘Great Leap Forward,’
the industrial ‘take-off’. And there was no doubt that the single
biggest factor contributing to the acceleration of the pace of
industrialization was construction of rail networks. The locomotive
176
of economic development was—the locomotive.16
The railroad was also one of the reasons King Frederick William
IV summoned the United Diet (Vereinigter Landtag) to Berlin in
February 1847. This assembly consisted of a curia of lords along
with three other curiae, one each from among the manorial
landowners, other rural landowners, and specially qualified urban
property owners, each body formed by delegates from the
respective curiae in the provincial parliaments. The United Diet
was supposed to have the right to approve new taxes and
government loans, but not the right of ‘periodicity’. That is, it was
not permitted to convene of its own accord, but only at the behest
of the king. In the speech that opened the Diet on 11 April 1847,
Frederick William wasted no time in making it abundantly clear
that he categorically rejected a Prussian constitution. Never would
he allow, he announced, ‘a piece of paper with writing on it to
interpose itself like a second providence between our Lord God in
Heaven and this country’.
The United Diet did not wish to acquiesce in the stunted version
of a parliament proposed by the king. Both sides remained
stubborn, with the result that the intention for which Frederick had
summoned the pseudo-parliament in the first place came to
nothing. The United Diet refused to underwrite state loans of 25
million thalers for the construction of the planned eastern railline
to Königsberg. Despite this project’s great economic attraction for
both western industry and East Prussian agriculture, the political
interest of both parties in a Prussian constitution and parliament
was greater still. This rejection of the government’s agenda marks
the end of Prussia’s renewal from ‘above’, led by the enlightened
177
bureaucracy. If the Hohenzollern state wished to continue the
process of modernization, it would no longer be able to avoid
granting Prussian society a considerable portion of political co-
determination.
Although the moderate liberals did not want to capitulate to the
king, at no time did they contemplate resolving the power struggle
through revolutionary means. Even after the experience with the
United Diet, the anti-revolutionary credo of liberalism held true,
such as it was formulated by the Marburg historian Heinrich von
Sybel (originally from Düsseldorf) in his essay Die politischen
Parteien der Rheinprovinz in ihrem Verhältnis zur preussischen
Verfassung* written at the beginning of 1847: ‘The revolution
awakens unbridled lust for power on all sides, a desire that can be
called the grave of the constitution as well as of all true liberty.’ If
the socialist and communist tendencies especially prevalent among
the youth and the working classes continued to spread as they had
during the previous ten years, they would soon terminate any
influence the government or the bourgeoisie might still have over
the fourth estate, that is, the proletariat.
For this there is only one remedy: tie the bourgeoisie firmly to the
power of the state by granting it political rights. This is the way,
and it is the only way, to restore the natural opposition it [i.e. the
bourgeoisie (H. A. W.)] feels, down to its least member, for the
aforementioned tendencies, the only way to create the spiritual
energy necessary to anchor public opinion in a salutary
contemplation of societal conditions.17
178
The other German great power, Austria, was also without a
constitution, but otherwise had little in common with Prussia.
Although the Vormärz witnessed a few reforms in the Habsburg
monarchy (such as in the judicial system), as a whole the Austrian
bureaucracy was far less effective, press censorship more complete,
and the surveillance system more extensive than their counterparts
in Prussia. In addition, the industrialization of the empire had
hardly even begun. In the 1840s Metternich rapidly distanced
himself from the goal of putting the government’s finances in
order. This led to growing debts and increasing dependence on the
banks, above all the Jewish Rothschilds. However well-disposed
the liberal Vienna bourgeoisie was towards demands for a liberal
constitution including the whole of the state, the actual addressee
of such demands, the crown and the government, were not.
German-speaking Austrians of all political camps continued to
consider themselves culturally part of the German nation. At the
same time, however, they looked upon themselves as the natural
rulers of the multi-ethnic Habsburg empire.
The non-German nations thought differently, of course. Yet
wherever resistance formed against the foreign rulers, as in the
Polish rebellion in Galicia in 1846, Vienna struck ruthlessly.
Cracow lost the independence it had gained in 1815 by treaty
agreement and, with the assent of Russia and Prussia, the other
powers involved in the partition, was annexed to Austria over the
protests of Britain, France, and all European liberals. In Hungary,
which enjoyed wide independence and whose king was the
emperor of Austria, Magyar nationalism turned against the
majority of the non-Magyar nationalities, including Croats,
179
Romanians, Slovaks, and even Germans. In Bohemia, a bourgeois
Czech nationalism began to take shape, and in much-partitioned
Italy, where (beginning in the 1820s and continuing after the 1830
July Revolution in Paris) regional revolutions had given vent to the
widespread dissatisfaction with the status quo, Mazzini’s secret
society Giovane Italia competed with other, more moderate groups
over the leadership role in preparing for the battle of national
independence.
Metternich responded to the national aspirations of the non-
German and non-Magyar nationalities in the same way as he
answered the demands of the liberals— with repressive measures.
One of the unintended consequences of the ‘Metternich system’ was
the psychological alienation of Austria from the rest of Germany, a
result of the continuing effort to keep all oppositional currents and
influences originating in the other members of the German
Confederation out of the Habsburg empire. The psychological
estrangement was accompanied by a lack of interest in the common
transportation network. Rail connections between Austria and the
rest of the federal territory were not high on the Vienna
government’s list of priorities. Only in 1849, after years of
negotiations, was the first line opened linking Austria with Prussia.
By the end of the Vormärz, it took a great deal of imagination to
believe that Austria, if faced with the choice between the
preservation of its empire and the unification of Germany, would
opt for the latter.18
In the southern German constitutional states, the separation of
the democrats from the moderate liberals continued to assume
more distinct organizational form in the period preceding the
180
Revolution of 1848. The Baden democrats initiated the process. On
12 September 1847, called together by the Mannheim lawyers
Gustav von Struve and Friedrich Hecker, the ‘Wholes’ (die Ganzen)
who sought to differentiate themselves from the moderate liberal
‘Halves’ (die Halben) met in Offenburg. The Offenburg platform
included the classical liberal basic rights with the freedom of the
press at the top of list, the election of a German parliament on the
basis of equal suffrage, military service in the form of a people’s
militia, a progressive income tax, and ‘redress of the imbalance
between capital and labor’. There was no mention of a republic,
but this was the result of tactical caution rather than a renunciation
of the goal.
The response of the ‘Halves’ followed on 10 October 1847,
ironically in the ‘Hotel of the Half-Moon’ in Heppenheim. Here,
representatives of the moderate parliamentary ‘chamber’
opposition met from Baden, Württemberg, Hesse-Darmstadt, and
the Electorate of Hesse. Several of the most famous participants
were Heinrich von Gagern from Hesse-Darmstadt, Friedrich Römer
from Württemberg, Carl Theodor Welcker, Friedrich Bassermann,
and Karl Mathy from Baden, as well as David Hansemann as
honorary guest from the Prussian Rhenish province. It was the
same circle whose common voice had been expressed in the
Deutsche Zeitung, published in Heidelberg, from 1846 onward.
Instead of a platform, the Heppenheim circle drafted a protocol. It
advocated the expansion of the German Customs Union, which was
to be supplemented with a representative body. Though it could
not participate in its entirety, Austria might join the Customs Union
through its territories belonging to the German Confederation and
181
thus contribute to German national unity. The protocol further
included demands for civil liberties and legal reforms very much
resembling those of the Offenburg circle, as well as a declaration of
support for a ‘just distribution of the public burdens to relieve the
smaller Mittelstand and workers’.*
The governments of the German Confederation were alarmed by
the announcements of both camps. The Baden government was the
least prone to agitation. In Austria and Prussia, on the other hand,
fear of a new wave of revolutions was great, and it grew still
greater when, at the end of November 1847, the ‘radicals’
controlling the Swiss government managed to decide the issue of
the war against the separatist union of seven Catholic cantons
(known as the Sonderbund) in their favour. Signs of approaching
turmoil increased during the first two months of 1848. At a large
public gathering in Stuttgart, demands for universal suffrage, the
general arming of the populace, and a German customs parliament
were announced. Anti-government demonstrations took place
around the same time in Munich.
A wave of unrest also shook the extreme north of Germany at
the beginning of the new year. On 20 January 1848 King Christian
VIII of Denmark died. In July 1846 he had announced in an ‘open
letter’, in which he appealed to the Danish law of succession, the
future incorporation of Schleswig into the Kingdom of Denmark
(i.e. the separation of Schleswig from Holstein). This proclamation
provoked a national protest movement in Germany, joined by
democrats and liberals with equal fervour. Christian’s successor
Frederick VII immediately convened a commission to consult over
the already existing draft of a constitution for the new integrated
182
Denmark. The nationalist party of the ‘Eider-Danes’ promptly
demanded that the government make the annexation official and
extend the national borders southward to the Eider river. The
breach of old historical law was obvious. The Treaty of Ripen of
1460 had declared that the duchies should remain united and
undivided in perpetuity. Energetic opposition immediately arose in
both territories, but was unable to prevent the new government,
controlled by the ‘Eider-Danes’, from proclaiming the annexation
on 21 March in Copenhagen. War was in the air, and at a time
when Germany had just crossed the threshold to open revolution.
Two months previously, on 23 January 1848, Friedrich Engels,
writing in the Deutsche Brüsseler Zeitung, had once again taken stock
of the ‘movements of 1847’ and arrived at a highly favourable
judgement concerning the situation of the European proletariat.
183
THE REVOLUTION OF 1848
‘When all the internal conditions have been fulfilled, the German
day of resurrection shall be announced by the crowing of the Gallic
rooster.’ This had been Marx’s prediction in February 1844 in the
Deutsch-Französische Jahrbücher. Four years later the time had
come. In Paris, the republican opposition to the July Monarchy
called for demonstrations for universal suffrage. The first
barricades were built on 22 February 1848. The next day, the
‘National Guard’ went over to the side of demonstrators. On 24
February, King Louis Philippe abdicated in favour of his grandson
and went into exile in England. But the masses were not satisfied
with a simple change of regent. They desired, and achieved, the
proclamation of the republic. On 25 February, the provisional
government passed a law guaranteeing workers the right to work.
Universal male suffrage followed on 4 March.20
The spark of revolution leaped quickly across the Rhine. On 27
February a public gathering in Mannheim, attended by the leading
Baden liberals and democrats, petitioned the Karlsruhe government
for freedom of the press, trial by jury, introduction of constitutions
in all German states, and a German parliament. The following day,
the delegate Heinrich von Gagern placed before the Hesse
parliament an official request for the convocation of a national
assembly and the ‘renewal of the federal presidency’. In this, he
was repeating demands made two weeks before (12 February) by
the Baden liberal Friedrich Bassermann in a speech to the Baden
parliament. The liberal agenda was thus staked out: political liberty
and national unity formed the core of the ‘March demands’, soon
184
made in all of Germany.
Moderate liberals had not turned into revolutionaries overnight.
What happened in February and March 1848 was that leading
liberals simply placed themselves at the forefront of a movement
that might easily have fallen into the hands of more radical forces
if left without the moderating influence of the educated and
propertied bourgeoisie. The threat was already evident by 1 March
in Karlsruhe, where a large band of armed men forced its way into
the plenary hall of the parliamentary chamber during the official
submission of the Mannheim petition. Only a few days later, a
peasant uprising began in large parts of south-west Germany, from
the Bodensee over the Kraichgau to the Odenwald. The revolt was
directed not only against the nobles and their officials, but also
against the Jews, in their capacity as creditors. Jewish houses were
destroyed in numerous locations, their inhabitants driven out of the
community. The catalyst for the pogrom was the decision of the
second chamber of the Baden parliament to grant the Jews political
and legal equality, which many communities opposed. Equal civil
status for the Jewish minority meant that the communal
governments would be responsible for the social welfare of a large
number of destitute Jews, something they felt they could not
afford. Thus, at the beginning of 1848, traditional Jew-hatred
combined with the rejection of feudal and fiscal burdens in an
explosive combination. The rural protest was sparked by the
revolution and expressed itself in a revolutionary manner. But its
goals were the opposite. It took aim at everything emancipatory,
liberal, and modern the revolution had to offer.
March 1848 was the month of new cabinets in the
185
administrations of most German states. The ‘March governments’
saw the entrance of many moderate liberals into office: Heinrich
von Gagern in Hesse, Friedrich Römer and Paul Pfizer in
Württemberg, Carl Stüve in Hanover. Only in Bavaria was a
monarch forced to renounce the throne. In consequence of his
affair with the dancer Lola Montez, King Ludwig I had so damaged
his reputation that he had no choice but to abdicate in favour of his
son, Maximilian II, after a violent uproar on 20 March. Only with
the two German great powers, however, would the fate of the
German revolution be decided. Both Prussia and Austria saw new
governments take power in March 1848. In Vienna, revolutionary
students together with bourgeois and workers created an agitation
that was finally too much for Chancellor Metternich. On 13 March
the man who embodied the system of censorship, surveillance, and
repression, like no other, stepped down. Not that the members of
the new government could be called ‘liberal’; almost all of them
had already occupied important positions under the former
chancellor.
The only possibly ‘liberal’ aspect of the new government was the
contents of the Austrian imperial constitution of 25 April 1848,
which was modelled after the Belgian constitution of 1831. Yet
because Emperor Ferdinand imposed it without any consultation
with representatives of the bourgeoisie, it suffered from a
considerable deficit in legitimacy. In mid-May the student and
proletarian left reacted against the imperial dictate with armed
violence, causing an ostensible reversal on the part of the court: the
April constitution was now to be revised by a newly formed
Imperial Parliament (Reichstag) elected by universal suffrage. In
186
reality, the ‘court party’ had by this time already resolved to defy
the revolution. The imperial family betook itself via Salzburg to
Innsbruck, from where it set about organizing resistance to the
Vienna radicals.
After another wave of revolutionary violence on 26 May, a sort
of dual government assumed power in the capital. The weak
administration of Franz von Pillersdorf was confronted with a
Security Committee controlled by the radicals. Tensions were only
eased somewhat when Archduke Johann, to whom the emperor
had granted full authority on 15 June during the period of his
absence from Vienna, fulfilled the demand of the Security
Committee for a reorganization of the administration. On 22 July
the newly elected Imperial Parliament convened in Vienna,
representing all territories of the Habsburg monarchy except
Hungary and the Lombardo-Venetian kingdom. The German
delegates were in the minority, and the radicals were forced to
accept defeat even in the capital, the centre of their power. The
distribution of parliamentary seats allowed the administration to
decide the most important issue in favour of the executive: the
Imperial Parliament recognized the emperor’s power of absolute
veto over decisions of the legislature.
The revolutionary movement was foiled on other fronts as well
during the summer of 1848. The Czech national movement was
suppressed by the middle of June. The troops of the Kingdom of
Sardinia-Piedmont, supporting the battle for liberation in
Lombardo-Venetia, suffered a series of serious defeats at the end of
July and beginning of August. Supported by Austria, a Croatian
opposition under ‘Banus’ Joseph von Jelačić formed against Lajos
187
Kossuth’s Hungarian revolution.
Because of these developments, Emperor Ferdinand did not have
a difficult time returning to Vienna on 12 August at the request of
the parliament. The court party did not have to fear that a majority
of its imperial subjects would turn on it when it came time to settle
accounts with the radicals.21
The events unfolding in Austria also had an effect in Prussia. In
Berlin the first disturbances occurred on 14 March. News of
Metternich’s resignation, reaching the Prussian capital two days
later, increased the agitation of the general populace. On 18 March
(a Saturday), King Frederick William IV drafted two proclamations,
the first lifting censorship, the second hastening the convocation of
the United Diet and demanding the reorganization of the German
Confederation to include federal representation. This
representation ‘necessarily requires a constitutional structure
[konstitutionelle Verfassung] of all German states’, which could only
mean that Prussia, too, was to become a constitutional state.
The rumour that great changes were imminent brought crowds
of thousands to the royal palace in the early afternoon. If the king
had not posted great numbers of soldiers in the square, the crowd
would probably only have demonstrated their gratitude for his
proclamations. In the event, the ostentatious display of military
power acted as a provocation and the mood of the demonstrators
changed. As the troops were clearing the square, two shots were
fired. The crowd, believing itself betrayed by the king, responded
by erecting barricades in the streets.
Civil war raged in Berlin until the next morning. The king,
188
whose disposition was about as far from that of a ‘soldier-king’ as
could be imagined, was deeply shaken. That night, he wrote an
appeal to his ‘dear Berlin subjects’, in which he promised, in
exchange for the dismantling of the barricades, to withdraw the
troops from the streets and public squares and to station them only
in a few select locations such as the palace and the armoury. And
in fact, the removal of the barricades had hardly commenced when
the troops began to withdraw. They left the city almost entirely,
which was much more than what the king had promised. Victory
appeared to belong to the insurgents, and this impression was
reinforced that afternoon when the king, with bared head, bowed
over the bodies of the more than 200 dead the rebels had
transported into the palace courtyard. The events of March 1848 in
Berlin—with the sovereign publicly humiliated and now protected
by nothing more than the newly formed revolutionary civil guard,
and the heir to the throne, Frederick William’s brother William,
champion of an unyielding anti-revolutionary course, in flight to
England—were for conservative officers and landowners, among
them the young Otto von Bismarck from Schönhausen in the
Altmark, a deep disgrace to old Prussia, one that would have to be
set to rights as quickly and effectively as possible.
Frederick William, on the other hand, sought to save himself by
taking the bull by the horns. Together with the princes of the royal
house, as well as several generals and ministers of the government
newly constituted two days previously (and which was to remain in
power only ten days), the king made a ceremonial tour through
Berlin, during which he and his entourage wore armbands sporting
the black, red, and gold of the German unification movement.
189
Addressing students of the University of Berlin, he not only
proclaimed his support for German unification, but declared it his
personal mission to lead the princes and people of Germany to the
realization of this goal. On the evening of the same day he issued
an ‘Appeal to My People and to the German Nation’, in which he
expressed his wish that the second United Diet, which would
convene on 2 April, be transformed into a temporary congress of
German estates by including delegates from other estate
assemblies. In the decisive final sentences of the proclamation, he
went so far as to pay tribute to the myth that black, red, and gold
originally had been the colours of the old Reich: ‘Today I have
assumed the old German colours, placing myself and my people
under the venerable banner of the German Empire. From now on
Prussia will merge into Germany.’
The king’s message met with a certain measure of sympathy
from well-meaning Prussian liberals, but not with unconditional
agreement. The assimilation of Prussia into Germany was not on
the liberal agenda. For the conservative faction, which began to
form in March 1848, the words and deeds of Frederick William IV
on 21 March represented yet another attempt to curry favour with
the revolution, if not moral high treason against the state. The
resolute left, for its part, divided over the question of Prussian
assimilation into Germany, saw no occasion to trust in the king’s
change of heart. Outside Prussia, the reaction to Frederick
William’s German initiative was uniformly negative. The events of
18 March in Berlin severely damaged the reputation of the
Hohenzollern king in the ‘third Germany’. The other German
governments had no intention of entrusting national unification to
190
this monarch, and none of the leading liberals of the ‘Trias’ in
spring 1848 was able to envision an expanded Prussian parliament
as the German national assembly.
And yet it was clear that the Hohenzollern state and its king
would continue to play a major part in German politics. Old Prussia
was as far from collapse in March 1848 as the Habsburg monarchy.
Their early successes deceived many liberals into thinking they had
already won the decisive battle. In reality, the aristocratic,
bureaucratic, and military foundations of the Prussian state
survived largely intact. While the dissatisfaction with the state of
things was strong enough to express itself in revolutionary form,
only a revolutionary minority sought a radical break with the past.
Moderate liberals still believed Prussia capable of reform, and
considered it intellectually and economically the most progressive
part of Germany. They knew they needed the Prussian state to
protect a unified Germany from external threat. Internally, too,
Prussia might still prove a valuable partner—when the moment
came to prevent the revolution, which no liberal had truly wanted,
from straying into political and social radicalism. In spring 1848
most of the Prussian liberals and their fellow travellers in the ‘third
Germany’ were convinced that they could change Prussia’s
government enough to serve their purposes—thus serving Germany
—within a reasonably short period of time. To dismantle the
Prussian state or destroy its military power was not part of their
programme.
On 31 March 1848, two days after a liberal ‘March ministry’ led
by Ludolf Camphausen as prime minister and David Hansemann as
minister of finance (both from the Rhineland) took office in Berlin,
191
the Preparliament (Vorparlament) convened in Frankfurt am Main.
This assembly of notables, composed of more than 500 liberal and
democratic delegates from all of Germany (though only two
Austrians attended), deliberated the political future of Germany for
four days in the old coronation city of the Holy Roman Empire. The
leftist minority around Hecker and Struve, if it could have had its
way, would have changed the Preparliament into a revolutionary
executive and made Germany into a federal republic like the
United States. But the majority, led by Heinrich von Gagern, was
able to block any move in this direction. Most of the delegates did
not want to continue the revolution, but rather to channel it as
quickly as possible into a process of peaceful evolution on the
foundations of monarchy. The epicentre of power was to be a
German parliament functioning as the representative of the
sovereign Volk and elected by general direct suffrage of all adult
German males. Schleswig, East Prussia, and West Prussia would
become members of the German Confederation and send delegates
to the German parliament. The status of the Grand Duchy of Posen,
inhabited primarily by Poles, remained uncertain for the time
being. Until the German parliament could convene, a ‘Committee
of Fifty’ appointed by the Preparliament was to cooperate with the
Federal Diet, where representatives of the ‘March governments’
were setting the tone at the present. The collaboration functioned
smoothly. The Preparliament and subsequently the Committee of
Fifty placed proposals before the Diet, which generally sanctioned
them.
The radical leaders, Hecker and Struve, were not elected into
the Committee of Fifty, and there were consequences. The extreme
192
left wing of the revolutionary movement came to the conclusion
that the counter-revolution was now in full swing and that the
moderate liberals had betrayed the revolution. They took aim
especially at the liberals dominating the government of the Grand
Duchy of Baden, who, for their part, entertained no illusions about
the radicals’ subversive intentions. On 13 April, starting from
Constance, Hecker and about fifty supporters began a march that
was to culminate in the proclamation of the German republic.
Within very few days, the number of participants grew to over a
thousand. For the Committee of Fifty in Frankfurt, Hecker’s putsch
represented an attack on the elections for the national assembly
planned in May. Moderate democrats failed in their attempts to
mediate. Troops from Baden and the German Confederation
inflicted several severe defeats on the rebels. On 24 April, the
Tuesday after Easter, Freiburg was occupied by government forces.
Three days later, the last battle was fought near Dosenbach on the
Rhine, ending in a debacle for the revolutionaries under the
leadership of the poet Georg Herwegh. Hecker and Struve had
already fled to Switzerland. For the left as a whole, the political
consequences of their enterprise were fatal. The idea of a German
republic fell into discredit for a long time to come. Among the
bourgeoisie the tendency grew to draw a sharp line of division
between themselves and the supporters of radical positions, and to
look upon a rapprochement with the existing powers as the only
viable path.
It was no fluke that the attempted coup from the left took place
in a region where, at the same time, peasants rioted against the
Jews. In the south of Baden, a poor region of small-scale
193
agriculture and domestic trades virtually untouched by
industrialization, economic backwardness provided fertile ground
for many different varieties of political radicalism. Thanks to the
relative liberality of the grand duchy, hard-line democrats enjoyed
a latitude they found nowhere else in Germany. The geographic
proximity of two progressive neighbour states was also an
important factor; without the active support of German emigrants
living in Switzerland and France, the disciples of Hecker and Struve
would hardly have been in a position to arm themselves and inflict
heavy casualties on the regular military forces of their opponents.
The extreme bourgeois left considered resistance to be a natural
right and rarely failed to invoke the laws of reason to justify their
actions. Nothing was more foreign to these ‘men of the people’ than
to think in the categories of historical development and national
individuality, as was natural for the moderate liberals representing
the educated and propertied bourgeoisie. The leaders of the
bourgeois democrats also rejected the tendency of liberals further
to the right to invoke German national interests in order to justify
cooperating with the princes, above all the king. They considered
such a partnership incompatible with the cause of democracy.
On 21 March the Kingdom of Denmark announced the
annexation of Schleswig, which it promptly effectuated by military
occupation. Here was a national interest upon which all Germans,
right and left, could agree. The provisional government of
Schleswig and Holstein, formed three days later in Kiel, demanded
the incorporation of Schleswig into the German Confederation. On
the same day, the duke of Augustenburg (the German heir to the
throne of Schleswig and Holstein), who had gone to Berlin,
194
requested Prussia’s military protection for the ‘eternally indivisible’
(up ewig ungedeelte) duchies. With the approval of the king, the new
Prussian foreign minister, Heinrich von Arnim, agreed, and
Prussian troops crossed the Eider on 10 April. On 12 April the
Federal Diet in Frankfurt resolved to make the Danish evacuation
of Schleswig a matter of Confederation policy. While it avoided
formal incorporation of Schleswig into the German Confederation
(in contrast to the Preparliament, which had taken this step
already, on 31 March), it nonetheless officially recognized the
provisional government in Kiel. On 3 May federal troops, under the
command of the Prussian General von Wrangel, crossed the border
of Denmark and advanced to Jutland. All German patriots were
elated. But Russia and Britain, both signatories of the Final Act of
the Congress of Vienna in 1815, perceived a threat to their interests
in the Baltic and North Sea. They made it clear that they would not
tolerate a German annexation of Schleswig. An international
conflict seemed to be in the making, one that Prussian’s difficult
relationship with Germany could also quickly turn into a major
domestic crisis.
What it eschewed in the case of Schleswig on 12 April, the
Federal Diet had already done the day before with regard to
another territory. In response to a petition from the Preparliament,
it accepted East and West Prussia into the German Confederation,
thereby making Prussia the largest German state. Since the
inhabitants of the annexed province (with the exception of parts of
West Prussia) considered themselves German or (like the
Evangelical Masurians and the Lithuanian-speaking Evangelical
inhabitants of the territory on the Neman* river) at least Prussian,
195
the expansion of the Confederation into these areas caused
virtually no controversy.
This was not true of the decisions taken in the matter of Posen.
On 31 March the Preparliament had spoken of the ‘disgraceful
injustice’ of the Polish partition and acknowledged the ‘sacred duty
of the German people’ to join in the restoration of Poland. But the
assembly refrained from committing itself to recognizing the
borders prior to the first partition of Poland in 1772. The question
of whether Posen should send delegates to the National Assembly
remained unresolved. On 22 April the Diet, at the behest of Prussia,
accepted the predominantly German territories of the Grand Duchy
of Posen into the German Confederation; the city of Posen (Poznań)
and the district of Santer followed on 2 May. In the months
following, the German National Assembly was responsible for two
further eastward shifts of the federal boundary into Polish territory.
The Poles, whose protests availed nothing, saw the expansion for
what it was: yet another partition of their historical territory and
the beginning of the attempt to transform the Polish subjects of the
Prussian king into ‘Germans’ against their will.
The Moravian and Bohemian Czech subjects of the Austrian
emperor were spared such a fate. It was true that the electoral law
passed by the Federal Diet on 7 April applied to the entire
population of the German Confederation, including the Czechs. Yet
when the Committee of Fifty invited the famous Czech historian
František PalackÝ to participate in its deliberations, the latter
refused. PalackÝ declared his allegiance to the supranational
Austrian state, which he considered necessary for the defense of
Europe against a universal Russian monarchy. In the interest of
196
Europe, indeed for the sake of humanity itself, PalackÝ said, the
Austrian empire would have to be invented if it did not already
exist. This ‘Bohemian of Slavic descent’ had no desire to be a
‘German’, and, since his compatriots either felt the same or else
sought an even more radical break with the Germans in the form of
a Czech national state, the May 1848 elections to the German
National Assembly did not take place in the Czech territories of
Bohemia and Moravia. It was no different in the Slovenian districts
of Carinthia, Carniola, and Styria. With the exception of Italian
Tyrol and Trieste, only the German-speaking provinces of Austria
sent delegates to Frankfurt.22
Thus the German National Assembly, which held its constituent
session in the Church of St Paul—the Paulskirche*—in Frankfurt on
18 May 1848, was a German parliament in a much stricter sense
than its founders had envisioned. In terms of social composition, it
was not a ‘parliament of professors’, as one often hears.
Nonetheless, it was more than anything else an assembly of the
educated bourgeoisie. Of the 585 members who took up their
mandates, 550 were in academic professions. As in the
Preparliament, the left, divided against itself, was in the minority,
but it was still considerably stronger than the conservative right.
The dominant tone was that of a moderate liberalism of various
shades.
The delegates considered their primary task to formulate a
constitution, especially the ‘fundamental rights of the German
people’. The ‘Draft of the Seventeen’ (Siebzehnerentwurf), an outline
drawn up by a committee appointed by the Federal Diet, was
197
largely ignored. Considering the delegates’ experience with
governmental repression in the three decades since the Congress of
Vienna, their decision to prioritize the constitution was
understandable. But it was unrealistic. The most important
questions thrown open by the German revolution were questions of
power, the resolution of which could not be postponed with
impunity. First, there was the matter of the relationship between
the future German national state and the supranational Habsburg
empire. Secondly, there was the relationship between the Frankfurt
parliament and the other German great power, the Kingdom of
Prussia, whose own freely elected National Assembly was
inaugurated on 22 May 1848 (four days after the convocation of
the German parliament) by Frederick William IV in a speech
proclaiming the king’s commitment to German unity.
On 28 July 1848 the German National Assembly made what,
four days previously, its president Heinrich von Gagern had called
a ‘bold move’: it passed a resolution setting up a provisional central
authority to replace the Federal Diet. The next day it elected the
Austrian archduke, Johann, to the office of imperial regent, thereby
expressing the wish of the Frankfurt parliament to include Austria
in the founding of the German national state. This act gravely
affronted Prussia and its king, albeit inadvertently. On 15 July
Prince Karl von Leiningen, a noble from the Odenwald and half-
brother of Queen Victoria, was appointed to head the imperial
ministry. But the new central government had no real power and
authority, a fact that became clear when the president of the
imperial ministry officially entered office on 6 August. The two
German great powers and several Mittelstaaten refused to permit
198
their troops to pay homage to the imperial regent, as the Prussian
General von Peucker, the imperial minister of war, had ordered
them to do.
The moment of truth for the provisional central government and
the German National Assembly came at the end of the Schleswig-
Holstein crisis. The parliamentarians themselves had helped to
make the international conflict worse by inviting the delegates
elected in Schleswig into the National Assembly, thereby accepting
—de facto if not de jure—the duchy into the German
Confederation. The protest from London and St Petersburg was
loud. On 26 August Prussia, under pressure from Russia and Britain
and ignoring the stipulations of the imperial ministry, signed an
armistice with Denmark in Malmö. The agreement called for the
evacuation of both Danish and Confederation troops from
Schleswig and Holstein as well as the replacement of the
provisional government in Kiel with a new joint administration
appointed by the Danish and Prussian sovereigns, the latter acting
as a sort of trustee of the German Confederation.
A storm of national protest shook Germany. When the imperial
ministry, which did not have the means to compel Prussia to
continue the hostilities, resigned itself to the inevitable and
expressed its willingness to accept the armistice despite all protest,
the National Assembly broke out in rebellion. The Bonn historian
Friedrich Christoph Dahlmann, member of the right-liberal ‘Casino’
and correspondent for the committees of international affairs and
the central authority, emphasized that not only Schleswig was at
issue, but also the very unity of Germany.
199
To nip ... the new German authority ... in the bud: that is the intent
here. It must be torn apart from all sides and finally smashed! In
this first test that is approaching, let us throw ourselves at the feet
of the foreign powers ... then, gentlemen, you will never again be
able to hold up your proud heads! Consider these my words: never!
200
the National Assembly voted 257 to 236 that the fulfilment of the
armistice should no longer be hindered. The Austrian Anton von
Schmerling, hitherto minister of the interior, became the new
prime minister.
The Frankfurt parliament had only itself to blame for its defeat.
Since it was clear from the start that Prussia would not revoke the
armistice of Malmö, the resolution of 5 September was little more
than an attempt to save face before the German public. Had the
German states, led by Prussia, complied with the demands of the
German National Assembly, a European war would have been the
result. If the National Assembly had not reversed itself, the German
governments would have been forced to break all ties with it.
But the self-correction came at a high price, drawing down upon
the parliament a great wave of popular anger and resentment. On
17 September an assembly of the Frankfurt radicals accused the
delegates who had voted in favour of the armistice of treason
against the German people and declared their mandates invalid. At
Schmerling’s request, the city senate ordered Prussian and Austrian
troops to march from the Confederation garrison in Mainz.
Barricades sprang up and furious street fighting began. Before
nightfall, however, the military had the situation under control.
Among the victims of the Frankfurt uprising were two conservative
delegates from Prussia, Prince Friedrich Lichnowsky and General
Hans von Auerswald, who had voted for the armistice of Malmö.
They were seized and murdered by insurgents during a scouting
patrol in the northern part of Frankfurt.
There was a warning sign in the fact that the leader of the mob
that killed the two delegates carried a red banner. Would the
201
proletariat now attempt in Germany, as in France the previous
June, to turn the political into a social revolution? The workers’
uprising had been brutally suppressed in the French capital. In
Frankfurt, too, the official forces won the day. But Frankfurt was
not Germany. On 21 September Struve left Basel, crossed the
border to Baden, and proclaimed the German republic from the
town of Lörrach. Among the active participants of the second
Baden rebellion was Wilhelm Liebknecht, later to become one of
the leaders of the German social democratic movement. The red
armbands of the insurgents caused a stir among both friends and
foes of the uprising. The same was true of their attitude toward
private property. The family of a wealthy Mühlheim vintner was
forced to pay a poll tax, and a ransom of a sort was extorted from
the Jews of Sulzburg. After four days, the putsch was at an end.
Baden troops won a clear victory at Staufen on 25 September, and
Struve and Liebknecht were soon arrested.
On the same 25 September Cologne, the city in which Marx had
been editing the Neue Rheinische Zeitung since 1 June and where the
Communist League had more supporters than anywhere else in
Germany, also appeared to be on the brink of a second revolution.
At the rumour of the approaching Prussian military, barricades
were erected, upon which red flags were flown during the whole
night. The next day, after soldiers failed to materialize, the
defenders left their posts. Order was restored, a state of emergency
declared, the civil defense dissolved, and the Neue Rheinische
Zeitung temporarily banned.
The cathedral city on the Rhine had made no serious
preparations for becoming the staging ground of the red revolution.
202
The chances were minimal to begin with. Only a small minority of
workers stood behind the Communist League. Much stronger was
the following of Stephan Born and his ‘General German Workers’
Brotherhood,’ which did not preach social revolution, but the
reform of the existing social order. Radical sloganeering enjoyed
even less support among the independent tradesmen, many of
whom had welcomed the revolution in the spring of 1848. In the
middle of July, a General Congress of Tradesmen and Craftsmen in
Frankfurt spoke out against the republic, freedom of trade,
socialism, and communism. Its positive demands included
cooperation between the bourgeoisie and the monarchy as well as
the restoration of the guild system. Although the Congress did not
speak for all the craft trades in Germany, its platform nonetheless
reflected the beliefs and wishes of a large part of the Mittelstand.
This state of affairs was far more satisfying to conservatives than
liberals.
Among the peasantry, too, there was a shift from ‘left’ to ‘right’.
In spring 1848 the peasants of south-west Germany had rebelled
against landowners and government officials. After their most
important demands (above all the abolition of feudal obligations)
had been met, they rejoined the ranks of those who wished to
safeguard the traditional order. Although the social protest of the
peasants and agricultural workers was generally less radical in the
manorial system of the Gutswirtschaft predominant east of the Elbe
than on the estates—Grundherrschaft—of the large western
landowners, even in the east there were changes. The abolition of
patrimonial jurisdiction and other concessions cleared the way for
a reconciliation between the agricultural lower classes and the
203
manorial lords. In Austria, too, timely reforms took the wind out of
the sails of peasant rebellion by September 1848. Throughout the
autumn of that year, with the exception of Silesia and Saxony,
revolution was all but forgotten in rural Germany. To the east of
the Elbe, the main tendency was more conservative than radical.
Of the conservative ruling classes, the Prussian landowners were
politically the most active in the revolutionary year of 1848. By
July, the core of a political party had taken shape from among
their ranks, the Association for the Protection of Property and the
Promotion of the General Welfare of All Classes of the People. At
its first general meeting, the organization gave itself a more honest
name, Association for the Protection of the Interests of the
Landowners, which, however, maintained the claim ‘Promotion of
the General Welfare of All Classes of the People’ in an addendum to
the new title. Much broader and more volkstümlich in character was
the societal catchment area of the more than 400 Catholic Pius
Associations, which could claim about 100,000 members by the
autumn of 1848. The petitions sent by these organizations to the
German National Assembly aimed at the complete emancipation of
the Church from the State as well as the preservation and support
of Catholic schools. Such demands were not strictly ‘conservative’,
but still less were they ‘revolutionary’. The political Catholic
movement, which began to form at this time, was conscious of the
fact that its ecclesiastical-political agenda would sooner find
sympathy among conservative Protestants than among the
emphatically secular liberals. The Evangelical churches remained
what they had been before 1848—pillars of the throne. Their
resolutely anti-revolutionary credo deepened the chasm separating
204
them from many of their politically active flock.
For the moderate liberals, the events of September 1848 were a
bitter lesson. The cooperation of some of their number with the left
during the protest against the armistice of Malmö had caused a
debacle. One demonstration of political impotence led to another
as the original opposition was retracted, further damaging the
reputation of the Frankfurt parliament and provoking the
radicalization of extra-parliamentary elements that led to street
fighting in Frankfurt and the carnage of the second Baden
rebellion. Their radicalization under the symbol of the red flag
isolated the extreme parliamentary left, the ultra-left ‘Donnersberg’
fraction (to which Arnold Ruge belonged) most of all, the more
moderate ‘Deutscher Hof’ of the Leipzig bookseller Robert Blum
somewhat less so. The liberal centre moved noticeably to the right,
a development that had started with the June Days in Paris and
gained ground among broad sections of the German populace,
including even the educated and propertied bourgeoisie, at the
signs of an approaching ‘second revolution’ in September 1848.23
The more the majority of the German National Assembly tried to
marginalize the radical minority, the more tense grew its
relationship to the Prussian National Assembly, where from the
start the left had been more strongly represented than in Frankfurt.
In July 1848 resolute democrats like Johann Jacoby from East
Prussia and Benedikt Waldeck from Westphalia lodged a protest
against the election of an imperial regent not responsible to the
German parliament, in which office they suspected the future
hereditary Habsburg emperor. ‘We desire to place the sword we
have so long wielded in victory for Germany in the lap of the
205
National Assembly. We gladly hand it over to the head of the
centralized German state,’ proclaimed Waldeck on 11 July. ‘But to
an imperial regent who could declare war on his own we will not
entrust the sword of Frederick the Great.’
When at the beginning of August the Prussian government, with
the approval of public opinion, refused to allow Prussian troops to
pay homage to the imperial regent as the imperial minister of war
had ordered, the opposition from the left was weak. Against the
armistice of Malmö, too, the democrats were reserved in their
criticism. But the September unrest in Frankfurt was a different
story. When the Frankfurt central authority, seeking to protect the
members of the National Assembly from insult and injury, took
measures restricting the freedom of association and assembly in
Frankfurt and vicinity, the protest from the staunch left was loud,
and not only in Prussia. The Frankfurt policies were compared to
the repressive system under the German Confederation. On 24
October Waldeck lodged a petition demanding that decrees from
the central authority touching upon matters of domestic policy in
the individual states be approved by the Prussian representative
assembly before entering into effect in that country. The democrat
Jodocus Temme accused the Frankfurt parliament of entering upon
a path that would lead back to the Vienna and Karlsbad Decrees.
To this accusation he added, with the applause of the left: ‘We have
not fought and earned our liberty only to throw it away on a
parliament in Frankfurt am Main.’ The vote to grant Waldeck’s
petition priority in the parliamentary protocol failed by only one
vote.24
The hostility of the Prussian left to the German National
206
Assembly in October 1848 was caused in part by a major political
development: the counter-revolution was gaining ground in
Austria, and the Frankfurt central authority was doing little to
counteract it. It began with the decision of the Vienna government
(supported by the majority of the Imperial Parliament) to assist the
Croatians in their struggle against the revolution in Hungary. The
response of the Austrian radicals was another Vienna rebellion (the
third of that name), in the course of which the minister of war,
Count Latour, was murdered. After the city had fallen into the
hands of the extreme left, the emperor and his court made their
escape to Olmütz in Bohemia. It was the monarch’s second flight
from Vienna within half a year.
From Bohemia, beginning on 11 October, Prince Alfred
Windischgrätz prepared for the coming battle with revolutionary
Vienna, assisted by political counsel from Prince Felix
Schwarzenberg. Emissaries from the German central authority,
attempting to mediate between the revolution and the counter-
revolution, were shown the door both at Windischgrätz’s
headquarters and in Olmütz. At the same time, representatives of
the parliamentary left, including Robert Blum, went to Vienna to
support the insurgents—not on behalf of the German National
Assembly, which rejected their activities, but only in the name of
the minority. On 26 October, imperial troops under Windischgrätz
and Jellačić, the governor of Croatia, commenced their attack on
the capital of the empire. Vienna fell on 31 October. Blum, who
had taken part in the fighting, was condemned to death and put
before the firing squad on 9 November.
By executing the German delegate, a flagrant violation of
207
parliamentary immunity, the Habsburg monarchy challenged the
German National Assembly in a way approaching an open
declaration of war. Once again, the Frankfurt parliament and the
German central authority, possessing no real power of their own,
had no recourse but to protest on paper, which availed nothing.
The new Austrian government, whose leadership Schwarzenberg
assumed on 21 November, wanted to break with Frankfurt, seeking
to return to the pre-revolutionary order both in the Habsburg
empire and in the non-Austrian part of the former German
Confederation. This aspiration was now supported by a large part
of the Slavic nationalities, namely the Czechs, Croatians, and
Slovenians.
The most radical among the revolutionaries reacted with
bitterness and hatred to what they considered the treason of the
Slavs. Only the Poles were excepted from this verdict. Writing in
the Neue Rheinische Zeitung in January 1849, Friedrich Engels wrote
of ‘nationlets’ (Natiönchen), ‘ruined fragments of peoples’, and
‘residual wastes of peoples’ representing the counter-revolution,
and he threatened: ‘The next world war will not only wipe
reactionary classes and dynasties from the face of the earth, but
also entire reactionary peoples. This, too, is progress.’ Against a
‘Slavdom that has betrayed the revolution’, the same author
proclaimed one month later, ‘ruthless warfare and terrorism—not
only in the interest of Germany, but also in the interest of the
revolution’.
The term ‘world war’ was borrowed from Karl Marx. In his new
year’s article for the Neue Rheinische Zeitung, Marx had come to the
conclusion that the revolution would only succeed if it assumed the
208
form of a European war, indeed a world war—a war that must
begin with the overthrow of the French bourgeoisie and then
spread both to capitalist Britain and to Russia, the pre-eminent
representative of eastern barbarism. ‘Revolutionary uprising of the
French working class, world war—this is the table of contents for the
year 1849.’25
By the time these articles appeared, the central European
counter-revolution had succeeded not only in Austria, but also in
Prussia. In Berlin, the key events took place in October. On 12
October the Prussian National Assembly struck the words ‘by the
grace of God’ from the royal title during the deliberations over a
draft for the constitution. Outraged, Frederick William IV swung
round to the course urged upon him by the ultra-conservative
camarilla of the brothers Leopold and Ludwig von Gerlach: he
decided to take up the struggle against the National Assembly and
to reject a liberal constitution of the kind taking shape in the
parliamentary deliberations.
In the middle of October, unrest broke out among the Berlin
workers, and barricades went up in the streets of the Prussian
capital. The civil defense forces soon had the situation in hand, but
when on 31 October the Prussian National Assembly rejected
Waldeck’s petition demanding Prussia’s intervention on behalf of
the revolution in Vienna, the radical democrats vented their fury in
a new round of tumults. The prime minister, the politically
moderate General Ernst von Pfuel (successor to the moderate
liberal Rudolf von Auerswald, who had resigned in September),
was not prepared to take responsibility for the declaration of a
state of siege the king had been demanding since 16 October,
209
whereupon he resigned. His successor, Count Frederick William of
Brandenburg, received a pronouncement of no confidence from the
National Assembly on 2 November. But neither this, nor the wish
of a delegation of representatives that he appoint a popular
cabinet, nor Johann Jacoby’s famous words to him on the occasion
(‘It is the misfortune of kings that they never want to hear the
truth’) could impress the king of Prussia.
On 9 November Frederick William ordered the Prussian National
Assembly to be moved to Brandenburg and adjourned until 27
November. When a majority of the parliament refused to comply
with the behest, which looked very much like a coup d’état, and
continued their deliberations on 10 November in the Berlin
Schauspielhaus, troops under General von Wrangel forced an end
to the meeting. The next session, in which a majority of the
delegates called for a general tax strike, experienced the same fate.
The Frankfurt central authority and the president of the German
National Assembly, Heinrich von Gagern, attempted to mediate,
but in vain. On 5 December 1848, the king dissolved the National
Assembly and, against the will of the camarilla, proclaimed a
constitution. Many, though certainly not all, of its significant
provisions followed the ‘Charte Waldeck’, the draft of the
constitutional commission of the National Assembly. It even
included general male suffrage for elections to the second chamber.
One of the main differences was in the king’s power of veto: the
constitution of 5 December granted the sovereign an absolute veto,
whereas the commission’s draft contained only a suspensive veto,
that is, the power to postpone legislative resolutions of the
chambers. The next day, on 6 December, the government
210
announced elections for both chambers of the new parliament for
January 1849. The most important task of the assembly was to be
the revision of the imposed constitution.
Frederick William’s coup, which in many ways can be
considered a ‘revolution from above’, brought the March
Revolution to an end in Prussia, at least for the time being. The
power of the old state had triumphed over the new understanding
of rights that had emerged from the revolution. It was a quick
victory, due mostly to the fact that large portions of the population,
in the cities as well as the country, had come to see the
consolidation of royal power as an effective antidote to political
and social radicalism. In any case, the call for a tax strike, itself a
violation of current law, found little resonance, and most moderate
liberals seem to have shared Gustav Mevissen’s analysis of the
situation. In a letter of 8 December 1848 the Rhenish businessman,
a member of the Casino fraction in the Frankfurt parliament, spoke
of a ‘bold move by the king’ and believed that the moment had
arrived ‘in which all men of political influence and political
courage must place themselves on the basis of the new legal order
and fight the impending anarchy’. Mevissen’s political colleague,
the historian Dahlmann, thought no differently when he spoke of
the ‘the deed of deliverance, which is its own justification’ (Recht
der rettenden Tat) in the plenum assembly of the Frankfurt
parliament on 15 December, arguing for the right of the future
German head of state to have absolute and not merely suspensive
veto power over the decisions of parliament.26
211
‘GROSSDEUTSCHLAND’ AND ‘KLEINDEUTSCHLAND’
212
constitution. According to Article 1, the new Reich was composed
of the territory of the former German Confederation, with the
status of the Duchy of Schleswig and the borders of the Grand
Duchy of Posen ‘deferred until definitive ordinance’. Article 2
proclaimed that no part of the Reich was to be united in statehood
with non-German lands. When a German and a non-German
country shared the same head of state, Article 3 determined that
the relationship between the two was to be structured along the
lines of a purely personal union, that is, as a dynastic arrangement
and not a matter of national law.
Thus the grossdeutsch solution advocated by most of the
delegates at that time was tantamount to the dissolution of the
Habsburg monarchy. Several of the speakers in the debate,
however, made no secret of the fact that they neither believed the
plan could be effected nor even desired it to be. The Vienna jurist
Eugen von Mühlfeld, member of the moderate conservative fraction
‘Café Milani’, supported the idea of an alliance between a German
federal state and a federative Austrian state, with which position he
—consciously or unconsciously—took up Paul Pfizer’s thoughts
from the early 1840s. His Prussian party colleague, the
Westphalian Georg von Vincke, articulated very similar sentiments:
‘We want a confederation with all of Austria, and a federal state for
ourselves without Austria.’ Finally, Heinrich von Gagern, president
of the National Assembly and member of the right-liberal Casino,
spoke out for Austria remaining ‘in permanent and indissoluble
confederation with the rest of Germany, in view of its
constitutional conjoinment with non-German lands and provinces’.
The precise nature of this confederate relationship was to be set
213
down in a special federal act.
Those who argued like Gagern or Vincke advocated the
kleindeutsch (‘Little German’) solution, that is, the unification of
Germany under Prussian leadership and excluding Austria. Neither
speaker could imagine any other way a German national state
might be established. Mühlfeld came essentially to the same
conclusion, although he was less interested in a German national
state than in the integrity of the Austrian empire. In order to
mitigate the inevitable separation, the narrower federation of non-
Austrian Germany was to be supplemented with a broader
federation between Germany and the Habsburg monarchy. In this
point, too, members of the kleindeutsch ‘party’ agreed with
parliamentary defenders of the multinational Austrian state like
Mühlfeld. Advocates of Little Germany were typically Evangelical,
belonging to the moderate liberals or, less frequently, to the
moderate conservatives. They came from the educated or
propertied bourgeoisie, generally showed a good deal of sympathy
for and confidence in Prussia, and could be found much more
frequently north of the Main than in the German south.
The grossdeutsch ‘party’, in contrast, came from a variety of
backgrounds. For Catholics, the thought of having to live in a
kleindeutsch national state dominated by Prussia and Protestants
was difficult to bear. This fact alone predisposed many Catholics to
the grossdeutsch idea. German democrats had no intention of
subordinating their conception of the German Volk to dynastic
interests and of renouncing German-speaking Austria simply
because it shared a monarch with other peoples. The poet Ludwig
Uhland, professor of German in Tübingen and member of the left-
214
moderate Deutscher Hof, designated Austria an ‘artery of the
German heart’. Speaking before the National Assembly on 26
October, he said that Austria had mixed its blood into the mortar of
the new edifice of German liberty. ‘Austria must be with us and
remain with us in the new political Church of St Paul.’
At the end of October 1848, many moderate liberals still
thought as Uhland did; otherwise there would not have been a
majority to support the commission’s draft. The grossdeutsch vision
did not confine itself to the annexation of German-speaking
Austria, however. Italian Tyrol and Trieste, which gave Germany
access to the Mediterranean, were also included. On 20 June the
Frankfurt parliament had warned the Piedmont-Lombard-Venetian
associations blockading the Adriatic, including the harbour of
Trieste, that an attack on the latter would be considered a
declaration of war on Germany. Bohemia and Moravia, too, were
parts of Germany, such as the constitutional commission defined it.
With regard to these territories, one of the founding fathers of
German nationalism, Ernst Moritz Arndt (elected, like Friedrich
Ludwig Jahn, to the Frankfurt parliament and a member of the
right-liberal Casino), exhorted the Germans, with a quote from
Klopstock, not to be ‘too just’. Germany must hold firm to the
principle ‘that whatever has belonged to us for a millenium and has
been a part of us, must continue to belong to us’.
The Czechs in Bohemia and Moravia were not, of course,
linguistically German, nor did they desire to become German in
any sense. For all his linguistic nationalism, Arndt dismissed these
realities with sovereign indifference. Yet he was articulating a
conviction held by a majority of the German National Assembly:
215
when push came to shove, the right to self-determination of
Czechs, Slovenians, and Italians must—like the political will of the
Danes in northern Schleswig—take second place to the historically
justified national interest of the Germans. ‘Germany ought rather to
die than to give in and renounce any of the Fatherland.’ This
pronouncement of the leftist Karl Vogt during the debate over the
status of Italian Tyrol on 12 August 1848 was greeted with a storm
of applause from nearly the whole house. The only concession the
National Assembly was prepared to set down in the imperial
constitution was a tolerance clause guaranteeing the ‘non-German-
speaking peoples of Germany’ the right to their ‘native
development’ (volkstümliche Entwicklung) as well as ‘equality of
language’ in religious practice, schools, local governmental
administration, and local law. In the case of Posen, too, the
Frankfurt parliament (with the exception of the extreme left,
eloquently led by Arnold Ruge) rejected any compromise of the
‘healthy egoism of the people’, to use a term coined on 24 July
1848 by the East Prussian Wilhelm Jordan, delegate for Berlin
(originally a member of the democratic Deutscher Hof, then the
centre fraction ‘Landsberg’).
Among the 90 representatives who rejected the third article of
the draft, and with it the dissolution of the Habsburg monarchy
spelled out there, were 41 of the 115 delegates from Austria. The
non-Austrians of this group were mostly conservatives and
Catholics. Speaking for a number of the Catholic delegates on 24
October, the Rhineland native August Reichensperger (who soon
left the right-liberal Casino for the grossdeutsch ‘Pariser Hof’)
warned that when the ring that held the whole Austrian monarchy
216
together was broken, Germany would lose its ‘bulwark against the
east’ and hence be guilty of ‘Russian-style politics’. The Vienna
historian Alfred von Arneth (who at the end of 1848 made the
same change of party membership as Reichensperger) argued in
very similar fashion: by preserving Austria, ‘empires on the eastern
border of Germany, be they Slavic or Hungarian, would be
prevented.’ In the logic of this justification of the Austrian empire
lay a ‘mitteleuropäisch’ solution to the German question, one that
could no longer be called national, since it included the whole of
the Austrian empire along with non-Austrian Germany. Count
Friedrich von Deym, a non-aligned representative from Bohemia,
spoke of ‘a giant empire of seventy, possibly eighty million souls’.
Neither the kleindeutsch nor the grossdeutsch camp had any
intention of giving up the possibility of a German national state,
but the former had an easier time coming to terms with the
strategic arguments of the Grossösterreich (‘Greater Austria’) party.
Explaining his conception of a narrower and a wider federation,
Gagern emphasized that a close relationship between a unified
Germany and the Austrian empire would consolidate the German
influence in south-east Europe. If unified Germany were to support
Austria’s mission of ‘spreading German culture, language, and
customs along the Danube to the Black Sea’, then the German
emigrants presently headed for the west, for America, would
instead turn to these regions. As a ‘great world-commanding
people’, it is the vocation of the Germans ‘to bring those peoples
along the Danube who have neither call nor claim to independence
like satellites into our solar system’. Clearly, the kleindeutsch
solution was no platform of national self-moderation. Gagern and
217
his political associates entertained as little doubt about Germany’s
and Austria’s joint mission to rule south-east Europe as before them
Friedrich List, Paul Pfizer, and other Vormärz liberals had done.27
The idea of a narrower and a wider federation initially found so
little support in the Frankfurt parliament that Gagern withdrew his
petition at the end of October 1848. A month later, however, the
situation changed drastically. On 27 November Schwarzenberg
responded to the resolutions of the German National Assembly
from 27 October. Speaking before the Austrian Imperial
Parliament, which at the behest of the emperor convened in the
Moravian town of Kremsier from that month onward,
Schwarzenberg described the preservation of the Austrian empire
as a German and European necessity. Only when a renewed Austria
and a renewed Germany had taken solid form would it be possible
to officially determine their mutual relationship. Until that time,
Austria would continue to fulfil the duties incumbent upon it by
virtue of its alliances.
Soberly considered, Schwarzenberg’s categoric rejection of
Grossdeutschland only left room for a kleindeutsch solution. Thus the
position of Schmerling, the president of the imperial ministry, was
now untenable, and he resigned on 15 December 1848. His office
was assumed by Gagern, and the presidency of the National
Assembly was taken over by Eduard Simson from Königsberg, a
professor of Roman law, baptized Jew, and, like Gagern, member
of the Casino. The new president immediately became active on
behalf of his interpretation of the narrower and wider federation,
but Schwarzenberg, at the urging of Schmerling, dismissed the
idea.
218
In the German National Assembly, there had never been a
majority for the project of a confederation between Germany and
the whole of the Austrian empire, Schwarzenberg’s ultimate goal.
On 13 January 1849 the Greifswald delegate and Casino member
Georg Beseler, who came from Schleswig-Holstein, called the
‘middle kingdom [Reich der Mitte], controlling Europe with seventy
million,’ a ‘political monster’. ‘We do not accept this middle
kingdom. Europe would not allow it, and it would not satisfy
Germany.’ However, a kleindeutsch solution without Austria, among
the most decided supporters of which were the delegates from
Schleswig-Holstein under the leadership of Beseler, was at this time
—the beginning of 1849—supported only by a relatively narrow
majority. A strong minority, joined by many who had recently
defended the Austrian cause, would in no way accept a break with
Austria, even though important constitutional questions continued
to divide the proponents of Grossdeutschland into democrats,
republicans, and conservatives. There were, to be sure, similar
contrasts of opinion among those who were now, for the time
being, prepared to be satisfied with a German national state
without Austria. Supporters of an elective emperorship stood at
loggerheads with proponents of the hereditary principle.
Consequently, Gagern could not be certain of success in canvassing
a sufficient majority for his core programme, the creation of a
kleindeutsch national state under a hereditary Prussian emperor, by
making concessions to the moderate left in constitutional matters.
Once again, the unresolved questions were cleared up by
Austria. At the urging of Schwarzenberg, Emperor Francis Joseph,
acceding to the throne at the beginning of December 1848 after the
219
abdication of Ferdinand I, dissolved the parliament at Kremsier on
7 March 1849 and imposed a constitution proclaiming the national
unity of the Habsburg empire. Two days later, Schwarzenberg
demanded that the imperial ministry in Frankfurt accept the whole
of the Austrian state into a renewed German Confederation and
establish both a centralized German authority in the form of a
directory, to be dominated by Austria and Prussia, and a House of
States (Staatenhaus) composed of representatives from the separate
national parliaments. As the Austrian prime minister saw it, the
majority of seats in this body would be occupied by delegates from
the Habsburg monarchy.
The Vienna ultimatum broke up the grossdeutsch party. One of
its leaders, Carl Theodor Welcker from Baden, went over to the
camp advocating a kleindeutsch solution with a hereditary
emperorship. Welcker did not succeed in obtaining a majority for
his motion of 12 March calling for the wholesale acceptance of the
draft of the imperial constitution in its second reading and the
conferral of the hereditary emperorship on the king of Prussia.
However, the representatives of the liberal centre, led by Heinrich
von Gagern, agreed to support the moderate democrats under
Heinrich Simon in voting for a restricted, suspensive veto on the
part of the head of state, as well as for universal suffrage for the
populace. After this, on 27 March, a diluted version of the first
articles (originally presented by the constitutional committee in
October 1848) succeeded in finding a majority. According to the
revised version, German countries sharing a head of state with a
non-German country were required to have their own separate
constitution, government, and administration. Since it was clear
220
from the start that Vienna would reject these terms, the resolution
of 27 March was a tacit vote for Kleindeutschland, that is, for
separation from Austria.
During the debates leading up to this vote, several speakers
warned the assembly against the dangers of civil war in the event
of a break with Austria. Among these were Joseph Maria von
Radowitz, a conservative delegate of Hungarian descent,
committed Catholic, general in the Prussian army, and personal
friend of Frederick William IV, as well as Moriz von Mohl, a
moderate democrat from Stuttgart, Protestant, ardent federalist,
and member of the Westendhall. At the end of his speech on 17
March, in words greeted with applause from the left and the
gallery, Mohl invoked the likelihood of a new thirty years war:
The left did not seem much disturbed by the prospect of a large
European war. The zoologist Karl Vogt, member of the Deutscher
Hof and proponent of a federation between the German and
221
Austrian empires in their entirety, gave a speech on 17 March in
which he proclaimed that the time had come to fight, along with
Poland and Hungary, the decisive battle between west and east:
‘Gentlemen, do not denigrate and poison this holy war of Western
culture against the barbarism of the East with a duel between the
House of Habsburg and the House of Hohenzollern ... No,
gentlemen, you must be resolved to let this war be what it should
be—a battle of nationalities.’ This speech was greeted with a storm
of applause from the delegates on the left.
The leftist call for a ‘holy war’ against the ‘military despotism’ of
Russia and Austria was, however, as little able to impress the
majority of the Frankfurt parliament as had been the similar
messages of Marx and Engels in the Neue Rheinische Zeitung. On 27
March the assembly voted with a narrow majority to confer the
imperial mantle on a reigning German prince and to make the
office hereditary. On the next day, King Frederick William IV of
Prussia was elected emperor. Of the 590 delegates 290 voted for
him; 248 abstained. The imperial constitution, proclaimed and
issued by President Simson, went into effect that same day, 28
March 1849.
When the parliamentary delegation under Simson met with
Frederick William IV in Berlin on 3 April 1849 to offer him the
emperorship, the Prussian sovereign had long since made his
decision. He had no intention of exchanging the office of king by
the grace of God for an imperial throne by the grace of the people.
What bothered him was not merely the fact that the new
constitution offered only suspensive, not absolute, power of veto
against decisions of both chambers of the imperial parliament, the
222
People’s House and the House of States; this alone was tantamount
to the introduction of a bona fide parliamentary system. He was
also repulsed by the ‘fictitious crown, baked from mud and clay’
because it exuded the ‘sordid stench of revolution’—because, in
other words, the democratic legitimization of the imperial office
would have separated him from his equals, the emperors in Vienna
and Russia and the kings in the capitals of Europe. It was easily
foreseeable that the Austrian emperor would have interpreted
Frederick William’s acquiescence as a hostile act. It was probable,
too, that in the event of war with Prussia and Germany, the
Russian tzar would have supported the Habsburgs as actively as he
had done in the summer of 1849 against the Hungarian revolution.
Consequently, Frederick William’s thinly disguised refusal on 3
April 1849 came as no surprise to anybody who knew him well.
The final rejection on 28 April also included the constitution,
which by that point had been recognized by twenty-eight
governments, though not by the larger Mittelstaaten of Bavaria,
Württemberg, Hanover, or Saxony.
The Prussian rejection spelled the end of the German National
Assembly, which followed shortly after. The imperial ministry
under Gagern, which had been acting in a purely caretaker
capacity since 22 March, stepped down on 10 May. Austria had
already withdrawn its delegates on 5 April. Prussia, Saxony, and
Hanover followed suit in May, Baden in June. Most of the
moderate liberal representatives resigned of their own accord
between 20 and 26 May. In their own justification, a group of sixty-
five delegates, including Dahlmann, Gagern, Simson, Droysen, and
Beseler, announced on 20 May that the decision not to put the
223
constitution into effect was a lesser evil than the propagation of a
civil war that had already begun. A rump parliament of about one
hundred members, dominated by the left, removed to Stuttgart on
30 May, where it was forcefully dispersed by Württemberg troops
on 18 June.28
224
forces stood under the supreme command of Frederick William’s
brother William, the ‘Kartätschenprinz’,* who would become
emperor of Germany somewhat more than two decades later.
Engels expressed the paradoxical nature of the
Reichsverfassungskampagne in a polemically exaggerated
formulation: ‘Those who were serious about the movement were
not serious about the constitution, and those who were serious
about the constitution were not serious about the movement.’ On
the one hand, the revolutionaries condemned the Prussian king for
letting the constitution fail; on the other hand, the republicans in
their ranks rejected one of its core provisions, the hereditary
emperorship. In 1849, the military balance of power was such that
the rebellion stood no chance of success. And outside of Baden, its
civil support was too weak. While the liberal bourgeoisie was
disappointed with the Prussian king’s attitude, only a small
minority decided to make common cause with the radicals against
the state. Ultimately, the movement lacked a centre. It pursued
national goals, but possessed no national organization, without
which these goals were unattainable.
On the heels of this revolutionary epilogue from ‘below’
followed a diplomatic coda from ‘above’. On 3 April 1849, the day
the Frankfurt delegation was received in Berlin, the Prussian
foreign minister, Count Arnim-Heinrichsdorff, had announced to
the governments of Germany that Frederick William IV was
prepared to assume leadership of a confederation of willing
participants. In a memorandum of 9 May, the Prussian government
once again took up Gagern’s idea of a narrower and a broader
federation (it had first done so in a circular on 23 January). This
225
initiative was rejected by Schwarzenberg that same month. Among
the other German governments, Saxony and Hanover associated
themselves with the Prussian proposals and signed a ‘Three Kings’
Alliance’ (Dreikönigsbündnis) on 26 May in Berlin. The three states
accepted the constitution proposed by Radowitz for a new German
‘Union’ (the so-called Unionsverfassung), a document that differed
from the Frankfurt constitution in two significant respects: the
emperor was to receive absolute veto power over parliamentary
decisions, and the lower house of the parliament was to be elected
not according to universal, direct, and equal suffrage, but rather by
the three-class system favouring the wealthy introduced in Prussia
in May 1849.
The democratic left rejected the Prussian project of a German
Union. Moderate liberal proponents of the heredity principle (the
so-called Erbkaiserlichen), on the other hand, declared their support
at a meeting in Gotha at the end of June 1849. In the months
following, most of the German states joined the Three Kings’
Alliance, though Bavaria and Württemberg declined. In mid-
October, after Prussian influence in the new Union’s ‘administrative
council’ carried the decision to set lower-house elections for
January 1850, Hanover and Saxony withdrew again from the
alliance, under pressure from Schwarzenberg. Austria had long
been pushing for a restoration of the German Confederation. It now
threatened war if the parliament of the Union should compromise
peace and order in Germany. The parliament was duly elected in
January 1850. It met in Erfurt in March and passed the now
completed draft of the constitution. However, without the
participation of the kingdoms of Bavaria, Württemberg, Saxony,
226
and Hanover (which had in their turn formed a ‘Four King’s
Alliance’ in February), the new union remained a pale reflection of
Prussia’s original plan.
Shortly after these events, in July 1850, Prussia was also forced
to suffer a painful reverse on the wider European political stage. In
April of the previous year, when Denmark renounced the Malmö
armistice, Prussian troops had once again entered Jutland. A
second armistice with Copenhagen in July 1849 was followed in
July 1850 by the Treaty of Berlin and the First London Protocol,
which restored the duchies of Schleswig and Holstein to Danish
rule. Prussia, under massive pressure from Tzar Nicholas I, bowed
to these arrangements. Ever since the summer of 1849, when it had
come to the aid of Austria against the Hungarian revolution, Russia
had advanced to become the greatest protector of all conservative
forces in Europe. A Prussian king who risked war with the
autocratic tzarist regime would automatically hazard his support
among the nobility and military. Frederick William IV had no
intention of purchasing the approval of the German patriots at such
a price.
The tension between Germany and Austria also entered a new
phase about this time. The catalyst of the crisis was Hesse-Kassel,
where the prince elector came into conflict with the estates over
the constitution. In May 1850 the government of Elector Frederick
William had renounced membership in the new Prussian-led Union.
At the beginning of September, he dissolved the chamber, decreed
unconstitutional emergency taxation measures, and placed the land
in a state of siege. Austria, which had reconvened the Federal Diet
in Frankfurt on 2 September with the backing of most of the
227
German Mittelstaaten, took the side of the elector. Prussia backed
the estates, claiming among other reasons that Electoral Hesse was
still a member of the Union.
When the Federal Diet proclaimed the federal execution in
Electoral Hesse on 16 October, war between Prussia and Austria
seemed only days away. Once again, Russian pressure was brought
to bear, and Prussia backed down. On 28 November 1850
Schwarzenberg and Otto von Manteuffel, the Prussian foreign
minister, signed the Punctation of Olmütz. The core of this treaty
was the agreement to resolve the crisis in Electoral Hesse in
harmony with all the German states and to join together against
Holstein, which had gone to war on its own against Denmark after
the conclusion of the Treaty of Berlin. The reform of the
Confederation was to be negotiated immediately at a conference of
ministers in Dresden. In a secret addendum, Manteuffel agreed to
reduce the Prussian military to peacetime status at once.
Schwarzenberg was able to achieve few of his aims during the
negotiations, neither the acceptance of the entire Habsburg empire
into the German Confederation nor the recognition of the rump
Federal Diet as successor to the former parliament at Frankfurt. To
this extent, Olmütz was not the Prussian ‘disgrace’ the liberals
would long continue to decry. Yet Manteuffel, for his part, failed to
secure the principle of equal partnership between Austria and
Prussia for the proposed renewal of the Confederation, a status
offered by Schwarzenberg as late as May 1850. Prussia had been
compelled to make more concessions than Austria, and it was
correspondingly more difficult for the Prussian government to find
support for its policies in the Second Chamber of parliament.
228
Among the most eloquent defenders of the Olmütz treaty was a
conservative delegate in the Prussian parliament, Otto von
Bismarck. On 3 December 1850, in a speech to the second
chamber, Bismarck declared that ‘the only healthy foundation for a
great state is national egotism, not romanticism.’ While he
conceded that a Prussian rejection of the Austrian demands would
have been popular, he found no plausible case for war:
229
A belligerent left and a peaceable right. Such a formula, applied to
German history between 1848 and 1850, would be too summary,
but less inaccurate than the reverse. From Karl Marx to Karl Vogt,
the left saw in tzarist Russia—and quite correctly—the mortal
enemy of the European revolution. Thus it was quite logical to
make a successful war against Russia the first condition of
revolutionary victory, or to claim, as Arnold Ruge of the
Donnersburg fraction had put it during the parliamentary debate
over the Posen question on 22 July 1848, that such a war would be
‘the last war, the war against war, the war against the barbarism
that is war’. Benedikt Waldeck pursued goals only seemingly less
ambitious when, at the end of October 1848, he appealed to
Prussia to intervene against the Weimar counterrevolution. If the
Prussian government had allowed itself to be guided by the
demands of the parliamentary left in Berlin, the German war would
inevitably have broadened into a pan-European conflict. Even
during the parliamentary uproar over the Malmö armistice a month
earlier, revolution in Germany might easily have given way to
European war—in the event that Prussia had bowed to the will of
the momentary majority in the Frankfurt parliament.
The conservative forces, for their part, were by no means as
inclined to peace as Bismarck’s defense of the Olmütz agreement
made them out to be. After all, the military suppression of the
revolution in Austria, in parts of Germany and Italy, and finally in
Hungary was a kind of warfare. It was, to be sure, a warfare within
existing political borders, not for the purpose of their radical
revision, and for the defense of the traditional order, not for its
overthrow. Nonetheless, the traditional order can itself be
230
described as a kind of structural violence. Not least among the
factors contributing to the success of the conservatives was the fact
that the forces desiring a complete break with the past were never
more than minorities. The political, economic, and social
consequences of such a break were unforeseeable in scope. For a
vast majority of Germans, the familiar way of things was more
bearable than the unknown in the form of a German republic, a
social revolution, a civil war, or even a world war. This was not
only true for the educated and propertied bourgeoisie, but also for
the great majority of the lower middle classes and peasants.
Moderate liberals, who had never wanted revolution, moved to the
right in the same measure that the left became radicalized. Nothing
contributed more to this radicalization than the suspicion that the
moderates were prepared to yield unconditionally to the old
powers. This was the dialectic of the German—and not only of the
German—Revolution of 1848–9.
Gustav Rümelin, former Württemberg representative and
member first of the ‘Württemberger Hof’, the left-centre fraction in
Frankfurt, then of the klein-deutsch ‘Augsburger Hof’, expressed the
dilemma of moderate liberalism in classic fashion several times
throughout 1849. ‘German unity cannot, and ought not, be realized
along the path of civil war,’ he wrote in the Schwäbische Merkur at
the beginning of May. In September 1849, in an essay for the same
newspaper, he wrote that the ‘way to salvation and deliverance’
still lay in the ‘Gagern program’, the cooperation of a narrower and
a broader federation, an alliance between Germany and Austria for
mutual protection. After all, this programme had enjoyed majority
support in the National Assembly. Rümelin warned against two
231
other paths. If Prussia were once again to join the troika of
absolutist eastern powers, the ‘restoration of the old system, both
domestically and abroad’, would be the result. If, on the other
hand, Prussia maintained its independent course, which for
Rümelin meant the execution of the kleindeutsch solution in open
confrontation with Austria, there would loom ‘a battle between
northern and southern Germany, a civil war that, like the Thirty
Years War, will summon foreign armies over the borders and once
again make Germany the battlefield of Europe.’
The utopian view of a great war for the liberation of peoples,
the militant side of the dream of the ‘springtime of nations’, lent
considerable political plausibility to the convergence of liberal and
conservative forces. On no account did this convergence proceed in
only one direction, the assimilation of liberalism to conservatism.
Prussia was a constitutional state by the end of 1848, and unlike
Austria not in name only. Thus the Hohenzollern state had already
accommodated to liberalism in no small measure. If before 1848 it
could be called a comparatively progressive German country only
in terms of economic and social developments, the imposed
constitution of December 1848 reduced the political gap between
Prussia and the southern German states. Many features of Prussian
absolutism outlasted the revolution, especially in the military. But
as a whole, post-revolutionary Prussia was, without question, an
absolutist state no longer.
The constitutionalization of Prussia is one of the reasons the
typical assessment of the German revolution of 1848—namely, that
it failed on all fronts—falls short of the mark. According to its twin
goals of political liberty and German unity, the revolution certainly
232
did fail. Germany did not became a liberal national state, nor could
liberalism successfully assert itself in the individual German states.
And yet, after 1848, the political and geographic meaning of
‘Germany’ was much clearer than before. It was now much more
manifest who ‘Germania’ really stood for, that goddess with the
black, red, and golden banner in her left hand, sword and freely
rendered olive branch in her right, whose giant image had presided
for nearly a year over the meetings of the delegates in the
Paulskirche. The kleindeutsch party, a small minority before 1848,
had gained considerable ground. ‘Waiting for Austria is the death
of German unity.’ At the time it was uttered during the
parliamentary sitting on 12 January 1849, many liberals disagreed
with this phrase of Hermann von Beckerath, a representative of the
Casino. Yet it expressed a fact that sober consideration was now no
longer able to deny.
The experiences of the 1848 Revolution were necessary in order
to impress upon moderate liberalism a sufficiently realistic idea of
the boundaries of a German national state. This happened only
gradually. Before realism could triumph over wishful liberal
thinking, the counter-revolution had won the day, both in the two
German great powers and in Europe as a whole. The chances of
unifying non-Austrian Germany without military conflict with the
European great powers of Austria and Russia were now much
smaller than in the spring of 1848. In short: when the kleindeutsch
solution was politically feasible, the Germans did not yet desire it,
and when they finally did desire it, it was no longer politically
feasible.
The revolution did much to weld together the forces that, in
233
future, would never again relinquish the goal of a liberal and
unified Germany. Liberals and democrats throughout all of
Germany developed closer associations than they had ever had
before. Their rapprochement also influenced their respective
political programmes. The common labour over the imperial
constitution, especially the sections treating of basic rights, had
established pan-German standards for the things to be achieved in
order to assure the victory of progress, both in the separate German
states and in a future unified nation. Equality before the law was
one of these basic German rights. A second was full freedom of
religion and conscience. A third would have ended the privileged
status of one religious community over another and abolished a
national church. Discrimination of the Jews would have thus come
to an end—if the fathers of the constitution of 1848–9 had had
their way.
The main reason the revolution failed was because German
liberalism had to deal with too many political desires and demands
within too short a space of time. It proved impossible to achieve
both unity and freedom simultaneously. In the older national states
of the European west, especially Britain and France, national
unification was achieved gradually, in the course of centuries,
through the agency of monarchs and assemblies of the estates.
Those desiring more freedom were able to work within the existing
political framework. In Germany, a framework for liberal and
democratic projects had to be created. The German liberals (in the
narrower political sense) labouring on these structures were fully
aware that they needed the power of the larger German states, led
by Prussia, in order to protect the work of national unification from
234
outside threat. For that very reason, and not only because they
feared social revolution, they sought to avoid a policy of
confrontation with the older powers—a policy promoted and
prosecuted by the left.
The left was correct when it claimed that the moderate liberals’
willingness to compromise had helped the forces of the old regime
recover quickly from the storms of March 1848. But to the question
of how Germany could be liberalized and unified at the same time,
the democrats and socialists had no answer. The leftist call to a
pan-European war of national liberation was an expression of
German intellectual wishful thinking, uninformed by any
consideration of the true balance of power, both domestically
within the individual states and between the states on the
international stage. Consequently, it was blind to the human costs
of the desperado politics it espoused. If the war clamoured for by
the left had indeed broken out, the counter-revolution would no
doubt still have won, but the victory would have been far more
sanguinary than the events that actually took place between
autumn 1848 and the closing months of 1850.
German political culture paid a disastrously heavy price for the
failure of the liberals and democrats to achieve their goals of unity
and liberty by their own power. The authoritarian deformities of
German political consciousness persisted and were reinforced. Yet
the fact remains that the ambitious twin goals of 1848 were
objectively unattainable. That they were was not entirely a
misfortune. For if the radical revolutionaries had gained the
opportunity of putting their programme into practice, the
consequences would probably have been a European catastrophe.30
235
4
236
date meant to be a ‘realist’, a ‘positivist’, or even a ‘materialist’. In
1856 the liberal Berlin National-Zeitung enthusiastically described
the new way of thinking:
237
the Erfurt Union once again participated in the Diet, thus restoring
the pre-revolutionary political order without any constitutional
reform. The eastern provinces of Prussia were still considered
outside Confederation territory, as they were before 1848. In the
preceding conferences in Dresden in spring 1851, Prussia had
sought in vain to win Austrian approval for the principle of parity
in the form of a rotating presidency in the Federal Diet. Austria
remained what it had been since 1815, the presidential power of
the German Confederation.
On 31 December 1851 Austria abolished its imposed
constitution of 1849, becoming an absolutist state once again.
Prussia kept its constitution. However, when the second chamber,
elected in January 1849, recognized the imperial constitution on
21 April, it overstepped the king’s and cabinet’s notion of
parliamentary authority, whereupon Frederick William IV dissolved
the assembly in May. That same month, the system of general
suffrage was abolished by emergency decree, replaced with a three-
class system, the Dreiklassenwahlrecht. As a gesture of protest, the
democratic left refused to vote in the elections of July 1849, as
well as in the two following elections of 1852 and 1855.
The new chambers revised the constitution, but almost entirely
in favour of the monarchical authority. The king returned to being
a sovereign ‘by the grace of God’. The administration depended
solely on the confidence of the king, not of the parliament. Nor did
the latter have any control over the army, bureaucracy, or foreign
affairs. While royal decrees still needed to be countersigned by the
ministry, the power to command military operations was explicitly
exempted from this requirement. And yet, despite all these
238
restrictions, Prussia continued to be a fundamentally constitutional
state after 1849–50. On 6 February 1850 the king swore to uphold
the constitution firmly and unswervingly, and to rule in accordance
with it. His power was great, but not limitless.
Among those who had advised Frederick William to swear this
oath was Friedrich Julius Stahl, the intellectual leader of the
conservative ‘Kreuzzeitung party’, founded in 1848 and named after
its leading newspaper. Stahl had also written a defence of the
‘monarchic principle’ in 1845. The revised Prussian constitution
was, in Stahl’s view, in harmony with this principle, and his
example allowed—indeed, compelled—his conservative colleagues
and followers to endorse constitutional monarchy. For Stahl, the
law existed exclusively through the state, and the state exclusively
through the law. The conservative commitment to the ‘inviolability
of the order of law’ made by Stahl in spring 1849 contained the
insight that this very order ‘is, at the present, a barrier against the
arbitrary will of the people, as it has been hitherto against the
arbitrary will of the prince’. Thus the Rechtsstaat, the state under
the rule of law, was an answer to the ‘permanent revolution’, the
constitution a bulwark against democracy. When it came to
adapting old agendas to new conditions, a not inconsiderable
number of post-revolutionary conservatives, led by Stahl and Ernst
Ludwig Gerlach, the president of the Magdeburg appeals court,
demonstrated a willingness to learn.
Prussian realpolitik was part bureaucratic, part feudal in
expression. The main goal of the conservative bureaucracy, which
had its most influential representative in Otto von Manteuffel, the
prime minister from 1850 to 1858, was to reverse the expansion of
239
the individual and societal rights and liberties gained during the
course of the revolutionary period. To this end, political and press
offences were withdrawn from trial by jury, and the freedoms of
assembly and association were curtailed. The Junkers succeeded in
restoring the manorial police, the right to guarantee the integrity of
their estates through entailment (Fideikommissen), and the old
system of district and provincial diets that guaranteed a strong
predominance of the manorial class.
Many conservatives, including the king himself, would gladly
have gone even further. Frederick William and several of his closest
advisers contemplated a radical redrafting of the constitution to
reflect corporative principles. But the king was cautious enough to
make all alterations of the constitution dependent on the approval
of both parliamentary houses, thus eschewing another coup d’état.
The first chamber, in any case, was amenable to his wishes. After
the House of Representatives had sanctioned a constitutional
amendment granting him the right to determine the composition of
the first chamber, the king created a House of Lords (Herrenhaus).
Its membership consisted exclusively of men of ‘birth’, no longer
elected, but proposed for nomination by certain privileged bodies
and then personally selected by the king. The restructuring of the
first chamber as a House of Lords was the high point of the
reaction in Prussia. Yet the new upper house was no guarantee that
the system would last, no more than the highly developed
surveillance apparatus and the numerous political trials the
Prussian state used to combat democrats and other opponents.2
Compared to Austrian neo-absolutism, Prussian
constitutionalism of 1850–8 could almost be called a ‘mild’
240
reaction. In this the Hohenzollern state resembled the German
Mittelstaaten, which also retained their constitutions (the Grand
Duchy of Baden was the only one that did not take a reactionary
turn in the 1850s, or did so in a very limited fashion) and with
whom Prussia continued to share economic interests in the
framework of the German Customs Union. When in 1849 the
Austrian minister of trade, Karl Ludwig von Bruck, began to build
upon Schwarzenberg’s central European plans by promoting the
project of a central European customs and economic union (i.e. a
merger between the Zollverein and the Habsburg monarchy),
Prussia necessarily saw this as a challenge. The contest for the
patronage of the other German states was decided in favour of the
northern power. In February 1853 Prussia concluded a separate
trade agreement with Austria, which was to last twelve years. In
April of the following year, the German Customs Union was
extended another twelve years. A few months earlier, on 1 January
1854, Hanover, which hitherto had lead a separate north German
customs union, the Steuerverein, joined the Zollverein. To be sure,
the battle to determine the relationship between the German
Customs Union and Austria was not decided at this time, but only
postponed until 1860, when a new round of discussions over
customs agreements would take place. Still, the contours of
Kleindeutschland had become much clearer in the course of 1853–4,
at least in terms of customs and trade policy.
In matters of foreign policy, too, Prussia was no longer content
to remain in the shadow of the Habsburg empire. The events in the
Crimea offered an opportunity to demonstrate initiative. Originally
(beginning in November 1853) a conflict between Russia and
241
Turkey, by March 1854 the Crimean War had drawn Britain and
France in on the side of the latter. Britain was primarily interested
in defending the security of its interests in the eastern
Mediterranean. Napoleon III, the new ‘Emperor of the French’ by
virtue of a plebiscite at the end of 1852, was primarily concerned
with gaining prestige through military and foreign policy successes.
In Prussia, opinions were extremely divided over the position
the country should take. A moderate conservative faction, the
‘Wochenblatt party’ around Moritz August von Bethmann-Hollweg,
wanted Prussia to join the western powers in order to gain their
support for a return to the policies of the Erfurt Union. Victory in
the war on the Black Sea could help Prussia to unify Germany at
the expense of Austria. The old-conservatives decisively rejected a
break with Russia, Prussia’s old ally. In their view, the real enemy
was still the revolution, embodied by Bonapartist France. The
prime minister, Manteuffel, held yet a third position, wanting to
keep Prussia neutral as long as possible. As it transpired, that was
not entirely possible. In April 1854 Prussia and Austria concluded a
defensive alliance, an ‘addendum’ to which demanded that Russia
clear out of the Danube princedoms it had occupied, Moldavia and
Wallachia.
Vienna had insisted upon this clause as a means of consolidating
Austria’s position in the Balkans. When, at the instigation of Count
Buol, the foreign minister, it went a step further and concluded a
treaty of alliance with France and Britain, but without renouncing
its neutrality, Prussia refused to go along. Otto von Bismarck,
Prussian ambassador to the Federal Diet in Frankfurt since 1851,
canvassed the relevant committees and brought a majority together
242
that took the anti-Russian sting out of the Austrian motion to
mobilize federal troops, thus neutralizing its intended effect. This,
the first major foreign policy success of the Junker from Mark
Brandenburg, was also one of the turning points in the history of
the relationship between the two German great powers.
The Crimean War ended in the defeat of Russia. In the Peace of
Paris (late March 1856), the tzarist regime was forced to agree to a
demilitarization of the Black Sea region, among other things. The
era of the ‘Holy Alliance’, the pact proclaimed in 1815 by the three
eastern powers for the defence of the old order, had finally come to
an end. For the foreseeable future, Russia was no longer in a
position to prosecute its conservative interventionist politics, the
high point of which had been the suppression of the Hungarian
revolution in summer 1849. The formerly good relationship
between Russia and Austria suffered lasting damage. Prussia, on
the other hand, which had thwarted Vienna’s anti-Russian
diplomatic offensive, could hope that the Russians might one day,
when occasion arose, show their gratitude for services rendered.
Austria suffered an additional setback that weakened its position:
the Kingdom of Sardinia-Piedmont, which had entered the war in
1855 on the side of Britain and France, succeeded in making the
Italian question part of the agenda at the peace conference in Paris.
The war changed the balance of power in Europe. The west had
won, the east had lost, and if one could speak of a hegemonic
power on the Continent, it was no longer Russia, but France.
There were winners and losers in Germany, too. The old
Prussian conservatives rightly considered the eclipse of autocratic
Russia and the strengthening of Bonapartist France as a personal
243
defeat. The plight of the liberals, in contrast, who were no less anti-
Bonapartist than the conservatives, improved to the extent that the
reactionary forces could no longer count on the tzarist empire. Yet
Bismarck was also among the winners. The Prussian ambassador to
the Federal Diet saw no occasion to regret what he considered the
most important result of the conflict: thanks to its neutrality,
Prussia now had more options and a greater scope of action than
before. It had liberated itself from the one-sided dependence on
Russia and Austria. Its status as a European great power had
increased, and along with it the status of Otto von Bismarck.
The distinctly non-ideological attitude Bismarck showed during
the Crimean War earned him the criticism of a staunch legitimist,
General Leopold von Gerlach, the older brother of the conservative
party leader Ernst Ludwig von Gerlach. But Bismarck knew how to
defend himself. ‘Sympathies and antipathies with regard to foreign
powers’, he wrote on 2 May 1857 to Gerlach,
244
be the only representative, or even the representative
κατ,έξоχńυ [kat’ exochén: ‘pre-eminently’ (H.A.W.)] of the
revolution ... How many entities still exist in the political world of
today without roots in revolutionary soil? Take Spain, Portugal,
Brazil, all the American republics, Belgium, Holland, Switzerland,
Greece, Sweden, and England, which even today is consciously
grounded in the ‘glorious revolution’ of 1688 ... Bonapartism is not
the father of the revolution; it is, like every other kind of
absolutism, only a fertile field for its seed. I am by no means
seeking here to place Bonapartism outside the purview of
revolutionary phenomena; I only wish to reveal it for what it is,
without all the accretions not necessarily intrinsic to its nature.
Bonapartist France as Prussia’s ally in a future confrontation
with Austria—for Leopold von Gerlach the idea was sacrilege, but
for Bismarck it was something reasons of state could conceivably
render plausible, if not imperative. The Prussian ambassador to the
Federal Diet did not cease to be a conservative when he confronted
conservative party ideology with the conservative interest in an
increase of Prussia’s power. ‘We must deal with realities and not
with fictions.’ The verdict of his letter to Gerlach of 2 May 1857
reflected his years of experience with Austria in the Federal Diet.
Before the Crimean War he would hardly have spoken so explicitly
about the long-term exigencies of Prussian foreign policy. Although
the Crimean War was not, as Heraclitus said of war in general, the
‘father of all things’, it was an extremely effective tutor in a
discipline that first received its name during this era: realpolitik.3
245
THE LESSONS OF 1848
246
‘Prussia must grow in order to survive, and Austria must not allow
Prussia to grow, lest she herself be destroyed. That is the reality
that gives the interrelationship between the two countries its true
character.’
Accordingly, realpolitik meant above all recognizing the
interests of the German states, especially the two great powers, as
clearly as the interests of the German nation.
247
that were to happen, her power and security would still not be fully
adequate. If we imagine her borders stretching all the way to the
borders of Austria, then we cannot but see even more vividly that
such a Prussia would not rest until she had finished the job by
annexing the German lands of Austria. Even if the ambition of the
Berlin cabinet were completely satisfied with the control of non-
Austrian Germany, the ambition of the nation would compel it to
decide the competition with Austria in a final either-or showdown.
248
constitutional movement at least a portion of its rights, and has
thereby sown the seeds of new battles.
249
perfectly conscious of the fact that the social question could only
be solved by policies of social reform, and not by suppressing
socialism. Yet the policies he proposed remained completely within
the boundaries of liberal notions of ‘helping people to help
themselves’, including ‘the greatest possible scope for the spirit of
association’, namely labour union activity.
‘Help yourself, and God will help you’: thus goes one of the wisest
sayings that passes from mouth to mouth among the people. And
vice versa: he who does not help himself cannot be helped by
either God or the state. ‘Help yourselves’ is the watchword of the
North American spirit of enterprise and the North American
worker, the magic formula that has created on the other side of the
ocean an economic power of the first order within two lifetimes
and a common prosperity that history has never yet known.5
250
The consequences were now plain to see. In the republic that
emerged from the February revolution in Paris, the opposition
between capital and labour was transformed into a broader
antagonism between property and poverty, between haves and
have-nots. The defenders of property closed ranks behind the
industrial reaction. Although Louis Napoleon owed his victory in
the presidential elections of December 1848 primarily to the rural
populace, he immediately went over to the side of the propertied
classes. The industrial reaction thus had the power of the state in
its hands, an autonomous power independent of the parties. The
contradictions inherent in industrialized society were not thereby
resolved, however. Far from it. Now a new battle commenced, the
issue of which was not yet decided. This was no longer simply the
old conflict between capital and labour, but ‘the battle between
social democracy and the industrial reaction’. If the latter emerged
as the victor, ‘then the result will be the definitive rule of capital
and the subjugation of labour, even in a legal sense. If social
democracy wins, then it will impose its own contrasting order upon
society—perhaps only after a very bloody interval.’6
It was still possible to avert a development of this kind in
Germany, including the ‘bloody interval’ that went along with it.
But in order for that to happen, the monarchy would have to recall
its historical task of embodying an idea of the state ‘elevated above
all societal interests’ and, for that very reason, standing for the
principle of liberty. Since the king was the representative of the
independent idea of the state, it followed ‘in a natural manner that
the part of society subject to rule will turn—now out of immediate
instinct, now consciously and intentionally—to the monarchy as its
251
natural protector’. In complying with this expectation, the
monarchy would simultaneously be fulfilling its own need for
independent action. It lay in the nature of this need ‘that the
monarchy, to whom the highest power of the state has been
entrusted, should use it in an independent manner, promoting
against the will and the natural inclination of the ruling class the
elevation of the lower class, hitherto subjugated in state and
society.’
In Stein’s view, the reward of such autonomy lay not only in the
possibility of gaining for the king the support of the entire lower
class. More importantly, the sovereign would link his very
existence with ‘the happiness of the state, the love and trust of the
genuine people’. In doing so, the monarchy would ‘identify the
throne itself with the idea of liberty, lending it the most secure
support humanly possible to achieve ... By thus fulfilling its truly
divine mission within its people, the monarchy shall wear a double
crown!’
This, then, was the lesson Germany and Prussia could learn from
the developments in France since 1848: in order to pre-empt the
dangers of industrial reaction and social revolution, the monarchy
must do everything it can to gain ‘an infinitely great social power
to complement its power of state’.
252
reform, then monarchy will, here and everywhere, henceforth
become either an empty shadow or a despotism, or else sink into
republic.
253
debate. But that such an institution could be considered the central
focus of politics, the thing holding together, ruling, mediating and
balancing the life of the state—in this sense it is indeed impossible.
254
internally through social reforms, externally through the
unification of Germany. If it was successful, then the idea of the
‘double crown’ was doubly meaningful. The Prussian state would
be a monarchy by both the grace of God and the will of the people,
and to the crown of the king of Prussia would be joined the crown
of the German emperor. Stein did not expand any further upon this
scenario, but the contours of his ultimate aspirations were plain to
see.7
The basic difference between Rochau and Stein was obvious.
The liberal publicist depended upon the reasoned insight of the
bourgeoisie and the self-help of the working class to solve the
social question. The conservative theorist of the state, in contrast,
expected the monarchy to do everything, and thus, unlike Rochau,
had no interest in increasing the influence of bourgeois-controlled
parliaments at the expense of governments. In one matter,
however, the two did agree. Both wanted to avoid social revolution
and considered social reform as the means of doing so. This was
what fundamentally differentiated Rochau and Stein from a radical
socialist like Karl Marx, who was motivated by the failure of the
1848 revolutions to undertake a critical review of his ideas
concerning the conditions for a successful proletarian revolution.
The result of these reflections was Die Klassenkämpfe in
Frankreich,* written in London, where Marx was living after his
exile from Prussia and after a brief sojourn in Paris in the late
summer of 1849 (the text appeared in 1850 in instalments in the
Neue Rheinische Zeitung. Politisch-ökonomische Revue, at that time
published in Hamburg). Marx’s historical and theoretical
255
reappraisal of the experiences of 1848 did not lead him to revise,
but to radicalize his ideas concerning the revolution of the working
class. Systematic suppression of the class enemy was going to be
the only way the proletariat could hold onto any power it would
gain—this was, for Marx, the most important lesson of 1848. His
view is reflected in his synopsis of the programme of ‘revolutionary
socialism’ or ‘communism’:
Marx did not refer in this text to the historical example of the
Jacobin terreur of 1793, although it was continually present in his
thoughts. In 1847 he had written that ‘the reign of terror, by its
mighty hammer blows, could only serve to spirit away, as it were,
the ruins of feudalism from the soil of France. The anxious and
considerate bourgeoisie could not have accomplished this task in
decades. The bloody acts of the people simply cleared its path.’ At
the beginning of November 1848, after the victory of the counter-
revolution in Vienna, Marx gave expression (in the Neue Rheinische
Zeitung) to his hope that ‘the cannibalism of the counter-revolution’
would convince the people that there was only one way ‘to shorten
the murderous death-throes of the old society and the bloody birth-
pangs of the new, to simplify them, to concentrate them’, namely
256
through ‘revolutionary terrorism’. The idea of the ‘dictatorship of the
proletariat’ as a proletarian reign of terror was not a secondary
element in Marx’s attempt to draw conclusions from the course of
the 1848 revolutions, especially the revolution in France. In a letter
to his friend Joseph Weydemeyer on 5 March 1852, Marx counted
the realization ‘that the class struggle necessarily leads to the
dictatorship of the proletariat’ among the core elements of his
theory.8
Marx’s momentous correlation between the ‘bourgeois’ and the
‘proletarian’ revolutions did not escape the criticism of his
contemporaries. In 1853 Rochau drew attention to a significant
difference between the ‘third’ and the ‘fourth estate’. ‘It is a
meaningless platitude,’ he wrote,
to speak of a fourth estate that will take the place of the middle
class, just as the middle class once took the place of the aristocracy.
There is no inner connection between the two ideas; the former is
prophecy, the latter actual historical process. The middle class
wrested power from the nobility not because it was more numerous
(indeed, the number of abused peasants was far greater still, and it
availed them nothing), but because it was intellectually and
morally superior, as well as more prosperous. These were the
qualities that gave the middle class both the claim to greater
political importance and the power to obtain it. The very lack of
such qualities, in comparison, is what best characterizes the so-
called fourth estate. However much sympathy ignorance,
uncouthness, and poverty deserve, only a complete fool could
accord to them the vocation of political leadership and rule. If we
257
strip away those negative qualities and educate it, then the so-
called fourth estate will simply blend together with the middle
class, and the only possible antagonism remaining will be the one
between the middle and the upper classes.
258
critique themselves, continually interrupt their own course and
return to what seemed already accomplished in order to begin it
anew. They deride with cruel thoroughness the half-measures,
weaknesses, and pitiful inadequacies of their first attempts. They
seem to hurl down their enemy only to let him draw new power
from the earth and rise up once again, even more colossal than
before. They quail again and again in the face of the indefinite
monstrousness of their own purposes, until a situation is created
that makes all turning back impossible, and the very conditions
themselves cry out:
259
the Crimean War, Marx began to take the political role of the army
seriously. In 1859 he spoke of the ‘establishment of the rule of the
army in place of rule through the army’, and shortly afterwards
even of the ‘fundamental antagonism between bourgeois society
and the coup d’état’. The materialist interpretation of history
seemed to be at the end of its rope. The societal basis, no longer
capable of determining the political superstructure, was now at the
mercy of the latter. Napoleon III had made the impossible possible.
The theoretical untenability of the prevailing conditions fostered
in Marx, and even more in his friend Friedrich Engels, an intense
longing for crisis, and, when crisis came, the inclination to
concentrate revolutionary hopes upon it. When the world economic
crisis broke out in 1857, Engels rejoiced: ‘The crisis will do me
good physically, as good as a swim in the sea, that I can already
tell. In 1848 we said, “Now our time has come,” and it came in a
certain sense. But this time it is coming fully and utterly. This time
it is life or death.’
Although the economic crisis did not fulfil its promise, signs of
danger soon increased on the political front, not least of all in
Prussia. In October 1858 Prince William assumed the regency on
behalf of his mentally ill brother, King Frederick William IV, and
within a short period of time formed a liberal-conservative ministry
under Prince Karl Anton von Hohenzollern-Sigmaringen. At first,
Marx cautioned against exaggerated expectations. After the victory
of the moderate liberals in the lower-house elections in November,
however, Engels was convinced of Prussia’s ‘political reawakening’,
a development he compared with similar events in other countries
and thus saw as part of a pan-European watershed. Ten years after
260
the revolution, the bourgeoisie seemed no longer prepared to
accept the consequences of its political self-incapacitation, its
subjugation under military and political despotism. Its societal
power had increased with its wealth, and now the bourgeoisie
began once again to feel the political fetters laid upon it.
Engels’s conclusion was simultaneously a prediction:
261
of the king, now quite deranged, that he refuse to take the oath
upon the constitution. The swearing of the oath before both
chambers on 26 October 1858 was followed on 8 November by a
sensational speech before the state ministry, composed of moderate
conservatives and ‘old-liberals’ in equal proportion. The most
famous passage spoke of the ‘moral conquests’ Prussia must make
in Germany. This revealed William’s intention of breaking with his
brother’s realpolitik and moving the unification of Germany
forward.
Another part of the speech received somewhat less attention—
William’s reference to changes in the armed forces, necessary in
order to give Prussia real clout in European power politics. The
‘New Era’, talk of which commenced immediately both within
Prussia and abroad, began amid great hopes, and these contributed
decisively to the overwhelming victory of the moderate liberals in
the Prussian parliamentary elections of November 1858, where
they gained an absolute majority of seats in the second chamber.
Many supporters of the democrats also returned to the polls for
the first time since 1849. Nonetheless, the staunch left was not yet
prepared to enter its own candidates, contenting itself for the time
being with recognizing the legality of the revised constitution and
announcing its willingness to support the government under certain
conditions. These conditions were spelled out in the Königsberg
‘Wahlaufruf der preussischen Demokratie’,* written by Johann
Jacoby: the left expected the state ministry under the prince of
Hohenzollern-Sigmaringen and his representative, the old-liberal
Rudolf von Auerswald, to ‘conscientiously observe the constitution
of the land and to further develop it in a liberal sense on the basis
262
of the law’.
The exaltation of autumn 1858 did not last long. The
government was slow to introduce reforms of the inner
administration, and, when it did so, it met with resistance and
delaying tactics from conservative ministers and provincial
governors. The House of Lords, too, was defiant, blocking liberal
bills in civil law as well as tax reform measures aimed at the
privileges of the Junker estates. Yet the old-liberals refrained from
using their parliamentary strength against the ministry and the first
chamber. The watchword of the leading fraction, led by and named
after Georg von Vincke, former delegate to the German National
Assembly, was ‘Let us not be pushy.’ Governmental liberals
adhered to this philosophy throughout all of 1859, to the
increasing discomfort of the democrats and many younger liberals.
In consequence of a major foreign event, the Italian war of
independence, the year 1859 was to mark a caesura in the history
of the liberal and national movement in Germany.10
In July 1858 Napoleon III and the prime minister of the
Kingdom of Sardinia-Piedmont, Count Camillo di Cavour, came to
an agreement that France would aid Sardinia in the conquest of
Lombardy and Venetia—that is, in a war with Austria—and, in
turn, receive from its ally the provinces of Nice and Savoy. This
secret agreement was completely in keeping with the policy of
close cooperation Cavour had begun with France in 1855, when he
led his country to war in the Crimea on the side of Britain and
France. At the end of April 1859, after Cavour ignored an
ultimatum from Vienna, Austrian troops invaded Sardinia-
Piedmont. The war for the unification of Italy had begun, and it
263
was only natural that this conflict would stir public opinion in
Germany to a far greater extent than the Crimean War a few years
previously.
The debate over the question of which side Prussia and Germany
should take in a war pitting Austria against Italy and France
prompted the emergence of completely new political ‘camps’ in
1859. The most unified of these was a group advocating Prussian
support for Austria in return for the latter’s recognition of Prussia’s
political and military leadership in non-Austrian Germany. To this
group belonged most of the kleindeutsch liberals of 1848, including
the historians Johann Gustav Droysen, Heinrich von Sybel, and
Georg Waitz. Far more mixed were the groups of unconditional
supporters and unconditional opponents of Prussian intervention
on behalf of the Habsburgs.
The group of unconditional supporters included old Prussian
conservatives like the Gerlachs, the grossdeutsch democrat Waldeck,
the socialists Marx and Engels, and, far more surprisingly, the anti-
Habsburg liberal Rochau and the former president of the German
National Assembly, Heinrich von Gagern, who went over from the
kleindeutsch to the grossdeutsch party at this juncture. Among those
who favoured a break with Austria, even in alliance with France,
were the socialist Ferdinand Lassalle, the ‘Forty-Eighter’ democrats
Arnold Ruge and Ludwig Bamberger, the liberal publicist
Konstantin Rössler, and the conservative political realist Otto von
Bismarck. The latter, upon Prince William’s assuming the regency,
had been compelled to give up his post as Prussian representative
at the Federal Diet; now, as Prussian envoy in St Petersburg, he had
been, as he saw it, ‘put on ice’ on the Neva.
264
As he wrote in a confidential letter to the general adjutant of the
Prince of Prussia on 5 May 1859, Bismarck would have preferred to
send the Prussian army to the south—with boundary posts in their
knapsacks, to be pounded in ‘either at Lake Constance, or wherever
the Protestant confession is no longer dominant’. The Germans,
once taken ‘into possession’ by Prussia, would gladly fight ‘for us’,
‘especially if the prince regent does them the favour of renaming
the Kingdom of Prussia the Kingdom of Germany’.
Ferdinand Lassalle saw things similarly. Born in Breslau in 1825,
son of a Jewish merchant, Lassalle was strongly influenced by both
Fichte and Hegel. He became an active follower of Marx during the
revolution; then, after serving a prison term for inciting resistance
against state authorities, he turned to political theory and writing.
In his pamphlet Der italienische Krieg und die Aufgabe Preussens. Eine
Stimme aus der Demokratie,* Lassalle considered Austria the very
embodiment of the ‘reactionary principle’ and consequently a more
dangerous enemy than Napoleon III, who was, after all, compelled
to base his rule on democratic principles like general suffrage.
Although a politics in the style of Frederick the Great—like the
conquest of Austria’s German territories and the proclamation of
the German Reich—was out of reach for Prussia at the moment,
nonetheless it could still lay claim to glory. ‘If Napoleon reworks
the map of Europe in the south according to the national principle,
so much the better; we shall do the same in the north. If Napoleon
liberates Italy, so much the better; we will take Schleswig-
Holstein!’
Marx and Engels, for their part, remained faithful in 1859 to
their belief that no regime stood more in the way of the revolution
265
than that Bonapartist government. Therefore it was in the
revolutionary interest to defeat or at least weaken Napoleon. For
that reason alone, Germany was obligated to ‘defend the Po on the
Rhine.’ In his pamphlet Po und Rhein,† written at the beginning of
the year and published in April, Engels urged the Germans to cause
whatever harm they could to France and to avoid moral reflections
about ‘whether such behaviour is in keeping with absolute
standards of justice and the national principle. One must save one’s
own skin.’ Once Germany achieved unity, then it would still be
able to forgo ‘all the Italian rubbish’.
The policy the Hohenzollern state actually pursued in the early
summer of 1859 disappointed all parties: it remained neutral. In
late June, after the Austrian debacle at Solferino, Prussia declared
itself willing to attempt an ‘armed mediation’ at the side of Britain
and Russia. But this proved superfluous when, in early July,
Emperor Francis Joseph and Napoleon III agreed to a compromise
peace. In the provisions to the treaty of Villafranca, Austria
renounced claims to the greater part of Lombardy, but was able to
retain Venetia, Mantua, and Peschiera. Thus the Italian national
state that emerged from the war and the ensuing revolutionary
popular movement was still incomplete (Venetia was not taken
until 1866, Rome four years later). Yet the Vienna system of 1815,
already shaken by the Crimean War, had lost one important
buttress, and that gave heart to all forces in other parts of Europe
fighting against the hegemony of monarchist legitimacy in the
name of the national principle.11
The leaders of Austria were aware of the possibility that the
266
Italian example might find imitators. In the year following the
1859 defeat, the Habsburg monarchy made another attempt to
become a constitutional state. The October Diploma of 1860
granted the nobility of both German and non-German crown lands
considerable autonomy within a federal structure. Yet this solution
provoked so much hostility among the Hungarians and the
German-Austrian liberals that it was replaced four months later by
a new constitution, the February Diploma.
The constitution of 1861, the work of the former imperial prime
minister and now newly appointed minister of state, Anton von
Schmerling, created an ‘Imperial Council’ (Reichsrat) as a
centralized legislature for the entire monarchy. Hungary was
partially excepted; all questions pertaining only to Cisleithan—that
is, non-Hungarian—territories of the empire were the province of a
‘smaller council’ (engerer Reichsrat). But the Hungarians, Czechs,
Poles, and Croats sent no delegates to the assembly out of protest
against the new system, and the Imperial Council had to begin as a
primarily German rump parliament. This was clearly not going to
be the way to fortify the cohesion of the empire. Still, the February
Diploma had one advantage: a ‘German’ and ‘liberal’ Imperial
Council in Vienna might well impress public opinion in Germany,
especially if Prussia continued to disappoint liberal expectations.
Prussia was still inspiring hopes as late as the autumn of 1859.
What had long been obvious to kleindeutsch liberals now came to be
accepted by many who had hitherto inclined in the grossdeutsch
direction: namely that after the defeat of the Habsburg empire,
only the Hohenzollern state could protect the Germans against
France. The organizational expression of this national-political
267
rapprochement between liberals and democrats was the German
National Union (Deutscher Nationalverein), founded in September
1859 in Frankfurt am Main after the example of the Italian Società
Nazionale of 1856. The leading personalities behind this
organization were the Hanover jurist Rudolf von Bennigsen on the
liberal side, and on the side of the moderate democrats the district
judge Hermann Schulze-Delitzsch, who in 1848–9 had been a
member of the Prussian National Assembly on the left centre.
For the new association, the only remaining solution to the
German question was the kleindeutsch project, a unification of
Germany under Prussian leadership. The promise to keep
membership in a German federal state open to Austria’s German
territories was little more than a tactical concession to southern
German predilections and animosities. The founding of the
National Union sent out a signal impossible not to hear. Liberals
and democrats put aside their remaining differences (for example
on the question of suffrage), since both parties were now convinced
that the creation of a liberal national state was more important
than everything else and would tolerate no further delay. The
Italian example began to catch on in Germany, too.
In terms of social composition and ideology, the new association
was dominated by the educated and propertied bourgeoisie, the
Bildungsbürgertum. Confessionally speaking, its orientation was
Protestant. With soon more than 25,000 members, however, the
National Union was far more than an assembly of dignitaries. It
stood in close contact with various other groups, including
organizations for the education of workers, gymnastics
associations, defensive alliances, and shooting clubs. Still, most of
268
these managed to retain their independence from the Union’s
aspiration to leadership, and many of them remained essentially
grossdeutsch in sentiment even throughout the 1860s. The National
Union was only one expression of the national idea, and certainly
not the one closest to the Volk. It played virtually no organizational
role in the celebrations in honour of the centenary of Friedrich
Schiller’s birth in November. Although it did take part in the
gymnastics and shooting festivals in Coburg, Gotha, and Berlin in
1860 and 1861, as well as the German Shooting Festival in
Frankfurt am Main in July 1862 and the German Gymnastics
Festival in Leipzig in August 1863, the Union did not succeed in
taking control of these events. It also failed, despite great efforts, in
its attempt to constrain the German gymnastics movement to
cultivate the practice of military gymnastics (Wehrturnen).
Nonetheless, on many occasions the German National Union was
able to set the tone. These included numerous celebrations in
honour of the centenary of Fichte’s birth in May 1862, as well as
the events within the framework of a Reichsverfassungskampagne in
the spring of 1863. It also played a significant role in the collection
of contributions for the reconstruction of the German fleet, first
created in 1848 by the provisional central authority, then broken
up in 1852 by the Federal Diet and ultimately sold in public
auction. The recipient of the donations was the Prussian ministry of
the navy, until internal developments in the Hohenzollern state
brought the campaign to a sudden end in spring 1862.
It was within the German academy that the Union was perhaps
most influential. Its organizational structures traversed all political
boundaries, and its major publication, the Wochenschrift (edited by
269
Rochau), provided a firm base for all varieties of kleindeutsch
nationalism. The Union was the organizational anticipation of an
all-German party, something not yet possible in reality. And
indeed, the National Union provided the impetus for the founding
of a new political party very soon after it came into existence.12
Nonetheless, a different kind of impetus was going to be
necessary before the system of political parties in Prussia would
experience a profound transformation. This catalyst was the reform
of the army. Upon assuming the regency, Prince William gave the
reorganization of the armed forces highest priority. That such
reform was necessary was uncontroversial. The last reform, the
Scharnhorst-inspired Heeresgesetz of 1814, now lay a half-century in
the past, and despite the fact that Prussia’s population had grown
from 11 to 18 million in the intervening years, the numerical
strength of the army had not changed since Frederick William III
fixed it in 1817. Of the 180, 000 annually eligible for military
service, only about 40, 000 were drafted.
But the increase in the number of recruits (by about 23,000) was
only one of the goals set at William’s behest by the new Prussian
minister of war, Albrecht von Roon, appointed in December 1859.
The reform plans also included a restructuring of the relationship
between the national guard (Landwehr) and the regular troops. The
three younger annual levies of the national guard would, in future,
be shifted to the reserves and thus incorporated into the army. This
was tantamount to a weakening of the ‘bourgeois’ element in the
Prussian military system. This effect was no less evident with
respect to another measure: Roon wanted to retain the three-year
term of military service, even though, from a purely military point
270
of view, the two-year period of training in effect since the
revolutionary era was perfectly adequate. From the point of view of
William and the generals, however, the issue had as much to do
with politics as with the military. Another year of training and
discipline would better serve to turn the recruit into a dependable
pillar of the Prussian soldier-state, a system that was to be
defended not only against external enemies, but also internally,
against overthrow and revolution.
The conflict with the bourgeoisie and its political
representatives was thus clear from the outset. Strictly speaking,
the Prussian House of Representatives was only authorized to
decide upon the reform’s budgetary dimension, not upon matters
pertaining to military organization. In effect, the budget was the
lever Roon’s opponents were constrained to use in their effort to
block the reform. This opposition was by no means particularly
resolute. On the contrary: despite its reservations, the parliament
approved a ‘provisorium’ permitting the government to move
ahead with the reorganization, despite the fact that the latter was
not clearly defined. Just less than a year later (18 January 1861, a
few weeks after the death of Frederick William IV), William had
himself crowned king in Königsberg. It was clear to all observers
that, as king, he would pursue the reform even more emphatically
than before. Nonetheless, in spring 1861, the parliament once
again approved ‘provisional’ funding—a designation the
government rejected on the quite appropriate grounds that the
increase in the armed forces was already a matter of established
fact.
The liberal majority of the chamber committed a great mistake
271
in letting slip the opportunity to tie the hands of the state ministry
in its use of the approved funds. The error was costly. On 6 June
1861, shortly after the passing of the second provisional finance
measure and the end of the parliamentary session, the German
Progressive Party (Deutsche Fortschrittspartei) was founded in Berlin.
This was an alliance of younger liberals who had long opposed the
all too compliant attitude of the ‘Vincke fraction’, as well as
democrats who had cut their political teeth in 1848. The regional
focal point of the opposition was in East and West Prussia, a fact
that earned the liberal rebels the mocking title ‘Young Lithuanians’.
The most active representatives of the group were the East Prussian
Junker Leopold von Hoverbeck and the Elbing lawyer Max von
Forckenbeck (originally from Westphalia). The party soon found
numerous supporters in the most eastern Prussian province, not
least among free-trade-oriented noblemen and merchants. Much
better known were several democrats who joined the Progressive
Party, foremost among them Johann Jacoby from Königsberg and
two other former members of the Prussian National Assembly of
1848, Benedikt Waldeck from Westphalia and Hermann Schulze-
Delitzsch from the province of Saxony, both of whom had recently
been elected to the lower chamber.
In the close cooperation of liberals and democrats, the
Progressives were following the footsteps of the National Union
from two years earlier. Indeed, the new party considered itself a
kind of executive of the latter. But it was more than that. The
German Progressive Party was Germany’s first really ‘modern’
political party. In terms of organization it went far beyond the
typical election committees, developing a firm party infrastructure
272
from the local to the national level. It put forward a binding party
platform emphasizing the common ground between liberals and
democrats, their mutual commitment to liberal reforms in Prussia,
and German national unification. Differences, though de-
emphasized, remained. On the controversial question of suffrage,
the democrats retained their commitment to the universal
franchise, whereas the liberals (in a narrower sense) preferred a
census that gave the propertied classes more political influence
than the propertyless masses. With new lower-house elections set
for December, however, the party’s foremost concern was the battle
for the consolidation of the constitutional state and thus against a
reorganization of the army that would effect the very opposite.
The parliamentary elections were a landslide victory for the
Progressives, who received over 100 seats and immediately became
the strongest fraction. Together with the left centre, the other
group of ‘resolute liberals’, they controlled more than half the
seats. In awareness of its strength, the Progressive Party put
forward the ‘Hagen motion’ at the beginning of 1862 in direct
challenge to the government. In this motion, the deputies furthest
to the left demanded more stringent itemization (the term used was
‘specialization’) of the state budget, including for the current year.
The purpose was clear: to prevent the state ministry from covering
the surplus costs of the army reform out of other budgetary titles,
as was the customary practice. The old-liberal finance minister,
Robert von Patow, spoke of a ‘vote of no confidence’, but could not
prevent the parliament from taking up the Hagen motion with 177
votes against 143. Five days later, on 11 March 1862, King William
I dissolved the lower chamber. On 14 March he dismissed the
273
ministry he had appointed in November 1858, at the beginning of
the New Era. The new cabinet, under the nominal leadership of
Prince Adolf von Hohenlohe-Ingelfingen and the actual leadership
of the finance minister, August von der Heydt, was now exclusively
conservative in composition.
The voters responded on 6 May 1862. The losers in the
parliamentary elections were the conservatives, the Catholic
fraction, and the old-liberals, the victors once again the staunch
liberals. The Progressive Party gained a further twenty seats.
Together, the liberal fractions controlled about four-fifths of the
seats. In the middle of September, the delegates Friedrich von
Stavenhagen and Heinrich von Sybel, both from the Left Centre
Party along with Karl Twesten of the Progressive Party, proposed a
last compromise measure on the basis of the two-year period of
military service. Although Roon himself and the entire ministry
regarded the compromise as a solution, the obstinacy of the king
spelled its doom. William briefly considered abdication, then
allowed Bismarck to talk him out of it. Bismarck, Prussian
ambassador in Paris since May 1862, had been summoned to Berlin
by Roon from his vacation in Biarritz. In long conversations in the
palace and gardens of Babelsberg on 22 September, Bismarck
managed to win the king over to a policy of confrontation with the
parliament. That same day, William conferred on Bismarck
provisional leadership of the state ministry.
In his Gedanken und Erinnerungen,* Bismarck wrote that he had
succeeded in convincing the king that for him ‘it was not an issue
of “conservative” or “liberal” in this or that nuance, but rather a
matter of monarchical vs. parliamentary rule,’ and that the latter
274
‘must unconditionally be averted, even at the price of a period of
dictatorship’. In speaking of a choice between monarchical and
parliamentary rule, as well as with the term ‘dictatorship’,
Bismarck was obviously referring to the British revolution of the
seventeenth century. Just as obvious was his rejection of the liberal
understanding of constitutionalism. From 1862 to 1866, when
Bismarck ruled without a budget approved by the parliament, the
Prussian government could no longer be described even in terms of
‘sham constitutionalism’. It was non-constitutionalism, pure and
simple.
In calling for clarification of the budget, the lower house had
brought up the question of power. The response of the Prussian
soldier-state was the ‘gap theory’ (Lückentheorie) as practised and
justified by Bismarck: whenever one of the two chambers upset the
balance between the three legislative powers (king, House of Lords,
House of Representatives) by withholding necessary budgetary
funds (a scenario for which the constitution made no provision),
then, according to the monarchic principle, it was the duty of the
royally appointed state ministry to rule without budgetary
legislation until the chamber in question retrospectively approved
the expenditures made in the interim.
On 23 September the House of Representatives pronounced its
final rejection of the army reform budget. A week later, on 30
September, Bismarck declared before the budget commission a
later oft-quoted maxim of his political philosophy:
275
nobody expects them to play Prussia’s role. Prussia must gather her
forces and hold them ready for the most propitious moment, a
moment that has slipped away several times already. Prussia’s
borders according to the Vienna treaties are not conducive to a
viable political existence. The great questions of the day will not be
settled by speeches and majority decisions—that was the great
mistake of 1848–9—but by iron and blood.
276
read by Bismarck, the delegates were told that the administration
could not roll back the reorganization of the army and must now
conduct the governmental budget in the absence of a basis in the
constitution. It was promised, however, that the non-budgetary
regime would not be extended beyond the fiscal limits set by law,
and the government was optimistic that, in due course, the
chambers would retroactively approve whatever expenditures had
proved necessary in the intervening period.13
277
unification of Germany will be brought about as surely as a law of
nature is fulfilled—not, however, by iron and blood, but rather by
iron and coal.’
Through all liberal factions ran the belief that the constitutional
conflict was ultimately rooted in a societal conflict, the ‘battle
between the bourgeoisie and the Junkers in alliance with absolutist
tendencies’ of which the Preussischen Jahrbücher spoke in 1862.
Karl Twesten (originally from Holstein), an official in the Berlin
city court and one of the most astute Progressives, had written in
1859—citing the sociologist Auguste Comte, founder of scientific
positivism—that the ‘dominant social classes’ must necessarily
become the ‘political ruling classes’. On 22 February, in one of his
last speeches before the chamber during the constitutional conflict,
Twesten called the struggle ‘between the Junkers and the people’
the ‘true conflict’. ‘The liberal majority of the House of
Representatives is not, as the conservatives claim, the bourgeoisie as
a small, particular class, living from large enterprises and filled
only with material interests’; but rather the Bürgertum, which
‘represents both the material and the ideal interests that fill the
working people and the thinkers, classes that have been steadily
rising since the end of the Middle Ages and that have the moral
authority always in hand and, sooner or later, will have the
political authority of our state in hand as well.’14
When in their own company, however, Prussian liberals often
expressed themselves more sceptically. The ‘whole opinion
concerning our moral successes’ is certainly only ‘a chimera’, wrote
Leopold von Hoverbeck to a friend at the end of June 1865.
278
The circles of society that read newspapers and occupy themselves
somewhat with politics formed their views on these questions long
ago and remain true to them to this day ... But upon the great mass
of the people, the third and even partly the second electoral class,
our debates have no influence, since they learn nothing of them—if
the official provincial press doesn’t make them even more
absolutistic than they are already, by virtue of their whole
education.
279
Bismarck in 1863 and the beginning of 1864. If they had, they
would have been able to claim with greater justification that the
socialist agitator was stabbing the liberal opposition in the back in
its confrontation with the ministry. On the other hand, however,
the Progressives themselves were hardly in a position to deny the
accusation Lassalle made against them in the progressive-liberal
district association of Friedrichstadt in Berlin during the election
campaign in the spring of 1862—namely, that the liberals ‘do not
dare to admit openly that constitutional questions arise not from
questions of law, but from questions of power’.15
The liberal publicist Heinrich Bernhard Oppenheim contradicted
Lassalle’s critique in the Deutsche Jahrbücher (which Oppenheim
edited) in 1864, arguing that, alongside the real relations of power,
the ‘public consciousness of the law’ was also a ‘power that must be
dealt with if lasting conditions are to be achieved. From the
conflict between such opposites proceeds the mediating character of
most constitutional documents.’ Hereafter, however, the author
expressed such strong doubts about the political maturity of the
bourgeoisie that critical readers were forced to wonder if
Oppenheim’s assertion concerning the power of consciousness of
the law even applied to Prussia.
280
For the Berlin National-Zeitung, it was established fact that the
weakness of the Prussian bourgeoisie had its roots in the specific
character and traditions of the Hohenzollern state. ‘In vain has the
Prussian bourgeoisie waited fifty years for institutions to be cleared
away that keep its political status below the measure it has been
granted in all other neighbouring states—with the exception of
those in the Slavic east,’ lamented the liberal newspaper in April
1862.
The gap separating our conditions in this respect from those of the
other German states has already become so wide as to threaten the
closer community in the most worrying manner. What is called the
‘Junker state’ appears so strange and repulsive there that every
sympathy immediately freezes at the mere mention of the name.
The Prussian liberals did not blame only their own bourgeoisie
for the fact that the old Prussian soldier-state and Junkerdom had
survived to the present day. In their opinion, the rest of Germany
shared responsibility for the conditions in Prussia by continuing to
accept the Bundeskriegsverfassung of 1821. In keeping with Article
VIII, which disallowed ‘even the appearance of the supremacy of
one federal state over another’, only three Prussian army corps
were permitted in the federal army. Despite the fact that three-
quarters of its territory belonged to the Confederation, however, as
a European great power Prussia maintained six further army corps
for reasons of security. As much as they despised Prussian
militarism, the other German states were not unhappy with this
arrangement. Ever since the wars of liberation, they believed they
could depend on Prussia in a crisis, and were able to restrict their
281
own military spending accordingly.
The consequences of this state of affairs for Prussian liberalism
were elucidated in the Deutsche Jahrbücher in 1861 by Wilhelm
Löwe-Calbe (democratic delegate to the German National Assembly
in 1848–9, last president of the Stuttgart rump parliament, and
Progressive Party delegate to the Prussian lower house since 1863).
The Prussian government, he wrote, needs the army ‘not so much
for Prussian purposes, but rather to fulfil her external obligations as
a German great power. Yet only from Prussia does the government
demand the means for these obligations as German great power,
whereas its duty is to demand them from Germany and to provide
only for Prussia’s own share.’ The high military expenditures in
Prussia were causing a ‘withering away and restriction of
prosperity and consequently of the available energies of the
people’. The most important goal was thus the creation of a
German army as the material guarantor of German unity.
Max von Forckenbeck expressed the same thought in a pithy
formulation: ‘Unless the German situation is given another form,
then in my opinion the existence of a reasonable and liberal
constitution is also, in the long term, an impossibility,’ he wrote in
a letter to Hoverbeck of 21 August 1859. ‘If the German situation
remains as it is, then only the military state will and must be
further developed in Prussia.’16
Under Bismarck’s rule, it became even more difficult for
Prussia’s ‘resolute liberals’ to gain an audience for such demands in
the rest of Germany. To be sure, the other liberal parties, which
also felt a close solidarity with the National Union, initially
282
applauded the Progressives in their parliamentary battle with the
non-budgetary regime of the new prime minister after October
1862. They also hailed the great victory of the Progressive and Left
Centre parties in the provincial elections at the end of October
1863, after the lower chamber had been dissolved once again. But
already in May 1863 the committee of the National Union was
expressing the wish that the House of Representatives would adopt
a more ‘resolute attitude’ toward the ‘insulting behaviour of the
government’. In the words of a resolution drafted by Karl Brater, a
liberal delegate in the Bavarian parliament, and edited by
Hoverbeck, Prussia had proved itself to be ‘the most dangerous
adversary of the national interests’. If its political leaders should
attempt to seize the ‘leadership of Germany’, then they would ‘find
the National Union among the first ranks of those to fight against
such temerity’.
Prussia suffered a dramatic loss of prestige in liberal Germany
after the autumn of 1862, at the same time the reputation of
Austria was on the rise again as a consequence of Schmerling’s
liberal policies. In addition, at the end October 1862, the National
Union had to deal with organized competition from the grossdeutsch
party. The German Reform Union (Deutscher Reformverein) united
Catholics and Protestants, conservatives, liberals, and democrats
who rejected Prussian hegemony. Among the founders of the new
grossdeutsch association were Julius Fröbel, the leftist delegate of
the Frankfurt parliament who in 1848 had hastened with Robert
Blum to embattled Vienna, where he barely escaped execution; the
Catholic historian Julius Ficker, the adversary of Heinrich von
Sybel in the debate over whether the medieval emperors would
283
have been better served by pursuing a nationalist policy instead of
Christian universalism; and finally the erstwhile leader of the
kleindeutsch party of 1848, Heinrich von Gagern, who had gone
over to the grossdeutsch camp in 1859 during the Italian war. Even
though the new organization was as unsuccessful as the National
Union in overcoming its internal antitheses, its founding in autumn
1862 was nonetheless a symptom. Kleindeutschland was now on the
defensive. Its situation worsened still further in summer 1863,
when Bismarck’s government, under the direction of the arch-
conservative interior minister, Count Eulenburg, began to use a
new press decree to shut down oppositional newspapers, some
temporarily, others permanently.
Austria tried to exploit this opportunity. At an assembly of the
German princes convened by Francis Joseph in Frankfurt, which
Bismarck induced King William I not to attend, the renewal of the
German Confederation was resolved in a reform act on 1
September 1863. The new constitutional draft proposed the
creation of a six-member directory with three permanent members
—Austria, Prussia, and Bavaria—flanked by a Federal Council
(Bundesrat) as a permanent congress of princely envoys as well as
an assembly of elected delegates from the parliaments of the
member states. But Prussian rejection doomed the project.
Bismarck insisted on veto powers for the two leading members in
matters of war, a presidency representing Prussia and Austria
equally, and a directly elected German national assembly.
In terms of the last two points, parity and direct elections, the
Prussian demands accorded with the resolutions of another body
convening at the same time and in the same place as the meeting of
284
the princes: the German Delegate Congress (deutscher
Abgeordnetentag), a congress of parliamentarians from non-Austrian
Germany. In mid-October the National Union rejected the Austrian
proposal as completely inadequate, but at the same time called into
question the credibility of the Prussian initiative. The Reform
Union supported Austria, but that was of little help to the Vienna
government; the German question could not be resolved against the
will of Prussia and a great part of German public opinion. Nor, for
that matter, could Prussia hope to achieve its version of the
national project as long as the government found itself in conflict
with the representatives of the people and with German
liberalism.17
After Bismarck’s nomination, there remained only one issue on
which Prussian liberals saw eye to eye with the administration:
trade policy. At the end of March 1862, the ministry under Prince
Hohenlohe-Ingelfingen and Count Bernstorff, the foreign minister,
concluded a liberal trade agreement with France, radically
reducing the tariffs between the two countries. Two years
previously, in the ‘Cobden Treaty’, France had largely dismantled
its customs barriers against trade with Britain. Thus the Franco-
Prussian agreement was tantamount to a fundamental decision for
economic integration in the free trade system of western Europe
and against the Austrian project of a central European customs
union built upon protective tariffs. With this decision, Prussia all
but put a stop to the negotiations between the German Customs
Union and Austria to coordinate or unify their respective tariff
policies. (Agreed upon in 1853, the negotiations were originally
planned for 1860, but then delayed in the face of Prussian
285
resistance.) The members of the Customs Union now had to make a
decision—between free trade and protectionism, between Prussia
and Austria, between west and east.
The liberal position was clear from the outset. ‘By means of the
French treaties’, it was announced in the report of the
parliamentary committees dealing with finance, tariffs, trade and
industry,
286
protectionist states collapse. In October 1864 the erstwhile
members agreed to a new customs treaty of twelve years’ duration.
A trade agreement with Austria was concluded in April 1865. In
terms of economic policy, the battle for the leadership of Germany
was decided in favour of Prussia.
The domestic repercussions of Bismarck’s successful trade policy
were great. The leaders of business and industry began to regard
the government in an increasingly positive light. For the liberals, at
that moment waging an embittered war against the prime minister,
this was hardly a good thing, since it meant that they were now
deprived of the active support of the very members of the
bourgeoisie for whose material interests they were so tenaciously
fighting. After the end of the constitutional conflict, the Preussische
Jahrbücher summed up:
287
necessitated German–Austrian cooperation. In May 1852 the five
European great powers, along with Sweden and Denmark, had
recognized (Second London Protocol) the integrity of the Danish
state and the right of Prince Christian of Schleswig-Holstein-
Sonderburg-Glücksburg to the throne of Denmark and its
neighbouring territories. Christian’s claim was through the female
line; through the male line, which was recognized in Schleswig and
Holstein, the rightful heir would have been Duke Christian August
von Augustenburg—who, however, renounced all claim to the
throne for himself and his descendants after a financial settlement.
Prussian and Austrian acquiescence in the London treaty was made
conditional upon the promise of the Danish king to respect the
special status of Schleswig and Holstein.
On 30 March 1863 King Frederick VII of Denmark issued letters
patent contesting this agreement. In July the Federal Diet
demanded their retraction and, when Denmark refused, proclaimed
the federal execution against Holstein on 1 October. Copenhagen
was unimpressed. On 13 November 1863 the government, a
cabinet controlled by the nationalistic Eider-Danes, presented the
parliament with a new constitution ordering the annexation of
Schleswig. Two days later Frederick VII died. He was succeeded by
Prince Christian, who became King Christian IX. One of the new
sovereign’s first acts was to sign the new constitution (18
November 1863). In doing so, he exploited the London Protocol’s
succession clause while simultaneously negating the whole basis of
the agreement.
The oldest son of Augustenburg, Frederick, who had already
attained his majority in 1852, had never accepted his father’s
288
renunciation of the throne. On 16 November 1863, the same day
Christian IX assumed the throne in Copenhagen, Duke Christian
conferred his right to succession in Schleswig and Holstein on his
son, who immediately proclaimed himself Duke Frederick VIII and
claimed the rulership of both duchies. All German nationalists,
kleindeutsch and grossdeutsch alike, applauded him. At the end of
December 1863 an assembly of almost 500 delegates from all parts
of Germany convened in Frankfurt, proclaimed the Augustenburg
cause the cause of the German nation, and set up a committee of
thirty-six representatives to coordinate their activities. Schleswig-
Holstein associations sprang up over all of Germany, organizing
demonstrations, proclamations, and collecting contributions for the
young duke.
The governments of the Mittelstaaten were also among Frederick
VIII’s supporters. Not so the two German great powers, who
insisted upon the renewal of the 1852 agreement, in light of which
Augustenburg’s position was untenable. Considering where the
other European great powers stood, this policy, developed by
Bismarck and accepted by the Austrian foreign minister, Count
Rechberg, was the only realistic one. Nevertheless, the Austro-
Prussian motion to proceed with the (thus far delayed) federal
execution against Holstein was not endorsed by the German
national movement, and most of the Mittelstaaten voted against it.
The measure was carried in the Diet on 7 December 1863, but by
an exceedingly narrow majority.
This tacit rejection of Frederick VIII enraged the supporters of
the National Union no less than those of the Reform Union.
Austria’s behaviour ‘has destroyed the Greater German party’,
289
lamented Gustav von Lerchenfeld, the president of the German
Reform Union and leader of the liberal chamber majority, in a
letter to the former ‘March minister’ of Saxony-Weimar, Oskar von
Wydenbrugk, on 25 December 1863. ‘If Prussia understands its own
advantage, it can now drive Austria out of Germany without so
much as a cock crowing afterwards.’
Federal troops entered Holstein on 16 January 1864, whereupon
the two German great powers issued an ultimatum: if Denmark did
not withdraw its new constitution, Schleswig would be taken as
security. When Copenhagen refused, war began—war for the return
to the London Protocol, not, as the southern German states and the
nationalists demanded, for the rights of Duke Frederick VIII. On 18
April Prussian troops took the Düppel entrenchments in the north
of Denmark, sealing the country’s fate. One week later, a European
conference began in London, but it was unable to come to an
agreement on any of the proposed solutions to the Danish problem,
including a simple personal union between Denmark and the ‘Elbe
duchies’ and a division of Schleswig along national lines. The
failure of the conference, which ended on 25 June, effectively
destroyed whatever political significance the London Protocol had
retained.
The ceasefire, effective since May, expired the next day. Austria
and Prussia immediately went on the offensive again, conquering
all of Jutland and the island From Düppel to Königgrätz 147 of Alsen.
Denmark was forced to sue for a ceasefire and peace. On 1 August
a preliminary peace agreement was concluded, the final treaty on
30 October. The Danish king renounced all rights in Schleswig,
Holstein, and Lauenburg to the emperor of Austria and the king of
290
Prussia, promising to abide by their decisions concerning the future
of the Elbe duchies.
The Vienna treaty made no mention of the hereditary claims of
Frederick VIII. To be sure, the foreign minister of Saxony, Friedrich
von Beust, had backed them at the London conference in the name
of the German Confederation, gaining the support of Austria and,
finally, of Prussia. But then, in an interview on 1 June, the heir
apparent learned the conditions under which Bismarck would be
willing to accept Schleswig-Holstein as an independent Mittelstaat.
They were terms calculated to make the Elbe duchies into a
Prussian protectorate: a Prussian naval base in Kiel, a federal
fortress in Rendsburg to be occupied by Prussian troops, military
and naval conventions with Prussia, shared construction of a canal
between the North Sea and the Baltic, and the separation of the
prince from liberal advisers. When Frederick VIII did not prove
immediately compliant, Bismarck dropped him. The Treaty of
Vienna left the final status of Schleswig-Holstein unresolved,
creating a construction that could, by its very nature, only be
temporary: an Austro-Prussian condominium over the Elbe
duchies.19
The controversy over Schleswig-Holstein’s future, which grew
especially heated in the summer of 1864, plunged the Prussian
parliamentary opposition into a violent conflict. From the
beginning of the Schleswig-Holstein crisis, the Prussian liberals
were in a more difficult position than their like-minded colleagues
in the other German states. While they hoped for military success
in the war against Denmark, the liberal delegates of the House of
Representatives also felt bound by their vow ‘not to grant this
291
ministry one cent’. They stuck to their clear budgetary standpoint,
despite the fact that one of the most outspoken of them, Karl
Twesten, had announced that if it were up to him to make a choice
between ‘whether the tenure of the Bismarck ministry should be
extended for a period or whether the duchies of Schleswig and
Holstein should be given up forever, I would not delay one moment
in choosing the former.’
While the great majority of ‘resolute liberals’ supported
Frederick VIII, a minority, under the leadership of the democrat
Waldeck, had no desire to aid in the accession of a new dynast.
They preferred the wholesale annexation of Schleswig-Holstein to
Prussia, which they considered the only ‘democratic monarchy’ in
Germany, the non-constitutional nature of the Bismarck regime
notwithstanding—a view Waldeck habitually justified with
reference to Frederick the Great’s ‘enlightened liberalism’. When a
opportunity to annex all of Schleswig-Holstein (including the north
of Schleswig with its predominantly Danish population) actually
did arise in summer 1864, however, the right wing of the
Progressive Party began to rethink its position.
At the end of May 1864 the National-Zeitung, the mouthpiece of
this faction of ‘resolute liberalism’, declared that ‘a Prussian
admission that she cannot liberate all of Schleswig would be
preferable to a war with several great powers.’ On 11 August, while
the negotiations over peace with Denmark were being conducted in
Vienna, the same newspaper published an article under the title
‘Glück und Macht’* delineating its fundamental stance: ‘To turn our
country into a barracks while making Schleswig-Holstein into a
peaceful farmstead—that is, for us, too one-sided, a too unequal
292
distribution of pleasure and pain.’
An appeal was sent out to the ‘friends of peace’ to no longer
look upon the Prussians as the ‘Spartans of Germany’.
that Germany cannot, and will never be able to, continue her
liberal development if Prussia regards herself as fundamentally a
military state, and is regarded as such by others ... For the exertion
of a country’s entire energies for the purpose of war is very rarely a
means of advancing the cause of domestic liberty at the same time
... Neither the Prussian people nor their representatives gain in
liberty when the state’s military duties are always seen as the
highest and most pressing. In this case, the highest taxes must be
paid according to the demands of the military offices, the highest
military expenditures must be defrayed while continually
strengthening the armed forces, and we know all too well what
then happens to the business of parliament. It has come to the
point where all those who recall the civic duties of the state receive
the honorary title of ‘wind-bag’, and only the accomplishments and
obedience of the soldier are seen as useful to the state.
293
At this time, the newspaper of the Progressive right wing was
still only concerned with Schleswig-Holstein’s military-political
integration with Prussia; it was not yet a question of annexation.
But the same arguments might have been used to justify the more
radical solution. Hitherto, the argument that the unequal
distribution of military expenditures as a result of the 1821
Bundeskriegsverfassung was preventing the liberal development of
Prussia had been more a matter of theoretical Progressive doctrine.
The victory over Denmark provided the first chance to examine its
practical consequences. Hitherto, Progressives had taken it for
granted that only a liberally minded Prussian government could
advance the cause of German unification. Now, a number of the
party conceded that a conservative ministry—indeed, even a
ministry in direct violation of the constitution—could take up this
task with equal success. If that was the case, then the relationship
between unity and liberty would have to be worked out anew. The
maxim with which the National-Zeitung brought its reflections over
the inadequacies of the Bundeskriegsverfassung to a close on 12
August 1864 was not merely a vote in the struggle over Schleswig-
Holstein’s future, but pointed far beyond the immediate occasion:
‘Yet every step forward in the attainment of the necessary German
power is, at the same time, a step forward for liberty. Conversely,
the prolonged neglect of power—power that must be seen as
indispensable for the attainment of nationhood—leads only further
into bondage.’
For the Progressive left wing, regardless of whether they were
‘Augustenburg’ advocates like Schulze-Delitzsch or supporters of
‘Greater Prussia’ like Waldeck, all such second thoughts about the
294
Bismarck administration were sacrilege. As long as the government
ruled without a budget approved by the parliament, it was to be
fought without any ifs and buts. In the spring and summer of 1864
the Berlin Volkszeitung, published by the democratic delegate Franz
Duncker and close in spirit to the viewpoint of Schulze-Delitzsch,
had declared as emphatically as the National-Zeitung against the
partition of Schleswig-Holstein according to the nationality
principle and on the basis of popular referendum. In contrast to the
Progressive right-wing newspaper, however, the democratic
Volkszeitung demanded that the citizens of the Elbe duchies be
guaranteed the right to determine their own domestic affairs.
Prussia was to impose nothing upon Schleswig-Holstein, but seek a
mutual agreement on liberal terms that took the duchies’ legitimate
military, economic, and political interests into consideration.
‘Unity and Liberty’ was the title of the democratic newspaper’s
response to the National-Zeitung on 16 August:
295
The conclusion to which the Volkszeitung came was no less
unambiguous than that of the National-Zeitung: ‘Germany’s unity
will only become possible through Germany’s liberty.’
The Austro-Prussian condominium over Schleswig-Holstein
proved to be so contentious that war seemed imminent in spring
1865. Under the leadership of Bavaria, the Mittelstaaten, themselves
under pressure from the German National Union, the thirty-six
representatives of the Delegate Congress, and the Schleswig-
Holstein associations, urged the Federal Diet to recognize the rights
of the duke of Augustenburg. Prussia responded with the ‘February
demands’, which went even further than the conditions Bismarck
had placed before Frederick VIII on 1 June 1864. Berlin now
required full military authority over the Elbe duchies, the
integration of their economies and transportation infrastructures
with Prussia, and the cession of territories around the mouths of
the planned North Sea–Baltic canal. On 6 April 1865 Austria,
seeking to counter the Prussian challenge, concurred with a federal
petition from the Mittelstaaten exhorting the two German great
powers to grant the duke of Augustenburg his rights. When at the
end of May King William I, with the support of the minister of war,
Count Roon, and the chief of the general staff, Helmuth von
Moltke, proclaimed his intention of annexing Schleswig-Holstein,
the outbreak of war between Austria and Prussia seemed only a
matter of days.
But it was Bismarck’s more flexible approach that prevailed in
the end. After lengthy negotiations in Bad Gastein from the end of
July to mid-August 1865, the Prussian prime minister came to an
agreement with Austria concerning the division and administration
296
of the duchies: Holstein fell to Austria, Schleswig to Prussia; Kiel
was to be a federal military port, though with Prussian
administration and police; Prussia was to construct fortifications in
the harbour and was granted the right to establish its own naval
station in the city. The duchy of Lauenburg was ceded to the king
of Prussia, who took it over in personal union.
The Gastein Convention of 14 August 1865—the event to which
Bismarck owed his elevation to the title of count—was as much a
provisional solution as the condominium it replaced, and explicitly
so. Yet Bismarck considered the time not yet ripe for war with
Austria, if indeed such a war was even necessary. There is much
evidence to suggest that, after the defeat of Denmark, the Prussian
prime minister was prepared to be satisfied for some time with a
Hohenzollern hegemony north of the Main. Such a solution to the
German problem might well have found the support of Austria,
provided it received compensation in northern Italy. However, the
German national movement would never, under any circumstances,
have consented to such a division of Germany, a grosspreussisch
alternative to Kleindeutschland, as it were. For that reason alone,
the much-debated ‘dualistic’ solutions of 1864–5 never had a
chance in the long term.20
As the 1865 crisis between Prussia and Austria was passing, the
schism within Prussian liberalism was deepening. In April Theodor
Mommsen, a historian from Garding in Schleswig-Holstein who had
been elected to the Prussian House of Representatives in 1863,
became the first right-wing Progressive parliamentarian to call for
the complete annexation of Schleswig-Holstein. The duchies’ right
to self-determination was in and of itself perfectly legitimate,
297
Mommsen explained in a circular letter to the delegates of the city
of Halle and the Saale district, ‘but it is not an absolute right. It is
limited by the general interests of the German nation. For there is
no distinct people in Schleswig-Holstein, only Germans, and when
the latter speak, the former must obey.’ Although Prussia, which
Mommsen considered a sort of trustee of the German nation, was
only justified in demanding partial—that is, military and maritime
—integration of the duchies, the people of Schleswig-Holstein were
well advised to assent to complete annexation.
Mommsen nonetheless rejected the naval bill presented to the
parliament in April 1865 requesting approval of funds for the
construction projects in the Kiel harbour.
298
and Germany against Austria, then things will go downhill. We will
become increasingly habituated to the actions and considerations
of small statehood, and one day we will find ourselves once again
under the hegemony of Austria. We will discover that the
population of Prussia, without an ideal goal, has lost the energy
necessary to fight the constitutional conflict with us to the end.
299
though not all for the same reasons. The Augustenburg party
deplored the disregard for Schleswig-Holstein’s right to self-
determination. The German Delegate Congress, convening on 1
October in Frankfurt, considered the entire lawful order and public
security in Germany fundamentally violated by Austria’s and
Prussia’s ‘breach of the law’. According to a statement by the
German National Union from the end of October, Schleswig-
Holstein had been liberated from arbitrary Danish rule only to be
‘raped’ by the German Confederation. Even the Prussian advocates
of annexation found fault with the convention. The National-Zeitung
complained that Prussia’s ‘ineradicable entitlement to both Holstein
and Schleswig’ was not reflected in the agreement. ‘No, this treaty
is not the work we have in mind; this solution has nothing in
common with our just demands.’
Only a few members of the Prussian parliamentary left took part
in the Delegate Congress in Frankfurt. The right wing of the
Progressive Party did not attend at all, wishing to pre-emptively
refute the expected proclamations of an anti-Prussian
‘particularism’. Mommsen and Twesten justified this decision in
open letters to the president of the Delegate Congress: ‘The
bankruptcy of particularism’, wrote Mommsen,
must reveal itself to each and every German with far more grievous
blows before a majority of the population of the middle and
smaller powers will cease only to be inspired by a nebulous idea of
German unity, a unity reserved for some distant future, and will
stop making every excuse—at the present moment, for example, by
draping their opposition to Prussia’s burgeoning power in the more
300
fashionable garments of opposition to Herr Bismarck’s carryings-on
—every excuse to run away from the beginnings of this unification,
wherever they present themselves in great earnest, and perhaps
under unpleasant and onerous conditions.
301
determination of Schleswig-Holstein. Afterwards, they had to come
to terms with a victory over Denmark that revealed the unbroken
power of the historical state. In both cases, in 1848 as well as in
1864, the Schleswig-Holstein question led to a ‘moment of truth’, to
which ‘left’ and ‘right’ liberals reacted in opposite ways. Whether
humiliating defeat as in 1848 or triumphant victory as in 1864, the
war with Denmark resulted in a polarization within the liberal
movement.21
In 1864–5 this crisis was not confined to Prussian liberalism.
After the victory over Denmark, the extreme right wing of the
National Union under the Göttingen attorney Johannes Miquel,
who had been a supporter of Marx in 1848, went over to the
annexationist line of Mommsen and Twesten, while most democrats
withdrew from the association after 1865. The adamantly
grossdeutsch democrats, the greater part of whom belonged neither
to the National nor to the Reform Union, remained hostile to
Prussia. When in April 1864 the Volkszeitung maintained that
Prussia needed only a firm national program to assume the
leadership of Germany, the Stuttgart Beobachter, the mouthpiece of
the Württemberg left, responded: ‘Derided, beaten, and kicked as
these Prussians are, they nonetheless want to conquer Germany.
Why don’t you conquer your own liberty instead? But if you are
slaves, do not demand that others share your lot!’
The Beobachter was speaking for the radically federalist
(‘particularist’ in the view of their opponents) faction of the
democratic movement in the smaller and medium-sized German
states when it announced on 10 February 1864 that wherever
nationality and liberty came into conflict, one must choose liberty.
302
The slogan that Germany’s future lay in a ‘confederation of free
states, a German Eidgenossenschaft’, derived from the same way of
thinking, influenced by the philosophy of natural law. When
Ludwig Pfau, one of leaders of the Württemberg left in both
parliament and press, sharply rejected the idea of a Prussian
empire in an article in the Beobachter at the end of April 1864, his
thoughts, paraphrasing the elder Cato, proceeded with a kind of
inner logic from the idea of a confederation: ‘Without the
dissolution of Prussia into its constituent peoples, it will be
absolutely impossible to form a unified and free Germany. Ceterum
censeo Borussiam esse delendam [‘Furthermore I think Prussia must
be destroyed’(H.A.W.)].’
The declarations of the Beobachter in spring 1864 were a kind of
platform, the founding documents of the ‘People’s party’ that began
at that time to detach itself from the Württemberg Progressive
Party, formed at the end of 1859. The political mobilization
brought about by the Schleswig-Holstein associations redounded to
the benefit of the democrats, who took advantage of the
widespread German anger at the Prussian and Austrian disregard
for Schleswig-Holstein’s right to self-determination. The old
Vormärz idea of the ‘Trias’—that the ‘third Germany’ of the smaller
and medium-sized states represented the authentic nation—was
once again revived in the language of the democratic movement, a
language that found fertile ground especially among the petty
bourgeoisie. With this vision, the democrats challenged moderate
Württemberg liberals—men like the Stuttgart attorney Julius
Hölder—who, despite their open criticism of the Bismarck
administration, held fast to the belief that Germany could not be
303
united without or against the will of Prussia. The views of the
moderates, who exercised strong influence over the second
chamber, were echoed by the industrialists and the educated
bourgeoisie, though not yet by the broad masses. In Württemberg
(and not only there), a politician could win considerable popularity
by attacking the two German great powers, especially if he
criticized Prussia more than Austria.22
Württemberg was the starting point of the German democratic
movement, which in 1865 began to organize itself throughout
Germany. The driving forces were not, however, natives of
Württemberg, but nationalist democrats from other south-west
German states, first and foremost the literary historian Ludwig
Eckardt from Baden and the physician and author of the much-read
materialist study Kraft und Stoff,* Ludwig Büchner from Hesse-
Darmstadt. It was in Darmstadt that the Democratic People’s Party
(Demokratische Volkspartei) was officially founded in September
1865—against considerable resistance from the Württemberg
democrats, who considered the new party insufficiently federalist
in outlook.
Eckardt and Büchner were the ‘liberal’ democrats who made the
greatest effort to include the workers in the People’s Party. Their
efforts were rebuffed by the General German Workers’ Association,
which had been plagued by leadership crises ever since its founder,
Lassalle, had died in August 1864 from wounds incurred in a duel.
The rival Union of German Workers’ Associations (Verband
Deutscher Arbeitervereine, founded in 1863 with the substantial
participation of Leopold Sonnemann, owner and editor of the Neue
304
Frankfurter Zeitung), on the other hand, joined the democratic
movement at its Stuttgart convention in September 1865. One of
the delegates at this congress was the man who shortly thereafter
would advance to become the leader of the German workers’
movement: August Bebel, a turner from Cologne, at that time
president of the workers’ education association in Leipzig.23
The founding of the People’s Party seemed to herald a shift to
the left, away from the alliance between liberals and democrats
and towards a greater left transcending class boundaries; away
from endeavours to reach a compromise with the government and
towards a resolutely oppositional course; away from realpolitik and
towards a politics of principle. However, the ‘third Germany’ also
saw developments tending in the opposite direction at the same
time, proceeding primarily from Baden. In Karlsruhe, the liberals
had been in power since 1860–1. Under August Lamey as minister
of the interior and Franz von Roggenbach as foreign minister, the
cabinet reformed the judiciary and bureaucracy, introduced
freedom of trade in 1862 (the same year as Württemberg, following
Saxony the year before), and removed the school system from the
controlling influence of the church. The latter reform incited a
bitter conflict with the ecclesiastic establishment that lasted years
and, while strengthening the cohesion of the liberal party, also
subjected it to growing pressure from a popular religious
movement.
In the first half of the 1860s, liberal Baden became the political
antipode to authoritarian Prussia and, paradoxical though it seems,
the latter’s partner of choice among the southern German
Mittelstaaten. On the issue of free trade, and thus on the basic
305
questions of customs union policy, the grand duchy and the
Hohenzollern state saw eye to eye. Yet the ruling liberals of Baden
advocated German unity under Prussian leadership not merely for
economic, but also for foreign policy reasons. The geographic
proximity to France created a need for security best served by
cooperation with Prussia. When it finally came to conflict between
the two German great powers, the decision to join the presidential
power of the German Confederation was, of all the middle powers,
the most difficult for Baden.
Among the governments of the other Mittelstaaten, sympathy for
Prussia was hardly to be found. Although Saxony acted in concert
with the north German hegemon on trade policy, in questions of
‘grand politics’ the foreign minister, Friedrich von Beust, tended to
side more with Vienna than Berlin. The Bavarian government, led
(since December 1864) by Ludwig von der Pfordten, a moderate
grossdeutsch liberal, wished to avoid a break with Austria at all
costs. The same was true of the foreign minister of Württemberg,
Karl von Varnbüler. Likewise, most of the other Mittelstaaten had as
yet given no grounds for the assumption that they would, if
necessary, support Prussia against Austria. Thus the Hohenzollern
state was forced to conclude that it would have the majority of the
Confederation members against it in any war with the Habsburg
empire. This was the state of things, as both powers saw it, at the
beginning of 1866, when the conflict over Schleswig-Holstein
entered a new phase.24
The Gastein convention already contained sufficient grounds for
dispute. The mere fact that Prussia had to pass through Austrian-
administered Holstein in order to reach Schleswig, its own
306
protectorate, caused great difficulties. The Austrian governor in
Holstein was manifestly uninterested in a good relationship with
Prussia; otherwise he would not have given the impression that he
approved of the pro-Augustenburg agitation in the south of the
Elbe duchies. The moment of crisis came at a demonstration in
Altona on 23 January 1866, which demanded the convention of an
assembly of all the estates in the duchies. Berlin protested, Vienna
refused to tolerate any Prussian interference in the administration
of Holstein, and the cooperation between the two powers came to
an end.
From that moment, politicians and military leaders in both
Prussia and Austria began to prepare for war. Prussia had good
reason to hope that Russia would not attack it from the rear in case
of armed conflict with Austria. The relationship between Vienna
and St Petersburg had not significantly improved since the Crimean
War, whereas the understanding between Russia and Prussia could
generally be considered good—especially since Bismarck, by means
of the ‘Alvensleben Convention’, had come to the aid of the tzar in
suppressing a new Polish rebellion at the beginning of 1863.
Likewise with the Bonapartist regime: Prussia could count on
France not supporting Austria as long as the latter held Venetia.
With the Kingdom of Italy, which sought to liberate this territory
from Habsburg control, Prussia concluded a secret offensive and
defensive alliance of three months’ duration on 8 April 1866. This
treaty violated the current statutes of the German Confederation, as
did the secret Franco-Austrian treaty of 12 June 1866, with which
Vienna assured itself of French neutrality in the event of war with
Prussia. This assurance came at a high price: if Austria was
307
victorious in Germany, Venetia was to be given up; in the event of
victory in Italy, the pre-war status quo was to be preserved. In
addition, Vienna agreed to the transformation of Prussia’s Rhenish
provinces into a new state, formally independent, but de facto
dependent on France.
The two German great powers were also active within Germany
itself. In mid-February Bismarck attempted to gain the support of
the Bavarian foreign minister, Ludwig von der Pfordten, for a
conjoint Prussian–Bavarian initiative in the Federal Diet calling for
the establishment of a German parliament to be elected by general,
equal, and direct suffrage, a voting system the Prussian prime
minister was convinced would actually strengthen the old powers.
‘Direct elections and general suffrage, however, I consider greater
guarantors of a conservative outlook than any artificial voting law
calculated to win ready-made majorities.’ Yet since the Bavarian
foreign minister wanted to preserve his country’s place at the head
of the ‘Trias’ as mediator between the two German great powers,
he rejected a common initiative.
The result was that Prussia was compelled to present its
constitutional reform motion alone to the Federal Diet, which it did
on 9 April 1866. Bismarck’s hope that his demand for a German
parliament and democratic suffrage would win public opinion for
Prussia went unfulfilled. Both within and without the Hohenzollern
state, the verdict was nearly unanimous: the Prussian government
ought first to look to the constitutional and parliamentary process
inside its own state before stepping onto the German stage with
progressive sounding language that, in the mouth of Otto von
Bismarck, made rather a caesarist, Bonapartist impression.
308
Even among the right wing of the Progressive Party, Bismarck’s
new course met with little approbation. To be sure, the National-
Zeitung began at the beginning of April to prepare its readership for
the possibility of war between Prussia and Austria. Such a war, it
wrote, must not perforce be called a ‘fratricidal war’ simply
because both states belong to the German Confederation. Prussia
and Austria only ever stood together in times of greatest peril, ‘in
conditions, under which two completely different and
heterogeneous peoples likewise tend to cooperate’. Thus it would
be a mistake to believe that Prussia ‘is bound with the holy bonds
of kinship and political community’ to the Austrian polity in its
present constitutional form. Nevertheless, the Berlin newspaper did
not trust the Bismarck ministry, at odds with the Prussian
parliament, to gain the sympathy of the rest of Germany, without
which no war could waged against Austria.
Bismarck’s constitutional reform proposal was commented on by
the National-Zeitung with a rhetorical question: ‘General and equal
suffrage, direct elections, a German parliament—these are, for us,
great and worthy goals. But what concern are they of Counts
Bismarck and Eulenburg? What concern are they of the Federal
Diet?’ Karl Twesten was of the same opinion. ‘Bismarck is not the
man who can rule with a true parliament,’ the liberal politician
announced on 17 April in a speech before the delegates and
primary electors of the first Berlin electoral district. ‘German
unification can only be brought about under Prussian leadership,
but this leadership must be in liberal hands.’
Notwithstanding the failure of its spring diplomatic offensive,
Prussia repeated the motion for Confederation reform on 10 June
309
1866. Yet despite the greater subtlety and more polished form of
the proposal, it found even less support this time. Immediately
before introducing the measure in the Federal Diet, Prussia had
itself flagrantly violated Confederation statutes. On 7 June Prussian
troops had entered Holstein. The occupation of the duchy was
Berlin’s response to a spectacular move by Vienna on 1 June: the
Austrian governor of Holstein was ordered to convene an assembly
of the estates, and Austria formally requested that the Federal Diet
decide the issue of the Elbe duchies’ future, since all efforts to come
to an understanding with Prussia had come to naught. This
manoeuvre breached the Gastein convention, yet for that very
reason was calculated to gain Austria a great deal of support in the
‘third Germany’.
On 9 June Austria countered the Prussian deployment in
Holstein with a petition to mobilize the federal army (excluding the
Prussian contingents). Although Prussia rejected the motion as
contrary to federal law, the vote nevertheless took place on 14
June. All the Mittelstaaten approved the motion, with one
exception: Baden first abstained, then gave in to public pressure
shortly thereafter and went over to the side of Austria. A few of the
smaller states sided with Prussia against the mobilization, including
Brunswick and the two grand duchies of Mecklenburg. Prussia itself
did not participate in the vote, but responded by declaring that the
decision rendered the Confederation accord null and void.
Since both Prussia and Austria had already commenced
mobilization in April, the military operations needed no further
preparation. On 16 June, after the expiration of ultimata that had
gone out to the governments in Dresden, Hanover, and Kassel two
310
days before, Prussian troops marched into Saxony, Hanover, and
Electoral Hesse. On 20 June Italy declared war on Austria. Prussia
followed suit the next day, immediately sending troops across the
border into Bohemia.25
Austrian liberals saw the war as an act of deliverance. ‘That for
which Austria is now to strike a blow’, wrote the Vienna Neue Freie
Presse on 19 June in a commentary to the imperial declaration of
war, ‘is the idea of liberty in the federation; that against which its
blows are aimed is detestable caesarism, the plague of our times,
poisoning all that is right and lawful.’ The right wing of Prussian
liberalism saw the situation in diametrically opposite terms. When
the war began, the National-Zeitung had no further recourse to its
earlier position that the domestic deficiencies of the Bismarck
ministry would prevent the successful prosecution of an Austrian
war. The north German population, the Berlin paper wrote on 30
June, would doubtless decide that the highest national purposes
were being defended in the Prussian ranks.
Even if the ministry currently at the helm is not one friendly to the
people, nonetheless, in her confrontation with Austria, Prussia
represents the cause of German liberty, just like the stubborn
Lutherans and Reformed represented and rescued the freedom of
the spirit during the Thirty Years War.
311
constitutional liberty. ‘In a war against Austria, a Prussian victory
surely means the victory of civil and religious liberty. It means the
preservation of northern Germany—in spiritual terms from the
Jesuits, in material terms from financial and economic ruin.’
Southern German Catholics and democrats naturally had a
different attitude. In Catholic upper Swabia, as Adolf Rapp wrote in
his 1910 study Die Württemberger und die nationale Frage,* the
people were galvanized by the feeling that their church, too, was
under attack in the war.
The Little Germany party had themselves given cause for this
reaction. How often could it be heard from them that a Protestant
Germany was the goal and that Austria must abide by the Peace of
Westphalia, by means of which it had explicitly excluded itself
from Germany. The Catholic clergy did not fail to point this out to
their communities.
312
the idea will once again make itself manifest in history.’
Things turned out much differently. On 3 July, the Prussians
inflicted a decisive defeat on the Austrian army at Königgrätz in
Bohemia. The allies of the Habsburg empire—those that had
intervened in the conflict (Baden was able to avoid doing so at the
last minute)—were defeated in short measure in a number of lesser
battles. Even prior to Königgrätz, Emperor Francis Joseph, despite
the fact that his troops were victorious in the Italian theatre, had
offered to cede Venetia to France and requested Napoleon III’s
arbitration with Italy, a step that immediately cost Austria the
sympathy of the ‘third Germany’ and encouraged the latter’s
readiness to cooperate with Prussia in mutual defence against the
expansionist intentions of France. On 5 July the French emperor
agreed to Francis Joseph’s request, but extended the mediation to
include Prussia.
Since Napoleon promised to accept an increase in Prussian
power north of the Main, Bismarck acquiesced. He was also
prepared, in the interest of the European balance of power (such as
Napoleon understood it), to accede to a number of French
demands, including the independence of a federation of southern
German states, the preservation of Austria’s territorial integrity
excluding Venetia, a referendum in northern Schleswig, and the
preservation of Saxony. France’s demand for the cession of left
Rhenish territory, however, met with Bismarck’s staunch rejection.
Instead, he directed Napoleon to Luxembourg, which was tied to
the Netherlands in personal union, and to Belgium. Yet for France
to go in this direction meant conflict with Britain.
The Prussian prime minister also had to skirmish with his own
313
king in the weeks following Königgrätz. William I wanted to
impose harsh peace terms on Austria. His leading minister, in
contrast, sought with mild terms to create the basis for future
cooperation. It was Bismarck who prevailed in the end. On 21 July
Prussia and Austria agreed to a provisional ceasefire of five days’
duration. On 26 July the ‘preliminary peace’ was signed in
Nikolsburg, becoming effective two days later. This agreement
accorded with the assurances the Prussian prime minister had
already given to Napoleon’s envoy, Vincent Benedetti. Bismarck
managed to deter the convention of a European peace conference,
such as Russia wanted. The final peace treaty could thus be
concluded with little delay, and was signed on 23 August in
Prague.
The Habsburg empire acquiesced in the dissolution of the
German Confederation and in the reshaping of Germany without
the participation of Austria. Vienna recognized the planned North
German Confederation, which included Prussia and fifteen northern
German states (seventeen after the two duchies of Mecklenburg
joined on 21 August). It also recognized the confederation of
southern German states, designed to have an independent
international existence, but permitted to have closer ties with the
North German Confederation. The southern alliance, however,
ultimately came to nothing. Article V of the peace treaty granted
the inhabitants of northern Schleswig the right to join Denmark,
provided that they expressed this wish in an open referendum.
The victorious state grew through annexation. Hanover,
Electoral Hesse, Nassau, and the Free City of Frankfurt, all of which
had stood with Austria in the war, were integrated into Prussia,
314
along with Schleswig-Holstein. Saxony became a member of the
North German Confederation, as did Hesse-Darmstadt, though only
with its territory situated north of the Main. With the southern
German states Prussia concluded not only the official peace
treaties, but also secret defensive alliances. These stipulated that, in
the case of war, the undersigned would place their armies under
Prussian supreme command and organize them according to the
Prussian model. Through these agreements, northern and southern
Germany became more closely linked than they had been before.
Nonetheless, the German war did not produce a German national
state, but rather a new kind of German dualism, now between
Greater Prussia on the one hand and southern Germany on the
other. Many contemporaries trusted in the finality of the border
along the Main, but most of them did not, though how and when it
would be set aside was uncertain. Only one thing could be ruled
out with near-total certainty: that the Germans would be able to
solve the German question, which was still very much open, with
as little outside intervention as the 1866 war had occasioned.26
The German war of 1866 was not a civil war. It was a war between
German states, fought by regular armies, not by guerrillas or
irregulars fighting in the streets. The Prussians and Austrians who
met at Königgrätz were fighting for king and country; they fought
against enemy soldiers, not against German brothers. The situation
of the soldiers from the Mittelstaaten who entered the war was very
315
little different. In the public opinion of the ‘third Germany’,
however, the feeling was widespread, among kleindeutsch and
grossdeutsch advocates alike, that the war was a ‘war between
brothers’ that might easily lead to the partition of Germany. In the
smaller and middle powers, state patriotism and German patriotism
blended together less problematically than in Prussia—not to
mention in Austria, which had been growing apart from Germany
for centuries.
Regardless of what politicians, publicists, and pastors wrote and
said, the conflict was also not a religious war. Rather, religion had
given rise to ideology, and to a far greater extent on the Protestant
and liberal side than among Catholics and conservatives. The
ideologization of religion was highly effective. When Prussian
liberals invoked the legacy of the Reformation, when they went so
far as to place the war with Austria in a ‘Schmalkaldic’ perspective
or in relation to the Thirty Years War, they were not making
arbitrary comparisons, but calculated arguments. They were
appealing to one of the fundamental existential and emotional
factors of Prussian identity, but which at the same time
transcended Prussia. The Evangelical ‘life-feeling’ embraced a
particular conception of Germany, one that excluded another, no
less particular conception—the Catholic idea of Germany, which
was oriented towards the old Reich and its succession in Austria.
The multinational Habsburg empire had been far less ‘German’
than the Hohenzollern state for a long time. When Count Belcredi,
the leader of the Vienna government since 1865, suspended the
February patent, the brief ‘German’ and liberal interlude of the
Schmerling era was at an end. Austria began to seek a compromise
316
with Hungary, leading in December 1867 to the establishment of
Austria-Hungary as an ‘imperial and royal’ (kaiserlich und königlich)
double monarchy. A compromise with Prussia, on the other hand,
had never stood on the agenda. As powerful as Prussia had become
after 1815, Austria refused to recognize its equal status until the
end of the German Confederation. However the course of German
history might have run if Vienna had not clung so adamantly to its
role as presidential power and agreed to an authentic reform of the
German Confederation, Austria’s policy vis-à-vis Prussia must be
considered one of the significant factors contributing to the war of
1866—indeed, one of its immediate causes. In Vienna no less than
in Berlin, there was a ‘war party’ in 1865–6 that clamoured for the
Gordian knot of German dualism to be cut through with the sword.
A German national state could not be formed with Austria.
Grossdeutschland would have meant the end of the Austrian empire.
Nor could a national state have arisen from any conceivable
renovation of the German Confederation. Consequently, the
German national movement was not to be satisfied with
Confederation reform. The nationalists wanted both unity and
liberty. Unity was only possible with Prussia—that was one of the
lessons of the German war. In the Mittelstaaten, liberty was possible
without German unity, but for the liberals in these countries, to
renounce the cause of unity was unthinkable, for both ideological
and material reasons. By definition, therefore, an arrangement with
Prussia was going to be necessary.
In Baden, a number of liberals had already come to this
conclusion before 1866. After Königgrätz, this group prevailed.
Under the new government, led by Karl Mathy, the grand duchy
317
consolidated its status as Prussia’s closest ally south of the Main. In
Württemberg moderate liberals, led by Julius Hölder, founded the
German Party (Deutsche Partei) in August 1866, thereby drawing a
clear line of separation between Prussia-friendly ‘realists’ and
democratic ‘particularists’. In Bavaria, the liberal Progressive Party
declared for the first time its unanimous support of the kleindeutsch
solution. In an announcement on 28 August, the Progressive
delegates of the second chamber demanded the establishment of a
close federation with Prussia, the political enlargement of the
German Customs Union, and Bavaria’s earliest possible entry into
the North German Confederation. Foreign intervention was, from
now on, to be met collectively. ‘Should a foreign power threaten
German territory, we demand that Bavaria immediately join the
leading northern German power for the purpose of common
defence under Prussian leadership.’
The man who took over leadership of the Bavarian government
at the end of 1866, Prince Chlodwig zu Hohenlohe-Schillingsfürst,
did not think much differently, although he expressed himself more
cautiously in public. ‘I look upon the present catastrophe with great
peace of mind,’ wrote the Prussia-friendly, liberal-conservative
Catholic on 13 July 1866, ten days after Königgrätz, in his journal.
318
misfortune for the dynasties, I admit, but a boon for the German
people.27
319
Only with this step have the Middle Ages and the feudalism of our
nation been overcome and left behind, finally and completely. By
separating ourselves from the House of Habsburg, which could not
divest itself of the ideas and pretensions of the Roman-German
empire—only through this separation are we now able to become
an independent nation and establish a German national state. We
can be more German than our ancestors were permitted to be.
such as has not been seen in a thousand years. The new way has
320
utterly triumphed over the old. But the old way does not date
merely from 1815; it goes all the way back to Charlemagne. The
idea of the Reich has fallen and been buried. If the German people
are ever again united in an empire, then it will be an empire that
can no longer look back upon a thousand-year history, but only a
history of three hundred years.
321
at hand necessitated conscious reflection upon the ‘force of reality’
and the ‘disconcerting world situation’. ‘We need a swift resolution
to the German question, and the only way that seems to present
itself at the moment is [for the southern states] to join the northern
federation and to cultivate a close alliance with Austria.’ For it
must not be forgotten that the coming ‘new federation’, Germany
unified under Prussian leadership,
only forms one, albeit the largest part of Germany, and that
another large part belongs to Austria. These two parts of the one
German nation must not look upon one other as strangers or
entertain merely diplomatic relations with each other as foreign
peoples, but must found such a indissoluble alliance as two parts of
the same nation by right and necessity of nature ought to do.28
322
spoke out against the removal of the ruling dynasties in Hanover,
Electoral Hesse, and Nassau, against the annexations of their
territories, and against Bismarck’s policies as a whole. ‘Mere
violence cannot be the foundation of legitimate right’; ‘the planned
north German federal parliament means basically a victory for
democracy’; ‘the beneficiary of the “tearing apart and weakening of
Germany” is none other than Napoleon III himself.’ In the ensuing
period, Gerlach drew closer to political Catholicism. At the
beginning of 1873, he was elected to the Prussian House of
Representatives as a delegate for the Catholic Centre, albeit
without converting to Catholicism.
Finally, 1866 also meant a defeat for the democrats. In Prussia
this was also the case for the representatives of the bourgeois left
who approved of the annexations, like Waldeck and now also
Schulze-Delitzsch. They remained opponents of Bismarck and thus
in the same ‘camp’ as Franz Duncker, the publisher of the
Volkszeitung, and Johann Jacoby, who, in the opening debate of the
newly elected House of Representatives on 23 August 1866,
declared unity without freedom to be the ‘unity of slaves’, and
compared Bismarck to Napoleon III. ‘Only in the service of right
and of liberty may the banner of the national principle be raised. In
the hands of a Louis Napoleon or one of his ilk, it only serves to
confound and destroy the peoples.’
The political sentiments of the Württemberg democrats were
given eloquent expression by the delegate August Österlen at a
public gathering in Schwäbisch Hall on 9 September 1866. ‘The
victory of violence’, declared the lawyer from Stuttgart, ‘makes a
deplorably strong impression on those whose character is less firm.
323
We, however, will not give up our principles, which are as eternal
as the stars. The power that has cast down thrones will not be able
to cast down our principles.’
This statement would also have been endorsed by August Bebel
and his friend Wilhelm Liebknecht in Leipzig—who were not,
however, in agreement with the particularism of the Württemberg
People’s Party, which had thus far stood in the way of a truly
national organization of German democrats. On 19 August 1866
Bebel and Liebknecht inaugurated in Chemnitz a new democratic
party, summarizing its national aspirations in the following words:
‘The unity of Germany in a democratic form of government. No
inherited central authority, no Little Germany under Prussian
hegemony, no Prussia expanded through annexations, no Greater
Germany under the aegis of Austria.’
This was nothing other that the resurrection of the anti-dynastic
pan-Germanism of the 1848 republican left. The Saxon People’s
Party consciously oriented itself toward the democratic middle and
working classes, thus renouncing a ‘proletarian’ image. This
distinguished it in two ways from the General German Workers’
Association, which had founded a new party in its own turn, the
Social Democratic Party of Germany (Sozialdemokratische Partei
Deutschlands), at the end of December 1866. The followers of
Lassalle, led after 1867 by the Frankfurt lawyer and writer Johann
Baptist von Schweitzer, were, on the one hand, firmly kleindeutsch
in outlook. On the other hand, they consciously sought to organize
only the working classes, distancing themselves from all varieties of
bourgeois liberalism.29
324
Marx and Engels had been no less grossdeutsch and anti-Prussian
during the German war than Bebel and Liebknecht. After
Königgrätz, however, they immediately accommodated to the new
reality. The most important thing for both was that the proletarian
revolution could profit from the suppression of German
particularism. ‘In my view, we simply have no other choice’, wrote
Engels from Manchester on 25 July 1866 to Marx in London, ‘than
to accept the fact, without countenancing it, and then to exploit as
well as we can the better opportunities that must now present
themselves for the national organization and unification of the
German proletariat.’
The tactical manoeuvre with the principle of general suffrage in
April 1866 had already revealed Bismarck to be an able pupil of
Napoleon III. For Engels, this was a confirmation of his argument
that Bonapartism was ‘the true religion of the modern bourgeoisie’.
325
Nearly two decades later, Engels would go so far as to ascribe
revolutionary significance to Bismarck’s 1866 policies. As he wrote
to August Bebel in November 1884, 1866
326
Prussian liberals saw no cause for such pessimism. When they
spoke of ‘revolution’ in the autumn of 1866, they meant the word
in a positive sense, appealing to the revolutionary legitimacy of the
‘nation’ against two competing notions: the dynastic legitimacy of
the rulers Prussia had deposed, and the democratic legitimacy of
the annexed populations. In the 20 December debate over the
annexation of Schleswig-Holstein, Karl Twesten declared to the
House of Representatives that although the ‘present revolutionary
course of Prussia’ violates old positive law, nonetheless ‘radical
changes in history simply do not conform to the paths of legality.
That which is rational is not brought about through reason ...
Prussia represents the victorious whole, to which the individual
parts must subordinate themselves.’ The events of 1866 had
changed Twesten from the self-confessed apostle of August Comte
of the late 1850s into a Hegelian. The ‘cunning of Reason’ was, in
any case, not a positivist, but rather a dialectical hypothesis.
On 1 December 1866 Hans Viktor von Unruh—originally a left-
centre, later right-centre delegate to the 1848 Prussian National
Assembly, in 1859 one of the founders of the German National
Union and in 1861 of the German Progressive Party—spoke of the
Prussian government’s dualism between conservative domestic and
revolutionary foreign policy agendas, embodied in ministers
Eulenburg and Bismarck, respectively. ‘I find the external policy of
the Herr Prime Minister, in contrast to the internal policy, to have
the same idea of state, the same guiding principle as the policy of
Frederick the Great—the goal of creating an independent,
powerful, enduring state inside Germany.’ Frederick’s policies,
according to Unruh, had not been conservative, but decidedly
327
revolutionary in character; they had not invoked historical law, but
rather the historical requirements of the era. In this sense, too, the
policies of Count Bismarck are
The year 1866 was a ‘revolution from above’. It was ‘the great
German revolution’ the Swiss historian Jacob Burckhardt spoke
about shortly thereafter, a radical overturning of the political order
in Germany with the help of the military might of Prussia, whose
success assured the northern power the military hegemony in
Germany and transformed Austria into non-German power. Thus
1866 was the answer to the Revolution of 1848—a failed
revolution, measured by its own goals, for the ‘mad year’ had
brought neither unity nor liberty. The German war, on the other
hand, by eliminating the grossdeutsch solution to the German
question and thus clearing away one of the major obstacles to the
kleindeutsch solution, had brought Germany at least a considerable
step closer to unity.
In bringing the issue to a forceful decision, the Prussian state
possessed the power the Paulskirche had never had. What Bismarck
achieved in 1866 answered a need of the educated and propertied
bourgeoisie, at least those who considered themselves liberal: the
desire to advance along the road to a German national state that
328
could hold its own, economically and politically, with the older
national states of the west. In principle, this desire was also in
accord with the aspirations of the young workers’ movement. It
was, in fact, an urge that permeated all of society, rendered
especially acute by the ongoing process of industrialization. Any
government that did not respond to this situation ran the risk of
precipitating a new period of revolutionary turmoil. Because he
knew this, Bismarck became a ‘revolutionary’. Yet he saw in the
‘revolution from above’ no break with the Prussian tradition, but
rather its very essence: ‘In Prussia, only the kings make
revolutions,’ he remarked to Napoleon when the latter warned him
of the dangers of revolution during the Prussian constitutional
conflict.
The achievements of the battlefield and the negotiating table in
1866 did not contradict the interests of the old Prussian ruling
classes from which Bismarck came. These interests, too, found
satisfaction in the war’s outcome. It is true that the liberals wanted
more than the Prague treaty actually spelled out. They wanted to
overcome the division between northern and southern Germany,
and they wanted to increase their political influence. As far as the
first goal was concerned, the liberals could be sure that it
corresponded to Bismarck’s own ideas. The outcome of the second
aspiration depended on the issue of the constitutional conflict. This
was the question that demanded resolution in Prussia as soon as
the German war had come to an end.30
329
CONSTITUTIONAL CONFLICT
330
The request for indemnity was not an admission of culpability.
Far from it: both the monarch and the ministry continued to insist
upon the absolute necessity of the non-budgetary regime. The only
thing the government was concerned about was obtaining the
parliament’s blessing for the facts it had created. Without such a
sanction, a return to constitutional normality was inconceivable,
not to mention the cooperation of the moderate factions of the
liberal party Bismarck required for his German agenda.31
In terms of constitutional legality, the position of the Bismarck
government was just as untenable in August 1866 as it had been in
October 1862. The king and his ministers were only permitted to
invoke a supra-legal obligation to preserve the essential functions
of the state in cases where the parliament was either non-
functional or resolved upon an obstructionist course. Neither was
the case here. The Prussian representative assembly was simply
availing itself of its classic right to approve the yearly budget. It
was also within its constitutional rights to demand, as it did on 22
May 1863, a new government from the king, an agreement with
the old administration having proved impossible. The appeal to the
monarch to reshuffle the ministry was a means of preserving
parliamentary rights when a fundamental conflict arose between
the legislative and the executive bodies. If he did not wish to
accede to such an appeal, the monarch in turn had the right to
dissolve parliament and set new elections, thus appealing directly
to the people (as William I had done in spring 1862 and again in
May 1863). Since the people were in support of the parliamentary
majority at the time, the king would have been compelled to come
to an understanding with the latter. To rule without an approved
331
budget and against the representative body was not permitted in
any constitutional system, not even in Prussia.
Even in the case of a true state of emergency, however, the costs
of an army reform à la Roon would not have been covered under
the category ‘absolutely necessary expenditures’. Bismarck himself,
as he openly admitted to Twesten at the beginning of October
1862, considered the three-year term of military service, the main
point of the reform conflict, to be completely dispensable, in
contrast to the king’s view. The prime minister’s ‘gap theory’ was
intended to justify what could not lawfully be justified, the break
with the constitution. In legal terms, the demand he placed before
the House of Representatives after the victory over Austria was a
demand for the squaring of the circle: a retrospective enabling act
that was annulled the moment it passed.32
Thus it seemed only logical when, in the parliamentary debate
over the indemnity measure at the beginning of September,
Benedikt Waldeck declared that a positive vote was tantamount to
a ‘disavowal of everything we have fought for’. Most of the
Progressive left argued similarly. The right wing of ‘resolute
liberalism’ came to a different conclusion. Several of its best-known
speakers, including Unruh, Michaelis, and Twesten, had left the
fraction of the Progressive Party in April out of protest against the
intransigence of the democrats. They now pressed for the approval
of the government’s indemnity bill, claiming it represented the only
way liberalism would be able to shape the future of Prussia and
Germany according to its own ideas.
The jurist Eduard Lasker, first elected to the House of
332
Representatives in 1865 in a by-election, also left the Progressive
Party shortly after the indemnity debate, claiming ‘inner reasons’
had convinced him that ‘the public consciousness of justice is
satisfied under the current conditions, despite the breach of the
constitution.’ The most important reason for Lasker was that the
unification of northern Germany represented Prussia’s first chance
to relieve its overstretched military finances. Michaelis exhorted
the representatives ‘to act as a factor of this state in the spirit of
this state’ and to make the decisions ‘that bring about the results we
desire’. Twesten, finally, declared that the representatives were not
to ‘renounce the development of liberty; but the development of
the power of our fatherland, the unification of Germany, that is the
true, the highest foundation we can create for the development of
liberty, and we can now have a hand in this great work.’ It was
true that the Bismarck administration had, in past years, committed
grave sins against the law and the people’s consciousness of justice.
‘But the history of the past year has granted him indemnity. Let us
pronounce it!’
The majority followed these appeals. The government’s bill was
approved with 230 votes against 75. Most conservatives and
moderate liberals voted positively. The negative votes came from
the ranks of the Progressive left wing, the Left Centre, and from a
few old-conservatives.
The delegates of the right wing of the Progressive Party had no
intention of capitulating to Bismarck. They also refused to
participate in the ‘idolatry of success’ that Rudolf Virchow, the
famous physician and delegate of the party left wing, had warned
against in the 23 August debate in which the assembly considered
333
its response to the king’s inaugural speech. These men made no
sudden political shift in the late summer and autumn of 1866. They
simply attempted to come to terms with a radically altered
situation, drawing logical conclusions from the differences of
opinion on the national question, differences that had been driving
the Progressive Party to the verge of rupture since 1864. By
promoting conciliation on the domestic front, they hoped to
achieve more for the cause of civil liberty and the consolidation of
parliamentary rights than had been accomplished by the stubborn
protests of the conflict era. And they remained deeply convinced
that national unification would redound to the benefit of
liberalism. ‘It is completely certain’, wrote the National-Zeitung on
19 August 1866 in an article on the eightieth anniversary of
Frederick the Great’s death, ‘that the Prussian government must act
in accordance with liberal principles, for its daily labour, the
unification of Germany, is per se a liberal and progressive goal. But
the liberal party, too, must dedicate itself to ideas that build up the
state, or else it will be completely left behind.’
Such ideas were not eccentric. Ever since 3 July 1866, the
chances that Prussia could transform itself single-handedly into a
liberal polity had grown even smaller than they had been before.
The politicians and publicists from the Progressive right were quite
justified in complaining about the unfair distribution of German
military burdens. The smaller German states had been living at
Prussia’s expense, militarily speaking. Thus it made a certain sense
for Prussian liberals to imagine that Germany’s unification would
help them in their battle against the old Prussia of the Junkers, at
least in the long term.
334
It also made sense to assume that the project of German
unification was still a question of power. Germany, as Lasker
pointed out in the indemnity debate, was not in the fortunate
position of England, which had been able to attain and consolidate
its liberty primarily because its insular geography protected it
against outside invasion.
‘The liberals must not again cast doubts upon the power of the
state!’ Twesten’s declaration before an assembly of the liberal
Berlin district association in November 1866 was an expression of
liberal self-critique. Outside of Prussia, this self-criticism sometimes
went even further. In spring 1866 the historian Hermann
Baumgarten, a native of Lower Saxony living and teaching in
Karlsruhe, wrote an essay entitled ‘Der deutsche Liberalismus. Eine
Selbstkritik’,* published in the Preussische Jahrbücher at the end of
the year. Baumgarten did not reproach the liberals for not being
resolute enough in their opposition to Bismarck. Rather, their
mistake had been to shun an understanding with the government
335
and, over the course of years, to employ the ‘strongest language’,
despite it being perfectly clear that the ‘general conditions’ of the
Prussian state precluded a violent resolution to the conflict. ‘A
people getting richer by the day does not make revolution.’
This argument was not easy to refute. But Baumgarten did not
leave it at that. He more or less denied the ability of the
bourgeoisie to engage in effective political action. ‘The bourgeois is
created to work, not to rule, and the fundamental task of the
statesman is to rule.’ For the nobility, on the other hand, the author
saw a great future.
336
the later National Liberal Party, they asserted their full support for
the government’s foreign policy, justifying this course with their
‘confidence’ that the efforts of the Bismarck ministry were directed
at German unity. On the domestic front, however, they still did not
consider fully secure the ‘new direction’ that would allow them to
go along with the government’s measures with trust. For that
reason Twesten, Michaelis, Unruh, Lasker, and twenty further
signatories declared it their task to be a ‘vigilant and loyal
opposition’. They summarized their domestic political demands in
the following statement:
337
budgetary authority.
The Prussian National Liberals remained firmly committed to
this programme, despite knowing that Bismarck, at least in the near
future, would not accommodate them as much as they wished.
Therefore, the goal of de facto parliamentary governance was
postponed until the unification of Germany was achieved, at which
time the constitutional demands of the liberals would be easier to
fulfil than at the present. Such expectations were given classic and
pithy expression by Ludwig Bamberger from Mainz, who had been
a member of the German National Assembly for a few months in
1849, thereafter a participant in the rebellion in the Palatinate and
an exile in France, where he had remained until 1866. In an appeal
to the voters in Rhenish Hesse during the December 1866
parliamentary elections in the grand duchy of Hesse-Darmstadt,
Bamberger asked the rhetorical question: ‘Is not unity itself a part
of liberty?’
Bamberger gave voice to a hope, one as easy or as difficult to
justify as the fear expressed by Friedrich Harkort during the
annexation debate in the Prussian House of Representatives on 7
September 1866: ‘What the sword has gained only the sword can
maintain, and as we were before, so we shall have to remain armed
against east and west. Thus it seems that the hoped-for relief in the
military burden will not happen, at least for a long time.’ In 1866 it
was generally expected that the unification of Germany and Italy
would liberate and lastingly preserve Europe from the threat of
Bonapartist France. But was it not rather presumptuous to add, as
Lasker did, that the nationality principle would also help the
principle of humanity to victory?
338
In 1866 Bismarck seemed to many liberals—not merely to
Baumgarten—to be a kind of German Cavour, whom one could
expect, while acting in the best interests of Prussia, to stride
forward toward German unification just as the prime minister of
Sardinia-Piedmont had worked for the unity of Italy until his death
in 1861. National Liberals now eschewed the comparison to
Napoleon III, and for good reason. While Bismarck employed
‘Bonapartist’ political tactics like general, equal, and direct male
suffrage (to be introduced for the elections to the North German
Reichstag), he was nonetheless too much the loyal Brandenburg
Junker to become a plebiscitary dictator in the style of the French
emperor. On the other hand, however, the prime minister had also
distanced himself considerably from the expectations of the old
Prussian conservatives—so considerably, in fact, that he was now
regarded by some of them as an apostate. For the National Liberals,
of course, this could only be a further recommendation.
In the view of many conservatives, Bismarck’s indemnity bill
already represented an unacceptable concession to the liberals.
Indeed Bismarck, unlike William I, had not stated that he would
again act as he had done during the constitutional conflict if a
similar situation arose in the future. Although he certainly reserved
for himself the right to rule again without a parliament-approved
budget, he much preferred to come to an agreement with the
elements of the liberal bourgeoisie willing to make a compromise—
provided the monarchy could preserve its independence from
parliamentary control in the core areas of the military and foreign
policy.
Jacob Burckhardt called the ‘great German revolution of 1866’ a
339
‘curtailed crisis of the first order’. With respect to the resolution of
the question of hegemony between Prussia and Austria, this
judgement was accurate. The internal question of power in Prussia,
however, was not settled once and for all with the passing of the
indemnity bill. Since it did not address the crucial problem of
whether the monarchical or the parliamentary authority had the
priority in cases of budgetary conflict, the political accord of 1866
could only be, in the language of the legal and political theorist
Carl Schmitt, an ‘inauthentic’ peace, a ‘dilatory formal
compromise’. If the House of Representatives had rejected the bill,
the consequence would hardly have been the beginning of a
revolution from below. None of the ‘nay-sayers’ had anything of
the kind in mind. Rather, there would probably have been open
counter-revolution in the form of a coup d’état. The alternative to
both scenarios was the revolution from above that actually
occurred in 1866.
Bismarck emerged triumphant from the constitutional conflict.
Nonetheless, the approval of the indemnity bill was not a defeat for
liberalism. Compared with the era of conflict and reaction, liberals
had gained in political influence, at least the groups who had
agreed to cooperate with Bismarck. To what extent liberalism, thus
divided, would be able to make its mark on Prussian and German
politics— that was an open question in autumn 1866.33
340
The year 1866 saw not only a split within Prussian liberalism, but
also among Prussian conservatives. At the end of July 1866,
approximately twenty conservative delegates who—in contrast to
the majority of the Conservative Party— unconditionally supported
Bismarck’s policies founded the ‘Free Conservative Union’ (Freie
Konservative Vereinigung). In addition to high state officials,
diplomats, and scholars, the membership of the new group
included Rhenish industrialists and Silesian magnates, the latter not
infrequently combining large landownership with involvement in
heavy industry. However, in the elections to the constituent North
German Reichstag on 12 February 1867, in which voting took place
according to general, equal, and direct male suffrage, the Free
Conservatives did not come off as well as the Conservative party,
with 39 and 59 seats respectively.
The actual winners of the elections were the National Liberals,
who captured 79 seats, in comparison to the Progressive party,
which gained only 19. In its constituent session on 28 February, the
fraction gave itself the name ‘National Liberal Party’
(Nationalliberale Partei). The Prussian representatives around
Twesten, who had disassociated themselves from the Progressive
party after August 1866, were joined by like-minded liberals from
the other members states of the North German Confederation,
including Rudolf von Bennigsen from Hanover, the president of the
German National Union (which had disbanded in the autumn of
1867, since, as Bennigsen explained, other organs had assumed its
tasks, most notably the North German Reichstag). Together with
the Free Conservatives, the old-liberals and several ‘wild’ delegates,
that is, those belonging to no fraction, the National Liberals formed
341
a parliamentary majority united in the conviction that the most
urgent task at hand was the swift adoption of a constitution.
In fact, a constitutional draft already lay before the assembly
when it first convened in Berlin on 24 February 1867. This
document, formulated by the Prussian government according to
Bismarck’s guidelines and in accordance with the governments of
the northern German states, proposed a Prussian-led federal polity
with a presidency to be occupied by the Prussian king. In the
Federal Council (Bundesrat), the organ of sovereign federal
authority composed of plenipotentiaries bound by orders from the
individual state governments, Prussia—which contained five-sixths
of the population of the North German Confederation—had at its
disposal a mere 17 of 43 seats. Nonetheless, the possibility of a
two-thirds majority against Prussia was thereby excluded, with the
consequence that the constitution could not be amended without
the consent of the presidential power. The federal chancellor,
appointed by the president, was the sole minister and, in the
original version of the draft, would function only as the executive
of the Federal Council. A catalogue of basic rights was not part of
the text, since, according to Bismarck, it would be incompatible
with the federal character of the North German Confederation.
The representatives of the Progressive party around Waldeck
were highly dissatisfied with the federal approach. They desired a
unitary north German state, indistinguishable from a Greater
Prussia. Many National Liberals thought similarly, like the historian
Heinrich von Treitschke (who was not a member of parliament).
But the majority was prepared to acquiesce in Bismarck’s solution.
The office of chancellor, however, was—with Bismarck’s blessing—
342
accorded considerably more power than in the government’s draft.
Although Bennigsen, the floor leader of the National Liberals, did
not succeed with a motion to create a collegial federal ministry
answerable to parliament, the Reichstag did approve his motion to
subject the office of chancellor to parliamentary review. This did
not mean the introduction of a legally enforceable form of
accountability, much less a bona fide parliamentary system.
Bismarck would never have consented to such things. The federal
chancellor simply assumed political responsibility before
parliament, to whom he was to account for his actions. Yet it was
by means of this provision that, in the words of the constitutional
historian Ernst Rudolf Huber, ‘the federal executive was conferred
upon the federal chancellor; the Federal Council was reduced to a
cooperating and controlling executive organ.’
While the international representation of the Confederation, as
well as all decisions concerning war and peace, were the sole
province of the president, the Reichstag managed to make the
enforcement of international treaties dependent upon its approval,
thereby gaining a voice in at least part of foreign policy. It also
succeeded—following a motion of Karl Twesten—in making the
armed forces subject to the legislative authority.
The representatives were unable to obtain unrestricted
recognition of the parliament’s budgetary competence in military
matters, however. Bismarck wanted the federal army to have a
peacetime strength of 1 per cent of the 1867 population and a
yearly total sum of 225 thalers per soldier, to be adjusted every ten
years—in effect, a permanent budgetary enactment or Äternat. The
Reichstag rejected this proposal. Following a motion by Max von
343
Forckenbeck, it approved Bismarck’s peacetime military strength,
but restricted it to a period of four years, until 31 December 1871.
The resolution of the power question was thus postponed.
Under massive pressure from Bismarck, a majority of the
Reichstag, led by National Liberals and Free Conservatives, finally
voted through another compromise: the 1867 resolution was to
stay in effect after 1872 until the planned federal law went into
effect. While the government’s military spending was to be
constrained by the budgetary law, the legislature, too, was required
to observe the legally codified provisions of the army organization
when deciding the military budget.
The ‘dilatory formal compromise’ that settled the Prussian
constitutional conflict thus became the cornerstone of the
constitution of the North German Confederation. It was a
constitution that favoured the executive branch of government,
which secured far greater powers than those granted to the
imperial government by the imperial constitution of 1849. The
military command authority of the Prussian king, the exercise of
which required no countersignature in the ministry, represented a
portion of absolutism, both in theory and in practice, in the
German constitution of 1867. The North German Confederation
was a constitutional federal state, but it was constitutional and
federal only within the limits set by the foreign and domestic
successes of old Prussia in the year 1866.
In one respect, however, the North German Confederation was
actually far more democratic than the contemporaneous
parliamentary monarchies of Europe, even those of Britain and
Belgium. General, equal, direct, and secret suffrage gave German
344
males a lawful right to political participation far beyond that
granted to the citizens of states with census-based suffrage. It also
went further than the moderate liberals in the separate German
states and communities approved, though it would have been
politically inconceivable for liberals to oppose Bismarck’s
successful campaign for the suffrage law of 1849. Even as early as
1867, the effects of these democratic achievements proved to have
a good deal less effect on government than the Prussian prime
minister had expected. But there was no going back on general
suffrage, not even for Bismarck. Thus the antithesis between
democratic voting law and authoritarian government persisted—an
antithesis that was to outlast the short-lived federal state of 1867
by half a century.
On 16 April 1867 the constituent assembly of the Reichstag
approved the revised constitutional draft with 230 votes against 53.
The nay-sayers consisted of the Progressive party; the Saxon
particularists; the Hanover ‘Guelfs’, including Ludwig Windthorst,
former minister and later leader of the Catholic Centre party, as
well as two further prominent Catholics, the Berlin Obertribunalrat
Peter Reichensperger (originally from Koblenz) and the
Westphalian landowner and governmental official Hermann von
Mallinckrodt; August Bebel, the only delegate from the Saxon
People’s party; and the Polish members of the Reichstag. In
consequence of the favourable vote, Bismarck’s ultima ratio, the
forceful imposition of the constitution through the member
governments, proved unnecessary. It was ratified by the
parliaments of the member states in the ensuing weeks and went
into effect on 1 July 1867. Two weeks later, on 14 July, King
345
William, as president of the North German Confederation,
appointed the prime minister of Prussia, Count Bismarck, to the
office of federal chancellor.
The founding of the North German Confederation took place at
the same time as a new customs union agreement was being
concluded. The new treaty included provisions for a federal
customs council (Zollbundesrat) and customs parliament
(Zollparlament)—state-like institutions that seemed, like the
defensive alliances made public in March 1867, to anticipate the
federal integration of northern and southern Germany. However,
the protectionist reservations south of the Main were no less
serious than during the controversy over the Franco-Prussian trade
agreement several years previously, and the military alignment of
the southern states with Prussia did not make the latter any more
popular. In Bavaria, the first elections to the customs parliament in
February and March 1868 were carried by the candidates from the
Catholic-conservative Patriots’ party around Edmund Jörg. In
Württemberg, the victors were the democrats of the People’s party.
Two short years after the German war, there was little trace of a
broad movement for German unification. The trend seemed, rather,
to be going in the opposite direction, towards a hardening of the
border along the Main, at least in the view of many southern
Germans.
The division between north and south was so deep that
Bismarck could not yet risk an open attempt to settle the
outstanding issues in the balance of power among the states of
Germany. When Napoleon III, with Bismarck’s discreet
encouragement, took steps in spring 1867 to purchase the grand
346
duchy of Luxembourg—a member of the German Customs Union—
from King William III, popular protest was much stronger to the
north of the Main than to the south. In view of the anti-Prussian
sentiments in the chambers and among the populace, the
governments in Munich and Stuttgart could not afford to agree
with Bismarck’s view that a war between Prussia and France over
Luxembourg would necessitate the activation of their defensive
alliances with Prussia. Consequently, the Prussian prime minister
found himself compelled to agree to a compromise. At the London
conference over Luxembourg in May 1867, the Prussian envoy gave
his consent to a guarantee of the grand duchy’s independence and
neutrality by the five great powers, the Netherlands, Belgium, and
Italy.34
But resolute opposition to Bismarck’s policies was not confined
to the south. In none of the states annexed by Prussia were the
loyalty to the old dynasty and the antipathy to the new masters as
strong as in Hanover. Eight ‘Guelf’ delegates had sat in the new
federal parliament’s constituent assembly where, as members of the
adamantly federalist ‘Federal-Constitutional Association’
(Bundesstaatlich-konstitutionelle Vereinigung) along with the pro-
Augustenburg party from Schleswig-Holstein and the Catholic
Westphalian Hermann von Mallinckrodt, they had voted in April
1867 against the ratification of the constitution.
Much more effective was the opposition on the left side of the
political spectrum, which began in Saxony. In August 1869 August
Bebel and Wilhelm Liebknecht, the leaders of the Saxon People’s
party, joined together with a minority of Lassalle’s followers,
including Wilhelm Bracke, to found the Social Democratic Workers’
347
Party (Sozialdemokratische Arbeiterpartei) in Eisenach. Whereas the
Saxon People’s party had been half proletarian, half petty
bourgeois in membership, the new party, in the spirit of Lassalle,
was devoted exclusively to the concerns of the workers. At the
same time, the majority of the Union of German Workers’
Associations, led by Bebel and Liebknecht, began at the 1868
Nuremberg congress to work towards a rapprochement with the
Marx-inspired International Workers’ Association, the First
International, founded in 1864. Despite the many differences
between the ‘social’ and the ‘particularist’ wings of the People’s
parties, however, these developments did not yet represent a
rupture in the democratic movement along ‘class lines’, that is,
between the proletarian and the bourgeois democrats.
That split took place a year later, when the new Social
Democratic party adopted the Basel resolutions of the International
from September 1869, wherein the party of Marx and Engels
demanded the immediate abolition of private property and the
radical restructuring of society. From this moment forward, the
‘proletarian’ Social Democrats and the ‘bourgeois’ democrats parted
company. This rift, which was a permanent one, represents one of
the most radical caesuras in the history of German political parties.
In their antipathy towards Prussia, the southern German
democrats were every bit as uncompromising as Bebel and
Liebknecht, and the more the governments in Stuttgart and
Karlsruhe adapted themselves to the northern German hegemon,
the fiercer the opposition of the People’s party became. The
differences between the two countries were nonetheless great. In
Württemberg, the July 1868 parliamentary elections had been
348
carried by the democrats and the grossdeutsch partisans. The new
parliament confronted a cautiously manoeuvring ministry, the
Varnbüler government, which took pains to prevent its de facto
dependence on Prussia from becoming all too evident. The Baden
parliament, in comparison, was dominated by moderate liberals,
and was just as interested as the government—led by Julius Jolly,
the interior minister, after Karl Mathy’s death in 1868—in making
the grand duchy a member of the North German Confederation as
quickly as possible.
The situation in Bavaria, the largest of the southern German
states, was different again. Like Baden, the moderate liberal
government under Hohenlohe-Schillingsfürst pursued
uncompromisingly anticlerical policies in the schools and a liberal
course in economic and judicial matters. Freedom of trade was
introduced at the beginning of 1868, followed by liberal municipal
ordinances a year later. But the educational reform was
successfully opposed by the Catholic protest movement. In two
parliamentary elections in 1869, the conservative Bavarian
Patriots’ party under Edmund Jörg achieved resounding victories.
In the second of these, in November, it managed to secure the
absolute majority of seats. In January and February 1870, both
chambers of parliament voted no confidence in the Hohenlohe
administration, and the prime minister stepped down at the
beginning of March, against the will of Bismarck. Hohenlohe’s
successor, Count Bray, hitherto the Bavarian envoy in Vienna,
made an effort to compromise with the parliamentary majority.
This was a serious setback for the chancellor of the North German
Confederation.35
349
After the election of the new Confederation’s first ‘regular’
Reichstag in August 1867, the chancellor and the National Liberals
developed, all in all, a fairly close cooperative relationship in
matters of domestic policy, far closer than even the optimists had
dared to imagine in autumn 1866. Along with the Progressive and
Conservative parties, the National Liberals had improved upon
their results in the elections of February 1867. They were now
more successful in the legislative process than they had been with
the constitution. The 1869 law mandating the equality of religious
confessions in civil and citizenship affairs, which concluded the
legal emancipation of the Jews, bore the mark of liberalism,
although it only proved practically necessary in Mecklenburg,
where legal discrimination against the Jews was still in effect. The
new trade ordinances, also passed in 1869, expanded commercial
and mercantile freedoms, though pressure from the Conservatives
kept them from allowing agricultural workers and manorial
servants the same freedom of association they granted to other
workers. The penal code of 1870, on the other hand, could not be
counted among the liberals’ successes, since they failed to obtain
their most important objective, the abolition of the death penalty.
The combined strength of the member governments and the
Conservatives proved too great on this issue. Still, when one added
the 1867 law concerning the free movement of labour and further
laws for the standardization of the economy, including the 1868
ordinance on weights and measures, the National Liberals had
every reason to be satisfied, on the whole, with the new federal
parliament’s year of legislative achievement. The interplay between
the Reichstag and the ministerial bureaucracy had led to the
350
development of a ‘portfolio liberalism’ (Ressortliberalismus) going
far beyond the parliament’s original authority in trade policy.
In their most important goal, however, the National Liberals had
made no progress since the election of the customs parliament. The
unification of Germany had flagged and was now in a phase of
stagnation. The liberals began to doubt whether Bismarck was even
serious about the issue and, consequently, whether further liberal
cooperation with him was justified. ‘If things continue like this,
then the task of the national party can no longer be only to stand
idly at the Herr Chancellor’s side,’ wrote the National-Zeitung on 22
February 1870. Two days later, on the occasion of the third reading
of the jurisdiction treaty between the North German Confederation
and Baden, the delegate Eduard Lasker presented a petition
expressing the disappointment and impatience of the National
Liberals with a hitherto unusual candour. The Reichstag, Lasker
declared, should thank Baden for its ‘unflagging national
aspirations’ and recognize as the state’s main goal the ‘most rapid
possible accession’ to the North German Confederation.
Bismarck was outraged, and for more than one reason. The
parliament’s interference in matters of foreign policy was bad
enough in itself. But he also suspected the Baden government to be
behind Lasker’s initiative, and he declared this openly in the
assembly. From a practical point of view, he thought it a mistake to
accept into the federation the southern German state most narrowly
associated with the national idea, since the north–south unification
of Germany would thereby be made more difficult as a whole. In
any case, any southern German key to German unity was not, for
the federal chancellor, to be found in Karlsruhe, but in Munich.
351
It was clear to all at the time, without it having to be spoken
openly, that the integration of Bavaria into the North German
Confederation would lead to war with France. In Bismarck’s view,
the accession of a single state could not justify such a war. If war
with France for the unification of Germany was going to be
unavoidable, then all of Germany, both north and south of the
Main, would have to share the conviction that it would be a just
and necessary war. That conviction did not yet exist in February
1870.
‘The policy of the North German Confederation does not require
expansion. The situation in the north has not yet been sufficiently
consolidated to allow for the increase of fractious elements with
impunity.’ Such were Bismarck’s words to the Württemberg envoy,
Baron Spitzemberg, in a confidential conversation on 27 March
1870. This was certainly no pretext. In order to contain the dangers
emanating from an ‘increase of fractious elements’, whether
Württemberg democrats or Bavarian Catholics, a national
movement of considerable strength was going to be necessary. In
spring 1870 nothing of the kind was in evidence. But even as he
spoke to Spitzemberg, Bismarck might have suspected that the
situation could change rapidly.36
352
news of the Prussian victory at Königgrätz. For militant Protestants
and liberals an occasion for triumph, the defeat of Austria was a
catastrophe for the curia. For the Protestantische Kirchenzeitung, a
newspaper closely associated with the liberal German Protestant
Association, the battle of 3 July 1866 had ‘finally brought the
Thirty Years War to an end’ and ‘destroyed ultramontanism [from
ultra montes, ‘beyond the mountains’ (H.A.W.)] once and for all in
Germany. For not only have Austria’s power and ambitions been
expelled from the land; the papacy, too, has therewith lost its
single remaining earthly support in Europe.’
In truth, Austria’s defeat was in more than one way a defeat for
the papacy. The ‘world’ that crumbled in the summer of 1866 was,
in the words of the historian Georg Franz,
353
1866 (after first being ceded to France), there was no longer any
doubt that the liberal Italian national movement would exert all its
effort to achieve the last great goal of the Risorgimento, the
annexation of the Papal States and the relocation of the kingdom’s
capital from Florence to Rome. The pope’s days of worldly rule
were numbered.
Pope Pius IX and his advisers had no intention of submitting to
the will of the liberals. In 1864, in Syllabus errorem, an addendum
to the encyclical Quanta cura, the pope had castigated as ‘modern’
errors not only pantheism and rationalism, socialism and
communism, but also classic liberal ideas like state authority in
education, separation of Church and State, and the legal equality of
Catholics and non-Catholics, demanding the unconditional
subordination of the state and the sciences under the authority of
the Catholic Church. Six years later, the ecclesiastic declaration of
war against the modern world was even more radically formulated.
The First Vatican Council, which convened in December 1869 and
ended in July 1870, promulgated the doctrine of ‘infallibility’,
according to which the pope, when speaking ex cathedra upon
matters of church teaching, could not err. In both Germany and
Austria, most bishops (including Ketteler) were opponents of the
new doctrine, but that had no effect upon the final vote on 18 July
1870. The critics submitted to ecclesiastic discipline, thus
inadvertently confirming the impression of the absolutist character
of the Roman Catholic Church.
The Catholic opposition lost ground outside of the Council, too.
Stubborn adversaries like the Munich theologian Ignaz von
Döllinger, a former delegate to the German National Assembly,
354
were excommunicated, while only a small majority of firmly liberal
‘old-Catholics’ (Altkatholiken) openly broke with Rome on their own
initiative. The governments of several primarily Catholic countries
reacted to the new doctrine of infallibility even more vehemently
than many Protestant states. Austria repealed the Concordat of
1855; Bavaria hindered the proclamation of the resolutions by
withholding official government sanction; Baden, which had come
into conflict with the church over school policy even before
Bavaria, rejected their legal validity. Prussia and the North German
Confederation, for their part, reacted with marked reservation. In
March 1870 Bismarck had informed the Prussian ambassador to the
Vatican, Harry von Arnim-Suckow, that Prussia considered the
doctrine of infallibility primarily a Catholic matter. On 16 July he
instructed Arnim to eschew any demonstrative protest.
There was good reason for such cautious manoeuvring. On 19
July 1870, the day after Pius IX proclaimed the new papal doctrine,
the Franco-Prussian War broke out—for Bismarck everything but
unexpectedly. Solicitude for the feelings of German Catholics was,
in such a situation, a simple imperative of Prussian self-interest.
The German liberals, who had always been among the most
relentless opponents of the Catholic Church and who now, in their
turn, felt the massive challenge of the clerical counter-offensive,
were forced as of 19 July 1870 to turn their undivided attention to
the war with the western neighbor. The historical confrontation
between the avant-garde of secular modernism and the guardians
of ecclesiastic tradition—the great Kulturkampf of which Rudolf
Virchow spoke before the Prussian House of Representatives on 17
January 1873—experienced only an interruption. It was already
355
clear in the summer of 1870 that this conflict would be the ruling
theme of the post-war period, and not only in Germany.37
356
motion once again. If Paris acted forcefully in the matter, a wave of
patriotic feeling was sure to sweep not only Prussia, but all of
Germany. Such a movement was the very thing to weaken not only
particularist sentiment in Bavaria and Württemberg, but also the
parliamentarism of the North German Confederation. In 1871 the
provisional settlement of the military budget, concluded three
years before, was set to expire. The National Liberals made no
secret of the fact that they would not consent to a further
provisorium, but were going to insist upon the yearly approval of
the army budget, as well as the strengthening of parliamentary
rights as a whole. If the conflict with France came to a head, then
there was every reason to believe that the forces sympathetic to the
government would once again gain the upper hand in the strongest
party. Thus, in the event of war, Bismarck stood a good chance of
being able to eliminate not only the French opposition to German
unification, but also, within Germany itself, the particularist and
parliamentary elements standing in the way of his German solution.
In the summer of 1892, two years after his dismissal, Bismarck
declared openly that the French war had been ‘necessary’. ‘Without
having defeated France, we were never going to be able to
establish a German Reich in the middle of Europe and elevate it to
the power it now possesses.’ Armed with this insight, the
chancellor of the North German Confederation refused to accept
Leopold’s rejection of the Spanish offer in April 1870, doing
everything in his power to change the prince’s mind. By June he
had succeeded, and King William, too, finally gave his consent to
Leopold’s decision, at which point—on 2 July—the Spanish prime
minister informed the French ambassador of the Hohenzoller’s
357
intention. The response from Paris was not long in coming. On 6
July Foreign Minister Gramont, speaking before the chamber,
threatened Prussia with war. France, he declared, would not suffer
‘a foreign power, by placing one of its princes on the throne of
Charles V, to disrupt the present balance of the powers of Europe
to our detriment and so to endanger the interests and the honour of
France.’
On Gramont’s orders, the French ambassador to the North
German Confederation, Count Benedetti, twice had an audience (on
9 and 11 July) with Wilhelm I, who was taking the waters at Bad
Ems. Pressured by the Prussian king, Leopold formally renounced
his candidacy for the Spanish throne on 12 July. But the French
government was not satisfied, insisting upon William’s official
endorsement of Leopold’s declaration and his promise to accept no
future candidacy on the part of his relative. When Benedetti
broached these terms with Wilhelm on the promenade in Bad Ems
on 13 July, the king rejected them out of hand. Bismarck, upon
receiving the communiqué on this interview from Heinrich Abeken
of the foreign office, radically shortened and edited the text in such
a way as to make the offensive character of the French diplomacy
all too clear. This was the form in which the report reached the
courts and the public of Germany. As the leader of Prussia had
intended, the ‘Ems dispatch’ provoked a national outcry. Now,
France was subject to the very humiliation it had thought to visit
upon Germany. In this situation, Napoleon III had no choice but to
do what Bismarck expected him to do. On 19 July, France declared
war on Prussia.
With the coup of the Ems dispatch, Bismarck managed to
358
outmanoeuvre Wilhelm I, disregarding all the constitutional
stipulations making the Prussian king the bearer of the highest
executive power of the North German Confederation. The war,
which the dispatch made inevitable, was Bismarck’s war. The
Prussian prime minister and federal chancellor of the North
German Confederation had wanted this war, for it offered the
unique opportunity to annul, with the support of all of Germany,
the French interdiction on German unity.
Nonetheless, it would be wrong to ascribe the responsibility for
the war entirely to Bismarck, Prussia, or Germany. As in
Copenhagen and Vienna in 1864 and 1866, there was a war party
on the other side, too, in Paris of 1870. Whereas France could only
justify its rejection of German unification in terms of power
politics, Germany could defend its aspiration to unity by appealing
to the fundamental principle of national self-determination—an
incipiently democratic principle, and one continually invoked by
Napoleon himself, though it had never occurred to the emperor of
the French that Germany, too, possessed such a right. By 1870, in
any case, Napoleon’s own authority inside France was already too
far weakened to permit him any room for a compromise with
Bismarck. ‘Revanche pour Sadowa!’ was a popular rallying cry in
France (Sadowa was the name for the battle of Königgrätz used on
the left side of the Rhine), and the policy to which it corresponded
set its sights on repairing the losses of 1866 and putting Prussia in
its place. Foreign Minister Gramont, who had been ambassador to
Vienna until May 1870, was a proponent of this broad current. To
employ foreign policy as a tool to deflect internal unrest had
always been a preferred tactic of Bonapartist rule. Thus the war of
359
1870–1 was also Napoleon III’s war—his last, as it soon turned
out.38
The German north was veritably unanimous in the conviction
that the war was in defence of a just cause. In the federal
Reichstag, the speech of the king and Bismarck’s announcement of
the declaration of war on 19 July were greeted with storms of
applause from members of all parties. Even the followers of
Lassalle, as well as one Social Democratic delegate, approved the
war credits. Bebel and Liebknecht abstained, believing the war to
be a dynastic conflict. In this they found themselves in conflict with
their German colleagues from the International Workers’
Association, who by this point had come to demand the annexation
of Alsace and Lorraine. Marx and Engels, living in England, did not
go this far. Yet even they had cause to go over to the side of Prussia
and Germany, since the war opened up the possibility of bringing
about the downfall of the hated Bonapartist regime. It was true,
wrote Marx in the ‘Erste Adresse des Generalrats’ of the
International, that Prussia had first put Germany in the position of
having to defend itself against Louis Napoleon. But that did not
change the decisive fact: ‘On the German side, the war is a war of
self-defence.’
South of the Main, the mood changed within a few days of the
incidents at Bad Ems. On 8 July the Reutlinger Neue Bürgerzeitung, a
democratic paper, had written that ‘the Frenchman’ could not and
would not tolerate ‘being surrounded by the Prussian satellites of
Muscovy; in this matter, the French people see eye to eye with
their government, and rightly so.’ After Napoleon’s declaration of
war, the same newspaper wrote that ‘the war that France is forcing
360
upon us may be a dynastic war; but since the people are bound to
dynasties, it is, in fact, a war of the French nation against the
German nation.’ During the debate over a war finance bill of the
Württemberg government in the second chamber on 21 July, Karl
Mayer, one of the leaders of the People’s party, declared his
support for fraternal allegiance with Prussia and for the victory of
the German banners, ‘which, at this moment, are the banners of
Prussia’. The speech was greeted with storms of applause from the
delegates and the listeners in the gallery. At the beginning of
August, the colours black, red, and gold were hardly to be met with
in Stuttgart any longer, having yielded their place at the side of the
Württemberg flag to the black, white, and red banner of the North
German Confederation’s navy and merchant marine—the flag in
which Bismarck united the Prussian black and white with the white
and red in the colours of the Hanse cities.
In Bavaria the war drove a powerful wedge into the Patriots’
party. Whereas Edmund Jörg wanted Bavaria to pursue a policy of
armed neutrality, his erstwhile comrade in arms, Johann Nepomuk
Sepp, ‘giving free reign to enthusiasm’, urged the deputies of the
second chamber to approve the war credits sought by the
government. The majority followed Sepp. On 19 July the
parliament, by 101 votes to 47, recognized the treaty obligation to
Prussia and approved the war credits.
The call for the annexation of Alsace and German-speaking
Lorraine was heard earlier and, at first, more emphatically in the
German south than in the north, uniting not only liberals and
democrats, but also, at the end of July 1870, representatives of
political Catholicism. The latter were guided, to be sure, not only
361
by the need for security against the western neighbor, but also by
the calculation that the inclusion of the Alsace-Lorrainers would
shift the confessional balance in Germany to the benefit of the
Catholics.
The sharpest sword in the annexationist cause was wielded by a
son of Saxony turned Prussian by choice—Heinrich von Treitschke.
Quoting Ernst Moritz Arndt, the national-liberal historian wrote in
the Preussische Jahrbücher that the land claimed by Germany ‘is
ours by virtue of nature and history’.
362
Lorraine promised, in Bismarck’s view, a certain security against
the repetition of the war. For that reason he agreed, albeit
reluctantly, to the annexation of the French territory around Metz
requested by the general staff for strategic reasons. The domestic-
political motivation for the annexation was also not insignificant.
Bismarck found it far more palatable to conciliate the national
movement through an expansion of German territory than through
the expansion of parliamentary rights.39
For this to happen, however, Prussia and its allies had first to
win the war. One of the preconditions for victory was already
fulfilled: the war was popular. Germans from north and south, who
had fought against each other only four years previously, now
knew themselves to be united in the goal of vanquishing the
historic enemy of German unity. Professors and publicists, pastors
and poets repeatedly invoked the memories of the wars of
liberation of 1813–15. Just as then Napoleon I, so now his nephew
Napoleon III was pilloried, and with him the nation that suffered
him as its leader. Whoever listened to Evangelical war sermons or
participated in religious services in the field learned that God
intended to punish the French through the agency of the Germans,
and that Germany was fated to win, since God would not permit
the just cause to go down in defeat.
To the French claim to be defending European civilization
against the German barbarians, German newspapers and journals
responded with articles and caricatures lampooning the ‘savages’,
that is, North African soldiers, fighting in the armies of the grande
nation. The Vossische Zeitung, a liberal Berlin paper, based its
assurance of victory on the insight that the Germans went ‘to war
363
as a whole nation with one mind, like never before’. On the other
hand, perhaps the war against France was the very precondition for
making a ‘whole nation’ of the Germans. The war poem ‘Wider
Bonaparte’* by Emil Rittershaus, at least, could hardly be
interpreted in any other way:
Ein einig Deutschland! Ach wie lang’ begehrt,
Wie oft erfleht in uns’rer Träume Dämmern! –
Und es ist da! Nun muss das Frankenschwert
Mit einem Schlage uns zusammenhämmern!
Doch seit der Mutter Schmach geboten ward,
Giebt’s keinen Grenzstrich mehr auf uns’rer
Karte,
Da kennen wir nur einen Schrei der Wut
Und einen Kampf auf’s Messer, bis auf’s Blut!
Nur einen Wahlspruch: Nieder Bonaparte!†
364
proclaimed the republic. Meanwhile the German armies continued
to advance. The siege of Paris began on 18 September. Toul
surrendered on 23 September, Strasbourg on 27 September. On 7
October Gambetta famously left the capital by hot air balloon and,
from Tours, called upon the nation to resist the German enemy.
In Germany, Sedan was interpreted as a divine judgement and as
a turning point in world history. On 3 September a victory
celebration took place in the New Theatre in Leipzig. The prologue
began with the words:
365
Bodelschwingh consciously invoked a particular historic
precedent: Ernst Moritz Arndt’s proposals—put into practice in
many places in Germany—to commemorate the October 1813
Battle of the Nations at Leipzig in an honourable and
simultaneously popular manner. In 1814 Arndt had given voice to
his conviction that the German people ‘knows no more festive days
—apart from those occasions made holy by divine revelation—than
the joyful days this autumn past when the Leipzig battle was
fought.’ Like Arndt, Bodelschwingh wanted to bring together
Church and State, throne and altar, uniting in one emotion love of
the Vaterland and Christian faith on Sedan Day. ‘On 2 September,
the hand of the living God intervened so visibly and powerfully in
history that it will be especially easy on this commemoration day
to get the people to reflect upon the great things the Lord has done
for us.’ Yet 2 September, as—beginning in Prussia in 1872—it came
to be celebrated in the predominantly Evangelical regions of
Germany, was not the civic-religious festival the liberal Protestants
had intended, but a day of military parades and acclamation for the
monarchy.
The reaction of the extreme left to the events of early September
was radically different. In a letter to the committee of the Social
Democratic Workers’ party at the end of August, Marx and Engels
drew from the German demands for Alsace and Lorraine the
conclusion ‘that the war of 1870 necessarily bears a war between
Germany and Russia in its womb, just like the war of 1866 bore the
war of 1870.’ In his ‘Zweite Adresse des Generalrats über den
Deutsch-Französischen Krieg’ of 9 September, Marx wrote that
‘with the surrender of Louis Napoleon, the capitulation of Sedan,
366
and the proclamation of the republic in Paris,’ Germany’s
‘defensive war’ was ended.
367
Belgium, urged the Prussian leader to negotiate with the new
provisional government in Paris.
Bismarck was not only fundamentally willing to negotiate; on 19
September he actually met twice with Favre. But a positive result
was a priori impossible. The chancellor of the North German
Confederation insisted upon the cession of Alsace and Lorraine, a
condition the French foreign minister was unable to accept. From
this point forward Bismarck, in order to forestall further
interference from foreign powers and avoid a widening of the war,
did everything in his power to see to it that the military operations
were brought to an close. To this end he required the capitulation
of Paris, and sought to have the city taken under bombardment.
At this point began a protracted and at times dramatic
confrontation between Bismarck and Moltke, the prima facie
subject of which was whether the bombardment should commence
at an earlier or later date. The larger matter at stake was whether
political or military goals should have the priority in the conduct of
the war. On 31 December German artillery finally opened fire on
the French capital. Yet another conflict between Bismarck and
Moltke over the conditions of capitulation was decided by the king
in favour of the chancellor. An agreement over a ceasefire of three
weeks and the surrender of Paris was signed on 28 January. On 26
February, after elections for a national assembly had been held
throughout France (even in Alsace and Lorraine), a provisional
peace treaty was signed in Versailles. France ceded to the newly
founded German empire Alsace and a part of Lorraine, including
the territory around Metz. In addition, it agreed to pay reparations
of 5 billion francs. The final peace agreement was signed on 10
368
May 1871 in Frankfurt am Main.
The founding of the Reich was the result of negotiations
Bismarck conducted with the representatives of the southern
German states in October and November 1870 in Versailles. With
regard to Baden and Hesse-Darmstadt, the chancellor prevailed in
his view that accession to the North German Confederation had
already all but been accomplished in a de facto sense. Both states
adopted the constitution without modification. To Württemberg
and Bavaria, on the other hand, he had to make far-reaching
concessions in terms of special rights (known as Reservatrechte),
from which Baden, in view of its state regulation of taxes on beer
and spirits, also profited in part. The most significant privilege was
the maintenance of separate postal and rail administrations in the
two states. Bavaria was also able to secure peacetime sovereignty
over its military forces and—per secret treaty—the right to be
separately represented at peace negotiations. In addition, the
constitution was modified in such a way as to strengthen the
federal element; federal executions and, within certain limits,
declarations of war would henceforth require the assent of the
Federal Council, which preserved its name. In this institution, the
three non-Prussian monarchies, Bavaria, Württemberg, and Saxony,
together possessed 14 out of 58 votes, which nominally represented
a political weight nearly as great as that of Prussia, which held 17.
According to Article 78, 14 votes were sufficient to prevent
amendments to the constitution (and only in the Federal Council
was a qualified majority necessary for constitutional amendments).
In the newly created Council committee of foreign affairs, the
second largest state, Bavaria, held the chairmanship.
369
Bavaria’s special privileges met with sharp criticism from the
National Liberal delegates, who were not federalists, but essentially
unitarians who would have preferred a homogeneous to a federal
state. Nonetheless, they did not seriously consider rejecting any of
the ‘November treaties’, nor the amended constitution as a whole;
to have done so would have risked the very national state they
sought to establish. For the Catholic fraction, on the other hand,
which began to acquire firmer organizational outlines in the
summer of 1870, the new constitution was not federal enough, and
for the Progressives and the Social Democrats not sufficiently
democratic in character. From these three parties, along with the
‘Guelfs’ (the Poles and the Danes did not participate) came the 32
negative votes in the final ballot over the treaty with Bavaria on 9
December 1870 in the Reichstag. Of the 227 delegates 195 present
voted positively, including the National Liberals, the Free
Conservatives, and the Conservatives (though the latter had a
difficult time overcoming their reservations about the lack of an
upper house). On the following day, the Reichstag adopted the first
amendment to the constitution, approved by the Federal Council on
9 December. The title German Confederation (der Deutsche Bund),
still used in the November treaties, was replaced with German
Empire (das Deutsche Reich). The federal presidency became ‘the
German Emperor’ (der Deutsche Kaiser).
It had cost Bismarck a great deal of effort (and the approval of
continuous discreet payments from the ‘Guelf fund’, the assets of
the king of Hanover, confiscated in 1866) to convince the Bavarian
king, Louis II, that the Prussian king would bear the name German
Emperor in the future and that he, Louis, should be the one to
370
propose the new title. William I himself wanted nothing to do with
it for a long time, believing that the glory of the Prussian monarchy
would thereby be eclipsed. When Bismarck refused to accept his
reservations against the ‘bogus emperorship’, William proposed the
title Emperor of Germany (Kaiser von Deutschland). This, too, was
impossible for Bismarck to accept, since it could be construed to
imply a claim to territorial supremacy unacceptable to most of the
other German princes. The variant Emperor of the Germans (Kaiser
der Deutschen), used in the imperial constitution of 1849, was
disqualified for other reasons—it sounded to ‘democratic’ in the
ears of the people who mattered. German Emperor was ultimately
the least problematic of all the proposed designations. It allowed
the Prussian king to stand forth as the first among the princes of
Germany, as in truth he was.
Only six Reichstag delegates voted against the terms Kaiser and
Reich —the Social Democrats and Lassalleans. Indeed, apart from
socialists, dyed-in-the-wool Great Germans, southern German
particularists, and old Prussian legitimists, the two words had
something to offer everybody. Germans in the smaller and middle
states were given the feeling of living in a community that was
more than just a Greater Prussia. Catholics could meditate upon the
tradition of the old Reich, democrats upon the legacy of 1848–9.
The National Liberals, of whom not a few initially objected to the
medieval and grossdeutsch connotations of the emperor title, were
optimistic that the Hohenzollern national monarchy would finally
be able to overcome German particularism. Most Prussian
conservatives were proud that their state and ruling house had
gained in power and prestige. That Prussia would ever be absorbed
371
into Germany seemed unthinkable to them.
In its last session on 10 December 1870, the North German
Reichstag not only approved the constitutional amendments
making the ‘German Confederation’ into the ‘German Empire’ and
the federal presidium into the ‘German Emperor’. It also adopted
(once again with the opposition of the six Social Democrats and
Lassalleans) a resolution brought forward by Eduard Lasker
requesting King William I to ‘consecrate the work of unification by
accepting the German imperial crown’. On 18 December 1870, the
imperial deputation of the Reichstag was received by William at
Versailles. At the head of the delegates stood the president of the
parliament, the National Liberal Eduard Simson, who once before,
on 3 April 1849, as president of the German National Assembly,
had asked a Prussian king to accept the title of German Emperor.
This time the delegates could be sure that their request would be
granted. For already on 2 December, William had received the
letter of King Ludwig II of Bavaria (drafted by Bismarck) asking the
Prussian king if he would consent to the common wish of the
princes of Germany that he bear the title of German Emperor in the
exercise of his presidential powers. Since the king had not refused
this request, which came from his peers, his answer to the
parliament’s entreaty could only be positive. William was in a
much different position than that of his brother Frederick William
IV in spring 1849. The commentary of the conservative Kreuz-
Zeitung hit the mark: ‘Today, the “emperor” makes the
“constitution”, not the constitution the emperor.’
The southern German parliaments had not yet ratified the
November treaties by the time the parliamentary deputation
372
received the royal affirmative. They only did so—with a single
exception—in the days between 21 and 29 December 1870. The
exception was Bavaria. The first chamber in Munich approved the
treaty with the North German Confederation by a large majority on
30 December. In the second chamber, however, the resistance was
strong enough to cast doubts on the feasibility of the necessary
two-thirds majority. On 13 January 1871 the Bayerisches Vaterland,
the mouthpiece of the extreme wing of the Patriots’ party, declared
its objections to the founding of a new Reich:
373
in the hall of mirrors at Versailles; they were represented by
princes of their dynasties. Though emissaries of the parliament
were not present, the military was. One of the young officers in
attendance was 23-year-old Paul von Benneckendorff und von
Hindenburg. The vexed question of title, which plagued William to
the end, was solved by Grand Duke Frederick von Baden, who
simply circumvented it by exhorting the assembly to hail ‘Emperor
William’.
The enthusiasm of those who had worked and waited for
German unity for decades knew no bounds. On 27 January 1871,
the day special editions of the newspapers were reporting the
imminent capitulation of Paris, the liberal Bonn historian Heinrich
von Sybel, then 53 years old, expressed his tears of joy in a letter to
his like-minded colleague, the Karlsruhe historian Hermann
Baumgarten:
374
5
375
would probably not have accepted it. As it was, they were just able
to tolerate the ‘half-hegemonic position of the Bismarck Reich on
the continent’, as the historian Ludwig Dehio put it in 1951. Events
in post-war France made their self-restraint more palatable. While
the country was still under German occupation, the French capital
undertook to give the world an example of a society in complete
political and social chaos in the form of a revolt of the Commune.
The shock that went out from Paris at the end of March 1871 was
as lasting as it was violent. In the face of the red revolution of the
Communards, which was finally and bloodily suppressed at the end
of May by troops of the French government, the fear of Bismarck’s
‘German revolution’ rapidly faded. The newly founded Reich was
able to present itself to Europe as a custodian of the public order.1
Disraeli was not the only one to compare 1871 with 1789.
Evangelical theologians and writers in Germany interpreted the
military defeat of France as the final defeat of the principles of the
French Revolution. ‘If we contemplate the forces that struggled
against each other in this war,’ wrote the Evangelische
Kirchenzeitung at the beginning of 1871, ‘it presents itself as a
victory of loyal subservience over the revolution, of the divine
order over anarchy, of the forces of morality over the dissoluteness
of the flesh, of rule from above over the sovereignty of the people,
of Christianity over modern paganism.’ Shortly afterwards the same
paper observed: ‘The revolution is France’s chronic illness. Not only
has it gone through the most sanguinary revolutions itself; it has
also carried the principle of revolution everywhere it has gained a
firm foothold.’
During the war, Evangelical theologians compared Paris with
376
sinful Babylon, interpreted the French Revolution and the two
Bonapartes as incarnations of the Antichrist, and designated the
French ‘the most satanic of peoples’. Now they saw in the German
triumph proof that God had predestined the Hohenzollerns to do
great things for Germany. It goes without saying that ‘Germany’
meant ‘Evangelical Germany’, an identification that brought the
founding of the Reich into close relation with the Reformation.
‘The paths of the Most High with our people are being made
straight, His thoughts concerning us are coming to their fulfilment,’
announced the Neue Evangelische Kirchenzeitung on 7 January 1871.
‘The epoch of German history that began in the year 1517 is
coming to a divinely appointed end with war and the cry of battle.’
‘The holy Evangelical empire of the German Nation is reaching its
fulfilment,’ rejoiced Adolf Stoecker, future court chaplain, three
weeks later, after the proclamation of the emperorship at
Versailles; ‘in such a way do we recognize the hand of God from
1517 to 1871!’ And on 18 March 1871, exactly twenty-three years
after the fighting in the streets of Berlin in 1848, the Neue
Evangelische Kirchenzeitung hailed Emperor William ‘as the initiator
of a new history, as the founder of the Evangelical empire of the
German Nation’.
The expression ‘Evangelical empire’ (evangelisches Kaisertum)
quickly caught on. On 6 March 1872 Bismarck, speaking before the
Prussian House of Lords, warned against the danger to the
confessional peace since the ‘Austrian war, after the power that had
formed the stronghold of Roman influence in Germany was
defeated and the future of an Evangelical empire clearly loomed on
the horizon’. The National Liberal party leader Rudolf von
377
Bennigsen took this thought further in the Prussian House of
Representatives on 26 January 1881 when he invoked the ‘bitter
enmity that prevailed and still prevails in Rome at the founding of
the Evangelical empire’—which comment drew from the ranks of
the Catholic Centre the corrective interjection ‘partitätisches
Kaisertum!’* Even after his dismissal, the founder of the new Reich
still insisted upon the its Protestant character. ‘I am committed to
the secular leadership of an Evangelical empire, and to this empire
I am faithful,’ Bismarck announced on 31 July 1892 in the town
square of Jena.2
The watchword of the ‘Evangelical empire’ would help create
national identity by dramatizing a double difference: between the
old Reich and the new, and between the German Reich and
Austria. Since the appeal to a common language, culture, and
history was not sufficient to legitimate a German national state, an
ideological justification was necessary, one that could give Prussian
leadership a meaning comprehensible even to non-Prussian
Germans. The invocation of the Reformation fulfilled this purpose,
but only for some Germans. The attempt to undergird the political
hegemony of Prussia with the cultural hegemony of Protestantism
united and divided Germany at the same time. The expression
‘Evangelical empire’ brought together Evangelical liberals and
Evangelical conservatives, Evangelical Prussians and Evangelical
non-Prussians. But it also drove a wedge between Protestants and
Catholics and thus impeded the progress of the national unification
it was intended to further.
The formation of the German nation did not begin with the
378
founding of the German national state, nor did it come to an end
with the conclusion of the latter. Rather, it entered a new phase. In
a larger sense, everything that gave the Germans the consciousness
of a community transcending the borders of their territorial states
was part of the process of national formation—first and foremost
language, culture, and history. Seen in this light, the beginnings of
the nation lay in the Middle Ages, regardless of what national-
liberal historians wrote and said about the Holy Roman Empire of
the German Nation. For the formation of a klein-deutsch nation,
such a common ground was a necessary but not sufficient
condition. The development of a Little German Staatsnation out of a
German Kulturnation was intimately connected with another
developmental history—that of Austria, whose centre had long ago
shifted from Germany to central Europe. Non-Austrian Germany, in
both its predominantly Evangelical and predominantly Catholic
territorial states, was marked by a history of confessional conflict.
Austria had successfully closed itself off from Protestantism, but in
doing so had separated itself from the rest of Germany. This
allowed not only the Habsburg empire to submit to the decision of
1866, but also the German states defeated by Prussia as allies of
Austria.
A gradual accommodation of the liberals of the ‘third Germany’
to Prussia can be traced back to the Vormärz period. Confessional
sympathies were as important as economic interests. In terms of
trade policy, the German Customs Union anticipated the
kleindeutsch solution, though it did not render it inevitable. The
Revolution of 1848–9 caused many, but by no means all, German
nationalists to realize that a part of the Habsburg empire could not
379
join a German nation without fatally compromising the whole.
Austria did everything in its power, both before and after 1848, to
keep the German national movement at a distance. After the die
had been cast on the fields of Königgrätz, there was still a great
deal of popular opposition to a Prussianled German national state,
above all south of the Main. But a realistic alternative no longer
existed.
Unlike the National Liberals, the federal chancellor had no
intention of waging a frontal war against the individual patriotic
movements of the German states outside of Prussia. ‘As a rule,
German patriotism requires the medium of dynastic loyalties in
order to be active and effective,’ Bismarck wrote in his memoirs’
most famous chapter, entitled ‘Dynastien und Stämme’.* He acted
according to this maxim when he organized the new Reich along
federal lines. His expectation that the war with France would
strengthen German national feeling was fulfilled. The National
Liberals emerged from the first Reichstag elections in March 1871
as the strongest party by far, gaining fully 30 per cent of the vote.
In Bavaria the liberal parties together, and in Württemberg and
Baden the National Liberals alone gained the absolute majority.
The results did not indicate the decline of particularist sentiment so
much as a broad approval of the new German Reich.
The strongest bastion of individual state patriotism continued to
be Prussia. There, the Conservative party did only marginally
worse than the National Liberals in the federal elections (20.8 and
23.3 per cent of the vote, respectively) and far better that the
emphatically empire-friendly Free Conservatives (12.2 per cent). In
the conservative view, the unification of Germany was primarily
380
Prussia’s achievement, not that of the national movement. In
consequence, Germany had every reason to become more Prussian,
that is, more military in spirit, but was not in a position to demand
the converse, that Prussia become more German, that is, more
bourgeois. This credo could be articulated both offensively and
defensively. Hans von Kleist-Retzow chose the second option when
he declared before the House of Lords on 21 December 1870:
381
Before the founding of the Reich, to be ‘national’ meant to be
against the dynastic principle of the Partikularstaat and for the civic
principle of the national state. Seen in this way, ‘national’ and
‘liberal’, ‘unity’ and ‘liberty’ were conceptually very nearly of a
piece. However, after Prussia, the largest Partikularstaat, placing
itself at the head of the national movement and, at the same time,
at the head of the German states, had unified Germany to the
enthusiastic applause of the National Liberals, there arose
conservative competition to the liberal claim to sole national
representation. What we today call ‘imperial nationalism’,
Reichsnationalismus, did not yet possess clear outlines in 1871. The
only thing that was clear was that the foundation of the German
Reich had set off a battle over what the concept ‘national’ would
now signify.3
Liberals and conservatives were not the only participants in this
struggle. German Catholics, many of whom voted for one of the
liberal parties in the first federal elections and not the newly
founded (1870) Catholic Centre Party (Zentrumspartei), were
fundamentally provoked by the Evangelical interpretation of the
Franco-Prussian War and the founding of the Reich. One of the
Catholic interpretations hurled the challenge back to the
Protestants. In his new year’s article for the first 1871 issue of the
Historisch-politische Blätter für das katholische Deutschland, Edmund
Jörg adapted the medieval doctrine of the translatio imperii to
describe—or at least suggest—the ‘transfer’ of the imperial crown
from Napoleon III to William I, thus highlighting the illegitimate
character of the new Reich, born of ‘the politics of revolutionary
nationalism’.
382
‘If we are soon to have a German Reich,’ wrote Jörg,
383
kleindeutsch solution after the German war of 1866 and who also
participated in the founding of the Centre party in 1870, justified
the necessity of concerted action in light of the shifted confessional
balance.
384
over the Catholic Church.4
385
summarized the typical view in October 1871: the battle against
France was only the ‘preliminary stage’ of the even more difficult
battle against the spirit of untruth and error, embodied in
international social democracy, inside the new Reich. Just as seven
years previously, after the battle with Denmark, many Prussian
conservatives had clamoured for war against the ‘inner Düppel’, so
now many Germans wanted to vanquish the ‘inner France’,
whether in the form of Catholicism or of social democracy.
It was true that Bebel made it easy for the enemies of the
German workers’ movement to draw a distorted picture of his
party. His declaration of unreserved solidarity with the Paris
Commune was indebted to Marx’s completely uncritical assessment
of the Paris events in ‘Der Bürgerkrieg in Frankreich’, formally
entitled ‘Adresse des Generalrats der Internationalen Arbeiter-
Assoziation’.* Two decades later, in the preface to the third German
edition of the text, Engels summarized this assessment in the
exhortation: ‘Look at the Paris Commune. That was the dictatorship
of the proletariat.’ Considering the putsch-like characteristics of the
rebellion, the excesses of violence committed also by the
Communards, the abolition of the traditional separation between
legislative, executive, and judicial powers, and the strident contrast
between the ideal of this supposedly higher form of democracy and
its reality, it is understandable even today why in spring 1871 the
great majority of Europeans looked towards the French capital with
fear and abhorrence. The German Social Democrats had little in
common with the Communards. Yet Bebel gave the impression that
his party was well on its way to eclipsing the French example. In
doing so, he set consequences in motion that he could not possibly
386
have intended and, in the interests of the German workers’
movement, should have avoided.5
In their protest against the annexation of Alsace and Lorraine,
the Social Democrats stood virtually alone in the new Reich. The
only well-known bourgeois democrat to speak out publicly against
this violation of the right to self-determination in September 1870
was Johann Jacoby of Königsberg, who was thereafter arrested. In
April 1872, shortly after the announcement of the verdicts in the
trial of Bebel and Liebknecht in Leipzig, Jacoby joined the Social
Democratic Workers’ party. Far more representative of the liberal
bourgeois attitude was the position taken by David Friedrich
Strauss (originally from Ludwigsburg), a scholar of religion and the
founder of the historical study of the life of Jesus Christ, in an open
correspondence with his no less famous Paris colleague Ernest
Renan in 1870–1. Strauss wanted to recover the two ‘German
provinces’ primarily for security reasons. We Germans, he wrote on
29 September 1870, would have to be ‘supreme fools, if we do not
seek to regain what was ours and what is necessary for our security
(but no further than is necessary, of course)’.
Renan, who saw in the Franco-Prussian War the greatest disaster
that could have befallen civilization, answered Strauss’s letter on
15 September 1871, nearly a year later.
387
upon scientific error, since only very few countries possess a really
pure race—can only lead to wars of annihilation, to ‘zoological’
wars... That would be the end of this fruitful mixture called
‘humanity,’ composed of such numerous and altogether necessary
elements. You have raised the banner of ethnographic and
anthropologic policy in the world in place of liberalism. This policy
will be your undoing ... How can you believe that the Slavs, who
emulate you in all things, will not do the same as you have done to
others?
The voice from Paris was the voice of reason. In Renan’s view,
the nation was a community of will, or, as he expressed it in a
lecture at the Sorbonne in 1882, ‘a plebiscite repeated daily’ (un
plébiscite de tous les jours). In imperial Germany, this concept of the
nation as a political idea based on the subjective decision of the
individual found no echo. Since no common German state had
existed before 1871, the seemingly objective factors of language,
extraction, and culture had always been fundamental to the
German conception of nation. It was true that, with the foundation
of the Reich, the long-desired national state had finally become
reality. But this state could not make membership in the German
nation conditional upon the free decision of its citizens—otherwise
the Alsace-Lorrainers in the west, the Poles in the east, and the
Danes in the north would have dissociated themselves from
Germany. On the other hand, there were millions of German-
speaking Austrians in the Habsburg monarchy, whom erstwhile
gross-deutsch-oriented Germans, whether Catholics or Social
Democrats, still considered part of the German nation. From their
388
perspective, the new German Reich was an ‘unfinished nation
state’.
For kleindeutsch Protestants, this point of view bordered on
spiritual treason. The Evangelical rationalization of the Reich after
1871 had its origins in a dilemma. Since the German nation could
be defined neither as a linguistic community nor as a community of
the will, it was necessary to look for German imperial identity on
another plane. The National Liberals and nationalist conservatives
were convinced they had found it: the hegemony of the Protestant
principle in German culture, society, and state. From the beginning,
this version of imperial nationalism implied both the continuing
secularization of German Protestantism, making religion
ideologically of very flexible utility, and the increased
theologization of German nationalism, separating the latter from
the purely secular nationalism of France.
In a 1935 book, written during his exile, the philosopher and
sociologist Helmuth Plessner called the Bismarck Reich a ‘great
power without an idea of the state’ (Grossmacht ohne Staatsidee).
Plessner discussed the belated nature of German national politics:
by the time Germany had become a nation, the time of the great
world-political visions was already at an end. In 1789 the French
were still able to link the revolutionary re-founding of their nation
with an appeal to the universal validity of ‘liberty, equality,
fraternity’. Bismarck’s ‘revolution from above’ brought forth a
‘power state’ (Machtstaat) that could no longer justify itself in terms
of universal ideas. German pastors, professors, and publicists
sensed this deficiency. Regardless of what they did to address it,
however, the result could only be compensatory in nature.
389
Still, this compensation did have a certain tradition. For the
early German nationalists after 1800, intellectual world leadership
served to make up for the political impotence of Germany. The
unwilling revolutionaries in the Frankfurt Paulskirche had created
a German-dominated central Europe in their imaginations before
their more modest plan for a kleindeutsch state foundered on the
hard reality of two German great powers. The later National
Liberals were ever more inclined during the era of constitutional
conflict to consider the expansion of the Hohenzollern state the
prerequisite for liberal domestic reforms.
Thinking in categories of the ‘power state’ as compensation for
internal political powerlessness—before 1871, liberal Germany had
gone so far along this path that it was very difficult to turn back
once the ‘power state’ had become reality in the form of the
German Reich. On the international stage, the founding of the
Reich represented the fulfilment of liberalism’s national goal. But
inside Germany the liberal programme remained in large part
unrealized. Since there was no lack of opponents, the great
question was now whether the thinkers and visionaries of the
‘power state’ would seek out a sphere of activity within the newly
created nation.6
390
on 17 January 1873) had long since begun. In the 1860s, two
primarily Catholic states, Bavaria and Baden, became the first in
Germany to initiate policies of educational reform, in response to
which they aroused the embittered and protracted opposition of
the Catholic clergy. Such battles over the boundaries between
Church and State also went on in other countries during the second
half of the nineteenth century, and it was not by chance that they
were most intense in predominantly Catholic countries, above all
France and Italy.
The conflict was inevitable, but the particular form it took was
not. From the viewpoint of the church, liberalism, heir of the
Enlightenment and embodiment of modernism, was the aggressor,
since it sought to abolish vested ecclesiastical rights like the
clerical supervision of the schools. For liberals in the broad sense of
the term, whether Catholic, Protestant, Jewish, agnostic, or atheist,
the papacy was the first to throw down the gauntlet in its quasi-
medieval declaration of war against the present age. When the First
Vatican Council promulgated the doctrine of infallibility, it seemed
to confirm this impression. Clearly, the Catholic Church’s will to
power was not going to be defeated by human reason alone.
Germany contributed indirectly but decisively to the termination
of the pope’s worldly rule only a few months later. The Kingdom of
Italy was only able to annex the Papal States in October 1870 after
the Germans had defeated its protector, Napoleon III, and brought
about his downfall. Although an Italian law of May 1871
guaranteed the pope’s sovereign rights as well as his possession of
the Church of St Peter, the Vatican, and the Lateran palaces, Pope
Pius IX rejected the offer and opted for the status of ‘Vatican
391
captivity’, which lasted nearly six decades, until the Lateran
treaties with fascist Italy in February 1929.
The restoration of the pope’s worldly rule was the first political
demand made by the Centre party after the foundation of the
Reich. On 18 February 1871, in the Prussian headquarters at
Versailles, Bismarck received from the hands of Bishop von Ketteler
a petition from fifty-six Catholic members of the Prussian House of
Representatives requesting that Emperor William do everything in
his power to make the pope once again the lord of the Papal States.
Bismarck angrily rejected the petition, since its acceptance would
have meant a rupture with Italy and most certainly war.
The next challenge from the Catholic Centre was a motion by
the party fraction in the Reichstag on 27 March 1871 to
incorporate (as in Prussia) articles on basic rights into the federal
constitution, including freedom of religion and the guarantee of
ecclesiastic self-government. The chancellor, whom the emperor
had just made a prince for his contribution to the founding of the
Reich, saw in the Catholic initiative an attack on the cultural
sovereignty of the individual states. For its part, the Reichstag
majority had no intention of promoting church affairs, and
promptly rejected the proposal. The complete draft of the
constitution, presented to the federal parliament by Bismarck on 23
March, represented a version of the North German constitution
revised and updated in only a few points. On 14 April 1871 the
great majority of the plenum adopted it against only seven negative
votes—from the two Social Democrats, four Guelfs, and one Dane.
Despite the rejection of its motion on basic rights, the Centre party
voted for the constitution.
392
In his Gedanken und Erinnerungen, Bismarck wrote that for him
the beginning of the Kulturkampf was defined by its Polish aspect.
He suspected the ‘Catholic section’ in the Prussian ministry of
education and ecclesiastical affairs of promoting the Polish national
movement in Posen and West Prussia through the Catholic clergy.
In July 1871, he succeeded in having the ministry dissolved. It was
only a short step from the allegation of the Catholic Church’s
complicity in the Polish cause to the claim that the Catholic Centre
was also allied to other ‘enemies of the Reich’ like the Guelfs and
the Alsace-Lorrainers—that is, that the Centre itself was hostile to
the Reich. Bismarck considered the attitude of the Centre and the
Catholic clergy a challenge to be met with measures strengthening
the authority and unity of the state. He was prepared to accept that
a sharply laical policy, which to a certain extent also affected the
Evangelical church, was rejected by the majority of old Prussian
conservatives, who broke with the chancellor on this occasion. For
his agenda in the Kulturkampf, Bismarck could depend on a
majority gathered from a Conservative splinter group, the Free
Conservatives, the National Liberals, the greater part of the
Progressives, and the newly founded (but short-lived) Liberal
Imperial Party (Liberale Reichspartei), whose parliamentary fraction
consisted in roughly equal parts of federally minded Protestants
and liberal Catholics. (The latter group included Baron
Roggenbach, former Baden foreign minister, and Prince Chlodwig
von Hohenlohe-Schillingsfürst, former prime minister of Bavaria.)
The first legislation of the Kulturkampf derived from a Bavarian
initiative. The ‘pulpit paragraph’ of December 1871 made it a
criminal offence for clerics to discuss affairs of state ‘in a manner
393
endangering public peace’ in the exercise of their offices. A law
dealing with the supervision of schools followed in March 1872.
Clerical inspection of city and district schools was abolished, and
all private schools were placed under state supervision. Four
months later, in July 1872, the federal parliament passed the so-
called ‘Jesuit law’ banning all establishments of the Society of Jesus
in the Reich and imposing residency restrictions on individual
members of the order. This law emerged from the campaigns of the
old-Catholics and the German Protestant Association, who sought
to outdo each other in the fierceness of their attacks on the Jesuits.
The violation of liberal principles was flagrant, but only a few
committed liberals like Bamberger and Lasker voted against the
measure. The liberal majority found itself entirely of one mind with
Bismarck as never before when, on 14 May 1872, the chancellor
declared to parliament: ‘To Canossa we shall not go—neither in
body nor in spirit!’ This promise was greeted with storms of
applause.
Prussia’s next blow fell in spring 1873. Buttressed by a
constitutional amendment, one of the so-called ‘May laws’ made
the appointment of any priest or minister to ecclesiastic office
conditional upon his getting a certificate from a German
gymnasium and passing a ‘cultural exam’ in the fields of
philosophy, history, and German literature. Another law created a
royal court of law for ecclesiastic affairs where appeals against
church penalties could be heard. In March the Prussian House of
Representatives passed a law subsequently adopted (February
1875) by the Reichstag in expanded form: the law concerning
obligatory civil marriage, which transferred the documentation of
394
marital status (birth, marriage, and death) from the church to
newly created register offices, the Standesämter. This law is the only
piece of Kulturkampf legislation that survives to the present day.
The ‘expatriation law’ of May 1874, on the other hand, was
nothing more than an instrument of repression. Disregarding the
most basic principles of the rule of law, it permitted the
governments to confine clerics to certain places of residence, or to
expatriate and ban them from the Reich. A Prussian law was passed
the same month empowering the minister of education and
ecclesiastic affairs to appoint a commissioner to temporarily
administer bishoprics that fell vacant owing to state action. Two
further Prussian laws of spring 1875 were aimed at the material
basis of the adversary. The April ‘bread basket law’ stopped all
payments of state funds to the Catholic Church. In May the
‘monastery law’ dissolved the monasteries of all the orders except
those devoted exclusively to the care of the sick. The annulment of
the Prussian constitution’s article on religion in June 1875
represented a kind of keystone of the Kulturkampf legislation—an
unwilling admission on the part of parliament and administration
that the constitutionality of the foregoing laws was not above all
suspicion.7
‘Germany has taken up the battle against the minions of Rome,
who have no fatherland, knowing full well that this battle will be
longer and more difficult than the one against our arch-enemy
across the Rhine.’ In expressing such sentiments in its leading
article on 21 October 1876, the national-liberal Berlin National-
Zeitung had no criticism to fear from its readers. A year later, in an
article on ‘Sedan Day’ in 1877, the same paper compared Catholics
395
to Social Democrats:
396
successful among the urban and rural lower classes than among the
educated and propertied bourgeoisie, which in many places—
Cologne and Bonn, for example—continued to favour liberal
candidates over those from the Centre party.
The clergy aided this development by encouraging a movement
of popular piety aimed at promoting cohesion among Catholics—
which, however, tended to appeal to women far more than to men.
Thus, the Kulturkampf contributed to Catholicism’s anti-intellectual
tendencies and to the widening of the chasm separating Catholics
from Protestants. And for a great number of Catholics the wounds
left by ostracism and repression had not healed even long after
most of the legislation had been softened or repealed. They had
been accused of national untrustworthiness, even of hostility to the
Reich, and they could not give themselves over to the illusion that
prejudices were shorter-lived than paragraphs.8
397
average share price; at the end of 1874, it was just over half of
what it had been at the end of 1872. But the low point was not
reached until 1878–9. The boom of the the new empire’s
Gründerzeit, the ‘founders’ era’, seemed increasingly like a
‘founders’ fraud’, and the long-term economic consequences of the
‘great crash’ were felt by many contemporaries to be a ‘Great
Depression’.
The social-psychological impact of the economic collapse has
been well described by the historian Hans Rosenberg, who speaks
of the ‘radical climate change in consciousness and in ways of
reacting’ throughout the 1870s and 1880s, all the way up to the
onset of a new economic upswing in 1896. The characteristics of
this ‘climate change’, according to Rosenberg, were
398
most spectacular bankruptcy of the crash was that of the ‘rail king’
Bethel Henry Strousberg, a converted Jew of East Prussian descent.
Strousberg owed his influence not to his own capital, but to
connections with members of the East Prussian nobility and high
bureaucracy, who counted in their turn upon the investments of
countless small shareholders—until Strousberg’s failed speculations
dragged everybody into ruin.
Strousberg was no liberal. He had served in the North German
parliament from 1867 to 1870 as a Conservative delegate. It was,
in fact, a liberal politician who revealed the ‘Strousberg system’ in
the Prussian House of Representatives on 7 February 1873—Eduard
Lasker. Along with Ludwig Bamberger, Lasker was the most
prominent Jew among the leading National Liberals. Heine’s
famous dictum that the baptismal certificate was the ‘entry ticket
to European culture’ had lost its universality; neither Lasker nor
Bamberger had been baptized. Nonetheless, they were among the
most distinguished and popular politicians of the National Liberal
party and two of the most convinced and convincing apostles of the
maxim that ‘the path of freedom lies through unity.’
One of the blessings of liberty many German Jews believed
national unity would bestow was progress along another path, the
one leading to authentic, not merely legal, equality between
Christians and Jews. In other words, German unity was supposed to
lead to the full emancipation of the Jews at the same time as it led
to the political emancipation of the bourgeoisie. The German
liberals of the new imperial era made the cause of Jewish
emancipation their own in a way one can almost call the
‘liberalization of liberalism’. There seemed to be little left of the
399
anti-Jewish reservations common among liberal politicians of the
Vormärz— despite the fact that anti-Jewish clichés could still be
found in the much-read novels of liberally minded authors of the
1850s and 1860s, writers like Gustav Freytag, Wilhelm Raabe, and
Felix Dahn.
The case of Richard Wagner was different. In May 1849 the
composer had fought for the Revolution in the streets of Dresden.
In 1850, while in exile in Switzerland, he wrote his essay ‘Das
Judentum in der Musik’,* in which he attacked the ‘Jewification of
modern art,’ professed his ‘instinctive antipathy towards
Jewishness’, and demanded ‘emancipation from the Jews’. Fifteen
years later, his essay ‘Was ist deutsch?’† (1865) went even further.
Wagner lamented that after the Vormärz period, ‘French-Jewish-
German democracy’ had been well received ‘by the misunderstood
and wounded spirit of the German people’, and sought to explain
the failure of the 1848 Revolution by asserting ‘that the authentic
German suddenly found himself and his name represented by a
kind of human being utterly foreign to him’.
The ‘great crash’ of 1873 brought to a close the short period of
general open-mindedness towards Jews that had begun in 1859
with the renaissance of the liberal movement. The National Liberals
were unable to preserve their national image in the face of
defamatory accusations of their being the henchmen of Jewish-
controlled international finance. It was no help to National Liberal
Jews that one of their own had been the first to attempt to expose
the ‘founders’ fraud’. In mid-1875, the journalist Franz Perrot
initiated with the so-called ‘era articles’ in the Kreuz-Zeitung a
400
campaign vilifying not only liberals and Jews in general, but also
Bismarck’s banker Gerson Bleichröder, the secret financier of the
Prussian war of 1866 and intimate of some of the most powerful
European statesmen. Perrot even went so far as to attack the
chancellor himself.
If the monetary and economic policies of the German Reich,
wrote Perrot, always give the impression of being Jewish policies,
that is, policies and legislation by Jews and for Jews, that is
Articles hostile to the Jews appeared in the mid 1870s not only
in the mouthpiece of the Prussian conservatives, but also in
Germania, the Centre party’s most important newspaper, and other
organs of political Catholicism. Edmund Jörg, at that time also a
Centre party delegate to the Reichstag for the Bavarian electoral
district of Schwaben, came at the beginning of 1877 to the
conclusion that the Kulturkampf was a ‘prime promotional tactic of
401
the new stock-market era’, since it had also drawn well-meaning
Protestants into its vortex and thus prevented a ‘reconstruction of
the conservative elements’. ‘In such a way, however, this unbloody
war of religion has also been greatly of service in assuring the
Mamelukes of money power in the parliamentary bodies their
majority and their influence.’ Shortly thereafter, Jörg spoke of
‘Jewified liberalism in German lands’. The Historisch-politische
Blätter für das katholische Deutschland equated the Jews with the
‘high cosmopolitan powers of finance’ and the actual ‘enemies of
the Reich’. Even Ketteler saw in the Kulturkampf a ‘Freemason-
Jewish-liberal conspiracy’ against the Catholic Church.9
The anti-Jewishness of the 1870s could no longer be considered
religious in motivation. It was ‘modern’ insofar as it took aim
against modern, emancipated Jewry using purely secular discourse.
The Jews were portrayed as agents of those aspects of modernism
by which various groups in society felt themselves threatened. For
many of its opponents, liberalism was nothing more than the
world-view of modern Jewry. Those who considered themselves
victims of capitalist speculation tended to see things like the
publicist Otto Glagau, who in the popular Gartenlaube contrasted
the Jews as representatives of unproductive, ‘grasping’ capital with
the productive, ‘creative’ capital of Christians. Many would have
echoed Glagau’s much-quoted assertion that ‘The social question
today is, basically, a Jewish question.’ Jewish bankers found
themselves stigmatized as members of a ‘Golden International’ (the
title of an 1876 text by the Berlin judicial official Carl Wilmanns),
which made it easy to suspect them of being nationally untrust-
worthy and in close contact with the—also putatively Jew-
402
dominated—‘red International’.
This was not the first time hostility toward the Jews and anti-
liberalism had come together. The origins of their partnership lay
back in the era of Prussian reform, when conservatives like
Friedrich August Ludwig von der Marwitz, seeking to mobilize
peasants and tradesmen, declared war on Jewish emancipation,
freedom of trade, and the free movement of labour. Emancipation
itself was always accompanied by an anti-Jewish counter-
movement. Tracts and pamphlets compared Jews to vampires and
disease-causing agents; demands for the expulsion, castration, and
even the destruction of the Jews were frequently heard. From the
late 1830s, and increasingly during the 1860s, ‘ultramontanist’
writers denounced the Jews as the beneficiaries and exploiters of
the process of secular emancipation threatening everything the
Catholic Church considered sacred. Immediately after founding of
the Reich, the event that sealed the minority position of the
German Catholics, this simultaneously anti-Jewish and anti-liberal
movement within Catholicism gained new force. After 1873, anti-
Jewish rhetoric was employed almost universally against
everything considered to be ‘modern’ and thus an attack on
tradition. Liberalism and socialism suffered the same fate, and
since Jews played a role in both movements, they were both
considered manifestations of the Jewish will to power.
In fact, if one gave credence to the accusations of popular
writers like Constantin Frantz and Wilhelm Marr, the Jews had
advanced to become the real rulers of Germany in the new imperial
era. Under the regime of national liberalism, as Frantz wrote in
1874, the Jews had entered into ‘the cynosure of our development’,
403
and whatever was called ‘progress’ was ultimately only the
‘progress of Jewification’. ‘A German Empire of the Jewish Nation is
thus rising before our eyes.’ ‘Who actually rules the new Empire?’
the same author asked two years later. ‘And to what purpose were
the victories of Sadowa and Sedan achieved, to what purpose the
billions carried off, to what purpose is the Kulturkampf, if not to
further the rule of the Jews?’ In his protest against Jewish
emancipation, Frantz did not hesitate to invoke the ‘short work’
that Fichte had wanted to make of the Jews in his 1793 work on
the French Revolution: before giving the Jews civil rights, their
heads were all to be cut off in one night and replaced with others
containing no Jewish ideas.
For Marr, baptism made absolutely no difference. In his
extraordinarily successful book Der Sieg des Judenthums über das
Germanentum. Vom nicht confessionellen Standpunkt aus betrachtet*
(1879), the writer fought not against the religion of the Jews, but
against the Jews as a race, one that had become a ‘world power of
the first order’ and strove for ‘world domination’. Membership in
the Jewish community thus became an unalterable fact of nature,
one that could not be changed by an act of volition like conversion
to a Christian confession. Throughout the 1870s, Jews were
increasingly referred to as ‘Semites’. The term ‘anti-Semitism’ first
arose in the autumn of 1879 in Marr’s circle. The ‘League of Anti-
Semites’ was founded in Berlin at approximately the same time,
and by followers of the same writer.10
The secularization of anti-Jewish hatred allowed the latter to
make pretences to a ‘scientific’ and ‘objective’ character.
404
Paradoxically, the ‘modernization’ of anti-Jewish hatred gave rise
to anti-modernism in its most extreme form. The ‘great crash’ of
1873 was merely the catalyst setting a momentarily retarded
protest movement in motion once again—protest not against the
founding of the Reich as such, but against everything that linked
the new German state to the philosophy and practice of liberalism.
Those strata of society who felt themselves threatened by
increasing industrialization—above all peasants, tradesmen, and
small merchants—proved especially susceptible of anti-liberal
mobilization.
Another sworn enemy of liberalism was the Catholic Church,
and not only since the Kulturkampf. The Catholic adversaries of the
Jews expressed the traditional anti-Judaism of the church in a
manner frequently indistinguishable from the political, economic,
and social discourse of ‘modern’ anti-Semitism. The reverse was
also true: ‘worldly’ anti-Semites like Wilmanns and Perrot invoked
the example of Christian Jew-haters like the Münster Catholic
theologian August Rohling, who in 1871 published Der
Talmudjude,* a classic work of religious anti-Semitism. The old
hatred of the ‘murderers of God’ and ‘usurers’, as common among
Protestants as Catholics, was not simply replaced by ‘modern’ anti-
Semitism. Christian anti-Judaism united with secular anti-Semitism
and lent it a historic ‘depth’ independent of any temporary
fluctuations in the state of the economy.
Most Christian anti-Semites, Protestants as well as Catholics,
were at pains to deny they countenanced any kind of racial hatred.
In his first major speech against ‘modern Jewry’, held on 19
September 1879 at an assembly of the Christian Social Workers’
405
party (which he founded) in Berlin, Adolf Stoecker, from 1874
chaplain of the royal Prussian court in Berlin, assured his audience
that he dedicated himself to the Jewish question ‘in full Christian
love’. This sentiment did not prevent him from describing ‘modern
Jewry’—which he sought to differentiate from both traditional-
orthodox and enlightened reform-Judaism—as ‘a great danger to
the life of the German people’. The Jews were and would remain a
‘people within a people, a state within a state, a tribe isolated
among a foreign race’. Here and there, noted Stoecker, hatred was
already flaming up against the Jews, a hatred irreconcilable with
the Gospel. ‘If modern Jewry continues to use the power of
The Rise of ‘Modern’ Anti-Semitism 209 capital as well as the
power of the press to the ruin of the nation, a catastrophe will be
unavoidable in the end. Israel must give up its aspiration to
become the lord of Germany.’
Among the remedies Stoecker proposed were the ‘reintroduction
of confessional statistics, in order that the imbalance between
Jewish wealth and Christian labour might be ascertained; reduction
of the appointment of Jewish judges to the proportion of Jews in
the population; removal of Jewish teachers from our public
schools.’ A week later the court chaplain fulminated against the
fact that Jews enjoyed a much greater representation among
employers, ‘directors and directresses’, in commerce, and among
the students of higher education—especially in Berlin—than was
warranted by the percentage of Jews in the population.
406
for survival of the most intensive kind. If Israel grows even further
in this direction, it will be more than we can cope with. For let us
harbour no illusions: on this soil, race stands against race and is
waging a racial conflict—not in the sense of hatred, but in the
sense of competition ... If Israel is united in socialpolitical activity
over the whole earth through the ‘Alliance Israelite’ [the Alliance
Israélite Universelle, an international Jewish association (H.A.W.)],
that is a state within the state, international within the nation.
407
indeed correctly perceived a great danger, a deeply disturbing
source of harm to the new life of the German people ... The number
of Jews in western Europe is so small that they cannot exercise a
perceptible influence on the national mores. But across our eastern
border, year after year, out of the inexhaustible cradle of Poland,
streams a horde of ambitious, pantaloon-peddling youths, whose
children and children’s children will one day control the stock
markets and newspapers of Germany ... Let us not deceive
ourselves: the movement is very deep and strong, and a few jokes
about the sage pronouncements of Christian-social stump-speakers
will not suffice to get it under control. Up to and including the
best-educated circles of society, among men who would reject with
disgust any thought of religious intolerance or national arrogance,
the German people of today speak as if with one voice: the Jews
are our misfortune.
408
Stoecker, supported the historian. Censure came from Jewish
scholars and politicians like the philosopher Hermann Cohen and
the Reichstag delegate Ludwig Bamberger, but also from prominent
Gentiles. Crown Prince Frederick publicly expressed his outrage
over anti-Semitism several times in the years 1879–81. On 12
November 1879 liberal personalities, led by Max von Forckenbeck,
the mayor of Berlin and former president of the Reichstag, joined
together to protest that in many parts of Germany ‘racial hatred
and the fanaticism of the Middle Ages has awakened to new life
and is being directed against our Jewish fellow citizens.’
Treitschke’s name was not mentioned, but he, too, was implied in
the observation that Lessing’s legacy of tolerance between Jews
and Christians was being undermined by men ‘who should be
proclaiming from pulpit and lectern that our culture has overcome
the isolation of the tribe that once gave the world the worship of
the one God’.
One of the seventy-five signatories of the declaration was the
world-famous historian of antiquity Theodor Mommsen, who
shortly afterwards wrote a pamphlet aimed directly at his long-time
political comrade in arms. Jews of German nationality, wrote
Mommsen, were forced to conclude from Treitschke’s article that
the author ‘considered them second-class citizens, a kind of penal
battalion that might, at best, be somewhat reformed. That is
tantamount to preaching civil war.’ The ‘feeling of alienation and
inequality’ with which, even today, many a Christian German
confronts the Jewish German, contains a danger for both: ‘the civil
war of a majority against a minority, even only as a possibility, is a
national calamity.’
409
It was not only in Germany that anti-Semitism proliferated as a
consequence of the stock market crash and the economic crisis. It
also spread in Austria, Hungary, and France, especially in the
1880s, and in Russia, where anti-Jewish pogroms were frequent
throughout the same decade. But when it came to ‘modern anti-
Semitism’, the battle against emancipated Jewry, Germany was the
master and its neighbors the pupils. It was, to be sure, a very
studious master, who learned what the British natural scientist
Charles Darwin had to teach concerning the struggle for survival
and absorbed the lessons of the French anthropologist Count
Joseph Arthur de Gobineau on the inequality of the human races
and the superiority of the ‘Aryan’ race.
Bamberger, in his response to Treitschke, speculated that the
deeper cause of German anti-Jewish sentiment lay in a peculiar
interrelationship of attraction and repulsion between Germans and
Jews.
410
Bamberger detected another German-Jewish spiritual affinity in
the ‘cosmopolitan disposition, which is closely allied to the ability
to break with the established order’. What separated the two
peoples he characterized in the following way: ‘A thoughtful,
solemn, reverent, serious, obedient character contrasts markedly
with a strangely agile, sarcastic, sceptical, undisciplinable spirit.’
Heinrich Heine and Eduard Lasker were two figures who, according
to Bamberger, had combined the German and Jewish spirit in
completely different but equally impressive ways: here the
profound poet with a humour verging on frivolity, there the
brilliant parliamentarian, who fought for the cause of German
idealism ‘with the most brilliant weapons of Jewish dialectic’
against the great realist Bismarck.11
A half-century before, in the year 1834, Heine had speculated in
his Paris exile over what would happen if Germany should one day
proceed from contemplation to practical action, from the
revolution in the realm of thought to revolution in the empire of
reality. At the end of his Geschichte der Religion und Philosophie in
Deutschland,* Heine wrote that Christianity had somewhat
moderated the brutal Germanic lust for battle, but had not been
able to destroy it,
and when one day the taming talisman, the cross, should crack,
then the ferocity of the old warriors will surge forth again, the
mindless berserker rage of which the Nordic poets speak and sing
so much. That talisman has grown brittle, and the day will come
when it shall break wretchedly asunder. The old stone gods will
then arise out of their forgotten ruin and rub the dust of a thousand
411
years from their eyes, and Thor with his giant hammer will finally
leap forth and pulverize the Gothic cathedrals ... The thought
precedes the deed like the lightning the thunder. Nonetheless
German thunder, being German, is not very agile, and comes
rolling up somewhat slowly. But come it will, and when you hear it
crash like it has never yet crashed in all of world history, know
this: the thunder of the Germans has finally reached its goal. At this
sound, eagles will fall dead from the sky, and the lions in the desert
of Africa will crawl into their kingly caves with their tails between
their legs. In Germany, a drama will be enacted that will make the
French revolution seem like a harmless idyll.
412
The strength and sustainability of the remedy must correspond to
the magnitude and the tenaciousness of the evil ... Now that the
race has been thoroughly understood, we will set ourselves from
the beginning a more distant goal, the way to which cannot be
cleared by any but the strongest remedies. The Jews are—this will
always be the conclusion of those who know the race—an inner
Carthage, whose might the modern nations must destroy, lest they
themselves be destroyed by it in their moral and material
foundations.
During the same period, in the years 1878–81, the first anti-
Semitic organizations and parties arose in Berlin and Saxony,
among them the German Reform Party (Deutsche Reformpartei),
established in Dresden in September 1881. The anti-Semitic
Association of German Students (Verein Deutscher Studenten) was
founded the same year in Berlin. On 13 April 1881 the chancellor
received an ‘anti-Semites’ petition’ with 255,000 signatures
demanding the prohibition of further Jewish immigration, the
exclusion of the Jews from all higher political office and from the
faculty of the public schools, the restriction of the number of Jews
in higher education and the judiciary, and the reintroduction of
confessional statistics. Five years later, in January 1886, the
Antisemitische Correspondenz (published in Leipzig by Theodor
Fritsch) stated the movement’s ultimate goal: the ‘elimination of
the Jewish race from international public life’. Finally, in 1887, the
famous orientalist Paul de Lagarde (whose real name was Paul
Anton Bötticher), despite repeated denial of a belief in racial anti-
Semitism, referred to the Jews as ‘rampant vermin’ (wucherndes
413
Ungeziefer) that had to be crushed underfoot. ‘One does not
negotiate with trichinae and bacilli. Trichinae and bacilli also
cannot be educated and improved. They are destroyed, as quickly
and completely as possible.’
Physical destruction was the most radical conceivable solution
to the ‘Jewish question’. For the extremist anti-Semites of Germany,
the idea that there was no more room on earth for the Jews seemed
to flow naturally from their conviction that the Jews were at home
in the whole world and therefore belonged nowhere. In this view,
the Jews were either not a nation or a nation within the nation.
Either way, they represented a foreign body. The Germans, for
their part, had left behind their own ‘cosmopolitanism’ with the
founding of their national state. Many of them thought the Jews
either unwilling or unable to do the same.
A more probable explanation is that the Germans, whom
Helmuth Plessner has called a ‘belated nation’, were profoundly
insecure about their national identity after 1871. Once the external
‘arch-enemy’ had been overcome, many were tempted to create an
internal arch-enemy who could help answer the question of what
was ‘German’ and what ‘un-German’. ‘International Jewry’ was
especially apt for this role. It could be connected to practically
everything by which Germans felt themselves to be threatened—to
international finance capital as well as to international socialism.
Since Protestants and Catholics, believing and non-believing
Christians could agree on this view of the Jews, the latter seemed
to afford an even better internal enemy—and thus support for
national cohesion— than ultramontanist Rome.
Only a small minority of Germans hearkened to the diatribes of
414
the radical anti-Semites after the mid-1870s. But those who fought
against anti-Semitism were a minority, too. The majority was
evidently not greatly disturbed by the activities of the anti-Semites
and not free of anti-Jewish prejudice. Bamberger declared in 1880
that ‘the people generally think in more human and less prejudicial
terms than a few academics.’ But he also felt that Treitschke had
done a service to the Jews
by making many of them whom the last decades had lulled into
illusion once again aware of how things really stand. He who
labours under self-deception always labours wrongly, and when he
discovers the truth, he falls into dismay, if not despair. It is better
for the Jews to acknowledge the feelings of antipathy hidden under
the outer reflexes of politeness.12
415
appoint new, regime-friendly members to the House of Lords.
On the federal level, the laws for the standardization of currency
and coinage, passed between 1871 and 1875, bore the mark of the
National Liberals. The Kulturkampf legislation was also ‘national-
liberal’ in inspiration, though in a completely different way. So
important were these measures to the largest party that it was
willing to assure Bismarck’s cooperation by accommodating the
chancellor in a different, highly controversial area—the military
budget. In 1871 the peacetime strength of the army was set at
401,000 troops for three years. In 1874 Bismarck sought a military
budget in perpetuum, the so-called Äternat. Since this would have
removed four-fifths of the federal budget from the control of the
Reichstag, the National Liberals could not vote for it without
renouncing the core of their political identity. The result of a long
process of conflict and negotiation was a compromise worked out
between Bismarck and Bennigsen, the Septennat, a seven-year
military budget. Since the legislative period lasted three years, only
every second or third Reichstag would be able to exercise its full
budgetary authority.
In the second half of 1877, it seemed for a time as though the
chancellor was going to accept a parliamentary governance of sorts
in Prussia and the Reich. He offered Bennigsen a ministerial
position in the Prussian government and, at the same time, the
office of an undersecretary to the imperial chancellor; in effect, he
proposed making Bennigsen his representative in both
governments. According to Article IX of the imperial constitution,
simultaneous membership in the Federal Council and the
parliament was prohibited; as a Prussian minister, Bennigsen would
416
have been compelled to give up his seat as a Reichstag delegate. If
the leader of the National Liberals, alone of his party, had joined
the government, he would have risked isolation, if not a total loss
of power. In order to avoid this danger, he proposed to Bismarck
that two further National Liberals, Max von Forckenbeck and the
Swabian-Bavarian landowner Franz Schenk von Stauffenberg (both
members of the party left) also receive governmental offices. If this
had happened, Bennigsen’s ‘ministerial candidacy’ would have
been transformed into a step towards a real parliamentary
monarchy. And that, of course, was precisely what Bismarck
wished to avoid.
His offer to Bennigsen had a rather different motivation. In 1875
Bismarck had decided to abandon economic liberalism and free
trade for the sake of a protectionist tariff policy. This change of
course was, for him, not a matter of doctrine, but primarily a
question of financial pragmatism. At that juncture, ‘matricular’
contributions from the individual states were the main source of
federal funds, making the Reich financially dependent on its
members. Protective tariffs, state monopolies (for example in
tobacco), indirect taxes (for example on beer, liquor, and coffee),
and a tax rise on tobacco would have ended or reduced this
dependence, and thus they seemed to Bismarck a good means to
reach this goal. Most National Liberals, however, fundamentally
rejected state-run monopolies and protective tariffs, and they were
prepared to approve new federal taxes only if the Reichstag were
granted the same yearly budgetary competence over a part of the
revenue it already possessed with respect to the allocations from
the individual states.
417
Thus, it was inconceivable that the whole of the National Liberal
party would support Bismarck’s economic about-face. But the
chancellor did not expect them to do so. Rather, his proposal to
Bennigsen was calculated to split the party and bring about the
formation of a new parliamentary majority, including the
conservatives, who, after their re-establishment as the German
Conservative party in 1876, had once again been aiming at a role
in government. In the event, Bennigsen was not willing to go along
with Bismarck’s scheme. For his part, the chancellor did not
seriously wish to have the leader of the National Liberals as a
minister. Thus Bennigsen’s ‘ministerial candidacy’ remained an
episode. Parliamentary rule never had a chance in Bismarck’s
Germany.
Bismarck’s 1877 initiative was not least an attempt at a pre-
emptive strike, with which the chancellor sought to prevent the left
wing of the National Liberal party (which he continually claimed
was under the control of Bamberger and Lasker) from gaining more
influence under the successor of the aged emperor. Indeed, the
political understanding between Crown Prince Frederick and the
resolute liberals was very much in evidence. This was one of the
main reasons Bismarck favoured any step that could bring the
conservatives closer to power. Moreover, when Pope Pius IX died
in February 1878 and was succeeded by the relatively ‘realistic’
Cardinal Pecci as Leo XIII, the opportunity presented itself for a
reconciliation between the Reich and the Catholic Church and
therewith the chance, sooner or later, to gain the support of the
Centre party for the government.
The conservatives and the Centre were opponents of economic
418
liberalism, above all free choice of trade or industry. In the middle
of the 1870s, the grain-exporting manorial lords of East Elbia were
not yet demanding the end of free trade. The Association of Tax
and Economic Reformers (Vereinigung der Steuerund
Wirtschaftsreformer), founded in 1876 and closely allied to the
German Conservative party, pronounced itself basically in support
of free trade and against protective tariffs, though it considered the
question of entrance tariffs and consumer taxes open to debate. The
most adamant supporters of protective tariffs sat in the rows of the
Free Conservatives and in the right wing of the National Liberals.
Their strongest backing came from the Rhenish-Westphalian iron
and steel industry and the southern German cotton industry—two
branches that felt themselves particularly threatened by the British
competition. In 1876 they played the decisive role in the founding
of the first leading association of German entrepreneurs, the
Central Association of German Industrialists (Centralverband
Deutscher Industrieller), which campaigned openly for protective
tariffs.
In February 1878 Bismarck presented to the Reichstag a bill to
raise the tax on tobacco. In exchange for their support, the National
Liberals demanded guarantees in law for the budgetary rights of
the federal parliament and the Prussian House of Representatives.
This was a price Bismarck was unwilling to pay. Moreover, he
prevented all possibility of compromise by stating to the Reichstag
on 22 February that he considered the tobacco tax merely a stop
along the way to state monopoly on tobacco. That destroyed the
bill. Bennigsen took the chancellor’s speech as an opportunity to
declare the negotiations over his joining the government officially
419
at an end (for his part, Bismarck considered them long since
closed). Shortly thereafter, the two last representatives of German
‘portfolio liberalism,’ Minister of Trade von Achenbach and Finance
Minister von Camphausen, stepped down, just as the president of
the imperial chancellery, Rudolf von Delbrück, had done in June
1876, when it became clear to him that Bismarck was turning away
from economic liberalism. ‘The system shift is becoming ever more
evident’—the balance drawn by Grand Duke Frederick I of Baden
in a private letter from the beginning of April 1878 represented the
view of most well-informed observers of the empire’s inner crisis.13
Even if it had not been for the two attempts on the life of
Emperor William, which took place in rapid succession in spring
1878, the open break between Bismarck and liberalism would still
have occurred. Nonetheless, the attacks fundamentally changed the
political situation in Germany and gave the chancellor an earlier
opportunity to revise the distribution of parliamentary power more
to his taste. On 11 May 1878 Max Hödel, a plumber’s journeyman,
fired at the emperor on Unter den Linden street without hitting
him; on 2 June, very near the first assassination attempt, William
was shot and seriously wounded by Dr Karl Eduard Nobiling, an
economist by profession. Hödel had earlier belonged to the
Socialist Workers’ Party (founded in Gotha in May 1875 as
successor to the hitherto separated Social Democratic and
Lassallean parties), though he had been excluded for embezzling
party funds. In Nobiling’s case, no connection with the socialists
could be proved. Bismarck, however, was not interested in legally
compelling evidence of a social-democratic motivation for the
assassination attempts; he was resolved to use the attacks as a
420
pretext for settling accounts with both socialism and liberalism.
The first draft of an anti-socialist law, presented to the
parliament on 20 May, a few days after Hödel’s attack, did not
outlaw the party, but made provisions for the possibility of banning
Social Democratic associations, gatherings, and printed material.
The great majority of the deputies rejected the bill, including most
of the National Liberals. On 11 June, after the second attack,
Bismarck dissolved the Reichstag—an act aimed primarily at the
National Liberals, to whom the Bismarck-inspired press ascribed a
‘moral complicity’ in the assassination attempts. Immediately
following the first attack, the Norddeutsche Allgemeine Zeitung,
Bismarck’s house organ, claimed that social democracy had
accustomed itself ‘to settling into the nests prepared for it by
liberalism’. After the second attempt on the emperor’s life, the
same paper accused the party of ‘forcing the second part of its
name into the foreground, at the expense of the first’, and
condemned it for having deserted the nation.
The Kreuz-Zeitung went much further, calling socialism the
‘logical development of liberalism’ and making, once again,
‘modern Jewry’ responsible for liberalism’s errors. ‘“Jewification” is
making great strides, and it is being pushed along by liberalism ...
The real enemies of the Reich are those who have taken from our
people their firm belief and firm support in the eternal Word of
God, thus destroying the foundations of a healthy national life.’
Several days later, the conservative paper declared that the German
people were becoming ever more dependent on money men, and
that these were, unfortunately, mostly Jews. ‘Liberalism does harm
to our people, both spiritually and materially ... And it cannot be
421
denied that, because of this, the national power of Germany suffers
harm.’
The appeal to the Germans’ conservative and national instincts
was successful. The Conservative and Free Conservative parties
emerged victorious from the federal elections on 30 July 1878. The
National Liberals, the Progressives, and the Social Democrats were
the losers. The Centre party managed to hold its own. In the wake
of their defeat (they lost 29 of their former 128 mandates), the
National Liberals gradually swung around in the very direction
Bismarck wished to see them go. The majority under Bennigsen
was now basically prepared to vote for the government’s draft of a
‘Law against the Dangerous Aspirations of Social-Democracy’. The
leftist minority around Lasker initially intended to persevere in the
rejection of an unlawful emergency measure, but then, after it had
pushed though several revisions and a restriction of the law to two
and a half years, decided to concur. On 18 October 1878 the
Reichstag passed the anti-socialist law (the so-called
Sozialistengesetz) with 221 against 149 votes from the ranks of the
Centre, the Progressive, and the Social Democratic parties.
The prohibition of socialist associations, assemblies, and press;
the expulsion of socialist agitators; the possibility of declaring a
year-long ‘minor state of siege’ in ‘endangered’ districts—all these
suppressive measures were rationalized with the unproved and
unprovable assertion that the emperor’s attackers in May and June
1878 were incited to their deeds by ‘social democracy’. The anti-
socialist law was an emergency measure, directed against
particular ideologies, not only against clearly defined acts. It
violated elementary principles of the rule of law, such as liberals
422
interpreted it. That the National Liberals nonetheless voted for it
marks a turning point in the history of German liberalism, a partial
capitulation before the power embodied by Bismarck.
For Social Democrats the law, continually extended until 1890,
meant danger and persecution. In the twelve years it was in effect,
some 1,300 writings in periodicals and elsewhere and 332 workers’
organizations, including unions, were banned. Approximately 900
individuals were expelled from ‘siege areas’, more than 500 of
whom were breadwinners of families. Some 1,500 persons were
imprisoned, the prison sentences reaching a combined total of
about 1,000 years.
Yet Bismarck did not manage to destroy social democracy. In
fact, his efforts contributed to the spreading of the social-
democratic workers’ movement in cultural form, as prohibited
organizations were replaced by workers’ sporting associations, glee
clubs, voluntary aid collections, and other similar groups. A further
paradoxical effect was a certain ‘parliamentarization’ of the party:
since participation in elections, agitation in the parliaments, and
the corresponding reportage were not banned, the actual party
leadership migrated from the party executive to the Reichstag
fraction. The new party organ, the Sozialdemokrat, as well as other
social-democratic writings, were published in Switzerland, whence
they were illegally brought into Germany and distributed via the
‘red army postal service’. And in Wyden, Switzerland, the Socialist
Workers’ party held a party congress in 1880, at which the 1875
Gotha platform was modified. Instead of aspiring toward its goals
‘by all legal means’, the statement now read ‘by all means’. This
was not a affirmation of illegality as a principle, but simply a
423
recognition of the fact that, under the anti-socialist law, nearly
everything the Socialist Workers’ party did was illegal.
During the anti-socialist years, there developed among the
members and supporters of the Social Democrats a consciousness of
living in a political ghetto. This experience helped transform
socialism into a kind of religion of secular redemption.
Consequently, social-democratic workers usually took from the
writings of Marx and Engels only as much as served their need to
compensate for a repressive reality with the belief in a bright
future. The world-view formed during those years was broadly
‘deterministic’ in nature: the conviction that the historical process,
by force of nature, would lead to the overcoming of capitalism by
socialism. Before the classless society could be realized, however, it
would be necessary to wage the class war of the proletariat and to
develop an ever clearer class consciousness. The anti-socialist law
demonstrated the reality of class society, the class state, and class
justice. It effected one thing above all: a broad absorption of basic
Marxist assumptions.14
With the anti-socialist law commenced the ‘system shift’
predicted by the grand duke of Baden, which contemporaries were
already referring to as an ‘inner founding of the Empire’ (innere
Reichsgründung). Bismarck had never expected the suppression of
social democracy to solve the social question. Already in 1871, he
had told the Prussian minister of commerce, Heinrich Friedrich
August von Itzenplitz, that ‘the action of the currently ruling state
authority’ was, in his view, the only means ‘to put a stop to the
socialist movement in its present aberration and to lead the same
along more salutary paths, notably by realizing what seems
424
justifiable in the socialist demands and what can be accommodated
in the framework of the present-day order of state and society.’
This was another reason Bismarck regarded the break with
economic liberalism as unavoidable. If he wished to make—
precisely as did Lorenz von Stein—the Hohenzollern monarchy into
a ‘monarchy of social reform’, he might depend upon the support of
conservatives who were open-minded about social policy, but not
upon devotees of the pure doctrine of ‘Manchester liberalism’.
Initially, however, everything depended on completing the
paradigm shift in finance and trade policy, so as to provide the
Reich with the material basis without which an active social policy
was inconceivable. The elections of 1878 had brought Bismarck
significantly closer to this goal. In the autumn, the ‘tariff
protectionists’, at the instigation of the Central Association of
German Industrialists, joined together in an inter-fractional
‘National Economic Association’ (Volkswirtschaftliche Vereinigung).
Of the total (397), 205 delegates participated, including 27
National Liberals in addition to Conservatives, Free Conservatives,
and almost all the members of the Centre.
Before the Reichstag could have its say, however, Bismarck still
had to overcome strong resistance to his tariff plans in the Federal
Council. The southern German states were against all protective
tariffs; Saxony desired duties on iron and textiles, but none on
grain. Bismarck, who sought to establish a lasting connection
between western heavy industry and the East Elbian manorial
system, insisted upon a package deal: no industrial tariffs without
agricultural tariffs. At the end of March 1879, the Council agreed
to a compromise, accepting the iron and textile tariffs at the level
425
the chancellor wished, but pushing through considerably lower
duties on grain than Bismarck’s bill envisioned.
In the Reichstag, too, the industrial protectionists were far more
numerous than the agrarian. During the second reading of the tariff
bill, the attempt by several Conservatives to raise the duties on rye
from those decided by the Council failed. But there was another
way of obtaining a majority for Bismarck’s original tariff project:
by linking the question of tariffs to the reform of state finances.
This new ‘package’ brought to prominence a party that had bitterly
fought against the chancellor until shortly beforehand—the Centre.
As a federalist party, the Centre wanted to maintain the
‘matricular’ contributions of the individual states; as a
constitutional party, it wanted to preserve the parliament’s right to
approve the budget. A Bavarian delegate, Georg von Franckenstein,
drafted a motion uniting both goals. The ‘Franckenstein clause’
limited state revenues from tariffs and the tobacco tax to 130
million marks annually; any surplus was to be distributed to the
federal states. Since the financial needs of the Reich would not be
fully covered by this measure, it would remain dependent on
allocations from the states, which now, however, were to be
financed by the funds transferred from the federal government.
Since surpluses would remain, the states, too, would profit from the
protective tariffs and higher tobacco tax. The Reichstag retained
the right to approve—together with the Federal Council—the
amount of the ‘matricular’ contributions, thus preserving its
budgetary authority.
In agreeing to Franckenstein’s motion (and rejecting a second
compromise by Bennigsen placing greater emphasis on the
426
budgetary rights of the parliament), Bismarck assured himself of
the Centre party’s compliance on higher agricultural duties. On 12
July 1879 the Reichstag formally approved the protective tariffs
and the increase in the tobacco tax with the votes of the two
conservative parties, the Centre, and sixteen National Liberal
delegates. This vote was an important turning point. In accepting
the ‘Franckenstein clause’, Bismarck ended his cooperation with the
National Liberals, begun twelve years previously, and took a sharp
turn to the right, in favour of a party constellation in which a
National Liberal party might find a place only if it, too, shifted
rightwards and ‘liberated’ itself from its left wing.
The economic and societal consequences of the shift to
protective tariffs in summer 1879 were even more momentous. The
East Elbian landowners, now shielded from international
competition (more as a result of Bismarck’s exertions on their
behalf than of their own), found themselves in a singular position,
which they exploited to the full: they could lead lives of privilege
at the expense of society and, at the same time, fully maintain their
societal influence and dominance. Within industry, the
protectionist policy benefited the old branches, above all coal and
steel, whose long-term competitiveness was open to doubt. The
new growth sectors—electricity, chemistry, and mechanics—paid
the price. The relation between ‘manor and blast furnace’
developed into the conservative axis of German politics, an alliance
against liberalism and democracy that, despite its inner
contradictions and frequent conflicts, always managed to hold
together in mutual defence against mutual enemies.15
The change from free trade to protectionism was accompanied
427
by a transformation in German nationalism. The opponents of free
trade spoke of the ‘defence of national labour’, employing both a
social and a nationalist argument—in this case the preservation of
German jobs—in order to isolate the German economy from the
effects of international competition. Before 1871, to be ‘national’
generally meant to be ‘anti-feudal’, both in a bourgeois-liberal and
in a proletarian-socialist sense. In the decade after the founding of
the Reich, ‘national’ and ‘anti-international’ began to mean the
same thing, transforming an originally liberal and leftist expression
into a battle cry of the political right.
‘National’ and ‘liberal’ were a contradictio in adjecto, an
oxymoron—thus remarked the formerly liberal Grenzboten
(financed from 1879 through Bismarck’s ‘reptile fund’ and
published by his intimate associate, Moritz Busch) on the crisis
occasioned within the National Liberal party by the shift in
economic policy:
That the advocates of free trade did not belong to the ‘national’
camp clearly emerged from the pages of the Grenzboten shortly
thereafter. ‘The Manchester radicalism is like the ultramontane and
the social-antinational. Its mania is the cosmopolitan free-trade
society, the atomistic world-fog with its nucleus of a sort in the
428
might of English capital, which preserves it from complete
evaporation.’
After the transformation on the domestic political front in 1878–
9, to be ‘national’ no longer meant to promote the emancipation of
the bourgeoisie or the workers. Rather, it meant to defend the
existing order against everybody engaged in the struggle for greater
societal openness, greater liberty, and greater equality.
Consequently, ‘national’ discourse, such as it was understood by
the right, involved aggressive attacks against those who were not
considered ‘national’ in this sense. The suspicion of ‘national
unreliability’, which the liberals had expressed against Catholics
and socialists, was now increasingly turned against them. Anti-
Semitism, which was always also a form of anti-liberalism, had
prepared the way for this anti-liberal nationalism. Nonetheless, one
did not have to be an anti-Semite to be ‘national’ in the new, post-
1879 meaning of the term.
This semantic shift was also accompanied by a transformation in
the identity of those who considered themselves ‘national’. Ever
since the Prussian conservatives had renamed themselves the
German Conservative party in 1876, they claimed to be the
authentic representatives of the national idea. At the same time,
they began increasingly to court the central groups of the urban
middle classes, tradesmen and small merchants, who had generally
voted liberal up to that time, but were now feeling more and more
threatened by liberalism in the shape of freedom of trade and free
movement of labour. These were the same groups to whom
organized anti-Semites sought to appeal after 1880. In 1888
Ludwig Bamberger subjected the rightward re-orientation of
429
nationalist discourse to an evocative and biting analysis:
The national banner in the hands of the Prussian ultras and the
Saxon guildsmen is the caricature of what it once symbolized, and
this caricature has come about quite simply in that the defeated
adversaries have donned the discarded garments of the victors and
turned them out in their own fashion, freshening up the colours
and tailoring them to fit, so as to be able to parade around in them
as the cheery heirs of the national movement.
430
democracy. They paid the price in 1879, when they had become
too weak to mount effective resistance to Bismarck’s swing to the
right.
In the course of the 1860s, German liberalism had liberalized
itself in many respects, for example with regard to the civil
equality of the Jews and the achievement of freedom of trade. The
following decade saw a reversal of this process, the de-
liberalization of liberalism. In the course of the campaigns for the
‘defence of national labour’ and a stringent anti-socialist law, press
organs close to the National Liberal right began to call for the
acquisition of colonies. The arguments used were not only
economic, but also ‘social’. In the summer of 1878, the Grenzboten
wrote that the state that considers domestic tranquillity its primary
civic duty gains a great deal ‘when unquiet spirits, who do no good
at home, leave the country ... A system, similar to the one used to
excellent effect with the deported Australians, seems to us the most
appropriate when, one day, the question presents itself to the
German Reich of getting rid of its wayward sons.’ The
Volkswirtschaftliche Correspondenz, the mouthpiece of the
protectionists, summarized the same idea in 1878 in the crisp
assessment: ‘As a safety valve for the rumbling volcano of the social
question, no country on earth is in such need of a nationally
organized emigration system as Germany.’
The National-Zeitung, at that time still oriented towards free
trade, objected: more dependable and useful than enticing pictures
of ‘flourishing factories in Africa and America’ were ‘enterprises
bringing the colonial chisel to bear on the periphery of our own
national territory’, so as to make (predominantly Polish) Posen into
431
‘German soil’. But this argument did not convince the protectionists
and advocates of colonialism, who saw no essential conflict
between the ‘Germanization of Posen’—which the National-Zeitung
claimed would be a blessing for both Germans and Poles—and the
acquisition of overseas colonies. In spring 1879 Treitschke’s
Preussische Jahrbücher demanded that the government use direct
measures to encourage German farmers to settle in Posen. ‘In the
west, it is imperative that we protect and secure the state, but we
do not need immigration [wir haben ... aber nicht volklich zu
erwerben]. The east, on the other hand, has always been the land of
promise for us, for as long as there has been German history.’
Treitschke and his friend Wilhelm Wehrenpfennig, co-editor of
the Preussische Jahrbücher, had left the National Liberal party the
day before the vote over the tariff and finance reform. The next
day, 12 July 1879, sixteen further protectionists followed their
example. The staunch advocates of free trade hesitated to take the
same step, but increasingly lost their confidence in Bennigsen’s
ability or desire to bring the party back to a resolute liberal
platform after he had gone so far to accommodate Bismarck on the
question of protective tariffs. On 15 March 1880 the first ‘leftist,’
Eduard Lasker, declared his resignation from the National Liberal
parliamentary fraction. In a letter to his constituents, he criticized
the assumption ‘that a reactionary economic and tax policy, so out
of step with the times, can be combined with a progressive policy
in other areas’. He believed the parliamentary tactics of the party
were guided by the intention of
432
conservative majority and instead working towards, at best, a
liberal-conservative majority. I, for one, consider such a
combination impossible, except at the price of sacrifices that would
damage a moderately liberal party in the present and endanger it
in the future.
433
by chance: most of those who now defied Bismarck had already
done so once before, in the years of the Prussian constitutional
conflict.16
434
third mitigating laws entered into effect, permitting a large number
of exceptions to the requirement of a ‘cultural exam’, and allowing
the king to pardon bishops who had been removed from their
posts. The so-called ‘peace laws’ of 1886 and 1887, which Bismarck
negotiated directly with the curia without any participation from
the Centre, formed the conclusion of the revision. The state
recognized the disciplinary authority of the pope, abolished the
cultural exam, and rehabilitated all religious orders with the
exception of the Jesuits. Three ‘achievements’ of the Kulturkampf
were preserved: the Jesuit ban (abolished in 1917), the pulpit
paragraph (abolished in 1953), and obligatory civil marriage.
The wounds left by the Kulturkampf could not be healed as
quickly as its laws could be changed or abolished. Most faithful
Catholics retained profound feelings of discrimination and
humiliation. This was one of the many reasons political
Catholicism, unlike Protestantism, remained a solidly unified force
even beyond the end of the imperial era. The dense network of
Catholic associations for peasants, workers, journeymen, craftsmen,
merchants, teachers, academics, and students, which together
formed the Catholic ‘milieu’, was a product of the Kulturkampf.
Thus, in its effects, the anti-Catholic legislation was much like the
anti-socialist law; pressure from the state promoted the formation
of a societal ‘subculture’ that sought to seal itself off from its
environment.
Nonetheless, there were considerable differences in the degree
and duration of state pressure. Unlike Social Democrats under the
anti-socialist law, Catholics were able to publish and engage in
politics relatively unhindered during the Kulturkampf. Anti-Catholic
435
discrimination was far less comprehensive than the persecution of
social democracy. The transformation of political Catholicism into
part of the governmental establishment began about a decade after
the official end of the anti-Catholic era. The Social Democrats, on
the other hand, continued to think of themselves as a fundamental
opposition, and continued to be perceived and treated as
fundamental adversaries of state and society, well after the
expiration of the anti-socialist law in 1890.17
In the first Reichstag elections after the 1878–9 transformation
of domestic policy, which were held in October 1881, the Social
Democrats suffered modest losses, from 7.6% of the previous vote
in 1878 down to 6.1%. With 23%, the Centre obtained almost
exactly the same amount of support as before. The biggest losers
were the National Liberals, who fell from 23.1% to 14.7%, the
biggest winners the left-liberals. The secessionists of the Liberal
Union straightway received 8.4%; the German Progressive party
was nearly able to double its mandate, climbing from 6.7% to
12.7%. Although the conservative camp, as a whole, declined in
influence, the losses were confined to the Free Conservatives
(called the Deutsche Reichspartei, German Empire Party, from 1871),
who fell from 13.6% to 7.4%; the German Conservative party
climbed from 13% to 16.3%.
Taken as a whole, the election could be interpreted as a
referendum against Bismarck’s swing to the right, since the
chancellor lost his governmental majority. Now, a bill could pass
the parliamentary hurdle only when either the Centre or the two
left-liberal parties voted with Conservatives, Free Conservatives,
and National Liberals. In this assembly, there was no majority for
436
the tobacco monopoly Bismarck had placed at the centre of the
electoral campaign.
The two left-liberal parties had so much in common that their
reunification seemed inevitable. It duly happened in March 1884:
the German Progressive Party and the Liberal Union united under
the leadership of the Berlin jurist Eugen Richter, head of the
Progressive party, forming the German Liberal party (Deutsche
Freisinnige Partei). With 110 seats, the new party formed the
strongest fraction for the time being. In order to gain as much
National Liberal support as possible, it opted for a deliberately
‘soft’ platform in the controversy over free trade and protectionism,
speaking out against a ‘tariff and economic policy in the service of
special interests’. The demand for legislation to create a responsible
governmental ministry, on the other hand, was fully clear:
socialism was to be fought ‘also in the form of state socialism’.
Around this same time, the southern German National Liberals
were turning towards what left-liberals termed ‘state socialism’. In
their ‘Heidelberg declaration’ of the end of March 1884, drafted by
the Frankfurt mayor Johannes Miquel, they expressly approved ‘the
aspirations of the Reich Chancellor towards heightened solicitude
for the well-being of the working class’ and the efforts of the
government ‘to improve the social situation of the working class’.
The rejection of Manchester liberalism was flanked by a statement
of support for the customs tariff of 1879, protection of agriculture,
renewal of the anti-socialist law, and the maintenance of a strong
German armed force. In May 1884 a National Liberal party
congress made the ‘Heidelberg declaration’ the official platform of
the entire party.
437
What the Liberals fought and the National Liberals welcomed
was, by this time, already in full swing—the erection of a German
system of social insurance. Bismarck had not broken with economic
liberalism for reasons of social policy, but without this break, he
would never have gone down in history as a socio-political
innovator. The ground for policies making it the task of the state to
mediate social conflicts had been prepared not only by
theoreticians like Lorenz von Stein and the so-called ‘lectern’ or
‘professorial socialists’ (Kathedersozialisten) around Gustav
Schmoller who founded the Association for Social Policy (Verein für
Sozialpolitik) in 1872, but also by close associates of Bismarck like
Theodor Lohmann. The chancellor himself was convinced of the
rectitude and necessity of the task. ‘It is possible that our policies
will one day perish, when I am no longer around,’ he said to the
writer Moritz Busch on 26 June 1881. ‘But we will push state
socialism through. Everybody who takes up this thought again will
have his chance at the helm.’
At first, Bismarck wanted government allocations to completely
replace insurance contributions by the workers, expecting that such
a programme would make workers loyal to the state. The anti-
parliamentary ulterior motives he pursued with this policy cannot
be denied. In the trade associations managing accident insurance
he saw the nucleus of a profession-based organization that,
according to his plan, was to culminate in a
Reichsvolkswirtschaftsrat, a corporative chamber competing with the
Reichstag. This plan was never realized, hampered partly by the
bureaucracy, partly by the Reichstag.
The reform that actually took place in the 1880s was far more
438
progressive than what Bismarck had conceived. The health
insurance law of 1883 obligated employees not belonging to a
voluntary insurance association to join a local provider. They were
responsible for two-thirds, employers for one-third of the costs. The
costs of accident insurance, created in 1884, were borne entirely by
the companies, organized into associations. The 1889 old-age and
disability insurance law divided the contributions evenly among
employers, employees, and the federal government.
What distinguished these social insurance programmes from
traditional organizations caring for the poor was the fact that they
gave the individual worker a lawful claim to social services. The
legislation of 1883, 1884, and 1889 laid state and society under the
obligation of providing assistance in emergency situations for
which the individual was not responsible and which he could not
prevent. These laws made Germany into a pioneer in social policy.
It achieved this status in the course of an inward political shift that
threw the country back in other areas. Protectionist tariffs and
progressive social legislation—these were two sides of the same
coin, the ‘inner founding of the Reich’.18
Bismarck did not consider it promising to attempt to win a
governmental majority on the strength of social policy alone. The
looming federal elections influenced another major policy
development in 1884: in this year, Germany became a colonial
power. From 1882, colonial acquisitions were on the agenda of the
newly founded German Colonial Association (Deutscher
Kolonialverein), in which, along with industrialists, merchants, and
bankers, politicians from the ranks of the National Liberals and the
Free Conservative German Empire party set the tone. These men
439
were not only interested in obtaining new markets for German
products, but also in the integrating effect the colonial movement
would have within Germany itself. At the association’s founding
congress in Frankfurt in December 1882, Miquel described colonies
as a task ‘at which the conflict of the confessions, the religious,
political, and social antagonisms, do not gnaw’. The president of
the association, the Free Conservative Reichstag delegate Hermann
von Hohenlohe-Langenburg, declared at the founding of the Berlin
branch in 1884 that ‘no aspiration is better suited to overcome the
social question, presently exercising all spirits, than the aspirations
of our association. Everyone who joins is doing his bit for the
resolution of the social question.’
Until well into the 1880s, Bismarck wanted nothing to do with
German colonies, for reasons of foreign policy as well as of cost. In
April 1884, however, he took advantage of British problems with
France and Russia, which temporarily distracted the maritime
power, to place a territory acquired by the Bremen merchant Adolf
Lüderitz on the bay of Angra Pequena in south-west Africa under
the protection of the Reich. In July 1884 Germany proclaimed its
sovereignty over Togo and Cameroon. In 1885 followed the
declaration of protective sovereignty over a large east African
territory acquired by the colonial agitator Carl Peters and the
founding of the colony of German New Guinea.
The chancellor became a momentary ally of the colonial
movement in 1884–5 because it seemed to have found a means of
giving Germany a national goal once again. If his countrymen
occupied themselves with a national enterprise like colonies, it
would be easier for Bismarck to keep them away from activities he
440
considered irreconcilable with the interests of the Reich. These
included the overthrow of the existing order, which the Social
Democrats were generally thought to be planning; regional and
confessional particularism, still very much alive; and, not least, the
pressure of the Liberals for parliamentary rule. The demand for
colonies, on the other hand, came from parties that assured
Bismarck of their support—the Free Conservatives and the National
Liberals. Accordingly, the strengthening of these parties lay in the
interest of the Reich, such as Bismarck interpreted it.
Their firm negation of colonial policy did the left-liberals as
little good as their rejection of the social insurance legislation. In
the 1884 Reichstag elections, the German Liberal party lost 39 of
their 106 seats. Yet it would be incorrect to speak of a resounding
victory for the parties supporting Bismarck in these areas. The Free
Conservatives held on to their 28 seats, but suffered minor losses in
their percentage of the vote; the National Liberals gained a further
2.9% and 4 seats. The German Conservatives profited from the fact
that far fewer votes were necessary to obtain a parliamentary
mandate in sparsely settled, agricultural East Elbia than in the
densely populated west. They climbed from 50 to 78 seats, even
though they lost approximately 1% of the total vote. Among the
clear winners were the Social Democrats, who grew from 6.1% to
9.7% and were thus able to double their number of seats from 12
to 24. The Centre party, Guelfs, Alsace-Lorraine independents,
Poles, and Danes experienced no significant changes. Since the new
Reichstag was hardly any closer to a government-friendly majority
than its predecessor, rule with ad hoc majorities continued.
If colonial policy was elevated to a national mission in the years
441
1884–5, the year 1886 saw a new focus of national attention: the
promotion of Deutschtum in the eastern part of the Reich. The
policy’s deeper roots were economic and social. The lack of
agricultural jobs caused a continuous decrease in the rural
population of Prussia’s eastern provinces; many emigrated abroad,
many migrated to the ‘industrial heart’ of Germany, the Ruhr area.
Since more Germans left than Poles, the ratio of Poles to Germans
increased in the home territories, especially in Posen and West
Prussia. The resulting fear of a ‘Polonization’ of the east led, in
1886, to renewed agitation for the ‘Germanization of the soil’. The
fact that Polish ownership of land had significantly declined in the
years prior to the campaign was deliberately ignored by its authors,
the foremost of whom was Christoph von Tiedemann, president of
the Bromberg government. In spring 1886, bolstered by a
memorandum from Tiedemann, the Prussian government drafted a
bill proposing the purchase of Polish-owned land through a fund of
100 million marks under the control of the Royal Settlement
Commission. The law went into effect on 26 April 1886.
Bismarck’s main goal was to weaken the Polish nobility
permanently. He initially weighed the thought of making the
purchased territory into domains. It was the National Liberals who
proposed giving the land over to peasant settlement, an idea
Bismarck quickly appropriated, thus helping to make the
‘Germanization of the soil’ into an official part of Prussian policy.
Between 1886 and 1914, Prussia spent about 1 billion marks in
gold bullion for the settlement of German peasants. Considering the
expectations, the time duration, and the size of the territory in
question, the practical results of the land-settlement law were
442
modest. Despite massive propaganda from the German Eastern
Marches Association (founded in 1894), only about 22,000
peasants—including their families some 120,000 Germans—could
be settled in Posen and West Prussia by 1914.
The ‘Germanization of the soil’ was accompanied by the
linguistic Germanization of the east. In the mid-1870s two pieces of
legislation (Prussia’s Geschäftssprachengesetz of 1876 and the federal
Gerichtsverfassungsgesetz of 1877) had made German into the
official language of public administration and law. ‘In the ensuing
decades, this wave of linguistic and national standardization made
its way into the most remote legal district,’ writes Hans-Ulrich
Wehler.
The laws of 1876 and 1877 not only affected the Poles, but also the
Danes of northern Schleswig. The 1877 federal law also affected
the French-speaking inhabitants of the territory around Metz. Only
in the east, however, could one speak of a ‘war between peoples’, a
Volkstumskampf between ‘Teutons’ and ‘Slavs’. Germans did not
bring to bear against Danes and Frenchmen the cultural superiority
they claimed to have over the Poles. The German relationship to
Poland was infused with racial prejudice. In this, it was more akin
443
to anti-Semitism than ‘normal’ expressions of nationalism. At the
same time, however, nobody would have accused the Poles, as they
did the Jews, of seeking to control Germany, much less to control
the world. Only anti-Semitism contained an admixture of an
inferiority complex. In this way, the German hatred of the Jews
was again different from ‘normal’ racism, such as that expressed
against the Poles.
The discrimination against the Alsace-Lorrainers was mild in
comparison. Some 60,000 of them, availing themselves of the right
of ‘option’, left their homes for France in the first two years of the
new German Reich. Those who stayed had fewer civil rights than
other Germans. At first, Alsace-Lorraine was administered as an
‘imperial territory’ (Reichsland) by an Oberpräsident under the
authority of the imperial chancellery. In 1879 it became a federal
state, but of a particular kind: its voice in the Federal Council was
merely advisory, and the governor in Strasbourg, who headed the
administration, was not under the authority of the chancellor, but
directly subject to the emperor. Up to 1911 Alsace-Lorraine had no
popular representation; the provincial parliament was replaced by
a provincial committee of notables. Nonetheless, a certain half-
hearted attempt at rapprochement with the other German states
could not be overlooked. The autonomists around August
Schneegans actively contributed to the imperial law of 1879
reorganizing the relationship between the Reichsland and the
Reich.19
In the year 1887, Alsace-Lorraine became the theme of a federal
electoral campaign, if only indirectly. At the beginning of 1886
General Boulanger, the idol of those who wanted a war of
444
vengeance against Germany in order to win back the lost provinces,
took over the French ministry of war. At the same time, a serious
crisis with Austria-Hungary led to an increase of pan-Slavist
influence in Russia. Bismarck feared Germany could quickly plunge
into a two-front war. The cauchemar des coalitions, the nightmare of
hostile alliances against Germany, was never so strong for the
German chancellor as in 1886–7.
When in January 1887 he demanded a renewal of the seven-
year military budget (the last Septennat, approved in 1880, having
expired), his main argument was the revanchist mood in France.
The rejection of the petition prompted Bismarck to dissolve the
Reichstag three-quarters of a year before the legislative term was to
come to an end. The ensuing electoral campaign was completely
dominated by a national-psychological mobilization against the
western neighbor. National Liberals, German Conservatives, and
Free Conservatives fought as a political ‘cartel’ at Bismarck’s side
for a new Septennat, and they were successful. The new parliament,
elected on 21 February, had a governmental majority again,
composed of the three above-mentioned parties together holding
more than 220 of 397 seats. In March 1887, the new ‘cartel
parliament’ passed its first law, the third Septennat.
That same year, Bismarck orchestrated the foreign policy coup
his posthumous admirers long considered to represent the summit
of his diplomatic mastery: the signing of the secret ‘Reinsurance
Treaty’ with Russia. The relationship between Germany and Russia
had deteriorated after the Congress of Berlin in 1878; Russia felt
that Bismarck had stripped it of the fruits of its victory against the
Turks. Germany’s response was an alliance with Austria-Hungary in
445
1879, a secret treaty of mutual assistance in the case of a Russian
attack against either partner. Additional security was the goal of
the 1882 triple alliance between Germany, Austria-Hungary, and
Italy, in which Berlin and Vienna promised to help Rome in the
event of an unprovoked French attack. If the attack went against
Germany, the latter could, in turn, count on Italian assistance. If
one partner went to war with another European power, the other
partners committed themselves to a policy of benevolent neutrality.
In the event of war with several powers, they were to provide
assistance.
In the meantime, Russia and Austria-Hungary had managed to
smooth out their differences in the Balkans sufficiently to conclude
a secret ‘Three Emperors’ League’ (Dreikaiservertrag) in 1881, a
three-year agreement in which St Petersburg, Berlin, and Vienna
promised to observe a policy of mutual benevolent neutrality in
case one of the partners got involved in a war with a fourth party.
This treaty was extended another three years in 1884. In 1885,
however, a new Balkan war broke out between Serbia and
Bulgaria, leading to a serious conflict between Austria and Russia.
The Bulgarians won. Without first discussing the matter with
Russia, Austria prevented Serbia from ceding territories. Russia
interpreted this action as a violation of the 1881 treaty, which
required consultations in such cases. The year following, the
relationship between Vienna and St Petersburg deteriorated still
further. With Russian help the prince of Bulgaria, Alexander von
Battenberg, was deposed. His successor—though it was not yet
decided who that would be—was to bring the country into closer
dependency on the tzarist empire. Under these circumstances, a
446
renewal of the Three Emperors’ League in 1887 was inconceivable.
After the coup in Sofia, Bismarck came under strong pressure
from the governments of Austria and Britain, who wanted the
Reich to commit itself in advance to counteracting any Russian
military measures against Bulgaria or on the Bosphorus. If the
chancellor had agreed, he would have provoked a complete break
with Russia and probably also a Franco-Russian alliance. This
would have been tantamount to abandoning the foreign policy
axiom he had formulated on 15 June 1877 in the so-called
‘Kissinger Diktat’, in which he declared that his view of Germany
‘has nothing to do with accumulating territories, but concerns a
total political situation in which all powers except France need us,
and are restrained as much as possible by their relationships with
each other from entering into coalitions against us.’
To promote rivalries between foreign powers for the purpose of
preventing coalitions against Germany was in keeping with the
Kissinger rule. In 1887, too, the chancellor remained faithful to his
maxim, endorsing Rome’s proposal of a ‘Mediterranean entente’
between Britain and Italy from February of that year, with Austria-
Hungary joining in March. The anti-Russian tenor of the secret
treaty was obvious. The three powers obligated themselves to
preserve the status quo in the entire Mediterranean and the Black
Sea as much as possible, as well as to prevent annexations,
occupations, and the erection of protectorates. Germany, the only
European great power informed of the agreement, was a kind of
secret sharer in the alliance. In Bismarck’s view, it was in
Germany’s interest for London both to cooperate with Vienna and
to become a greater adversary of Russia than it had been in the
447
past. What the chancellor could not want, on the other hand, was a
war between Russia and Austria-Hungary, Germany’s closest ally.
Yet with the British now backing the Austrians in the
Mediterranean, the danger of such a war grew.
The Reinsurance Treaty with Russia, which Bismarck signed on
18 June 1887, contradicted the Mediterranean entente (and also
the Treaty of London of March 1871, which closed the straits
between the Bosphorus and the Dardanelles to all military vessels)
in one important respect: in an ‘extremely secret supplementary
protocol’, Germany promised to assist the tzarist empire if the
latter should decide to defend its interests at the outlet of the Black
Sea. The chancellor accepted this contradiction as a price for the
advantages contained in the many body of the treaty: Germany
committed itself to neutrality in the case of an unprovoked
Austrian attack on Russia, and Russia promised to remain neutral
in the case of an unprovoked French attack on Germany.
The treaty with St Petersburg lessened the risk of a two-front
war—as did the fall of Boulanger in Paris shortly before, in May
1887. Still, Bismarck was not at all convinced that the agreement
had forever averted the danger of a Franco-Russian alliance. At the
end of December 1887, he remarked to the Prussian minister of
war, Bronsart von Schellendorf, that, to judge by the course of
European politics, it was likely ‘that in the not too distant future
we will have to deal with a war against both France and Russia’.
Herbert von Bismarck, the son of the chancellor and secretary of
the foreign office, noted soberly in June 1887 that the treaty would
keep ‘the Russians off our backs probably six to eight weeks longer
than without it, when push comes to shove’. But European politics
448
was not the only thing keeping the fear of a two-front war alive.
Bismarck’s tariff policy, too, strained the relationship with Russia.
In 1887, the 1879 duties on grain were drastically increased for the
second time, which meant virtual economic war with Russia. The
consequences of the conservative swing in domestic policy put
increasing pressure on the alliance between Germany and the great
conservative power in the east.
Bismarck’s biographer Lothar Gall has called the chancellor’s
domestic policy after 1881 a ‘system of expedients’. The same can
be said for his foreign policy in 1887. The conflicts between the
alliances in which Germany was—directly or indirectly—involved
were impossible to resolve. Only the relationship with Austria-
Hungary was comparatively stable—not unlike, after the 1879
alliance, the ‘wider federation’ of which Heinrich Gagern had
spoken three decades before. Bismarck had entered into the pact in
consciousness of the centuries-old common bond between Germany
and Austria, but he was not blind to the fact that this alliance, too,
had its dangers for Germany.
The Reinsurance Treaty did not violate the letter of the
agreement with Austria, but it violated its spirit. In turn, the
‘Lombard prohibition’ of November 1887, by means of which
Bismarck almost completely cut Russia off from German sources of
credit, violated the spirit of the Reinsurance Treaty. In the second
half of the 1880s, the foreign policy of the Reich founder had
reached an impasse, a predicament for which he himself, through
his domestic policies, bore no small share of responsibility.20
449
THE ACCESSION OF WILLIAM II AND
THE FALL OF BISMARCK
450
1898, the year of his and Bismarck’s death, the eponymous hero,
the old Junker Dubslav von Stechlin, a moderate conservative,
explains to the rather liberal Count Barby why Emperor Frederick
most certainly would have failed:
451
against Bismarck’s successor, General von Caprivi, a resolute
opponent of high agricultural tariffs. Just as Caprivi had failed not
least owing to the resistance of the Junkers, so too would Emperor
Frederick have failed, if he had attempted to pursue a policy
counter to the central interests of the Prussian manorial lords. That
was the upshot of old Stechlin’s reflections, and he was probably
right.
William II, who succeeded Frederick on 15 June 1888, was just
29 old at the time and the precise opposite of his father in nearly
all important respects: no man of liberal convictions, but rather
profoundly authoritarian; at times closely allied with leading
representatives of the anti-Semitic movement like the court
chaplain, Stoecker; widely talented, but superficial; a vain, pomp-
loving blusterer, who sought to compensate for inner insecurity
and a physical defect—his left arm had been crippled since birth—
with tough talk. That he would clash with Bismarck was clear from
the outset. In 1887–8, the prince sided with the ‘war party’ around
the emperor’s general aide-de-camp, Count Waldersee, which
pushed for a swift conquest of Russia and France. Reinforced in his
tendency to overestimate his abilities by ambitious favourites like
the amateur poet-composer Count Philipp von Eulenburg and
enemies of Bismarck like the anti-Semite and Russian-hater
Waldersee, William, when he assumed the throne, was determined
to step out from behind the shadow of the Reich founder and rule
by himself.
The antagonism between the young emperor and the old
chancellor was especially marked in the social field. At the
beginning of 1890 William sought to undermine social democracy
452
by adopting a number of labour-protection measures, among them
a prohibition of Sunday work. Bismarck feared such measures
would have the opposite effect and advised the emperor against a
proclamation of social policy. When William insisted, Bismarck
relented—but only tactically, hoping that the legislative process
would thwart William’s plans.
The simultaneous conflict over a further renewal of the anti-
socialist law was even more serious. The National Liberals would
not vote for the bill unless the paragraph authorizing expulsion was
eliminated. William was prepared to make this concession in order
to salvage the law. When the conservatives demanded that
Bismarck condemn this move through an open declaration before
parliament, Bismarck refused. On 25 January 1890 the
Conservatives, Centre, Liberals, Social Democrats, the Alsace-
Lorraine party, and the national minorities voted against the third
reading of the amended bill. The 169 negatives were opposed by
only 98 positive votes from National Liberals and Free
Conservatives. The anti-socialist law thus expired on 30 September
1890, and there could be no doubt about result: the ‘cartel’ of 1887
was destroyed and the power of the chancellor permanently
weakened.
On 20 February 1890 a new Reichstag was elected, with run-off
elections on 28 February. The parties of the former ‘cartel’ suffered
a decisive defeat, losing a combined total of 85 seats and therewith
the majority. The winners were the Social Democrats, now the
strongest party, climbing from 10.1% of the vote to 19.7% and
from 11 to 35 seats; and the Liberals, who gained 16% of the vote,
up from 12.9%, won 66 seats, up from 32.
453
It was impossible for Bismarck to rule with the new parliament.
To be sure, he could attempt to break the Centre party out of the
oppositional front, and to that purpose held a conversation
(mediated by Bleichröder) with his long-time parliamentary
adversary, Ludwig Windthorst, on 12 March 1890. But a
governmental majority of the former cartel parties—German
Conservatives, Free Conservatives, National Liberals, and Centre—
was inconceivable. The antagonisms between the parties had
become too great, and the small but numerically possible
conservative-clerical majority was a long way from being a
political majority for the same reason.
Bismarck was resolved upon an even fiercer battle against social
democracy, and this now meant battle against the Reichstag—both
the one currently sitting and, in all likelihood, against the one soon
to be elected anew. In his efforts to paralyse parliament the
chancellor, in a meeting of the Prussian ministry on 2 March, went
so far as to consider the renunciation of the imperial office by the
Prussian king and even the dissolution of the Reich by the princes.
At first the emperor seemed willing to attempt the coup, but then
refused, intimidated by the consequences.
This spelled the failure of the anti-parliamentary strategy of the
chancellor. Yet there can be no doubt about the seriousness of
Bismarck’s attempt to replace the imperial constitution with a new
one more to his taste. His 1890 scheme for a coup d’état suddenly
revealed the merely provisional nature of the compromise that had
brought the Prussian constitutional conflict to an end in 1866. In a
power struggle between parliament and executive, one that would
come sooner or later, the government was going to have to decide
454
whether it would adapt itself to the representative majority or
violate the constitution. In 1890 Bismarck was clearly prepared to
take the second path.
The interview with Windthorst (which is occasionally and
questionably cited as evidence that Bismarck ultimately preferred a
parliamentary and thus constitutional solution to the crisis)
contributed significantly to the fall of the first chancellor of the
Reich. The emperor was enraged that Bismarck had not sought his
permission for the talks. It provoked William still more that, on 4
March, Bismarck sent the Prussian ministers a copy of a cabinet
order of Frederick William IV from the year 1852 prohibiting them
from consulting directly with the monarch without the approval of
the head of government. When William demanded that Bismarck
abolish the old order, the latter refused to present a draft for a new
one.
As the conflict between emperor and chancellor reached a head,
Bismarck sought to negotiate an extension of the Reinsurance
Treaty. The Russian ambassador, Count Shuvalov, had received
instructions in St Petersburg to conclude the negotiations with
Bismarck, and only with him. He came too late. On 17 March 1890,
the day Shuvalov met Bismarck, the emperor requested the
chancellor’s resignation. Bismarck obliged on 18 March. Two days
later he was dismissed from the offices of Reich chancellor, prime
minister, and foreign minister of Prussia. The Reinsurance Treaty
was not extended. The enemies of Bismarck’s Russian policy, led by
Count Waldersee and the éminence grise in the foreign office, the
councillor Friedrich von Holstein, had exploited the situation and
convinced William that it would be better for Germany to reject the
455
Russian wish.21
456
foreign to most Germans. They liked to quote a passage from his
parliamentary address of 6 February 1888: ‘We Germans fear God,
but nothing else on earth’. But that was only half of the sentence.
The second part, ‘and the fear of God is the thing that makes us
love and preserve peace,’ did not become proverbial.
A particular warning issued by the chancellor in this same
speech had even less of an effect on contemporaries and posterity.
457
with the other world powers, above all Britain, became
overpowering only after Bismarck’s dismissal.
Bismarck’s historical greatness was confined to foreign policy.
Virtually no historian would rank Bismarck’s achievements on the
domestic front in anywhere close to the same category. On 10 July
1879 Albert Hänel, professor of constitutional law in Kiel and
delegate of the German Progressive party, expressed what by that
time was the common opinion of all critical observers:
It has turned out that the Herr Chancellor has brought ministerial
dictatorship to a hight at which all parliamentary life is indeed
illusory ... With this Chancellor, every constitutional assembly is
more or less an ornament. All that is happening is that the Herr
Chancellor is using the constitutional system as a way of avoiding
to some extent the responsibility for his dictatorial plans.
458
reason the parliamentarization of Germany and Prussia never came
about. It also failed because of the party that supported it in
principle, but identified it so completely with its own political
hegemony that a ‘legitimate’ shift in parliamentary power was out
of the question.
And yet, despite all that, the nation grew closer together—so
much so, in fact, that the founder of the Reich was probably the
only German in 1890 to brood over the possibility of its
dissolution. The only adversaries of the Reich as a national state
were those belonging to it by compulsion, the Poles in the Prussian
eastern provinces and the Danes in northern Schleswig, who could
no longer look forward to a popular referendum over their
nationality after an October 1878 treaty between the German Reich
and Austria-Hungary had abolished the corresponding clause from
the 1866 treaty of Prague. Concerning the Alsace-Lorrainers, on the
other hand, by 1890 it could no longer be assumed that they were
all ‘Germans by force’. First the Protestants, then the Catholics of
the imperial territory had come to accept their place in the Reich.
No more ‘protesters’ were elected to the parliament after 1890. In
the ‘old German’ areas, nobody seriously questioned the existence
of the Reich after 1871. By 1890, hardly anyone still felt a conflict
between loyalty to a particular state and loyalty to the Reich. Two
decades after the empire’s foundation, the feeling of German
national community was at least as strong as the sense of
individual regional differences. By the time Bismarck was
compelled to leave the government, the formation of the German
nation state could be regarded as complete.
One and a half years before his dismissal, at the end of
459
September 1888, Bismarck had Heinrich Geffcken, a jurist from
Hamburg and university friend of Emperor Frederick, arrested on
the charge of disseminating state secrets through the publication of
the crown prince’s war journal from the years 1870–1. Geffcken
was acquitted in January 1889 in a federal court. Fontane, as well
as the liberal press, considered Bismarck’s behaviour not only
damaging to his reputation, but also to the reputation of Germany.
‘Perhaps it is only an outraged sense of patriotism, but it gives the
impression of a mean personal hatefulness,’ Fontane wrote to the
court official Georg Friedlaender on 7 January 1889.
The great man was indeed a great hater, and in his hatred was
often very ignoble. When Eduard Lasker died in New York on a
visit to the United States in January 1884, the chancellor saw to it
that no member of government took part in the funeral services in
Berlin, and he did not forward a message of condolence from the
American House of Representatives to the Reichstag. When he tried
to justify his behaviour before the delegates on 13 March 1884, he
did not refrain from yet another attack on his deceased opponent.
To be sure, Bismarck also made other, more positive
contributions to what we today call ‘political culture’. Many of the
great speeches he gave in the Prussian House of Representatives
and the Reichstag have gone down in the annals of German
parliamentary history for their rhetorical brilliance, the clarity of
their language, and the keenness of their argumentation. Yet
stylistic virtuosity could not counterbalance the political
unscrupulousness that was also part and parcel of Bismarck’s
character. Theodor Mommsen, one of the liberals who first fought,
then supported, then finally fought Bismarck again, came in 1902
460
—the year before his death—to a deeply pessimistic conclusion
about the effects the founder of the Reich had had on Germany. In
a letter to the economist Lujo Brentano of 3 January, Mommsen
wrote: ‘Bismarck broke the back of the nation’.
Some three decades before, in the year 1873, Nietzsche had
written in the first of his Unzeitgemäße Betrachtungen* about the
error committed by public opinion and by all those who publicly
opined that German culture was one of the victors in the war
against France. ‘This madness is highly pernicious: not because it is
madness (for some errors are very salutary and blessed), but
because it threatens to turn our victory into a signal defeat: into the
defeat, indeed the extirpation, of the German spirit for the sake of the
“German Empire.” ’
At about the same time, Jacob Burckhardt in Basel was already
predicting that within a few years, thanks to the ‘moulting process’
taking place within German historical scholarship, ‘all of world
history from Adam onwards will be given a new coat of German
victory paint and oriented towards the years 1870–1.’ This goal
had not yet been reached by 1890, nor was it the goal of all
German historians. But it is true that the dominant schools of
German historiography, as well as the writers of school textbooks,
made a concerted effort to place the rise and ‘German mission of
Prussia’ in the centre of all modern historical study.
In an academic address given shortly after the outbreak of the
Franco-Prussian War, Emil Du Bois Reymond, the famous Berlin
physician, referred to his university as the ‘intellectual royal guard’
(geistiges Leibregiment) of the House of Hohenzollern. As it
461
happened, neither Du Bois Reymond’s alma mater on Unter den
Linden street, nor the German university system as a whole, ever
lived up (or rather down) to that distinction. In fact, German
higher education experienced a new renaissance during the
Wilhelmine empire, gaining prestige and admiration throughout
the whole world. Yet the phrase ‘intellectual royal guard of the
House of Hohenzollern’ had been pronounced, and it certainly did
apply to a part of public opinion. Professors and politicians, pastors
and publicists created the cult and the mythology around Bismarck,
Prussia, and the Hohenzollerns, which became inalienable
components of German imperial nationalism.
‘Rarely has a dynasty ruled so completely over the hearts of its
subjects as the Hohenzollerns,’ wrote the National-Zeitung at the
beginning of December 1878, when Emperor William I, recovered
from Nobiling’s assassination attempt, returned to Berlin. ‘Already
something of the radiance of myth surrounds the majestic head.
Charlemagne and Frederick Barbarossa seem like the only worthy
figures of history whom we might compare with him as equals.’
Educated Germany celebrated the country’s belatedly achieved
national unity. The continuing absence of the political liberties for
which the liberals had fought before 1871 was not something the
majority of Germans felt to be a lack. German liberalism had failed.
Mommsen’s verdict on Bismarck’s twenty-eight years of rule in
Prussia and Germany cannot be interpreted in any other way.22
462
6
463
Three decades before, Otto Michaelis, delegate of the right wing of
the German Progressive party, had argued in very similar fashion.
Prussia, the liberal delegate declared before the House of
Representatives on 13 June 1865, must set itself ‘great goals’ and
seize the ‘banner of Prussia and Germany against Austria’; for
without such an ‘ideal goal’, the people would lose the energy
necessary ‘to fight the constitutional conflict with us to the end’. In
1895 the Freiburg economist and sociologist found himself up
against the same inner adversary Michaelis had confronted in
1865, and like the latter Weber believed that the means to supply
the power struggle against the Prussian Junkers with new energy
lay in foreign policy. Since the national goal of German liberalism
had been achieved with German unification, however, the foreign
policy objective could not be the same as thirty years before.
Austria had been replaced with the world—not as a target of
attack, but nonetheless as a sphere of activity.
The nationalist thrust of Weber’s critique of Prussian Junkerdom
lay in his reference to the growing significance of migrant Polish
labour. Seasonal, low-cost agricultural workers from Russian
Poland were increasingly replacing German day laborers fed up
with conditions in the old patriarchal manors. ‘Large-scale
operations, which can only be maintained at the cost of
Germandom, ought, from the point of view of the nation, to go
under,’ Weber concluded. In other words, they should be left to
fend for themselves without their life support of protective tariffs
burdening society.
Though condemned to economic decline, politically the Junkers
were still the most influential class in society. And more than that.
464
Like his fellow economist Werner Sombart and the constitutional
scholar Hugo Preuss, Weber saw that the entrepreneurial class of
the German bourgeoisie had gone far in adapting itself to
aristocratic values. The desire to acquire noble titles and landed
estates, imitate a seigneurial lifestyle, achieve the status of reserve
officer, cultivate dueling—there was a plethora of evidence to
support the thesis of the Verjunkerung or ‘feudalization’ of the
upper German bourgeoisie. Yet the social emulation of the nobility
was not confined to Germany, and within Germany was not
everywhere as strong as in the old Prussian territories. The
‘aristocratization’ of parts of the bourgeoisie was accompanied by a
Verbürgerlichung or ‘bourgeoisification’ of parts of the nobility, and
even when ennobled, a bourgeois—for example an industrialist—
could not renounce his heritage. The ‘feudalization’ was not an
invention of bourgeois scholars, but it was also not as profound as
its critics imagined. Ultimately the counter-forces, working
consciously and unconsciously towards the Verbürgerlichung of
Wilhelmine Germany, would prove the stronger.1
It was no coincidence that the new debate over Prussian
Junkerdom began in the last decade of the nineteenth century. It
was set off by the manner in which the East Elbian landowners
reacted to the about-face in commercial policy by Bismarck’s
successor. Caprivi had correctly recognized that Germany’s
transformation from an agricultural to an industrial state was
irreversible and that the country’s economic future depended on an
increase in exports. ‘We must export,’ he announced to the
Reichstag on 10 December 1891. ‘Either we export goods, or we
export people. Without an industry growing at the same rate as our
465
population, we cannot survive.’ The commercial treaties of the
Caprivi era (in 1891 with Austria-Hungary, Italy, Belgium, and
Switzerland, and in 1893–4 with Spain, Serbia, Romania, and
Russia) were guided by this insight. Germany promoted the export
of industrial goods and, in exchange, dismantled its tariff barriers,
including the duties on grain so important to agriculture east of the
Elbe.
The opposition of conservatives and landowners began
immediately in 1891, and it grew more intense when the
chancellor extended his commercial system to agricultural
countries like Romania and Russia. The founding of the Federation
of Agriculturalists (Bund der Landwirte) in February 1893 gave
organizational expression to the protest against the government’s
policies. Although the great landowners were the driving force
behind this association, it sought from the beginning to appeal to
all groups who felt threatened by the continuing industrialization
of society, especially the peasants and the small and medium-sized
tradesmen and business owners, the Bauernstand and the
Mittelstand. The success of this effort was not very great among
craftsmen, but among small farmers it was resounding: by the turn
of the century, they formed nine-tenths of the organization’s more
than 200,000 members.
Nonetheless, the leadership of the Federation was dominated by
the great Prussian landowners, who did not shrink from promoting
their interests with a propagandistic excess hitherto unknown in
Germany. An early example of the new style of agitation was the
call to establish the Federation, sent out to German agriculturalists
in December 1892 by the Silesian Generalpächter* Alfred Ruprecht-
466
Ransern:
In fact, it would take some time before the screams were heard.
But their opponents in the chancellery, without knowing it, were
playing into the hands of the landowners. The commercial treaties
and a number of domestic reforms—such as the prohibition of
Sunday and child labour, the institution of industrial tribunals
(Gewerbegerichte, consisting of a state official as chairman and one
representative each from among the employers and the workers),
and a progressive income tax in Prussia—demonstrated the earnest
desire of Caprivi and his colleagues in the federal offices and
Prussian ministries to promote the much-discussed ‘new direction’
in economic and social policy. When it came to turning his plans
into practical realities, however, the chancellor made one bad
mistake after another. In March 1892, his failure to push through
an education reform measure friendly to church interests had led to
his resignation as Prussian prime minister. The bill had outraged
the Liberals, with whom Caprivi had cooperated well up to that
point; its retraction incensed the Catholics. On 6 May 1893, during
467
the hearing of a federal bill to increase the armed forces, the
majority of both fractions voted against both the governmental
draft and a compromise proposed by the Centre delegate Huene
with the government’s agreement, thus terminating the project.
Caprivi responded by dissolving the Reichstag (which otherwise
would not have been re-elected until February 1895, according to
the five-year legislative period introduced in March 1888).
The May 1893 vote had far-reaching consequences for the
Liberals. Caprivi’s bill had gone far to accommodate them in one
important point: the term of military service was reduced from
three to two years. Consequently, six Liberal delegates voted for
Huene’s compromise proposal. When the party leader, Eugen
Richter, demanded that the rebels receive a sharp reprimand and
canvassed a narrow fractional majority to that purpose, the German
Liberal party split. The majority, under Richter, renamed itself the
Liberal People’s Party (Freisinnige Volkspartei); the minority,
including the ‘secessionists’ Bamberger, Mommsen, and
Stauffenberg, founded the Liberal Association (Freisinnige
Vereinigung). The social profiles of the two left-liberal parties
differed considerably. The People’s party was strongly petty
bourgeois in character; in the Union, representatives of the
propertied and educated classes set the tone.
In the federal elections of June 1893, the proponents of military
enlargement, the former ‘cartel’ parties, performed well, the
opponents of the measure poorly, with the exception of the Social
Democrats. The two liberal parties did worst of all, together falling
from 66 to 37 seats (24 for the Liberal People’s party, 13 for the
Liberal Association). The new Reichstag passed the military bill.
468
But the chancellor’s victory strengthened his position only
temporarily. In the next year, the antagonism between Caprivi and
the arch-conservative Prussian prime minister Count Botho zu
Eulenburg came to such a head that further cooperation between
the two was impossible. The leader of the Prussian government
favoured a new emergency law against social democracy, as
William II was demanding, as well as the emperor’s plans for a coup
d’état; the chancellor rejected both schemes. The conflict ended in
October 1894 when the emperor dismissed the two adversaries and
appointed the former Bavarian prime minister and then-governor of
the imperial territory of Alsace-Lorraine, Prince Chlodwig von
Hohenlohe-Schillingsfürst, to the offices of federal chancellor and
Prussian prime minister.
Hohenlohe, 75 years old at the time, was a Catholic and a
moderate in the southern German liberal tradition. As chancellor,
however, he bowed to the will of the emperor, who intended to
exploit Caprivi’s dismissal for a sharp change of course to the right.
To the extent that this involved more forceful action against social
democracy, a particular matter of concern for William II, success
continued to elude the government. To the so-called ‘subversion
bill’ (Umsturzvorlage) of December 1894, which contained strong
punitive measures against inciting class hatred and against insults
to the monarchy, religion, marriage, family, and property, a further
clause was added, in order to secure the support of the Centre,
punishing slander against doctrines of the church. For National
Liberals, however, the added clause made the bill completely
unacceptable, and the Reichstag rejected it by a large majority in
May 1895.
469
After the attempt to pass a new—albeit disguised—anti-socialist
law had failed on the federal level, it was the turn of the individual
states. But even in Prussia the government did not manage to find a
House majority for a so-called ‘little anti-socialist law’ granting the
police wide authority to disband associations and disperse
gatherings. Saxony’s efforts met with more success. In March 1896
the Saxon parliament replaced its moderate census-based voting
law with a three-class system (Dreiklassenwahlrecht) favourable to
propertied interests. The consequence was that, at least for the time
being, the Social Democrats could send no delegates to the
parliament of a state that was a stronghold of social democracy.
The Saxon proletariat reacted with bitterness and a marked shift to
the left.
Three years later, the Reich leadership undertook a renewed
legislative effort against social democracy. The so-called
‘penitentiary bill’ (Zuchthausvorlage) of May 1899 increased the
penalties for ‘compulsion to coalition’ (Koalitionszwang), that is,
attempting to force workers to participate in a strike or labour
union. Except for the two conservative parties, however, no
fraction was prepared to vote for the new emergency measure. The
‘penitentiary bill’ was the last legislative attack on social
democracy at the federal level. Nonetheless, legal and de facto
discrimination of members of the social-democratic worker’s
movement continued to be very frequent, for example the Prussian
Lex Arons of June 1898 (for which also the Centre voted) excluding
Social Democrats from all academic teaching positions.3
Far-sighted people of all political camps knew that the war
against social democracy could not be won with suppressive
470
measures alone. But they also saw that the social policy of the
government was unable to stop the rapid growth of the movement.
In the federal elections of 1893, the Social Democratic Party of
Germany (SPD; Sozialdemokratische Partei Deutschlands, as it was
called after the 1891 party congress at Erfurt) had increased its
share of the vote from 19.7% to 23.3% and its seats in the
Reichstag from 35 to 44. This sobering experience was partly
responsible for the lack of major social policy initiatives during the
‘Hohenlohe era’ from 1894 to 1900.
Instead, the containment of social democracy would be
attempted by other means—through the ‘gathering together’
(Sammlung) of all forces wishing to uphold the prevailing order of
state and society. The actual author of this Sammlungspolitik was
Johannes von Miquel, Prussian minister of finance from 1890 and
in 1897 ennobled by the emperor. A former National Liberal
delegate to the Reichstag and mayor first of Osnabrück, then of
Frankfurt am Main, Miquel had long since turned his back on
economic liberalism, both freedom of commerce and freedom of
trade. In 1879 he demanded an end to the neglect of the craft
trades by the government and the ruling classes, believing this state
of things ‘a very dangerous one, especially dangerous in times of
revolutionary agitation against the foundations of our societal
order’. In the 1884 ‘Heidelberg declaration’ of the southern German
National Liberals, which he penned, he advocated the protection of
German agriculture. Both groups played an important role in the
Sammlungspolitik of the 1890s. Miquel aimed at a grand alliance of
elites uniting industry and agriculture, liberalism and conservatism,
in their entirety. The Mittelstand was to form the broad base of
471
society, with the independent craft trades as its solid nucleus.
It was relatively easy to accommodate the craft trades to a
certain extent. In 1897 an innovation in the trade ordinances
created ‘handicraft chambers’ (Handwerkskammern) as corporations
covered by public law, as well as instituting ‘elective compulsory
guild membership’ (fakultative Zwangsinnung); that is, when the
independent craftsmen of a given trade within a particular
handicraft chamber district opted to establish a guild, membership
was obligatory. The attempt to reconcile the interests of agriculture
and industry proved much more difficult. In September 1897, in
accordance with a demand from the Central Association of German
Industrialists taken up by Miquel, an Economic Committee was
formed in the federal ministry of the interior for the purpose of
preparing commercial policy measures. This council, which
contained representatives from both agricultural and industrial
organizations, was to prepare a new customs tariff for
implementation after the commercial treaties from the Caprivi era
expired.
Miquel intended the East Elbian agricultural protectionists and
the heavy industrialists to have a dominant voice in this body. It
was so dominant, however, that the export and finished goods
branches (organized in the League of Industrialists after 1895 and
the Association for the Protection of the Interests of the Chemical
Industry) immediately began to form an anti-protectionist
defensive front. In an appeal of March 1898, the free traders
confronted the old protectionist shibboleth for the ‘protection of
national labour’—which Emperor William II, at Miquel’s
instigation, adopted in June 1897—with the accusation that the
472
‘fulfilment of the demands of the agricultural special interests
necessarily causes inflation in the cost of living for the broad strata
of society’ and represents a ‘preference for the few at the expense
of the many’. In November 1900 the Commercial Treaty
Association (Handelsvertragsverein), an anti-agrarianist forum, was
founded in Berlin. In terms of customs policy, therefore, Miquel’s
Sammlungspolitik was a failure. Protective tariffs were not going to
be a platform upon which all of agriculture and industry could be
brought together, at least for the time being.4
By the turn of the century, however, a promising solution to the
conflict of economic interests had been discovered: the German
naval fleet. The ‘father’ of the German navy, Rear Admiral Alfred
von Tirpitz, head of the imperial navy since June 1897, knew that
he would have a good chance of convincing the educated and
propertied bourgeoisie of the merits of a navy. The memory of the
first German fleet, created by the provisional central authority in
1848 only to be auctioned off disgracefully in 1852, was still alive,
and ever since the revolutionary era, Germans were widely
disposed to regard a German navy as the armed branch of the
merchant marine—a ‘bourgeois’ fighting force, in contrast to the
aristocrat-dominated army.
In the Reichstag, however, there were still serious reservations
about the kind of large-scale naval expansion envisioned by the
emperor and Tirpitz. The modest plan establishing the navy in
1873, which considered the protection of German marine
commerce and the German coast the navy’s most important task,
was still in effect. Since a longer-term programme meant longer-
term budgetary decisions, it would be unworkable unless
473
parliament voluntarily restricted its own budgetary powers.
Moreover, large-scale fleet construction would require the creation
of offensive capabilities, and thus threatened to have a negative
impact on Germany’s relationship to the leading naval power,
Great Britain. In light of such considerations the Reichstag, which
had agreed to the construction of three new cruisers in March
1896, rejected the proposal for three more the following year. The
immediate consequence of the government’s defeat was the
resignation of the federal naval secretary, Admiral Hollmann. The
indirect consequence was the nomination of Tirpitz as his
successor.
The watchword for the new chapter of German naval policy that
began with Tirpitz’s appointment was pronounced by Baron von
Marschall, secretary of the foreign office, in the Reichstag debate
over the construction of the cruisers on 18 March: ‘The question of
whether Germany should pursue a world policy is directly
connected to another question: whether or not Germany has world
interests. This question has long since been answered’. To pursue a
‘world policy’ (Weltpolitik) meant the desire to compete with
England on an equal basis, a goal inconceivable without a strong
navy. Thus a German fleet of the dimensions Tirpitz had in mind
was necessarily a fleet against Britain. Simultaneously, however,
the construction programme was aimed against an inner adversary,
social democracy. A letter from Tirpitz to the leader of the imperial
admiralty, General von Stosch, written at the end of 1895, makes
no secret of this motivation. Germany, wrote Tirpitz, must
undertake a world policy ‘in no small measure because in the great
new national undertaking and the economic boon associated
474
therewith lies a strong palliative against educated and uneducated
social democrats’.
Tirpitz’s first naval bill, passed by the Reichstag at the end of
March against the votes of the Social Democrats, the Liberal
People’s party, and a minority of the Centre, as well as Guelfs,
Alsace-Lorrainers, and Poles, provided for the increase of the naval
fleet to nineteen battleships, eight coastal defence ships, twelve
heavy and thirty light cruisers. Two years later, in June 1900, the
Reichstag—newly elected in June 1898—consented to a new bill
from the imperial naval office practically doubling the size of the
fleet. At the conclusion of the programme, the strength of the
German navy was to be comparable to that of the British in a ratio
of two to three. This would mean parity in the North Sea. After
1900, therefore, there could be no doubt that the law of 1898 had
been only the first stage of a comprehensive plan.
The parliamentary majorities for the naval bills were not
fortuitous. Tirpitz revealed himself to be a master at gaining the
necessary public support for his policies. He was aided in this effort
by an organization founded in 1898 with his collaboration: the
German Navy League (Deutscher Flottenverein), a modern ‘pressure
group’ comparable to the Federation of Agriculturalists. Active in
the League were the direct and indirect interests of heavy industry,
shipyards, wholesale and foreign commerce, and the export
branches; a broad spectrum of political parties from the Free
Conservatives through the National Liberals to the Liberal
Association; and wide strata of the bourgeoisie and petty
bourgeoisie, whose support of the fleet construction gave them the
feeling of participating in a great national enterprise. By 1900 the
475
association numbered some 270,000 individual members. If we
include corporate members (i.e. members of organizations who had
joined the League), by 1908 the number had grown to over a
million.
With one group of the ‘power elite’, however, Tirpitz’s naval
policy was not especially popular. From the point of view of the
Prussian conservatives and Junkers, everything that had to do with
the German navy was suspect. They instinctively regarded it as the
competitor of the Prussian army. Its economic beneficiaries could
only be industry and commerce and thus the same ‘modern’ world
by which rural East Elbian Germany saw itself threatened. To be
sure, the conservatives could not easily vote against the naval bills;
to do so would have brought the party of Count Kanitz and Ernst
von Heydebrand und der Lasa into close proximity with the party
of Eugen Richter and August Bebel. For their assent to the second
naval bill (1900), however, the conservatives demanded ‘payment’:
a promise to increase the grain tariff. This promise was fulfilled in
December 1902 by Chancellor Bernhard von Bülow (in office from
October 1900). The so-called ‘Bülow tariff’, passed by the Reichstag
with a large majority and in effect from 1 March 1906, awarded
the landowners higher duties on wheat, rye, and oats (albeit not as
high as the conservatives and the Federation of Agriculturalists had
demanded). Thus the most important goal of Miquel’s
Sammlungspolitik, an understanding between industry and
agriculture over customs policy, was achieved by means of a
‘detour’ through the naval programme. Miquel himself, who died in
September 1901, did not live to see this triumph. The success was
above all the work of Bülow—and Tirpitz.
476
In 1928 Eckart Kehr, author of the first critical study of the
societal foundations of German Sammlungspolitik and naval policy,
summarized the provisional conclusion to the political struggle:
‘Industry and agriculture came to an agreement that each would
not seek alone to control the state and to exclude the loser from the
use of the legislative process; rather, they would together erect an
agro-industrial condominium directed against the proletariat.’ The
first phase of German Weltpolitik thus occasioned a result
completely different than the one Max Weber had put forward in
1895. Not only had it not broken the political power of the
Prussian Junkers, it had actually helped to consolidate their
societal base.5
477
behind a policy outlined by Bernhard von Bülow (at that time
Prussian foreign minister and secretary of the foreign office) in the
Reichstag on 6 December 1897 to the effect that Germany did not
seek ‘to put anyone in the shade, but we do demand our place in
the sun’.
When Bismarck temporarily overcame his aversion to colonies in
the mid-1880s, Germany was in the middle of a depression.
Colonial acquisitions could be seen as a means of leading the
nation out of the depths of its economic woe. By the time William
II, Bülow, and Tirpitz took up the cause of a German Weltpolitik,
the country was in the midst of an economic upswing (beginning in
1895 and lasting until about 1900) comparable to the foundation
years of the Reich. Conscious of their economic strength, the
political leaders of Germany issued demands that aimed at a
dramatic change in the international balance of power.
In principle, their actions were no different from the imperialism
of other great powers like Britain and France. In terms of its impact
on Europe, however, the German bid for overseas acquisitions did
indeed have a different quality. For France, a colonial empire was
also a compensation for the losses in power and prestige the
country had experienced in 1870–1. The British empire supported
Britain’s claim to be the world’s leading maritime power. For
Germany, already semi-hegemonic on the Continent by virtue of
the founding of the Reich, the decision to pursue a ‘world policy’
could only mean that it was no longer satisfied with its Continental
status, that it intended to become a sea power as strong as British
(at least in the North Sea)—that it sought, in other words, to
advance from semi- to outright hegemony. It was inevitable that
478
the other great powers affected by this policy would attempt to
counter it. These included, in addition to Britain and France,
Russia, which was just as unwilling to see its world-political status
reduced by Germany.
In the autumn of 1900, after the defeat of the Chinese ‘Boxer
rebellion’ by the cooperative efforts of all the European great
powers, America, and Japan, the German–British relationship
seemed about to take a turn for the better. Both powers came to an
agreement that all nations should have unrestricted trade access to
all the rivers and coasts of China. However, although an alliance
was discussed between London and Berlin in 1901, neither side was
truly interested. A return to some version of Bismarck’s
‘reinsurance’ policy with Russia was now also out of the question.
A 1893 military convention between Paris and St Petersburg— the
Russian reaction to a trade war initiated by Caprivi—had outlasted
the German–Russian rapprochement under Hohenlohe. When
Russia, after the conclusion of a British–Japanese alliance at the
beginning of 1902, once again showed an interest in a treaty with
Germany, the Wilhelmstrasse refused. Berlin suspected that St
Petersburg was only interested in German backing for its
expansionist plans in the Far East.
The June 1902 renewal of the Triple Alliance with Austria-
Hungary and Italy initially seemed to be a foreign policy success
for Germany. Five months later, however, Rome concluded a secret
agreement with Paris, committing both countries to strict neutrality
in the event a third party attacked either. The rapprochement
between Italy and France, which undermined the Triple Alliance,
was unmistakable. By the end of 1902, Germany had only one
479
remaining ally on whom it could depend, but with which it could
also easily find itself drawn into entanglements with unpredictable
outcomes: Austria-Hungary. Bülow’s ‘free hand’ policy had
manoeuvred Germany into a dangerous isolation. The increased
national power German Weltpolitik had generated was nothing
more than appearance. In reality, Germany was more vulnerable
twelve years after Bismarck’s dismissal than it had ever been
before, and it had only itself to blame.6
480
The Pan-German League, a product of the colonial movement
and closely connected with Rhenish-Westphalian heavy industry,
declared war on all groups within Germany that opposed ‘national
development’, that is, leftists of every shade. Outside Germany, the
League supported the ‘German-national aspirations’ of Germans
living abroad throughout the world. In 1894 the Pan-Germans
began calling for a German-controlled Mitteleuropa and the
conversion of Germany from ‘great power status’ to ‘world power
status’. Of the political parties, the National Liberals and Free
Conservatives were initially closest to the Pan-Germans (for a time,
in fact, the Pan-German League sought to transform both into one
‘National party’). After 1911, when they became advocates of an
expansionist German world policy, the German Conservatives, too,
drew closer to the Pan-German agenda. At all times, the Pan-
German League looked upon itself as the avant-garde of all forces
pushing for an aggressive German foreign policy. For this reason, it
was careful to maintain its distance from official government
policy, which was seldom aggressive enough to find favour in the
eyes of the Pan-Germans.
The Eastern Marches Association saw its main task in the
‘strengthening and gathering together of Germandom in the Polish-
dominated eastern marches of the Reich through the uplifting and
consolidation of German national feeling as well as through the
increase and economic enhancement of the German population’. In
other words, the agenda of the ‘Hakatist Association’ (the unofficial
name, composed from the beginning letters of the names of the
organization’s founders, Ferdinand von Hansemann, Hermann
Kennemann, and Heinrich von Tiedemann-Seeheim) was the
481
rigorous Germanization and de-Polonization of the empire’s eastern
provinces. Place names and surnames were to be Germanized, and
the Poles replaced with German settlers.
Measured by the number of Germans who actually migrated to
the east, the success of the Association was modest. Far more
significant was its impact on ‘informed opinion’. The ‘Hakatists’
transformed the German prejudice against the allegedly racially
inferior Poles into a weapon of ethno-cultural warfare, and their
efforts were endorsed by the parties of the political right, the
German Conservatives, the Free Conservative Empire party, and the
National Liberals. The Hakatist attitude toward the government
was more favourable than that of the Pan-Germans. Official
government policy was fundamentally no less anti-Polish than the
Hakatist agenda; the only real differences were tactical in nature.7
In contrast to the Eastern Marches Association and the Pan-
German League, the ‘warriors’ associations’ (Kriegervereine) formed
a veritable mass movement. They were dominated not by members
of the educated and propertied bourgeoisie, nor by the ‘one-year
volunteers’ (Einjährig-Freiwilligen, holders of a patent of mittlere
Reife who, after six years in a middle or higher school, served only
one year in the military, not two, but in exchange had to pay for
their own lodging, board, and equipment). Rather, their
membership was predominantly craftsmen, shop owners, peasants,
day labourers, lower civil servants, employees, and industrial
workers—that is, ‘ordinary folk’, who cherished the memories of
their years of military service and, if they were old enough, of the
Franco-Prussian War, thereby holding up an idea of the nation as
the community-in-arms of all German men, even of the least
482
wealthy and educated. The umbrella organization, the Kyffhäuser
Union of German Warrior Associations (Kyffhäuserbund der
Deutschen Landeskriegerverbände), founded in 1899, had a country-
wide membership of 1.8 million at the beginning of 1900, and
nearly 2.6 million in 1910. Its name referred to a national symbol,
in the cause of which the most important ‘warriors’ associations’ in
the individual states had rendered great services in the 1890s: the
construction of a monument for William I on Kyffhäuser
Mountain.8
This monument was built between 1892 and 1896 by Bruno
Schmitz, the architect of the ‘official’ emperors’ monuments at the
Porta Westfalica and Am Deutschen Eck near Koblenz; it was
dedicated by Emperor William II on 18 June 1896. The location
was chosen carefully. In the words of the inscription on the
foundation stone: ‘On Kyffhäuser Mountain, under which, as legend
tells, Emperor Frederick Redbeard awaits the renewal of the
Empire, Kaiser William Whitebeard shall rise up, having fulfilled
the legend.’ In his famous 1968 study Nationalidee und
Nationaldenkmal in Deutschland im 19. Jahrhundert,* Thomas
Nipperdey wrote that the Kyffhäuser monument represented a
moment of transition from the ‘monarchical monument’ to the
‘monument of national concentration’ (Denkmal der nationalen
Sammlung):
483
monarchical nation, but above all the powerful and closed nation,
that celebrates itself here in the fulfilment of a mythic history.
484
showed the founder of the Reich either in the uniform of his
cuirassier regiment (in which he had so often appeared before the
Reichstag), or, less frequently, in the freely rendered symbolic
armour of a knight. The man to whom bourgeois and petty-
bourgeois Germany dedicated these monuments was the man of
‘blood and iron’, not the master of diplomacy. Occasionally Roon
and Moltke, representatives of the Prussian soldier-state, appeared
in Bismarck’s company—not only on the Avenue of Victory in
Berlin and in other cities of old Prussia, but elsewhere in the Reich,
too, for example at the imperial palace in Goslar and Am
Friedensengel in Munich.
These monuments to the emperor and chancellor reflected
changes in public consciousness, changes nobody had contributed
more to bring about than Bismarck himself. The concepts of
‘monarchization’ and ‘militarization’ summarize the essentials. The
monuments erected by the Wilhelmine bourgeoisie in honour of the
founding hero of the Reich no longer betrayed anything of the
bourgeoisie’s own liberal origins. It was not within liberalism that
the legacy of 1848 lived on, but within social democracy.
The historian Wolfgang Hardtwig speaks in this context of a
‘deformation of bourgeois political consciousness’ that
accompanied the ‘bourgeoisification of society, culture, and
political culture in the Reich’. Part of this ‘deformation’ was to be
found in the fact that the Volk, whom general and equal suffrage in
parliamentary elections had made into a political factor of no
inconsiderable weight, simply did not appear in the political
iconography of the Reich—or, as in the case of the Niederwald
monument near Rüdesheim (dedicated in 1883), only in the form
485
of victorious or fallen soldiers. A popular cause from the Vormärz
period, the construction of the Hermann monument in the
Teutoburg Forest, was not completed until 1875, and only after
Emperor William I provided the necessary funds. The result was
that this national symbol, too, lacked a specific ‘bourgeois’
character.
When, on 5 December 1894, the keystone was placed in the new
Reichstag building (built by Paul Wallot) on the Königsplatz in the
presence of William II (who referred to the building as the
Reichsaffenhaus, ‘imperial monkey house’, in a personal letter to his
friend ‘Phili’ Eulenburg), the ceremony—just like the laying of the
cornerstone ten years previously—took on the character of a
military spectacle. True to this spirit, the president of the
Reichstag, the conservative delegate Albert von Levetzow,
appeared in the uniform of a Landwehr major. However, the
commentary in the liberal Vossische Zeitung made it clear that this
kind of demonstration would not necessarily come off without
criticism in Wilhelmine Germany. ‘The President of the Reichstag is
the master of the house,’ wrote the ‘Vossin’,
and the highest authority within his chambers. But the major is
subservient to every last lieutenant colonel, to whom he must bow.
The military uniform symbolizes the relation of service and
hierarchical subservience. It has its importance and does honour to
the one who wears it. But everything in its proper place. If the
army were mobilized tomorrow and Major von Levetzow rushed off
to the standards, nobody would take issue with his uniform ...
Yesterday, however, the Major had no business to attend to ... only
486
the President of the German Reichstag was occupied, the freely
elected representative of the people, and therefore we might have
wished that he had done justice to this high honor in the attire of a
free man.
The emperor did not like new parliament building. The large glass
dome bothered him (probably because he saw in it the attempt to
outdo the dome of the palace, the official seat of his rule). Nor was
he happy with the inscription Dem deutschen Volke (‘To the German
People’) that Wallot had chosen for placement over the main
entrance (it was not put up until December 1916, in the middle of
the First World War). When, on a visit to Rome in spring 1893,
William referred to Wallot’s building—under construction at the
time—as the ‘pinnacle of tastelessness’. German architects were not
the only ones who were outraged. The verdict of the self-
proclaimed expert was one of the many comments that ended up
damaging William’s reputation.
As an institution, however, the emperor remained a national
symbol, not least because the Reich did not have many other
symbols. Imperial Germany had no national anthem of its own, but
only the monarchical anthem ‘Heil dir im Siegerkranz’,* sung to the
same melody as ‘God Save the King’. The black, white, and red flag
of the navy and merchant marine was declared the national flag in
1892, but it only achieved a certain popularity in the course of the
propaganda campaign for the new naval policy. Germany also had
no national holiday like the French 14 July, the day of the
storming of the Bastille. The day the emperorship was proclaimed
at Versailles, 18 January, remained first and foremost the
487
coronation day of the Prussian kings; 2 September, ‘Sedan Day’,
was a festival of Evangelical Germany, and 27 January, the
birthday of William II, was too intimately connected to the person
of the reigning sovereign to unite Germany in national celebration.
Nor did the symbols of the Reich overshadow those of the
individual German states, but coexisted with them. This was not
only true of flags, anthems, and monuments, but also of dynasties.
All in all, this hodgepodge of relationships and loyalties was, as
Bismarck had correctly recognized, more beneficial than
detrimental to the cohesion of the new nation. It was true that the
federalism of Bismarck’s Reich was not coequal, owing to Prussia’s
dominance. Yet Bismarck made Prussian hegemony bearable for
non-Prussian Germans. And despite the centralization that set in
under the first chancellor, the Reich did not cease to be
meaningfully federal after 1890. A completely centralized state,
such as envisioned by Heinrich von Treitschke and other National
Liberals, would no doubt have achieved far less national
integration than the federal state created by Bismarck.9
488
this right and utilizing its results, all parties and associations,
regardless of whether or not they considered themselves ‘national’,
were instrumental in bolstering Germans’ sense of themselves as a
single national agent, though one of many voices. This
consciousness grew stronger as the activities and competencies of
the central government—and with them the importance of the
Reichstag—increased. William II’s animosity against the House was
not unfounded.
Parliamentary power struggles depended on many factors, not
least of which was the economy. For example, the fact that the
anti-Semitic parties were able to increase their share of the vote in
the 1893 federal elections from 0.7% (in 1890) to 3.4% and their
number of seats from 5 to 16 was closely connected to the
economic depression of the years 1890–5. In the election of 1898,
an economically good year, they again managed a modest gain of
0.3%. In the next two elections, 1903 and 1907, which took place
during an economic upswing, the anti-Semites fell to 2.6% and
2.2% of the vote, respectively. Party-based anti-Semitism was a
social protest movement. It was divided within itself into, on the
one hand, a radical, aggressively anti-conservative wing around the
former Berlin school director Hermann Ahlwardt, originator of the
battle cry ‘against Junkers and Jews’, and the Hessian archivist and
‘peasant king’ Otto Boeckel, and, on the other hand, a faction
amenable to cooperation with the conservatives and led by Max
Liebermann, former Prussian officer and functionary of the
Federation of Agriculturalists. Political anti-Semitism found support
among peasants, small merchants, craftsmen, and company
employees, who felt themselves equally threatened by large-scale
489
capitalism and the Marxist worker’s movement. As the economic
situation improved, the attractiveness of this kind of anti-Semitism
decreased.
However, the decline of the anti-Semitic parties is not to be
confused with the extinction of anti-Jewish sentiments. Far from it:
anti-Semitism had also been absorbed into the platforms of other
parties and associations. These had the advantage of offering their
constituents much more than the single-issue focus of the ‘pure’
anti-Semitic parties. At the same time, they recognized anti-
Semitism as legitimate, thereby contributing to its social
respectability. For example, in its first official party platform, the
openly agrarian-conservative ‘Tivoli platform’ of 1892, the German
Conservative party committed itself to the battle against ‘the multi-
form, obtrusive, and pernicious Jewish influence on the life of our
people’. To this statement were linked a number of concrete anti-
Semitic demands: ‘a Christian leadership for a Christian people and
Christian teachers for Christian schools’, as well as ‘effective
governmental intervention against any economic activity harmful
to the community and against un-German offences to good faith
and credit in business’.
In conservative discourse, declarations of war against modern
Jewry were accompanied by disavowals of Radauantisemitismus, the
violent, rowdy anti-Semitism preached by the likes of Hermann
Ahlwardt and all purely anti-Semitic parties. According to the
fourth edition of the German Conservative party handbook, which
appeared in 1911, these parties, with their ‘diatribes against all of
Jewry’, only cause Jews to unite more closely together, even
though there are also ‘very many honourable and patriotic Jews—
490
many more than the radical anti-Semites care to admit’. Thus the
anti-Semitism of the German Conservatives was not ‘racist’ in the
‘biologistic’ sense. But that only contributed to its appeal. With
their ‘differentiating’, seemingly moderate Jew-hatred, with its
pretence of cultural scruple, the German Conservatives gave voice
to a widespread feeling.
The Federation of Agriculturalists, closely connected to the
Conservatives, were far less scrupulous in appealing to the anti-
Jewish prejudices of the rural populace and the urban middle
classes. In June 1895, for example, the group’s main publication
described agriculture as the main embodiment of Germandom and,
as such, the irreconcilable enemy of the Jews: ‘Thus it is a natural
fact that agriculture and Jewry must fight against each other, fight
to the death, until the one party is lying lifeless, or at least
powerless, on the ground.’
The German League of Commercial Employees (Deutscher
Handlungsgehülfen-Verband; founded in 1893, the same year as the
Federation of Agriculturalists), which renamed itself the German
National League of Commercial Employees (Deutschnationaler
Handlungsgehilfenverband) in 1896 and initially maintained close
contact with the anti-Semitic German Social Reform Party
(Deutschsoziale Reformpartei) of Liebermann von Sonnenberg and
the Austrian Pan-Germans around Georg Ritter von Schönerer,
expressly excluded ‘Jews and persons shown to be descended from
Jews’ from membership. The societal group targeted by the League
was mockingly referred to as ‘proletarians in stand-up collars’
(Stehkragenproletarier) by many on the left. For their own part, the
majority of commercial, industrial, and bank employees saw
491
themselves as members of a ‘new middle class’ (neuer Mittelstand, a
concept first appearing in the 1890s) or, in part, as ‘private
officials’, but in no way as proletarians. As ‘brain workers’, nothing
was more important to them than emphasizing the distance that
separated them from those who worked with their hands. Since
most ‘blue-collar’ workers had joined the Marxist and
internationalist social democratic movement, it seemed clear to
many ‘white collar’ workers that proletarian internationalism was
part of a Jewish conspiracy. Thus, for the early employees’
organizations, patriotism and animosity towards ‘international
Jewry’ were first and foremost a means of promoting the claim of
their members to a higher social status—through a strict
dissociation from a class movement that, for its part, placed great
value on not being ‘nationally’ minded.10
Yet anti-Semitism was not necessary for a ‘bourgeois’ alternative
to social democracy in dealing with the social question. The
Evangelical minister Friedrich Naumann, originally a follower of
Adolf Stoecker and his Christian Social Workers’ Party
(Christlichsoziale Arbeiterpartei, founded in 1879, renamed the
Christlichsoziale Partei after 1881), finally broke with the latter in
1896 by establishing the National Social Association
(Nationalsozialer Verein). The anti-Semitism of the former court
chaplain (Stoecker had been dismissed from his post by William II
at the end of 1890) played a smaller role in the separation than the
fact that Stoecker’s party had never ceased to form a part of the
German Conservative party, which, ever since the 1892 ‘Tivoli
platform’, sought to become the political arm of the East Elbian
agrarian interests.
492
Naumann had long been the leading voice of the socially
engaged younger members (referred to as die Jungen) of the
Evangelical-Social Congress (co-founded by Stoecker), and a large
number of social reformers followed him when he left. The
National Social Association did not aspire to a radical
transformation of the existing order of society, but it did express its
expectation ‘that the representatives of German education in the
service of the community support German labor in its political
struggle against the predominance of existing vested interests’ and,
on the other hand, that the ‘representatives of German labor be
willing to promote patriotism in child-rearing, education, and the
arts’. The workers were to receive a greater share in the total
production of the German economy; women were to be permitted
access to professions, both private and public, in which they could
‘effectively demonstrate nurturing and educational abilities for
their own sex’. These reform demands were made from ‘the soil of
the nation’, and, as such, were accompanied by a declaration of
support for a ‘reasonable expansion of the German navy’ as well as
for the preservation and extension of the German colonies.
The ‘basic principles’ of the National Social Association, which
were drafted by Naumann, made no mention of the restriction of
Jewish influence. In fact, the manifesto expressly called for the
‘unrestricted preservation of the civil rights of all citizens of the
state’. General suffrage was not only not to be interfered with, it
was to be extended to the parliaments of the individual German
states and organs of communal representation. If there is one
phrase that summarizes the Association’s platform, it is the appeal
for a ‘vigorous cooperation between the monarchy and the people’s
493
representatives’—a distant reminder of Lorenz von Stein’s
‘monarchy of social reform’.
Naumann further developed this idea in his book Demokratie und
Kaisertum,* which appeared in 1900. He had come to set all his
hopes on William II, the ‘emperor of navy and industry’, whom he
considered willing and able to prepare the way for the ‘social
empire’ (soziales Kaisertum) of the future. ‘As Prussian king he has
assumed the legacy of the old tradition; as emperor he is the
national imperator, incarnation of the general will, personal leader
out of an old and into a new age.’ The emperor as executor of the
popular will, overcoming the dominance of agriculture by means of
a ‘dictatorship of industrialism’ and using national power politics to
create the basis for a dissolution of class antagonisms—in
connecting imperialist world politics to the modernization of
society, Naumann showed himself to be an avid pupil of Max
Weber. For his part, however, the sociologist did not share
Naumann’s belief that the emperor was the one called upon to lead
Germany on the path to renewal. For Weber, the democratic
proving ground of national leadership was the parliament, and
Germany should be parliamentarized. Naumann was not long in
joining him in this demand.
Social democratic workers were not won over by Naumann’s
synthesis of national power politics and Christian socialism.
Consequently, they remained deaf to his appeal that social
democracy endorse the emperor’s military and naval policies and
prove ‘that Germany can be governed without the East Elbian and
clerical parties’. This well-intended product of political wishful
thinking resonated strongly with intellectuals, but it found no echo
494
in the electorate. In the federal elections of 1898, the candidates of
the National Social Association, which was running as a party for
the first time, obtained a mere 27,000 votes. In 1903 only one
National Social candidate succeeded in getting into the Reichstag:
the former anti-Semite Hellmut von Gerlach, who joined the
Liberal Association as an auditor.
These failures prompted Naumann to take radical steps. In
August 1903 he caused the National Social Association to be
dissolved, urging the active members to join the Liberal
Association, which most of them did. The two groups did indeed
have a great deal in common. Under the influence of Theodor
Barth, the editor of the Berlin weekly Die Nation, the former
secessionists had abandoned their doctrinaire Manchester
liberalism and turned toward social reform and colonialism.
Naumann’s transformation into one of the leading representatives
of German left-liberalism formed the temporary conclusion to a
development that had begun with a split in the left wing of the
conservatives.11
The turn of the century also saw a ‘national’ turn among the
Catholic part of the political centre. Like the Liberal Association,
the Catholic party voted for Tirpitz’s first navy bill in April 1898. It
had already given its support the year before to a bill authorizing
176 million marks to re-equip the artillery; in 1899 it agreed to a
further increase in the peacetime strength of the army. The Centre’s
motives had nothing to do with military policy. Rather, it hoped its
cooperation with the government in matters affecting the prestige
of the Reich would help bring about the abolition of the remaining
Kulturkampf statutes and enable the German Catholics to be
495
recognized as a ‘national’ party. When the first reading of the
second navy bill was about to take place at the beginning of
February 1900, Cardinal Georg Kopp, prince bishop of Breslau,
demanded in a letter to the delegate Carl Bachem that the Centre
tell Tirpitz it was prepared ‘to treat the bill as favourably as
possible’, but on the condition that ‘the Jesuit law be completely
abolished beforehand, in order to sooth the spirits of the Catholic
voters’.
The cardinal’s plan did not come off. The navy bill was passed
on 12 June 1900 with the votes of the Centre. The only thing the
fraction was able to achieve was a six-year delay in the
construction of six cruisers destined for service abroad. When the
third navy bill went up for a vote in June 1906, the Navy League
exerted such massive public pressure that the Centre, no longer
daring to propose its usual package deal (assent to the naval
office’s bill in exchange for abolition of the Jesuit law), simply
voted in unconditional agreement. To have turned its back on the
‘national’ voting record it had compiled in the previous years
would have cost the Centre its political credibility.
This danger loomed over the Centre above all from the right. As
far as the bourgeois left and the Social Democrats were concerned,
the credibility of the Catholic party had been deeply compromised
ever since it had gone over to the side of the government ‘with beat
of drums and flourish of trumpets’, as Bebel remarked in 1897.
Mommsen went even further in 1902, rounding off his assessment
of the Bismarck years (‘Bismarck broke the back of the nation’)
with the comment: ‘the Centre has taken over the inheritance.’ And
in truth, the political opportunism of the Catholic party was
496
obvious, if not its most characteristic feature. There was no
important political question the Centre did not consider primarily
in tactical terms, deciding its parliamentary behaviour according to
the advantages or disadvantages a given measure would bring for
political Catholicism.
The short-term benefits of cooperating with the government
were obvious. Under Hohenlohe, and under his successor Bernhard
von Bülow until 1906, the Centre party became so powerful that it
began to nurture the belief that, without it, Germany could not be
governed. From this perspective, it made little sense to campaign
for the parliamentarization of the state; a majority government
without the participation of the Centre would clearly achieve far
less for German Catholics than direct negotiations between the
party and the leaders of the Reich, who were appointed by the
emperor and not beholden to the Reichstag. The political cost of
the Centre’s tactics only became evident at a later date, when the
threat of war grew as a result of policies it had supported from the
late 1890s. The parliamentarization of Germany did not depend on
the Catholic party alone, but it was not possible without the
Catholic party. The constitutional-authoritarian German state was
(at least for the time being) strong and sure of itself, and one of the
sources of its security was the fact that the Centre did not question
the status quo.
Under the leadership of Ernst Lieber, the successor of
Windthorst (who died in 1891), the Centre endorsed German
colonialism. Whenever an opportunity presented itself, it
emphasized its loyalty to emperor and Reich. Yet political
Catholicism was never a driving force behind nationalism and
497
colonialism; it simply helped enable German Weltpolitik by
supporting the relevant legislation for its own domestic-political
reasons. With respect to anti-Semitism, the situation was different
and yet similar. The Centre rejected legislation that would have
resulted in the reversal of Jewish emancipation. In November
1880, when the Prussian House of Representatives, in response to a
motion from the Progressive party, debated the collecting of
signatures for the ‘anti-Semite petition’, Windthorst had committed
his party to an unambiguously negative vote on the restriction of
Jewish civil rights. However, this did not mean that politicians and
publicists of the Centre did not share Catholic prejudices against
the Jews. The speeches of Peter Reichensperger and Julius Bachem
during the debate left no room for doubt that they did, and during
the 1890s, the Historisch-politische Blätter demonstrated the same
vehemence in the anti-Semitic campaign as they had done two
decades before.
The decline of party anti-Semitism at the turn of the century
also had an effect on the Centre. To be sure, Catholics continued to
hold Jews in great suspicion, and there was a specifically Catholic
discourse of anti-Semitism, which included the dictum that the
Catholic position represented the only legitimate opposition to the
Jews—opposition based on the Christian religion. Yet at no time
did the Centre consider making anti-Semitism part of its political
platform, as did the emphatically Evangelical German Conservative
party. Catholics knew only too well what it felt like to be a
disadvantaged minority; their own experiences warned them
against active participation in discrimination towards another.
Compared with the Kulturkampf era, the Catholic situation at the
498
turn of the century was comfortable. Catholics were politically
influential and no longer had to feel like they were living in a
fortress under siege. Liberalism was still an enemy, but had lost a
good deal of its former menace. The domestic-political détente
came at a price, however. The last year the Centre managed to gain
more that 20 per cent of the federal vote was 1887. Thereafter it
stagnated at about 19 per cent. The ‘electoral discipline’ of German
Catholics was no longer as high as it had been during the era of
legislative persecution. The Catholic microcosm had lost some of its
cohesive force. Catholic workers, peasants, academics, small
tradesmen, entrepreneurs, and landowners all lived in different
social worlds, held together by a common faith and by Evangelical
anti-Catholicism, which was no less strong—indeed, militant—after
the Kulturkampf had formally passed. (A defensive nationalist
Protestant organization, the Evangelical Union for the Protection of
German-Protestant Interests, was founded with the expressed goal
of keeping up the anti-Catholic political pressure. Its membership
increased from 81,000 in 1891 to 510,000 in 1913.)
It was only in comparison with the fragmented Protestant
political landscape that turn-of-the-century German Catholicism
represented any kind of unified whole, the so-called ‘Catholic
milieu’. In terms of social composition, the Centre was more of a
‘people’s party’ than any other. From the 1890s, bourgeois
politicians became increasingly dominant in the party leadership,
pushing out the nobility. While continuing to support the Catholic
workers’ associations, after 1894 the Centre began to promote the
Christian Trade Unions (Christliche Gewerkschaften). These were not
exclusively Catholic in membership, but inter-confessional, and as
499
such they marked the beginning of a conflict in Catholic political
identity. The Centre had to decide if it saw itself as a confessional
or supraconfessional party, whether it wished to remain a purely
Catholic organization or open itself up to Protestant
participation.12
In the meantime, social democracy was moving—or at first
glance seemed to be moving—in a very different direction. The
expiration of the anti-socialist law on 30 September 1890 restored
to the party of August Bebel the full range of political activities it
had done without for twelve years. However, the vast majority of
social democrats had no intention of renouncing their opposition to
capitalist society and authoritarian imperial government, just as,
for their part, the leaders of the Reich and the federal states did not
cease considering social democracy a force of subversion. Nor did
the proletarian left abandon its anticipation of what Bebel called
the ‘great crash-bang-wallop’ (grosser Kladderadatsch), that is, the
approaching collapse of exploitative bourgeois society, and the
socialist ‘state of the future’ that was to replace it. At the 1891
party congress in Erfurt, Bebel went so far as to predict that the
‘realization of our last goals is so near that there are few people in
this hall who will not live to see it.’
The 1891 Erfurt platform of the Social Democratic Party of
Germany, the basic principles of which were drafted by the editor
of the theoretical party organ Neue Zeit, Karl Kautsky (born in
Prague in 1854, active in the socialist movement from his first
semester at the University of Vienna), was much more ‘Marxist’
than its predecessor, the 1875 Gotha platform, which, to Marx’s
chagrin, had contained a strong residual Lassallean influence. In its
500
very first paragraph, the new platform took up the thesis of the
1848 Communist Manifesto, according to which the development of
the economy would, ‘by necessity of nature’, lead to the downfall
of small private business, the concentration of the means of
production in the hands of a few large capitalists and landowners,
and the swelling of the ranks of the propertyless proletariat with
former small proprietors. Marxist, too, was the claim that class
struggle between bourgeoisie and proletariat would become more
bitter, dividing society into two hostile camps. The same was true
of the demand to make the means of production the common
property of the community at large—a goal that, according to the
text, the working class could not achieve ‘without first obtaining
the requisite political power’.
The pragmatic part of the Erfurt platform, primarily the work of
Eduard Bernstein (born in Berlin in 1850, son of a Jewish railway
engineer; trained as a bank employee; during the anti-socialist era
editor of the weekly Sozialdemokrat, which was published first in
Zurich, then in London), presented the SPD more as a radical
democratic than socialist party. Its demands included general,
equal, and direct suffrage in all elections and ballots, ‘irrespective
of sex’; a ‘general legislature of the people by means of the right to
propose and reject legislation’; a ‘people’s army in place of the
standing armies’; the ‘authority of the people’s representatives to
decide matters of peace and war’; secular schools; the restriction of
religion to the private sphere; the abolition of all laws
‘discriminating against women in public and private law’; the
abolition of the death penalty; and the guarantee of the freedom of
association.
501
One particular concept that played a key role in the thinking of
Marx and Engels did not appear in the Erfurt platform, the
‘dictatorship of the proletariat’. So as to provide no pretext for
further ‘legal’ persecution of social democracy, the document’s
authors also refrained from referring to their goal of the
‘democratic republic’. Nor did they adopt the alternative proposed
by Engels that they demand the concentration ‘of all political
power in the hands of the people’s representatives’. That the text
should contain no mention of ‘revolution’ went without saying.
Yet it was not merely a tactical move when, two years after
Erfurt, Kautsky, who had replaced Wilhelm Liebknecht as party
theoretician by this time, sought to liberate the concept of
‘revolution’ from its connotations of revolt, urban warfare, and
bloodshed. In the Neue Zeit (December 1893), he described social
democracy as ‘a revolutionary, but not a revolution-making party’.
It had no intention ‘of inciting or preparing a revolution. And since
we cannot wilfully make one, we also cannot say anything about
when, under what conditions, and in what form the revolution will
come.’
Kautsky remained firm in his insistence that the radical social
changes envisioned by social democracy could only be achieved ‘by
means of a political revolution, through the conquest of political
power by the fighting proletariat’. Unlike its situation in backward
Russia, however, the proletariat in the ‘modern states’, thanks to
the freedom of association, freedom of the press, and general
suffrage, now had other and better weapons at its disposal than the
revolutionary bourgeoisie had in the eighteenth century. When the
proletariat availed itself of democratic institutions, it did not cease
502
to be revolutionary.
503
done so would have risked the unity of the party and given impetus
to ‘anarchists’ or a new anti-parliamentary opposition like the
‘young’ fraction (die Jungen) thrown out of the party in 1891. This
was the reason Kautsky sharply contradicted his friend Eduard
Bernstein in 1897 when the latter, writing from London (where he
had moved in 1888 after his expulsion from Switzerland), began to
openly criticize the teachings of Marx and Engels.
Inspired by debates in the London Fabian Society, a circle of
intellectual social reformers around George Bernard Shaw and
Sidney and Beatrice Webb, Bernstein called fundamental premises
of Marx’s and Engel’s ‘scientific socialism’ into question, first in a
series of articles entitled ‘Probleme des Sozialismus’* in the Neue
Zeit, then in a summarizing communication to the Social
Democratic party congress in Stuttgart in October 1898. Citing a
wide range of evidence, Bernstein argued that the ‘catastrophe
theory’, the prognosis of the inevitable collapse of bourgeois
society, as well as the doctrines concerning the deterioration of
societal conditions, the end of small business ownership, and the
disappearance of the middle classes, had all been disproved. His
political conclusions were radical:
The more ... the political institutions of the modern nations are
democratized, the more the need and opportunity for great political
catastrophes are decreased ... The conquest of political power by
the working class and the expropriation of the capitalists are not
ultimate ends in themselves, but only a means to the attainment of
certain goals and aspirations.
504
In response to the first objections of Kautsky and other
theoreticians, among them Viktor Adler in Vienna, Bernstein, in a
letter to the party organ Vorwärts, further sharpened his theses:
505
invoked the spirit of Kant, the critic of pure reason, against the
‘comfortable shelter’ of Hegelian dialectic. Then, on the final pages
of his study, Bernstein once again made clear what, more than
anything else, separated him from the left of his party:
506
‘objectively necessary’. Bernstein’s revisionist theory was a ‘theory
of socialist stagnation, justified with a vulgar-economic theory of
capitalist stagnation’—the ‘first, but also the last attempt to provide
opportunism with a theoretical basis’.
Luxemburg considered Bernstein’s understanding of reform and
revolution, democracy and dictatorship of the proletariat
completely undialectical.
Legislative reform and revolution are ... not two different methods
of historical progress one can select like hot sausages or cold
sausages from the dinner table of history. They are different
moments in the development of class society, conditioning and
completing, yet at the same time excluding each other, like, for
example, south pole and north pole, like bourgeoisie and
proletariat.
507
reformers, Rosa Luxemburg saw them with the eyes of a
revolutionary coloured by the experience of tzarist Russia.
Bernstein had a tendency to underestimate the resistance faced by
parliamentary and democratic reformers in Hohenzollern Germany.
Luxemburg overestimated the revolutionary potential in the
international and German proletariat, and she underestimated not
only the opportunities opened up to social democracy by general
suffrage, parliamentary politics, the right to strike, freedom of
speech, and freedom of the press, but also the consciousness-
transforming power of these achievements.
Karl Kautsky, the third important participant in the great debate
over democracy and dictatorship, reform and revolution (a debate
that was to have a great influence well into the twentieth century),
found himself in a dilemma. On the one hand, he had begun to
relativize the revolutionary aspect of Marxism even before
Bernstein. As far as the possibility and desirability of an
evolutionary transition from capitalism to socialism was concerned,
he was far closer to ‘rightist’ Bernstein than ‘leftist’ Luxemburg. But
on the other hand, as the foremost theoretician of the SPD, Kautsky
considered it his duty to hold the social democratic movement
together as a party. A particular understanding of Marxism had
established itself in the party, based, in not inconsiderable part, on
the reading of Engels’s Anti-Dühring and Bebel’s Die Frau und der
Sozialismus,* which had appeared in 1878 and 1879, respectively.
Bernstein was now calling the fundamentals of that understanding
into question, and Kautsky had to take steps against him. He could
do so in good conscience, since Bernstein’s account of Marxism
was, in many ways, more like a caricature. Kautsky interpreted
508
Marx in a different way than Rosa Luxemburg, but he agreed with
her that Marx’s theory was still the correct one and that it should
therefore still be the intellectual and practical guide for the
international workers’ movement.
In his 1899 book Bernstein und das Sozialdemokratische
Programm,† Kautsky accused his colleague of fabricating a Marxist
‘theory of collapse’ that had little to do with what Marx had
actually written; even the ‘theory of immiseration’ did not appear
in the writings of Marx and Engels, but was invented by the
opponents of Marxism. Kautsky held firm to the thesis concerning
the progressive concentration in industry and agriculture, and, in
consequence, to the idea of the long-term inviability of small
business. In order to be victorious, therefore, social democracy
would have to remain faithful to Marx’s clear historical theory and
the ultimate goal of social revolution. The alternative was
‘muddling through on a case by case basis’, which would
necessarily result in the proletariat losing the consciousness of its
great task. Nonetheless, social revolution and rebellion were to be
sharply differentiated: ‘Social revolution is a goal we can set by
way of principle; rebellion is a means to an end, a means we can
only ever evaluate according to practical purposes.’
In contrast to Bernstein, Kautsky drew a clear line of distinction
between socialism and liberalism. ‘A progressive democracy is, in
an industrial state, now only possible as a proletarian democracy.
Hence the decline of progressive bourgeois democracy ... Today,
only a firm belief in the necessity of the rule of the proletariat and
in its political maturity can lend power of attraction to the
democratic idea.’ For the rest, Kautsky exhorted social democracy
509
to have hope: ‘What has become the strongest party within three
decades can, within another three decades, become the ruling
party, maybe even sooner.’15
Kautsky’s position represented that of the ‘party centre’ around
Bebel and became the official viewpoint of the SPD. In October
1899, even before the appearance of Kautsky’s ‘anti-critique’, the
Hanover party congress had overwhelmingly passed a resolution
proposed by Bebel rejecting any change in platform or tactics and
specifically targeted at any endeavour ‘attempting to conceal and
shift its stance on the existing order of state and society and on the
bourgeois parties’. The resolution did not oppose cooperation with
bourgeois parties ‘on a case by case basis’, as long as the goal was
‘to strengthen the party in elections, or to extend the political
rights and liberties of the people, or to seriously improve the social
position of the working class and promote cultural tasks, or to
combat activities aimed against the workers and the people’. Even
in such cases, however, the SPD insisted on maintaining ‘its
complete autonomy and independence’ and on regarding every
partial success merely as one step ‘bringing it closer to its final
goal’.
Neither the Hanover resolution nor Kautsky’s book brought the
revisionist conflict to an end. It occupied future party congresses
and reached, with the sharp repudiation of revisionism at the
Dresden congress in 1903, more of a climax than a denouement.
Bernstein, who returned to Germany in 1901 and was elected to
the Reichstag in a by-election the following year, along with his
comrades-in-arms, among them Joseph Bloch, publisher of the
Sozialistische Monatshefte, the agrarian revisionist Eduard David,
510
and the Berlin attorney Wolfgang Heine, had no intention of
renouncing arguments that had, in part, long been scientifically
confirmed—like the thesis concerning the growth of the middle
classes and the emergence of a new kind of small business, the
repair shop.
More important than the small group of avowed ‘revisionists’,
however, was the even greater number of influential ‘reformists’,
pragmatic men like the leaders of the Free Trade Unions (Freie
Gewerkschaften) under Carl Legien, head of the general commission,
or the chairman of the Bavarian Social Democrats, Georg von
Vollmar. It was Vollmar who, in his ‘Eldorado’ speeches in Munich
in June 1891, was the first to exhort the party to move ‘from the
theoretical to the practical, from the general more to the particular’
and not to forget ‘present-day, immediate, urgent concerns’ in a
single-minded focus on the ‘future’—that is, to engage in
realpolitik. Conflicts over principle, like the one between Kautsky
and Bernstein, held little interest for the ‘reformists’. What they
were interested in was exploiting the political options available to
them without undue concern for the purity of tradition and dogma.
The majority of the party delegates, officials, and members
stood with Bebel. Most Social Democrats approved of the
parliamentary fraction’s decision to renounce its right to a vice-
presidential position, since part of the protocol of the Reichstag
presidency was ‘going to court’, that is, official audiences visits
with the emperor. At times, the party was even more doctrinaire
than its head. At the 1895 party congress in Breslau, an agrarian
programme advocated by Bebel, which would have opened the
party to small farmers, was voted down by a large majority of
511
delegates. The industrial workers feared an attenuation of the
party’s ‘proletarian’ identity. Unlike the Catholic milieu, social
democracy was socially homogeneous. It had emerged from the era
of anti-socialist legislation more firmly unified than Catholic
Germany during or after the Kulturkampf. The social democratic
workers had developed a proletarian class consciousness. And if
class consciousness is a necessary part of social class, then the
working class was the only one in Germany.
In shielding itself from the rest of society, the social democratic
milieu was, to a certain extent, immune from the Wilhelmine
zeitgeist. It was anti-militaristic, anti-nationalist, and—at least in
principle—anti-colonialist. Anti-Semitism was uncommon, more so
than in any other social stratum in Germany. Yet social democrats
did not only reject and combat anti-Semitism, they also habitually
underestimated it. The dictum that anti-Semitism was ‘socialism for
imbeciles’ gave expression to a foolish optimism. A party pamphlet
of the early 1890s put it in the following way:
512
The issue of anti-Semitism was connected with a crisis in French
socialism that also rapidly affected the German Social Democrats,
the uncontested leaders of the Second International founded in
1889. In May 1899 the Socialist Alexandre Millerand entered the
radical-republican Waldeck-Rousseau cabinet, leading to the
formation of a leftist government. The cas Millerand, the first
instance of ‘ministerialism’ in the history of the European workers’
movement, was a crisis ancillary to the Dreyfus affair, the scandal
around the Jewish Captain Alfred Dreyfus, who was falsely accused
of being a German spy, demoted, and exiled from France for life.
The reform-minded forces within French socialism, led by Jean
Jaurès, defended Millerand’s decision, pointing to the danger from
the right. For the supporters of the orthodox Marxist Jules Guesde,
on the other hand, Socialist participation in a bourgeois
government was absolutely irreconcilable with the doctrine of class
struggle.
In September 1900 the congress of the International in Paris
dealt with the French conflict. It passed a resolution drafted by
Kautsky declaring the ‘dangerous experiment’ of Socialist
participation in a bourgeois government acceptable only as a
‘temporary and exceptional emergency measure’. But this
formulation was not strict enough for Guesde and his friends. At
the Amsterdam congress of the International in 1904, they
presented another resolution (which had already been adopted by
the Dresden congress of the SPD the year before), according to
which the party, in the spirit of the ‘Kautsky resolution’, ‘cannot
aspire to share in the authority of the government within bourgeois
society’. The congress adopted this resolution by 25 to 5 votes and
513
12 abstentions.
Unlike in the Third Republic, formal coalitions were not possible
in the constitutional German Reich. Ever since the commercial
treaties of the Caprivi era, it was considered acceptable for the
Social Democrats to support individual government bills or
proposals by the bourgeois parties, if such a decision could be
justified in terms of the interests of the workers or the
improvement of society. To approve budgetary measures, on the
other hand, was treated by the party leadership as tantamount to a
vote of confidence in the government, and was therefore rejected.
In 1894, when Social Democrats in Bavaria, Baden, and Hesse
nonetheless voted for a state budget for the first time, they were
severely reprimanded by Bebel. The Dresden resolution of 1903,
which was ‘internationalized’ a year later by the Amsterdam
congress, went still further. By rejecting participation in any kind
of coalition, it made the parliamentarization of the German Reich
even less of a possibility than it already was.
The decision highlighted one of the main contradictions within
the politics of German social democracy. The combination of a
radical theory and a practice aimed at achieving small
improvements had led the party to an impasse. With its militant
refusal to participate in government, the SPD blocked precisely the
kind of constitutional transformation it should have been its
highest interest to promote. A revolutionary alternative to the
parliamentarization of the Reich existed only in the imaginations of
the party left around Rosa Luxembourg, Karl Liebknecht, Clara
Zetkin, and Franz Mehring. It did not exist in reality. Social
democracy was in the process furthering the isolation in which its
514
adversaries sought to confine it.17
515
constituency. The conservatives benefited from the parliamentary
support of the chancellor not only through higher duties on grain
(the ‘Bülow tariff’ of 1902); in 1905, after a bitter struggle, they
also managed to block the construction of the decisive connecting
section in the canal between the Elbe and Hanover. This was the
greatest victory for the agrarian interests of the Reich to date. The
majority of the parliament bowed to the pressure of the East Elbian
Junkers, who opposed the canal for no other reason than to keep
transportation costs for cheaper foreign grain as high as possible.
In December 1906 the governmental majority collapsed over a
controversial issue of colonial policy. A native rebellion in German
South-west Africa—first by the Herero, then by the Hottentots—
prompted the government to petition the Reichstag for additional
expenditures for the colonial troops. In the meantime, however, the
behaviour of the colonial administration and military had come in
for massive criticism at home. The battle against the Herero had
turned into a campaign of annihilation. Since not only the Social
Democrats, Guelfs, Alsace-Lorrainers, and Poles spoke out against
the government’s request, but also the Centre, under the influence
of its left wing around the Württemberg delegate Matthias
Erzberger, a narrow majority came together against the bill.
Bülow responded by dissolving the Reichstag. The ensuing
election campaign was conducted by the government, German
Conservatives, Free Conservatives, National Liberals, and the two
liberal parties with nationalist slogans and a common front against
the Social Democrats and Centre party. Since the ‘national’ forces
supported each other in the run-off elections, they achieved a
resounding victory in the second ballot of the ‘Hottentot election’
516
on 5 February 1905. The parties of the new liberal-conservative
‘Bülow bloc’ gained a total of 220 seats, the parties opposing the
colonial policy only 177. The real loser was the SPD. Despite small
increases in the number of votes they obtained, the Social
Democrats lost nearly half of their power, falling from 81 to 44
seats.
The pact with the ‘cartel parties’ of 1887 allowed the left-
liberals to achieve their primary tactical goal, the removal of the
Centre party from the governmental camp. Liberalism and
conservatism came together over a minimal ‘national’ consensus
consciously excluding Social Democrats and Centre. Yet
nationalism could not serve as the basis of a true reform policy.
The most important legislative achievement of the ‘Bülow bloc’, the
federal law of associations (Reichsvereinsgesetz) of April 1908, was a
compromise. Liberals scored important successes: police authority
in the dissolution of assemblies and associations was abolished, and
most restrictions on political associations were removed. Women
and young people over 18 years of age now finally had the right to
become members of political associations and to participate in
political assemblies. But the law also contained the controversial
‘languages paragraph’ (Sprachenparagraph), which mandated the
use of German for all public occasions, with the exception of
international conferences. In areas of mixed nationality (Posen,
northern Schleswig, and parts of Alsace-Lorraine), an interim
regulation applied, allowing the use of the native language for a
period of twenty years.
On the international stage, the German Reich under Bülow
drifted yet further into political isolation. The construction of the
517
Baghdad railroad, a consortium of German and French banks and
industrial concerns, provoked Great Britain and Russia. In 1904
London concluded an entente cordiale with Paris. On the face of it,
the two powers did nothing more than agree that Morocco would
be in the French, Egypt in the English sphere of influence.
Germany, however, felt passed over and affronted. When, shortly
afterwards, Russia suffered a serious defeat in its war with the
Japanese, the leaders of Germany decided that the hour had come
to teach France, the most important ally of the tzarist regime, a
lesson. At the end of March 1905 William II, wishing to emphasize
the German interest in Morocco, landed at Tangier. When two
months later the French foreign minister, Delcassé, the architect of
the anti-German policy of the Quai d’Orsay, was removed from
office, many Germans saw the high-stakes Weltpolitik of their
government confirmed.
But then Britain, the other ally of France, entered the scene.
When Germany demanded an international conference on Morocco
(held in April 1906 in Spanish Algeciras), British diplomacy saw to
it that Berlin was completely isolated from the start. Germany
reacted by stepping up naval production, which provoked Britain
anew. London responded by concluding a treaty with St Petersburg
in which the two powers came to an understanding about their
respective spheres of interest in the Near and Middle East. What
Germans felt to be ‘encirclement’ was the result of Bülow’s ‘free
hand’ policy—the Reich had done everything in its power to bring
the other great powers (except for Austria-Hungary) closer
together. The alliances of 1904 and 1907 were, first and foremost,
the fruits of German foreign policy.
518
When Britain proposed to Germany a mutual limitation of naval
expenditures in 1908, the emperor saw to it that the overture was
brusquely rejected. At the same time, the situation in the Balkans
was reaching a crisis point. In the autumn of 1908 Serbia
responded to the annexation of Bosnia and Herzegovina by Austria-
Hungary (which had occupied these still formally Ottoman
territories since 1878, after a decision by the Congress of Berlin)
with the mobilization of its armed forces. Vienna, with the firm
support of Berlin, demanded good behaviour from Belgrade. Serbia
and its ally Russia were forced to bow to the pressure of the central
powers and recognize the annexations. Thus the Bosnia crisis
seemed to end with a German success. In the long term, however,
Germany’s position deteriorated. Russia now increasingly sought
the support of the western powers, and Wilhelmstrasse had
committed itself to Austria-Hungary’s Balkan interests in a way that
was bound to compromise and endanger German diplomacy.
The most serious crisis of the Wilhelmine empire, the Daily
Telegraph affair, also occurred during the era of the ‘Bülow bloc’.
On 28 October 1908, the London Daily Telegraph published an
interview William II had granted to the English colonel Sir Edward
James Stewart-Wortley during German military manoeuvres in
Alsace. The emperor asserted in this conversation that the majority
of Germans did not share his Anglophilia; that he had prevented a
coalition of the Continental powers from forming against Britain
during the Boer War of 1899 and had sent to his ‘revered
grandmother’ Queen Victoria a war plan that the British had then
obviously followed; and, finally, that the German fleet that so
exercised the British was only a means of gaining worldwide
519
respect for Germany, above all from Japan and China, as well as in
the Pacific. ‘It may even be that England herself will be glad that
Germany has a fleet when they speak together on the same side in
the great debates of the future.’
Following the strict constitutional and bureaucratic protocols to
the letter, William II, in East Prussian Rominten at the time, had
sent the text of the interview to the chancellor, who was taking the
waters on the island of Norderney. Bülow, without reading the
document, sent it to the foreign office, where a privy councillor by
the name of Klehmet—the secretary and the chief of press were
absent—dealt with it. Klehmet did not see fit to question the
propriety of the emperor’s words, making only a few minor
corrections. From the undersecretary, who also did not think it
necessary to read the text himself, the manuscript travelled back to
the chancellor. According to his own testimony, Bülow once again
refrained from reading the interview, returning it to the emperor
along with Klehmet’s emendations. William then sent it off for
publication.
520
Like a punch in the ribs, the English conversations of William II
violently reminded the nation of all the political errors the Emperor
had committed in the twenty years since the beginning of his rule,
and of all the wrathful prophecies of the deposed Prince Bismarck.
Something like a dark foreboding was felt in the widest circles, a
premonition that such careless, brash, imprudent, indeed childish
talk on the part of the head of the Reich could ultimately lead to
catastrophe.
521
the Federal Chancellor in the Federal Parliament and assures Prince
von Bülow of his continuing confidence.’
Regardless of how William really felt about his chancellor, in
the autumn of 1908 he was unable to get rid of him. Bülow
therefore had the opportunity to try his hand at the long-overdue
‘great reform’ of the imperial finances—and to fail. On 24 Juny
1909 the German Conservatives, Centre, and several smaller groups
successfully blocked the centrepiece of the reform, the estate tax
bill. There were, to be sure, good conservative as well as Catholic
reasons to claim that the estate tax was hostile to property and
family. But the rejection of the bill also had to do with factors that
went far beyond the immediate legislative context. The Centre had
not forgotten Bülow’s role in the ‘Hottentot election’, and the
Conservative vote was influenced by a strong ultra-royalist faction
around the West Prussian landowner Elard von Oldenburg-
Januschau that accused the chancellor of disloyalty to the emperor.
On 26 June, two days after the parliamentary failure of the
Reich leadership, Bülow asked William II to allow him to resign
from his post. The emperor, who had momentarily considered
dissolving the Reichstag and even staging a coup d’état, gladly
fulfilled his request, but wished Bülow to remain in office until the
end of parliamentary debate over the other finance bills. An early
dismissal would have made it seem as though the emperor were
merely fulfilling the will of the Reichstag majority and tacitly
assenting to parliamentary monarchy. In order to avoid this
undesirable impression, Bülow was not dismissed until 14 July
1909, two days after the Reichstag had passed the remaining tax
laws.18
522
THE PARTIES AND THE QUESTION OF
PARLIAMENTARY RULE
523
constitution. Thus a veto from Prussia would have sufficed, and
there was no doubt about where Prussia stood on the question of
parliamentary rule. The Hohenzollern dynasty with the emperor
and king on the throne, the army, the Junkers, and the high
bureaucracy were all resolved to defend their power and interests.
If the Reichstag had really forced the issue, the consequences
would have been very serious indeed.
A conflict of similar proportions would have resulted if the
opponents of Prussian three-class suffrage had banded together to
fight for the introduction of the federal voting law in the largest
German state. The injustice of the Prussian system was evident in
the relationship between the number of votes and the number of
parliamentary seats received by the Social Democrats. In 1903, the
second time it ran for election to the Prussian House of
Representatives, the SPD obtained 18.8% of the vote, but no seats.
In 1908 the party, supported by 23.9% of the vote, entered the
House for the first time, but received a mere 7 of 443 seats. It went
without saying that the Social Democrats were the most vocal
proponents of general direct suffrage in Prussia. They were joined
only by the left-liberals. The Centre vacillated between the federal
model and plural suffrage system friendly to propertied interests.
Among the Prussian Free Conservatives and National Liberals there
were probably supporters of plural suffrage, but no politicians who
would have considered replacing the three-class with the federal
system. The German Conservatives were uncompromising
champions of the status quo and its main beneficiaries. With 14%
of the vote in 1908, they obtained 143 seats, 34% of the total.
As progressive as federal suffrage was in comparison to the
524
system in Prussia, it had one great deficiency: it only applied to
men. Of all the political parties, only one fully endorsed equal
political and social rights for women, the Social Democratic Party.
Left-liberals were divided on the question. The followers of
Friedrich Naumann were firm supporters of equal rights, but they
were in the minority. In its March 1910 platform, the Progressive
People’s Party demanded the ‘extension of the rights of women and
their employment opportunities’ as well as ‘the right to vote and to
stand for election to shop and commercial tribunals’. Significantly,
however, nothing was said about parliamentary suffrage. The
National Liberals had even greater reservations about women’s
rights, and the Centre Party and the Conservatives generally
considered politics an area in which women had no business.
Things were not very much different within the women’s
movement itself. Women in the socialist movement, led by Clara
Zetkin and Luise Zietz, fought for equal rights for women in all
areas of social and political life, including the right to vote and to
be candidates for election (active and passive suffrage). The
bourgeois women’s movement was divided. For the left wing
around Minna Cauer, women’s suffrage was an indispensable part
of democracy. The moderates, led by Helene Lange, saw the vote
not as part of their immediate political agenda, but as a goal to be
achieved by women in the distant future and as a reward for
especial merit. When, in 1907, the non-partisan League of German
Women’s Associations (Bund Deutscher Frauenverbände) demanded
active and passive women’s suffrage, it did not address the question
of the legal framework (Reichstag system, three-class, or plural
suffrage) in which the goal was to be realized. Of all the bourgeois
525
women’s movements, none was less radical than the German.19
By the first decade of the twentieth century, however, it could
also no longer be said that the German workers’ movement was
especially radical. When revolution broke out in Russia in 1905
(first in St Petersburg, then in large parts of the country) in the
wake of the Russo-Japanese war, leftist socialists in the west
believed it to be the beginning of a new cycle of revolutions. It was
true that Russia was considered the most backward country in
Europe. For that very reason, however, Engels had predicted in the
1880s that ‘the country is approaching its 1789’ and that ‘when the
year 1789 has once begun there ... the year 1793 will not be far
off.’ Engels predicted that the ‘avant-garde of the revolution will go
on the offensive’ in Russia; indeed, he considered the tzarist empire
to be one of the ‘exceptional cases’ ‘in which a handful of people
manage to effect a revolution’. If there was one place ‘Blanquism’
might conceivably work (the ‘fantasy’ of the French Communard
Louis Auguste Blanqui, according to which a small group of
conspirators could overturn a whole society), it would be,
according to Engels, St Petersburg.
What the young Marx had expected from Vormärz Germany—
that its backwardness would bring forth an especially radical
revolution—the late Engels thus hoped for, at least momentarily,
from Russia. Russia would make up for the lack of a revolutionary
situation in the developed west, thereby driving forward the
revolutionary agenda on the international stage. Once again, the
revolutionary idea migrated from west to east. In 1843–4, Marx
had proclaimed Germany the heir of the French revolutionary
tradition. Four decades later, Engels continued the tradition of the
526
‘translatio revolutionis’ in ascribing a special revolutionary mission
to Russia.
After 1903 there really did exist a Marxist ‘revolutionary avant-
garde’ in organized form in Russia, the group of ‘Bolshevist’
professional revolutionaries around Lenin. The party was formed
when the Russian Social Democratic Workers’ Party split into a
majority (Bolsheviki) and a minority (Mensheviki) faction. However,
Lenin and the Bolsheviks were not the ones who fascinated the
German and west European left in 1905, but rather the Russian
proletariat and its mass strikes. Rosa Luxemburg—who in 1904 had
accused Lenin of seeking to eliminate the initiative of the masses
for the sake of the tightly organized, centralized rule of a
‘Blanquist’ coterie—concluded from the revolution of 1905 (she
was writing about a year later) that the mass strike was the ‘form of
movement of the proletarian masses, the manifestation of the
proletarian battle in the revolution’, and thus that it was ‘inseparable
from the revolution’.
From her point of view, the backwardness of Russia was no
reason not to learn from the Russian experience. On the contrary:
‘Precisely because it has delayed its bourgeois revolution so
unforgivably long, the most backward country can reveal to the
proletariat of Germany and the most advanced capitalist countries
ways and means of extending class struggle.’ Unlike in Russia, the
bourgeois order had had a long history in Germany; but since it
had been completely exhausted in the meantime and since
democracy and liberalism had had time ‘to become extinct’, then,
‘in a period of open political warfare in Germany’, it could ‘only
come down to the dictatorship of the proletariat as the ultimate,
527
historically necessary outcome’.
Rosa Luxemburg’s staunchest adversary within the party, Eduard
Bernstein (whom she called an ‘opportunist’), neither accepted her
dialectics of backwardness nor had any interest in preparing the
way for a ‘dictatorship of the proletariat’. He, too, regarded the
mass strike as a legitimate political weapon, the modern alternative
to old revolutionary barricade-warfare, which he considered
‘obsolete’ (Engels wrote something very similar in 1895, shortly
before his death, in the introduction to the new edition of Marx’s
Klassenkämpfe in Frankreich). Unlike the radical left around
Luxemburg, however, Bernstein did not think of the political mass
strike as an instrument to set the proletarian revolution in motion,
but as a means of achieving democratic goals (such as the abolition
of the Prussian House of Lords or the reform of Prussian suffrage)
or defending against assaults on already existing democratic
achievements (like the Reichstag suffrage law).
In an address to the party at the Bremen congress of the SPD in
1904, Bernstein said that although he had the reputation of being a
‘moderate comrade’ and considered the name a designation of
honour, nonetheless moderation was not to be confused with
weakness or lack of energy. If an attempt on the Reichstag voting
law should be made, then the thought that the German workers
‘may under no circumstances renounce this right must be so
powerful that they use all forms of resistance at their disposal. And
if they should be defeated, then it is better to go down with honour
than to let the vote be taken away without any attempt to resist.’
With this statement, the revisionist Bernstein came very close to
the Engels of 1895. Engels, too, regarded general, direct suffrage as
528
an well-nigh revolutionary achievement allowing the workers’
movement—at least in developed societies— to wage class struggle
in new and more ‘civil’ forms and to proclaim an end to the ‘era of
surprise attacks, of revolutions made by a small, conscious minority
at the head of unconscious masses’. The Free Trade Unions,
however, who were as indispensable for political mass strikes as
they were for industrial action, had no intention of letting
politicians and theoreticians dictate what they should or should not
do, no matter whether they were called Luxemburg or Bernstein,
Engels or Bebel. With a membership of 1.6 million in 1906, nearly
five times as large as the SPD (which had 384,000 members at the
time), the social democratic trade unions were self-confident
enough to give highest priority to preserving their own
organization and avoid subordination to the party. At a trade union
congress in Cologne in May 1905, an overwhelming majority of the
delegates declared it ‘reprehensible’ ‘to attempt to prescribe a
particular tactic by agitating for political mass strikes’. A general
strike, such as was endorsed by ‘anarchists and people with no
experience in the business of economic warfare’, was ‘out of the
question’.
Since the Social Democratic left responded to the Cologne
resolution with aggressive attacks against the Free Trade Unions,
the SPD could also no longer avoid taking a position on the issue of
mass strikes. In September 1905 a large majority of the Jena party
congress passed a resolution drafted by Bebel declaring it the duty
of the entire working class, ‘in particular in the case of an attack on
the general, equal, direct, and secret suffrage law or on the right to
coalition ... to employ any means that seems appropriate for their
529
defence’. The congress designated the ‘comprehensive mass
walkout’ as ‘one of the most effective weapons to prevent such a
political crime against the working class or to secure an important
basic right for its liberation’. However, in order that this weapon
might be made available and used to greatest effect, ‘the greatest
possible extension of the political and trade union organization of
the working class is absolutely necessary, as well as vocal
agitation.’
With the expression ‘most effective weapon’, the party
leadership of the left sought to emphasize the defensive purpose of
the political mass strike, thus accommodating moderates and the
trade unions. But the Free Trade Unions, seeing the Jena resolution
as a paternalistic attempt to co-opt their voice, resisted. After a
series of confidential negotiations in February 1906, the leadership
of the SPD agreed to a near-total self-correction: a general strike
was not called for at the present time, and the party leadership
would not call for one without the assent of the general
commission of the Free Trade Unions.
That was not yet the end of the debate, however. Through an
indiscretion, the confidential agreement was leaked to the public,
causing a storm of outrage among the left wing of the party. The
following party convention, which took place in September 1906 in
Mannheim, passed, after heated exchanges, a resolution drafted by
Bebel and Carl Legien, the head of the general commission of the
Free Trade Unions. The so-called ‘Mannheim agreement’ confirmed
the Jena resolution, but added that the latter did not contradict the
Cologne resolution, and obliged the party leadership to consult the
general commission ‘as soon as it considered the necessity of a
530
political mass strike to exist’. In sum, the result was a victory for
the unions. The losers were not only the left-wing radicals, but also
committed revisionists like Bernstein. After the 1906 SPD
convention, political strikes for the democratization of Prussia were
off the agenda for the time being.
The nature of the Mannheim ‘compromise’ corresponded to the
distribution of power within the social democratic workers’
movement. The party centre under Bebel, trying to reconcile the
differences between ‘left’ and ‘right’, between the party and the
unions, often could only do so with the help of a rhetoric that
sounded more radical than was intended. The party’s theoretical
basis remained Marxist in the sense of the Erfurt platform of 1891.
The day to day practice, however, was completely unrevolutionary,
though in no way consistently reformist. The SPD fractions in the
southern German state parliaments, repeatedly voting for state
budgets against the party line, were regularly criticized, and when
in 1909 the Baden Social Democrats formed a ‘grand bloc’ with the
National Liberals and Liberals in order to push the Conservatives
and Centre into the minority, their actions were regarded by most
of the federal SPD as a betrayal of sacred principle.
In fact, under the prevailing conditions, the policy of the Baden
Social Democrats was realistic and successful. It could not easily be
transferred to the federal level, however, and to Prussia, so
intimately linked to the Reich, not at all. A strategy with the goal
of moving Germany and Prussia closer to democracy could not be
discerned anywhere in the SPD. Everything undertaken by the
party leadership was calculated to improve the material
circumstances of the workers, to hold the party together, and to
531
improve its performance in the next election. Consequently, the
mixture of revolutionary doctrine and practical renunciation of
revolution made a certain sense. Any real decision, whether for
revolution or reform, would have split the party.20
The Centre found itself in a similar predicament at around this
same time. After the turn of the century, differences of opinion
with regard to the confessional character of the Christian Trade
Unions, which had been supported by the Centre since their
founding in the 1890s, led to a conflict (the so-called Zentrumsstreit)
over the question of whether the Centre should consider itself a
Catholic-clerical party or an interconfessional political party open
to Protestants and independent of the clergy. The former position
was represented by the Berlin ‘integralists’ around the Reichstag
delegate Hermann Roeren, the latter view by the reformers of the
‘Cologne school’ around Julius Bachem, publicist and editor of the
Kölnischer Volkszeitung. The integralists could count on the support
of the pope, Pius X, the German episcopate, and the Berlin branch
of the league of Catholic workers’ associations. The reformers
found allies in the Goerres Society for the Cultivation of Science in
Catholic Germany (founded in 1876) under its leader, the Munich
philosophy professor Georg Baron von Hertling, the Christian Trade
Unions, and in the People’s Association for Catholic Germany,
which counted over 650,000 members in 1910, the tenth year of its
existence.
The conflict escalated in spring 1906. Julius Bachem published
in the Historisch-politische Blätter für das katholische Deutschland an
essay entitled ‘Wir müssen aus dem Turm heraus!’* calling for an
532
end to the sequestration and confinement implicit in the popular
image of the ‘Centre tower’.
533
but confirmed the status quo.
The battle over the self-image of the Centre party died away
after the autumn of 1909, but the trade union controversy
continued. In the encyclical Singulari quadam of September 1912,
Pope Pius X did not expressly forbid interconfessional Christian
Unions, but he so clearly stressed Catholic unions as the rule that it
was tantamount to an endorsement of the integralist position. The
Christian Trade Unions, which counted 340,000 members by 1913,
responded with massive protests, eventually forcing the German
bishops to reinterpret the encyclical to accord with the prevailing
conditions. The beneficiaries of the conflict between integralists
and workers were the bourgeois politicians of the Centre, to whom
Matthias Erzberger now also went over. The ‘Catholic milieu’ had
become further subdivided in the course of the conflict—so much
so, in fact, that the curia and the episcopate could no longer
compel acceptance of the integralist programme. The conservatives
had lost influence. They were, however, not yet defeated.21
While Centre and Social Democrats were dealing with factional
disputes, liberalism was enjoying a political renaissance of sorts. In
June 1909 the liberal forces within German industry, weary of
constant concessions to East Elbian agriculture, joined together to
form the Hanse League for Commerce, Trade, and Industry (Hansa-
Bund für Handel, Gewerbe und Industrie). Three-quarters of a year
later, the three left-liberal parties united to form the Progressive
People’s Party. Among the National Liberals, too, there were signs
of a reinvigorated bourgeois self-consciousness. Younger delegates
like Gustav Stresemann (born in 1878 in Berlin, Ph.D. in
economics, lawyer for the Association of Saxon Industrialists, and
534
from 1911 member of the committee of the League of
Industrialists) questioned the close relationship between their party
and the Conservatives. Many National Liberal mayors became
pioneers in cooperative economic policy, and such things as
municipal transport offices, gas, electric, and water utilities were
possible only because liberals of all political stripes had turned
away from the ‘Manchester’ interpretation of laissez-faire
economics and recognized the necessity of publicly funded social
services. Liberalism thus became more reformist. Ironically,
however, this transformation was not brought about by means of
general, direct federal suffrage, but through local elections based
on voting laws friendly to property. An ‘opening towards the left’,
that is, to social democracy, was not part of the liberal renewal.
Even in 1910, the vast majority of liberals and social democrats
were still worlds apart.
Only a minority on the left-liberal fringe went so far as to
categorically reject any kind of liberal–conservative joint policy
and openly endorse cooperation with the Social Democrats. The old
historian and Nobel Prize winner in literature, Theodor Mommsen,
was still a lone voice in the wilderness when, at the end of 1902
(about a year before his death), he wrote an article in the liberal
Nation calling for the collaboration of liberals and Social Democrats
against the ‘absolutism of a league of interest between the Junkers
and the chaplainocracy’ in order to prevent the ‘overthrow of the
federal constitution’. And that was not all. Mommsen proclaimed
the Social Democrats the ‘only large party at the present time that
commands political respect’ and wrote that everyone in Germany
knew ‘that equipped with a single mind like Bebel’s between them,
535
a dozen East Elbian Junkers would stand head and shoulders above
their confreres.’ It was no coincidence that the title of the piece,
‘Was uns noch retten kann’,* was the same as that of a pamphlet by
Mommsen’s friend Karl Twesten (died in 1870) from the year 1861,
right before the beginning of the Prussian constitutional conflict.
Five and a half years later, in May 1908, the former editor of the
Nation, Theodor Barth, together with two former members of the
National Social Association, Hellmut von Gerlach and Rudolf
Breitscheid, founded a party that put into practice Mommsen’s
appeal for a collusion between the liberal bourgeoisie and the
social democratic workers: the Democratic Union (Demokratische
Vereinigung). To be sure, with about 11,000 members in 1911 and
about 30,000 votes in the federal elections of 1912, the new party
never amounted to more than a splinter group. And it was only a
minority of that minority that took the logical step of joining the
Social Democratic Party. Rudolf Breitscheid was the first, in 1912.
For this economics Ph.D. and former Burschenschaft member, this
was the real beginning of his political career, which was to lead
him to the leadership of the party in the Reichstag sixteen years
later.
Within the Progressive People’s Party, too, there were
proponents of cooperation with the Social Democrats along the
lines of the Baden ‘grand bloc’. The advocates of an anti-feudal and
anticlerical bloc ‘from Bassermann to Bebel’, among them
Naumann, looked to the reformists and revisionists in the SPD and
to the ‘young liberals’ in the National Liberal Party, who were led
by the Baden jurist Ernst Bassermann after 1904. But the new left–
liberal fusion party was by no means united in this matter, and, for
536
the time being, only a few National Liberal and Social Democratic
outsiders were prepared to take the idea of a ‘social–liberal bloc’
seriously. Thus, at the end of the Bülow era, there was no
potentially majoritarian party constellation in sight that could have
replaced the shattered ‘Bülow bloc’.22
537
1910 in the struggle for a modest reform of the Prussian three-class
voting system. The centrepiece of the proposal was an education
bonus for ‘bearers of culture’ (Kulturträger), including retired non-
commissioned officers. The chancellor summarily withdrew the bill
after the House of Representatives, with the votes of the
Conservatives and the Centre, rejected an upper-house amendment
to the draft.
He was more successful a year later in his attempt—vigorously
contested by the Conservatives—to provide the province of Alsace-
Lorraine with a constitution. The Alsace-Lorrainers now obtained a
parliament with two chambers, the second of which they could
elect. The fact that suffrage was general, equal, and direct was not
the chancellor’s doing, however (he had preferred a plural system),
but that of the Reichstag. The province received three votes in the
Federal Council, which, however, would not be counted if Prussia
could find a majority only by means of them. The constitution of
1911 did not grant the Alsace-Lorrainers full equal rights with the
other Germans, but it was an important step in this direction.
At the beginning of Bethmann Hollweg’s chancellorship, the
reins of foreign policy lay in the hands of Alfred von Kiderlen-
Waechter (originally from Württemberg), the secretary of the
foreign office. It was this man who, in the summer of 1911,
brought Germany to the brink of a great war, though
unintentionally. As in 1905, the catalyst of the international crisis
was Morocco. Disturbances in the north African sultanate prompted
France to occupy Fez and Rabat in spring 1911. Although the
Wilhelmstrasse was prepared to leave Morocco to the French, it
sought to make them pay for the concession by relinquishing
538
extensive territories in the Congo to Germany. In order to
underscore the German demands, Kiderlen sent the gunboat
Panther to the port of Agadir, where it anchored on 1 July 1911.
The main instrument in the journalistic accompaniment to the
‘Panther’s leap to Agadir’ was the Pan-German League. Its leader,
the judicial official Heinrich Class, had been personally asked by
Kiderlen for help with propaganda. The idea of war with France,
which, given the state of international affairs, could rapidly expand
into a world war, held no fear for Class and the Pan-Germans.
Indeed a German West Morocco, such as the Pan-Germans
demanded in one of their pamphlets, was not to be brought about
without war. Numerous conservative, national-liberal, heavy-
industrialist, and Evangelical newspapers expressly endorsed this
idea. The chief of the general staff, Count Helmuth von Moltke, did
the same. The ‘younger Moltke’ considered a great war unavoidable
in any case and believed the present moment more favourable for it
than later, when Germany’s enemies would be stronger.
By 22 July it was clear that France would have the backing of
Britain in the case of armed conflict with Germany. On this date
Lloyd George, the chancellor of the exchequer, after a confidential
agreement with the foreign minister, Sir Edward Grey, gave his
support to Paris in a speech at the Mansion House. Neither William
II nor Bethmann Hollweg wanted war, however, and even Kiderlen-
Waechter sought a gain in prestige without violence. On 4
November 1911, after Russia, Austria-Hungary, and Italy had told
their respective allies that they did not consider a war for Morocco
to constitute a threat to their mutual defence agreements, Germany
and France negotiated an end to the conflict. Germany renounced
539
all forms of political influence in Morocco (which became a French
protectorate in 1912) and contented itself with the assurance of
most-favoured-nation treatment. France ceded (not especially
valuable) parts of its Congo territory to Germany, which granted
France a strip of land in Togo in exchange.
The result of the Franco-German negotiations was the precise
opposite of the hoped-for gain in prestige for the Reich. With his
‘Panther’s leap’, Kiderlen-Waechter had demonstrated a classic
example of unreflective Wilhelmine brinkmanship and, through the
frivolous recourse to the Pan-German League, made himself
hostage to extremist nationalism. Nonetheless, the Pan-Germans
and the right-wing parties were not the only ones who stoked the
fires of outrage over the ‘cowardice’ of the Reich leadership in the
second Morocco crisis. Voices from the Centre were also heard
lamenting the humiliation Germany received at the hands of
Britain and France. Even left-liberal publications were not immune
from invocations of ‘Olmütz’, Prussia’s retreat in the face of
Austria’s threat of war in November 1850.
When, speaking before the Reichstag on 9 and 10 November
1911, Bethmann Hollweg defended the government’s position and
sharply rejected a pre-emptive war, the assembly responded with
massive criticism. The leader of the Centre fraction, Count Georg
von Hertling, concluded his speech with the remark that it would
‘possibly’ do no harm ‘if at some point the authorities were to say
that preservation of the peace is certainly a great good, but that it
is too dearly bought if it can happen only at the price of our
standing in the world’. The leader of the National Liberals, Ernst
Bassermann, accused the chancellor of pursuing a ‘politics of
540
illusion’, spoke of the ‘defeat’ Germany suffered in the Morocco
crisis, and warned foreign nations to be ‘cognizant of the fact that
we shall not let our national honour be offended and that, if it
should prove necessary to take up arms in defence of Germany, we
shall be found a united nation’.
Even more extreme were the words of Ernst von Heydebrand
und der Lasa, the leader of the German Conservative party, which
up to that point had left ‘world policy’ mainly in the hands of the
National Liberals and Free Conservatives. Yet the ‘uncrowned king
of Prussia’, as Heydebrand liked to be called, had apparently come
to the conclusion that old Prussiandom could successfully defend
its endangered position only by placing itself at the forefront of the
national movement demanding a better position in the world for
Germany. ‘What secures peace for us’, said Heydebrand, ‘is not this
compliance, these agreements, these understandings. It is our good
German sword. And it is also the feeling—one the French will have,
with good reason—that we, too, hope to behold a government
willing, when the time has come, to not let that sword go to rust.’
At Britain the speaker aimed the comment that it had now been
revealed to the whole German people ‘where the enemy is. The
German Volk, if it wishes to spread itself out in the world, if it
wishes to find that place in the sun its right and destiny have
appointed to it—then it now knows the location of the one who
pretends to the right to pass judgment on whether it shall be
permitted.’
The counterpoint to the belligerent language of the
Conservatives and National Liberals was provided by the leader of
the Social Democrats, the only party to call upon its supporters to
541
protest over the war in the foregoing weeks. The end of Bebel’s
speech was prophetic. All sides will now arm themselves further
until the moment when one or other of the parties says one day:
better an end with horror than a horror without end ... Then the
catastrophe will come. Then, the general march will be struck up
throughout Europe, at which point sixteen to eighteen million men,
the flower of manhood of all the different nations, equipped with
the latest instruments of murder, will confront each other as
enemies on the field of battle. But in my view, hard on the heels of
the great general march shall come the great crash-bang-wallop ...
It will not come through us, it will come through you yourselves.
You are bringing things to a head. You are leading us into
catastrophe ... The twilight of the gods of the bourgeois world is
upon us. Be assured: it is upon us! You are today in the process of
undercutting your own political and societal order, sounding the
death knell of your own political and societal order.
542
unbearable for us that our competitors on the world market shut
the doors of the last free areas before our noses. That must drive a
nation like Germany, which must expand if she does not wish to
suffocate on her own population surplus, to the point where only
war remains as ultima ratio. This is the feeling of the thinking
circles of the nation.23
543
One of the reasons the Tripolitan war was popular in Italy was
because it could be understood as an attempt to divert mass
emigration away from South and North America towards a more
proximate area on the north African coast, an area that
considerably expanded Italian national territory and strengthened
Italy’s strategic position in the Mediterranean. Germany could
claim no such interest in Morocco. As in other examples of
Wilhelmine Weltpolitik, the second Morocco crisis was a case of
prestige above interest. Germany was economically far stronger
than France, not to mention Italy. Unlike its neighbour to the west,
it did not need colonies and protectorates to compensate for the
trauma of military defeat and the attendant territorial losses. The
German economy was not far from overtaking that of Britain, the
motherland of the Industrial Revolution and imperialism. The
German share of world exports was continually rising, the British
share just as continually sinking. The German Reich was among the
world’s great intellectual powerhouses, perhaps the greatest of
them all. Yet all this was not enough for the political right. It was
not enough that Germany should remain a mere ‘great power’; it
had to become the leading power of the world. This was the
unequivocal message of the second Morocco crisis.
While the political events were still unfolding, Friedrich von
Bernhardi, a former general and Pan-German writer on military
affairs, was drawing his conclusions. Deutschland und der nächste
Krieg* was the title of his book, which drew great attention when it
appeared at the beginning of 1912, was reprinted several times in
rapid succession, and was immediately translated into the most
important languages. Appealing to Darwin and his ‘social
544
Darwinist’ disciples, the author declared the ‘struggle for existence’
to be the ‘basis of all healthy development’. War was a means of
selection, preventing inferior or decaying races from choking the
growth of the healthy ones. But war was ‘not merely a biological
necessity, but also a moral requirement and, as such, an
indispensable factor of culture’. The result was what Bernhardi
termed the ‘right to war’ and the ‘duty to war’.
‘A deep rift has opened between the feeling of the nation and
the diplomatic action of the government,’ proclaimed the book’s
foreword. The only thing that could close this rift was a
thoroughgoing re-education of the German Volk with the goal of
preparing it for a hard decision—the decision ‘whether we, too,
wish to develop ourselves into a World Power, to assert ourselves as
such, and to attain for the German Spirit and German Way of Life
the worldwide respect they still lack to this day’. In consequence,
movements such as social democracy and pacifism, which rejected
this insight, were to be fought. Since it was a matter of ‘World
Power or downfall’, ‘existence or non-existence’, it was imperative
that the ‘confusion of the political scene’ and ‘spiritual
fragmentation’ finally be overcome. Bernhardi thus resolutely
opposed parliamentary government. ‘No people is so little suited to
the direction of its own fate as the German People, either in a
purely parliamentary or even republican government. For none
does the commonplace liberal stencil fit so ill as for us.’ Germany
needed a strong government that would bring to an end the ‘policy
of peace and renunciation’ pursued hitherto, putting in its place a
‘propaganda of the deed’. It needed to strengthen the army and the
fleet and to train the youth in military service from an early age.
545
These goals could only be reached in conjunction with a strong
popular press dedicated to the task of maintaining the warlike
disposition of the people. ‘It must constantly draw attention to the
importance and the necessity of war as an indispensable tool of
politics and culture, as well as to the duty of sacrifice and personal
dedication to state and Fatherland.’ As their preceptors in these
endeavours, Bernhardi held up to the Germans first and foremost
Frederick the Great, then Bismarck, and, for their intellectual
training for war, primarily Fichte, Arndt, and Treitschke. The book
concluded with lines from a poem by Arndt from the year 1844:
546
preached, was a logical imperative for those who accepted the
author’s aims. Other contemporaries expressed themselves even
more clearly on this point, hoping that war would have a salutary
effect on the domestic political front. In the Reichstag debate on 9
November 1911, Bebel quoted two such voices. The Deutsches
Armeeblatt had written during the second Morocco crisis: ‘A
generous passage at arms would also be a very good thing for the
internal situation, even though it would bring tears and pain to
individual families.’ In the 26 August 1911 edition of the
conservative Post one could read: ‘In wide circles, the conviction is
prevalent that a war can only be advantageous, since it would
clarify our precarious political position and bring about the
convalescence of many political and social conditions.’ Bebel
commented on these statements with the observation: ‘Nobody
knows what to do about social democracy anymore. A foreign war
would have been such an excellent distraction.’
There was no lack of groups pursuing the same goals as
Bernhardi. The Colonial Association, the Eastern Marches
Association, the Pan-German League, and the Navy League were
joined in 1904 by the Imperial League against Social Democracy
(Reichsverband gegen die Sozialdemokratie), endorsed by heavy
industry and the parties of the right. By 1909 the organization
counted over 200,000 members. Its pamphlets against the SPD
were countless, but its success was modest. At best, its participation
in the ‘Hottentot election’ of 1907 helped slow the increase in votes
obtained by the Social Democrats.
In January 1912, in the wake of the second Morocco crisis, yet
another nationalist organization was founded, whose primary aim
547
was the strengthening of the German army: the German Army
League (Deutscher Wehrverein), led by August Keim, another former
general. The organization’s top members came from the ranks of
the Pan-German and Navy Leagues; it was no less bürgerlich in
character than that of the other nationalist associations. Its idol
was Bismarck. The Army League, which numbered 78,000
members in May 1912, was markedly anti-governmental, since the
Bethmann Hollweg administration did not seem militaristic enough
to its taste. It courted the masses and endeavoured to be populist.
But it was too much of an assembly of notables to become the
nucleus of mass movement transcending social class.
The Army League did not represent, any more than the Navy
League or the Pan-German League, a new type of ‘populist
nationalism’. It also did not stand for a new kind of ‘militarism
from below’. Rather, it was a further expression of the nationalism
of the ‘right’ that had existed from the first decade of the German
Reich, spreading among bourgeois and petty-bourgeois Germany in
several phases and becoming increasingly radical. The last phase,
which began around 1911, was characterized by unprecedented
militarism. Politicians, publicists, and propagandists heedlessly
preached world war as a way out of foreign and domestic crisis.24
548
and 25 January. The Social Democrats achieved a dramatic victory.
Their tally of votes rose from 3.26 million in 1907 (29% of the
total vote) to 4.25 million (34.8%). Their increase in parliamentary
seats was even more dramatic. The SPD grew from 43 to 110 seats,
primarily because of a run-off agreement with the Progressive
People’s Party. The Social Democrats had been the strongest party
ever since 1890; now, for the first time, they represented the
strongest fraction in the Reichstag.
A governmental majority without the Social Democrats was now
possible only if the conservative parties, the National Liberals, the
Progressives, and the Centre all voted together. A constellation of
this kind was extraordinarily difficult to bring together. Yet the
obstacles to a fixed alliance of parties including the Social
Democrats were even greater. A centre-left bloc would have
required the participation not only of the 42 delegates of the
Progressive People’s Party, but also of the 91 members of the
Centre fraction. Considering the ideological antagonisms between
the three parties, such a ‘coalition’ was, for the time being, pure
fantasy.
The depth of the chasm still dividing the bourgeois parties from
the Social Democrats was revealed in the struggle for the Reichstag
presidency. On 9 February the Social Democrat Philipp
Scheidemann was elected temporary first vice-president, provoking
sharp public protest from the right and prompting the
demonstrative resignations of the president, Peter Spahn from the
Centre Party, and the National Liberal vice-president, Hermann
Paasche. When Scheidemann, true to the Social Democratic ‘code
of conduct’, refused to ‘go to court’ (i.e. participate in an audience
549
of the presidency with the emperor), he lost the confidence of a
majority and therewith his office. The final presidency was made
up exclusively of members of the liberal fractions.
Shortly after the elections, a book appeared that seemed like the
right’s answer to the victory of the left. It bore the title Wenn ich
der Kaiser wär’,* and was written by an author who named himself
‘Daniel Frymann’. Behind the pseudonym stood Heinrich Class,
after February 1908 the leader of the Pan-German League. The
Mainz attorney embodied like no other the radical nationalism of
the Wilhelmine era. Already in 1904, the year before the first
Morocco crisis, Class was among the League officials calling for a
war to conquer West Morocco. During the second Morocco crisis,
he further demanded the cession of large parts of eastern France.
Together with Alfred Hugenberg, the president of the directory of
the Friedrich Krupp company in Essen, who had been the driving
force behind the founding of the Pan-German League in 1890–1,
Class helped the populist and anti-Semitic elements gain the
ascendancy within the group. But the message of his 1912 book
went far beyond what might have been called the organization’s
official opinion. There were good reasons for the pseudonym.
The Kaiserbuch,† as it came to be called, which had gone
through five printings by 1914, had many points in common with
Bernhardi’s manifesto, which had appeared shortly before and won
the praise of Class. ‘Let war be holy to us like purifying fate’,
implored the Pan-German leader. ‘Let it be welcome like the doctor
of our souls, who shall heal us with the strongest of medicines.’
Germany did not need to fear war either with Britain or with
550
Russia. France was to be crushed, Belgium and the Netherlands
annexed to the Reich under preservation of a limited
independence. ‘The Gordian knot must be cut through; it cannot be
undone amicably.’
In his domestic politics, Class was far more radical than
Bernhardi. He demanded the abolition of general, equal suffrage
and its replacement with a five-class system restricted to taxpayers,
the exclusion of women from political life, inflexible resistance to
the democratization of Prussia, a tenacious battle against the Poles,
a readiness to a coup d’état against social democracy, this ‘enemy of
the Fatherland’—a step that was, to be sure, only to be expected
under an ‘emperor as leader’. Class directed his strongest attacks
against the Jews, who, according to their inmost nature, were to
the Germans as water to fire. The Jews were the ‘agents and
teachers of the materialism that rules the day’, controlling theatres,
the press, and the outcome of the recent ‘Jew elections’. In order to
avert the Jewish danger, the author called for a stop to Jewish
immigration, the exclusion of the ‘resident’ (landansässig) Jews
from public life, abolition of their active and passive suffrage, their
placement under alien law, and the doubling of their tax burden. In
addition, Jews were to be excluded from service in the army and
navy, from leadership positions in the theatre and banks, and from
the professions of attorney and teacher. Finally, newspapers with
Jewish employees were to be designated as such.
Class sought to prove what he saw as the disproportionate
societal influence of the Jews primarily with statistical data on the
confessional background of university students. Measured
according to their share of the population, five times as many Jews
551
as Evangelicals studied at Prussian universities, and seven times as
many as Catholics. Class’s conclusion was ‘that the life of our
people, to the degree that it rests on, is connected with, or depends
upon the activities of the learned professions, is falling victim at a
rapidly increasing rate to a Jewish participation far beyond that
which would normally be their right.’ From his findings the author
generalized: ‘Never in history has a great, gifted, industrious people
fallen so quickly and with so little resistance under the influence
and intellectual sway of a foreign people so different in nature, as
the Germans have with the Jews.’ The battle cry Class hurled
against the Jews was ‘Germany to the Germans!’ This challenge
was to be turned into action by the ‘leader’. The emperor was
certainly the most desirable incarnation of the idea of the leader,
but he was not necessarily the only conceivable one. ‘If the leader
were to arise this day, he will be amazed at how many faithful
followers he has—and how valuable, selfless men rally round him.’
The Kaiserbuch amply demonstrated the demagogic qualities of
its author. Nonetheless, it cannot easily be described as ‘populist’.
Class did not address the German Volk so much as the higher
classes, especially the ‘academically educated’, who were to be the
‘intellectual leaders of the people’ and the ‘backbone of political
life’. The assertion that an ‘aristocracy of merit’ should take its
place at the side of the ‘aristocracy of birth’ was quite as
‘reactionary’ as the wish to replace general direct suffrage with a
class system. Class’s anti-Semitism, too, was calculated to appeal
primarily to the educated. The fact that the number of Jewish
students, doctors, lawyers, and journalists was far greater than the
proportion of Jews in the population was excellent fodder for an
552
argument intended to stir the envy, inferiority complexes, and
socio-economic anxieties of non-Jewish students and academics.
Class used this method to gain support for goals that can best be
epitomized in the term ‘radical bourgeois conservatism’. This was a
world-view that had long been on the ideological defensive. The
outcome of the federal elections in January 1912 easily disposed it
to panic.
German anti-Semitism seemed to have largely declined in
political importance by the time Class’s book came out. In the
federal elections of 1907, only 5.5 per cent of the vote had gone to
the fragmented anti-Semitic parties. Five years later, in January
1912, they managed only 2.5 per cent, their worst result since
1893. The public resonance to organized Jew-hatred had indeed
decreased. Nonetheless, anti-Semitism was far from extinction. It
lived on among the German Conservatives, in special interest
associations like the Federation of Agriculturalists and the German
League of Commercial Employees, in numerous student
organizations, and in the Evangelical and Catholic churches. There
were also a number of highly influential books by respected anti-
Semitic authors, for example the Deutsche Schriften* (1878 and
1881) by Paul de Lagarde, Rembrandt als Erzieher† by Julius
Langbehn (first published 1890, but reworked and expanded a year
later to include radically anti-Semitic arguments inspired by the
fanatic Jew-hater Theodor Fritsch), and the 1899 two-volume work
of Houston Stewart Chamberlain, an Englishman by birth but
German by choice, entitled Die Grundlagen des neunzehnten
Jahrhunderts,† which counted Emperor William II among its
553
enthusiastic readership.25
Apart from the resolute liberals, the bourgeois culture of
Germany had long been saturated with anti-Semitism. Only an
external stimulus was required to bring it to the fore anew. The
1912 electoral victory of the left was one such stimulus, and Class’s
Kaiserbuch was only one of the first symptoms of the movement
that now began. Shortly after the January elections, an ‘Association
against Jewish Arrogance’ (Verband gegen die Überhebung des
Judentums) was founded, including among its members not only
well-known anti-Semitic publicists, but also the head of the
German Gymnastics Association and the Association of German
Gymnasts. In May 1912, at the instigation of Theodor Fritsch, the
‘Hammer League’ (Reichshammerbund) was formed, the inmost
circle of which was a ‘Germanic Order’ (Germanenorden) organized
as a system of lodges and using the swastika as its symbol. The
Hammer League considered its most important task to be the
coordination of all anti-Semitic activities and the infiltration of its
agents into all sorts of different movements and associations.
One of the organizations that drew Fritsch’s special attention
was the club that began the German youth movement, the
Wandervogel,* founded in Berlin-Steglitz in 1901. In 1912, a
campaign began to exclude Jews from membership. Only a
minority of the Wandervögel resisted. The controversy came to a
provisional end in a compromise passed by the organization’s
federal congress in Frankfurt an der Oder during Easter 1914: it
was left to the local groups to decide whether Jewish candidates
would be accepted.
554
In 1913 came a new proposal for the ‘solution to the Jewish
question’. In a memorandum sent to two hundred prominent
Germans, including Crown Prince William, Konstantin Baron von
Gebsattel, a former Bavarian cavalry general (and at that time not
yet a member of the Pan-German League), not only repeated the
demands of the Kaiserbuch, but even outdid them. Any mixing of
‘the Jewish and the German race’ was to be punished, and all
discriminatory measures were to be coordinated in such a manner
as to leave the Jews only one course of action, emigration from
Germany. Before they left the country, however, they were to be
compelled to transfer the greater part of their property to the state.
It was true that William II and Bethmann Hollweg, to whom the
crown prince (himself approached by Class in a letter) approvingly
sent the memorandum, contradicted the general’s scheme with
economic and foreign policy arguments (and rejected his call for a
coup d’état to abolish equal suffrage on ‘practical’ grounds, as too
little likely to succeed). Off the record, however, both the emperor
and the chancellor lamented the prominent role Jews played in the
press. William also announced his firm intention ‘to exclude Jewish
influence from the army with all decisiveness and to limit it in all
literary and artistic activities as much as possible’.
Only a minority of the German right went as far as Gebsattel
and Class, and the membership in German racialist or ‘völkisch’
organizations like the Hammer League remained small (nothing
dependable is known of the numbers before 1914; the Hammer, the
journal published by Fritsch, had an average circulation of 10,000).
But the converse was also true. The resolute opponents of anti-
Semitism enjoyed no wide support among the middle classes. The
555
Association against Anti-Semitism (Verein zur Abwehr des
Antisemitismus, founded in 1890), an initiative of liberal politicians,
scholars, and publicists like Rudolf von Gneist, Heinrich Rickert,
Theodor Barth, and Theodor Mommsen, numbered 18,000
members in 1897. There were probably not many more in 1912.
The organization’s most active members were ‘cultural
Protestants’ and left-liberals. In close alliance with the Central
Association of German Citizens of Jewish Faith (Centralverein
deutscher Staatsbürger jüdischen Glaubens, founded in 1893), it
attempted to counteract the malignant effects of anti-Semitic
horror stories of Jewish ritual murder (or later, after the First
World War, forgeries like the Protocols of the Elders of Zion) by
means of educational campaigns and appeals to reason. There was,
on the other hand, no cooperation between the Association against
Anti-Semitism and the Zionist Union for Germany (Zionistische
Vereinigung für Deutschland), an organization established in 1897–8
by the German followers of Theodor Herzl, the Jewish publicist
from Vienna. The Zionist response to anti-Semitism, the call for a
Jewish national state in Palestine, was far more popular in eastern
Europe than in Germany. Regardless of what the German Jew-
haters said or did, the vast majority of German Jews considered
Germany to be a country of the Enlightenment, of the rule of law,
and of cultural progress—their Vaterland.
German Jews (i.e. those who were unbaptized) could find their
political home only in parties firmly opposed to anti-Semitism, the
liberal, especially left-liberal parties, and increasingly the Social
Democrats. The workers’ movement owed a great part of its
ideological character to Jewish intellectuals, who had joined the
556
Social Democratic Party in great numbers from the beginning. The
organized left-liberalism of Wilhelmine Germany, for its part, is
utterly inconceivable without the prosperous and educated Jewish
bourgeoisie of the great cities like Berlin, Hamburg, Frankfurt am
Main, or Breslau. The Jewish share in total tax revenues was many
times the proportion of Jews in the population. In Berlin and
environs, where the number of Jews in the population (5 per cent)
was about five times greater than in the rest of the Reich, their
contribution to the 1905 tax revenues amounted to 30 per cent of
the total. The liberal press was financially just as dependent on the
Jewish middle classes as it was intellectually dependent on the
productivity of Jewish journalists. In contrast to the bureaucracy,
judicial system, and armed forces, where massive discrimination
against Jews continued, journalism was one of the fields in which
Jews could freely engage, as they could in freelance academic and
artistic professions, commerce, and banking. The great liberal
newspapers in Germany—the Frankfurter Zeitung, Berliner Tageblatt,
Vossische Zeitung, and so on—all showed the influence of Jewish
spirit and wealth. The anti-Semitic activists knew full well why
they attacked ‘Jewish’ liberalism and the ‘Golden International’ just
as forcefully as ‘Jewish’ Marxism and the ‘Red International’.
War against the ‘Golden’ and the ‘Red International’ was the
rallying cry of a new interest group that presented itself to the
public in Dresden in September 1911: the Imperial German Middle
Class Association (Reichsdeutscher Mittelstandsverband), whose
initiator was the indefatigable Theodor Fritsch, leader of the
Middle Class Union for the Kingdom of Saxony
(Mittelstandsvereinigung für das Königreich Sachsen). The new,
557
militantly antisocial-democratic organization was established in
order to draw craftsmen and small tradesmen out of the liberal
Hanse League, bring them together under its own banner, and lead
them into a firm alliance with the conservative forces in German
society. To a certain extent, these efforts were successful. The
Mittelstand organizations that joined the new federal association
numbered over half a million members by 1913.
On 19 February 1912, four weeks after the Reichstag elections,
the Association informed Chancellor Bethmann Hollweg that it was
working ‘in cooperation with a leading industrial association in
order to create a community of interest among all independent
productive estates’. One and a half years later, this task had largely
been accomplished. At the third federal congress of the Mittelstand
in Leipzig in August 1913, a ‘Cartel of the Productive Estates’
(Kartell der schaffenden Stände) was brought into existence
(immediately branded ‘cartel of the rapacious hands’, Kartell der
raffenden Hände, by its enemies), an alliance between the young
federal Mittelstand association, the Federation of Agriculturalists,
the Central Association of German Industrialists, and the Union of
Christian Farmers’ Associations (Vereinigung der Christlichen
Bauernvereine). The Leipzig platform culminated in the call for the
‘preservation of authority in all economic concerns’, ‘the protection
of national labour, the guarantee of fair prices, and the protection
of those willing to work’ as well as the ‘battle against Social
Democracy and the false doctrines of socialism’.
The Cartel was a new attempt to reach an old goal, the
reinforcement of the existing order through conservative
Sammlungspolitik. Apart from declarations of common intention,
558
however, the unequal partners achieved little. The textile
industrialists of the Central Association protested against collusion
with the agriculturalists. The heavy industrialists would also have
refused to back the anti-Semitic position for which Fritsch
campaigned. In the end, the August 1913 ‘gathering of all state-
upholding forces’ remained a mere demonstration, underscoring
the desire of East Elbian agriculture, Rhenish-Westphalian industry,
and the conservative wing of the old craft-industrial Mittelstand to
fight anything and everything that amounted to a liberalization of
society and a democratization of the body politic.26
Two months after the proclamation of the Cartel of the
Productive Estates, Leipzig became the stage of a much more
spectacular event. On 18 October 1913, the centennial of the
victory over Napoleon, the monument to the Battle of the Nations
was consecrated. Ernst Moritz Arndt had been the first to propose
such a monument in 1814. In 1894 prominent members of the
‘nationally’ minded Leipzig bourgeoisie, led by the architect
Clemens Thieme, took up the idea, and their organization, the
German Patriots’ League for the Erection of a Monument to the
Battle of the Nations near Leipzig (Deutscher Patriotenbund zur
Errichtung eines Völkerschlachtsdenkmals bei Leipzig) collected
donations made by numerous groups, including associations of
singers, marksmen, veterans, gymnasts, and cyclists, as well as by
the emperor himself, the princes, and many cities and
communities.
Proclaiming the Battle of the Nations at Leipzig the ‘birthday of
the German people’ and the Battle of Sedan the ‘birthday of the
German Reich’, the text of the 1913 dedication interpreted the
559
1871 founding of the Reich as the completion of the work
undertaken by the German people in 1813. The nation was seen as
a close-knit Volksgemeinschaft and, as such, was made into a
standard the Germans of 1913 felt no less bound to live up to than
their forebears a century before. The commemoration of 1813 had
one overarching purpose: to convince the Germans that
materialism and trivialization, cosmopolitanism and socialism,
confessional and class antagonisms had to be overcome, and that
they must elevate themselves again to the ‘pure heights of German
idealism’ and to ‘pure Germandom’. Seen in this way, the new
monument was directed against internal no less than against
external enemies. Like the Kyffhäuser monument of 1896, the
monument to the Battle of the Nations was a monolithic bourgeois
declaration of war on the ‘the forces without Fatherland’.
In reference to the giant, sword-supported warrior figures on the
peak of the dome and the archangel Michael on the main relief,
Thomas Nipperdey has said that they are all ‘characterized by a
stern and ponderous seriousness, indeed of melancholy’. The pathos
of sacrifice and suffering in the dedication literature is also present
in the structure.
560
The dedicatory speech, given by Thieme in the name of the
Patriots’ League in the presence of the German emperor, the king of
Saxony, many German princes, the chancellor, ambassadors from
the Russian, Swedish, and Austrian courts, military delegations,
student corporations, and tens of thousands of onlookers, was in
keeping with the dedication text. Thieme described the monument
as a symbol of the ‘spirit of sacrifice, courage, strength of faith, and
the power of the German people’; it was the ‘celebratory deed of
the German people’, commissioned to
561
intellectual and member of the board of directors of the General
Electricity Society, composed a cycle of poems entitled 1813. Ein
Festgang zur Jahrhundertfeier.* One of them ran thus:
562
Thee!” ’
There were voices in opposition to all this. In spring 1913 Social
Democrats and progressive liberals drew attention to the obvious
connections between the patriotic celebrations and the current
policies of the government, namely its new military bill. At the first
congress of the Free German Youth on the Hoher Meissner near
Kassel in October 1913, the school reformer Gustav Wyneken
warned against the ‘mechanization of enthusiasm, which is now
everywhere on the increase. Has it gone so far that they only need
to shout certain words at you like “Germandom” and “national” in
order to win your applause?’
Nonetheless, the prevalent opinion was certainly more like the
one Ernst Müller-Meiningen, delegate from the Progressive People’s
Party, lamented before the Reichstag on 8 April 1913:
The army bill that had caused such consternation among the
left-liberal politicians in April 1913 was passed by the Reichstag
with a large majority in June. The Progressive People’s Party, too,
563
agreed to the stepwise increase of the army by 136,000 men. Even
the Social Democrats, though rejecting the bill in its final form,
helped pass earlier legislation that made the army increase
financially possible; the capital gains tax, passionately opposed by
the Conservatives, was, according to the Social Democrats, a first
step towards the federal income, wealth, and inheritance tax they
had long desired.
About this same time, France introduced the three-year term of
military service in response to the enlargement of the German
army. The latter was, in its turn, a reaction to the peacetime
strength of the Russian army as well as to the offensive strategy the
French general staff had worked out with Russia after the German
‘Panther’s leap to Agadir’. The international arms race of 1912–13
took place against the background of two Balkan wars that once
again brought Europe to the brink of a great war. The first began in
summer 1912, when the Albanians revolted against Turkish rule
and received the support of Bulgaria, Serbia, and Greece, which
shortly before had formed a ‘Balkan League’ under Russian
auspices. Within a few months, the allies had almost completely
conquered Turkey’s European territories. Russian and Austrio-
Hungarian intervention was again held back—by Berlin’s warnings
to Vienna and London’s warnings to St Petersburg.
The following year, the victors of the first Balkan war came into
conflict over the partition of Macedonia. Bulgaria was on the one
side, Serbia and Greece on the other, joined by Romania, which
sought to take southern Dobrudja away from Bulgaria. Once again,
Austria-Hungary threatened to intervene. Once again Germany, this
time in cooperation with Rome, was able to convince its ally to
564
back down. The Treaty of Bucharest of August 1913 led to the
partition of Macedonia between Serbia and Greece; Albania gained
its independence, though without the predominantly Albanian
province of Kosovo, which fell to Serbia; Romania acquired
southern Dobrudja. Austria-Hungary was the first great power to
order the enlargement of its armed forces, which began while the
crisis was still in progress. This was the start of the spiral that
reached its—temporary—conclusion with the German army bill of
1912 and the French ‘loi de trois ans’.
Not until more than half a century later did the world learn how
close Europe came to a great war in the Balkan crisis. The
enlightenment came through the work of German historians
evaluating documents detailing a conversation in Berlin on 8
December 1912. Alarmed by the announcement of the British
foreign secretary, Sir Edward Grey, that his country would not
remain neutral if the Balkan crisis led to a German attack on
France, William II had invited Moltke, chief of the general staff,
Tirpitz, secretary of the imperial navy, Heeringen, chief of the
admiralty, and Müller, chief of the navy cabinet, to a meeting. He
did not invite the chancellor, the Prussian minister of war, or the
secretary of the foreign office. The emperor was not the only one to
urge an immediate war with France and Russia that day; Moltke
agreed with him, saying: ‘I consider war inevitable, and the sooner
the better.’ To Tirpitz’s objection that the navy was not strong
enough for a war with Britain and that the great battle should be
postponed a year, Moltke responded that the navy would not be
ready even at that point and that the army would be at a greater
disadvantage, ‘for our enemies are arming more than we, who are
565
so restricted with respect to finances.’
The deliberations of 8 December had no immediate effects. Nor
did William II’s evaluation of the situation, such as he propounded
a week later to his friend Albert Ballin, director of the joint-stock
company Hamburg-American Paketfahrt (‘Hapag’) and most
prominent of the ‘emperor’s Jews’ (Kaiserjuden): if Germany were
forced to resort to arms, then it would not be merely a matter of
coming to the aid of Austria and fighting off Russia, ‘but of fending
off Slavdom in its entirety and remaining German. Id est, we were
on the brink of a racial war of the Germanic peoples against the
upstart Slavs. A racial war that we will not be spared, for the future
of the Habsburg monarchy and the existence of our Fatherland are
at stake.’
This was not Bethmann Hollweg’s approach. For the present, the
chancellor remained committed to avoiding a great war and to
working out an understanding with Britain, and his policy proved
decisive, at least temporarily. It was not at all certain that the
chancellor and the foreign office would prevail against the emperor
and the military in coming crises. There was a war party, both
among the public and at the head of the Reich. Not only the pan-
German, conservative, and national-liberal press, but also Catholic
newspapers spoke, like the emperor, of an unavoidable war
between the Germanic peoples and the Slavs. On 8 March 1913
Germania, the leading Centre paper, went so far as to reduce the
whole of the ‘oriental question’ to a decision between ‘the Teutonic
world or the Slavic world’ and to endorse the patriots of the
‘Austrian war party’ who had been ready in November 1912 ‘to
strike against Serbia and Russia’. The German war party was a
566
factor the civil leadership of the Reich could not ignore.28
The issue of ‘race’ also played a role in the domestic policies of
Bethmann Hollweg’s administration. On 25 June 1913 the
Reichstag, by a large majority, passed a law on state and imperial
citizenship against the votes of the Social Democrats and the Poles.
According to the old law, which dated from the era of the North
German Confederation (1 June 1870), imperial citizenship was
attained through citizenship in one of the federal states. The new
law added a ‘direct imperial citizenship’, the purpose of which was
to bind expatriate Germans to the Reich. The precondition for
citizenship in a German state remained, as a rule, descent from at
least one German parent. At birth, the legitimate child of a German
male assumed citizenship through the father; an illegitimate child
acquired the citizenship of the mother. Under certain conditions,
foreigners living in Germany could be naturalized, but there was no
right to that status. When the Social Democrats (and, less
emphatically, the progressive liberals) demanded one, the secretary
of the federal ministry of the interior, Clemens von Delbrück,
rejected the idea as an attempt ‘to make it easier for alien elements
to acquire citizenship’.
This comment was probably more anti-Polish than anti-Semitic
in intent. But it was seriously meant. Delbrück explicitly endorsed
the jus sanguinis, the law of descent through blood, and rejected the
jus soli, which made place of birth decisive for citizenship. He thus
defended an old German tradition, the nation defined as an
objective community of fate, not a subjective community of will of
the western type. During the deliberations, the Conservatives were
the most eloquent advocates of this view of the nation. But the
567
broad majority that voted the bill through made it clear that the
traditional idea of ‘Germanness’ was still the predominant one.
After 1871 the Alsace-Lorrainers had never been asked if they
wanted to be German. They were simply considered German, just as
the law on state and imperial citizenship considered the imperial
territory of Alsace-Lorraine a German state. Their initial rejection
of the Reich had, over the years, given way to acceptance (as the
federal elections demonstrate), and the constitution of 1911
seemed calculated to promote the organic integration of the
province into the empire. On 6 November 1913 this development
was suddenly interrupted. The Zaberner Anzeiger reported on hostile
and coarsely derogatory statements a Prussian, Lieutenant von
Forstner, had made to a group of recruits about the Alsace-
Lorrainers (whom he referred to by the insulting name of Wackes).
The matter would probably have been quickly resolved if the
commander, General von Deimling, had fulfilled the demand of
Count Wedel, the governor, and not only punished Forstner with
six days of house arrest, but sent him to another regiment. Since he
failed to do so, the problem not only did not go away, it was
exacerbated. Protests, clashes between soldiers and civilians, and
military arrests increased in Zabern (Saverne).
The Reichstag turned its attention to the problem at the end of
November and beginning of December. The Prussian minister of
war, Falkenhayn, told the delegates that neither he himself nor the
parliament had any authority in matters of disciplinary
punishment, that this was the province of the military command,
that is, of the monarch. Chancellor Bethmann Hollweg was so
cautious in distancing himself from the behaviour of the military
568
administration that the delegates took it as a provocation. When,
on 4 December, he explicitly highlighted his agreement with the
minister of war, it was the straw that broke the camel’s back. An
overwhelming majority of 193 to 34 votes with 3 abstentions
agreed to a motion—lodged the day before by the Progressive
People’s Party—that the Reichstag officially declare itself in
disagreement with chancellor’s treatment of the Zabern affair. Only
the two Conservative fractions voted against the first motion of
disapproval in the history of the German Reichstag.
Political consequences there were none. The chancellor was not
dependent upon the confidence of the delegates, a fact he stated
openly after the vote. Behind the scenes, a few steps were taken
toward restricting the high-handed behaviour of the military
(including an expression of disapproval from the emperor, a
punitive transfer of the two battalions stationed in Zabern, and a
never-published decree from 19 March 1914 concerning the
military’s use of armed force). However, the military courts in
Strasbourg decided, as a rule, in favour of the inculpated officers
and against the soldiers who had informed the press. Lieutenant
von Forstner, who, on 2 December, had struck a shoemaker’s
journeyman with the flat of his sabre when the latter was arrested
for a public insult, was exonerated on 10 January 1914 after appeal
to the military high tribunal, which recognized a situation of
‘punitive self-defence’.
The real scandal of the ‘Zabern affair’ lay not in the events
themselves, their public resonance, and their juridical
consequences, but in the constitutional situation, upon which the
crisis had shed a glaring light. The federal parliament, chancellor,
569
and minister of war were all powerless in questions of royal
military command. When necessary, the Prussian soldier-state
could show the constitutional state its limits. In civil life,
absolutism had been overcome. It lived on in the military.29
570
remain a local conflict, that it would almost certainly bring in
Russia and France, and possibly Britain, on the one side, and
Germany on the other, was a probability the moderate forces in the
country, led by the chancellor, accepted. For the radicals within the
military, political, and publicist war party, the prospect of a great
war was the very reason to urge Vienna towards harsh treatment of
Belgrade.
The proponents of a military solution were also swayed by other
considerations, both foreign and domestic. For Moltke and the
leading figures of the military, the arms race was the decisive
factor: in a few years the adversaries would be even stronger, so
the time to strike was now. The civilian war party, dominated by
the Pan-German League, regarded war as an opportunity to resolve
the internal situation: if the Social Democrats were to become any
stronger, Germany would not be capable of waging any war at all;
if Germany won the war, it would then be strong enough to bring
the Social Democrats and pacifists to reason.
This was not the opinion of Bethmann Hollweg. He believed that
if the Social Democrats supported the war, they would demand a
high price in terms of domestic reforms. The cooperation of the
SPD could only be secured if Russia could be made to appear as the
real aggressor and warmonger. The more Russia sided with Serbia,
the more plausible this scheme became.
In all other respects the chancellor was pessimistic. On 14 July
he spoke to his adviser Kurt Riezler of a ‘leap in the dark’ that was
nonetheless ‘the most earnest of duties’. Since Italy, Germany’s
other partner in the Triple Alliance, could hardly be counted on
any longer, everything would have to be done to keep at least
571
Austria-Hungary, this multinational empire in a perpetual state of
crisis, in firm partnership. By this time the chancellor, too, had
come round to the military’s view that within two or three years
Germany would no longer be in a position to defend itself against
an attack from France and Russia. For him, it would be a ‘pre-
emptive war’, as he would tell Conrad Haussmann, delegate of the
Progressive People’s Party, in the summer of 1917.
Bethmann Hollweg was fully aware that Britain would not long
remain neutral in a war that pitted Germany against France and
Russia. After all, the 1905 ‘Schlieffen Plan’ was still in effect, which
called for an invasion through neutral Belgium in the case of a two-
front war—a provocation that would amount to a declaration of
war on Britain. But the military had made its decision, and a
chancellor who had dared to confront the Prussian soldier-state
after Sarajevo would probably soon have found himself out of
office.
On 25 July Serbia responded with such accommodation to the
remarkably far-reaching catalogue of demands Austria-Hungary
had submitted to Belgrade on 23 July that even William II, who
had supported a firm anti-Serbian line, decided on 28 July, upon
his return from a ‘trip to the north land’, that there was no longer
any reason for war. But it took a long time for Bethmann Hollweg
to become active on behalf of the emperor’s wishes, and he
represented them less forcefully than they were intended.
Consequently, William’s change of opinion no longer had any effect
in Vienna. At eleven o’clock in the morning of 28 July, Austria-
Hungary declared war on Serbia. Russia reacted the following day
with a partial mobilization against Austria in several military
572
districts. On 30 July Tzar Nicholas II, urged by his generals,
ordered general mobilization.
When, on 30 July, the German chancellor (aware that Britain
would join France and Russia in the event of war, but still ignorant
of the general mobilization in Russia) pressured Vienna to accept
an offer of mediation from Grey, the British foreign secretary, he
was probably not motivated by a desire to rescue peace in Europe
at the last minute; rather, he sought to make Russia appear as the
aggressor. The next day Moltke encouraged his Austrian colleague,
Conrad von Hötzendorf, to begin general mobilization
immediately. This prompted the Austrian foreign minister, Count
Berchtold, to ask sardonically: ‘Who is in charge, Moltke or
Bethmann?’ The Habsburg monarchy followed the recommendation
of the chief of the Prussian general staff. At noon on 31 July it
declared general mobilization.
One hour later, Germany proclaimed a ‘state of acute danger of
war’. Since Russia failed to reply to a German ultimatum of that
same day, Germany declared general mobilization and war against
Russia on 1 July. Two days later, after Paris, responding to a
German ‘enquiry’ of 31 August, said that it would act in keeping
with its interests, Germany declared war on France. On 4 August
Britain, which had mobilized its fleet three days before, demanded
that Germany promise to respect Belgian neutrality by midnight.
By this time, however, German troops had already crossed the
border into Belgium, with the consequence that Britain declared
war on the same day. Negatives came from two potential allies:
Italy stated that Austria-Hungary was the aggressor, and that the
situation did not therefore represent an attack against the Triple
573
Alliance. Romania’s position was similar. Only Turkey joined the
central powers, and Bulgaria a year later. Thus began the First
World War, which in 1979, after an interval of sixty-five years, the
American historian and diplomat George F. Kennan would call ‘the
great seminal catastrophe of this century’.
Although it is not correct to say that Germany was alone
responsible for the war, Germany’s was nonetheless the principal
fault. Without German backing, Austria-Hungary could not have
declared war on Serbia. The war parties in Berlin and Vienna
added fuel to each other’s fires, and both bolstered the war party in
St Petersburg. The pan-Slavic movement demanded unflinching
solidarity with Serbia, putting the Russian government under a
pressure similar to that exerted by the rightist parties and
associations of nationalist agitation on the leadership of the
German Reich. But it was not initially successful: Russia did not
encourage Serbia to remain inflexible to Vienna’s demands of 23
July, but rather to fulfil them as much as possible.
In France, the ruling right did not want to endanger the alliance
with the tzarist empire by urging St Petersburg to counsel
moderation to Serbia. In addition Paris, much like Berlin, feared a
further strengthening of the domestic left and, as a result, a
majority against the three-year term of service. Whether the war
would have been prevented if Britain had contradicted German
speculations on its neutrality at an early date and with sufficient
emphasis, is questionable. Such a policy might have had the
opposite effect, making Russia and France even more willing to go
to war.
And yet whatever the other participants could have done
574
differently, one thing is certain: no great power worked for the
escalation of the conflict more persistently than Germany. The fear
that the military capabilities of the Reich would decline in the
coming years does not alone justify the term ‘pre-emptive war’. In
summer 1914 no great power had any concrete plan to wage a war
of aggression against Germany. The decisive actors in Germany, the
leaders of the army and navy, were not concerned about preserving
the status quo. They were after the hegemony of Europe. And their
aspirations were shared by a strong current within the German
bourgeoisie.
And yet, broad public support for the war—including the
support of the Social Democrats—was only to be had if the
government’s aggressive intentions could be disguised as national
defence. The Russian general mobilization permitted the Reich
leadership to create this impression. On 1 August 1914 the head of
the navy cabinet, Admiral von Müller, noted in his journal:
575
Balkan crisis in November 1912, when the Social Democrats,
together with other parties of the Second International, gave
expression to the common will for peace among the European
proletariat at the cathedral in Basel. However, there were no clear
indications of what the working class was actually to do in case of
war. At the Stuttgart congress of the International in summer 1907,
August Bebel (who died on 13 August 1913 at the age of 73) had
impeded the passage of a motion explicitly designating ‘mass strike’
and ‘uprising’ as a means of averting and preventing war. His
argument that ‘mass strike amusements’ made no sense after the
outbreak of a war, since by then the masses would already have
been called up, reflected the view of the majority of German Social
Democrats.
At the end of July 1914, the SPD sponsored anti-war
demonstrations in most of the large urban centres in Germany. The
largest took place at Treptow Park in Berlin on 28 July. At the
same time, however, the Reich leadership, attempting to secure the
loyalty of the workers, was conducting negotiations with the
leaders of the SPD, employing Albert Südekum, a ‘revisionist’, as
intermediary. On 29 July Südekum was able to inform the
chancellor that actions on the part of the SPD were neither planned
nor to be feared in the event of war. Two days later a state of
emergency was declared throughout the Reich. Not only the
demonstrations against the war stopped suddenly, but all public
criticism of the federal government—which, a few days before, the
Social Democratic press (above all the Vorwärts and the Leipziger
Volkszeitung) was accusing of being deeply complicit in the
dramatic deterioration of the international situation.
576
The fear of government suppression and persecution of the party
played a key role in the decision of the SPD leadership to ask the
Social Democratic newspapers to ‘exercise all due caution’. And in
truth, had it called for resistance to the war, the party would have
been subject to the harshest measures of the state, and it could not
have been certain that the masses of its supporters would have
obeyed such an appeal. The SPD and the labour unions had grown
into mass organizations, upon whose preservation numberless lives
depended. Even an August Bebel could not have afforded to
jeopardize their existence.
On the other hand, the party leadership did not have to deal
with any nationalist pressure from ‘below’ on the eve of the First
World War. There was little ‘hurrah patriotism’ among the social
democratic workers—but also little enthusiasm for the cause of the
international proletariat. Neither the leaders nor the masses of the
members were in any doubt about the fact that the looming war
was going to be a war against Russia. To encourage the victory of
the tzarist regime, considered, as in the time of Marx and Engels,
the supreme reactionary power of Europe, was the last thing the
SPD could wish. This fact, not a belief in the innocence of the Reich
leadership, was decisive for the behaviour of the Social Democrats
at the end of July and the beginning of August 1914.
The Russian general mobilization on 30 July made it easier for
the party leadership to find support for its position and to move
beyond its initial decision to abstain from the vote on war credits.
On 3 August the Reichstag fraction of the SPD voted 78 to 14 for
the approval of the government bill. On 4 August the head of the
party, the Königsberg lawyer Hugo Haase, himself one of the
577
defeated minority, read the fraction’s statement before the plenum
assembly of the Reichstag. It was a compromise between ‘realists’
and ‘internationalists’, attempting to do justice to both sides. Still,
it was an honest statement, faithfully representing the motivations
and the expectations of the party fraction.
Haase first attacked the imperialist policy of the arms race,
which the SPD had ‘fought with all its powers, but ultimately in
vain’. He then spoke of the ‘iron fact of war’ confronting all, and of
the ‘looming horrors of enemy invasion’. He emphasized that the
‘victory of Russian despotism’ would place the German people and
their future freedom in great danger, if not bring about their utter
destruction; ‘in harmony with the International’, he invoked the
right of every people to national independence and self-defence. He
condemned all wars of conquest, demanding ‘that, once the goal of
security has been achieved and the adversary is prepared for peace,
an end be made of war through a peace that permits of friendship
with the neighbouring peoples.’ The two most significant sentences
of the speech were the following: ‘We shall put into practice what
we have always preached: in the hour of danger, we will not leave
our own Fatherland in the lurch ... Guided by these principles, we
herewith approve the desired war credits.’
The minutes record ‘spirited cheers’ after the first sentence,
‘spirited applause by the Social Democrats’ after the second. When
the Social Democrats stood up from their seats after the final vote
on the war credits, there was ‘repeated stormy applause and
clapping in the house and in the stands’. At the salutation of the
‘Emperor, People, and Fatherland’, the Social Democrats, too, stood
up. But only few of them joined in vocally. In doing so, they
578
ignored the party resolution that Haase, incensed over the applause
with which several of his party colleagues on the ‘right’ greeted the
chancellor’s speech, had brought about during a pause in the
parliamentary session. Bethmann Hollweg had also made a
demonstrative gesture towards the Social Democrats. Singling out
the largest fraction, he proclaimed that the whole German people,
‘united to the last man’, stood behind the army and navy—a
statement that set off ‘joyous applause lasting several minutes’.
In November 1914, three months after the Social Democratic
fraction had voted unanimously for the war credits, thus enabling
their unanimous passage through the Reichstag, Lenin, living in
exile in Switzerland, identified the ‘transformation of the present
imperialist war into a civil war’ as ‘the only authentic proletarian
slogan’. In this sense, the SPD was no longer a ‘proletarian’ party, if
indeed it had ever been one. It could not desire a civil war, which
threatened to destroy everything it stood for. To prevent armed
conflict was beyond its power, but it shared in the German war
effort without making its support conditional upon far-reaching
domestic reforms. And of course, there was absolutely no guarantee
that the leadership of the Reich and the military would keep the
promise made by the emperor in his speech on 4 August, a promise
the Social Democrats referred to in order to justify their support for
the war: ‘We are not driven by a desire for conquest!’
When the war began, the social democratic movement had
become far more integrated into German society than it and its
adversaries had hitherto been willing to admit. In voting for the
war credits and the ‘truce’ (Burgfriede), a kind of domestic-political
ceasefire, the movement relinquished its commitment to class
579
struggle, at least for the time being. This could have consequences
for its relationship to the bourgeois centre, though such
implications were not foreseeable on 4 August 1914. The
consequences for the relationship between the SPD and the other
socialist parties of western Europe, on the other hand, were already
in evidence. The German Social Democrats had to accept the
charge that they had opportunistically betrayed the international
solidarity of the working class they themselves had invoked. In
truth, the last phase of the July crisis witnessed the most radical
transformation of the SPD in the party’s history. Whether it would
have survived the year 1914 politically intact without the new
‘social patriotism’ (so-called by leftist critics) is an open question.31
580
beggars description war has been forced upon us, a war for whose
justification the rational mind searches in vain. The Emperor, with
untiring solicitude, has sought to spare the world nameless misery.
It was in vain! The die has been cast. The mobilization has been
proclaimed.
581
the military clergy, Militäroberpfarrer Balthasar Poertner, to his
hometown in September 1914, ‘and Germany’s cause in this war,
forced upon us with such wanton perversity, is the cause of justice,
therefore of God’. In 1915 the Paderborn Old Testament scholar
Norbert Peters called the war a ‘holy war’ and a battle ‘for God
against Satan’, ‘for the old commandments of Sinai against the
hellish moral doctrine of modernity from beyond good and evil’.
The bishop of Speyer, Michael von Faulhaber, declared that Jesus’s
teaching ‘Give to Caesar what is Caesar’s’ was to be considered the
‘eternal precept of the civil conscience, even for the sacrificial
journey to the fields of death ... for the necessary holy war of the
nations’.32
Sermons in the nations of the Entente were certainly no less
patriotic than in the German Reich, and what was true of the pulpit
was true of the lectern. However, the shrill tone of one particular
early manifesto of German war nationalism provoked worldwide
outrage. The October 1914 ‘Aufruf an die Kulturwelt’,* signed by
well-known artists and many of the most famous scholars, justified
the attack on neutral Belgium without reservation, denied German
crimes against the civil population of the country, and culminated
in the statement: ‘Without German militarism, German culture
would long ago have been wiped from the face of the earth.’
German professors confronted the French-promoted ‘ideas of
1789’ with the ‘ideas of 1914’. This term was coined in 1915 by the
Münster economist Johann Plenge. Its most important popularizer
was Rudolf Kjellén, a Swedish constitutional scholar whose
advocacy of the German cause made him very popular in Germany.
582
The ‘ideas of 1914’ represented a rejection of liberalism and
individualism, democracy and universal human rights—in short,
the values of the Western world. German values were duty, order,
and justice, which could only be guaranteed by a strong state.
‘Since 1789, there has been no revolution on earth like the German
Revolution of 1914,’ wrote Plenge.
583
extent merely a temporary life-condition of our people ... It is,
nonetheless, the first authentic ‘socialist’ society and its spirit is the
first authentically active emergence of a socialist spirit, as opposed
to one made up exclusively of ill-defined demands. The pressure of
war has driven the socialist idea into German economic life. Its
organization grew together in a new spirit, and in this way the self-
assertion of our nation gave birth to the new idea of 1914 for
humanity, the idea of German organization, the people’s
cooperative of national socialism.
584
model, as admired as it was envied. This led to a kind of love–hate
relationship and made a dramatization of the German–English
antagonism, which found popular expression in the salutation ‘God
punish England!’, seem necessary. Secondly, Russia was ‘used up’
as Germany’s most deadly enemy only one year after the beginning
of the war, after the Eighth Army, under the nominal command of
the reactivated General Paul von Hindenburg, confronted the
numerically greatly superior Russian forces that had invaded East
Prussia at the end of August 1914, defeating them first at
Tannenberg, then, about two weeks later, on the Masurian Lakes.
After the German military, in summer 1915, succeeded in
conquering Lithuania, Courland, and Poland, the nation directed its
undivided attention to the western theatre. There, after the
indecisive outcome of the battle of the Marne in September 1914,
the front stabilized and the war turned into a ‘war of position’
(Stellungskrieg).
The Catholic philosopher Max Scheler was one of the first to
maintain that the war was ‘first and last a German–English war’.
His explanation, detailed in the book Der Genius des Krieges und der
deutsche Krieg,* made it clear that the author was especially
interested in convincing social democrats of the unavoidability of
Britain’s defeat. Every war against Britain, as the ‘motherland of
capitalism’, Scheler wrote, was also a ‘war against capitalism and
its outgrowths as a whole’. The book appeared in 1915. This was
the year Germany commemorated the centenary of the birth of
Otto von Bismarck and the 500th anniversary of the Hohenzollern
accession to power in Brandenburg.
That same year, the economist Werner Sombart characterized
585
the war between Britain and Germany as a war between
‘businessmen and heroes’ (Händler und Helden; this was the title of
his patriotische Besinnungen*). He contrasted British
‘commercialism’, which he described as execrable, with the positive
example of German ‘militarism’. ‘Militarism is the heroic spirit
elevated to the spirit of the warrior. It is the highest unity of
Potsdam and Weimar. It is Faust and Zarathustra and Beethoven-
score in the trenches. For the Eroica and the Egmont-overture, too,
are purest militarism.’
For both scholars, however, justifying the war as such was more
important than the emphasis on the German–British antagonism.
Scheler quoted the statement, attributed to Treitschke, that war is
an ‘examen rigorosum of states’. He himself called war the ‘large-
scale psychotherapeutics of peoples’ and claimed that the ‘genius of
war’ would become ‘a religion of its own—a guide toward God’.
Sombart believed that war could provide salvation from ‘cultural
pessimism’, from ‘the life of the vulgar, the life of an ant’. To raise
‘German heroes’, ‘heroic men and heroic women’, was to be the
highest priority of all education, an education designed to
harden the body and to develop all its powers in harmony, in order
that we might see grow up a generation of courageous, broad-
chested, bright-eyed young people. For the Fatherland needs them.
Broad-hipped women to breed strong children, strong-boned, wiry,
courageous men with endurance who will make good warriors.
586
of these centuries’, the ‘representative of divine thought on earth’,
‘the people of God’.
Less martial, but no less patriotic, was the ‘intellectual service
under arms’ that Thomas Mann felt to be his duty. In the autumn of
1914, Mann compared the war to the battle Frederick the Great
had to wage against a powerful alliance of enemies from 1756 to
1763. From 1916 to the end of 1917, the author of Buddenbrooks
worked on his Betrachtungen eines Unpolitischen.† He was finishing
this book when, in December 1917, the peace negotiations began
between the German Reich and Bolshevik Russia in Brest-Litovsk.
Thomas Mann was relieved that war now no longer had to be
waged against the land of Dostoevsky, to whom he felt a kind of
mystical connection, but only ‘against the West alone, against the
trois pays libres [‘the three free countries’ (H.A.W.)], against
“civilization”, “literature”, politics, the rhetorical bourgeois’.
Against the West and its democracy, no matter whether in
French, English, or American form, Mann sought to defend what he
considered the profound essence of Germany: music, poetry, and
philosophy. German Kultur, which he opposed to Western
civilization, required the authoritarian state criticized in the West
in order to protect it from the deleterious influence of politics. ‘The
politicization of the German concept of art would mean its
democratization, an important feature of the democratic levelling
and realignment of Germany.’ Since Mann saw the progress of
democracy in Germany, too, and did not think it could be stopped,
his protest was, at heart, without hope, an expression of the
educated bourgeoisie’s discontent with Western civilization and, at
the same time, a farewell to that world ‘of a resigned inwardness,
587
sheltered by power’ the poet was to commemorate many years later
in his Munich speech on the ‘Leiden und Grösse Richard Wagners’*
of 10 February 1933.33
Initially, German intellectual war propaganda made no mention
of concrete military goals. An open discussion of the goals the
country pursued would have contradicted the notion of the war’s
purely defensive character; thus, such a discussion was forbidden
until November 1916. Behind the scenes, however, plans of
conquest were drawn up, both by the government and by political
associations and business interests. In September 1914 the
chancellor set down his ideas in a programme that amounted to a
German-controlled central Europe and thus a German hegemony
over Europe. His catalogue of demands included the annexation of
the north Lorraine ore basin of Longwy-Briey, the city of Belfort,
the transformation of Luxembourg into a German federal state, and
the demotion of Belgium to a German ‘vassal state’. The new
Belgium was to cede Lüttich (Liège) and Verviers to Prussia and
receive in return the French part of Flanders ‘with Dunkirk, Calais,
and Boulogne, mainly Flemish in population’. Concerning Russia,
all that was said for the moment was that it had to be ‘driven from
the German border and its power over the non-Russian vassal
nations broken’. The neighbour states, including Austria-Hungary,
France, and possibly ‘Poland’ were to join a central European
economic league, ‘with its members in nominally equal
partnership, but de facto under German leadership’.
It was no coincidence that Bethmann Hollweg’s goals were in
conspicuous agreement with those of Walther Rathenau, who in
588
August 1914 had been given leadership of the war commodities
department in the ministry of war. The demand for a central
African colonial empire came from the imperial colonial office at
about this same time. The government’s war aims found the
support of the most important branches of industry as well as the
Deutsche Bank. The ‘September programme’ was thus a kind a
lowest common denominator among the parties who thought
clearly about Germany’s war goals in the autumn of 1914.
The Pan-Germans and individual heavy industrialists went much
further. Already at the end of August 1914 Heinrich Class,
president of the Pan-German League, was demanding territorial
concessions from France much more comprehensive than those of
the chancellor shortly afterward. The face of Russia was to be
‘powerfully redirected to the east’ and ‘basically thrown back into
the borders of the time before Peter the Great’; the Baltic region, as
well as parts of Russian Poland, White Russia, and northwest
Russia, were to be settled by Germans, and the Russian Jews
transferred to Palestine. In September 1915 the Centre delegate
Matthias Erzberger, at that time a political confidant of the Thyssen
concern, went so far as to demand the ‘splitting up of the Russian
colossus’. August Thyssen himself, in September 1914, had spoken
out in support of the annexation of Belgium, certain départements of
France, and, in the east, Russia’s Baltic provinces. He also desired,
if possible, German control of the territory around Odessa, the
Crimea, the territory around Azov, and the Caucasus, in order to
secure Germany’s raw material needs for the future.
In May 1915, in separately published memoranda, the leading
economic associations, among them the pillars of the ‘Cartel of the
589
Productive Estates’ of 1913 (the Federation of Agriculturalists, the
Central Association of German Industrialists, and the Imperial
German Middle Class Association) and the export-oriented League
of Industrialists, adopted the Pan-German programme. They were
joined a month later by a large number of university teachers, civil
servants, and artists. The ‘professors’ petition’, inspired by the
Berlin theologian and scholar Reinhold Seeberg, a native of the
Baltic region, called for the expulsion of large segments of the
native population from the territories to be ceded by Russia and,
for the purposes of ‘Germanization’, their replacement with
German farmers from other regions of the tzarist empire, or even
with people from overpopulated Germany. Estonians, Latvians, and
Lithuanians were to be employed as migrant labour in German
agriculture. A more moderate group of German intellectuals, led by
Theodor Wolff, publisher of the Berliner Tageblatt, as well as the
historian Hans Delbrück, responded with another memorandum in
July 1915. They rejected territorial gains in the west with the
argument that ‘the incorporation or annexation of politically
independent peoples, who are long accustomed to independence, is
reprehensible.’ The path to expansion in the east, however, the
moderate imperialists also left open.
Prominent signatures stood on both documents. Classical
philologist Ulrich von Wilamowitz-Moellendorff, historians Eduard
Meyer and Dietrich Schäfer, economist Adolf Wagner, and jurist
Otto von Gierke signed the Seeberg petition; the Wolff–Delbrück
text was endorsed by physicists Albert Einstein and Max Planck,
theologians Adolf von Harnack and Ernst Troeltsch, sociologists
Max and Alfred Weber, constititional scholar Gerhard Anschütz,
590
historian Max Lehmann, and economist Gustav von Schmoller.
There was, however, a conspicuous difference in the respective
number of signatories. The moderate petition counted only 191
personalities from German intellectual life; the Pan-German
document was signed by 1,347 people, including 352 professors.
The longer the war lasted, the more dissatisfied the Pan-
Germans and with them the whole of the German right became
with chancellor’s policy, which Bethmann Hollweg himself later
was to call a ‘policy of the diagonal’, meaning an attempt to square
the circle. On the one hand, he had to make an effort to retain the
support of the Social Democrats, who, at the end of June 1915,
clearly reiterated their refusal to countenance German annexations.
On the other hand, he had to make the expansionist military and
the corresponding part of the German public believe that he was no
less committed than the military leaders to enlarging and
strengthening Germany through the war.
The chancellor’s actual position was most closely represented by
the advocates of a German-dominated Mitteleuropa. For this group,
the hegemony of the Reich was to be primarily a material one
secured by economic treaties, only secondarily through territorial
expansions. Friedrich Naumann’s 1915 book Mitteleuropa, which
returned to Heinrich von Gagern’s old idea of a ‘wider federation’
between Germany and Austria-Hungary (without referring
explicitly to Gagern), rapidly became a kind of Bible for the
moderate school of German war imperialism. In calling for the
transformation of the federation into an ‘existential partnership
from top to bottom’, the liberal parliamentarian and publicist not
only took up the legacy of grossdeutsch foreign policy; he also
591
forged domestic links to the forces that had resisted the kleindeutsch
solution most tenaciously: Catholics, southern German democrats,
and Social Democrats.
Naumann’s appeal for a Mitteleuropa that was ‘German at the
core’, organized around a federally structured German-Austrian-
Hungarian economic area, had qualities that seemed to make it
into a kind of ‘Columbus’s egg’. However, the author gave his plan
a higher historical respectability by pointing out that a ‘world-
power of Mitteleuropa’ had once existed in the past.
The Germans filled the middle of Europe, but on all their borders
they drew neighbouring peoples to themselves: the Holy Roman
Empire of the German Nation. Now, during the war, this old
Empire is shifting and pressing up again under the earth, for it
wants to rise again after a long slumber.
592
Poland, which the central powers had transformed into a formally
independent ‘Kingdom of Poland’ in November 1916. And for their
part, the German military and political right were less inclined than
ever to agree to a ‘separate peace’ in the east without massive
Russian territorial concessions.
In the west, the war of position continued after the failure of the
Anglo-French Somme offensive and the bloody battles near Verdun
in autumn 1916. The decision to conduct unrestricted submarine
warfare, however, taken by the crown council on 9 July 1917
against the opposition of Bethmann Hollweg, proved fatal for the
central powers. Torpedo attacks against British and French
passenger liners by German submarines had already caused the
death of a large number of American citizens since the start of the
war. Now, the new resolution seemed like a calculated challenge to
the United States, which responded by breaking off diplomatic
relations and declaring war on Germany on 6 April 1917. The
central powers had no realistic chance of winning the war after
that.
By this time, the nationalist emotions and ‘war enthusiasm’ that
had run high among the urban bourgeoisie and petty bourgeoisie at
the war’s start were gone. The disappointment of the hopes for a
quick victory; widespread hunger in the course of the allied
blockade of German ports, which cut off the import of foodstuffs;
illicit trading and profiteering, which led to the sharpening of class
antagonisms—all these things contributed to political discontent.
This was not only true of the workers and farmers, who had rarely
exhibited any surfeit of nationalist sentiment even at the beginning;
it was also true of the broadest strata of the bourgeoisie, who
593
identified themselves as thoroughgoingly vaterländisch. On the
political right, Pan-Germans and anti-Semites tried to make the
Jews into scapegoats. Rathenau’s appointment to head the office of
war commodities was seen as proof of their claim that Germany
was falling increasingly under Jewish control. Already in 1915
Hans von Liebig, a chemistry professor in Giessen, was referring to
Bethmann Hollweg as ‘chancellor of German Jewry’.
On 19 October 1916 the Centre party delegate Matthias
Erzberger called upon the chancellor to ‘provide a detailed
summary of the entire personnel of all war associations, divided
according to sex, age of military service, salaries, confession’. A few
days before, on 11 October, the Prussian ministry of war, reacting
to a flood of complaints about supposed Jewish ‘shirking of duty’,
had ordered ‘statistics on the Jews’ (Judenstatistik) to be compiled
for the army. This survey represented nothing less than the official
governmental recognition and legitimation of anti-Semitism. This
was a profound historical caesura, and it shocked not only the
Jews.
The results could only disappoint the initiators of the measure.
The proportion of Jews among those who participated in the war,
in both its achievements and its sacrifices, corresponded to the
proportion of Jews in the population. To be sure, the statistics only
became known after the war. Even if they had been published at
the time, however, they would probably not have affected the
views of the anti-Semites. When, in September 1914, the national
economist Franz Oppenheimer paid tribute to the Jewish Social
Democrat Ludwig Frank (originally from Baden), the first Reichstag
delegate to fall in the war, the anti-Semitic press responded in a
594
manner he summarized in one sentence: ‘Don’t get your hopes up;
you are and remain the pariahs of Germany.’34
On the political left, protest against the policies of the Reich
leadership and the Social Democrats supporting them began
already in the autumn of 1914. On 2 December Karl Liebknecht,
son of the party founder Wilhelm Liebknecht and a lawyer in
Berlin, became the first Reichstag delegate to vote against new war
credits. Nineteen other Social Democrats followed his example on
21 December 1915, among them Hugo Haase, the party head. In
January 1916 Liebknecht was excluded from the parliamentary
fraction. Following another ‘breakdown of discipline’, which the
majority punished by excluding the offender from the fraction,
eighteen opposition delegates united to form the Social Democratic
Working Group (Sozialdemokratische Arbeitsgemeinschaft) in March
1916. Most of them belonged to the old pre-war left. Criticism of
the official claim to a ‘defensive war’ was heard far beyond the left
wing, however. The ‘centrist’ Kautsky, who held no parliamentary
mandate, and the ‘rightist’ Bernstein also both rejected the majority
line.
But the ‘left’ was by no means all of one mind. The delegates
Haase and Wilhelm Dittmann, along with Rudolf Hilferding
(originally from Vienna), a physician, Marxist theorist, editor of
Vorwärts, and author of the popular 1910 book Das Finanzkapital,
belonged to the moderate camp. Liebknecht, Rosa Luxemburg,
party historian Franz Mehring, and Clara Zetkin, champion of the
socialist women’s movement, were among the radicals (though
only Liebknecht was a parliamentary delegate). From the spring of
1915 the extreme left had its own organization, the Gruppe
595
Internationale, which renamed itself the ‘Spartacus League’
(Spartakusgruppe) in the course of 1916. Other radical leftist groups
included the ‘Left Radicals’ (Linksradikalen) in Hamburg and
Bremen and the ‘International Socialists of Germany’ (Internationale
Sozialisten Deutschlands) around the publisher of the journal
Lichtstrahlen, Julian Borchardt.
The radicalization of the German left was closely connected to
what happened among the left wing of the Second International. In
September 1915 leftist socialists, some from countries at war
(including Germany, France, Russia, and Italy), others from neutral
states, convened in Swiss Zimmerwald. From there they issued a
unanimous condemnation of socialist parties supporting the
‘imperialist war’, declaring them guilty of opportunistic treason
against the principles of the International. The conference had been
announced by Italian and Swiss socialists in mid-May 1915, one
week before Italy joined the war on the side of the Entente.
Moderate adversaries of the ‘truce’ policy, men like Bernstein,
Kautsky, and Haase, whom Lenin disparagingly referred to as
‘social pacifists’, were not invited to Zimmerwald. But the extreme
left around Lenin, which called for the ‘truce’ to be ended by a civil
war, itself represented only a small minority of the participants. A
unanimously passed resolution demanded a rapid end to the war, a
peace without annexations and indemnities, and the right of
national self-determination.
At the next conference, which took place in April 1916 in
Kienthal outside Bern, Lenin and his supporters again remained in
the minority. On two points, however, the Kienthal resolutions
went beyond those of Zimmerwald. In the first place, the
596
participants called for a ‘rejection of all support for war policies by
socialist parties’ and a refusal to endorse war credits. Secondly,
they accused the executive of the International of complete failure
and of having ‘collaborated in the policy of betraying its principles,
the policy of the so-called “defence of the fatherland”, and the
policy of the “truce”’. The result fell short of Lenin’s goal of
splitting the Second International, and it was still very far away
from laying the foundations for a new, revolutionary Third
International. Nonetheless, it was a sign that the international
workers’ movement had polarized even more.
In Germany leftist opposition to the ‘truce’ policy of the Social
Democrats increased measurably after the ‘swede’ or ‘rutabaga
winter’ of 1916–17. Famine was not the only thing that radicalized
the proletariat, however. Another factor was the increased
exploitation of workers and the role the trade unions played in it.
On 5 December 1916 the Reichstag, with the votes of the Social
Democrats, passed the Patriotic Labour Service Act (Vaterländisches
Hilfsdienstgesetz), which introduced national service for all men
between the ages of 17 and 60 who had not already been called up
for military service. It also required companies of more than fifty
employees to institute workers’ and employees’ committees and
created mediation commissions above the company level composed
of representatives from labour and management in equal
proportion.
The Labour Service Act was the Magna Carta of a programme
that forced German industry to convert to the mass production of
armaments. This was the so-called ‘Hindenburg programme’, which
was designed by Erich Ludendorff, the quartermaster-general of the
597
German army, and named after the ‘victor of Tannenberg’, Field
Marshal Paul von Hindenburg, from the end of August 1916 chief
of the Army High Command (Oberste Heeresleitung or OHL). While
the act gave the unions more influence, they became so closely
involved with the state and company management that, in the view
of many workers, they ceased to represent the interests of the
proletariat.
In April 1917 a wave of ‘wildcat’ mass strikes broke out in many
large German cities, partly in response to the February revolution
in Russia and the recent founding of the Independent German
Social Democratic Party in Gotha (Unabhängige Sozialdemokratische
Partei Deutschlands or USPD, formerly the Sozialdemokratische
Arbeitsgemeinschaft). In Berlin, the ‘Revolutionary Shop Stewards’
(Revolutionäre Obleute) from the metal industry, who came from the
left wing of the new USPD, first appeared on the scene. The prima
facie goal of the uprising was an increase in the bread ration. In
reality, however, it was the first large-scale workers’ protest against
the war, and it took place throughout the Reich. Within a few days,
labour unions from all political camps—‘free’ social-democratic,
Christian, and the liberal Hirsch–Dunker groups—managed to bring
about the end of the demonstrations. But the antipathy to the
‘truce’ did not weaken. And the workers were not the only ones
protesting against the war at this time; soldiers, too, became
increasingly restless throughout 1917. In the German navy, hunger
strikes and cases of absence without leave multiplied. The military
courts responded with draconian—and legally untenable—
punishments against the ‘ringleaders’. Some ten sailors were
condemned to death. Two, Max Reichpietsch and Albin Köbis, were
598
executed on 5 September 1917. The outrage and bitterness of the
sailors reached a new high.
The government’s efforts to pacify the domestic scene in the
spring of 1917 were insufficient. The ‘Easter message’ of 7 April,
which Bethmann Hollweg had wrung from the emperor, promised
constitutional reforms after the end of the war, including a
restructuring of the Prussian House of Lords and House of
Representatives. William also renounced the three-class voting
system, but left the decision about whether it would be replaced by
a plural system or general direct suffrage to the constitutional
organs. This was not enough to satisfy the Social Democrats. It also
did nothing to strengthen the chancellor’s position on the ‘right’.
Around this same time, Germany’s most important ally showed
increasing symptoms of crisis. Between January and April 1917
Emperor Charles, the successor of Francis Joseph, who had died in
November 1916, had tried in vain to conclude a separate peace
with France (and had brought a German renunciation of Alsace-
Lorraine into the discussion). The failure of these peace feelers
made the Habsburg monarchy even more dependent on Germany,
but did not alter the wishes of the Vienna government to bring the
war to an end as quickly as possible in order to preserve the
multinational empire.
For this purpose Charles and his foreign minister, Count
Czernin, attempted to act upon one of the most influential
members of the German Reichstag, Matthias Erzberger, who, in
agreement with Bethmann Hollweg, had gone to Vienna at the end
of April. By this time the Centre delegate was acquainted with
Czernin’s extraordinarily bleak assessment of the military situation,
599
which the latter had set down in a memorandum on 12 April (the
foreign office in Berlin had permitted Erzberger to look at the
document). The Vienna pessimism was partly responsible for
Erzberger’s transformation during those months from an
annexationist to an advocate of a peace agreement and critic of
unrestricted submarine warfare, the failure of which was obvious
by that time. Erzberger’s change of position influenced the attitude
of his party, and the Centre moved somewhat to the left.
The Majority Social Democrats (the MSPD) did the same thing in
June of 1917, but for other reasons. At an international conference
in Stockholm, convened at the prompting of Dutch and
Scandinavian socialists and attended by representatives from both
German Social Democratic parties, the MSPD adopted the maxim of
the Petrograd Workers’ and Peasants’ Soviet, ‘peace without
annexations and contributions’. Shortly thereafter, on 26 June
1917, the Social Democratic Party committee gave Bethmann
Hollweg an ultimatum: the party would not take a final position on
the new war credits bill until the administration had made a clear
statement concerning both its war aims and its domestic plans. The
Russian February revolution, the founding of the USPD, the April
strikes— all these things came together to provoke the majority
SPD into threatening an end to the ‘truce’.35
Prisoner to his own ‘policy of the diagonal’, the chancellor could
not satisfy the demands of the Social Democrats—neither in regard
to a peace treaty nor with respect to the constitution—without
provoking storms of protest from the right, where the Pan-Germans
had long been waiting for an opportunity to call for a general
mobilization against Bethmann Hollweg. On 2 July, having just
600
returned to Berlin from the General Headquarters, the chancellor,
in the presence of the most important ministerial secretaries,
informed the leaders of the larger parliamentary fractions that he
did not intend to comply with the wishes of the Social Democrats.
In so doing, he stripped himself of the parliamentary majority that
had supported him since 4 August 1914.
Ludendorff, too, went against Bethmann Hollweg, but for
entirely different reasons than the Social Democrats. The strong
man of the OHL wanted to replace Bethmann Hollweg with a head
of government who would follow his instructions. Through Colonel
Max Bauer, his most trusted confidant, the First Quartermaster-
General informed Erzberger in June about the seriousness of the
military situation. The politician could only interpret this
communiqué as an exhortation to work for the chancellor’s ouster.
Erzberger also knew about Pope Benedict XV’s intention to appeal
to the warring countries for peace. When on 6 July, after
consultation with the Majority Social Democrats and Progressives,
the Centre delegate directed an impassioned attack against the
policy of unrestricted submarine warfare and demanded a peace
initiative from the parliament, he knew what he was doing.
Erzberger’s speech triggered the ‘July crisis of 1917’.
That same day, National Liberals, Centre, Progressive People’s
Party, and Majority Social Democrats—who had been cooperating
since April 1917 in the Reichstag’s newly formed constitutional
committee with the goal of reforming the federal constitution—
founded a new body, the Interparty Committee (Interfraktioneller
Ausschuss). But it was quickly evident that there were still sharp
antagonisms between the fractions over the main questions. While
601
the National Liberals had changed their mind about the
appointment of parliamentarians to governmental office, they still
had serious reservations against general direct suffrage for Prussia
and rejected a peace resolution on the basis of the ‘Peterograd
formula’. Nonetheless, the chancellor’s removal was a priority,
especially for their delegate Gustav Stresemann, who had advanced
to ever greater power and influence.
The three other parties considered the peace initiative more
important. Out of consideration for its fraction in the Prussian
House of Representatives, the Centre kept a low profile on the
suffrage question. And despite certain remarks from Erzberger to
the contrary, the majority of Centre delegates still found
themselves unable to endorse the idea of parliamentary rule in the
Reich. The Majority Social Democrats backed both the introduction
of the Reichstag suffrage law in Prussia and, in principle,
parliamentary rule. But with the exception of the party right wing,
they did not desire to participate in a coalition government
themselves, wishing to avoid conflict with the parliamentary and
extraparliamentary left. The Progressive People’s Party was
unreservedly for both equal suffrage and parliamentarization and,
in the unlikely event that Bethmann Hollweg did adopt these goals,
for the chancellor’s continuing in office. A part of the Centre
around Erzberger and the National Liberals, on the other hand, was
of one mind with the OHL on the chancellor issue: Bethmann
Hollweg’s predecessor, Prince Bernhard von Bülow, should also be
his successor. Least concerned about the chancellor was the MSPD,
although virtually nobody in the party trusted him any longer.
Then, for a short period, Bethmann Hollweg’s position seemed
602
to stabilize. William II was outraged over the machinations of the
military and had no intention of appointing Bülow again. Prompted
by Bethmann Hollweg, he even decreed (on 11 July) the
introduction of equal suffrage in Prussia. After this, the anti-
Bethmann Hollweg bloc began to crumble. But the next day,
induced by Ludendorff, Crown Prince William intervened. The most
important result of his conversations with selected opponents of the
chancellor from almost all fractions was that Erzberger was able to
commit the Centre to a statement against Bethmann Hollweg.
The dramatic climax of the crisis came on the afternoon of 12
July when Hindenburg and Ludendorff submitted their letters of
resignation. The telegram in which Ludendorff announced his lack
of confidence in Bethmann Hollweg was a historically
unprecedented challenge to the monarch. But it had the desired
effect. William II submitted to the will of the two generals, whom
the great majority of the people still considered to be the
guarantors of German victory.
On that same 12 July, the Majority Social Democrats, Centre,
and Progressive People’s Party agreed to the text of a peace
resolution in the Interparty Committee. The Reichstag sought a
peace through ‘understanding and lasting reconciliation among the
nations’, the statement read. ‘Forcible territorial expansion is
incompatible with such a peace, as are other kinds of political,
economic, and financial rape.’ The language was chosen to allow
an interpretative freedom sufficient to accommodate an expansion
of the German sphere of influence. This was decidedly too little for
the National Liberals, however, who not only did not endorse the
resolution, but walked out on the committee. Nor did Bethmann
603
Hollweg, who was still in office at this time, show any inclination
to adopt the position of the three fractions (called ‘majority parties’
from this point forward). The OHL protested by telegram and
demanded that the chancellor—in whom it had just officially
declared its lack of confidence—prevent the adoption of the
resolution. Bethmann Hollweg knew that he was no longer able to
do this. On the evening of 12 July he personally asked the emperor
to accept his resignation, submitting his request in writing the next
day.
Hindenburg and Ludendorff, ordered to Berlin by William II,
conducted interviews with the party leaders on the same day, 13
July. But there was no solution to the conflict over the peace
resolution. The Social Democrats, Centre, and Progressives insisted
on the text they had agreed to the day before. Shortly thereafter,
William II accepted Bethmann Hollweg’s resignation, appointing a
successor on 14 July: Georg Michaelis, a government jurist,
Prussian state commissioner for food procurement, and a complete
novice to politics. Ludendorff had approved his appointment in
advance. Michaelis himself reports in his memoirs that on the
morning of 13 July, when he received the emperor’s request, he
had read in the ‘Watchwords’ of the Moravian Church (Herrnhuter
Brüdergemeinde) the passage from Joshua 1: 9: ‘Have not I
commanded thee? Be strong and of a good courage; be not afraid,
neither be thou dismayed’. He understood the passage as a sign
that he ought not to refuse the entreaty of his highest superior.
The outcome of the July crisis of 1917 was not a success for the
majority parties. They had allowed themselves—in part
consciously, in part unconsciously—to be used as tools of the Army
604
High Command, and now they had to suffer the consequences.
After 14 July they were forced to deal with a chancellor far more
remote from them than his luckless predecessor. In an article for
the Frankfurter Zeitung at the beginning of September 1917, Max
Weber described the crisis as a ‘classic paradigm’ of ‘how the lack
of normal parliamentary rule operates in crisis situations. There
will never be any change unless the Reichstag parties are
continually forced to express a clear position both on the questions
of policy and with regard to the participants.’ The ‘German
domestic question’ was thus as follows: ‘How may the parliament,
condemned to a negative politics by dint of its current structure, be
transformed into an agent of political responsibility?’
When Bethmann Hollweg’s successor took office, the cause of
parliamentary rule seemed, at first glance, to have advanced
considerably. A significant number of politicians from the ranks of
the majority parties and the National Liberals entered high
positions in the federal administration and the Prussian ministries.
The Centre politician Peter Spahn, for example, became Prussian
minister of justice; Rudolf Schwander, the mayor of Strasbourg and
close associate of the Progressive People’s Party, became
undersecretary, then, in October 1917, secretary of the newly
formed Imperial Economic Office; the majority Social Democrat
August Müller was appointed undersecretary of the War Nutrition
Office. Nonetheless, these appointments did not make the majority
parties and National Liberals into ‘agents of political responsibility’
in Weber’s sense. It would be more accurate to speak of an attempt
by the Reich leadership to discipline the parliamentary majority by
drawing it into the bureaucracy. As far as the key question was
605
concerned, Michaelis wasted no time in making it clear—though in
somewhat qualified form—that he was no man of the majority. In
the Reichstag debate on 19 July, which preceded the adoption of
the peace resolution, he announced that the aims of the Reich
leadership could be ‘achieved within the framework of your
resolution, as I interpret it’.
The majority parties seemed not to notice—or else they
deliberately ignored—this reservation, which could be taken to
turn the resolution into the very opposite of what it was intended
to be. The parliamentary minutes record calls of ‘“bravo!” and
“very good”! from the Centre, Progressive People’s Party, and
Social Democrats’. The assembly passed the peace resolution by
212 to 126 votes with 17 abstentions. The MSPD delegates all
voted positively; the Progressive People’s Party did so with one
exception, and the Centre with five. On the next day, the Majority
Social Democrats assisted in the passage of the new war credits bill,
approving 15 billion marks’ worth of government expenditures.
The peace resolution was greeted with approval by the
constituents of the majority parties. The right reacted with
hostility. The nationalists responded by establishing the German
Fatherland Party (Deutsche Vaterlandspartei) on 2 September 1917
in York Hall of the East Prussian Landschaft in Königsberg. The
moving spirit behind the party’s founding was the jurist Wolfgang
Kapp, head of the Landschaft, a public credit office for agriculture.
Despite its name, the new party did not look upon itself as a party
among other parties, but as a supra-party association uniting all
‘patriotic’ forces. Its main catchment area was in Prussia east of the
Elbe. The party activists came primarily from among the educated
606
bourgeoisie employed in state offices, or from the manorial class,
and they were mostly Evangelical in confession. The main
constituents came from the ranks of the conservatives and National
Liberals. Numerous nationalist associations joined the Fatherland
Party as corporations, effectively doubling its direct and indirect
membership. This fact must be taken into consideration when
evaluating the party’s own records; the claims of 450,000 members
in March 1918 and 800,000 at the beginning of September 1918
both represent significant exaggerations.
It says much about the social character of the new association
that its leadership was assumed by a member of the ruling house,
Duke John Albert of Mecklenburg, co-founder of the Pan-German
League and president of the German Colonial Society after 1895,
and by the former secretary of the imperial naval office and creator
of the German navy, Admiral of the Fleet von Tirpitz. The
Fatherland Party was the culmination of Sammlungspolitik from
above, called into life by servants of the state, represented by
members of the old ruling elite, and supported by bourgeois and
petty bourgeois devotees of the alliance of throne and altar, the
two most important symbols of the authoritarian state. Despite the
‘populism’ of its platform and propaganda, however, it was not a
mass organization of the new type. Among the lower classes of
society, the ‘party’ that claimed it was not a party had virtually no
support.
In its founding declaration, the Fatherland Party did emphasize
that ‘without strong backing from the people’ the government
would not be able ‘to master the situation alone’. But the
explanation that the government required ‘a powerful tool for a
607
powerful federal policy’, a tool in the shape of ‘a great people’s
party, supported by the broadest patriotic circles’, quickly made
clear that the party had no interest in the will of the people.
Democracy was unequivocally rejected.
The only thing that could save Germany was, according to the
Fatherland Party, a ‘Hindenburg peace’, embracing ‘enormous
sacrifices and efforts as the price of victory’. While the founding
declaration did not go into detail on this matter, countless speeches
and petitions by the speakers of the national ‘unity party’ did. They
wanted Germany to achieve a peace of conquests and
aggrandizement at the expense of Belgium, France, Russia, and
Britain. A ‘Hindenburg peace’ was the diametric opposite of the
‘peace without annexations and contributions’ endorsed by the
Social Democrats—the ‘Scheidemann peace’ as the right called it,
after its most eloquent proponent, the delegate Philipp
Scheidemann.
If the founding of the German Fatherland Party was a response
to the peace resolution of the moderate parliamentary majority, the
608
founding of the People’s Federation for Liberty and Fatherland
(Volksbund für Freiheit und Vaterland) on 14 November 1917 was a
response to the nationalist propaganda of the Fatherland Party. The
new Federation sought to organize a movement for moderate war
aims, redress of social conflicts and disparities, and inner reforms.
Its participants included well-known professors like Friedrich
Meinecke, Hans Delbrück, Max Weber, Hugo Preuss, Lujo Brentano,
and Ernst Troeltsch; politicians and publicists from the ranks of the
Progressive People’s Party around Friedrich Naumann and Theodor
Heuss; and eminent representatives of the trade unions, among
them the Catholic labour leaders Johann Giesberts and Adam
Stegerwald, and the Social Democrats Carl Liegen and Gustav
Bauer.
The founding declaration of the People’s Federation contained
an appeal for the ‘maximal concentration of our energies’,
necessary ‘to break the destructive will of our enemies’. Domestic-
political restructuring in the sense of a ‘liberal consolidation of our
state institutions’ was to begin immediately, not wait until after the
end of the war. It was essential ‘to give a government willing to
reform firm popular support, and to draw the necessary conclusions
from the nature of the modern state, conclusions that today every
nation must draw in the context of its development’. Finally,
Germany needed a ‘clear foreign policy, sustained by the people
and the government, aspiring to a lasting peace, securing raw
material imports and foreign markets, and placing the life, honour,
and self-determination of the nations upon the basis of morality
and law’.
The People’s Federation did not explicitly endorse parliamentary
609
rule for Germany, but it supported the call for more democracy. It
did not demand annexations, but it also did not exclude them. The
increase in German economic power was one of the common goals
of all the signatories of the founding declaration. All in all, the
proclamations of the Federation and its spokesmen gave the
impression that the peace it sought was more the parliament’s
‘peace through understanding’ than the Fatherland Party’s ‘peace
through victory’.
The formation of a new socially inclusive, supra-confessional
political centre met with a weaker public response than the shrill
and arrogant tones of the Fatherland Party. The Federation’s
individual membership remained very modest, allegedly increasing
from 1,000 in May to 2,800 by October 1918. If we count affiliate
organizations, including all unions from diverse political camps, we
arrive at the impressive but misleading figure of 4 million
members. In reality the Federation was far too heterogeneous ever
to become a serious factor in domestic politics. Still, the
cooperation between moderate representatives of the bourgeoisie
and labour made it important. The People’s Federation for Liberty
and Fatherland (like the Association for Social Reform, which was
involved in its founding) was an expression of the class
compromise that had developed under the ‘truce’. Without such a
compromise, parliamentary democracy was inconceivable in
Germany.
At the time of the Federation’s founding, Michaelis was no
longer the chancellor. The second chancellorship crisis of 1917 was
provoked by extreme and untenable attacks by Eduard von Capelle,
secretary of the naval office, against the leaders of the USPD,
610
whom he accused on 9 October in the Reichstag of abetting a plan
of revolt in the high seas fleet. The accusations enraged not only
the party directly affected and the Majority Social Democrats, who,
through their party and floor leader Friedrich Ebert, declared open
war on the chancellor that same day. Among the Centre and the
Progressive People’s Party, too, Michaelis had lost all support by
this time. The belief was widespread that he was proceeding
against the Independent Social Democrats in the same way as the
pre-war government had pursued the movement as a whole. Even
for the National Liberals, who had been cooperating in the
Interparty Committee again as of the end of August, it was clear by
9 October that Michaelis could no longer be supported as
chancellor.
His successor, though a member of the Centre party, was not a
man after the heart of the parliamentary majority from July. The
Bavarian prime minister, Count Georg von Hertling, who assumed
the offices of chancellor of the Reich and Prussian prime minister
on 1 November 1917, rejected any further development towards
parliamentary rule; as a resolute federalist, he feared that it would
lead to a unified central government. He also gave no indication of
being any more sympathetic to the July peace resolution than
Michaelis. But his appointment also dissatisfied the political right.
Militantly Evangelical circles among the military, the National
Liberals, and the Fatherland Party felt the choice of a Catholic
chancellor during the height of the ‘Luther year’ of 1917, the 400th
anniversary of the Reformation, to be downright insulting.
Nonetheless, the change of chancellor was accompanied by signs
that could be interpreted as a creeping parliamentarization of the
611
state. For the second time in 1917, the Reichstag had brought
about the collapse of the administration, this time without the help
of the OHL, as in the case of Bethmann Hollweg. The vice-
chancellorship was assumed not by the incumbent Karl Helfferich,
favourite of the emperor and chancellor, but by the nominee of the
Interparty Committee, Progressive delegate Friedrich von Payer
from Stuttgart. The National Liberal parliamentarian Robert
Friedberg from Berlin became vice-president of the Prussian state
ministry. With these appointments, three of the Interparty
Committee’s four parties were represented in the executive of the
Reich and Prussia.
But it was not yet clear in late 1917 whether the November
reshuffle of the government would lead to an increase in the power
of the July parliamentary majority to affect matters of ‘grand
policy’. It was no longer even certain that the July majority was
still tenable. The Centre, for one, clearly distanced itself from the
peace resolution in November. On the other hand, in the debate
over Hertling’s inaugural speech, the Majority Social Democrats
expressly reserved the right to move against the new
administration if it provoked them, the largest German party,
through its actions or inaction. To begin the history of German
parliamentary monarchy in November 1917 would thus be to
exaggerate the significance of the caesura between Michaelis and
Hertling.36
612
THE MILITARY DEFEAT AND THE
‘REVOLUTION FROM ABOVE’
613
Bismarck had played a similar game with revolution before and
after the outbreak of the war with Austria in 1866, when he took
steps to mobilize the Hungarians, Czechs, and southern Slavic
peoples against the Habsburg monarchy. After Prussia’s military
and diplomatic victories, he did not find it necessary to continue
along this path. In 1917, faced with the prospect of an American
landing in Europe, the political leaders of Germany saw themselves
justified in acting according to a maxim Bethmann Hollweg had
pronounced already on 4 August 1914 with regard to the invasion
(which he himself considered ‘unjust’) of neutral Belgium:
‘Necessity knows no law!’
At first, news of the Russian October Revolution was greeted in
Germany with great relief. The ‘decree on peace’, resolved at
Lenin’s behest by the Second All-Russian Congress of the Soviets on
8 November and calling all belligerent nations to cease fire and to
start peace negotiations, impressed the war-weary working class in
Germany and Austria-Hungary, since it seemed to promise a swift
end to the war. The other appeal from Lenin’s Russia—that the
international working class lead the proletarian revolution to
victory everywhere—showed, as yet, no tangible effect.
Even the Majority Social Democrats, whom Lenin castigated as
‘social patriots’, temporarily joined the chorus in praise of the
peaceable Bolsheviks. Lenin and his followers had brought down
without much bloodshed the government of Prime Minister
Kerensky, which was obviously no longer politically tenable, but
still not prepared to negotiate a peace. Now, for the first time,
there was a chance for a separate agreement in the east. As early as
the end of November, however, an increasing number of Social
614
Democrats—both from the Majority party and among the moderate
Independents—began to accuse the Bolsheviks of employing
terrorist methods. After 19 January 1918 virtually no social
democrat in Germany expressed sympathy with the Russian path.
On this day, Lenin’s revolutionary troops forcibly dissolved the
Constituent Assembly (elected on 25 November), in which the
Bolsheviks controlled only a quarter of the seats. With this act,
Russia’s path toward dictatorship was irreversible.
Not only the Majority Social Democrats, but also the moderate
elements within the USPD were outraged at this violation of the
democratic tradition within the European workers’ movement. Karl
Kautsky, an outspoken critic of the Bolsheviks even before 19
January 1918, wrote, condemning Lenin, that the dictatorship of
one among several parties was not ‘the dictatorship of the
proletariat’ according to Marx and Engels, ‘but the dictatorship of
one part of the proletariat over another’. The dictatorship of a
minority, however, finds its strongest support in a loyal army, and
the more the minority substitutes armed force for the majority, ‘the
more it compels all opposition to avail itself of the appeal to the
bayonet and the fist instead of the appeal to the ballot, from which
it is cut off. Then, civil war becomes the form in which political and
social antagonisms are worked out.’
For Kautsky, civil war was the most brutal form of war, and
therefore a catastrophe.
In a civil war, every party fights for its existence, and the loser
faces complete annihilation ... Many people confuse civil war with
social revolution, consider it the form social revolution takes, and
615
are disposed to excuse the unavoidable violence of civil war by
thinking that without such violence revolution is not possible ... If,
according to the example of the bourgeois revolution, we should
say that the revolution is the same thing as civil war and
dictatorship, then we must be consistent and say also that the
revolution necessarily terminates in the rule of a Cromwell or
Napoleon.
Even on the far left, in the ‘Spartacus League’, the opinions on the
Bolsheviks’ January coup d’état were divided. Clara Zetkin and
Franz Mehring defended the dissolution of the Constituent
Assembly without reservation. The champion of the socialist
women’s movement went so far as to claim that not taking this step
‘would have been a crime, paired with foolishness.’ Rosa
Luxemburg, on the other hand, in ‘preventive detention’ since July
1916, considered it inexcusable that the Bolsheviks, after dissolving
the Constituent Assembly (an act she, too, approved), did not
immediately call for new elections. ‘Freedom only for the
supporters of the government, only for the members of one party—
be it ever so numerous—is not freedom. Freedom is only ever the
freedom of those who think differently.’ These are the two most-
quoted sentences of Luxemburg’s posthumously published ‘Die
Russische Revolution’. They were not, however, intended as an
argument for a liberal pluralism, but for a revolutionary, socialist
pluralism. When she wrote about the ‘freedom of those who think
differently’, the author was not thinking of ‘class enemies’ and
‘class traitors’.37
By the time Lenin broke up the Constituent Assembly, Germany
616
had come a good bit closer to a ‘separate peace’ in the east. The
guns had been silent since 5 December 1917. On 22 January peace
negotiations between Russia and the central powers commenced in
Brest-Litovsk. The main point of contention was the right of
national self-determination, which the Bolsheviks advocated, but
which, in practice, they were resolved to subordinate to the
exigencies of revolutionary class struggle, such as they interpreted
it. The German side sought not only to permanently withdraw the
entire Baltic region, Poland, and Finland from Russian rule, but
also to secure German control over the Ukraine, its raw materials
and grain fields, and these interests took precedence over the self-
determination of the respective peoples. It took an interruption of
the peace negotiations and a new German-Austrian offensive in
February 1918 to force the Russian government to yield.
On 3 March 1918 the two sides signed the Treaty of Brest-
Litovsk. The modern world had not yet known such a peace built
on conquest and violence. Russia was compelled to give up not
only Finland, Poland, Lithuania, and Courland,* but even the
Ukraine. Shortly thereafter, the Transcaucasian states declared
their independence. This cost Russia a third of its population and
agricultural land, and more than half of its coal reserves. The
formation of an independent Ukraine and the cession of Armenian
territory to Turkey considerably weakened Russia’s position on the
Black Sea. It preserved only a narrow access to the Baltic. Further
treaties at the end of August 1918 imposed 6 billion gold marks of
reparations and cost Russia the north Baltic provinces of Livonia
and Estonia. As before in the case of Courland, the greatest
pressure for the cessions came from the German-Baltic upper class
617
(as well as their representatives in Germany), who demanded that
‘their’ territories be separated from Russia. But even without this
pressure, it would not have been difficult to persuade German
public opinion in the spring and summer of 1918 that it was
necessary to create a buffer zone between Germany and the
Bolsheviks out of the ‘border states’ of east central Europe.
The Treaty of Brest-Litovsk flagrantly violated the peace
resolution of 19 July 1917. Nonetheless, the Reichstag accepted it
on 22 March 1918 by a great majority. Among the votes in favour
were those of the two bourgeois majority parties, the Centre and
the Progressive People’s Party. The MSPD, internally split into
opponents, advocates, and ‘neutrals’, abstained from the vote. The
USPD voted against the treaty. Before the vote the Majority Social
Democrats, together with the two other majority parties, had
passed a resolution expressing the expectation that the Reich would
observe the right to national self-determination in the case of
Poland, Lithuania, and Courland.
This resolution was also an attempt to overcome the tensions
that had been building for months between the Majority Social
Democrats and the bourgeois centre. The crisis between the
majority parties came to a head in January 1918 in a great wave of
strikes, provoked by the arrogant triumphalism that General Max
Hoffmann, chief of the General Staff East, had demonstrated in his
dealings with the Russian delegation in Brest-Litovsk under Leon
Trotsky. The uprising began on 28 January in the greater Berlin
area, where the Revolutionary Shop Stewards under Richard Müller
succeeded in convincing at least 180,000 workers to down their
tools. Most of the strikers worked in the metal industry, which was
618
crucial for the production of armaments. Within a few days,
workers from many German cities—Kiel, Hamburg, Leipzig,
Brunswick, Cologne, Bochum, and others—joined the strike. On 30
January Vorwärts, the main Social Democratic Party organ, was
banned. The next day saw the proclamation of a heightened state
of siege and the arrest of Wilhelm Dittmann, Reichstag delegate of
the USPD, as he spoke to striking workers in Treptow Park. A
military tribunal condemned him to five years in prison on 4
February 1918.
The two leaders of the MSPD, Friedrich Ebert and Philipp
Scheidemann, as well as the Prussian parliamentary delegate Otto
Braun from Königsberg, had themselves elected to the strike
committee on 28 January in order to bring the industrial action—
in which the unions did not participate—to an end as quickly as
possible. They achieved this goal on 4 February, when the strike
was ended in Berlin (in most other parts of the Reich it was over
earlier). Had it lasted any longer, the strike of the munitions
workers would have threatened Germany’s military power. For this
reason, the counter-measures of the military, police, and judicial
system were rigorous. Striking workers were arrested in large
numbers or conscripted into the army.
In his 1928 book Entstehung der Deutschen Republik, 1871–1918,*
the classical historian Arthur Rosenberg, a Communist party
delegate in the Reichstag until his break with Stalin in 1927,
described the January strike as a ‘great mass uprising against the
military dictatorship and General Ludendorff’. After the 1917
unrest in the navy, it represented the ‘second act of direct mass
action against the military dictatorship’, followed by a third act,
619
the revolution, in November 1918. The workers were rebelling
against hunger, the ‘state of siege’, the militarization of industry,
and the policy of a ‘peace through victory’. The USPD joined in the
strike, since, like the workers, it wanted to bring the war to an end
as quickly as possible, without annexations and indemnities and on
the basis of national self-determination. The MSPD joined the strike
leadership in order to prevent the workers from falling entirely
under the influence of the USPD, to avoid further deterioration in
the domestic-political situation, and to counter the accusation that
the social democratic proletariat was stabbing the armies on the
battlefront in the back.
Such accusations were made during the Berlin strike of the
munitions workers. In a sermon on 3 February Bruno Doehring,
court chaplain and pastor of the Berlin cathedral, called the
instigators of the strike ‘worthless and cowardly creatures’ who had
‘treacherously defiled the altar of the Fatherland with fraternal
blood’ and ‘poisoned the good blood of our people’. Then,
foreshadowing the later myth of the ‘stab in the back’ (the so-called
Dolchstosslegende), Doehring declared that the strike’s initiators had
rushed the misguided workers ‘from the places of calm productive
labour onto the streets, pressing the murder weapon into their
hands’ and ordering them to attack ‘their brothers from the rear,
even as they are lying before the enemy’.
The word ‘treason’ was also already part of Doehring’s
vocabulary. His sermon has the fallen crying from their graves,
‘such that it screeches to high heaven, “Treason! Treason against
our own people!”’ It seemed like an echo of the chaplain’s
accusation when, on 26 February 1918, the Conservative Party
620
leader Ernst von Heydebrand und der Lasa declared that the strike
was ‘nothing other than a simple matter of treason to country,
instigated and influenced by foreign agents and, unfortunately,
sustained partly by the influence of the German Social Democrats’.
It was this atmosphere, poisoned by accusations of treason, that
made the Majority Social Democrats fear the worst, not only for
themselves as a party and for the workers’ movement, but for
Germany as a whole. To end the war quickly was also their wish.
But the peace at the end of all battles was, they hoped, not going to
be only an external peace, but an internal one, too, as much as
possible. That the ‘truce’ would survive the war was neither to be
expected nor even desired. Without a certain amount of mutual
understanding between the classes, however, Germany threatened
to fall into civil war, such as was already happening in the Ukraine,
Finland, and parts of Russia in the winter of 1917–18.
For this reason, the leaders of the Majority Social Democrats
sought to build bridges between, on the one hand, the striking
workers and, on the other, the bourgeois parties of the
parliamentary majority and the leadership of the Reich. They did
not entirely fail. Despite their disapproval of the Social Democratic
strike tactics, the spokesmen of the bourgeois parties did not take
up the accusations of treason. The Reichstag majority of July 1917
managed to hold together until 22 March 1918, when it was re-
established after a fashion by means of a new resolution. In the
end, the common front against the extreme nationalism of the Pan-
Germans, the Conservatives, and the Fatherland Party proved
stronger than what divided the bourgeois centre from the moderate
left. The same was true with regard to another shared antagonism:
621
the majority parties of July 1917 were united in rejecting the
radicalism of the extreme left, regardless of whether it came from
the Revolutionary Shop Stewards or from the German followers of
the Bolsheviks in the Spartacus League. From this point of view,
too, the centre parties could ill afford a rupture with the Majority
Social Democrats.38
After the beginning of 1918, the inner development in Germany
was far more affected by a power other than the Russia of Vladimir
Ilyich Lenin—the America of Woodrow Wilson. The ‘Fourteen
Points’ of 8 January and the ‘Four Principles’ of 11 February 1918,
in which the president of the United States set down his ideas of a
just international peace settlement, were not only an answer to the
Bolsheviks, but also seemed to offer an alternative to the
retributive peace proposed by the heads of government in Paris and
London, Georges Clemenceau and David Lloyd George. Wilson’s
aspiration ‘to make the world safe for democracy’ appealed to
many more Germans than did Lenin’s call for a proletarian ‘world
revolution’. The idea of a ‘League of Nations’ was incomparably
more attractive than the project of a ‘Third International’. Even a
victorious Entente might be made to accept the right to national
self-determination for which the American president was
campaigning. To be sure, Wilson’s specific demands to Germany
included several that seemed totally unacceptable to a majority of
Germans, including the cession of Alsace-Lorraine to France and
the formation of an independent Polish state to include the
territories where the population was clearly Polish and with free
access to the Baltic. The evacuation and restoration of Belgium, on
the other hand, was far more difficult to baulk at—which did not
622
prevent the political right from doing exactly that.
In terms of militarily power, too, the United States became
increasingly important as the year 1918 progressed. American
troops played a significant role in stopping the last German western
offensive (begun at the end of March) in summer 1918. The
decisive moment came on 8 August—the ‘dark day of the German
Army’, as Ludendorff called it—when British tank battalions broke
through at Amiens. Thereafter the much-lauded ‘troop spirit’
deteriorated with each successive day. Cases of refusal to obey
orders, voluntary surrender to the enemy, absence without leave,
and desertion increased rapidly. The recollections in Ludendorff’s
Kriegserinnerungen* were not exaggerated and represented typical
incidents: ‘Retreating troops, passing by a fresh division that was
fighting bravely, called out: “strike breakers!” and “war
prolongers” ... In many places the officers no longer had any
influence, and allowed themselves to be swept along.’
‘War morale’ was in little better condition at home.
Dissatisfaction had long become general, not confined only to the
working class. Already in August 1917 Ritter von Brettreich, a
minister in the Bavarian government, declared that the Mittelstand
was in ‘a worse mood than all other circles at the moment’. A year
later, in August 1918, Ernst Troeltsch had occasion to observe even
among ‘patriotic’ and ‘war-believing’ farmers and cheese
manufacturers of the Allgäu a ‘veritably fanatical hatred breaking
out against the whole of the officers’ corps as the epitome of
injustice and privilege’. On 23 September Oskar Geck, a Reichstag
delegate from Baden, reported in a meeting between the SPD
fraction and party committee that in southern Germany there was
623
‘enormous bitterness against Prussia; not against the Prussian
people, but against the Junkers and the military caste. The
dominant feeling among us is that Prussia must bite the dust, and if
Prussia does not bite the dust, Germany will bite the dust because
of Prussia.’39
At that same meeting, SPD party chairman Friedrich Ebert
discussed the practical implications of the traditional elite’s loss of
power. He placed before his party colleagues a clear choice:
624
Before the war, a majority of the party would have condemned
such a policy as a betrayal of the fundamental Marxist principle of
class struggle. The experience of class cooperation during the
‘truce’ and within the Interparty Committee had changed the
situation. Nonetheless, the shift from parliamentary cooperation
with the bourgeois centre to official participation in government
was still going to require a political decision. This decision was
made easier—indeed, made possible—by the fact that the dogmatic
Marxists had left the party after the majority had voted to approve
the war credits. This split did represent a burden on the social
democratic movement and a handicap for the parliamentary
democracy desired by a majority of Social Democrats. At the same
time, however, it was something quite different. It was the very
precondition of parliamentary governance, which was
inconceivable without a coalition between the moderate elements
within the bourgeoisie and the working class. It took the war, the
splitting of the party, the Russian October Revolution, and the
certainty of Germany’s military defeat to convince the MSPD that it
was necessary to take the decisive step towards a democratic
Germany. The party committee and the parliamentary fraction took
this step on 23 September by approving Ebert’s course with a clear
majority.
Ebert’s speech to his party colleagues was based on
consultations between Majority Social Democrats and Progressives
in the Interparty Committee the day before. The Centre went along
with most of the arrangements, including the agreement to general
direct suffrage for all federal states and the evacuation of all
occupied territories. But against the decisive point, the change to a
625
parliamentary monarchy, the party of political Catholicism was still
stubbornly resistant, even against the will of Erzberger. All too
deep, for many delegates, was the fear that the rule of the
parliamentary majority would cost the Catholic minority more
political influence than under the constitutional system.
It also did not help that the Social Democrats wanted a change
of chancellor. Hertling had shown himself to be as weak as
Michaelis when dealing with the Army High Command and
therefore not the right man to pursue the kind of peace agenda the
MSPD wanted. Only on 28 September, when the National Liberals
joined the call for full parliamentary rule, did the Centre yield,
promising to accept parliamentarization and the appointment of a
new chancellor endorsed by the majority.
Before the Reichstag majority could turn words into deeds,
however, the Army High Command intervened and compelled both
a change of chancellor and the shift to parliamentary government.
On 14 September Austria-Hungary, acting on its own, had offered
peace negotiations to the Allies. On 29 September Bulgaria,
another German ally, accepted the Entente’s ceasefire conditions.
That same day, Hindenburg and Ludendorff convinced Emperor
William II that Germany had definitively lost the war and should
immediately approach Wilson with an offer of ceasefire and peace
negotiations. The responsible party was not to be the OHL,
however, but a new government endorsed by the parliamentary
majority.
Ludendorff combined his initiative with a Dolchstosslegende. ‘I
have asked H.M. [‘His Majesty’ (H.A.W.)] to now bring those
circles into the government whom we have to thank in great part
626
for the present situation,’ the First Quartermaster-General
announced to a group of high officers on 1 October. ‘Accordingly,
we shall now see these gentlemen moving into the ministries. Let
them conclude the peace that must now be made. Let them lie in
the bed they made for us.’
William II accepted Ludendorff’s solution. Since Hertling was
not willing to introduce the parliamentary system himself, the
emperor accepted his resignation on 30 September and, on 3
October, appointed as his successor Prince Max of Baden, who had
already been approved by Ludendorff and the majority parties. The
new chancellor, cousin and successor of the childless Duke
Frederick II of Baden, had not distinguished himself as a supporter
of parliamentary government or of the peace resolution of July
1917. Nonetheless the majority parties endorsed him, confident in
their ability to influence his policymaking.
The Reichstag majority fractions were, in fact, much more
strongly represented in the new Reich and Prussian governments
than under Hertling. The Centre provided the new secretary of the
federal ministry of the interior, Karl Trimborn; Friedrich von Payer
from the Progressive People’s Party continued as vice-chancellor;
and the Social Democrats, too, now attained cabinet rank when
Gustav Bauer, acting president of the general commission of the
Free Trade Unions, took over leadership of newly created imperial
labour office. Philipp Scheidemann became secretary without
portfolio, a function also assumed by other leading
parliamentarians such as Adolf Gröber and Matthias Erzberger from
the Centre, as well as Conrad Haussmann, one of the ‘discoverers’
of Prince Max, from the Progressive People’s Party. Otto Fischbeck,
627
Haussmann’s party colleague and president of the Interparty
Committee, became Prussian minister of commerce. From the ranks
of the National Liberals came the new secretary of the office of the
judiciary, Paul von Krause, and Robert Friedberg, vice-president
and president incumbent of the Prussian ministry, who regularly
participated in the meetings of the federal cabinet.
In early autumn 1918, shortly before the change to the new
government and governmental system, Paul von Hintze, secretary
of the foreign office from the beginning of July to the beginning of
August, spoke to Vice-Chancellor von Payer about a ‘revolution
from above’. According to Hintze, a ‘revolution from above’ was
the only way to prevent a ‘revolution from below’. For those
involved at the time, the closest thing to a ‘revolution from above’
was what the OHL set in motion at the end of September.
Parliamentarization, at least in a de facto sense, would no doubt
have occurred in October 1918 even without the interference of the
military, since the Reichstag majority was pursuing the same
objective at the same time, albeit for different reasons.
Nonetheless, the concept of a preventive ‘revolution from above’
can also be used to describe the actions of the party leaders,
including those of the Majority Social Democrats.
The Russian example showed what a radical and determined
revolutionary minority could do and the disastrous consequences of
a coup d’état by such a group, an event well-schooled Marxists like
Kautsky regarded as a relapse into ‘Blanquism’, which was
generally thought of as historically superseded. Violent acts like the
murder of the German ambassador Mirbach by anti-Bolshevist
‘Leftist Social Revolutionaries’ on 6 July 1918 and the murder of
628
the tzar’s family by the Bolsheviks ten days later cast a glaring light
on the dynamics of terror and counter-terror. Lenin’s statement
‘Whoever resists is to be shot’ on 21 February 1918, originally
referring to men and women of the bourgeois class who refused to
join special contingents digging trenches under the supervision of
the Red Guard, had long become quotidian practice. Revolutionary
rhetoric and action in Russia created fear of revolution in Germany,
and they strengthened the resolve of the moderate forces within
the working class and bourgeoisie to pre-empt a revolution from
below through a peace settlement and a democracy.
The far right reacted to the formation of the first parliamentary
government by declaring war against democracy—and against the
Jews. On 3 October 1918 the president of the Pan-German League,
Heinrich Class, called for the establishment of a ‘great, courageous
and dashing national party and the most ruthless war against
Jewry, upon whom all of the only too justified anger of our good
and misled people must be deflected.’ In an article co-authored
with Class for the League’s official (since 1917) Deutsche Zeitung on
15 October, Baron von Gebsattel, Class’s deputy, wrote that he and
many like-minded people considered the new administration
illegitimate and would do everything to bring it down. After the
Jews had been so long active ‘on behalf of the most extreme
radicalization of Prussian suffrage’, they had now once again been
the real ‘driving force’ behind the ‘bloodless coup’ in Berlin.
Democratic ideas were ‘poison’, and they were poison ‘of Jewish
origin’. The Jewish spirit of ‘corrosion’ (Zersetzung) was ‘a real
thing, not bound by good or evil will, a veritable law of nature
that, as the legacy of a centuries-long development, is inseparably
629
bound up with Jewry’.
The Pan-German diatribes against the Jews reached a climax at
the congress of the group’s leadership and executive committee on
19 and 20 October 1918 in Berlin. In the name of his friend
Gebsattel, who was ill, Class called upon the audience ‘to use the
situation for fanfares against Jewry and the Jews themselves as
lightning rods for all injustices’. Only when all available forces and
means were exploited could ‘fear and terror’ be successfully stirred
up among the Jews. Near the end of his speech, the president
assured his listeners: ‘I will not shrink from any means and direct
myself in this respect according to the words of Heinrich von Kleist
...: Schlagt sie tot, das Weltgericht | fragt Euch nach den Gründen
nicht!’*
The formation of the cabinet of Prince Max of Baden represented
a de facto parliamentarization, not yet the legal-constitutional
introduction of parliamentary rule. The formal change from
constitutional to parliamentary monarchy took place on 28 October
1918 through constitutional amendment. From this moment
forward, the chancellor was dependent on the confidence of the
Reichstag. If the Reichstag expressed its lack of confidence in him,
he was compelled to resign. The chancellor became responsible for
all acts of political significance performed by the emperor in the
exercise of his constitutional powers. Thus the emperor’s military
command authority, to the extent it involved acts of ‘political
significance’, was also placed under the control of parliament.
Another constitutional amendment was hardly less important: no
war could be declared nor peace concluded without the consent of
the Reichstag.
630
While still preserving a monarchical form of government, the
‘October reforms’ transformed Germany into a democracy of the
Western type, comparable to Great Britain, Belgium, the
Netherlands, and the Scandinavian monarchies. The empire’s
fundamental contradiction, the antinomy between economic and
cultural modernity on the one hand and the backwardness of its
pre-parliamentary system of government on the other, was
abolished on 28 October 1918. At least on paper.40
631
and de facto head of the OHL was General Groener, a native of
Württemberg.
On 29 October, one day after he had signed the laws amending
the constitution, William II, on the advice of Hindenburg, left
Berlin for the General Headquarters in the Belgian town of Spa.
Ernst Troeltsch, a keen observer of the events of the time, regarded
this step as nothing more and nothing less than the final splitting of
the government. ‘The monarchic-military and the parliamentary-
bureaucratic authorities were utterly divided and at war.’ Indeed,
there can be no doubt about the meaning of the emperor’s
departure. ‘At the moment, it might well have been simply an
instinctive reaction—the Hohenzollern monarchy, in the hour of
danger, seeking to return to its military origins,’ writes the
historian Wolfgang Sauer. ‘But, at the same time, this decision
meant that the old powers severed their bond, formed so recently
and with such difficulty, with the people’s representatives and
made the rash attempt to restore the old military monarchy.’
The most serious challenge to the new parliamentary
government came not from the army, however, but from the navy.
The leadership of the naval forces took the cessation of submarine
warfare on 20 October as an opportunity to announce that the
measure restored its ‘operative freedom’. When the chancellor
received news of this announcement from Admiral Scheer, he was
unable to assess its implications. Nor was it intended that he do so.
The plan was for the navy, which had not seen action since the
battle of the Skaggerak (Jutland) at the end of May 1916, to write
history once again and deal Britain a severe blow. It was a question
of ‘honour’. Heavy German losses, as well as conflict with the Reich
632
leadership and the Reichstag majority sustaining it, were accepted
as the necessary price. The navy command pursued its own foreign
policy, and in a way that can be seen as a coup attempt.41
That the attempt failed was the result of resistance among the
sailors. The revolt began on 29 October in a row of ships lying in
the roads off Wilhelmshaven. The navy command took harsh
counter-measures, but only succeeded in making things worse. On
1 November Kiel became the centre of the sailors’ uprising.
Dockyard workers joined the protests on 3 November. On the
following day, at the request of Admiral Souchon, the commanding
officer and governor of Kiel, the government intervened, sending
Secretary Conrad Haussmann and the military expert of the Social
Democratic Party fraction, Gustav Noske, to the city on the Baltic.
Noske was able to pacify the rebellious sailors with a promise of
amnesty, but he did not succeed in containing the unrest. On 4
November only Kiel was in the sailors’ hands. Two days later
Lübeck, Brunsbüttel, Hamburg, Bremen, and Cuxhaven were, too.
On 7 November the mutiny turned into revolution. The
Wittelsbach throne was the first to fall. The Independent Social
Democrat Kurt Eisner, a journalist from Berlin, seized power in
Munich as the head of that city’s workers’ and soldiers’ council,
declaring Bavaria a ‘free state’ on the next day. Brunswick fell—
also on 7 November—into the hands of sailors from Kiel and
soldiers from the vicinity who had joined them. On 8 November
Cologne, too, was taken over by a workers’ and soldiers’ council.*
The mayor, Konrad Adenauer, was the first member of the city
administration to ‘accept the facts of the new situation without
633
hesitation’. On the evening of the same day, the Prussian ministry
of war listed the names of nine more cities that had gone ‘red’,
among them Halle, Leipzig, Düsseldorf, Osnabrück, and Stuttgart.
While revolutionary unrest spread throughout the country,
Berlin was the scene of a struggle over the emperor’s abdication.
The Majority Social Democrats, though republicans in principle,
had long since become ‘reasoned’ or ‘practical monarchists’
(Vernunftmonarchisten). Distinguishing between the person of the
monarch, who stood in the way of peace, and his office, they were
willing to preserve the monarchical form of government for the
time being. On 5 November Vorwärts declared its support for this
social-democratic pragmatism: ‘The prospect of having to mess
around with royalist Don Quixotes for perhaps thirty years in a
new republic, thereby setting back necessary developments, is not
among the most pleasant of notions.’
William II’s abdication was demanded not only—circuitously—
by Wilson, but also—openly—by the sailors in Kiel. The first
prominent Social Democrat to declare on the matter was Philipp
Scheidemann, who had written to the chancellor on 29 October
requesting that the emperor voluntarily renounce the throne as a
way of gaining more tolerable conditions for a ceasefire and peace.
On 6 November party leader Friedrich Ebert, in an interview with
Groener, urged William’s abdication and the assignment of the
regency to one of the imperial princes. To the chancellor he
declared (if Prince Max quotes him correctly) on 7 November that,
if the emperor did not step down, social revolution was
unavoidable. ‘But I don’t want it; in fact, I hate it like sin.’
Ebert may have expressed himself somewhat differently in
634
reality, but there was no doubt about his basic attitude. Once
revolution broke out in Germany, there was no guarantee that it
would not take a course similar to the revolution in Russia. As in
Russia, moreover, radical social upheaval in Germany would
prompt an allied intervention. Ebert saw things no differently than
Groener, who, on the evening of 8 November, succeeded in
persuading Hindenburg that the ‘plan of a march on the home
country’, as contemplated by a minority of officers, was hopeless.
‘In addition to certain civil war, there would also be the
continuance of the bloody struggle with the Entente, which would
be sure to press in from the west.’42
The ‘emperor question’ was not the only unresolved problem
burdening the parliamentary majority and cabinet in the first week
of November. On 24 October the Prussian House of Lords had
adopted at first reading the bill introducing equal suffrage in
Prussia, but it was uncertain that the House of Representatives
would do the same. Under pressure from the majority parties, the
mitigation of the war conditions proceeded apace. Karl Liebknecht,
who had been condemned to four years in prison by the high
military tribunal in July 1916 for attempted treason, was released
on 23 October, and Rosa Luxemburg was released from preventive
detention on 8 November. On the same day, the war cabinet
followed the suggestion of Secretary Haussmann to release the
sailors arrested during the 1917 navy mutiny. But the relaxation of
the freedom of assembly, decreed on 2 November by the Prussian
minister of war, Heinrich Scheüch, proved to be nothing more than
a piece of paper. On 7 November General von Linsingen, regional
commander of the Marks of Brandenburg, issued a ban on the
635
formation of workers’ and soldiers’ councils as well as on
gatherings of the USPD.
For the Social Democrats, this ban was the last straw. On the
evening of 7 November the executive committees of the party and
fraction placed before the war cabinet an ultimatum demanding
the withdrawal of the ban, extreme forbearance on the part of the
police and military, a reshuffle of the Prussian government to
accord with the Reichstag majority, the strengthening of Social
Democratic influence in the government of the Reich, and, finally,
the abdication of the emperor and the renunciation of succession
on the part of the crown prince.
On the evening of 8 November William II was still emperor of
Germany and king of Prussia. Nonetheless, the Social Democrats
announced publicly what Secretary Scheidemann had declared in
the war cabinet the evening before: that the SPD did not wish to
leave the government before the conclusion of a ceasefire
agreement (the German negotiators, led by Matthias Erzberger, had
left Berlin on 6 November but had not yet reached Allied
headquarters; the armistice was not signed until 11 November in
the forest of Compiègne). In extending its ultimatum, however, the
MSPD could point to several successes in negotiations with the
government and the majority parties. Equal suffrage on the basis of
proportional representation was to be introduced per federal law in
Prussia and the other states of the Reich; Prussia was to be
immediately parliamentarized; and Social Democratic influence in
the federal government was to be increased. In addition, the
Majority Social Democrats were able to announce that the latest
call-up of soldiers, which was stirring a great deal of public
636
controversy, had been cancelled.
A further, last-minute concession came from the bourgeois
majority parties on the evening of 8 November: the Centre and the
Progressive People’s Party agreed to the introduction of female
suffrage. Both parties now also joined in the call for the emperor to
step down, and even the National Liberals let it be understood that
they would endorse an abdication. Thus the Social Democrats
found themselves in a key strategic position, and they were able to
take advantage of it. As an anonymous delegate from the
Progressive People’s Party frankly admitted to the midday edition
of the Berliner Zeitung: ‘In these times, Germany cannot be governed
without the Social Democrats; otherwise the revolution would not
proceed along an orderly and peaceful path, but along a Bolshevist
path with all the horrors of civil war.’
The revolution reached the capital of the Reich on 9 November.
Once again, developments were accelerated by the actions of
General Linsingen, who arrested one of the leaders of the
Revolutionary Shop Stewards, Ernst Däumig, and set up security
posts in the large factories. Informed by party confidants in the
plants that the Berlin workers were now also heading for the
streets, Otto Wels, the district secretary, declared a general strike at
9 a.m. on 9 November in the name of the SPD and called the
workers to a ‘decisive battle under the old common banner’. One
hour later Scheidemann resigned from his office as secretary. In a
meeting of the Social Democratic parliamentary fraction, which
commenced at the same time, Ebert could report that negotiations
with the Independent Social Democrats and the workers had
already been conducted. The SPD wished to cooperate with the
637
workers and the soldiers in any necessary undertaking and then
‘take over the government, completely and totally, like in Munich,
but without bloodshed, if at all possible’.
The MSPD profited greatly from the fact that competing
organizations among the workers had only limited power to act at
this moment. The USPD was unwilling to make any agreement
without its party leader Hugo Haase, who was in Kiel, and the
Revolutionary Shop Stewards on the party left wing were not
planning their move until 11 November. This temporary
organizational and strategic vacuum to the left of the SPD was
accompanied by a veritable vacuum in the military planning. There
were almost no front-line troops in Berlin on 9 November, only
three rifle battalions, and of these, it was none other than the
Naumburg Rifle Battalion, considered especially loyal to the
emperor, that took the initiative in seeking contact with the MSPD.
Otto Wels seized the chance, and with resounding success. The
soldiers responded enthusiastically to his call to join the people and
the Social Democratic Party.
The news that the Naumburg Rifle Battalion had gone over to
the revolution fell like a bomb on the chancellor’s palace and the
General Headquarters. Prince Max now knew that his government
was no longer tenable. At about 11 a.m. he learned by telephone
from distant Spa that the emperor had decided to abdicate. When,
half an hour later, William’s official declaration had still not been
received, the chancellor took the initiative and announced the
intention of the emperor and king over the Wolff Telegraph
Agency. He himself, Prince Max, would remain in office until the
question of the regency was resolved, would propose to the regent
638
that Ebert be appointed to the chancellorship and that a law be
passed for immediate elections to a Constituent National Assembly,
which would then settle the German people’s future form of
government.
By 9 November, however, it was too late to rescue the
monarchy, even through a regency. At 12.35, a delegation from the
SPD led by Ebert appeared at the chancellor’s palace, where Prince
Max was meeting with the cabinet, and demanded the handover of
power. This was necessary, the leader of the MSPD explained, in
order to preserve the public order and avoid bloodshed. The
Independent Social Democrats were united behind the majority in
this question, he said, and it was possible that they would
participate in the new government. Representatives of the
bourgeoisie might also join, though the predominance of the Social
Democrats would have to be assured. When the chancellor
remarked that the question of the regency still had to be settled,
Ebert replied that it was too late. Then Prince Max proposed that
Ebert assume the chancellorship, whereupon the latter, after a
short moment of hesitation, accepted.
At first, the Social Democrats also demanded the office of
minister of war and the military command in the Marks of
Brandenburg. But when Minister Scheüch, pointing out that the
armistice negotiations were under way, proposed that he remain in
office with a Social Democratic undersecretary at his side and a
Social Democratic official at the side of the commander-in-chief of
the Marks, the leaders of the delegation accepted. They also agreed
that the secretary of the foreign office, Wilhelm Heinrich Solf,
should remain in office. Then, after a short private consultation
639
with the state secretaries, Prince Max of Baden took the
revolutionary step of handing over the power of government,
transferring to Ebert, as the latter put it in his first proclamation to
the German people, ‘the pursuance of the duties of imperial
chancellor with the approval of all the ministerial secretaries’.
Thus, for the first time, a ‘man of the people’ stood at the head
of the German Reich. Ebert was born in Heidelberg on 4 February
1871, the son of a master tailor. After a basic education at the
primary school, he became a saddle-maker’s apprentice, then
travelled throughout Germany as a journeyman saddler, joining a
trade union and the Social Democratic Party. In 1893 he became
the editor of the party newspaper in Bremen, beginning a political
career that would lead him, a gifted organizer, to the Reichstag by
1912 and to the head of the SPD (an office he shared with Hugo
Haase) after Bebel’s death the following year. Ebert was neither an
ideologue nor a man of bold vision. He was a sober pragmatist, and
one of the first within his party to realize that Germany was
headed for a catastrophe unless the moderate forces within the
labour movement and the bourgeoisie made an effort towards
compromise and mutual understanding.
Philipp Scheidemann, born in Kassel in 1865, a printer by trade,
from 1913 one of the three heads of the SPD Reichstag fraction and
from October 1917 joint chairman of the SPD, did not
fundamentally differ from Ebert in his view of the situation. On 2
October 1918, however, the gifted orator warned his party against
‘entering into a bankrupt enterprise in this moment of greatest
desperation’—that is, against joining Prince Max of Baden’s
cabinet. And on 9 November, too, he demonstrated greater
640
sensitivity to the mood of the masses than Ebert. At about two in
the afternoon, he proclaimed the German Republic from a
balustrade of the Reichstag building, two hours before Karl
Liebknecht, standing on the balcony over the portal of the royal
palace in Berlin, proclaimed the ‘free socialist republic of
Germany’. Ebert was outraged at his colleague’s announcement,
wanting the freely elected constituent assembly to decide the
question of Germany’s form of government. But the jubilation of
the crowd proved Scheidemann correct. The masses expected a
demonstrative break with the old system of rule, which had led
Germany to its current pass.
Ebert emphasized more the continuity than the break. Such
signs were important above all to the government officials, judges,
officers, and with them the German bourgeoisie. In an
announcement on 9 November, the newly ‘appointed’ chancellor
assured the government agencies and bureaucrats that the new
government had ‘taken over the leadership and duties in order to
protect the German people from civil war and famine and to carry
out its just demands for self-determination’. The government could
fulfil these tasks only ‘if all civil servants and agencies in the city
and countryside lend a helping hand’. He, Ebert, was aware that
many would find it difficult to work with the new men.
641
The new government of which Ebert spoke did not yet exist at this
time. Delegations from the MSPD and USPD were negotiating over
its formation during the afternoon of 9 November while sporadic
but intense fighting took place at the Marstall and the university.
The starting point for the talks was Ebert’s proposal that both
parties be equally represented in the cabinet, to which members of
the bourgeois parties could be admitted as special ministers.
Among the Independents, however, scruples against cooperating
with the ‘government socialists’ were still very strong. Liebknecht
attempted to commit the fraction to the demand for ‘all executive,
all legislative, all judicial power to the workers’ and soldiers’
councils’, which meant a system similar to that of the Russian
soviets and a rejection of a constituent national assembly.
The USPD initially adopted this call, but the majority party
rejected it out of hand: ‘If this demand is aiming at a dictatorship
of one part of one class, behind which the majority of the people
does not stand, then we must reject it, since it does not accord with
our democratic principles.’ The USPD’s wish to exclude the
bourgeoisie was also denied, with the justification that such a step
would considerably endanger the national food supply, if not lead
to its complete interruption. The MSPD also declined the USPD’s
desire to participate in the government only until the armistice was
concluded, insisting that cooperation between the socialist camps
was necessary at least until the constituent assembly was in place.
The USPD’s catalogue of demands bore the mark of Liebknecht and
the Revolutionary Shop Stewards. After Haase returned from Kiel
on the evening of 9 November, more moderate voices prevailed. On
the morning of the next day, the Independents agreed to accept
642
bourgeois special ministers as technical assistants, on condition
that one member from each of the Social Democratic parties
monitor the activities of each special minister. The USPD gave up
the idea of a provisional cabinet and declared that the question of
the constituent assembly should only be addressed after the
consolidation of the post-revolutionary situation. But the political
power ‘must lie in the hands of the workers’ and soldiers’ councils’,
which were ‘to be straightway convened in a general assembly
representing the whole Reich’.
The MSPD agreed to these conditions, since the leadership of the
USPD no longer fundamentally rejected a constituent assembly and
believed it would gain a majority in the planned assembly of
workers’ and soldiers’ councils. The majority party also accepted
the USPD’s appointees to the government, the moderates Hugo
Haase and Wilhelm Dittmann (who had been released from prison
in October), as well as Emil Barth as representative of the
Revolutionary Shop Stewards. The Majority Social Democrat
members of the new ‘Council of People’s Commissars’ (Rat der
Volksbeauftragten) were Ebert, Scheidemann, and Otto Landsberg, a
lawyer originally from Upper Silesia and delegate to the Reichstag
since 1912.
Before entering upon its duties, the new Council had to be
confirmed by an assembly of some 3,000 representatives of the
workers’ and soldiers’ councils from the greater Berlin area. This
gathering took place on the afternoon of 10 November in the Busch
Circus. Even though the radicals, led by Karl Liebknecht, did not
have a majority, the event was nearly cut short when Barth called
for the government to submit to supervision by an action
643
committee from the Revolutionary Shop Stewards and members of
the Spartacus League threatened Ebert. The latter considered the
situation so serious that, after leaving the assembly, he secured the
assent of the war minister, Scheüch, to protect the new government
if it proved necessary.
It did not. The soldiers’ representatives, whom Otto Wels had
committed to the MSPD line, warned the Revolutionary Shop
Stewards that they alone would form the government if the
principle of parity were not also observed in the action committee.
Thereupon the Revolutionary Shop Stewards submitted. By a great
majority, the assembly elected a fourteen-member Action
Committee of the Workers’ Council (Aktionskomitee des
Arbeiterrates), composed of seven representatives each from the
MSPD and USPD. Most members of the Action Committee of the
Soldiers (Aktionskomitee der Soldaten), also fourteen in number, had
no party affiliation. The next day representatives from both
committees formed the Executive Council of the Greater Berlin
Workers’ and Soldiers’ Council (Vollzugsrat des Arbeiter- und
Soldatenrates Grossberlin), which was to monitor the Council of
People’s Commissars until the convention of the general assembly
of workers’ and soldiers’ councils. Finally, at Müller’s motion, the
assembly confirmed the Council of People’s Commissars in the line-
up agreed upon early that afternoon. Late in the evening on 10
November, the MSPD and USPD reaffirmed their coalition
agreement in the imperial chancellery. Germany once again had a
government.44
644
THE ORIGINS OF THE REVOLUTION FROM BELOW
the greatest of all revolutions has brought down the regime of the
Emperor with everything that belonged to it, above and below. We
can call it the greatest of all revolutions, for never has such a firmly
built bastille, ringed with such strong walls, been taken in one
attack. Only one week ago there was a military and civil
administrative apparatus so ramified, so enmeshed and deeply
rooted, that its rule seemed secure for many ages. The grey
automobiles of the officers raced through the streets of Berlin. The
guards stood in the city squares like columns of power. A giant
military organization seemed to embrace everything, and an only
seemingly undefeated bureaucracy sat enthroned in the
government offices and ministries. Yesterday morning that was all
still there, at least in Berlin. Yesterday afternoon it was all gone.
645
order for the immediate requirements of life. On all faces was
written: ‘Salaries are still being paid.’
646
monarchy by force of arms was minuscule. But royalists did exist.
There were more of them among the Protestants than among the
Catholics, and nowhere were they as strong as in Prussia east of the
Elbe. Although sovereigns had everywhere stood at the head of the
church, an inner affinity to the unity of throne and altar and to the
prince as the summus episcopus was above all a characteristic of
German and especially northern and eastern German Lutheranism.
It was not coincidentally Bruno Doehring, court chaplain and
pastor of the cathedral in Berlin, who, in probably his last war
sermon on 27 October 1918, called Wilson’s demand for William
II’s abdication a ‘satanic idea’ and intoned: ‘The monarchy in
Prussia is a thousand times more than a political question for us
Evangelicals. For us, it is a question of faith.’
Speaking of Germany after the First World War, Max Weber
remarked that the ‘history of the collapse of the authority
legitimate until 1918’ revealed ‘how the war’s destruction of
traditional ways of life and the loss of prestige through defeat, in
connection with the systematic habituation to illegal behaviour,
contributed equally to undermining obedience to army and labour
discipline and thus prepared the overthrow of the regime.’ The
sociological situation can be expressed in the argument that, by
autumn 1918, the German Reich had largely lost the ‘most
common form of legitimacy today’, namely ‘belief in legality’
(Legalitätsglaube), a resource of authority Weber defined as
‘conformity to formally correct rules that have been imposed by
accepted procedure’.
The war’s erosion of traditional values, the looming military
defeat of the central powers, and the spread of ‘black markets’ in
647
the wake of economic and monetary policy failure—this was the
triad of factors that caused, in Weber’s astute analysis, the collapse
of the Wilhelmine empire. The emperor was the embodiment of the
old system. In the view of the broad masses, he was the one who
bore ultimate responsibility for the length and the catastrophic
outcome of the war as well as for the material deprivations of the
people. And since he refused to take responsibility, he had to go.
Wilson’s ‘Fourteen Points’ nourished the hope that Germany could
look forward to a just peace if it democratized its government. The
longing for peace thus promoted the desire for democracy. By
autumn 1918 these goals had the support of a broad majority. This
majority formed the core of a societal consensus on the eve of 9
November 1918 and in the weeks following. It was not an all-
inclusive consensus, to be sure, but nonetheless one that included
all classes and confessions in Germany.
A good deal of democracy had, in fact, been achieved by this
date. General direct male suffrage had been introduced in Germany
at the founding of the Reich in 1871 (and even earlier in the North
German Confederation, in 1867). On 8 November 1918 the
majority parties agreed to extend this system to all states in the
Reich, including Prussia, and to grant the right to vote and to stand
for election to women. Furthermore, parliamentary government
existed in a de facto sense after 3 October, and was made official
on 28 October. However, the high-handed behaviour of the
emperor, army, and navy command in the days after the
constitutional reform made it clear that the new parliamentary
system existed only on paper, and the inter-party agreements on 8
November came too late to alter the course of events.
648
The revolution from below broke out because the revolution
from above, the October regime-change, had failed. This failure
was the result of obstruction by the military. This obstruction,
above all that of the navy command, made it impossible to
preserve the institution of monarchy. Collapse, obstruction, and
revolution led to the proclamation of the German Republic on 9
November 1918. This date did not mark the end of the German
revolution, but only the beginning of a new period in its history.45
649
7
650
already in 1914 to get caught up in the ‘war enthusiasm’, or who
had turned against the war in the course of the conflict, could opt
in 1918 for the most radical leftist position, calling for a civil war
to end the capitalist order from which the ‘imperialist’ war had
supposedly arisen. For those who affirmed the war as a necessity of
nature and maintained to the end that a war for one’s country
could only be just, peace was little more than an interlude until the
next armed conflict between the nations. At the end of 1918, both
of these positions represented only small minorities, that of the
extreme right no less than that of radical left. The vast majority in
all countries, though not converted to principled pacifism, had had
enough of war. How long the peace would last depended above all
on how just or unjust it was perceived to be.1
In expectation of a just peace, Germany had separated itself
from its monarchical system at the beginning of November 1918
and become a republic. What a just peace meant, however, was a
matter of controversy. All Germans hoped that Germany would
remain a unified state with its borders as intact as possible, and
with no or only minor war reparations to pay. At the same time,
most were aware that Alsace-Lorraine would be restored to France
and the Polish-speaking parts of Prussia to the newly created Polish
nation. Many looked to the Anschluss of Austria to provide a certain
compensation for these losses. After the collapse of the Habsburg
monarchy, Germany’s closest ally in the war, a Provisional National
Assembly in Vienna had proclaimed German-speaking Austria a
republic and part of the German Republic. This was a solution
according to Wilson’s idea of the right to national self-
determination. In both Germany and Austria the Social Democrats,
651
who justly considered themselves the true heirs of the 1848
revolution, were the most eloquent proponents of a grossdeutsch
national state. Realistically speaking, however, there could be no
reason to expect the victors to permit such a large territorial
increase on the part of their defeated foe.
It was more realistic to think that a common antipathy might
bridge the gap between victors and vanquished—the antipathy to
Bolshevist Russia. On 5 November, when Max of Baden’s
government was still in power, Germany had broken off diplomatic
relations with Russia in protest over Russian payments to German
revolutionaries. The armistice agreement of 11 November
contained a point that fitted in with Berlin’s calculations: with the
consent of the Allies, German troops were to remain temporarily in
the Baltic region in order to prevent further incursion by the
Bolsheviks. The MSPD People’s Commissars had more than one
reason to emphasize the German–Allied consensus against
Bolshevism. In the first place, only a moderate German government
could expect a timely lifting of the blockade that still cut off food
imports into Germany. Secondly, this very expectation necessarily
strengthened the domestic political position of the moderate
elements.
And in truth, there were good domestic as well as foreign policy
reasons for Germany to disassociate itself from Bolshevism. When
Karl Liebknecht—the leader of the Spartacus League, at that time
still part of the USPD—attempted on 9 November to commit the
Independent parliamentary fraction to the Russian agenda of ‘all
power to the councils’, Eduard Bernstein, who was in attendance,
‘was struck as if by lightning: “he will bring us counter-
652
revolution.”’ A counter-revolution would have provoked Allied
intervention no less than a communist coup, and the war would
have recommenced. For the moderates, therefore, democracy, and
the Constituent Assembly that was to lay its foundations,
represented both an earnest of peace abroad and the alternative to
civil war.
Moreover, parliamentary government was a logical consequence
of German history. Nobody worked this out as clearly as Bernstein,
who, though still a member of the USPD, returned to the bosom of
the majority party in December 1918 in order to set an example for
the reunification of the Social Democrats. (He did not leave the
Independents until March 1919, when the USPD banned such
double memberships.) In his 1921 book Die deutsche Revolution, ihr
Ursprung, ihr Verlauf und ihr Werk,* Bernstein, then 71 years old,
sought to understand and explain to his contemporaries why the
overthrow of the German government was so different from—that
is, far less radical than—all great historical revolutions.
He saw two main reasons. The first was the high level of societal
development. According to Bernstein’s argument, the less
developed societies are, the better they are able to cope with
measures aimed at their radical restructuring.
But the more intricate their inner subdivisions, the more elaborate
the division of labour and the cooperation of their inner organs, the
greater is the risk of great harm to their existential possibilities
when the attempt is made to radically alter their form and content
within a short time and with violent means. Regardless of how they
account for it in theoretical terms, the authoritative leaders of the
653
Social Democratic movement have comprehended this reality and
modified their revolutionary praxis accordingly.
654
The inhabitants of these societies were far less dependent on the
services provided by the state and the communities than were the
inhabitants of industrialized nations with a complex division of
labour. Consequently, the need for administrative continuity was
far higher in developed than undeveloped societies.
The degree of democracy in Germany had a similarly inhibitive
effect on revolutionary activity. Since Germany had already known
general male suffrage for some fifty years, in 1918 it could only be
a matter of extending the vote to women and all the separate states
and introducing parliamentary rule in the government. The order
of the day was more democracy, not less. The Social Democrats,
the empire’s staunchest advocates for democratization, would have
sacrificed their political credibility if they had deviated from this
course and fallen back on the orthodox Marxist ‘dictatorship of the
proletariat’.2
In contrast to Liebknecht, moderate Independents like the
People’s Commissars Haase and Dittmann did not fundamentally
oppose the idea of a Constituent National Assembly. However, they
did not want to hold elections until April or May 1919, three or
four months later than what the Majority Social Democrats had in
mind. They intended to use the intervening period to take
precautionary measures securing democracy and socialism. The
provisional government, as Rudolf Hilferding explained in the party
organ Freiheit on 18 November 1918, must summon all its energies
655
must not serve as a hotbed for counter-revolutionary activities. But
above all, we must prove that we are not merely democrats, but
also socialists. Pushing through a series of provisional measures is
completely possible. They must be enacted so that here, too,
positions can be set up that any capitalist counter-attack will be
unable to take.
656
were forced to work together with the old elites to a certain extent.
In order to achieve rapid demobilization, they could not avoid
cooperating with the Army High Command; in order to avoid
administrative collapse, they had to leave the greater part of the
bureaucratic system intact; and they were compelled to rely on the
help of the business community in getting the economy going
again.
Still, the cooperation was greater, and the changes fewer, than
conditions demanded. It was certainly difficult to find workers for a
republican people’s militia whose task, in a crisis situation, it might
have been to confront ‘class brothers’ from the ranks of the
Spartacists. But virtually no effort was made to set up an armed
force loyal to the republic. Even modest reform proposals by
moderate workers’ councils—the abolition of the compulsory salute
when off-duty, the closing of officers’ casinos, and the same
provisions for officers and ranks, for example—were not
undertaken. When Groener and Hindenburg’s protested against
these suggestions in December 1918, Ebert gave in almost
immediately.
In the civil administration, too, nearly everything remained as it
had been. Arch-conservative bureaucrats, making no secret of their
contempt for the republic (they were especially common in
agricultural East Elbia), remained in office despite the complaints
of local workers’ councils. Even when conservative provincial
officials themselves wished to resign, Wolfgang Heine, Prussia’s
Majority Social Democratic interior minister, exhorted them to
remain at their posts for the sake of public order. Eight months
passed after the fall of the monarchy before the self-administrative
657
bodies of the three-class suffrage system were replaced by new,
democratically elected bodies.
Neither the Majority Social Democrats nor the Free Trade
Unions had any intention of interfering with the existing system of
private property. Workers’ and employers’ organizations, headed
by Carl Liegen, president of the general commission of the Free
Trade Unions, and Hugo Stinnes, a heavy industrialist, agreed on
15 November 1918 to form the Central Cooperative Union of the
Industrial and Commercial Employers’ and Employees’ Federations
of Germany (Zentralarbeitsgemeinschaft der industriellen und
gewerblichen Arbeitgeber- und Arbeitnehmerverbände Deutschlands).
The ‘Stinnes–Liegen agreement’ recognized the unions as wage-
partners and introduced the eight-hour day (which, however, was
to remain in place only if other civilized countries followed the
German example). This agreement meant the de facto
abandonment of nationalization, at least for the time being. This
was true even in the coal-mining industry, where the socialization
commission put in place by the People’s Commissars called for
nationalization in February 1919.
At no time in 1918–19 was the social dominance of the of East
Elbian manorial lords seriously threatened—an elite that had
resisted all efforts at democratic and parliamentary reform during
the imperial era, perhaps even more tenaciously than the heavy
industrialists. No movement arose among the agricultural workers
and poor farmers to demand the expropriation of the Junkers. In
fact, the Council of People’s Commissars agreed to the formation of
common councils for landowners, medium and small farmers, and
agricultural labourers, which all but guaranteed the existing system
658
of property relations. In January 1919 agricultural labourers were
granted the right to engage in union activity and collective
wagebargaining. Nonetheless the manorial districts, one of the
administrative foundations of Junker rule, remained in place until
1927. If the revolution cost the Junkers their access to state power,
they preserved enough of their societal influence to work
effectively towards the restoration of their political power.
The obstacles to land reform were very great in 1918–19, and
probably insurmountable. The conversion of large estates into
individual farms, as the agrarian reformers around Eduard David
had long demanded, would have put the food supply at risk. For a
‘state capitalist’ solution, that is, the appropriation of the large
estates by the governments of Prussia, Mecklenburg-Schwerin, and
Mecklenburg-Strelitz and their cultivation by qualified tenant
farmers, virtually no preliminary work had been done, and there
was little awareness of the urgency of the problem. The proposal of
the socialization commission under Karl Kautsky in the middle of
February 1919 was no simple nationalization (which would
immediately have raised the objection that the Allies, invoking the
armistice agreement, could have used state property as a
productive pledge to compel reparations). Rather, the experts
proposed the creation of an independent economic body called the
‘German Coal Community’ (Deutsche Kohlengemeinschaft), in whose
controlling board the management, workforces, federal
government, and purchasers of coal were to have equal
representation. However, the Social Democrats and Free Trade
Unions believed that the time for a change in property relations
had not yet come. Instead they followed the motto:
659
‘Reconstruction, and then, where it makes sense, nationalization’.
The result was that the coal mines, the bastion of those
representing the ‘lord in the manor’ point of view, remained an
anti-democratic bulwark after 1918.
The ruling Social Democrats justified their non-interference in
the state bureaucracy by claiming they did not have personnel
qualified to replace the current officials. This was true, but it is not
sufficient to justify the ‘hyper-continuity’ that characterizes the
German revolution of 1918–19. In taking it for granted that a state
agency was to be headed by ‘unpolitical’ officials, not politicians,
the Social Democrats showed that they had basically adopted the
logic of the old imperial bureaucracy. And wherever expertise was
really required (usually of a juristic nature), the new governments
might have had recourse to a reservoir of competence among the
liberal forces—a reservoir greater than the Social Democrats were
willing to admit. To a certain extent the same was true of the
military. Younger officers, prepared for reform, were certainly
available, for example in the circles surrounding the
Republikanischer Führerbund. But they were approached neither by
the Council of People’s Commissars nor by the coalition
governments that succeeded it.
There is no need to speculate about the deeper reason why the
Social Democratic People’s Commissars concentrated on the most
immediate tasks—preservation of the empire’s unity, public order,
restoration of the economy—and left everything else to the
Constituent Assembly: they did not believe they had the democratic
mandate to undertake drastic political and social change. ‘We were
literally the bankruptcy trustees of the old regime’, as Friedrich
660
Ebert put it in his report to the national assembly on 6 February
1919, to the lively applause of the Social Democrats.
All barns, all stockrooms were empty, all provisions were fast
becoming exhausted, credit was shaken, morale had sunk low.
Supported and aided by the Central Council of the Workers’ and
Soldiers’ Councils, we devoted all our energies to fighting the
dangers and afflictions of the interim period. We did not pre-empt
the National Assembly. But where time was of the essence and the
situation required it, we strove to accomplish the most pressing
demands of the workers. We did everything we could to get the
economy going again… If our success was not commensurate with
our wishes, the circumstances that prevented it must be justly
recognized.
661
the mouthpiece of the Spartacus League: ‘The “civil war”, which
some have anxiously tried to banish from the revolution, cannot be
banished. For civil war is but another name for class struggle, and
the notion of introducing socialism without class struggle, by
decision of a parliamentary majority, is a ridiculous petty
bourgeois illusion.’3
662
19 December the delegates took the step that Ernst Däumig,
spokesman of the Revolutionary Shop Stewards, had beforehand
called the ‘death sentence’ of the revolution: by 344 against 98
votes, they rejected the proposal to base the new republican
constitution on the council system, endorsing instead—by about
400 against 50 votes—MSPD delegate Max Cohen-Reuss’s motion
to hold elections to the national assembly on 19 January 1919.
This was an even earlier date than the one (16 February) the
Council of People’s Commissars had agreed to on 29 November.
In his arguments, which contributed strongly to the
unambiguous outcome of the vote, Cohen-Reuss used the Russian
revolution as an example of the domestic and international
consequences of rule by council in Germany. In truth, the devotees
of a ‘pure council system’ were labouring under an illusion. It was
nothing but wishful thinking to imagine that the masses of a
complex industrial society could be mobilized to keep permanent
watch over their political representatives. Since only a privileged
minority enjoyed the requisite free time, the transformation of
council into dictatorial rule was foreordained. The great majority of
German workers did not want a dictatorship, not even a
‘dictatorship of the proletariat’, which they foresaw would quickly
become a dictatorship over the proletariat. They wanted to expand
democracy, not abolish it. The decision of the council congress to
set elections for 19 January 1919 made it clear that there was a
broad, class-transcending consensus for a constituent assembly in
Germany.
On two other questions, however, the ‘leftist’ tendencies of the
congress emerged clearly. In the first place, a large majority of
663
delegates called upon the Council of People’s Commissars to
immediately commence the nationalization of qualified branches of
industry, especially mining. Secondly, they unanimously passed the
so-called ‘Hamburg points’ on military policy. According to this
resolution the Council of People’s Commissars, under the auspices
of a yet-to-be-elected Central Council (Zentralrat) of the workers’
and soldiers’ councils, was to function as commander-in-chief of
the armed forces. All insignia of rank were eliminated ‘as a symbol
of the demolition of militarism and the abolition of blind
obedience’, and the wearing of arms while off-duty was banned.
The soldiers elected their own leaders, and the enforcement of
discipline became the responsibility of the soldiers’ councils.
Finally, the standing army was to be disbanded and a people’s
militia set up as expeditiously as possible.
The ‘Hamburg points’ were a reaction to negligence on the part
of the government. If, in the weeks before, the People’s Commissars
had accommodated the moderate reform proposals of the soldiers’
councils, the programme of the Hamburg soldiers’ councils, parts of
which were utopian, would hardly have been adopted. The
delegates would probably also have responded to reasonable
objections on Ebert’s part if the latter had pushed for more realism.
But the leader of the Council of People’s Commissars took action
only later, under the pressure of an ultimatum from the Army High
Command. The ‘executive resolutions’ decreed by the People’s
Commissars on 19 January 1919 already showed the influence of
the military, especially in matters of command authority. The law
concerning the institution of a provisional army (Reichswehr),
passed by the national assembly on 6 March 1919, no longer
664
contained any trace of the ‘Hamburg points’.
The last serious conflict fought out in the council congress was
over the distribution of authority between the Council of People’s
Commissars and the Central Council. The MSPD proposed that,
pending a final decision by the national assembly, legislative and
executive authority be conferred on the Council of People’s
Commissars, the task of parliamentary supervision on the Central
Council. Questioned by his party colleagues, Haase defined
‘parliamentary supervision’ to the effect that all laws were to be
placed before the Central Council, the more important of them
requiring consultation with that body. But the USPD delegates
wanted more, namely the ‘full right’ of the Central Council to
accept or reject bills before they became law. The MSPD, believing
that such an arrangement would be fatal to the political
effectiveness of the Council of People’s Commissars, responded
with an ultimatum: if the Independents’ motion were accepted, the
Majority People’s Commissars, secretaries of state, and Prussian
ministers would resign. After the congress had adopted Haase’s
concept of ‘parliamentary supervision’, the USPD’s extreme left
wing pushed through a boycott of the elections to the Central
Council. As a result, only Majority Social Democrats were elected
to the twenty-seven member Central Council of the German
Socialist Republic.
Thus the People’s Commissars of the USPD lost their voice in the
government. The formal dissolution of the coalition of 10
November was ushered in by the ‘Berlin Christmas battles’, the
dramatic climax to a two-week conflict over the payment of the
‘people’s naval division’, which occupied the royal palace. On 23
665
December the rebellious sailors detained the government and
‘arrested’ Otto Wels, the commander of Berlin, in the Marstall. The
ensuing bloody fighting in the vicinity of the palace and royal
stables ended in a military defeat for the regular soldiers and a
political defeat for the government. The USPD People’s Commissars
rightly complained that their MSPD colleagues, calling upon the
Prussian minister of war for help, had given him carte blanche (and
thus put the life of Otto Wels at risk). When the Central Council
nonetheless approved the actions of Ebert and his colleagues on 28
December, Haase, Dittmann, and Barth decided to resign from the
Council of People’s Commissars.
Two days after the collapse of the coalition, the founding
congress of the Communist Party of Germany (Kommunistische
Partei Deutschlands; KPD) commenced in the Prussian House of
Representatives in Berlin. The new party was a union of two
political camps: the Spartacus League, hitherto the far left of the
USPD, and the International Communists of Germany, formed by
leftist radicals in Hamburg and Bremen. The mood in the chamber
was radical in the extreme. Rosa Luxemburg sought in vain to
convince the delegates that it was senseless and dangerous to
endorse the motion of Otto Rühle from Pirna committing the party
to a boycott of the elections to the national assembly. Endorse it
they did, by 62 against 23 votes. The anti-parliamentary thrust of
the measure was obvious. Arthur Rosenberg rightly called it ‘an
indirect call to coup-like ventures’.
On 4 January 1919, three days after the KPD’s founding
congress came to an end, Paul Hirsch, Prussia’s Majority Social
Democratic prime minister, dismissed the president of the Berlin
666
police, Emil Eichhorn, who belonged to the left wing of the USPD.
Since Eichhorn’s security force had joined the rebellious ‘people’s
naval division’ during the Christmas battles, his removal was
unavoidable. No government could entrust the capital’s police force
to a man who himself worked for that government’s overthrow.
The radical left saw things differently, and interpreted Eichhorn’s
dismissal as an intentional provocation. On the evening of the 4
January, leaders of the Berlin USPD and the Revolutionary Shop
Stewards decided to hold demonstrations the next day in protest
against the government’s move. The manifesto was also signed by
the KPD leadership.
The size and spirit of the masses that gathered on 5 January far
exceeded the organizers’ expectations. But events got out of control
on the first day. While the Berlin USPD, the KPD, and the
Revolutionary Shop Stewards were discussing strategy in police
headquarters, armed demonstrators occupied the press centres of
the Social Democratic Vorwärts and the left-liberal Berliner Tageblatt
as well as the publishing houses of Mosse, Ullstein, and Scherl, the
Büxenstein printing office, and the Wolff Telegraph Agency. At the
same time, police headquarters received reports that all the Berlin
troops and even garrisons from other parts of Germany like
Frankfurt an der Oder stood behind the Revolutionary Shop
Stewards and were prepared to overthrow the Ebert–Scheidemann
government by force.
These reports—all false, as soon became evident—sent Karl
Liebknecht into a revolutionary transport. He proclaimed that
‘given this state of affairs, not only must the strike against Eichhorn
be deflected, but the overthrow of the Ebert–Scheidemann
667
government is also possible, and absolutely necessary.’ Against the
protests of Richard Müller and Ernst Däumig, a majority of the
demonstrators called for the continued occupation of the
newspaper buildings, a general strike of the Berlin workers, and
battle against the government until it collapsed.
The January uprising—often called by the dubious name of
‘Spartacus revolt’—was leaderless from the beginning. But it was
not without goal. The cry to ‘Overthrow the Ebert–Scheidemann
government!’ had no lesser aim than to block the elections to the
Constituent National Assembly and set up the dictatorship of the
proletariat. What the Russian Bolsheviks had accomplished in
January 1918 by forcibly dissolving the Constituent Assembly,
their German followers and sympathizers hoped to achieve before
such a body was even elected. Consequently, the Council of
People’s Commissars had no choice but to accept the challenge of
the radical minority of the Berlin proletariat and meet the attack
on democracy with force.
This task fell to Gustav Noske, who had joined the Council of
People’s Commissars only on 29 December, one day after the
Independents had abandoned the coalition. Noske, a woodworker
by trade, later editor of a party newspaper and, as Reichstag
delegate, the naval expert of the SPD, initially had at his disposal
only a few Berlin reserve battalions, parts of the Republican
Soldiers’ Militia, and the Charlottenburg security forces, as well as
the newly created ‘auxiliary service’ (Helferdienst) of the Social
Democratic Party. These predominantly social democratic groups
were joined by right-leaning ‘free corps’ (Freikorps), formed in
response to an appeal from the government on 7 January
668
(‘Volunteers, come forward!’), and, after 8 January, by volunteer
troops from the Army High Command.
At the beginning, armed conflict did not seem inevitable. On 6
January the government, urged by the USPD leadership, initiated
talks with the insurgents. The MSPD demanded the immediate
evacuation of the occupied newspaper buildings. Kautsky,
representing the moderate forces within the USPD, proposed a
compromise: the negotiations would be considered to have failed if
they did not succeed in restoring full freedom of the press. It is not
very probable that the rebels would have accepted this solution;
their counter-proposal, the reinstatement of Eichhorn, was
impossible. But the attempt was not even made, since the MSPD,
and on 7 January the Central Council, spoke against it. The die was
cast for violence.
There could be no doubt about the outcome of the struggle. On
11 January, after a bombardment lasting several hours, the
occupiers of the Vorwärts building surrendered. Government troops
stormed the other occupied buildings on the same day, and, on
Noske’s orders, the free corps organized by the OHL and
commanded by General von Lüttwitz began to march on Berlin.
With the uprising over by 12 January, there was no compelling
military reason for them to enter the capital. But Noske and the
OHL, thinking to prevent further coup attempts by setting a
warning example, let them in.
Among their first victims were Karl Liebknecht and Rosa
Luxemburg, murdered by officers on 15 January. The next day’s
newspapers claimed that Liebknecht was shot while trying to flee
and that Luxemburg was killed by the crowd. In May 1919 a
669
military tribunal acquitted several of the officers who had
participated directly in the murders; two other accomplices were
given light prison sentences. Those who had actually ordered the
killings, the ‘murderer central’ of Waldemar Pabst at the Eden
hotel, remained untouched.
The murdered Communist leaders were partly responsible for
the blood spilled in the January fighting. This was especially true
of Karl Liebknecht, whose call for the overthrow of the government
flew in the face of all reason. Rosa Luxemburg praised the
revolutionary masses, although their actions were diametrically
opposed to the socialist theoretician’s better insight. The January
uprising in Berlin was an attempted coup d’état by a revolutionary
minority. Had it not been defeated, civil war would have spread
throughout Germany and provoked an Allied intervention.
However, there was no justification for the excesses of violence,
which transformed the division between the moderate and radical
forces in the workers’ movement into a yawning abyss. The
Majority Social Democrats availed themselves of the free corps—
many of whom were no less ready than the Communists for civil
war—more than was necessary. The young officers and students
who dominated these volunteer forces had no interest in saving the
republic. What drove them on was hatred of everything on the
‘left’. It made sense to them to continue on the domestic front the
war they had been waging against the external enemy. It was the
Marxist left, after all, whom they considered responsible for
Germany’s defeat.4
By 12 January what the far left sought to prevent with their
670
uprising could no longer be stopped: elections to the Constituent
German National Assembly. For the bourgeoisie, there could be no
more pressing goal than the prevention of a socialist majority. But
this did not yet mean that the bourgeois parties had to work to
overcome their fragmentation. The newly introduced proportional
suffrage gave them the chance to be represented in the national
assembly in a strength corresponding closely to their share of the
vote. There was little motivation to pool their resources.
Of the two centre parties that had worked closely with the
MSPD since 1917, one split immediately following the collapse of
the imperial government. On 12 November 1918 the Bavarian
Centre Party became the Bavarian People’s Party (BVP; Bayerische
Volkspartei). The founding of the BVP represented an attempt to
thwart the shift to the centre that was to be feared from a coalition
between Social Democrats, Centre, and left-liberals, and above all
from Matthias Erzberger’s agenda. For a few weeks after 9
November, the Centre Party itself considered becoming an
interconfessional Christian popular party open to Protestants. But
then Adolph Hoffmann, a sharply anti-clerical Independent Social
Democrat and Prussian minister of education, came unwittingly to
the aid of political Catholicism’s conservative faction. Hoffmann’s
radically anti-ecclesiastical educational reforms had an effect
opposite to the one intended, making for good propaganda for the
traditional Centre.
In its appeal of 30 December 1918, the German Centre Party did
emphasize that, as a Christian people’s party, its membership was
not confessionally restricted. Rather, it formed a common base for
the political aspirations of all citizens of Christian faith. In terms of
671
its personnel and platform, however, the party was the same as it
had been under the imperial government. It addressed the question
of the desirable form of government in language so reserved as to
read like a veritable declaration of neutrality: ‘Toppled in a violent
overthrow, the old order of Germany has been destroyed. Some of
the old authorities have been eradicated, others have been
paralysed’, reads the 30 December appeal. ‘A new order is to be
created on the basis of the present realities. After the collapse of
the monarchy, this order must not take the form of a socialist
republic, but must become a democratic republic.’
The Social Democrats’ earlier liberal partner, the Progressive
People’s Party, was absorbed into the German Democratic Party
(Deutsche Demokratische Partei; DDP). The call to found the new
organization was published on 16 November 1918 by a circle
around the editor-in-chief of the Berliner Tageblatt, Theodor Wolff,
and the sociologist Alfred Weber, brother of the more famous Max.
The architects of the DDP clearly proclaimed their commitment to
the republican form of government and to the renewal of society.
They even recommended ‘taking up the idea of nationalization for
monopolistically developed branches of the economy’, and declared
war on all forms of terror, ‘Bolshevist’ as well as ‘reactionary’.
Several National Liberals also joined the new party, among them
Eugen Schiffer, but not including Gustav Stresemann, who had
taken over leadership of the National Liberal Reichstag fraction
after the death of Ernst Bassermann in 1917. Since Stresemann had
advocated large-scale German annexations during the war, the
Wolff circle regarded him as unacceptable. For his part, Stresemann
mistrusted the ‘leftist’ proclivities of the intellectuals around Wolff
672
and the Berliner Tageblatt.
On 15 December 1918 Stresemann, together with other National
Liberals, founded the German People’s Party (Deutsche Volkspartei;
DVP). The DVP differed from the DDP in its stronger nationalist
emphasis, the sharp delimitation of its platform vis-À-vis the Social
Democrats, and its promotion of farmers’ interests. With regard to
nationalization, the party’s 15 December election announcement
stated that it was prepared ‘to endorse… the transfer to the public
authority of appropriate branches of industry, provided that a
higher yield for society in general and better living conditions for
the workers thereby be created’. On the question of form of
government, the DVP kept all options open, but considered
imperial monarchy, provided it could be re-established by popular
will and through legal means, ‘the most appropriate form of
government for our people, according to our history and character’.
At the same time, however, the party announced its willingness to
participate ‘in the current form of government, within the
framework of its political principles’.
The two liberal parties had similar social agendas. They
appealed primarily to the educated strata, independent
businessmen, craftsmen, merchants, civil servants, and salaried
employees. The right-liberal DVP gained strong and financially
powerful backing from heavy industry, while the DDP enjoyed the
support of leading entrepreneurs in the electrical industry and
commerce, as well as of several banks. The Democratic Party was
the party of the liberal Jewish bourgeoisie and had the support of
several large Berlin and national newspapers like the Berliner
Tageblatt, the Vossische Zeitung, and the Frankfurter Zeitung.
673
What liberalism failed to do at the end of 1918, conservatism
accomplished right away: the unification of its forces. On 24
November 1918 the German National People’s Party
(Deutschnationale Volkspartei; DNVP) was founded, absorbing the
German Conservative Party and the Free Conservative German
Empire Party, as well as two anti-Semitic organizations, the
Christian Social Party (Christlich-soziale Partei) and the German
Ethnic Party (Deutschvölkische Partei). The DNVP avoided an open
endorsement of monarchy at first; its inaugural platform demanded
simply the ‘return from the dictatorship of one class of the
population to the parliamentary form of government, the only one
possible after the recent events’.
Then, in its electoral announcement of 22 December, the
German Nationalists proclaimed themselves convinced ‘that even
under the new democratic government, a monarchical head,
standing above the parties as a factor of stability in political life’,
best accorded with the particular nature of the German people,
such as it had developed historically. Nonetheless, the party was
prepared to cooperate ‘in any form of government created by the
National Assembly’. The DNVP had its strongest backing in the
Evangelical areas of old Prussia. It was the party of the large East
Elbian landowners and the far right wing of heavy industry. It also
found supporters among monarchist academics, above all pastors
and higher government officials, farmers, small merchants, and
‘nationalist’ white-collar and blue-collar workers.5
In the elections to the Constituent National Assembly on 19
January 1919, the bourgeois parties achieved their most important
common goal: the two socialist parties failed to gain a majority of
674
votes and seats. With 37.9%, the MSPD received 3.1% more of the
vote than the undivided party had attained in 1912 in the last
elections under the imperial government. The USPD achieved
7.6%. The majority party owed its electoral gains not least to
agricultural workers in rural east Elbia; its increase in seats was the
result of the proportional suffrage system and a far more just
division of electoral districts than under the Reich.
The most successful bourgeois party was the DDP, which
received 18.5%. This was 6.2% more than its predecessor, the
Progressive People’s Party, had managed in 1912. The party of left-
liberalism profited from the fact that bourgeois voters looked upon
it as the future coalition partner of the Social Democrats, whom the
DDP would, when necessary, put on the path of economic reason.
The two parties of political Catholicism, the Centre and the
Bavarian People’s Party, also made electoral gains. Together they
managed 19.7%, 3.3% more than the Centre had reached in 1912.
According to widespread belief, the Centre’s gains were primarily
the result of Adolph Hoffmann’s Kulturkampf.
The German Nationalists, the party of monarchic restoration,
was among the losers. With 10.3% of the vote, it achieved
considerably less than its predecessors, which together had attained
15.1%. The greatest losses, however, were suffered by the right-
liberals. The DVP received 4.4%, compared with the National
Liberals’ 13.6% seven years before. To be sure, Stresemann’s party
was just getting organized in January 1919; it did not even contest
the ballot in fifteen of the thirty-seven electoral districts.
Of the eligible population 83% voted in the elections, down
somewhat from the 84.9% of 1912. At the same time, however, the
675
introduction of female suffrage and the decrease in the minimum
voting age from 25 to 20 greatly swelled the ranks of eligible
voters, adding nearly 20 million, an increase of 136%. The Social
Democrats, resolute champions of women’s right to vote, did not
benefit from their doing so. In the districts where men and women
voted separately, many more men voted SPD than women. In
Cologne, for example, the Social Democrats received 46% of the
male vote, but only 32.2% of the female vote. The beneficiaries of
women’s suffrage were, along with the DDP, the explicitly church-
friendly parties: the Centre and BVP among the Catholic
constituency, the German Nationalists among Evangelicals.
There could be no doubt about the will of the majority. Most
Germans wanted social reforms within the framework of
parliamentary democracy, not political revolution and a radical
restructuring of society. Revolution was to be guided as quickly as
possible into the channel of peaceful evolution. It was not only
bourgeois and farmers who thought so, but also most industrial
workers.
The parallels to the last German revolution, that of 1848–9,
were obvious. Both were to a certain extent ‘unintentional’; the
Social Democrats of 1918 were as much accidental revolutionaries
as the liberals of 1848. When, at the end of February 1848, the
liberals placed themselves at the head of the revolutionary
movement, they did not suddenly become revolutionaries. They
only wanted to make sure that they, and not the radicals, would be
the ones to determine the course of events. Seven decades later, the
Social Democrats were guided by the same thought, and, being
democrats, they had to act accordingly. In both cases, a minority
676
tried to prevent the election of a constituent national assembly,
justifying its actions by claiming that the moderates had betrayed
the revolution. Hecker’s putsch of 1848 returned, after a fashion, in
the 1919 January uprising in Berlin. Lenin’s world-revolution was
the latter-day echo of Marx’s call to revolutionary world war.
The geographic directions of revolution and counter-revolution
had, of course, changed. In 1848 the wave of revolution that spread
across Europe began with the February Revolution in Paris. At the
end of the First World War, it was the Russian October Revolution
that provided the impetus. Just as a continuance of the 1848
revolution in Germany would have led to war with autocratic
Russia, the Russian solution in post-1918 Germany would have
provoked military intervention by the Western democracies. In
neither case did things go so far. But the mere thought that they
could was an important political factor in both revolutions, evoking
fear and serving as a means to stir fear up—fear of the red
revolution, of chaos and civil war.6
677
with that of the BVP. On 8 February the Centre opted to enter into
a coalition with the SPD and DDP.
One particular solution, conceivable in theory, never came up
for debate—a minority government of the two Social Democratic
parties. Writing in 1921, Eduard Bernstein explained why there
was no alternative to the coalition between the Social Democrats
and the bourgeois centre. ‘The Republic’, he wrote in his book on
the German revolution,
could fight against particular bourgeois parties and classes, but not
against all of them at once without putting itself in an untenable
situation. It would not be able to bear the great burden that had
fallen to its lot unless it managed to interest considerable portions
of the bourgeoisie in its preservation and successful development.
Even if the Social Democrats had gained a numerical majority in
the elections to the National Assembly, their enlistment of the
bourgeois-republican parties in government would have been
imperative for the survival of the Republic. But it was also an
existential necessity for Germany as a nation.
678
Friedrich Ebert told the Council of People’s Commissars on 14
January: ‘If we link the spirit of Weimar with the building of the
new Germany, it will be received with pleasure the world over.’
Philipp Scheidemann agreed with him: ‘The city of Goethe is a
good symbol for the young German Republic.’
On 10 February the National Assembly passed the provisional
government law—that is, the provisional constitution—drafted by
the secretary of the interior, Hugo Preuss, a liberal constitutional
scholar from Berlin. The next day, the delegates elected Ebert as
provisional president, who then assigned Scheidemann the task of
forming the government. On 13 February the prime minister’s
coalition cabinet, consisting of ministers from the SPD, the Centre
Party, and the DDP, commenced its work.
The new government’s greatest domestic challenge was the
wave of strikes that shook Germany in the first months of 1919. It
began in the Ruhr area at the end of December 1918, spreading
into the middle of the country by February. Its goal was the
nationalization of the mining industry, although opinions were
widely divided as to how this could best be accomplished.
Syndicalist workers wanted to take the industry into their own
hands; the socialist left wanted ‘only’ to set up works committees to
deprive the management of power, at least for the time being. The
strike movement in central Germany ended on 8 March after
Scheidemann’s government promised legislation introducing works
committees and the nationalization of the coal and potash industry.
In the Ruhr area, the miners’ strike soon widened into a general
strike, to which the government responded by sending troops. The
most serious fighting took place in Berlin in March. On 9 March
679
Noske, now minister of defence, issued an order without any legal
backing: ‘Any armed person met fighting against government
troops is to be immediately shot.’ Some thousand people died in
the March uprising in Berlin.
The great strikes in spring 1919 formed the second wave of the
German revolution. The radical part of the proletariat tried through
industrial action to force the societal changes the revolution’s first
phase—between the fall of the monarchy and the national elections
—had failed to bring. The immediate results of the workers’ efforts
were modest, compared to the expectations. The ‘nationalization’
that the government had promised their delegations, and that took
shape in several laws in March and April 1919, made no changes in
the property relations. In the coal and potash mining industry,
syndicates with compulsory membership were set up whose
supervisory boards, the Imperial Coal Council (Reichskohlenrat) and
Imperial Potash Council (Reichskalirat), contained representatives
from the Reich leadership, the states, mining entrepreneurs,
processing industries, commerce, and workers. This kind of
‘cooperative economy’ did not entail a restriction of the
entrepreneurs’ power.
Of far greater importance to the workers were the works
committees (Betriebsräte), introduced by a bitterly controversial law
of 4 February 1920. Although the radical left was highly
dissatisfied with the measure, it became the Magna Carta of
employer–worker codetermination. The law on works committees
made Germany into a pioneer of economic democracy.
The two Munich ‘council republics’ (Räterepubliken)* also
680
belonged to the second phase of the revolution. The prehistory of
the first began with the murder on 21 February 1919 of the
Bavarian prime minister Kurt Eisner by Count Anton Arco-Valley, a
law student and lieutenant on leave. Eisner, the head of the
Bavarian USPD, was on his way to the Bavarian parliament, where
he was intending to announce his resignation as head of the
government, since his party had suffered a crushing defeat in the
Bavarian elections on 12 January. On the day following the
murder, a general Munich council assembly elected a Central
Council of the Bavarian Republic, composed of delegates from the
MSPD, USPD, KPD, and farmers’ councils, and appointed Ernst
Niekisch, a teacher from Augsburg and leftist Social Democrat, to
lead it.
On 3 April the Augsburg councils, inspired by the proclamation
of a Hungarian Council Republic by the Communist Béla Kun, came
out in support of a Bavarian Council Republic. In the night of 6–7
April the central council in Munich decided to take up this demand.
Its announcement, signed by Niekisch, declared the Bavarian
parliament, ‘this sterile assemblage from the outworn age of
bourgeois capitalism’, dissolved and the government of the
Majority Social Democrat Johannes Hoffmann dismissed.
Within a very few days, the Munich Council Republic managed
to make itself into an object of general derision. The breaking off of
‘diplomatic relations’ with the Reich, a telegraph message to Lenin
reporting the unification of the upper Bavarian proletariat, and the
announcement of ‘free money’ towards the abolition of capitalism
—these were the highlights of the short-lived regime of the
Schwabing literati. On 13 April, Palm Sunday, the Republican
681
Soldiers’ Militia, cooperating with the Hoffmann government in
Bamberg, marched against Munich and engaged the leftist
revolutionaries in heavy fighting. The victors were the ‘Red Army’
and the Communists, who had initially declined to join what they
called the ‘sham council republic’. On the evening of 13 April the
Munich KPD, led by Eugen Leviné, who came from Russia, took
over the leadership of what then became the Second Council
Republic. Leviné was acting on his own, without orders from the
Berlin party headquarters. After he commenced fighting, however,
he received approval and support directly from Lenin. The latter
even enquired on 27 April whether the Munich revolutionaries had
assumed control of all the banks and taken hostages from among
the bourgeoisie.
The attempt to impose on a predominantly agrarian, Catholic,
and conservative country the dictatorship of a revolutionary clique
—which had little backing even among the inhabitants of the
capital—was foolish and doomed from the start. The red terror of
the Communists was followed in the first days of May by the white
terror of the Württemberg free corps, which came to the aid of the
legitimate Bavarian government on Noske’s orders and dealt out
bloody punishment to the ‘Spartacists’, both real and imagined. By
3 May the Second Council Republic had been crushed, with a total
of 606 killed in the fighting, including 38 government troops and
335 civilians. Eugen Leviné was accused of high treason,
condemned to death, and executed on 5 June 1919. Ernst Niekisch
got off with a two-year prison sentence.
The Munich council republics were one of the main reasons the
Bavarian capital became a bulwark of the extreme right soon after
682
the spring of 1919. Anti-Semitism, which was already strong, was
made even stronger by the prominence of Jews in the Bavarian
revolution: Kurt Eisner was a Prussian Jew; Eugen Leviné and Max
Levien, another of the Communist council leaders, were eastern
Jews; and numerous intellectual leaders of both council republics—
including the writers Ernst Toller, Erich Mühsam, and Gustav
Landauer, who was killed by free corps soldiers—came from Jewish
families. For the most skilled of the anti-Semitic activists, Adolf
Hitler, who began his political career in the summer of 1919 as an
agent of the Reichswehr district command, the conditions in post-
revolutionary Munich were highly opportune. Nowhere else would
his propaganda have found such a strong echo.7
683
and the greater part of West Prussia, thereby cutting off East
Prussia from the rest of the Reich. Danzig became a Free City under
a commissioner to be appointed by the League of Nations when
that assembly was established. The Memel district was to be
administered by the Entente. In two further areas, East Prussian
Masuria and the West Prussian territories east of the Vistula around
Marienburg and Marienwerder, the population was to decide
whether they wished to remain in Germany or join Poland. The
right to self-determination was also accorded to North Schleswig
with its part Danish-speaking, part German-speaking population.
In the west, the Eupen-Malmedy district fell to Belgium. The
Saar district was not ceded to France, as Paris wished, but placed
under the direction of the League of Nations for a period of fifteen
years, after which the population was to be permitted to exercise
its right to self-determination. In the Rhineland, too, French
ambitions were downsized. The peace terms did not strip Germany
of its territory left of the Rhine, but divided it into zones, which
were to be occupied by the Allies for five, ten, and fifteen years,
respectively. In addition, left-Rhenish Germany was to be
permanently ‘demilitarized’. Germany had to recognize Austria’s
independence as permanent, a status that could only be changed
with the assent of the League of Nations. An article forbidding an
Anschluss was written into the Peace Treaty of St Germain, which
the Allies concluded with Austria on 10 September 1919. This
treaty also forbade Austria to send representatives in an advisory
capacity to the Reich Council (Reichsrat) until the future ‘union
with the German Reich’, as Article 61 of the Weimar constitution
had it. In a protocol of 22 September 1919, the Allies forced the
684
Reich government to declare this passage invalid. The National
Assembly agreed to this protocol on 18 December 1919.
The terms affecting the German military were also very strict.
Conscription was abolished, the army reduced to 100,000 and the
navy to 15,000 professional soldiers with extended periods of
service. Germany was allowed no air force, submarines, tanks, or
poison gas. The general staff was dissolved. With the exception of a
few ships, the high seas fleet was to be turned over to the Allies (a
condition the German navy skirted by scuttling the fleet at Scapa
Flow on 21 June 1919).
The peace agreement cost Germany a seventh of its territory and
a tenth of its population, as well as its colonies. If we take into
consideration the partition of Upper Silesia in 1921, it lost a third
of its coal reserves and three-quarters of its iron ore. The Allies had
not yet agreed on a final figure for reparations. For the time being,
Germany had to hand over its undersea telegraph cables, nine-
tenths of its commercial fleet, and more than a tenth of its
livestock. Moreover, it was to deliver some 40 million tons of coal
yearly to France, Belgium, Luxembourg, and Italy for a period of
ten years. The reparation demands were written into the Versailles
Treaty in the fiercely controversial ‘war guilt article’, which
compelled Germany to recognize, along with its allies, that it had
been responsible for ‘causing all the loss and damage to which the
Allied and Associated Governments and their nationals have been
subjected as a consequence of the war imposed on them by the
aggression of Germany and her allies’.
At first, the governing parties were inclined to reject the terms
of the peace treaty. At a demonstration held by the National
685
Assembly at the University of Berlin on 12 May, Prime Minister
Scheidemann asked a rhetorical question: ‘What hand shall not
wither that binds itself and us in these fetters?’ The Prussian prime
minister, Paul Hirsch, also a Majority Social Democrat, started the
slogan ‘Better dead than a slave!’ The president of the National
Assembly, the Centre delegate Konstantin Fehrenbach, called the
treaty the ‘eternalization of the war’ and threatened the Allies with
a second world war, first in Latin, then in German: ‘Memores estote,
inimici, ex ossibus ultor. [The Latin phrase translates as: ‘Remember,
o enemies, from the bones [of the fallen] an avenger shall arise’
(H.A.W.).] In the future, too, German women will give birth to
children, and these children will smash the chains of bondage and
cleanse the disgrace that is to be smeared on our German face.’
Nonetheless, only one party within the ‘Weimar coalition’
adopted an almost completely rejectionist position with regard to
the Treaty of Versailles, the DDP. SPD and Centre were divided
within themselves. The realists, among them Erzberger, David, and
Noske, were alive to the fact that a German rejection would lead to
an Allied occupation of Germany, which the nation’s weakened
military was in no position to prevent. This assessment was shared
by Wilhelm Groener, who still held the office of quartermaster-
general. The German peace delegation did manage to wring certain
concessions from the Allies. On 16 June the victors agreed to a
referendum in Upper Silesia to decide the nationality question.
They also held out the prospect of an earlier end to the occupation
of the Rhineland in the case of good behaviour on the part of
Germany. With regard to the question of war guilt, however, the
Allies were inflexible, sharply and comprehensively rejecting the
686
German view on the matter.
Before an Allied ultimatum forced the National Assembly to
decide whether it wanted to accept the treaty unconditionally, a
new government came to power. Having committed himself to
rejection, Scheidemann resigned on 21 June. He was replaced by a
politically colourless party colleague, Gustav Bauer, the minister of
labour, former second chairman of the general commission of the
Free Trade Unions, and a personal confidant of Ebert. The DDP,
solicitous of its newly acquired nationalist image, did not
participate in Bauer’s cabinet. The Centre, which did, vacillated
between acceptance and rejection of the treaty. On 22 June the
Allies refused to accept Germany’s contention that it could neither
accept sole culpability for the war nor the obligation of extraditing
German war criminals and the politicians the Allies considered
responsible for the outbreak of hostilities. Once again, the outcome
of the decisive vote, which was to take place the next day, seemed
completely open.
Two things tipped the scales. The first was a telegram from
Groener emphasizing the hopelessness of the military situation.
Secondly, the DNVP and DVP declared their willingness to
accommodate the Centre by recognizing the ‘patriotic motivations’
of those—most of the Centre delegates—wishing to vote for the
treaty. The National Assembly then voted by a show of hands to
empower the government to sign. In addition to the Centre
majority, the two Social Democratic parties and a minority of the
DDP voted in favour. The German Nationalists, the DVP, a majority
of the DDP, and a minority of the Centre opposed the signing. On
28 June Foreign Minister Hermann Müller, the newly elected head
687
of the SPD, and the minister of transport, Johannes Bell of the
Centre Party, affixed their signatures to the Treaty of Versailles in
the Hall of Mirrors of the Versailles palace, the same place William
I of Prussia had been proclaimed German emperor on 18 January
1871.
One of the main reasons German outrage over the ‘diktat of
Versailles’ was so profound and lasting was that the Scheidemann
government, against the entreaties of President Ebert, had
consciously chosen not to inform the German public concerning the
events leading up to the war. This it might well have done, given
the availability of the German documents from the July crisis of
1914, which the Council of People’s Commissars had entrusted to
Karl Kautsky, at that time USPD representative in the foreign office,
and Max Quark, MSPD representative in the ministry of the
interior. The foreign office papers tell so heavily against the Reich
leadership and the Austrian government that one can hardly avoid
pronouncing Germany and its main ally primarily responsible for
the outbreak of the war. This was precisely the reason a number of
ministers, led by Scheidemann, refused to allow Kautsky’s
documents to be published. The fear was too great that Germany
might give the Allies further justification for a hard peace.
In mid-June, shortly before the decisive ballot on the Treaty of
Versailles, the Majority Social Democrats, gathering in Weimar for
their first post-war party congress, closed ranks behind the prime
minister. It was in vain that Eduard Bernstein urged his party
colleagues to do justice to historical truth and free themselves from
the prison of 4 August 1914. ‘When I say the old system is guilty, I
am not saying that we, the German people, are guilty. Rather, I am
688
saying that the ones who lied to and deceived the German people
at that time are guilty. From the German people I remove the
blame.’ The debate turned into a tribunal against those deviating
from the party line. The sharpest attack came from Scheidemann.
He called Bernstein ‘an advocate of the devil’ who, in his
overdeveloped sense of justice, would defend even the imperialist
enemies.
The majority of Germans not only repressed the German war
guilt of 1914, but also the terms Germany dictated to Russia in
spring 1918. The territorial and economic conditions of the Treaty
of Versailles were milder than those of Brest-Litovsk. Neither
treaty, of course, was just or sensible. The victors’ representatives,
while working on the preliminary agreements, were under pressure
from their nations, who demanded punishment for the former
central powers, Germany most of all, and compensation for the
losses they had suffered. The victors violated the principle of
national self-determination to the detriment of the vanquished. But
had the Germans not done the same, when they were victorious?
Had they not denied Poland the right to a state ever since the late
eighteenth century? And was a viable Polish state even conceivable
without access to the Baltic, and thus at the expense of German
territory?
Versailles was harsh. But virtually nobody in Germany stopped
to think that it might have been a great deal worse. The Reich
remained in existence and the Rhineland in Germany. Germany
was still the most populous country west of the Russian border and
economically the most powerful nation in Europe. In a certain way,
its international position had even improved over the period before
689
1914. The conflict between the western powers and communist
Russia meant that Germany no longer had cause to feel ‘encircled’.
And even in Versailles, the first cracks in the western alliance—
between France on the one hand, Britain and the United States on
the other—had become visible. Membership in the League of
Nations was not open to Germany, but it did not have to stay that
way. Germany had good prospects of becoming a European great
power again. Sober reflection on the new situation was all that was
required to see ‘Versailles’ in realistic proportions.
But sober thinking was seldom to be found in Germany in the
summer of 1919. Versailles provided grist for two historical legends
that severely burdened the new republic from its very beginning
and prevented a clean moral break with Wilhelmine Germany. The
first was the myth of war innocence, which ‘nationalist’ Germany,
with the backing of prominent historians, used to counter the
putative ‘war guilt lie’ of the victors. In the view of the ‘right’,
Germany was not to blame for the war, or at least no more to
blame than any of the other nations that had taken part in it. The
other myth was the so-called Dolchstosslegende, the legend of the
‘stab in the back’ that the German army, ‘undefeated in the field’,
had received at the hands of the homeland. This story can be
traced back to the last year of the war. It was given classic
expression by Hindenburg, the last head of the Army High
Command, who left office together with Groener on 25 June 1919.
On 18 November 1919, speaking before a parliamentary committee
investigating the causes of the German collapse, Hindenburg
quoted an unknown British general as having said that the German
army had been ‘stabbed in the back’. The twin myths were already
690
working their poison in the first days of ‘Weimar’. There were
many views on how Versailles could be revised. That it had to be
revised was the unanimous opinion of all Germans from the day
the treaty was signed.8
691
government officials, half to delegates from the provincial
assemblies.
Preuss, who was secretary of the interior from 15 November
until 13 February 1919, then minister of the interior until 20 June
1919, aspired to turn Germany into a decentralized unitary state.
This provoked the federalists, from the Bavarian conservatives to
the Social Democrats now in power in many of the German states.
Ultimately, the Reich became more federal than the unitarians
wished, and more unitarian than the federalists wished. The
Weimar Republic was not a unitary state like the third French
republic. The Länder were more than mere administrative units.
Prussia’s separation from the federal government had the effect of
destroying the former’s privileged status. On the other hand, the
Reich Council had far less influence than the Federal Council under
the monarchy. The southern German states lost the ‘special rights’
in matters of military, postal, and tax policy granted to them by
Bismarck. Bavaria was the hardest hit by these losses, and there
was little indication that the southern German Free State would
long accept the diminution of its status.
In another area Preuss was much more successful. On the
recommendation of Max Weber, he introduced the office of a Reich
presidency. Democratically legitimated through direct election by
the people, a strong president would prevent ‘parliamentary
absolutism’ by providing a check on the Reichstag. As a non-
partisan, integrating force, he would also both counterbalance the
tendency of the party system towards fragmentation and represent
the natural point of reference for the professional bureaucracy,
itself obligated to be non-partisan, thus securing governmental
692
continuity between the monarchy and the republic.
The conservative aspects of the institution initially provoked
sharp criticism from the Social Democrats. In a meeting of the
party fraction on 25 February 1919, the SPD veteran Hermann
Molkenbuhr referred to the president as a ‘replacement emperor’
and spoke of the ‘truly Napoleonic trick of the people electing the
president’. On 28 February, at the first plenary reading of the draft
of the constitution, the Social Democratic speaker, Richard Fischer,
warned that the proposal gave the president greater and less
restricted powers than those possessed by the presidents of the
Republic of France or the United States of America. Furthermore,
the new constitution should not be tailor-made to fit President
Ebert, as the earlier constitution had been tailor-made to fit
Bismarck. ‘We must consider the fact that one day another man
from another party—perhaps a reactionary, coup-plotting party—
will stand in this place. We have to prepare against such chances,
especially since the history of other republics provides very
instructive examples of this kind of thing.’
Such reservations, articulated even more strongly by the USPD,
found no resonance within the bourgeois parties. In response to the
civil war-like conflicts in spring 1919, the desire for a powerful
presidency grew strong, even within the MSPD. Contrary to their
original intention, the Social Democrats eventually accepted a long,
seven-year term of office, and they did not even resist when the
bourgeois parties extended the presidency’s extraordinary powers
in emergency situations. The third reading on 30 July 1919
scrapped the drafting committee’s decision to make measures
pursuant to Emergency Article 48 dependent on the parliament’s
693
agreement. Instead, it was considered sufficient that the president
immediately notify the Reichstag of his decisions. The right of that
body to annul the president’s measures was not affected by this
modification, however.
The principle of parliamentary and representative democracy
was not only restricted by the Reich president as a replacement
legislative, but also by the possibility of direct popular legislation.
The Social Democrats had demanded popular legislation in their
1869, 1875, and 1891 platforms. Consequently, they were now the
driving force—despite second thoughts from some party members
—behind the introduction of a referendum and a petition for a
referendum. The DDP was less enthusiastic about such things than
the MSPD and USPD, but nonetheless lent its support, along with
some of the DNVP delegates. The DVP was strictly opposed.
Long debates ended in compromise. The Social Democrats’ wish
for popular ratification of the constitution and a petition for a
referendum on the dissolution of the Reichstag were not adopted.
But a majority did agree that a referendum should be held when a
tenth of the electorate petitioned for one. A parliamentary bill was
to be subject to referendum when a twentieth of the electorate so
desired and a third of the Reichstag had opposed its passage. In
order to annul a parliamentary decision, however, a majority of the
electorate was required to participate in the referendum. The SPD,
Centre, and DDP erected these hurdles in the third reading in order
to make it more difficult for the president to oppose the Reichstag,
since the constitutional committee—against the advice of SPD
delegate Wilhelm Keil—had granted the president the right to call
for a referendum against a parliamentary law.
694
For a time it looked as though the Reichstag would have yet
another competitor in the legislative function, an ‘economic
parliament’ (Wirtschaftsparlament). Such an organ was demanded
by the Second Congress of the Workers’, Farmers’, and Soldiers’
Councils of Germany in Berlin in April, as well as by the German
Nationalists, who revived Bismarck’s plan for a profession-based
corporative chamber. But the majority of the National Assembly
had no interest in such ideas, irrespective of their ‘leftist’ or
‘rightist’ provenance. Article 165 (the so-called Räteartikel) did
provide for an federal economic council with a right to initiate and
examine legislation. But lacking veto power, this body had no
ability to limit the authority of the Reichstag.
The most important thing about Article 165 was not the
economic council, but the codification of the principle of parity
between capital and labour, and government recognition of
Tarifautonomie, the right to free collective regulation of terms and
conditions of employment. Generally, however, the National
Assembly treated issues concerning the social order with great
caution. The constitution made ‘nationalization’ optional, not
obligatory. It allowed freedom of coalition, but not the right to
strike, since then it would have had to define the limits of that
right.
Several controversial matters of great symbolic importance
could not be left open, however. One of these was the title of the
constitution. The two Social Democratic parties favoured the
‘Constitution of the German Republic’, but were unable to prevail
against the united front of the bourgeois parties, who insisted upon
the ‘Constitution of the German Reich’. On the question of the
695
national flag, the USPD wanted red, the MSPD, invoking the legacy
of 1848, black, red, and gold. The majority of the Centre and a
minority of the DDP also favoured the colours of the German
unification movement; the DNVP and DVP, along with the majority
of the DDP and a minority of the Centre, desired to preserve the
black, white, and red, the colours of the Bismarck empire. After
heated debate a compromise was worked out here, too. The Reich
colours would be black, red, and gold, but a separate flag was
created for the commercial fleet (supposedly for reasons of ‘better
visibility’) with the colours black, white, and red ‘with a jack in
black, red, and gold in the upper inner corner’. Clearly, not only
the political right, but also elements within the ‘Weimar’ parties
mourned the loss of the old Reich.
The most passionate battle was fought over the article dealing
with schools. The Social Democrats sought to improve the social
position of the lower classes through reform of the school system;
the Centre desired to preserve confessional schools; the liberals
wanted to curtail church influence in education. These antagonisms
proved so tenacious that the SPD temporarily considered
abandoning the whole section dealing with basic rights. This the
Centre could not do, however, since the catalogue of basic rights
included those dealing with matters of religion and religious
organizations, rights of paramount importance to the Catholic
Church. It was not until shortly before the third reading that the
‘Weimar’ parties agreed on the solution that made its way into the
constitution: the interdenominational school (the so-called
Simultanschule) was to be the ordinary school type, but it could be
replaced by a confessional or non-religious school if the parent or
696
guardian so desired.
This compromise saved the constitutional ‘Basic Rights and
Obligations of the German People’ (which updated and expanded
the ‘Basic Rights of the German People’ in the imperial constitution
of 28 March 1849) and assured the Weimar constitution a broad
majority in the final vote on 31 July 1919. Of the 420 members of
the National Assembly 338 participated, 262 voting for adoption,
75 opposed, and one delegate abstaining. The votes in favour came
from the SPD, Centre, and DDP (the ‘Weimar’ parties), the
negatives from the USPD, DNVP, and DVP. President Friedrich
Ebert signed the constitution on 11 August, and it took effect on 14
August through proclamation in the Reichsgesetzblatt. One week
later, on 21 August, the president, National Assembly, and cabinet
departed from Weimar. Henceforth Germany was governed from
Berlin.
On 31 July Eduard David, the Social Democratic minister of the
interior, hailed the adoption of the Weimar constitution: Germany
was now the ‘most democratic democracy in the world’; nowhere
else had democracy been ‘as thoroughly realized as in this
constitution’. He was thinking primarily of its provisions for direct
democracy. The attitude of the public was more one of passive
acceptance than of active adoption. It was only in the course of the
campaigns of hatred and violence waged by the far right that the
constitution was to become a symbol of the republic. The document
itself contained no guarantee against its own abolition, provided
that the necessary majorities voted accordingly. To the fathers and
mothers of the 1919 constitution, clauses limiting the will of the
majority would have seemed like a relapse into the authoritarian
697
state.
And yet the authoritarian state was by no means a thing of the
past. It lived on not only apart from the new constitution, but also
in it and through it. By granting the president legislative powers in
vaguely defined emergency situations, the National Assembly
provided ample scope for parliamentary opportunism. If the ruling
parties were having difficulty reaching a compromise, the
temptation was great to shift responsibility ‘upstairs’ to the head of
state. An emergency measure on the basis of Article 48 then took
the place of the regular legislative process, the parliamentary
constitution giving way to a ‘presidential reserve constitution’ and
the ‘provisional dictatorship of the president’. While the Reichstag
was authorized to dismiss the chancellor and any minister without
naming a successor, the constitution made no provision for it to
elect a new chancellor. All that was necessary was that the
president’s choice of a chancellor enjoy the confidence of the
Reichstag. Since the president could dissolve the Reichstag, the
parliamentary vote of no confidence was a weapon that would
blunt rapidly. The president, not the parliament, had the whip
hand.
The new constitution brought the Germans a great increase in
personal liberty. But it did not guarantee the preservation of that
liberty in difficult times. The ‘most democratic democracy in the
world’ was not only endangered by the powers that rejected and
fought it. It was also designed in such a way as to permit its own
self-abolition.9
698
THE KAPP–LÜTTWITZ PUTSCH
AND THE RUHR UPRISING
699
confiscatory taxes did not work. In fact, since businesses shifted the
costs to consumers in the form of higher prices, the new revenues
exacerbated the devaluation of the currency. Higher prices, in turn,
stymied Erzberger’s efforts towards social justice. All that survived
of his finance reform was the unification of the tax system and
finance administration, along with the political consequences that
went along with it. The financial disempowerment of the Länder
placed considerable strain on their relationship to the federal
government, and the high indebtedness of the cities and
communities, a result of the drastic curtailment of their sources of
revenue, was to prove one of the main causes of Germany’s
financial weakness in the second half of the 1920s.
The political right had long hated Erzberger, author of the July
1917 peace resolution and signatory of the November 1918
armistice agreement. His levies on wealth made the hatred even
more virulent. In January 1920, the finance minister found himself
compelled to prefer charges against Karl Helfferich, former
secretary of the interior, for accusing him of ‘habitual
untruthfulness’ and the constant mingling of his business interests
with his political duties. On 26 January Erzberger, leaving the
courtroom in Moabit, was shot and badly wounded by a discharged
cadet. At his trial on 21 February, the culprit was charged not with
murder, but only with causing ‘grievous bodily harm’. The next
day, the campaign against Erzberger escalated even further with a
newspaper article accusing him of tax evasion. Two days later,
Erzberger requested to be temporarily relieved of his duties of
office.
The trail against Helfferich was concluded on 12 March 1920,
700
the former secretary of the interior condemned for defamation and
libel. But the real loser, morally and politically, was Erzberger,
whom the by no means impartial judges considered guilty of two
instances of perjury and seven instances of corruption. Though he
was later cleared of the charges of perjury and tax evasion, the
public impact of the sentence was catastrophic. Later that day
Erzberger resigned from his office as minister of finance.10
That same day, the Reich cabinet learned from Gustav Noske
that a coup d’état was in the works. According to the defence
minister, the wirepullers of the operation were Wolfgang Kapp,
general director of the East Prussian Landschaft, and Captain
Waldemar Pabst. In fact, considerable portions of the Reichswehr
were behind the plan, which went down in history as the ‘Kapp
putsch’. Ever since the Treaty of Versailles went into effect on 10
January 1920, prominent officers, led by General Walter von
Lüttwitz, commander of the District Command I in Berlin, had been
steering towards a conflict with the government. That the Reich
would fulfil the Allied demand to try German war criminals before
German courts, if not extradite them, was an idea many officers
found intolerable. The reduction of the army to 100,000 men, not
yet complete, was an additional grievance. The free corps were
most strongly affected, above all the so-called Baltikumer, who had
fought against the Bolsheviks in Estonia and Latvia after the end of
the war with the approbation of the Allies. Noske named one of the
free corps by name: Marine Brigade Erhardt.
The civilian wing of the conspiracy consisted of politicians from
the far right supported by Junkers and monarchist officials in the
old Prussian provinces. Its planning centre was the National
701
Alliance (Nationale Vereinigung) in Berlin, founded in October 1919
under Ludendorff’s patronage. Its aim was an authoritarian—
though not yet monarchical—regime that would actively pursue
revisionist policies on the international stage.
Noske’s defensive measures proved insufficient. On the morning
of 13 March the Erhardt brigade entered Berlin. Kapp seized the
chancellery towards seven o’clock. Since most of the generals,
including the chief of the office of troops, Hans von Seeckt,
considered military counter-measures hopeless, President Ebert,
Chancellor Bauer, and most of the ministers had shortly before left
for Dresden (whose commanding officer, General Maercker, Noske
believed to be loyal). In the meantime, an appeal went out in
Berlin proclaiming a ‘general strike along the whole line’ and
calling for the unification of the proletariat. The author was the
government chief of the press, the Social Democrat Ulrich
Rauscher. The manifesto bore the name of Ebert and the Social
Democratic ministers was well as that of Otto Wels, acting
chairman of the SPD. At Maercker’s reproach, the president and the
Social Democratic ministers distanced themselves from the appeal.
In fact, probably only Noske had read and approved the text in
advance. We can also assume that Wels knew about and had agreed
to it.
The danger of a general strike was that it could easily get out of
control and turn into civil war. It was all but certain that
Communists and syndicalists would not limit themselves merely to
fighting for the restoration of the Bauer government. On the other
hand, however, a general strike had a good chance of success in
spring 1920. An inflation-induced economic boom had brought
702
Germany to near-full employment. Striking workers did not have to
worry that their jobs would be taken by the unemployed. A general
strike against a military coup and for the constitutional authority of
the state possessed an undeniable democratic legitimacy. And there
was every reason to believe that a strong signal from workers and
employees was necessary in order to convince the bureaucracy to
unite against the putsch and to force its leaders to capitulate.
The leadership of the strike in March 1920 was assumed by the
Free Trade Unions, in which Majority and Independent Social
Democrats were still cooperating. The General German Trade
Union Association (Allgemeiner Deutscher Gewerkschaftsbund or
ADGB) and the Cooperative Union of Free Employees’ Federations
(Arbeitsgemeinschaft freier Angestelltenverbände) were able to
organize the deeply divided workers’ movement in common action.
It was hoped that the Communists, if they could not be moved to
cooperate, would at least be held in check. For its part, the KPD
first announced on 13 March—the text was drafted by Ernst Reuter,
the Berlin party leader—that the revolutionary proletariat would
‘not lift a finger for the government of Karl Liebknecht’s and Rosa
Luxemburg’s murderers, which has collapsed in shame and
disgrace’. But since many Communists were already participating
in the strike, the KPD made a new announcement the next day,
now calling the strike the commencement of the battle against the
military dictatorship and exhorting Communists ‘to be bound and
limited in their actions by the goals the majority of the workers
provisionally sets for itself’. This was nothing less that an appeal to
form the kind of unified proletarian front the party had rejected the
day before.
703
The coup’s supporters were limited from the beginning to the
conservative forces in East Prussia. Since most of the ministerial
bureaucracy refused to cooperate with the Kapp ‘government’, and
since the protesting workers and employees had brought the whole
economy to a standstill, the failure of the coup was already
foreseeable on 14 March. Thus it is all the more astonishing that
the German People’s Party, despite all its reservations against a
violent overthrow of the government, refused to actually condemn
it. Instead, in a 14 March announcement drafted by its leader,
Gustav Stresemann, the DVP blamed the ‘recent government’ for
‘disrupting the path of organic development to which we are
committed’. As a solution, the party limited itself to calling for the
legalization of the ‘current provisional government’ and early
elections. In the ensuing days Stresemann, who saw himself as a
‘mediator’ between the hostile camps, even spoke personally with
Lüttwitz and Kapp.
The republican forces were even more exasperated by the fact
that Vice Chancellor Schiffer, who had remained in Berlin (his
party, the DDP, had joined the Bauer government again in October
1919), made far-reaching concessions to the coup leaders on 16
March in the presence and with the agreement of several Prussian
ministers, including the Social Democrats Hirsch and Südekum: if
‘Chancellor’ Kapp and ‘Commander-in-Chief’ von Lüttwitz resigned,
a new coalition government would be formed and new Reichstag
elections held, including a direct election for president. This would
have meant a partial success for the rebels.
The Bauer government, which had now retreated to Stuttgart,
did well not to seek compromises. On 17 March, under pressure
704
from the military, Kapp resigned, followed by Lüttwitz. Marching
from the government district to the tune of the ‘Deutschlandlied’,
the Erhardt brigade, clad in its usual gear (‘Hakenkreuz am
Stahlhelm, schwarz-weiss-rotes Band | Die Brigade Erhardt werden wir
genannt’),* staged a parting massacre among the civilians gathered
to protest. Twelve were killed, thirty wounded.
The end of the Kapp putsch was not the end of the general
strike. The General German Trade Union Association, the
Cooperative Union of Free Employees’ Federations, and the German
Civil Servants’ Association (Deutscher Beamtenbund) decided to
continue the action until a number of conditions were met. Noske,
who was seen as responsible for alienating large portions of the
Reichswehr, was not to return to Berlin as commander-in-chief of
the troops. The undependable units were to be dissolved, disarmed,
and the troops reorganized to make any future coup attempt
impossible. In addition, the three groups demanded ‘a key role in
organizing the new situation’.
This announcement was followed on the same day by a nine-
point programme by the workers’ organization demanding, among
other things, the punishment of all those involved in the coup; the
fundamental democratization of the administration; the
nationalization of the mining and energy industries; a security
service organized and staffed by workers; and the resignation of
two Prussian ministers, the Social Democratic minister of the
interior, Wolfgang Heine, and the Democratic minister of transport,
Rudolf Oeser. Heine, it was said, had been too tolerant of
reactionary elements, and Oeser too soft on the rebels.
705
Negotiations between the unions, representatives of the majority
parties, and the Reich and Prussian governments led to a partial
agreement on 20 March. By this time, Noske and Heine had handed
in their resignations. Units of the security police disloyal to the
constitution were to be broken up and replaced by dependably
republican groups. The socialization commission was to reconvene
and prepare for the nationalization of selected industries. The Free
Trade Unions then announced an end to the strike. The USPD did
not do so until 23 March, after the Bauer administration had
granted several further concessions.
The reorganization of the government was completed by 27
March. The colourless Gustav Bauer, whose reputation had sunk
even further during the crisis, was replaced by Hermann Müller,
one of the two chairmen of the SPD (the other being Otto Wels).
Müller, who was born Mannheim in 1876, had learned several
languages as a commercial employee, which helped his political
career. Long before he assumed the office of German foreign
minister in June 1919, he was the informal foreign minister of the
SPD, speaking for the party in its negotiations with sister parties in
western Europe. To replace the discredited minister of defence with
another politician from their own ranks seemed to the Social
Democrats neither important nor opportune. After Wels refused the
post, it went to Otto Gessler, former mayor of Nuremberg and
hitherto minister of reconstruction. A member of the DDP right
wing, Gessler made no secret of his monarchist proclivities. General
von Seeckt, who on 13 March had very forcefully refused to allow
Reichswehr troops to fire on their own, took over leadership of the
army.
706
The new government was not as leftist as the Free Trade Unions
desired, and Müller’s cabinet bore little resemblance to the kind of
‘workers’ government’ Carl Legien, head of the ADGB, had called
for. But such an administration was never more than a chimerical
hope. The Weimar coalition could not support a ‘trade union state’,
and it was wishful thinking to imagine that the social reforms left
undone in the winter of 1918–19 could be made up in the spring of
1920.
Still, it was not too late to make the administration more
democratic, as the new Prussian government under Otto Braun
(hitherto minister of agriculture) proved. This agile Social
Democrat from Königsberg, a printer by trade, appointed Carl
Severing to the office of interior minister. Severing, a metalworker
from Westphalia, was a long-time SPD delegate in the Reichstag
and had shown great ability as federal and Prussian commissioner
in the troubled Ruhr district. As minister of the interior, Severing
initiated a large-scale reshuffle of offices in the provincial and
police administrations. Officials who had collaborated with the
putsch were replaced by men Severing trusted to stand firm in the
republic’s defence. This reorganization opened a new chapter in the
history of Prussia. Within a few short years, the former
Hohenzollern state would become a bulwark of the German
Republic.
The opposite happened in Bavaria. On 14 March 1920 Munich
experienced its own kind of coup. General Arnold Ritter von Möhl,
commander of Group IV of the Reichswehr, after reaching an
understanding with monarchist politicians like Georg Escherich,
captain of the paramilitary civilian police forces (the so-called
707
Einwohnerwehren), Gustav Ritter von Kahr, district president of
Upper Bavaria, and the Munich police chief Ernst Pöhner,
demanded that the Social Democratic prime minister, Johannes
Hoffmann, transfer the executive authority to him, Möhl, in the
interest of peace and the public order. The Bavarian coalition
government—a minority cabinet made up of SPD, Bavarian
Peasants’ League (Bayerischer Bauernbund), and non-aligned
politicians—tolerated by the BVP and DDP—submitted to the
ultimatum. Hoffmann resigned, refusing to go along. Two days
later, the Bavarian parliament elected Gustav von Kahr to the office
of prime minister by a one-vote majority. His government was
composed of members of the BVP, DDP, and the Bavarian Peasants’
League. The Social Democrats went into the opposition, from which
they did not emerge again during the life of the Weimar Republic.
Bavaria became a rightist ‘cell of order’ (Ordnungszelle), a
stronghold of all the forces seeking to redirect the Reich towards
the right and replace parliamentary democracy with an
authoritarian system of government.
It was not until the suppression of the Ruhr uprising that the
Kapp–Lüttwitz putsch was finally over, however. The ‘Red Ruhr
Army’, the paramilitary arm of a proletarian mass movement far
larger than the KPD and its followers, had seized power in the
industrial district of Rhineland-Westphalia after the rightist coup.
Its leaders had no intention of relinquishing their position after the
return to parliamentary government in Berlin. The most radical
groups were the executive councils in the mining areas of the ‘wild
west’, where syndicalists and leftist Communists were in command.
In the eastern and southern parts of the district, where
708
metalworking was the primary industry and the USPD was
dominant, the federal and Prussian governments found a much
greater willingness to cooperate.
Berlin took advantage of this political difference. The ‘Bielefeld
agreement’ of 24 March (mainly a product of Severing’s
negotiation), which among other things provided for the workers to
turn in their weapons under the common supervision of executive
councils and local authorities, succeeded in convincing the
moderate elements to stop fighting. But the more radical councils
in Mülheim and Hamborn and the leaders of the Red Ruhr Army
refused a ceasefire, and the Reichswehr and free corps were
burning to settle accounts with the far left. The chaotic conditions
among the anarchic leaders in Duisburg convinced the Reich
government of the necessity for rapid action. The military
repression of the rebellion took place between 30 March and 3
April. Among the forces deployed were units that shortly before
had backed the Kapp–Lüttwitz putsch. The exact number of those
killed in the Ruhr civil war has never been ascertained. The
workers numbered well above 1,000 dead, the Reichswehr 208
dead and 123 missing, and the security police 41 dead.
The Ruhr uprising was the last of the great proletarian mass
movements that had begun with the wildcat strikes of the year
1917. There is good reason to think of it as the third phase of the
German revolution, which had entered a latent stage after the
suppression of the Second Munich Council Republic in May 1919.
The radical workers were protesting, on the one hand, against the
political and social system they held responsible for the war and
against those who wanted to restore this system after 1918, and, on
709
the other hand, against the traditional workers’ organizations,
which they had come to regard as a part of the capitalist order. Of
course, the unions and the Social Democrats still had far more and
better qualified workers in their ranks than syndicalists,
Communists, and independents. The political divide went right
through the middle of the working class, and no one party and
trade union could speak for it as a whole any longer.
The desire for a radical restructuring of the societal order
survived the end of the revolutionary period in spring 1920. But
the experiences of those weeks was sobering. The general strike
had revealed itself to be a two-edged sword. To the extent that it
helped bring the Kapp regime quickly to heel, it was certainly a
success. Yet it developed a momentum of its own, one the unions
and Social Democrats were powerless to stop. Against the will of
the moderates, the radical left transformed the political strike into
an armed conflict from which not the workers, but rather the
military emerged victorious. Although the Ruhr uprising was
followed by Communist rebellions like the ‘March action’ of 1921
in central Germany, the age of proletarian mass demonstrations
was over, and there were no more general strikes in the Weimar
Republic after 1920.
Those involved in the Ruhr uprising were punished far more
severely than the participants in the recent putsch, most of whom
were able to escape abroad, including Kapp and Lüttwitz. Pursued
by the law, Lieutenant Commander Ehrhardt took refuge in the
Bavarian Ordnungszelle, where, under bureaucratic cover, he set
about preparing the next stage of the counter-revolution. The Reich
government granted only a minimal part of the concessions it had
710
made to the striking workers. The efforts of the new socialization
commission remained as inconsequential as in 1919. Undependable
police units were broken up only in places where the Social
Democrats had sufficient power. Few workers joined the
Ortswehren, the new local security forces. The Reichswehr, in order
to avoid being implicated in anti-constitutional activities, was
careful to display political restraint. At the same time, however, an
amnesty in August 1920 allowed free corps officers who had
supported the coup to be taken into the final-status army and navy.
A staunch anti-republican attitude was no bar to professional
advancement in the ‘state within the state’, which is what the
German military in the Seeckt era in effect became.
The first Reichstag election after the revolution took place in
Germany on 6 June 1920 (though ‘Germany’ must be qualified
here; open border questions delayed the elections in Schleswig-
Holstein, East Prussia, and Upper Silesia). They turned into a
debacle for the republican forces. The parties of the Weimar
coalition, who had possessed a two-thirds majority in the National
Assembly, lost the majority of votes and seats. The MSPD sank
from 37.9% to 21.6% of the vote, while the USPD was able to
increase its share from 7.6% to 18.6%. The KPD, on the ballot for
the first time, achieved only 1.7%. The DDP fell from 18.5% to
8.4%, while the DVP grew by almost the same amount, from 4.4%
to 13.9%. The Centre Party’s losses were comparatively minimal. In
January 1919 it had gained 15.1% (without the BVP); now it
managed 13.6%. The DNVP improved its share of the vote from
10.3% to 14.4%.
A shift to the right in the bourgeoisie, a shift to the left among
711
the workers: this phrase sums up the election results. The elements
that had refused to endorse the founding compromise of the
German Republic were politically rewarded. The moderates were
punished for what they had done—or neglected to do—after the
beginning of 1919. The left blamed the republican governments for
the resurgence of the ‘reaction’. The right condemned the Weimar
parties for everything that injured the national honour and harmed
the interests of property. Versailles and the tax reform, the Kapp
putsch and the ensuing battles: all this had an effect on the
outcome of the election, which was, in effect, a vote of no
confidence in the Weimar Republic.
Not that a new majority was anywhere in sight. A Grand
Coalition from the Social Democrats to the German People’s Party
was not yet an imaginable way out of the crisis; the memories of
the tactical opportunism of Stresemann and his party during the
putsch were still too fresh among the SPD, and the ‘anti-Marxist’
sentiments of the DVP too strong. A bourgeois bloc including the
German Nationalists was even less conceivable. The only solution
was minority governments the opposition parties were willing to
tolerate, either a Weimar coalition supported by the DVP and USPD
or else a bourgeois cabinet shored up by the SPD. The SPD
preferred the second variation, since the first seemed to offer even
less scope for the party to display its distinctive image.
On 25 June 1920 Ebert appointed the Centre Party politician
Konstantin Fehrenbach, president of the National Assembly, as
chancellor. He formed a cabinet composed of members of the
Centre, DDP, and DVP, along with two independent ministers. For
the first time since October 1918, Germany had a government
712
without the SPD. But the Reich could not be governed without the
Social Democrats. This they knew, and the bourgeois minority
cabinet adapted itself accordingly.11
The spring of 1921 saw two crises, both with a common root in
questions left unresolved by the Treaty of Versailles. The first was
in Upper Silesia. On 20 March, the plebiscite called for in the treaty
took place. Those voting for Germany were 60% of the population,
those for Poland 40%, the former predominantly in the industrial,
the latter in agricultural areas. Thereupon the Reich government
demanded all of Upper Silesia for Germany, while Poland and the
Allies endorsed a partition. In order to underscore its demands, the
Warsaw government secretly sponsored an uprising, in the course
of which Polish insurgents occupied large parts of the plebiscite
territory.
The Reich and Prussian governments responded by delivering
arms to the Upper Silesian militia, a paramilitary formation
established in 1920. Together with the Oberland, a Bavarian free
corps, the militia stormed the Annaberg, the highest point in Upper
Silesia, on 23 May. At the end of June an Allied commission
brought about the withdrawal of the armed parties. On 20 October
1921 the Allied High Council decided the border dispute according
to a League of Nations study: four-fifths of the industrial territory
in Upper Silesia was to go to Poland, including the cities of
713
Kattowitz and Königshütte, in which overwhelming majorities had
voted for Germany. With no means to compel a more favourable
resolution, Germany could do nothing but protest at this violation
of the right to self-determination.
The second crisis in the spring of 1921 had to do with the war
reparations. The payments were so high that they could not be
raised in ‘normal’ fashion, by taxation, and thus made inflation
worse. The fact that the peace treaty had not set the total amount
Germany owed had fatal consequences. Perpetual uncertainty over
the extent of the payments made it impossible for potential private
creditors to realistically assess the creditworthiness of the country.
Under such conditions, Germany could get no more long-term
loans.
On 5 May 1921 the British prime minister, Lloyd George,
handed the German ambassador in London an Allied ultimatum
requiring reparations in the amount of 132 billion gold marks, to
be paid in several stages at the current exchange value, that is, not
including future interest, and with an additional 6 billion for
Belgium, which had been attacked by Germany in 1914. A payment
of 1 billion gold marks was required within twenty five days, by 30
May. Furthermore, the Allies demanded the payment of the 12
billion gold marks still outstanding from the total of 20 billion due,
according to the Treaty of Versailles, on 1 May, disarmament in
keeping with the Allied prescriptions, and the condemnation of
German war criminals. In the case of non-fulfilment of Germany’s
obligations, the Allied threatened to begin occupation of the whole
Ruhr area on 12 May (Düsseldorf, Duisburg, and Ruhrort had
already been occupied on 8 March 1921 in punishment for non-
714
compliance with the previous ultimatum).
The day before the London ultimatum was issued, Fehrenbach’s
cabinet, having failed to enlist the mediation of the United States,
had announced its intention to resign. The reparations crisis thus
coincided with a crisis of government, and both could only be
solved in tandem. The DNVP, DVP, and KPD demanded that the
ultimatum be rejected. SPD, Centre, and USPD wished to accept it,
in view of the looming sanctions. The DDP was split.
A victory of the hard line would have meant the economic
collapse of Germany. The rightist parties knew this, but they could
assume, as with the vote over the Treaty of Versailles in June 1919,
that there would be a majority for the lesser evil even without their
support. Their calculations proved correct. SPD, Centre, and DDP
assumed the responsibility of accepting the ultimatum and formed
a new government, the first minority cabinet of the Weimar
coalition. Its leader was Joseph Wirth of the Centre Party,
appointed by Ebert on 10 May. Wirth, a former mathematics
teacher from Baden, had succeeded Erzberger as finance minister in
March 1920. He was a brilliant speaker and ardent nationalist, but
also a committed republican and, as far as domestic issues were
concerned, a ‘leftist’ within his party. Wirth’s appointment to the
chancellorship was the beginning of what came to be known as the
‘policy of fulfilment’ or Erfüllungspolitik.
Erfüllungspolitik meant that Germany would do its utmost to
fulfil the requirements imposed upon it, thereby demonstrating the
absurdity of the reparations policy. Since the reparations
overburdened the country’s economic capacities, catastrophic
consequences could be expected. These would convince the Allies
715
of the necessity of revising the London payment plan. Wirth was
not the only one to adopt this logic; the Reichstag majority did,
too. On 10 May 1921 it accepted the London ultimatum with 220
votes against 172. MSPD, USPD, and Centre voted unanimously in
favour, along with a strong minority of the DDP and small
minorities from the DVP and BVP. The Wirth government had
passed its first test.
Among the London ultimatum’s non-material requirements, one
went all but unfulfilled—the condemnation of war criminals. Nine
trials against twelve defendants did take place between May and
July 1921 before the federal court in Leipzig, but only half of these
led to convictions. The greatest stir was caused by the conviction of
two lieutenants who had participated in the sinking of rescue boats
from a torpedoed steamer. Both were condemned to four-year
prison terms. This outraged the navy. In January 1922 the
imprisonment came to an abrupt end when members of the right-
wing radical ‘Consul’ organization, led by Lieutenant Commander
Ehrhardt, freed the two officers from their prison cells. The Allied
registered protest against the small number of convictions and the
lenient sentences, but they undertook no action. Apart from the six
1921 convictions, German war crimes went unatoned.
The demand for disarmament was formally fulfilled by the early
summer of 1921. It primarily affected the Bavarian
Einwohnerwehren, which the Munich government had stubbornly
refused to dissolve the year before. Under massive Allied pressure,
Prime Minister Kahr finally had to order their disarmament in June
1921. Three weeks later, on 24 June, the federal government
declared the Bavarian Einwohnerwehren, the East Prussian local and
716
border militias (Ortswehren and Grenzwehren), and Bavarian Forest
Steward Georg Escherich’s paramilitary ‘Escherich Organization’
(nicknamed Orgesch) disbanded throughout the whole Reich.
But that was not the end of the paramilitary influence on
German politics. The Ordnungszelle Bavaria remained the El Dorado
of numerous ‘Patriotic Leagues’ (Vaterländische Verbände) far more
radical than the Einwohnerwehren. And in the rest of Germany, too,
the Weimar state failed to secure the ‘monopoly of legitimate use of
physical force in enforcing its rules’ that Max Weber identified as
the main characteristic of the state as a compulsory political
association with continuous organization (politischer
Anstaltsbetrieb). This monopoly had already been undermined by
the war. Widespread arms ownership, as Weber correctly noted,
was one of the things that made the revolution of 1918–19
possible. To a certain extent, the transfer of arms from the
Reichswehr to the free corps, local, and citizen’s militias was
understandable as a reaction to the forced, one-sided German
disarmament and to attempted Communist takeovers. But it also
contributed to the militarization of public life, and its consequences
outlasted the civil war-like conflicts of the first five years after the
war. Paramilitary organizations and party armies of the widest
political description continued to block the development of a civil
society, and a literature glorifying war did its part in keeping the
spirit alive—a spirit that would construct its body, a militarily
powerful Germany, capable of seeking vengeance for 1918.12
The hard core of the London ultimatum could not be softened.
Already in 1921, Germany had to pay 3.3 billion gold marks in
reparations, of which 1 billion was due on 30 May. The Reich was
717
able to come up with only 150 million in cash for the first
instalment. The remainder was financed through three-month
treasury bills, which the government had the greatest difficulty
redeeming by the due date. This operation clearly made inflation
worse and induced the Social Democratic minister of trade and
commerce, Robert Schmidt, to call on 19 May 1921 for
fundamental financial reforms: the expropriation of 20 per cent of
capital assets in agriculture, industry, commerce, banking, and
housing.
Schmidt’s call for the ‘capture’ of asset values spelled the end of
the silent ‘inflation consensus’ that had characterized German
economic, finance, and social policy since 1919. In the spring of
1921, the Social Democrats and Free Trade Unions began to realize
that the devaluation of the currency steadily shifted the societal
balance of power in favour of private property and against the
workers. They also recognized that the reform of Germany’s
finances would not be possible without massive government
intervention in existing property relations. Business interests and
the bourgeois parties rejected this notion. Walther Rathenau,
former president of the board of directors of the General Electricity
Society, now minister of reconstruction and close to the DDP, said
that Schmidt’s proposals robbed Germany of economic freedom and
that consumption could produce higher numbers than property.
Joseph Wirth, who held both the chancellorship and the office of
finance minister, agreed with his experts that ‘the forces necessary
to put these plans into action cannot be found.’
The political right morally condemned Erfüllungspolitik for the
simple reason that ‘Marxists’—that is, MSPD and USPD—had
718
helped bring about the acceptance of the London ultimatum. The
political centre, too, drew fire from the right, on account of its
cooperation with the moderate left. In the Miesbacher Anzeiger, a
provincial newspaper very popular in Bavaria, the writer Ludwig
Thoma (who remained anonymous) called Chancellor Wirth the
‘confidant of the shyster from Biberach’, a reference to Erzberger.
Thoma characterized Georg Gradnauer, the Social Democratic
minister of the interior, in the following manner: ‘Narrow, Saxon
eyes; Hebrew nose and chin; the giant ears even more Hebrew.’
When it was announced in the middle of June 1921 that Defence
Minister Gessler would relocate the Reichswehr captain Franz
Xaver Ritter von Epp, a former Bavarian free corps leader, to
Prussia, Thoma commented: ‘We will neither be ruled nor bullied
by those Jewish swine on the Spree, and if Berlin has not yet been
completely forsaken by God, they will throw Gessler out of the
Reichswehr as quickly as possible.’
With their hate campaigns against the republic and its
representatives, newspapers like the Miesbacher Anzeiger created an
atmosphere that could discharge violently at any moment. On 9
June 1921 Carl Gareis, leader of the USPD fraction in the Bavarian
parliament, was murdered in Munich by an unknown assailant,
who shot him four times. The perpetrator of the next political
murder was quickly identified by the police. On 26 August 1921,
shortly after a judicial investigation had exonerated him of all
charges of tax evasion and illegal flight of capital, Matthias
Erzberger, the former minister of finance, was gunned down by two
members of the ‘Consul’ organization and the Munich ‘Germanic
Order’ while taking a walk in Griesbach in the northern Black
719
Forest. The murderers, Heinrich Tillessen, a lieutenant, and
Heinrich Schulz, a reserve lieutenant, escaped via Munich to
Hungary. They were not sentenced until 1950, and even then
served only two years of their prison terms (which were twelve and
fifteen years, respectively). The leader of the ‘Germanic Order,’
Lieutenant Commander Manfred Killinger, who had ordered the
murder, was acquitted by the Offenburg court of the charge of
abetment to murder.
The commentary in the right-wing press amounted to a
justification of the murder. The Oletzkoer Zeitung, a German
Nationalist newspaper from East Prussia, wrote that the former
minister had been overtaken by the fate that probably all national-
minded Germans had wished upon him. ‘A man who, like
Erzberger, bore the main responsibility for the misfortune of our
Fatherland, was a continual danger to Germany while he remained
alive.’ The like-minded Berliner Lokalanzeiger wrote that any other
country would show ‘limitless understanding’ for conspirators like
the officers who had shot Erzberger. The German Nationalist Kreuz-
Zeitung compared the conspirators to Brutus, William Tell, and
Charlotte Corday, who had killed the Jacobin Marat in 1793, and
accused ‘Erzberger’s present-day eulogists’ of ‘completely forgetting
that the war waged against Erzberger was a war of defence’.
The trade unions and the Majority and Independent Social
Democrats responded to Erzberger’s murder and the right-wing
glorification of violence with large demonstrations, which the KPD
also joined. On 29 August the Reich government, at the behest of
the president, passed an emergency measure on the basis of Article
48, Paragraph 2 of the Weimar constitution, which gave the
720
interior minister the authority to ban all press, assemblies, and
associations hostile to the republic. This provoked a new conflict
with Bavaria, whose administration refused to enforce Gradnauer’s
ban against the Miesbacher Anzeiger, the Münchner Beobachter, and
the Völkischer Beobachter, the central organ of the National Socialist
German Workers’ Party (Nationalsozialistische Deutsche
Arbeiterpartei; NSDAP), founded in 1919.
On 28 September the president issued a second emergency
measure in defence of the republic. This was the result of
negotiations with the Bavarian government, now led by Count
Lerchenfeld, a fairly moderate politician of the BVP who had
replaced Kahr as prime minister the week before. The new
ordinance promised to protect not only ‘representatives of the
republican-democratic form of government’, but also, in
accordance with the wishes of the Bavarian government, ‘all
persons in public life’. The authority to enforce prohibitions and
confiscations in defence of the republic was transferred to the
individual state agencies. In return, the Bavarian government
committed itself to lifting the state of emergency, in effect in
Bavaria since November 1919, by 6 October 1921 at the latest.
At the end of October 1921 another crisis, as superfluous as it
was serious, struck the Wirth cabinet. The cause was the Allied
High Council’s decision regarding the partition of Upper Silesia.
The DDP and—somewhat less forcefully—the Centre insisted that
the Reich government resign immediately in order to demonstrate
to the whole world Germany’s protest against the violation of its
right to national self-determination. The Social Democrats
considered such a step both risky and useless, but they did not
721
prevail. On 22 October Wirth informed the president of his
cabinet’s resignation.
There followed negotiations over the formation of a ‘Grand
Coalition’, a crisis solution that even the Social Democrats now
endorsed. This time, however, the DVP was unwilling, ostensibly
because it doubted the SPD’s commitment to a ‘national defence
front’ on the Upper Silesia question. In reality, however, the
business-friendly party was afraid of being outvoted by the other
parties on controversial matters of tax policy. The DVP’s rejection
induced the DDP to withdraw from governmental responsibility.
The only remaining solution was a ‘black–red’ minority cabinet
under the current chancellor. The president had to threaten his
resignation in order to overcome the Centre Party’s resistance to a
coalition with the SPD alone. The DDP, which no longer considered
itself a coalition partner, nonetheless kept Gessler in the cabinet as
a ‘special minister’. This made the resignation of the government
look like a farce. On 31 January 1922 Wirth transferred the office
of foreign minister, which he himself had intermittently held, to
the former reconstruction minister, Walther Rathenau. After a
three-month interim, the DDP was once again a formal participant
in the government.13
Two weeks before Rathenau assumed office, the Allied High
Council invited Germany to attend an international conference in
Genoa at which the victors and vanquished of the late war were to
discuss the problems of economic reconstruction for the first time.
Soviet Russia was also invited. It seemed logical for Berlin and
Moscow, both ‘have nots’ in international affairs, to coordinate
their policies in advance. While their diplomatic relations had not
722
yet been restored, each country had a trade delegation in the
other’s capital after May 1921. Ideologically speaking, relations
remained tense. In March 1921 the Communist International,
founded two years previously in Moscow, had even tried to start a
revolution in Germany, beginning in the industrial heartland—a
plan Severing’s Prussian police managed to stop in time. However,
the Cominern’s policy and the official policy of the Soviet
government were not the same thing. While the ‘internationalists’
were preparing the way for world revolution, the ‘realists’ were
seeking to strengthen Russia’s position in cooperation with certain
capitalist states, above all Germany.
This was especially the case in military affairs. In September
1921 the Red Army and the Reichswehr began top secret and
increasingly systematic cooperation. The Russians were interested
in superior German technology, and Germany sought Russia’s
support against the restrictions of the Treaty of Versailles,
especially in the areas of air force and poison gas production. The
common antipathy to Poland was also a factor. In 1920, Russia had
suffered defeat in a war against Poland; the Treaty of Riga in
March 1921 had forced it to recognize the Polish eastern border,
which lay 200 to 300 kilometres east of the ‘Curzon line’ set by the
Allies at the end of 1919.
Germany was as unwilling as Russia to accept its territorial
losses to the new Polish national state. At the beginning of
February 1920, on the eve of the Russo-Polish war, General von
Seeckt, chief of the army command, had expressed the view that
only in ‘firm alliance with Great Russia’ could Germany hope to
recuperate its territories lost to Poland and its ‘world power status’.
723
Chancellor Wirth, an active promoter of secret military cooperation
in his capacity as finance minister, shared Seeckt’s opinion.
Throughout 1922, he called for Poland to be crushed and for
Germany and Russia to become neighbors once again.
Ago von Maltzan, the director of the eastern department of the
foreign office, was not only a proponent of the ‘eastern orientation’
like Wirth and Seeckt, but the policy’s actual architect. In the
middle of January 1922 this diplomat from the school of Alfred
von Kiderlen-Waechter negotiated with Karl Radek, the Soviet
expert on Germany, working out the basis of a treaty that
accommodated the Russian desire for closer economic cooperation
with Germany. The plan did not include supervision by an
international syndicate, as the Allies had suggested for the
reconstruction of Russia.
However, the new foreign minister did not adopt this course. In
contrast to Wirth, Seeckt, and Maltzan, Walther Rathenau was
decidedly ‘western’ in orientation, wished to avoid German–
Russian entanglements, and advocated an international economic
consortium. Consequently, the German–Russian negotiations
stalled, and were not revived until the Russian delegation under
Foreign Minister Georgy Chicherin stopped in Berlin at the
beginning of April on the way to Genoa. No treaty was concluded
on this occasion, but there were so many points of convergence
that one seemed possible in the near future.
On 5 April, before Wirth and the German delegation left for
Genoa, President Ebert made his constitutional powers and political
desires very clear to the cabinet. Since he was responsible for
representing the Reich internationally, he insisted that no
724
substantive agreements and commitments be made without his
approval. It turned out differently. In Genoa the German
representatives in the financial commission did succeed on one
point: the Allied experts agreed that the reparations had
contributed to the devaluation of the German currency and must
not be permitted to overburden the country’s economy. At the
same time, however, a disconcerting rumour made the rounds that
separate Allied–Russian negotiations might lead to an agreement at
Germany’s expense. Influenced by these reports, Rathenau finally
submitted to pressure from Maltzan and authorized the latter to
resume talks with the Russians.
As it turned out, there was no danger of a breakthrough in the
Allied–Russian negotiations. Though Maltzan soon learned as
much, he still did everything in his power to bring about a separate
treaty with the Russians. An ambiguously worded telegram
informing the president about the German–Russian talks was
postponed by Maltzan for one day so that Ebert would have no
opportunity to strengthen the already hesitant Rathenau in his
resistance to a Russian treaty.
The decision was made in the night of 15–16 April at the
legendary ‘pyjama party’ in Rathenau’s hotel room. Maltzan
informed those present, including the chancellor, about a phone
conversation in which Chicherin had declared his willingness to
conclude a treaty with the Germans immediately, and on their
terms. Wirth and Maltzan managed to talk Rathenau out of
informing the British prime minister, Lloyd George, of the latest
developments. On the following day, Easter Sunday, the German
delegation travelled to Rapallo, a seaside resort in northern Italy.
725
That evening, Chicherin and Rathenau signed the treaty that bears
this name, which very soon acquired quasi-mythical status. In the
Treaty of Rapallo, Russia and Germany renounced all mutual
claims to compensation for wartime damages, restored full
diplomatic relations with each other, and committed themselves to
a most-favoured-nation clause whereby any commercial advantages
either country accorded to a third nation were automatically
granted to the treaty partner.
The Berlin response to the treaty was mixed, but mostly
positive. The president was—and remained—very upset that the
chancellor and foreign minister had disregarded his instructions,
but he supported the government in public. The Reichstag adopted
the treaty at third reading on 4 July 1922 against a small number
of votes from the DNVP. Among the warning voices was that of the
USPD delegate Rudolf Breitscheid. At the end of April he called the
Treaty of Rapallo the ‘greatest possible disservice to German
interests for the near future’, since it stood to harm the emerging
economic understanding with the West.
And so it happened. The manner in which Germany and Russia
reached their agreement in April 1922 could not but cause the
highest alarm among the Western powers, especially France. It is
uncertain whether, absent Rapallo, the Genoa conference would
have brought substantial progress on the reparations problem.
After Rapallo there could be no hope for progress, at least for a
long time.
And not only that. In a speech in Bar-le-Duc on 24 April, only
one week after the signing of the treaty, Raymond Poincaré, the
prime minister of France, hinted at the possibility of French
726
military intervention. The commander-in-chief of the Allied forces
in the Rhineland, General Degoutte, wrote in a letter to Defence
Minister Maginot on 2 May that the German–Soviet agreement in
Rapallo meant that France could not afford to lose any more time if
it intended to occupy the Ruhr basin. The Treaty of Rapallo was a
relapse into Wilhelmine brinkmanship, and it was guided by forces
that thought in ‘Wilhelmine’ categories in more ways than one.
When Wirth, speaking with Chicherin in Genoa, came out in
support of a ‘restoration of the 1914 borders’, he knew he was also
speaking for large portions of the German ruling class.14
The one reluctant German co-signatory of Rapallo did not live to
see the treaty ratified by the parliament. In the late morning of 24
June 1922, Foreign Minister Walther Rathenau was shot to death
by two men who overtook his automobile as he was driving to the
foreign office from his villa in Grunewald. The murderers, quickly
identified as Erwin Kern, a retired navy lieutenant, and Herman
Fischer, a reserve lieutenant, were caught by the police on 17 July
at Saaleck Castle near Kösen. Kern was shot while fleeing,
whereupon Fischer committed suicide. Both were members of the
‘German National Protective and Defensive League’
(Deutschvölkischer Schutz- und Trutzbund), a 170,000-strong,
militantly anti-Semitic group, and the ‘Consul’ organization, which
had also planned Erzberger’s assassination. The police were able to
catch several other men involved in Rathenau plot, several of
whom came from the same secret society.
The assassination of Rathenau was meant as an attack against
the Erfüllungspolitik and the republic as a whole, and to a certain
extent Rathenau really did stand for everything the murderers
727
hated. He was a critic of the old Germany, who, as a Jew, could not
have become foreign minister without the revolution. He
represented the Erfüllungspolitik vis-À-vis the west without Joseph
Wirth’s ulterior motives with respect to the east. At the same time,
however, he was a product of the Wilhelmine era and a German
patriot, exhorting the Germans to a levée en masse in October 1918
and working to overcome the Versailles system from the summer of
1919. In a sense, Rathenau’s own contradictions made him into a
symbol of the young republic and into a target for the hatred of all
who were out to bring Weimar down through a revolution on the
right. According to the plans of the ‘Consul’ conspirators, the far
left was to play the role of catalyst. In reaction to the assassination,
the Communists would unleash the violence of the proletarian
masses, which would be too great for the weak Weimar
government to deal with. Only a national dictatorship would be
able to restore order.
Rathenau’s murder shook the republic like no other event since
the Kapp–Lüttwitz putsch. But the ‘Consul’ conspirators waited in
vain for the escalation of violence. The Communists joined
Majority and Independent Social Democrats in the massive
demonstrations sponsored by the General German Trade Union
Association, and Chancellor Wirth, after paying tribute to the slain
minister in the Reichstag on 25 June, hurled words at the political
right the latter never forgave: ‘There [on the right] the enemy is
standing, dripping poison into the wounds of a nation. There he
stands, and there can be no question about it—the enemy stands on
the right.’ The minutes record here ‘tumultuous, incessant applause
and clapping in the centre and left, in all the stands. Great,
728
incessant commotion.’
As before, after Erzberger’s assassination in 1921, the president
and the government took legal action against the extreme right,
and once again this led to a serious conflict with Bavaria. Two new
ordinances pursuant to Article 48 were followed by a law ‘in
defence of the Republic’ (Republikschutzgesetz), adopted by the
Reichstag at third reading on 18 July 1922 with the two-thirds
majority required for constitutional amendments. The support of
the DVP made this possible. Actions hostile to the republic, from
insulting the national flag to murdering the republic’s official
representatives, were sanctioned with severe punishments, and a
special ‘state court for the protection of the Republic’ was
established in Leipzig to try such crimes.
In response, Bavaria struck an unprecedented blow against the
federal government. It abolished the new law on 24 July, one day
after it took effect, and replaced it with an ordinance that, while
adopting the material points of the Republikschutzgesetz, transferred
the new court’s jurisdiction to Bavarian courts. The Reich answered
this violation of the constitution by proposing negotiations, which
led to a compromise on 11 August (‘Constitution Day’). To the new
federal court was added a second senate with jurisdiction over
crimes committed in southern Germany and staffed with southern
German judges. The Bavarian government reciprocated on 25
August by abolishing its ordinance of 24 July. The rightist majority
in the Bavarian parliament, considering this compromise an
unacceptable retreat, forced Count Lerchenfeld to resign on 2
November. He was succeeded a week later by Eugen Ritter von
Knilling, who was far more sympathetic to the ‘Patriotic Leagues’
729
and Adolf Hitler’s National Socialists than Lerchenfeld had been.
The new federal law was a good deal less effective than its
proponents had envisaged. The authoritarian-minded judiciary had
no intention of strictly enforcing the new provisions. And when it
did do so, then sooner against offenders from the left than from the
right. For example, a Communist who used the expression ‘republic
of thieves’ (Räuberrepublik) was sentenced to four weeks in prison,
while a defendant from the völkisch scene got off with a fine of 70
marks for ‘Jew republic’ (Judenrepublik). In Bavarian courts, the
term Saurepublik (‘damned republic’, literally ‘sow republic’) was
not considered a term of opprobrium. Saying ‘black-red-mustard’
(Schwarz-Rot-Mostrich) or ‘black-red-eggyolk-yellow’ (Schwarz-Rot-
Hühnereigelb) for ‘black-red-gold’ was also not usually punished.
One citizen who used the expression ‘black-red-shit’ (Schwarz-Rot-
Scheisse) was first acquitted, then made to pay a fine of 30 marks
after the case was appealed.
The religious community, too, continued to have strong feelings
against the Weimar Republic. This was especially true of
committed Protestants, many of whom had not come to terms with
the fall of the monarchy. The official church attitude found
expression in a popular rhyme: ‘Die Kirche ist politisch neutral, aber
sie wählt deutschnational’ (‘The Church is politically neutral, but it
votes German-National’). Like the DNVP, the German Evangelical
Church Committee (Deutscher Evangelischer Kirchenausschuss)
condemned the 1922 murder of Rathenau as a ‘heinous crime’, but
it saw the Allied powers as the real perpetrators: ‘We accuse our
enemies, whose blindness has cast our people into disgrace and
misery, out of which arise all of the demons of the abyss.’
730
The Catholic Church had not associated itself as intimately with
the Hohenzollern state as the Evangelical church. Nonetheless, the
German Catholic Congress in Munich at the end of 1922 showed
how strong anti-republican attitudes still were among Catholics,
too. Cardinal Faulhaber, the archbishop of Munich and Freising,
condemned the November 1918 revolution as ‘perjury and high
treason’, calling it an ‘atrocity’ that was not to be ‘sanctified’
simply because it had brought Catholics a few successes. When the
president of the Catholic Congress, Cologne mayor Konrad
Adenauer, distanced himself from Faulhaber in his closing address,
the division within German Catholicism was plain to see.
The republican forces were well aware that anti-republican
suspicions and antipathies could not be overcome by bans and
punishments. For this reason, the president and government sought
to promote and consolidate pro-republican attitudes through a
‘positive’ campaign. But success was modest here, too. ‘Rightist’
governments in the Länder torpedoed Interior Minister Adolf
Köster’s plan to make 11 August, the day the Weimar constitution
was signed, into a national holiday. On another issue Köster had
better success. It was his doing that the ‘Deutschlandlied’ was
proclaimed the German national anthem by President Ebert on 11
August 1922. The words of Hoffmann von Fallersleben were not to
serve as an expression of national arrogance, Ebert announced. ‘But
as once the poet did, so too do we today love “Germany above all”.
In fulfilment of his longing, let the song of unity and right and
freedom under the banners of black, red, and gold be the festive
expression of our patriotic feelings.’
Ebert’s hope was not realized. The ‘Deutschlandlied’ did not
731
make the national colours any more popular. To sing the first
stanza was not sufficient to turn one into a republican. The
republicans among the working class, for their part, found it
difficult to join in singing a song that, despite its democratic, black-
red-gold origins, was now more often sung on the black-white-red
right than by the devotees of Weimar. ‘Deutschland, Deutschland
über alles’ had become the veritable signature tune of ‘national
Germany’ ever since, on 10 November 1914, young volunteers—
with this song on their lips, according to the official legend—had
stormed enemy lines and fallen near Langemarck in Flanders. For
the victors of the war, 11 November was the day of
commemoration for the signing of the armistice in 1918. On the
same day, to the strains of the ‘Deutschlandlied’, the German right
cultivated the myth of Langemarck, a patriotic cult of sacrifice with
a strong admixture of desire to wipe away the disgrace of 1918.15
Attempts to counter the kind of fanatical anti-Semitism that
found expression in the campaign against and murder of the Jewish
Walther Rathenau had virtually no chance of success. Three weeks
after the murder, the Prussian government found it necessary to
ban a general student congress in Marburg, since statements by the
organizers seemed to indicate that public justifications for the
assassination were in the offing. The völkisch camp of the German
Students’ Association (Deutsche Studentenschaft), with the Deutscher
Hochschulring as its main organization, went to Bavarian Würzburg
instead. The constitution drawn up there extended membership in
the student body to German-Austrians and Germans from abroad
‘of German heritage and mother tongue’. This was the first step
along the way to an ‘Aryan paragraph’, which was already in effect
732
in Austria.
The far right considered the Jews the authors of Germany’s
military defeat, believing they had systematically undermined the
morale of the German workers with Marxist or Bolshevist ideas or
had enriched themselves at the expense of the German people. The
Jews were represented as instigators and beneficiaries of
revolution, inflation, and the Erfüllungspolitik. They became, in
effect, scapegoats for everything Germany suffered—or believed to
be suffering—after November 1918. Rosa Luxemburg, Paul Levi,
Kurt Eisner, Hugo Haase, Rudolf Hilferding, Eduard Bernstein, Otto
Landsberg, Paul Hirsch, Ernst Heilmann, Hugo Preuss, Eugen
Schiffer, Theodor Wolff, and Walther Rathenau—in the eyes of the
fanatical anti-Semites, it made no difference whether these
politicians and publicists were radical or moderate, Marxists or
liberals, unbaptized or baptized. They were Jews, and thus
enemies.
Anti-Semitism was especially virulent in higher education, since
many academics and students perceived Jews mainly as
competitors for prestigious positions in society. The fact that the
‘Marxist’ workers’ movement had advanced to political power in
1918 was taken by many in the academy as a personal insult. They
saw their claim to the leadership of the country called into
question by people whom they believed to be lacking in the
necessary intellectual and moral aptitude. The role of Jews in the
politics of the left was enough to turn this feeling of downgraded
status and prestige in an anti-Semitic direction. The völkisch
students and young academics saw themselves in the tradition of
the wars of liberation. In Fichte, especially, they found what they
733
were looking for: the idea of the eternal German Volk, which they
promptly put to use against the Jews as agents of a ‘foreign’ ethnic
culture and against the Weimar state, which they saw as shaped by
the Jews.
The favourite theme of anti-Semitic propaganda was the so-
called Ostjuden, the ‘Eastern Jews’. Names of Jews from Russia or
Poland who had played a prominent role in the revolution of 1918–
19 were as easy to learn as the names of eastern Jewish
‘speculators’ who profited from inflation. In contrast to the German
Jews, who had long been settled and integrated in German society,
the orthodox eastern Jews were of distinctly foreign appearance
and formed, on account of their general destitution, a social
problem group. What the anti-Semites studiously avoided
mentioning was that the ‘question of the Eastern Jews’ was in good
part the result of OHL policies in the occupied territories of Russian
Poland in 1914. First largely stripping them of their material basis
of existence, the Army High Command had then recruited Jews as
labour for the German armaments industry. Some 35,000 came to
Germany in this way, more or less under duress. About the same
number were taken as prisoners of war or caught in Germany by
the sudden outbreak of hostilities and interned. Added to the
80,000 eastern Jews living in Germany before 1914, the total
figure at war’s end was about 150,000.
Most of the eastern Jews working in the armaments industry lost
their jobs after the war was over. But they could not easily return
to their homes. The new states in eastern central Europe showed
little initial willingness to take in large numbers of unemployed
Jews. This was especially true of Poland, which had a strong
734
tradition of anti-Semitism. In 1920–1 a considerable number—
some 30,000—left for America, where eastern Jews had long found
a new home. This emigration continued in the following years. In
1925, the year of the largest eastern Jewish immigration, there
were about 108,000 living in Germany. That was 30,000 more than
in 1910, but certainly no justification for the anti-Semitic claim
that Germany was being flooded with eastern Jews.16
Even under the emperors, anti-Semitic parties and organizations
had never held a monopoly on anti-Jewish sentiments. The German
Conservative party had been unequivocally anti-Jewish from the
moment of its inception and had made anti-Semitism an official
part of its ‘Tivoli platform’ in 1892. The German Nationalists
consciously adopted this tradition and offered a political home
even to fanatical anti-Semites like the delegates Wilhelm Henning,
Reinhold Wulle, and Albrecht Graefe. One of these, Henning,
played a prominent role in the political and press campaign against
Walther Rathenau. In the June 1922 edition of the Konservative
Monatsschrift, the organ of the völkisch wing of the DNVP, Henning
an article claiming, among other things, that ‘German honour’ was
‘no barter commodity for international Jew hands’, and that
Rathenau and his agents would be called to account.
After the murder of the foreign minister, the party leadership
under former Prussian finance minister Oskar Hergt thought it
advisable to make a clear distinction between the DNVP and the
extremist völkisch elements on the right. Henning’s removal from
the party was calculated to demonstrate to the other bourgeois
parties the DNVP’s capacity to govern. The German Nationalist
Reichstag fraction, which was given the final decision, decided it
735
was enough to expel Henning from its own ranks; to exclude him
from the party was not necessary. Shortly thereafter, Henning took
this step himself, along with Wulle und Graefe. In September 1922
they founded the German Ethnic Cooperative Union
(Deutschvölkische Arbeitsgemeinschaft), which gave rise to the
German Ethnic Freedom Party (Deutschvölkische Freiheitspartei) in
December.
One of the new party’s strongest bastions was Munich, where
the district chapter of the DNVP merged with it. The Bavarian
capital provided a particularly favourable climate for the völkisch
nationalists, though they also found a competitor there whose anti-
Semitism was not to be outdone: the National Socialist German
Workers’ Party, led by Adolf Hitler. For the still very anti-Semitic
DNVP, the departure of the radically völkisch elements had more
advantages than disadvantages. Since the autumn of 1922, the
German Nationalists had come a good distance towards their most
important goal, participation in a bourgeois coalition that could
govern without—and against—the Social Democrats.
On the left, too, Rathenau’s assassination led to a realignment of
political forces. In July 1922 the Reichstag fractions of the Majority
and Independent Social Democrats formed a cooperative union. In
September they reunited as one party. The USPD of 1922 was no
longer the USPD of 1917. At the party congress in Halle in October
1920, a clear majority of the delegates had advocated joining the
Communist International and merging with the KPD. To join the
‘Third International’ meant submitting to the ‘twenty one
conditions’ the Comintern had set down at its Second World
Congress in Moscow in summer 1920. According to these, all
736
member parties were required to recognize the Bolshevist party
type and method of obtaining political power. The moderate
minority of the USPD, led by Wilhelm Dittmann and Artur Crispien
along with the party intellectuals Rudolf Hilferding and Rudolf
Breitscheid, refused to make such a radical break with the
traditions of democratic socialism and remained independent.
This did not mean that a reunification with the Majority party
would be automatic, however, especially since the MSPD took what
many Independents considered a step to the right the following
year. In its 1921 ‘Görlitz platform’, co-authored by Eduard
Bernstein, the MSPD defined itself as a ‘party of the working people
in the city and country’ that aspired to fundamental societal
reforms, was open to the middle classes, and no longer considered
socialism the result of a natural and ineluctable economic
development, but as a question of political will. It took the rise of
right-wing radicalism and, finally, Rathenau’s murder to convince
the two Social Democratic parties that they could no longer afford
to be divided.
The 1922 reunification considerably strengthened the Social
Democrats, especially in parliament. But it had its disadvantages.
The MSPD’s reformist Görlitz platform, designed for a leftist
people’s party, was invalid after only one year. The new
‘Heidelberg platform’ of September 1925, mainly the work of
Hilferding, began with the old Marxist argument about the
unavoidable and progressive destruction of small by big business.
This effectively closed the party to the middle classes. The re-
ideologization was accompanied by a more dogmatic party attitude
towards participation in government. The union of the
737
parliamentary fractions meant that ‘rightist’ notions of a Grand
Coalition had to be abandoned, at least for the time being. Most of
the USPD delegates, as well as the Majority left wing, would not
have agreed to a coalition with the business-friendly DVP.
Nonetheless, Stresemann’s party became a tacit participant in
the Wirth administration. On 19 July 1922, five days after the
union of the Social Democratic Reichstag fractions, the DVP, DDP,
and Centre formed a ‘cooperative union of the constitutional
centre’ (Arbeitsgemeinschaft der verfassungstreuen Mitte), which was
intended to counter the new weight of the Social Democrats. With
its endorsement of the Republikschutzgesetz the day before, the DVP
had positioned itself in the political centre. On 24 October it
facilitated the two-thirds majority necessary to amend the
constitution and extend the term in office of (still only provisional)
President Ebert until 30 June 1925. This step made the direct
popular election set for December 1922 superfluous—an election
the two liberal parties wished to avoid, fearing the consequences
for public order.
There were good, indeed compelling reasons for a Grand
Coalition in the autumn of 1922. Rathenau’s murder destroyed in
one blow all remaining confidence in the currency. Germans and
non-Germans alike dumped their credits in marks on the market in
a panic. The flight of capital took on gigantic proportions. At about
the same time, the inflationary boom that had protected the Reich
from the world economic crisis in the early 1920s came to an end.
The interest in cheap German imports decreased in the same
measure as German domestic industries regained strength. German
exports lost the ‘bonus’ they had enjoyed when global production
738
decreased after 1920. Inflation, which turned into hyperinflation in
the autumn of 1922, lost its economic ‘attraction’. Objectively
speaking, this increased the chances for a currency reform.
Politically, however, a reform was only possible through close
cooperation between business and organized labour, between the
moderate bourgeois parties and the Social Democrats.
This realization was far from widespread among the business
community in autumn 1922. Hugo Stinnes, who had exploited
inflation to erect a veritable industrial empire, wanted to
strengthen the currency by making German workers work two
hours longer per day without increased pay for ten to fifteen years.
The proposals of the big industrialists, submitted to the federal
economic council on 9 November, the fourth anniversary of the
revolution, outraged the trade unions, Social Democrats, and
Communists. Since Stinnes sat in the Reichstag for the DVP, his
statements could also be taken as an indication that the heavy
industrialist wing of the German People’s Party would not be
amenable to Wirth’s and the bourgeois centre’s efforts towards a
Grand Coalition.
Yet Stinnes did not speak for all of German industry, nor even
all of the DVP. The party head, Stresemann, was convinced of the
necessity of reaching an understanding with the moderate forces in
the workers’ movement. This made it easier for Wirth, on 26
October, to talk the governing coalition and the DVP into forming a
commission to look into a common platform for pressing economic
questions, above all that of reparations. One of the architects of the
Central Cooperative Union of November 1918, Hans von Raumer,
an electrical industrialist, joined the commission for the DVP. The
739
SPD was represented by Rudolf Hilferding. Internationally
recognized as one of the leading Marxist theoreticians ever since
the publication of his 1910 book Das Finanzkapital, Hilferding had
been a member of the USPD before the reunification of the parties
in September 1922.
The commission achieved what seemed like a ‘miracle’. The
experts agreed to a number of economic and fiscal measures, and
the government was able to use these in its note on reparations to
the Allies on 13 November 1922. Reductions in state expenditures
and revenue increases were to help balance the budget. The real
sensation, however, was one of the proposals for increasing
productivity: a reform of the legislation governing working hours
‘establishing the eight-hour day as normal working day and
permitting legally restricted exceptions, which are to be worked
out by contract or by the administration’. Without fundamentally
abandoning one of the most important social achievements of
November 1918, the commission was nonetheless suggesting, at
least for certain branches of the economy, a temporary increase in
working hours in order to make possible the reform of the
currency, the economic reconstruction of the country, and the
peaceful settling of differences with Germany’s neighbors.
The Wirth government’s note to the Allies was based primarily
on the commission’s suggestions. In response to an Allied demand,
it also promised, for the first time, large-scale measures by the
federal bank (Reichsbank) in support of the mark: if Germany were
to receive an international loan of 500 gold marks, the federal bank
would provide matching funds to stabilize the currency. The note
was endorsed not only by the heads of the SPD, Centre, and DDP
740
coalition fractions, but also by the representatives of the DVP. The
foundations of a Grand Coalition thus seemed to be in place. But
the appearance was deceptive. On the following day, 14 November,
the fraction of the United Social Democratic Party of Germany
(Vereinigte Sozialdemokratische Partei Deutschlands, the name of the
party after its September unification congress in Nuremberg) voted
against a Grand Coalition by three-quarters majority. Otto Braun,
the prime minister of Prussia, would have supported such an
alliance, but his was a lost cause. The party leadership did not want
to risk another split with the former Independents, most of whom,
in contrast to Hilferding, were still strictly opposed to joining the
DVP, the ‘business party’.
Joseph Wirth, in accordance with an agreement with the centre
parties, resigned from the chancellorship on the same day. On 22
November the president named as his successor Wilhelm Cuno, the
general director of the Hamburg-America Line. Cuno, a Catholic
born in 1876 in Thuringian Suhl, stood clearly on the centre-right,
though he did not belong to a political party. In appointing him,
Ebert hoped that an experienced economics expert at the head of
the cabinet would bring the German business community closer to
the republic as well as make a good impression internationally. In
addition to Cuno, four further non-aligned politicians joined the
new administration, including the former mayor of Essen, Hans
Luther, as minister of agriculture and the former quartermaster
general, Wilhelm Groener, as minister of transport, which position
he had already held under Fehrenbach and Wirth. The other
ministers were members of the Centre, BVP, DDP, and DVP.
None of the republican governments hitherto had so resembled a
741
cabinet of imperial bureaucrats. And never before had the selection
of the chancellor been so exclusively the decision of the president.
It would be only a slight exaggeration to say that the Cuno
administration was a presidential cabinet in disguise. This relapse
into the authoritarian state was not only Ebert’s mistake. The main
responsibility lay with the real governmental party of the Weimar
Republic, the Social Democratic Party. Fearing for its internal
unity, it turned away from a parliamentary solution to the crisis
and thus made possible the presidential pathway.17
Under the Cuno government, Germany began the year 1923, in the
course of which the republic was to totter more than once on the
brink of the abyss. On 11 January French and Belgian troops
occupied the Ruhr district. The reason was mere pretence.
According to the Allied Reparations Commission, Germany had
failed to fulfil its obligations to deliver wood, telegraph poles, and
coal. This neglect was the fault of the previous administration
under Wirth, which after August 1922 had been following the
popular maxim ‘Bread first, then reparations’. This was foolish, for
France had been waiting for an opportunity to occupy the Ruhr
district ever since the Treaty of Rapallo. The purpose of the
occupation was to obtain for France the kind of security vis-À-vis
its eastern neighbor it had not been able to achieve in the Treaty of
Versailles, owing to British and American resistance. Behind this
742
security interest, however, lay more. The French wished to
underscore their claim to hegemony over the European continent.
Germany responded to this act of aggression—for such the Ruhr
invasion was—with a policy of ‘passive resistance’, that is, by
refusing to follow the orders of the occupiers. The Cuno minority
government, which hitherto had been able to rule only thanks to
the acquiescence of the SPD, found a large Reichstag majority for
this approach, as well as the active support of the unions. Only the
extreme left and right refused to join the unified national front. On
22 January the Communists promulgated their solution: ‘Strike
Poincaré and Cuno on the Ruhr and on the Spree!’ In the weeks
following, however, out of consideration for the ‘anti-imperialist’,
anti-French policy of the Union of Soviet Socialist Republics
(officially founded in December 1922), the German Communists
placed greater emphasis on the external enemy. The attitude of the
National Socialists was more extreme. On 11 January Hitler
declared to his followers in the Zirkus Krone in Munich that the
proper response was not ‘Down with France!’ but ‘Down with the
November criminals!’ He called the National Socialists the
‘avenging army of the Fatherland’.18
The policy of passive resistance was generally successful until
March 1923. Because of the German boycott, the French and
Belgians were unable to force reparations payments. After that
date, however, the occupiers began to confiscate coal mines and
coking plants and to take over the rail system. The Reich not only
had to keep paying the salaries of the rail employees, who were
expelled from the occupied zone; it also gave millions in credit to
the coal mining, iron, and steel industries for the continued
743
payment of wages after the facilities were shut down. Financially
speaking, the Ruhr occupation became a bottomless pit.
Hyperinflation spiralled out of control. The mark’s foreign value,
which the federal bank had temporarily stabilized at about 21,000
to the dollar through the sale of gold reserves and foreign exchange
from February through April 1923, fell in May to 48,000 and in
June to 110,000 marks.
As the failure of passive resistance became clearer, the radical
right went over to active resistance in the form of sabotage. In
March and April 1923 a commando troop under Heinz Hauenstein,
a former free corps leader, blew up several rail facilities in the
occupied area. One of the commando’s lower leaders, Albert Leo
Schlageter, a National Socialist, was apprehended by the French
criminal police in Essen on 2 April. On 9 May a French military
tribunal in Düsseldorf convicted him of espionage and sabotage and
condemned him to death. The sentence was carried out by firing
squad on 26 May.
Schlageter’s execution set off a storm of protest in Germany that
echoed even in distant Moscow. In a speech before the expanded
executive of the Communist International on 20 June 1923, Karl
Radek, the organization’s Germany expert, called the ‘fascist’
Schlageter a ‘martyr of German nationalism’ and ‘courageous
soldier of the counter-revolution’, who deserved ‘honest and manly
recognition from us, soldiers of the revolution’.
If the circles of the German fascists who sincerely wish to serve the
German people will not understand the meaning of Schlageter’s
fate, then Schlageter has fallen in vain, and they should write on
744
his memorial: ‘The Wanderer into the Void’… We desire to do
everything so that men like Schlageter, who were prepared to die
for a general cause, do not become wanderers into the void, but
wanderers into a better future for all of humanity; so that they do
not shed their hot, selfless blood for the profits of the coal and iron
barons, but for the cause of the great working German people, who
are a member of the family of nations fighting for their liberty.
745
nationalist right in summer 1923. Despite remarkable rhetorical
concessions to anti-Semitism, however, the campaign generated
few practical results.
Far greater was the KPD’s success among the workers. Though
not the actual initiators of the ‘wildcat strikes’ in the Ruhr district
in mid-May, the Communists exploited them effectively, gaining
strong support in the summer 1923 elections on all levels—works
committees, trade unions, municipalities, and state parliaments.
The membership of the KPD grew from 225,000 in September 1922
to 295,000 in September 1923. The number of local branches
increased from some 2,500 to more than 3,300. By August 1923 a
political explosion seemed to be in the offing. Deteriorating social
conditions had created an atmosphere of desperation, which found
expression in the so-called ‘Cuno strikes’. The Free Trade Unions
sought to protect the shops of the federal press, where money was
printed, from the industrial action. But in vain. For one day, 10
August 1923, the press was shut down, and the resultant shortage
of paper money was immediately felt.
The Social Democrats had been tolerating the Cuno
administration up to this point. They did so, even though it stood
further right than any other Weimar government, maintained
contacts among the radical free corps, made no serious effort
towards a diplomatic resolution of the Ruhr conflict, and undertook
nothing to counter the growing destitution of the masses. In mid-
April 1923 Theodor Leipart, president of the General German Trade
Union Association, demanded the formation of a Grand Coalition, if
necessary under Stresemann as chancellor. But the Social
Democrats refused. With left-wing resistance to cooperation with
746
the DVP still so strong, the party leadership saw no alternative to
continuing their toleration of the rightist cabinet. What is more,
they feared that if the SPD were to assume governmental
responsibility at the height of the crisis and break off the
catastrophic passive resistance policy, ‘nationalist’ Germany would
once again accuse the party of ‘back-stabbing’.
It took the ‘Cuno strikes’ to convince the SPD party leaders that
further toleration of the government was not a lesser, but a greater
evil than a Grand Coalition. This remedy, long overdue, came
together within a few short days, since not only the moderate
bourgeois parties, the Centre and the DDP, but also powerful voices
within the DVP and the business community had come to regard
the policies of the minority cabinet as mistaken and dangerous. In
its negotiations with the government parties, the SPD pushed
through a number of measures, including the speedy containment
of inflation, preparation of a gold currency, separation of the
Reichswehr from all illegal organizations, ‘foreign policy activity
towards a solution of the reparations question while maintaining
the complete unity of the nation and the sovereignty of the
republic’, and an application for German membership in the League
of Nations. A Stresemann chancellorship, long in public discussion,
was acceptable to the Social Democrats not least because they
themselves, for domestic political reasons, were intent on avoiding
the top position. Though they formed the strongest party, they
contented themselves with the ministries of finance, economics,
justice, and the interior. The remaining cabinet members came
from the ranks of the DVP, Centre, and DDP, or were non-aligned,
like the agriculture minister, Luther, who remained in office.
747
On 13 August, the day after Cuno’s resignation, President Ebert
appointed Gustav Stresemann chancellor and foreign minister. The
following day, the new head of government won the vote of
confidence from the Reichstag. A good third of the DVP and SPD
delegates did not show up to vote, a clear sign that the Grand
Coalition was still a highly controversial matter in the two wing
parties. The political right, in Bavaria no less than in the occupied
territories, was outraged by the SPD’s return to government. Two
cabinet members were targets of particular hostility: Rudolf
Hilferding, the finance minister, since he was a Jew, and Gustav
Radbruch, the minister of justice, who, holding the same office
under Wirth, was seen as the embodiment of the hated
Republikschutzgesetz. For the working class, on the other hand, the
formation of the Grand Coalition had a calming effect. The ‘Cuno
strikes’ subsided. There was no longer any immediate danger of a
revolutionary situation in Germany.19
The leaders of the Third International were of a different
opinion. In the middle of August, under the impact of the ‘Cuno
strikes’, Grigory Zinoviev, the general secretary of the Comintern,
called on the KPD to ready itself for the impending revolutionary
crisis. On 23 August the politburo of the Russian Communist Party
held a secret meeting. Against a hesitant party secretary Stalin,
Zinoviev, Radek, and the people’s commissar for defence, Leon
Trotsky, pushed through a resolution to set up a special committee
with the task of systematically preparing for a communist
revolution in Germany.
In September came the final decision for the ‘German October’,
which Trotsky wished to set for 9 November 1923, the fifth
748
anniversary of the German revolution. On 1 October Zinoviev
directed the KPD leadership to involve the party as much as
possible in the minority government of Saxony under the leftist
Social Democrat Erich Zeigner, which the Communists had been
tolerating since March. The next step was to be the arming of the
Saxon proletariat. Saxony would be the starting point of the
German revolution, a civil war that would end with the triumph of
the Communists over the fascists and the bourgeois republic.
While the Communists were planning the revolution, the
political crisis in Germany came to a head. On 26 September, after
long hesitation, the president and government announced the end
of the policy of passive resistance. The Bavarian government
responded that same day by declaring a state of emergency and
transferring executive power to the president of Upper Bavaria,
Gustav Ritter von Kahr. To this the Reich reacted on 26 September
with its own decree declaring a state of emergency in all of
Germany and transferring the executive authority to the minister of
defence, who could delegate it in turn to the military commanders.
Legally speaking, the Bavarian government was obligated to annul
its measures at the behest of the Reich president or parliament. But
Stresemann and the bourgeois ministers in his cabinet, believing
that Bavaria would not obey such a request, thought it better to
avoid a trip to Munich in the first place.
Shortly thereafter, the weakness of the federal government
became even more clearly manifest. On 27 September the
Völkischer Beobachter, the mouthpiece of the NSDAP, directed sharp
anti-Semitic attacks against ‘dictators Stresemann and Seeckt’, since
the latter was married to a ‘half-Jewess’, the former to a ‘Jewess’.
749
Defence Minister Gessler ordered Kahr to ban the newspaper. Kahr
refused, and he was backed up by the commander of the federal
troops in Bavaria, General von Lossow. This was a clear case of
refusal to obey orders. Nonetheless Seeckt, the head of the
Reichswehr, was as little inclined as he had been in the spring of
1921 to order federal troops to fire on their own. Instead, he
prepared himself for a role on the federal level similar to Kahr’s in
Bavaria, and there were many who supported him. A ‘national
dictatorship’ under a ‘directory’ led by Seeckt—this was the
demand of prominent heavy industrialists like Hugo Stinnes and all
those whose political home was the German National People’s
Party.
Even Ebert and Stresemann, in their conversation on 22
September (Defence Minister Gessler and Sollmann, the Social
Democratic interior minister, were also present), discussed the
possibility of a temporary Seeckt dictatorship as a last resort to
protect the unity of the Reich against Bavarian particularism and
the separatist aspirations in the Rhineland. Another kind of
dictatorship was proposed at the cabinet meeting on 30 September
by Labour Minister Brauns, a Centre politician, and Finance
Minister Hilferding: an ‘enabling act’ (Ermächtigungsgesetz) allowing
the government to take whatever financial and political steps were
necessary. One such necessary step, in the view of both, was the
extension of the working day, a measure supported by business but
opposed by the Free Trade Unions.
The SPD, however, was not prepared to countenance the kind of
executive authority over working hours Hilferding and Brauns
desired. Hermann Müller, the party head, made this clear in a
750
meeting of the coalition party leaders on 2 October. The
governmental right wing was far less ceremonious in its rejection of
the Stresemann cabinet’s policies. At that same meeting Ernst
Scholz, head of the DVP fraction, demanded the complete
elimination of the eight-hour day, a ‘break with France’, and the
inclusion of the German Nationalists in the Grand Coalition. He
was acting as the mouthpiece of Stinnes, who sought nothing less
than to bring down Stresemann and erect a ‘national dictatorship’.
The next day saw the failure of final attempts to arrive at a
compromise on working hours, since the Social Democratic
Reichstag fraction refused to do battle with the Free Trade Unions.
With no options left, Stresemann handed in his resignation.
By 6 October, however, the chancellor was once again Gustav
Stresemann, and once again he presided over a Grand Coalition
cabinet. The rapid end to the government crisis was the work of
President Ebert, who charged Stresemann with the task of forming
a new government immediately after the old cabinet’s resignation.
In addition, the majority of the DVP refused to uphold the hard-line
policy of its industry-dominated right wing. The decisive
breakthrough came in the night of 5–6 October. The party leaders
agreed to the compromise formula on working hours their experts
had worked out eleven months previously, on 13 November 1922:
the eight-hour day was to be fundamentally preserved, but could
be extended by contractual agreement or legislation.
At this juncture the SPD, too, was ready to endorse an enabling
act that did not affect working hours and that would remain in
force only for the duration of the current coalition. On 13 October
the Reichstag passed this law with the necessary constitutional
751
two-thirds majority, and it formed the basis of ordinances dealing
with unemployment assistance, staff cuts in public services, and
introducing compulsory government mediation in wage
disagreements. The latter measure, which made the state the final
arbiter of industrial action, brought about what Hilferding
summarized in 1927 in the concept of the ‘political wage’ and
celebrated as the expression of a higher, ‘organized capitalism’,
indeed as a step towards socialism: the veritable abolition of wage
autonomy and market forces. The parallels to Article 48 were
obvious. Just as the parties could shift responsibility ‘upstairs’ to
the president in difficult situations, the wage partners could now
relegate their conflicts to the minister of labour. This was the first
time changes of such magnitude had been initiated in Germany by
government ordinance.
While the fate of the Grand Coalition was being decided in
Berlin, rightistauthoritarian forces and Communists were working
for regime change. In Bavaria, Hitler had himself elected on 25
September as the leader of the ‘German Combat League’ (Deutscher
Kampfbund), a new umbrella organization of the ‘Patriotic Leagues’.
Four days later, Kahr abolished the enforcement of the
Republikschutzgesetz. Another measure by the general state
commissar was even more clearly intended to attract the support of
the National Socialists: considerable numbers of eastern Jews were
deported from Bavaria after the middle of October. However,
Kahr’s heaviest blow against the Reich came on 20 October, when
Defence Minister Gessler ordered the long-overdue dismissal of von
Lossow, the Munich district commander. Kahr promptly appointed
him commander-in-chief of the Bavarian forces and declared the
752
Seventh Division of the Reichswehr under the authority of the Free
State.
Neither Kahr, nor Lossow, nor their ally, Colonel von Seisser,
commander of the Bavarian police, had any intention of seceding
from the Reich. Rather, the Bavarian triumvirate wished to model
the Reich after Bavaria. What they had in mind was a ‘march on
Berlin’ culminating in the establishment of a ‘national dictatorship’.
The National Socialists were permitted to participate in the march.
The role of the German Mussolini was not intended for Hitler,
however, but reserved for Kahr and, later, on the federal level, for
a man of comparable outlook. Seeckt was considered a candidate,
even after the events of 20 October. Yet despite his sympathies for
Kahr’s political vision, the chief of the army command was a
legalist. It was improbable that he would launch an attack against
the expressed will of the Reich president.
The main theatre of Communist activities was central Germany.
On 10 October the KPD was able to report that it had carried out
the orders of the general secretary of the Comintern issued nine
days before: the Communists had entered Erich Zeigner’s Saxon
government, where they assumed the posts of the minister of
finance, minister of economics, and, in the person of party head
Heinrich Brandler, director of the state chancellery. In Thuringia,
too, an SPD–KPD coalition government under the Social Democrat
August Frölich was formed on 16 October, the Communists
acquiring the economics and justice ministries.
The Dresden and Weimar governments of the unified leftist front
were formed legally, and both were endorsed by parliamentary
majorities. The Zeigner and Frölich cabinets undertook no steps
753
that could be considered hostile to the Reich. But nobody in Berlin,
including the ruling Social Democrats, had any doubt about the
intention of the Communists to use Saxony and Thuringia to launch
the struggle for power in Germany. The Reich reacted accordingly.
On 13 October the Saxon district commander, General Müller, who
from 27 September held the executive power, banned the
paramilitary ‘Proletarian Hundreds’ (Proletarische Hundertschaften)
of the KPD. On 16 October, after consultation with Defence
Minister Gessler, Müller announced that the Saxon police was to be
immediately placed under the command of the Reichswehr. This
move stripped the Dresden government of its only instrument of
power.
Until 21 October, the danger of a communist uprising in Saxony
was quite real. The KPD had planned for this day a workers’
conference in Chemnitz that was supposed to call a general strike
and give the signal for the uprising. But when Brandler, speaking at
this conference, demanded the immediate commencement of the
general strike in response to the dictatorship of the Reichswehr, he
met with no agreement. The Social Democratic minister of labour,
Graupe, threatened to desert the meeting with his party colleagues
if the Communists persisted in this direction. He met with no
resistance. The timetable for the ‘German October’ was thus
frustrated, and with it the attempt to repeat in Germany the
example set in Russia by the Bolsheviks in October (and November)
1917. Only Hamburg witnessed a putsch-like communist rebellion.
After three days of bloodshed, the police were once again in
control of the Hanse city by 25 October.
At about this same time, between 21 and 27 October, the
754
Reichswehr took control throughout Saxony. Armed clashes cost
lives in several cities. On 27 October Stresemann sent Zeigner an
ultimatum. Pointing to the revolutionary propaganda of
Communist members of the administration, including Brandler, he
demanded that the Saxon prime minister form a government
without Communists. If Zeigner refused, Stresemann would appoint
a commissar with the task of governing Saxony until a
constitutional government was put in place. Zeigner refused.
The process of federal executive action began the next day.
Invoking Article 48 of the Weimar constitution, the Reich president
empowered the chancellor to remove the members of the Saxon
government from office and to entrust other persons with the
business of state. Stresemann then appointed the DVP Reichstag
delegate Karl Rudolf Heinze, former federal minister of justice, as
the commissar for Saxony. On the afternoon of 29 October, the
Reichswehr expelled the Saxon ministers from the government
buildings. Zeigner resigned the next day. In the presence of Otto
Wels, head of the Social Democratic Party, and Wilhelm Dittmann,
a member of the party executive, the SPD fraction of the Saxon
parliament nominated Alfred Fellisch, former economics minister,
to lead a Social Democratic minority cabinet. The DDP promised to
tolerate the new government. On 31 October, the Saxon parliament
elected Fellisch as Zeigner’s successor, whereupon President Ebert,
at Stresemann’s request, withdrew Heinze as federal commissar.
The federal executive action against Saxony proceeded
differently from what the Reichswehr had had in mind.
Stresemann’s solution was a state of emergency in its most civil
form, preserving the primacy of politics. Nonetheless the Social
755
Democrats, though they had fundamentally endorsed the
chancellor’s 27 October ultimatum, raised serious objections after
the fact. Although they were critical of their Saxon and Thuringian
comrades, what really upset both the leadership and the rank and
file of the SPD was the discrepancy between the treatment of
Saxony and the treatment of Bavaria. In a meeting of the Reichstag
fraction on 31 October, Stresemann’s critics prevailed. They now
issued an ultimatum in their own right, demanding that the Reich
government lift the military state of emergency, declare the
behaviour of the Bavarian leaders unconstitutional, and
immediately undertake the necessary steps against the Free State.
The bourgeois ministers, convinced of the political and military
impossibility of a civil war over Bavaria, rejected these demands.
On 2 November the SPD withdrew from the government.
Stresemann’s second cabinet was now a bourgeois minority
government.
In leaving the Grand Coalition, the Social Democrats were acting
out of the fear that many of their supporters would turn to the
Communists if they continued to countenance what were felt to be
‘rightist’ policies. But the retreat from government was a risk; it left
a power vacuum that could be exploited by the plotters of
dictatorship. In a fraction meeting on 31 October the Prussian
interior minister, Carl Severing, who represented Münster in the
Reichstag, warned against the danger that the French could
conspire with German separatists to establish a ‘Rhenish Republic’
and the forces of the extreme right start a war with France. ‘Back to
pure class struggle!’, the solution of Reichstag president Paul Löbe,
was not a realistic alternative for the Social Democrats. Even if
756
they were not part of the government, Germany’s largest party was
obligated to see to it that the country remained governable.
Anything else would have been a politics of catastrophe, an option
not open to the party that had founded the Weimar state.
Four days after the end of the Grand Coalition the Reichswehr,
with Ebert’s authorization, moved into central and eastern
Thuringia, forcibly dissolving the Proletarian Hundreds over the
ensuing days. On 12 November the Thuringian SPD, submitting to
the pressure from Berlin, terminated the coalition with the KPD.
Prime Minister Frölich stayed on as head of a minority cabinet
until early Landtag elections on 24 February 1924. These elections
took the SPD out of power; for the next three years, Thuringia was
governed by a bourgeois minority cabinet tolerated by the DVP. In
Saxony, the interim of the SPD minority government lasted until
January 1924, when it was succeeded by a SPD–DDP–DVP cabinet
bitterly opposed by the Social Democratic left wing. Two years
later the Saxon SPD split, and the Social Democratic Party
thereafter stood in opposition to a government headed by a prime
minister from the rightist ‘Old Social Democrats’ until 1929.
At the beginning of November 1923 very few people imagined
that Stresemann’s rump cabinet would last very long. Ebert was not
one of them. In a letter of 4 November, the president authorized
General von Seeckt to find out if Otto Wiedfeldt, the German
ambassador in Washington, would be prepared to assume the
chancellorship at the head of a ‘small, directory-type cabinet with
emergency powers’. Wiedfeldt’s letter of refusal had not yet arrived
in Berlin by the time the Bavarian crisis reached its spectacular
peak. On the evening of 8 November, Adolf Hitler took advantage
757
of an assembly of Kahr supporters in the Bürgerbräukeller, one of
Munich’s large beer halls, to call for a ‘national revolution’.
Wielding a pistol, the National Socialist leader forced Kahr, Lossow,
and Seisser to participate. Erich Ludendorff, however, whom Hitler
appointed on the spot as commander-in-chief of the national army,
soon restored the triumvirate’s freedom of action, and Hitler’s coup
was rapidly countered. At noon on 9 November the bullets of the
Bavarian police brought the putsch to an end at the Munich
Feldherrnhalle. Hitler himself managed to make his escape, but was
arrested two days later. Sixteen of his followers paid for the
‘national revolution’ with their lives.
The Munich events led to dramatic changes in Berlin, too. In the
night of 8 November, Ebert transferred to the chief of the army
leadership the command over the Reichswehr and, in modification
of the ordinance of 26 September 1923, the executive authority.
The president and parliament were seemingly convinced that
Seeckt’s empowerment was the only way to get the Bavarian
Reichswehr to oppose the rebels. There was, of course, no
guarantee that Seeckt himself would not try to take over. Ebert
probably thought that the army chief would be less dangerous to
the republic as his, the president’s, direct subordinate than in his
previous position, which was virtually free from control.
The Hitler putsch was a watershed, not only for Bavaria, but for
the Reich as a whole. The events of 8 November thoroughly
discredited the ‘serious’ dictatorial plans of Kahr and his associates
and deeply compromised the authority of the general state
commissar. But without the firm backing of the Bavarian political
leadership, a ‘national’ revolution was virtually inconceivable in
758
Germany. As a result, Hitler’s putsch achieved the exact opposite of
what the National Socialist leader had sought, contributing greatly
to the consolidation of the endangered republic.20
On 15 November, one week after the Munich putsch,
Stresemann’s rump cabinet succeeded in performing the
‘Rentenmark miracle’. The new currency, introduced on this day,
was conceived as a provisional measure. According to the proposal
of Hans Luther, who had succeeded Rudolf Hilferding as finance
minister on 6 October, debenture bonds and mortgages on
agricultural and industrial land would guarantee the market value
of the Rentenmark until a permanent, gold-supported currency
could be introduced. On 20 November the mark’s exchange rate,
which had been 1.26 trillion to the dollar on 14 November, was
stabilized at 4.2 trillion. At that point, the federal bank set an
analogous rate of 1 trillion paper marks to 1 Rentenmark, thereby
achieving once again the pre-war exchange rate between mark and
dollar.
The victim of the currency reform was the Rhineland. Until the
introduction of the gold-backed Reichsmark on 30 August 1924, the
occupied territory was all but abandoned by the Reich and forced
to get along with municipal emergency money as a means of
payment. On 13 November the mayor of Cologne, Konrad
Adenauer, protested that ‘the Rhineland must be worth more than
one or two or even three new currencies’, but in vain. From the
perspective of Berlin, the danger that the Rhineland would
temporarily become politically independent, in whatever form,
seemed like the lesser of two evils compared with the economic
collapse threatening all of Germany if the new currency were to be
759
ruined by the continued complete subsidization of the occupied
territories.
Another kind of ‘miracle’ became manifest after 25 October. On
this day Poincaré, the French prime minister, informed the British
government that he was prepared, under certain conditions, to
agree to a re-examination of the reparations question. He was
responding to a suggestion the American secretary of state, Charles
Hughes, had originally made at the end of December 1922, and
which the British government had adopted on 12 October, that the
economic aspects of the reparations question be discussed at an
international conference. The French conditions were the
following: the panel of experts was to be appointed by the Allied
Reparations Commission; the amount of Germany’s reparation
debt, imposed on the Reich by the May 1921 London ultimatum,
was to be independent of the examination’s findings; and a second
commission was to determine the amount and the whereabouts of
German foreign currency assets. After America also agreed to this
proposal, Paris officially petitioned the Reparations Commission on
13 November to set both commissions to work. This set the course
for the Dawes Plan, the 1924 reparations treaty to which the
economic upswing of the middle Weimar years was inextricably
linked.
There were many reasons for Poincaré’s about-face. The
occupation of the Rhine had become too much of a burden and was
endangering the French currency; the domestic opposition of the
left had grown stronger, and France had increasingly isolated itself
on the international stage. But more important was another reason.
On 23 October Hughes, the American secretary of state, informed
760
the French prime minister that the United States would accept
France’s participation in the inter-Allied expert commission and
link the discussion of the reparations question with the problem of
debts between the Allies. By accommodating Germany to a certain
extent, France could thus expect to improve its position as
America’s debtor.
Paris’s new policy on reparations did not yet mean that France
had renounced its aim of separating the Rhineland from Germany.
On 25 October, the same day he informed the British of his change
of course, Poincaré also decided to adopt a policy of active and
official support for the autonomy of the occupied territories. After
21 October there were movements afoot in a number of cities—
including Aachen, Trier, Koblenz, Bonn, and Wiesbaden—to
proclaim a ‘Rhenish Republic.’ These separatist aspirations were
supported by the French and Belgian occupying authorities, but not
by larger circles of the population. By November 1923 it was clear
both in the Prussian Rhineland and in the Bavarian Palatinate that
a voluntary secession from the Reich would not occur.
Thus the chances of a gradual de-escalation of tensions were
looking favourable, both internally and externally, when a new
governmental crisis broke out in Berlin. On 22 November, against
all of Ebert’s warnings, the SPD brought a motion of no confidence
in Stresemann’s minority cabinet. The justification was that the
government had proceeded with utmost harshness against Saxony
and Thuringia but had done nothing decisive about the anti-
constitutional situation in Bavaria. The motion was consciously
formulated such that the German National People’s Party, on
whose vote everything depended, would not be able to endorse it.
761
The Social Democrats did not want to bring down Stresemann’s
government, but to make a statement and thus to calm emotions on
the party’s left wing. Stresemann, however, was not about to allow
his position to be further undermined, and countered the SPD
initiative by calling for a vote of confidence. On 23 November, by
231 against 156 votes and 7 abstentions, the Reichstag rejected the
motion of the governmental parties. For the first time in the history
of the German Republic, as the deposed chancellor afterwards told
foreign correspondents, a government had fallen ‘on the open field
of battle’.
The settling of the crisis was mostly the work of Ebert. On 30
November 1923 Stresemann was succeeded by the head of the
Centre Party and parliamentary fraction, Wilhelm Marx, a judge
from Cologne and a somewhat colourless personality. He led a
bourgeois cabinet composed of members from the Centre, DVP,
DDP, and BVP, including the former chancellor as foreign minister.
The new administration was subject to the toleration of the SPD,
which, again under massive pressure from the president, even
assisted in the passage of another enabling act. This law, in effect
until 14 February 1924, empowered the government to ‘take the
measures it considers necessary and urgent with regard to the
distress of the people and the state’. One of the things now subject
to government control was the length of the working day. A new
policy was a matter of pressing concern, since the demobilization
ordinances from the revolutionary period, which had been
extended several times, had expired on 17 November. Thereafter,
according to law, the pre-war working day was in effect wherever
hours were not regulated by contract.
762
By the time the enabling law expired a quarter of a year later,
many things had changed in Germany. Although the eight-hour day
continued to be considered the norm, the ten-hour day was legally
permitted in large portions of the economy. The Free Trade Unions
answered this defeat in January 1924 by terminating the Central
Cooperative Union of November 1918. But this was little more than
a symbolic protest. In December 1923 the salaries of civil servants
were set at a level far below that of pre-war Germany. On 14
February 1924 an emergency tax measure initiated the dismantling
of state control in the housing market. This was an important step
towards ending the ‘war socialism’ that had outlasted the war by
more than half a decade.
The same ordinance regulated the fiercely controversial
revaluation of outstanding debts from certain kinds of capital
investments destroyed by inflation, such as mortgages, bonds,
debentures, savings accounts, and life insurance policies. The flat
revaluation rate of 15 per cent of the value in gold marks contained
the tacit admission that Finance Minister Luther’s principle of
‘mark for mark’ simply could not be reconciled with an elementary
sense of justice, since it would have meant complete dispossession.
However, repayment of revaluation debts was extended until 1932,
the repayment of the principal and interest on public loans,
including the war loans, until the final amortization of the
reparations—that is, for an indefinite period of time.
The embittered protest of the millions who were affected had no
effect on the measure. The Marx government could do nothing else,
at the risk of jeopardizing the new currency. The real victims of the
inflation were people with savings and the underwriters of war
763
loans. ‘A concentration of property in a few but powerful hands has
taken place,’ wrote the economist Franz Eulenburg in 1924.
The capital assets of the middle classes were destroyed, and with
them the claim to a part of the other properties. The appropriation
has occurred above all in industry. The smaller and medium-sized
companies have not been expropriated, but are now more strongly
dependent on the concerns. Accordingly, the distribution of wealth
is significantly less equitable than before.
764
which had initiated the inflation in 1914, appeared to many
Germans in a rosy light once more.
The inflation had had a levelling effect. The income difference
between the higher and lower government officials had shrunk,
and the same was true of the gap between the bureaucracy as a
whole and the working classes. But the workers by no means
benefited. According to the calculations of the federal statistics
bureau, real weekly wages in December 1923 amounted to 70 per
cent of the pre-war level. Unemployment was also high. In
December 1923, 28 per cent of union members were without work.
The membership of the ADGB fell from 7.7 million in September
1923 to 4.8 million in March 1924. The role the unions had played
in the Ruhr conflict, where they had shored up the state, was
disregarded by large numbers of workers. In fact, at the beginning
of 1924 everything seemed to indicate that the potential for mass
proletarian protest was much higher than a year before.
Nonetheless, there were also unmistakable signs of an easing of
tensions on the domestic front. In the Ruhr district, normal work
resumed at the end of November after an agreement between the
Mining Association (Bergbaulicher Verein) and the MICUM (Mission
interalliée de contrôle des usines et des mines) on 23 November 1923.
The stabilization of the economy in the Rhineland and Ruhr district
increasingly undermined the project of a Rhenish federal state
loosely connected with the Reich, such as Adenauer had been
pursuing with the support of Stinnes in autumn 1923. The plan was
sharply rejected by Foreign Minister Stresemann in January 1924,
whereupon the mayor of Cologne put it aside.
On 28 February the military state of emergency was ended at
765
Seeckt’s initiative. On the one hand, the general wanted to avoid
eroding the authority of the Reichswehr in a running battle with
civil agencies, above all in Saxony and Thuringia, but also in
Prussia. On the other hand, Seeckt feared an infiltration by right-
wing radical organizations. Internal consolidation of the army was
more important to him than an exercise of power with no political
payoff.
There was some initial controversy about whether the bans
against the KPD, NSDAP, and the German Ethnic Freedom Party,
which the army chief had pronounced on 23 November 1923 as
head of the executive, should remain in effect. Seeckt wanted to
maintain them; Severing demanded they be lifted. The Prussian
interior minister was able to prevail, for the most part. The party
bans came to an end at the same time as the state of emergency.
Public demonstrations in the open air were still forbidden for the
time being, though the central agencies of the Länder were
permitted to authorize exceptions. This ‘civil’ state of emergency
lasted eight months, ended by presidential decree on 25 October
1924.
The conflict between Bavaria and the Reich was officially
brought to an end in February 1924. As per a settlement of 14
February, the Reichswehr commander in Bavaria could only be
recalled in agreement with the Bavarian government and in accord
with its legitimate wishes. Furthermore, the wording of the oaths
for the army and navy were modified to include a pledge of loyalty
to the constitution of the recruit’s home state. This settled the right
of the Munich government to use the troops of the Reichswehr
stationed in Bavaria. Four days later, Kahr resigned as general state
766
commissar and Lossow as commander-in-chief in Bavaria. Their
anti-Reich and anti-constitutional activities in the autumn of 1923
had no legal consequences of any kind.21
On 1 April 1924 the Munich court handed down the judgments
in the cases against the rebels of 8–9 November 1923. General
Ludendorff was acquitted of the charge of high treason. Five other
participants in the putsch, among them Ernst Röhm, organizer of
the National Socialist ‘Storm Troopers’ (Sturmabteilungen or SA),
were sentenced to three months in prison and a fine of 100 marks.
Hitler himself, along with three other conspirators, was condemned
to five years in prison and a fine of 200 marks. For this latter
group, too, probation was possible, once six months of the sentence
had been served. (In the event, Hitler was released at Christmas
1924 from his imprisonment at Landsberg, which he had used to
write his book Mein Kampf.*) All of the accused were recognized by
the court as having ‘acted in a purely patriotic spirit and according
to the noblest and most selfless will’ and in the sincere belief ‘that
they were compelled to act in order to save the Fatherland and that
they did exactly what the intentions of the leading Bavarian men
had been shortly before.’ Morally speaking, this amounted to an
acquittal, and it was understood as such throughout all of
Germany.
The controversy of the Munich decision had not yet passed
when an event that was to decisively affect the further
development of the republic made headlines. On 9 April 1924 the
commission of experts under the American banker Charles Dawes,
which had gone to work in Paris in January, published its report on
767
the reparations question. While no total amount owed by Germany
was named, the study evidently assumed that the 132 million gold
marks demanded by the London ultimatum of May 1921 was
beyond the capacity of the German economy. In order not to
endanger the currency, it was recommended that the creditor
nations appoint a reparations agent to arrange for ‘transfer
protection’, a payment method that sought to safeguard the
stability of the mark. The yearly instalments, or annuities, began
with a billion gold marks, to be increased to 2.5 billion over the
course of five years. In order to accommodate the French desire for
guarantees, the rail system of the Reich was changed into a
corporation with specific obligations to fulfil and with a board of
directors including representatives from the creditor nations (an
internationally composed general council was also installed in the
Reichsbank). Several other sources of federal revenue and a 5
billion mark, interest-bearing mortgage of German industrial
corporations were to provide further securities.
These restrictions on German sovereignty were far-reaching, and
yet considerably easier to bear than the territorial guarantees
France and Belgium had seized in January 1923 in the Ruhr
occupation. The Dawes Plan also contained an element of good
news for the German economy: a 800 million mark foreign loan,
which was to provide the foundation for a new bank of issue and
secure the stability of the currency. The proceeds were to be
directed exclusively towards the payment of domestic obligations
to the Allies like materials shipments and occupation costs. But the
provision contained the prospect of future American loans and
investments, and this prospect had a stimulating effect. Germany,
768
which had been one of the most important markets for American
goods before 1914, could well assume that the United States had
recognized the opportunity that lay in engagement with its
productive, albeit capital-hungry economy.
The Dawes Plan was America’s contribution towards the
stabilization of Germany. The Union of Soviet Socialist Republics
made its own contribution at about the same time. After Lenin’s
death in January 1924, Moscow’s drive towards world revolution
abated. As Stalin’s position grew stronger, the ‘development of
socialism in one country’—that is, in the Soviet Union—became the
priority. The improvised revolutions attempted in Germany by the
Comintern in March 1921 and again in the autumn of 1923 did not
accord with the new policy, proclaimed by Stalin in 1925 but
already in practice earlier.
The world-political change of scene in 1923–4 also included
new developments in London and Paris. In Great Britain, the
Labour Party and the Liberals triumphed over the Conservatives in
the elections for the House of Commons on 6 December 1923. In
January 1924 Ramsay MacDonald became the first British head of
government from the Labour Party. In France, Poincaré’s Bloc
national lost the majority to the Cartel des gauches, an electoral
alliance between the Socialists and the bourgeois Radical Socialists,
on 11 May 1924. The Radical Socialist Édouard Herriot, an admirer
of German Idealist philosophy, became prime minister without
opposition from the Socialists. Germany could expect a friendlier
attitude from the new British and French governments than from
the previous rightist cabinets.
It was evident by the spring of 1924 that France’s attempt to
769
forcibly revise the post-war order in its favour had failed. Germany
emerged from the Ruhr conflict economically weakened but, thanks
to the intervention of the United States, politically stronger. The
post-war period had come to an end between November 1923 and
April 1924. A relative stabilization in Germany and in the relations
between the most important states was unmistakable.22
770
vote. In May 1924 they managed only 34%. The SPD fell from
21.7% to 20.5%. This seems like only a small loss at first glance. In
reality, however, it was a disaster: the reunited SPD gained fewer
votes in May 1924 than the Majority party alone had obtained in
1920. Clearly, a considerable number of the 17.9% of Germans
who had voted USPD in 1920 had migrated to the KPD, which,
with 12.6% of the 1924 vote, became a proletarian mass party in
the Reich for the first time.
Some of the parties of the bourgeois centre and the moderate
right took considerable losses—losses which became the gains of
the German Nationalists and the German Ethnic Freedom Party.
The DVP fell from 13.9% to 9.2%, the DDP from 8.3% to 5.7%. The
Catholic parties suffered comparatively little: the Centre Party sank
from 13.6% to 13.4%, the BVP from 4.2% to 3.2%. Bourgeois
splinter groups garnered a total of 8.5% of the vote, a 5.3% gain
from the previous elections.
After the elections, the DNVP demanded the office of chancellor
and presented the former Admiral of the Fleet von Tirpitz as its
candidate. The centre parties rejected the Father of the German
Navy with the same resolution the German Nationalists showed
with regard to the centre’s exhortation that they embrace the
Dawes Plan, the rejection of which had been one of the main
themes of their electoral campaign. The upshot was that, on 3
June, Wilhelm Marx formed another bourgeois minority cabinet,
which, however, the BVP refused to join on account of acute
tensions.
After the defeat of the SPD, a Grand Coalition was no longer
seen as a serious option. At the Social Democratic Party congress in
771
Berlin from 11 to 14 June 1924, the left wing denounced the
leadership’s coalition policy to date. The president of the
Metalworkers’ Association, the former Independent Robert
Dissmann, confronted the ‘solicitude for the state and bourgeois
coalition parties’ with a ‘policy of uncompromising class struggle’,
which Dissmann saw as the only way to win back the proletarian
voters who had gone over to the Communists. Müller, the party
head, expressed the mitigating view that if the coalitions of the
recent years were examined, it would be seen that ‘we have only
been in the government when we had to be in the government. The
reasons that compelled us were almost always reasons of foreign
policy.’ In a proposal submitted by Müller and adopted by the
delegates with a large majority, the congress characterized the
coalition policy as a matter of tactics and not of principle.
Participation in government was to be undertaken only ‘after an
examination of all advantages and disadvantages for the interests
of the less well-off, so that we can be sure that the working class
will not alone be called upon to make sacrifices’. The message was
clear: non-participation in the government of the Reich was to be
considered the norm for the Social Democratic Party.
Non-participation did not mean unconditional opposition, of
course. In matters of foreign policy, Chancellor Marx and Foreign
Minister Stresemann could regularly count on the votes of the SPD.
This was also true with regard to the Dawes Plan and the
supplementary agreements worked out at a conference in London
in August 1924. The German negotiators were able to achieve a
notable success there: the French and Belgians promised to
withdraw from the areas they had occupied in 1921 and 1923—not
772
immediately, as the Germans had demanded, but within a year’s
time. The Dortmund-Hörde zone, however, as well as all territories
on the right bank of the Rhine occupied in January 1923, would be
restored to Germany the day after the final signing of the
agreement. This made it officially clear that the Rhineland would
remain part of Germany and would be economically and financially
fully reintegrated into the Reich.
The votes of the SPD were not sufficient to pass all the laws of
the Dawes Plan. The railway bill, which restricted German
sovereignty and thus entailed modifying the constitution, required
a two-thirds majority. This could only be achieved with the support
of a considerable number of German Nationalist votes. In order to
soften DNVP opposition, the government made a clearly ‘national’
statement about the question of war guilt on 29 August, one day
before the signing of the London agreement. But even this gesture
failed to achieve the hoped-for effect. Several influential interest
groups pressed the German Nationalists to accept the bill. These
included the National Association of German Industry
(Reichsverband der Deutschen Industrie, which united in 1919 the
Central Association of German Industrialists, the Federation of
Industrialists, and the Association for the Protection of the Interests
of the Chemical Industry), the Christian-national Labour unions,
and, for a time, though in a qualified manner, even the National
Land League (Reichslandbund), the powerful successor of the
Federation of Agriculturalists. In the case of the bill’s rejection, the
president and the chancellor threatened to dissolve the Reichstag.
The DVP held out the prospect of a share in the government if the
DNVP helped it pass. In the decisive ballot on 29 August, 52
773
German Nationalist delegates voted against, 48 in favour. That was
sufficient to pass the railroad law with a two-thirds majority and
the London agreement as a whole.
The next day, the ‘good behaviour’ of the DNVP was to be
rewarded. On the parliamentary agenda stood the first reading of a
bill proposed by the government to reintroduce the agricultural
customs of the 1902 ‘Bülow tariff’, which had expired in 1914. The
law was to take effect on 25 January 1925, the day Germany’s
Versailles Treaty obligation to grant the victorious powers most-
favoured-nation privileges expired, restoring the Reich’s
commercial freedom. But the plan foundered on the opposition of
the left: the SPD and KPD walked out of the plenum assembly,
leaving the Reichstag without a quorum. It adjourned until 15
October. Three days before it reconvened, the government
withdrew the bill. The attempt to restore the privileges of the large
landowners at the expense of the consumers had failed, at least for
the moment.
The impossibility of finding a dependable parliamentary
majority for the business of government was so obvious that the
Marx cabinet decided, on 20 October, to petition the president for
the dissolution of the Reichstag. Ebert agreed that same day,
setting new elections for 7 December. The second parliamentary
electoral campaign was influenced by the economic recovery the
country was experiencing. On 30 August, the provisional
Rentenmark gave way to the new Reichsmark, which was covered to
40% of its value by gold or foreign currencies. Once the London
agreement had been concluded, foreign credit poured into
Germany. The unemployment rate fell from 12.4% of union labour
774
in July to 7.3% in November. Contract wages climbed from an
average of 57 pfennigs per hour in January 1924 to 72.5 in
January 1925 (according to information from twelve selected
branches of industry). At the same time, the percentage of workers
who had to work more than 48 hours per week fell from 54.7% in
May to 45.5% in November.
The economic upswing led to a political de-radicalization. The
extremists on the outer party wings—the German Ethnic Freedom
Party, which now called itself the National Socialist Freedom Party
(Nationalsozialistische Freiheitspartei), and the Communists—
emerged from the 7 December elections weaker than before. The
winners were the Social Democrats and, to a lesser extent, the
German Nationalists. The SPD climbed from 20.5% to 26% of the
vote, the DNVP from 19.5% to 20.5%. The Communists fell from
12.6% to 9%, the union of the National Socialists and the German
Ethnic Freedom Party from 6.5% to 3%. The changes in the centre
and moderate right were fairly minor. Of the small parties, the
Economic Party (Wirtschaftspartei), representing the small and
medium-sized business community, was the most successful with
3.3%. The revaluation parties achieved a mere 0.4%, mainly
because the German Nationalists had once again campaigned
strongly as the party of those harmed by the inflation.
The election results allowed only two possibilities for a majority
government: either a Grand Coalition or a bourgeois centre-right
cabinet. The DVP declared its unwillingness to work with the Social
Democrats, and the DDP refused to join with the German
Nationalists—which did not make a bourgeois bloc impossible,
however, since the latter already had a majority without the left-
775
liberal party. On 15 January 1925, after arduous negotiations, the
first Reich government with DNVP participation came together
under the non-aligned chancellor Hans Luther. Stresemann
remained foreign minister, the Centre politician Heinrich Brauns
minister of labour. Despite the fact that his party, the DDP, was not
in the government, Otto Gessler remained the director of the
defence department as a ‘special minister’. The German Nationalists
assumed the leadership of the interior, finance, and economics
ministries. Right at the beginning of his term, the new minister of
economics, Karl Neuhaus, found himself compelled to bitterly
disappoint a large portion of the DNVP constituency. Supported by
the unanimous vote of the leading organizations in agriculture,
industry, commerce, and banking, he announced in a memorandum
at the end of January that a revaluation of more than 15 per cent
(that is, the rate set by the emergency tax measure of 14 February
1924 and vehemently attacked by the DNVP) would be intolerable
for the owners of material assets and thus unworkable.23
776
levelled at him by—among others—a völkisch journalist by the
name of Erwin Rothardt, editor of the Mitteldeutsche Presse, for
Ebert’s role during the strike of the Berlin munitions workers in
January 1918. On 23 December 1924 an enlarged jury of the
Magdeburg district court handed down its judgment in the libel
action Ebert had undertaken against the journalist. Although the
court sentenced the accused to three months in prison for libel
against the president, the opinion noted that Rothardt’s claim that
Ebert’s participation in the strike represented treason was, in a
legal sense, valid. Accordingly, defamation of character had not
taken place.
Powerful voices were immediately raised against this character
assassination on the part of the Magdeburg judges, a typical case of
judiciary anti-republicanism. The lawyers for the prosecution
appealed the case. The Reich government, the Marx cabinet, issued
a statement in defence of the president’s honour. Famous scholars,
including the historians Friedrich Meinecke and Hans Delbrück as
well as the jurists Gerhard Anschütz and Wilhelm Kahl, made an
official statement on his behalf. In an open letter to Ebert printed
in the Berliner Tageblatt, the Evangelical theologian Adolf von
Harnack (who was also president of the Kaiser Wilhelm Society for
the Advancement of Science, the later Max Planck Society, and had
written one of the drafts for the emperor’s appeal to the German
people in August 1914) spoke of a ‘miscarriage of justice’ that filled
him with outrage:
A disgraceful thing has taken place here, and we are saddened and
alarmed. All the more do I, along with all good Germans, feel the
777
gratitude the Fatherland owes you, most honoured Herr President
of the Reich, for all of your patriotic work and efforts, especially
during the years 1918 and 1919. And since this gratitude lives on
today in thousands of hearts, the judgement of history shall
confirm it for all future. Nonetheless, such voices were the
exception and became, in their turn, the target of hateful
commentaries from the right. The Magdeburg decision had a
powerful effect, and it was directed against Ebert and the republic
for which he stood.
778
Those who had hated and despised Ebert while he was alive
continued to do so after his death. This was equally true of the
National Socialist, German Nationalist, and völkisch right as of the
Communists, for whom the delegate Hermann Remmele spoke
when he told the Reichstag on 1 March 1925 that the deceased
president had ‘gone to the grave with the curse of the German
proletariat’. In retrospect, this reproach shows precisely where
Ebert’s greatest historical achievement lay: in his untiring efforts
towards rapprochement and cooperation between the moderate
forces in the labour movement and the bourgeoisie. Ebert had
recognized earlier than many of his party colleagues that this
compromise formed the existential core of the republic, and no one
matched him in his persistent devotion to the goal of achieving it.
At the same time, however, Ebert’s limitations were clear. He
depended all too often on the judgement of military and
bureaucratic advisers, whom he had every reason to mistrust. Much
to the consternation of his friend Otto Braun, he had no instinct for
the dangers inherent in frequent recourse to Article 48 (in 1923
alone Ebert signed forty two emergency measures, most of them
aimed at economic crises). By the time of his death, the law
concerning the formal execution of this article, required by the
constitution, still had not been passed. It never was.
We would thus probably not be justified in calling Friedrich
Ebert a great German statesman. He was a staunch democrat, a
German patriot, and an exponent of balance and conciliation both
at home and abroad. His knowledge of politics and humanity made
him, who had no more than a basic school education, far superior
to the many academics who condescended to him. Only after he
779
was gone did many of his political friends realize what it meant to
have a republican at the head of the republic. His death, a few
short months before the end of his term and the election of his
successor, marks a deep caesura in the history of the first German
Republic.24
The first ballot in the first popular election for a Reich president
took place on 29 March 1925. The candidate for the governmental
right was Karl Jarres, former interior minister and at that time
mayor of Duisburg, endorsed by the DVP (his own party), the
German Nationalists, and the Economic Party. The SPD put forward
Otto Braun, who had just resigned from the office of Prussian
prime minister in the wake of a governmental crisis. The Centre
was represented by the former chancellor, Wilhelm Marx, the DDP
by Willy Hellpach, president of Baden, and the BVP by Heinrich
Held, prime minister of the Free State of Bavaria since April 1924.
The Communists ran their party head, Ernst Thälmann, a former
shipyard worker who had cut his political teeth during the uprising
in his home town of Hamburg at the end of October 1923. The
National Socialist candidate was Erich Ludendorff.
Nobody achieved the absolute majority necessary in the first
ballot. Jarres did best with 38.8% of the vote, followed by Braun
with 29% and Marx with 14.5%. The rest of the field was far
behind: Thälmann managed 7%, Hellpach 5.8%, Held 3.9%, and
Ludendorff 1.1%.
The ‘Weimar’ parties realized they could not defeat the right
unless they agreed to a common candidate. They chose Marx,
among other reasons because the SPD knew from its experience
780
with runoff agreements during the monarchy that its adherents
followed party recommendations in a far more disciplined manner
than bourgeois voters. The Centre, in turn, agreed to elect Braun as
Prussian prime minister (which occurred on 3 April 1925).
Since Jarres would have no chance against Marx backed by a
unified republican Volksblock, the right looked around for a more
attractive option. The law on the election of the Reich president (of
4 April 1920) permitted a candidate who had not taken part in the
first ballot. This gave ‘nationalist’ circles around the former
Admiral of the Fleet von Tirpitz the chance to court a popular non-
politician who was already a living legend: Field Marshal Paul von
Beneckendorff und von Hindenburg. At 77 years of age (he was
born in Posen on 2 October 1847), Hindenburg had been living in
retirement in Hanover since his departure from the Army High
Command in the summer of 1919.
The ‘victor of Tannenberg’ had become a kind of ‘substitute
emperor’ already during the First World War. The nationalist right
had first identified him as a possible future Reich president in
spring 1920, at the time of the Kapp–Lüttwitz putsch. Once again,
German Nationalists from the old Prussian provinces, big
landowners from the high councils of the National Land League,
and former senior officers were the most active on behalf of his
candidacy. There was, to be sure, strong initial resistance in the
committee of the conservative Reichsbürgerrat around the former
Prussian interior minister Friedrich Wilhelm von Loebell, which
had backed Jarres in the first ballot. Big industry saw Hindenburg
as the spokesman for agrarian interests. Stresemann feared that a
President Hindenburg would provoke a negative reaction from
781
abroad. After Jarres withdrew his candidacy, however, such
reservations lost force. As a loyal monarchist, Hindenburg first
sought the blessing of the former emperor, living in exile in
Holland, before announcing on 7 April his willingness to run as the
candidate of the Reichsblock.
Hindenburg’s electoral prospects were good. He could depend
not only on the votes of staunch monarchists and most churchgoing
Protestants; the Bavarian People’s Party, too, backed him, a
Prussian Lutheran, mainly because Marx, though a Bavarian
Catholic, was supported by the despised Social Democrats. It also
helped Hindenburg that the KPD leadership decided on 11 April to
keep Thälmann in the running—a decision the Communists
justified by saying that it was not the duty of the proletariat ‘to
seek out the cleverest representative of bourgeois interests or
choose the lesser of two evils between the civilian dictatorship of
Marx and the military dictatorship of Hindenburg’. Accordingly,
Marx could not count on making up on the ‘left’, among the
Communist voters, what he would lose on the ‘right’ in
conservative Bavaria.
The Volksblock kept warning until the end about the dangers
Hindenburg’s election would represent for the interests of the
republic and of peace. But it was of no use. On 26 April 1925 the
Reichsblock candidate emerged from the second ballot with a
margin of 900,000 votes over Marx. With electoral participation at
77.6%, 8.7% higher than in the first ballot, Hindenburg achieved
48.3% of votes cast, followed by Marx with 45.3% and Thälmann
at 6.4%. The field marshal missed an absolute majority by only a
small margin—but an absolute majority was not necessary this
782
time.
The Communists could have prevented Hindenburg’s election.
The Vorwärts headline ‘Hindenburg, President by the Grace of
Thälmann’ accurately described the effect of the KPD leader’s
candidacy. But the BVP was at least equally responsible. Other
‘fathers’ and ‘mothers’ of Hindenburg’s success were anti-Catholic
liberals in Württemberg and anticlerical Social Democrats in
Saxony; the former had voted for Hindenburg, the latter for
Thälmann as a ‘lesser evil’ than Marx, a ‘vassal of Rome’. The shift
from one of the ‘black, red, and gold’ candidates to Hindenburg
was especially marked in East Prussia. Nowhere was the myth of
the ‘victor of Tannenberg’ as strong as in the province where the
battle had taken place.
Hindenburg’s triumph was not a plebiscite for the restoration of
the monarchy, but it did represent a referendum against
parliamentary democracy, such as the Germans had become
acquainted with it in the time since 1919. Disillusionment with the
grey republican daily routine went hand-in-hand with a nostalgic
transfiguration of the past. The liberal Frankfurter Zeitung wrote
that it was mainly unpolitical Germans who were responsible for
the outcome of the election:
783
imperial governance and warfare, as whose representative they
honour that commander. The romantic longing for the glory and
greatness of the past—that is what brought these unpolitical strata
to the polls and Hindenburg the victory.
The Berliner Tageblatt, also a liberal paper, felt ‘shame over the
political immaturity of so many millions’; the Social Democratic
Vorwärts spoke of a ‘shock victory of the reaction, won by
Communist treason to the republic’. Both newspapers—along with
the writer Heinrich Mann at about the same time—compared
Hindenburg’s election to an event early in the history of the third
French republic, the victory of the clerical monarchist Marshal
MacMahon in the presidential election of 1873. Vorwärts saw a
modicum of comfort in the historical analogy:
784
and gold standard of the Reich president), ‘Hindenburg will now
make the republic acceptable, including the Black, Red, and Gold,
which will now appear everywhere together with Hindenburg as
his personal colours. Some of the admiration directed at him must
necessarily rub off on it.’
Such hopes were not completely without foundation. The fact
that Hindenburg promised to respect the constitution made it
difficult for many of the republic’s enemies to persist in their
inveterate hatred of the new state. The ‘realist’ shift in the
Evangelical church was a characteristic example; only in 1925 did
it finally accept the republic as a fact, albeit an unpleasant one. If it
had been legally possible to elect a replacement emperor, the
German people might one day have accepted or even sought
outright a return to the monarchy. For the time being, however,
the main goal of the black, white, and red camp seemed to be
within reach—the development of a strong state, as capable and
effective as imperial Germany had been in putting the parliament
and the parties in their place.
The milieu with the greatest cause for satisfaction after the
second presidential ballot was the one from which Hindenburg
himself came and with which he continued to be on intimate terms,
the world of the military and the Prussian nobility. It was of
singular importance to the army and the great manorial lords to
have, once again, direct contact to the head of state, one who held
supreme executive authority in times of crisis. The relations of
power in society and politics did not suddenly change after 26
April 1925. But, as of that day, the old Prussian power elite of pre-
republican Germany had the whip hand again, which it could use
785
anytime the parliament was unwilling to recognize the demand of
the hour. From the perspective of the ‘right’, this was a great step
forward. It was certainly a step away from the Weimar of 1919.
The events of spring 1925 represented nothing less than a silent
constitutional metamorphosis, a conservative re-founding of the
German Republic.25
786
can be inspected at any time, though not without risk. Not without
risk if, for example, the ‘Berliner’ tried energetically to turn off the
terror, dictatorship, and insolence of the local bourgeoisie. No
court supports him there, no administrative agency, no newspaper.
He is forsaken and must quit the field.
787
which had no lack of symbolic targets. The incessant campaign
against the Bauhaus, the bastion of modern architecture, was one
example (Tucholsky also referred to it). In 1925 the Bauhaus was
forced to leave its original home in Weimar after the Thuringian
parliament cut its funds by half in the autumn of 1924, making it
de facto impossible for the institution to keep running. But even
the new location in Dessau, the capital of Anhalt, where a Social
Democratic prime minister was in office almost continually from
1918 to May 1932, the Bauhaus was a blot on the landscape for the
forces of the right. In 1929, at the dedication of a colony Gropius
had designed for the workers and employees of the Junker’s factory
in Dessau-Törten, National Socialists and German Nationalists
protested against the ‘Moroccan huts’ of the ‘nigger colony’. The
main cause for alarm was the fact that the houses did not have the
characteristic German pointed roofs, but were flat-roofed in the
style of the ‘new objectivity’.
The war against the Weimar spirit also expressed itself in more
elevated forms. The intellectual critics on the right regarded the
republic as a product of collectivist levelling, in which the masses
triumphed over the individual personality. One of these critics was
the philosopher Martin Heidegger. In Sein und Zeit,* his 1927
magnum opus, Heidegger spoke of a dictatorship of ‘the they’ (das
Man):
The they is always present, but in such a manner that it has always
stolen away whenever existence presses for a decision. Since the
they presents every judgement and decision as its own, however, it
strips the individual existence of responsibility. The they can
788
afford, so to speak, to have ‘them’ constantly appealing to its
authority. It can most easily take responsibility for everything,
because it is not someone who needs to take a stand for anything.
It ‘was’ always the they who did it, and yet it can be said that it
was ‘no one’. In the daily routine of existence, most things come
about in such a way that we must say, ‘it was no one.’
789
the general will. The directly elected head of state as the
incarnation of the Rousseauist volonté générale had to be made
stronger in confrontation with a parliament articulating the volonté
de tous, the sum of the many individual wills. According to the
Weimar constitution, as Schmitt wrote in his 1928 book
Verfassungslehre,† there were two political leaders, the chancellor
and the president.
790
most intellectuals of the ‘conservative revolution’, the term
‘revolution’ was more a metaphor than a concrete programme. It
stood for a radical break with bourgeois liberalism and Western
democracy. The ‘revolution from the right’—the title of a 1931
book by the sociologist Hans Freyer—set the Volk against society,
the order of the whole against the pluralism of competing interests,
the political decision binding for all against uncontrolled
parliamentary debate, the leader against the masses.
Their opposition to the capitalist and democratic west brought a
number of ‘conservative revolutionaries’ into a certain proximity
with the communist and dictatorial east. Cooperation between
Germany and the Soviet Union, both strongly revisionist powers,
had its supporters among communists and nationalists alike.
Moreover, among the so-called ‘national Bolshevists’ (a
questionable name) around Ernst Niekisch, former leader of the
Augsburg and Munich councils of 1919, this foreign policy
orientation found its counterpart in a sharply anti-capitalist
rhetoric, albeit of a more ‘Prussian’ than ‘Russian’ colouring.
Even staunchly anti-Marxist intellectuals like Oswald Spengler,
author of Der Untergang des Abendlandes,* advocated for a ‘German’
or ‘Prussian socialism’. The great question for the world, as
Spengler wrote in his 1920 book Preussentum und Sozialismus,† was
the choice between the Prussian and the English idea, socialism
and capitalism, state and parliament.
791
itself from the illusions of Marxism. Marx is dead. As a form of
existence, socialism is just beginning, but the socialism of the
German proletariat is at an end. For the worker, there is only Prussian
socialism or nothing… For conservatives, there is only conscious
socialism or destruction. But we need liberation from the forms of
Anglo-French democracy. We have our own.
792
society was to be replaced by the unified nation. In this way,
integral nationalism was a response to Marxism and liberalism. The
latter was seen as the precondition for the former, and both were
enemies. Nobody understood the domestic-political serviceability
of nationalism as a weapon against all forms of the Marxist left as
well as Hitler. At the beginning of 1924 the leader of the National
Socialists wrote in an essay in defence of the putsch of 8–9
November 1923:
793
called German ‘consensus anti-Semitism’. It was not at the centre of
a quasi-religion, as for the National Socialists.
In general, anti-Semitism was not as widespread and intense
during the relatively ‘good’ Weimar years between 1924 and 1929
as it had been during the republic’s first half-decade. But it was still
strong in Germany society, especially with respect to the putative
cultural hegemony of the Jews, that is, in journalism, the
publishing industry, theatre, and film. Those who saw the Jewish
spirit at work in corrosive, subversive intellectualism and the
decadent civilization of the metropolis could count on an audience
not only on the right, but even in the environs of the political
centre. Otto Dibelius, Evangelical superintendent of the Kurmark,
delegate for the DNVP, and confessed anti-Semite, wrote in a
‘confidential’ letter to the pastors of his parish in 1928 that it was
undeniable that the Jews had always ‘played a leading role… in all
the subversive phenomena of modern civilization’. Der grosse
Herder, a Catholic reference work, observed in a 1926 article under
the correspondng rubric that anti-Semitism was ‘an antipathy felt
by the majority against what it considers to be a foreign minority,
one that closes itself off to a certain extent but is nonetheless
unusually influential, demonstrating high spiritual values and, at
the same time, an exaggerated self-importance.’ That same year, an
advertising brochure for the Vitte seaside resort on the island of
Hiddensee, distributed by island management, stated: ‘It must be
said that the Jews make a point of avoiding Vitte.’ Anti-Semitism
was a quotidian affair in Germany of the Weimar Republic, and
among conservatives it was socially acceptable, provided it did not
go beyond the limits imposed by traditional notions of ‘decency’.
794
Jews played a very prominent role in the circles normally
associated with the spirit of Weimar. They were not present on the
political right simply because the right was anti-Semitic. To oppose
the discrimination of minorities was, by definition, to be a leftist or
a liberal. One of the main reasons Jews were strongly involved in
the labour movement was because nowhere else did there exist a
mass movement fighting for a society of equal rights. But since
Weimar society was very far from being egalitarian, it could not
satisfy the left. Criticism of the prevalent order thus became one of
the primary characteristics of the leftists intellectuals of the
Weimar Republic, Jews and non-Jews alike.
A considerable number of these thinkers went so far in their
critique as to condemn the new Weimar state in its entirety. For the
Communists and their adherents, the ‘bourgeois’ republic was not
worth defending. The intellectuals acting under the aegis of Willy
Münzenberg, the press and propaganda chief of the KPD, aspired to
what the party proclaimed: the revolutionary destruction of the
existing system and the creation of ‘Soviet Germany’. Even the
most prominent among them, people like Bertolt Brecht, Arnold
Zweig, Anna Seghers, Johannes R. Becher, and Kurt Weill, were
‘Weimar’ intellectuals only in the period sense, not because they
had any inner attachment to the republic of 1919.
One such leftist Jewish intellectual was Kurt Tucholsky. Though
he was certainly no KPD propagandist, Tucholsky’s view of Weimar
was nonetheless radically critical. The main target of his mockery
and contempt was the SPD. When the Social Democrats adopted a
reformist platform at their party congress in Görlitz in 1921,
Tucholsky wrote a poem referring to them as ‘Skat brothers… who
795
have read Marx’. Five years later he compared ‘the completely
pigheaded dear good SPD’ with ‘modest radishes: red on the
outside, white on the inside’. For the Weltbühne author, the
necessity of compromise, which the SPD could not avoid even
when they were out of power, was nothing more than ‘parliament
rote’ (Parlamentsroutinendreh). In terms of its effects, the battle
Tucholsky and his colleagues waged against the Social Democrats
was a battle against parliamentary democracy. In this respect, the
intellectuals of the Weltbühne circle were much closer to the anti-
parliamentarians of the ‘conservative revolution’ than both sides
were aware.
Most of the intellectuals who supported Weimar were conscious
of its inner weakness. Thomas Mann, who had still been defending
the authoritarian German state at the end of the war, made a very
famous speech in defence of the German Republic to an auditorium
of partially hostile Berlin students in October 1922, on the occasion
of Gerhart Hauptmann’s sixtieth birthday. At the end of November,
at an event sponsored by the DDP in Munich, Mann, who strongly
identified with the Bavarian capital, expressed his anger and
sadness over the alienation that had been growing between Munich
and Berlin since the pre-war years. Before the war, Munich had
been democratic and Berlin feudal-military, but now it was
practically the reverse.
796
kinds of sinister foolishness. We were forced to look on as Munich
was decried in Germany and beyond Germany as the refuge of
reaction, as the headquarters of all obstinacy and contumacy
against the will of the age. We were forced to listen to it being
referred to as a stupid city, as the stupid city.
797
Association in Berlin in January 1925. ‘Social discord no longer
exists between the working class and the bourgeoisie in general.
The rift has shifted to the right and now goes right through the
middle of the bourgeoisie itself.’
Meinecke could also have said that the rift had shifted to the left
as well as the right, going through both the middle of the working
class and the middle of the bourgeoisie. Indeed, the political lines
of division reflected those in society less than ever before. Between
the bourgeois Vernunftrepublikaner and the far right yawned a
veritable abyss. But the same was true of the relationship between
Social Democrats and Communists. Both workers’ parties still
employed some of the same Marxist concepts, but understood
completely different things by them. ‘Class struggle’, for example,
meant for the Communists the sharpening of social conflict towards
the ultimate goal of proletarian revolution, whereas for the Social
Democrats and the Free Trade Unions it meant a pluralist politics
in the interest of the workers.
The republic still rested on the shoulders of the moderate forces
within the bourgeoisie and the working class. In the middle of the
1920s there were both signs pointing toward a renewal of the ‘class
compromise’ of 1918–19 and developments that seemed to indicate
political polarization. Only one thing was certain: the stabilization
of Weimar after 1923 was only relative, measured according to the
instability of the previous years. The internal threat to democracy
had not ceased to exist, but only ebbed somewhat.27
798
THE CONSERVATIVE REPUBLIC
The year 1925 has gone down in history not only because of
Hindenburg’s election to the Reich presidency. A second event also
ranks high in historical importance: the conclusion of the Locarno
accords on 26 October, which sealed Germany’s readmission into
the circle of European great powers. This treaty system was
designed to consolidate the post-war order, but it did so, in
accordance with German wishes, in an asymmetric manner. Only
the German western borders were secured by international law.
Germany, France, and Belgium renounced all use of force to alter
the existing borders, and their inviolability was further guaranteed
by Britain and Italy. With its eastern neighbors Poland and
Czechoslovakia, on the other hand, Germany only concluded
arbitration treaties. However, France promised military support for
both countries in the event of a German attack.
Locarno in no way closed off the possibility of a peaceful
revision of the German eastern border. Foreign Minister
Stresemann left no doubt that he was working towards this goal, in
full agreement with German public opinion. As he informed the
German embassy in London on 19 April, a peaceful solution to the
Polish border question would not be achievable
799
until such time as the country is ready for a border agreement
corresponding to our wishes and until our own position of power is
strong enough… Only the unrestricted return of sovereignty over
the territories in question can satisfy us.
800
canvassed for League membership earlier and more consistently
than any other German political party, the achievement was a
cause for great celebration. ‘Germany and Europe are progressing
from the crisis of international anarchy to a condition of
international order, in which the liberty of all peoples will be
realized in due course,’ wrote Vorwärts, which called the event a
veritable ‘world-historical leap’.28
German’s entrance into the League of Nations represented the
zenith of the ‘Stresemann era’. Ardent annexationist of the war
years, opportunistic tactician during the Kapp–Lüttwitz putsch,
Gustav Stresemann had since matured into a Vernunftrepublikaner
and a statesman. As chancellor during the autumn 1923 crisis, he
had done more than anyone else to preserve the unity of the Reich
and the democratic government of the republic. As foreign
minister, he was the champion of peaceful understanding with the
west—though to Germany’s eastern neighbor Poland he was, it is
true, no less ‘nationalist’ than most German politicians on the right
and the left. The foreign minister of the years 1923–9 was an
enlightened representative of German great power politics and an
advocate of a narrower union among the states of Europe. He could
be both, since in his perspective there was no contradiction
between the two. No German politician of the post-war period was
so highly respected abroad. On 10 December 1926 he was awarded
the Nobel peace prize along with his French colleague, Aristide
Briand.
The firmest pillar in Stresemann’s foreign policy was the Social
Democratic Party. When the German Nationalists, dissatisfied with
the western concessions in Locarno, abandoned the Luther
801
government in October 1925, the SPD stepped into the breach and
helped pass the treaties in the Reichstag on 27 November. It would
have been natural to make this support contingent upon
participation in the administration, but the SPD made no such
demand.
In June 1926 it let another chance for a Grand Coalition slip by.
On 12 May 1926, Luther’s minority bourgeois cabinet was
compelled to step down after a conflict over flags. (The cause was a
cabinet decision on 1 May allowing the embassy and consular
agencies to fly the black, white, and red commercial flag along
with the black, red, and gold of the Reich.) The successor, a
bourgeois minority cabinet under Marx, was open to SPD
participation. But now the Social Democrats paid the price for their
active support during the previous months of a KPD-sponsored
plebiscite—the first on the federal level—involving a petition and
referendum on the expropriation without compensation of the
former German princes. The referendum, which took place on 20
June 1926, fell short of its goal, since only 36.4 per cent of eligible
Germans voted for the bill, well below the necessary majority.
After this episode of common extra-parliamentary action with the
Communists, the SPD did not find it within its power to return
immediately to a politics of ‘class compromise’ with the bourgeois
centre. The Grand Coalition, which Stresemann had also supported,
did not come off.
In mid-December 1926 arose yet a third opportunity to replace
bourgeois minority governance with a majority government in the
shape of a centre–left alliance. At Stresemann’s behest, the Marx
administration offered the Social Democrats a Grand Coalition as a
802
way of avoiding a debate over the armed forces, which the SPD
was pushing for. The latter refused. On 16 December 1926 Philipp
Scheidemann gave a speech before the Reichstag that went down in
the annals of German parliamentary history. To the outrage of all
the bourgeois parties, the former prime minister of the Reich spoke
of the secret weapons production going on in the country and
described how its financing was being covered up. He detailed the
interplay between the Reichswehr and extremist right-wing
organizations, and mentioned the so-called ‘black Reichswehr’, the
numerous small-calibre shooting clubs the army used to circumvent
the limit of 100,000 troops. And he shocked the Communists by
disclosing that their cell at the port of Stettin had been fully
informed about the arms and munitions shipments Soviet ships had
unloaded there in September and October.
On 17 December, the day after this spectacular speech, the
parliament brought down the Marx government by 249 against 171
votes. The DVP, DNVP, and KPD joined the SPD vote of no
confidence. The question of the Grand Coalition was now
definitively put to rest; after Scheidemann’s speech, no bourgeois
politician would have supported or even seriously considered
one.29
The outcome of the governmental crisis in the winter of 1926–7
was a centre-right administration under Wilhelm Marx, which took
office on 29 January 1927. In this fourth Marx cabinet, which
included the Centre Party, BVP, DVP, and DNVP, the German
Nationalists provided the ministers of the interior, justice,
agriculture, and transport. Interior Minister von Keudell, who, as
Landrat of Königsberg in the Neumark (East Brandenburg), had
803
cooperated with the Kapp–Lüttwitz government in March 1920,
was the main target of criticism from the left, a target he offered on
numerous occasions.
On 27 November 1927, for example, the DNVP interior minister
sent a telegram to the German Students’ Association assuring them
of his ‘inner solidarity’. This was a demonstration of sympathy in
the conflict between the umbrella organization of the German
Students’ Associations and the Prussian government. The German
Student Association also included the Austrian organizations that
had expressly excluded Jews from membership. In September 1927
the non-aligned Prussian minister of education and ecclesiastic
affairs, Carl Heinrich Becker, one of the great university reformers
of the Weimar Republic, had prevailed upon the Braun
administration to strip Prussian student associations of their
government recognition, since they persistently refused to leave the
German Students’ Association. In a letter to Chancellor Marx on 27
November, Braun threatened to break off all contact between the
Prussian government and the federal interior minister if incidents
like the latter’s supportive letter to the German Students’
Association were repeated. Motions of no confidence, brought
forward by the SPD and KPD on account of the ‘Keudell case’,
failed on 6 December 1927 against a majority of the government
coalition.
The new government’s agricultural policy also bore the German
Nationalist trademark. The DNVP had already effected a return to
the grain tariffs of 1902 in August 1925, under Luther’s
chancellorship. Amendments in July 1927 extended and, in some
cases, increased the existing tariffs for a number of agricultural
804
products, including potatoes, sugar, and pork. On the whole,
however, the government of the bourgeois bloc was in no way
unambiguously ‘reactionary’. In May 1927 the Republikschutzgesetz
was extended for two years—though in somewhat milder form—
with the votes of the German Nationalist Party. In July of the same
year the Reichstag passed, by an overwhelming majority, the most
important social policy reform measure of the Weimar Republic,
unemployment insurance. As the unions had demanded, the law
changed the previous welfare programme into an insurance fund,
into which employers and employees paid equal contributions, 3
per cent of the wage. This insurance programme, along with an
employment agency, was administered by an independent federal
office, the Reichsanstalt für Arbeitsvermittlung und
Arbeitslosenversicherung, with district and local branches. There
were also standardized administrative boards on all levels,
composed of representatives of the insured, the employers, and the
public corporations, all with an equal number of votes.
The unemployment insurance programme represented the
greatest increase in social security achieved by workers and
employees during the Weimar Republic. The 1927 law took up one
of the central ideas of Bismarck’s social policy, social responsibility
as the joint obligation of business and government. For the costs of
the unemployment insurance were not borne by employers and
workers alone, but by the state, too. In emergency situations, the
Reich would have to provide a loan to the federal unemployment
agency if the latter’s ‘emergency funds’ proved insufficient. There
was no provision for direct state allocations, however. The
government, parliament, and organizations apparently did not
805
consider mass unemployment on a gigantic scale to be a threat. If
such a situation occurred, the 1927 system was doomed.
Indeed, there would hardly have been such broad support for
the reform if unemployment had been high in 1927. The numbers
were especially low in July, with 630,000 unemployed workers
drawing state support. The year 1927 was also the high water mark
for German industrial profits during the Weimar Republic. Not only
workers and employees benefited, but also civil servants. The
December 1927 salary reform raised state pay by an average of 16–
17 per cent, with the greatest increases on the lower levels. With
this reform, Finance Minister Köhler, a Centre politician, sought to
address the fact that state salaries had increased much less than
workers’ wages since 1924. Nonetheless, it caused serious problems
for public budgets, and Heinrich Brüning, the Centre Party’s budget
expert, abstained from the final vote on 15 December 1927. His
concerns about the financial consequences of the law were shared
by the employers’ associations and the reparations agent, the
American Parker Gilbert, who had written a memo in October
accusing municipalities of irresponsible financial behaviour. But
1928 was an election year, and to insist on financial solidity was to
swim against the current.
By the time the salary reform law was passed, the disintegration
of the bourgeois bloc was already in sight. Ever since July 1927,
the coalition partners had been fighting over an education measure
introduced by Interior Minister von Keudell. The bill proposed
placing Christian interdenominational schools (the Simultanschule)
and confessional schools on the same legal footing. It was endorsed
by the Centre, BVP, and DNVP. The DVP, the heirs of the National
806
Liberals of the Kulturkampf, rejected it, pointing out that the
Weimar constitution had codified the priority of the
interdenominational school. On 15 February 1928 Count Westarp,
head of the DNVP parliamentary fraction and leader of the
coalition council meetings, found himself forced to announce that
no agreement on the controversial question was possible and that
the coalition was thus dissolved.
It seemed as though every kind of parliamentary majority in
Weimar contained the seeds of its own destruction. For a Grand
Coalition, the crisis zone was social policy. For a coalition of the
right, it was foreign policy and cultural issues. The parties, which
the era of constitutional monarchy had not habituated to the
necessity of compromise, tended continually to regard their
individual goals as non-negotiable. Even ‘state’ parties behaved this
way again and again, as if the decisive line of division still ran
between the government and the parliament, as in the time before
October 1918, instead of between the parliamentary majority and
the opposition, as the logic of the parliamentary system dictated.
The government was often considered the ‘adversary’ even by
parties playing an important role in it. This legacy from the
Wilhelmine era goes far towards explaining the weakness that
characterized German parliamentary rule even during the relatively
peaceful years of the first republic.
On 31 March 1928 President Hindenburg dissolved the
Reichstag and set new elections for 20 May. But 31 March was
important for another reason, too. On this day, the Reich Council
decided concerning a project that would plunge the next
government into its first crisis: ‘battleship A’. The navy planned this
807
ship as the beginning of a series of replacement vessels and wanted
to commit the legislature to a long-term programme lasting several
legislative periods. In December 1927 the Reich Council, led by
Prussia, had come out against the spending for the project. In the
Reichstag, however, the bourgeois bloc put together a majority that
approved the first budgetary instalment at the end of March. On 31
March the Reich Council responded by requesting that the acting
cabinet not authorize work on the battleship until further
examination of the finances and in any case until after 1 September
1928. Since the government had to rely on the cooperation of the
Reich Council more than usual in the ensuing weeks, Groener, the
non-aligned minister of defence (who had replaced the worn-out
Gessler on 19 January 1928), saw himself forced to agree to this
stipulation.
‘Battleship A’ provided the leftists parties with a rousing
campaign theme. The KPD, which, under Ernst Thälmanns
leadership, had become more and more the obedient tool of Stalin,
opposed the battleship with the popular demand for free meals for
school children. (The bourgeois parliamentary majority had
rejected the 5 million marks intended for this programme.) The
slogan ‘Food for kids, not battleships!’ was also used by the Social
Democrats, who thus made themselves seem more radical than
they actually were. At its party congress in Kiel in May 1927, the
SPD had left no doubt about its resolve to prevent a new rightist
cabinet and, election results permitting, to assume governmental
power.
On the far right margin of the political spectrum, a consolidated
NSDAP entered the campaign in the spring of 1928. Adolf Hitler
808
was the uncontested Führer of the National Socialists. Ever since
the ‘Führer congress’ in February 1926, the left wing around the
brothers Otto and Gregor Strasser, which was strong in northern
Germany, no longer provided a counterweight to the party
headquarters in Munich. Though the NSDAP still represented itself
as a ‘socialist’ party friendly to the worker, it was clear already
before the election that its greatest resonance was not in the large
cities, but in hard-hit rural areas, especially after the collapse of
pork prices in 1927, the prelude to a worldwide agricultural crisis.
It was the rural population Hitler had in mind with his new,
binding interpretation of point 17 of the 1920 party platform: the
demand for uncompensated expropriation of land for community
purposes referred, in the new reading, only to unlawfully acquired
property, above all that of ‘Jewish land speculation companies’.
In society as a whole, however, there was little sense of an
impending crisis on the eve of the election. The economic data
were positive, and the unemployment figures were below those of
the previous year. The democratic forces of Germany had never
entered a Weimar parliamentary election with such reason for
optimism as they did the election of 20 May 1928.30
The radiant victors of the fourth Reichstag election were the Social
Democrats with 29.8% of the vote, a 3.8% increase over the
previous election in December 1924. The greatest losers were the
809
German National People’s Party, which fell from 20.5% to 14.3%.
Of the moderate bourgeois parties, the Centre’s loss of 1.5% was
the worst, the two liberal parties each falling by 1.4%. If there had
been a 5 per cent hurdle in the Weimar Republic, the DDP would
have been out of the parliament; it managed only 4.9%. Pure
interest-group parties, on the other hand, did comparatively well.
The national party of small and medium-sized businesses, called
the Economic Party for short, grew from 3.3% of the vote to 4.6%.
The newly founded Christian National Farmers’ and Rural People’s
Party (Christlich-Nationale Bauern- und Landvolkpartei) achieved
2.9% in its first election. On the far left, the KPD increased its share
of the vote from 9% to 10.6%, while on the extreme right, the
NSDAP had to accept a national average of only 2.6%. In several
zones of agricultural crisis on the west coast of Schleswig-Holstein,
however, the popularity of the National Socialists was truly
sensational. In North Dithmarschen they gained 28.9%, in South
Dithmarschen no less than 36.8%.
The election outcome allowed practically only one kind of
parliamentary majority: a Grand Coalition, which finally came
together on 28 June 1928 after arduous negotiations. This was not
yet a formal coalition government, but at first only a ‘cabinet of
personalities’. The political independence of the ministers
suggested by this term was an illusion. In fact, the objections of the
DVP to the coalition—led by the head of the SPD, Hermann Müller
—were so strong that it required an ultimatum from Stresemann
before his party would temporarily agree to two DVP cabinet
members (himself as foreign minister and Julius Curtius as minister
of the economy).
810
The government had only been in power a few weeks when it
was plunged into a serious crisis. On 10 August 1928, the cabinet
approved the construction of ‘battleship A’, against which the SPD
had actively campaigned before the election. The Social
Democratic finance minister, Hilferding, was unable to raise fiscal
objections, since the project’s costs were covered by cuts in other
areas of the military budget. Thus the confirmation of the previous
parliament’s decision was correct. What is more, a negative vote
from the Social Democratic ministers would have meant the
immediate end of the Müller administration. But many members
and supporters of Germany’s largest party saw things differently,
and they had a powerful ally: Otto Wels, the de facto leader of the
SPD as long as party co-chair Müller was chancellor. On 31
October, after the end of the parliament’s summer break, Wels
brought forward a motion by the SPD fraction to halt construction
on the ship and use the savings for school lunch programmes.
If this motion was a slap in the face for Hermann Müller, even
more so was the notion that the chancellor and SPD ministers
should vote with the fraction and against the cabinet decision of 10
August. But this is exactly what happened on 16 November 1928 in
the Reichstag plenum. By endorsing the motion, Chancellor Müller,
Interior Minister Carl Severing, Finance Minister Rudolf Hilferding,
and Labour Minister Rudolf Wissell all but registered a vote of no
confidence in their own administration.
The effect on the public was disastrous. Though the government
was spared an actual defeat, since all the bourgeois parties and the
National Socialists voted against the SPD motion and were able to
bring it down, the strongest party in the government had
811
nonetheless caused great harm to the image of parliamentary
governance in Germany. The Vossische Zeitung rightly accused the
Social Democrats of lacking credibility. Wels had given ‘an
opposition speech of the heaviest calibre’, the liberal Berlin paper
wrote.
812
governing parties were interested in reducing this burden as
quickly as possible. But the reparations agent himself wanted to
revise the Dawes Plan. As long as Parker Gilbert was authorized to
determine whether Germany’s balance of payments and currency
situation justified a transfer of the reparations, the Germans could
hide behind him, as it were. Gilbert considered this state of affairs
harmful and wanted to use a new treaty to force Germany to
become economically independent.
The Paris negotiations resulted in the Young Plan, named after
Owen D. Young, the American director of the conference of experts
that ended on 7 June 1929. According to this plan, Germany was to
pay reparations for nearly six decades, until 1988. During the first
ten years, the annuities would be below their average amount of 2
billion marks. Thereafter they would rise, and not decrease again
for thirty-seven years. No further foreign monitoring of German
finances was planned, nor mortgages on industry and federal
revenue. The German government was to take over responsibility
for the transfer from the reparations agent. It was granted the right
to distinguish between the ‘protected’ and ‘unprotected’ portions of
the reparations, the latter to be paid unconditionally and on time,
the former subject, upon application, to deferment of up to two
years. The payments would be received by a new agency, the Bank
of International Settlements in Basel. If Germany had difficulty
making the payments, it could have recourse to an international
board of experts. This body would have to make suggestions for the
revision of the Young Plan in the event that the German economy
was unable to fulfil its obligations. Another eventuality was also
covered: if the United States granted its Allied debtors a debt
813
reduction, two-thirds of that amount was to be subtracted from the
German reparations.
Compared to the Dawes Plan, the Young Plan had one great
advantage for Germany: it restored the country’s economic
sovereignty. The termination of transfer protection was a
disadvantage, however, since it meant that the Reich would now
have to keep paying reparations during an economic depression.
And the prospect of having to pay them for fifty-eight years was
bleak. Still, there was a kind of political recompense for this
harshness. The German government’s acceptance of the Young Plan
made France willing to accommodate Germany on the Rhineland
question. On 30 August 1929 a conference in The Hague between
Great Britain, France, Italy, Belgium, Japan, and Germany came to
an end with the signing of an agreement for early withdrawal from
the Rhineland. The Allied troops were to pull out of the second
zone by 30 November 1929 (the occupation of the first zone had
been ended in the winter of 1925–6); the third and last zone would
be vacated on 30 June 1930, five years before the date fixed in the
Treaty of Versailles.
The staunch right did not wait for the outcome of the
negotiations before mobilizing its followers against the Young Plan.
On 6 July the National Committee of German Agriculture
(Reichsausschuss der deutschen Landwirtschaft) declared the results of
the negotiations economically unacceptable. Two days later the
heavy-industrial Langnam association, whose actual name was the
Association for the Protection of the Common Economic Interests in
Rhineland and Westphalia (Verein zur Wahrung der gemeinsamen
wirtschaftlichen Interessen in Rheinland und Westfalen), declared that
814
the experts’ report would overburden the German economy. On 9
July a National Committee for the German Referendum
(Reichsausschuss für das Deutsche Volksbegehren) convened in Berlin.
The Pan-German League was represented by Heinrich Class, the
Steel Helmet (Stahlhelm), a paramilitary League of Front Soldiers
(Bund der Frontsoldaten) founded at the end of 1918, by its national
leader Franz Seldte, the DNVP by the film and press magnate
Alfred Hugenberg, who had assumed the party leadership in
October 1928, and the NSDAP by Adolf Hitler. This meeting
produced a manifesto calling on the German people to fight against
the Young Plan and the ‘war guilt lie’ and launching the drive for a
plebiscite.
While the right was gathering its forces, the gulf between the
moderate and the radical left was growing deeper. In the summer
of 1928 the Sixth World Congress of the Communist International
in Moscow had set a course sharply to the left, defining the new
general party line in terms of a theory of the ‘third period’.
According to this doctrine, the relative stability of capitalism,
which had superseded the acute revolutionary post-war crisis in the
autumn of 1923, was now at an end throughout the world. The
hallmarks of the ‘third period’ of post-war development were the
increasing contradictions within the capitalist order and the
preparation of an imperialist war against the Soviet Union. The
outcome of the new crisis was historically inevitable. Led by the
Communist parties, the labouring masses in the capitalist countries
and their colonies would rise up against the capitalist systems of
exploitation and bring about a proletarian victory through
revolutionary war. The most pressing task for the Communist
815
International was to fulfil the precondition for the final battle: to
destroy the ‘bourgeoisified’ Social Democratic movement, whose
ideology of class cooperation now demonstrated many points of
contact with fascism.
The reasons for the Comintern’s ‘ultra-leftist’ turn were to be
found both within the Soviet Union itself and within Germany. In
the Soviet Union, a power struggle was raging between Stalin, the
general secretary of the Communist Party, and a group around
Nikolai Bukharin. The latter group, stamped as ‘rightist’ by its
enemies, spoke against Stalin’s programme of forced
collectivization of agriculture and accelerated industrialization. If
the other parties of the Third International could be committed to
an offensive against rightist tendencies, this would help Stalin in
his battle against Bukharin. The German cause of the leftward shift
lay in the fact that a Grand Coalition under a Social Democratic
chancellor was in power after June 1928. The SPD was regarded as
the party that had done more than any other to promote
rapprochement with the western powers, especially with France.
This alone made it, in Stalin’s view, a dangerous political enemy of
the Soviet Union.
Were it not for Bloody May Day in Berlin, the Sixth World
Congress’s declaration that the Social Democratic parties were
growing closer to the fascists and would therefore have to be more
strongly fought might have remained mere abstract theory. But in
the spring of 1929 the Social Democratic president of the Berlin
police, Karl Friedrich Zörgiebel, unintentionally came to Stalin’s
aid. Zörgiebel upheld a ban on open-air assemblies and
demonstrations, decreed by him the previous December, for 1 May,
816
the workers’ traditional ‘day of struggle’. When the Communists
ignored the ban and erected several barricades, the police promptly
moved against them with armoured vehicles and firearms. The
outcome of the battle was 32 dead, all civilians, almost 200
wounded, and far more than 1,000 arrests.
The police action was followed by an administrative measure, a
country-wide ban of the League of Red Front Fighters (Roter
Frontkämpferbund), a paramilitary group founded in 1924. The KPD
responded in June in Wedding, the part of the capital that had seen
the heaviest fighting at the beginning of May. A party congress
originally planned for Dresden was convened there. Bloody May
Day and the ban of the Red Front Fighters were proof to the party
leadership that the Social Democrats were becoming ‘social
fascists’. Ernst Thälmann went so far as to call the ‘social fascism’
of the SPD an especially dangerous form of fascist development.
The delegates hailed the party head in a manner that can only be
described as a ‘cult of the leader’. According to the minutes of the
meeting, he was received with calls of ‘bravo!’ and prolonged
applause even before commencing his twohour speech. ‘The Party
Congress welcomes Comrade Thälmann with tumultuous applause.
The delegates rise to their feet and sing the “International”. The
youth delegation greets the First Chairman of the Party with a
threefold “Hail Moscow!”’
The political radicalization on the left was closely connected
with increasing unemployment. In February 1929 economic
recession brought the unemployment rate above 3 million for the
first time, and the usual spring recovery was weak; there were still
2.7 million people out of work in March. The federal
817
unemployment agency was only able to cover 800,000 workers
with ‘primary support’ from its contribution fund and was forced to
take out a loan from the government. Since the government lacked
the money, the finance minister had to seek help from a bank
consortium. Only this unorthodox solution prevented the collapse
of the federal unemployment agency.
At this point it was clear to all that the federal finances could
not be straightened out without a reform of the unemployment
insurance programme. Unfortunately, on no issue were the two
wing parties of the Grand Coalition so divided as on social
insurance policy. The SPD, in conjunction with the Free Trade
Unions, supported increased contributions from employers and
workers. The DVP, with a view to the business community, rejected
this solution, demanding that contributions be reduced.
At the end of September, despite countless conferences of
experts, still no agreement was in sight. On 1 October Chancellor
Müller indicated for the first time that he might resign if the
government could not resolve the problem. But the DVP yielded
that same day: if the SPD and Centre were willing to postpone their
proposed half per cent increase in the contributions until December
1929, the People’s Party would enable (by abstaining from the
ballot) the passage of a bill lowering the rates of unemployment
support and correcting defects in the insurance programme. The
Social Democrats and Centre agreed, and the bill passed on 3
October. The Grand Coalition had passed its most difficult test to
date.
By the time the president of the Reichstag announced the results
of the ballot, the man who had done the most to save the Müller
818
government was no longer alive. In the early morning of 3 October
1929 Gustav Stresemann died of a stroke. Long in a declining state
of health, the foreign minister had spent his last reserves
attempting to prevent any change of government that threatened to
undermine his rapprochement policy in the parliament. In order to
secure his foreign policy on the ‘right’, Stresemann had
occasionally behaved more nationalistically than he actually was.
But he firmly held to his belief that the desired revision of
Versailles did not justify starting a new war. The prerequisite for a
foreign policy based on this belief was a cooperation between the
moderate forces within the bourgeoisie and the working class.
Since Stresemann knew this, he was the staunchest advocate of the
Grand Coalition within his party, and his death considerably
weakened this alliance. The only statesman the Weimar Republic
produced was soon to prove—both domestically and in terms of
foreign policy—irreplaceable.31
819
representatives, who, contrary to § 3, sign treaties with foreign
powers, are subject to the punishments as per § 92, no. 3 of the
Criminal Code.’ This paragraph sanctioned treason with a prison
term of no less than two years. In its first draft, § 4 had read
somewhat differently, threatening imprisonment to ‘Chancellor,
Ministers, and authorized representatives of the Reich’—that is,
including the Reich president. Since Hindenburg was an honorary
member of the Stahlhelm (as of 1924), Franz Seldte, the
organization’s national leader, pushed through an amendment with
the help of the German Nationalists excepting the head of state
from the threat of punishment.
The ‘liberty law’ cleared the first hurdle, though barely: 10.02%
of the voting population participated in the referendum, just 0.02%
more than required by the constitution. The Reichstag was thus
forced to deal with the bill, and did so between 27 and 30
November 1929. That it would fail was clear from the start, given
the overall distribution of votes. How the DNVP would act—that is,
to what extent Hugenberg would be able to impose his will on the
fraction—was very much an open question, however. The vote on
the imprisonment paragraph revealed that he by no means enjoyed
the full support of his party; only 53 out of the 72 DNVP delegates
voted in favour. Hugenberg’s sharp counter-measures split the
fraction. At the beginning of December twelve delegates, including
former interior minister von Keudell, landowner Hans Schlange-
Schöningen, Walter Lambach, director of the German Nationalist
Commercial Employees’ Federation, and the retired lieutenant
commander Gottfried Treviranus, left the DNVP and joined
together to form the German Nationalist Cooperative Union
820
(Deutschnationale Arbeitsgemeinschaft). The DNVP floor leader,
Count Westarp, resigned from his post in protest against
Hugenberg.
The referendum on the ‘law against the enslavement of the
German people’ took place on 22 December 1929, when 5.8 million
Germans, or 13.8% of eligible voters, came out in support of the
bill. Since 21 million were necessary to pass it, the failure of the
National Committee was obvious. But it was significant that more
than a fifth of the electorate voted in favour of the bill in nine of
the thirty-five electoral districts and that Hitler was well on his
way to being recognized by ‘decent society’ as a political partner.
With the participation of the NSDAP in the National Committee,
the leader of the 1923 putsch had achieved an important
intermediate goal: the established right reckoned with him and
granted him access to financial means that benefited the further
advancement of the NSDAP.
By the autumn of 1929 there was no more mistaking that the
National Socialists were on the rise. They achieved great gains in
all elections they entered in November and December: for the
parliaments of Baden and Thuringia, for the city parliament of
Lübeck and the provincial assemblies of Prussia, for the
municipalities in Hesse and Berlin. The latter election, which was
held on 17 November, saw the Social Democrats fall from 73 to 64
seats and the DDP from 21 to 14, whereas the NSDAP, which had
not yet been represented in the city council assembly, obtained 13
seats right away. The National Socialist conquest of German
universities began at about the same time. The National Socialist
German Students’ League (Nationalsozialistischer Deutscher
821
Studentenbund) was the big winner in the General Student
Committee (Allgemeiner Studentenausschuss; AStA) elections of the
1929–30 winter semester. In Würzburg it achieved 30%, at the
Technical University in Berlin 38%, and in Greifswald no less than
53%.
The shift to the right among the students was an expression of
social protest. A young generation of academics was rebelling
against its ‘proletarization’ and declaring war against the ‘system’,
which it held responsible for its financial difficulties and uncertain
career prospects. Hatred of the Weimar state and animosity
towards the Jews went hand in hand. While the Jews represented
only 1 per cent of the population, they formed 4–5 per cent of the
student body. The percentages were even higher in certain fields,
like medicine and law, and at a number of universities, such as
Frankfurt am Main and Berlin. For many of their non-Jewish fellow
students, this meant that the Jews were taking advantage of unfair
privileges. The advance of the National Socialist student
organizations was based in part on the mass mobilization of social
envy.
We need not search long for the economic causes of the far
right’s popularity in the autumn of 1929. The agricultural crisis
had grown even more serious and radicalized the rural population
in northern Germany. Bomb attacks on finance agencies and
district administration offices had been making headlines
repeatedly since the spring, above all in Schleswig-Holstein. The
number of those looking for work in Germany rose from 1.5 million
in September to 2.9 million in December, 350,000 more than the
same month the previous year. Share prices had reached their
822
highest point in the boom year of 1927. If we equate their level
between 1924 and 1926 to 100 points, 1927 would represent 158.
They fell to 148 points in 1928 and to 134 in 1929. But the most
alarming news came from America. On 24 October 1929, the
infamous Black Friday, the prices on the New York stock exchange
suffered a massive drop. They continued to fall in the ensuing days,
and within a short time the gains of an entire year had been wiped
out.
The cause of the stock market crash was a sustained period of
over-speculation. Trusting in the longevity of the boom, small
shareholders and large investment corporations had invested ever
more money in industry, thus raising production. In October 1929
it became clear that supply had far outpaced demand. The decrease
in stock prices suffered by companies such as General Electric and
investment firms like the Goldman Sachs Trading Company caused
a panic among the shareholders.
The consequences were immediately felt on the other side of the
Atlantic. In order to remain liquid, American banks began to
demand the return of funds they had placed in short-term European
investments. Hardest hit was Germany, where the sum total of
short-term foreign—and that meant primarily American—credit
amounted to 15.7 billion marks in 1929. However, some three-
quarters of short-term and medium-term loans, a considerable
portion of which came directly or indirectly from abroad, were
regularly used for long-term investments. The municipalities, in
particular, tended to be involved in this kind of activity, which had
long drawn the criticism of Parker Gilbert, the reparations agent.
Thus diverted, the funds were practically frozen. They could not be
823
made liquid on demand, but repaid only through new loans.
The Reich, too, found it increasingly difficult to obtain foreign
credit. For the president of the federal bank, Hjalmar Schacht (a
founding member of the DDP at the end of 1918, but now leaning
more to the right), this was a welcome predicament. He used the
looming December deficit as a means of pressuring the government
into committing itself to a long-term reform of national finances.
The cabinet and coalition considered this goal achieved on 14
December 1929, when the Reichstag agreed—despite uncertainty
with regard to its final legislative form—to a fiscal programme
containing an increase in unemployment contributions from 3 to
3.5 per cent, an increase in the tobacco tax, a decrease in direct
taxes for the purpose of shoring up capital formation, and the
announcement of a law dealing with the national debt.
Two days later, however, Schacht declared the cabinet’s short-
term measures to be inadequate. Backed by the reparations agent,
he demanded a sum of 500 million marks for debt repayment in
the 1930 budget (an amount he reduced by 50 million marks in the
negotiations during the ensuing days). On 22 December the
Reichstag passed the necessary legislation, and the government
received a bridging loan from a domestic banking consortium
headed by the federal bank, allowing the Reich to remain solvent.
By this time, Rudolf Hilferding was no longer finance minister.
The Social Democrat had handed in his resignation on 20
December, declaring that ‘outside intervention’ had prevented him
from pursuing his policies. Schacht was indeed the victor in his
duel with Hilferding. But the coalition, government, and finance
minister had only themselves to blame. Their attempt to
824
manipulate the revenue estimates in the spring of 1929 had only
been a trick to avoid real finance reform—not to mention the
questionable finance policies of the previous governments,
especially the fourth Marx cabinet.32
By the end of 1929 and the beginning of 1930, there could no
longer be any doubt about the fact that parliamentary democracy
was in deep crisis in Germany. The power struggle between the
government and the president of the federal bank was not the only
indication. There were also signs that large portions of the ‘power
elite’ were beginning to turn away from the government, if not
entirely from parliamentary rule. The large-scale agricultural
interests, represented by the National Land League, had been
against the Grand Coalition from the start. In December 1929 the
National Association of German Industry sent the Müller cabinet a
memorandum entitled Aufstieg oder Niedergang?,* which demanded
—in language tantamount to an ultimatum—that social
expenditures be adjusted to the productivity of the economy and
that the government have the right to veto spending increases
authorized by the parliament. By the end of 1929 at the latest,
Defence Minister Groener and his closest adviser, General Kurt von
Schleicher, head of the new Ministerial Office (Ministeramt), were
working together with Otto Meissner, the secretary in the office of
the Reich president, for a government without Social Democrats.
Given the state of affairs at the time, this could only have been a
presidential cabinet. Hindenburg himself had come out in favour of
such a reform in spring 1929. Count Westarp, at that time still floor
leader of the DNVP, was one of the first to learn of his intentions.
825
Hindenburg made himself clearer at the beginning of 1930. On 6
January he enquired of Hugenberg, and on 15 January of Westarp,
whether the German Nationalists would support, either directly or
indirectly, a cabinet put together by the president if the financial
reform in February or March should lead to a governmental crisis.
‘There is a great concern’, wrote Westarp, paraphrasing the head of
state, ‘that the opportunity to form an anti-parliamentary and anti-
Marxist administration would then fail because of the DNVP and
that Hindenburg would not be able to free himself from governing
with the Social Democrats.’ The president received two different
answers from his German Nationalist interlocutors: the party head
responded negatively, the former floor leader favorably. At the
beginning of 1930 it was clear that the path from parliamentary to
presidential rule still contained many obstacles.
What held the government together during this period was the
interest all the coalition parties shared in passing the Young laws.
On 20 January, after the details had been discussed for months in
subcommittees of the expert commission, the Young Plan was
passed in The Hague. The most important thing for Germany was
that the payment schedule and total amount remained what the
experts had proposed in June 1929. Eight days later began the last
chapter in the history of the Grand Coalition. On 28 January, at the
suggestion of Heinrich Brüning, who had been elected floor leader
in December, the Centre decided to make its support of the Young
Plan conditional upon an agreement on financial reform. Brüning’s
proposal was neither a repudiation of the Grand Coalition nor of
the new reparations agreement. It was an attempt to use the
coalition’s foreign policy goals as a means of leveraging a reform of
826
the national finances.
Though a small number of SPD delegates wanted to put forward
the counterproposal of a financial reform in the spirit of the Social
Democrats, they were unable to prevail. The great majority of the
fraction staunchly rejected linking domestic and foreign policy
measures in this way. But this attitude had the unintended effect of
weakening the SPD’s negotiating position. On the right wing of the
coalition, the DVP refused any further concessions on
unemployment insurance, increases in direct taxes, and an
emergency levy from civil servants and others with fixed salaries.
At the end of February, it looked like an agreement was no longer
possible.
The president shared this assessment. In a personal interview on
1 March, Hindenburg asked Brüning if the Centre would be willing
to support a different government. Brüning rejected the idea. It was
the shared opinion of the fraction leadership, he said, that the
present coalition was to be maintained as long as possible, in order
to pass the Young Plan and a series of important domestic reforms.
An attempt to carry through these plans without the Social
Democrats would lead to great turmoil, and, in any case, without
the SPD a majority was very uncertain. Towards the end of the
conversation, Brüning summarized the position of the Centre by
telling Hindenburg ‘that we at least demand a commitment of the
party leaders to the finance laws and have all of us the wish that
the current coalition be preserved for a while’.
On 5 March, against the expectations of nearly all observers, the
cabinet of the Grand Coalition came to an agreement on the
proposals to cover the 1930 federal budget. One of the most
827
significant points was the increase in the ‘industry burden’—which
was actually slated to be abolished after the passage of the Young
Plan—from 300 to 350 million marks in January 1930. This
fulfilled the SPD’s demand for a direct property tax, albeit only for
a year. Equally important was another concession by Rudolf
Hilferding’s successor as finance minister, Paul Moldenhauer of the
DVP: the board of directors of the unemployment office was
authorized to raise the contributions from 3.5 to 4 per cent. In
exchange, the Social Democratic ministers agreed that there would
be no refund of the payroll tax in 1931.
The cabinet’s agreement was a triumph of the moderate forces
in all camps, but it was built on sand. On 6 March the DVP
fraction, supported by the Association of German Employers’
Associations (Vereinigung der deutschen Arbeitgeberverbände) and the
National Association of German Industry, rejected key provisions of
the compromise. The BVP, which occupied the postal ministry in
the Müller cabinet, refused to go along with an increase in the beer
tax. On 11 March the president intervened again. In conversations
with Brüning and Müller, Hindenburg declared his willingness to
grant the government the plenary powers of Article 48. This
seemed to accomplish Brüning’s purpose of linking the finance
reform to the reparations agreement. On 12 March the Young laws
were passed at the third reading by 265 against 192 votes and 3
abstentions. Almost all Centre delegates voted in favour.
Whatever Hindenburg’s intentions were on 11 March, however,
his ‘camarilla’ was determined to exploit the new situation
presenting itself after the passage of the Young laws for a decisive
shift to the right, away from a parliamentary and towards a
828
presidential system. Already on 18 March, heavy-industry circles in
the German People’s Party learned that the president, ‘apparently
at the instigation of Groener and Schleicher’, had decided not to
permit the Müller cabinet to use Emergency Article 48 after all. On
19 March Hindenburg, in an imperious tone, demanded that the
government pass measures to help east German agriculture. His
state secretary, Otto Meissner, remarked on this step to General
von Schleicher in the following words: ‘This is the first stage of
your solution! It also lays the foundation for the best thing we can
have, “for the Hindenburg leadership”.’33
Knowing the president’s plans, the DVP could afford to adopt a
relatively moderate stance vis-À-vis the Social Democrats at its
party congress in Mannheim on 21 and 22 March. On 26 and 27
March Brüning again tried to work out a compromise, the main
effect of which would have been to postpone the battle over
unemployment insurance reform. The unemployment office was to
introduce austerity measures, and the government was to decide at
a later date whether it wished to raise the level of the
contributions, lower the level of payments, or increase indirect
taxes for the purpose of financing federal loans. This proposal,
which weakened the cabinet decision of 5 March to the detriment
of the unemployed, obtained a majority of People’s Party votes on
27 March. At the meeting of the SPD fraction, however, union
representatives and Labour Minister Wissell were especially vocal
in their rejection of the ‘Brüning compromise’. Chancellor Müller
and the other Social Democratic members of the cabinet were
among the small minority endorsing the Centre’s proposal. The
cabinet had no choice but to record its failure and tender its
829
resignation to the president.
The decisions made on 27 March 1930 mark one of the deepest
ruptures in the history of the Weimar Republic. In retrospect, it is
clear that the period of relative stability came to an end on this day
and that the first phase in the dissolution of the first German
democracy began. But many contemporaries, too, were conscious
of the significance of the day’s events. On 28 March the Frankfurter
Zeitung spoke of a ‘black day… doubly ominous, because the trivial
cause of the battle stands in such a grotesque disproportion to the
disastrous consequences that may arise from it’. Criticism was also
soon heard in the ranks of the SPD, whose decision had sealed the
fate of the Müller government. In the May issue of Die Gesellschaft,
the theory-oriented journal he edited, Rudolf Hilferding explained
why he could not follow the argument of the party majority that a
reduction in unemployment compensation would have been
unavoidable in the autumn if Brüning’s proposal had been
accepted:
830
other way to vanquish the Weimar social welfare state. This was
the immediate objective of the pioneers of the presidential solution,
not merely the defeat of a minor increase in the unemployment
contributions. It was therefore the right that bore the main
responsibility for what occurred after the collapse of the Müller
government.
The moderate left cheerily accepted the departure from
parliamentary democracy and thus shared the blame for the shift to
a presidential system. The Social Democrats could have prevented
the disintegration of the Grand Coalition at the end of March—
though only at the price of a party crisis and probably only for a
short time, since the coalition would hardly have outlasted the
passage of the Young laws, its most important goal, in the autumn
of 1930. Nonetheless, it would have been the right thing for the
SPD to have stepped onto the bridge Brüning had thrown, for the
self-reproach the Social Democrats now had to suffer was bitter. At
the decisive moment, they had not done everything within their
power to preserve parliamentary democracy and to prevent a
relapse into the authoritarian state.34
831
had been an officer at the front in 1915 after a broad course of
study in history and political science, which he completed with a
doctoral thesis in national economy. He had been wounded and
decorated in the war. The fact that he had been the director of the
Christian-national trade unions after 1920 preserved him from
party-internal accusations of being a ‘rightist’. On the other hand,
he had also gained high regard in Conservative circles as a finance
expert in the Reichstag, to which he had belonged since 1924.
While his Catholicism was a liability in the view of Kulturkampf
liberals, for the architects of the presidential state it was, if
anything, an asset. Through Brüning’s mediation, political
Catholicism became one of the buttresses of the silent
constitutional revolution Hindenburg and his circle sought to
effect.
Initially, the Brüning government was not an openly presidential
cabinet, but a covert one. In addition to Groener, the non-aligned
minister of defence, it included representatives from the bourgeois
parties that had participated in the previous administration—the
Centre Party, DVP, DDP, and BVP—as well as one minister each
from the DNVP, the Economic Party, and the People’s National
Reich Association (Volksnationale Reichsvereinigung), formed by
former members of the DNVP and the Farmers’ and Rural People’s
Party. But since Martin Schiele, the German Nationalist minister of
agriculture, resigned his seat in the Reichstag upon entering office,
it was not clear at first how his party would react to the Brüning
cabinet. And in fact the DNVP voted inconsistently several times in
April 1930, delivering narrow parliamentary minorities to the
government—very much against the will of Hugenberg.
832
Then, about a quarter of a year after Brüning’s appointment,
there arose the kind of situation Hindenburg had had in mind when
he made him chancellor: a government bill to cover the budget was
rejected in the parliament’s tax committee. Thereafter, on 16 July,
the president made it officially known that he had authorized the
chancellor to put the budgetary programme into effect on the basis
of Article 48 in case it should fail to pass the Reichstag, and to
dissolve the latter if it should call for the annulment of the
emergency measures or pronounce a vote of no confidence in the
chancellor.
Speaking for his party, Rudolf Breitscheid, the SPD floor leader,
immediately protested that the purpose of Article 48 was ‘to help
the state and to protect the state when need arises, not to help a
particular government out of its predicament when it cannot find
the majority it is looking for’. But government by emergency
decree could now no longer be averted. On 16 July, after the
rejection of the budget bill in the plenum assembly of the
Reichstag, Brüning announced that the government had no interest
in further debate. The first two emergency measures went into
effect that same day, though they lasted only two days. On 18 July
the Reichstag accepted the SPD’s motion to lift the measures,
whereupon it was immediately dissolved by the president. New
elections were set for 14 September. On 26 July Hindenburg
decreed a new ‘emergency measure to remedy financial, economic,
and social crises’. Among other things, it introduced a ‘civic tax’
(Bürgersteuer) that, unlike in the previous emergency measures, was
graduated to a certain extent. Moreover, it formed the legal basis
for a programme of federal assistance for people on fixed incomes,
833
an income tax supplement, a tax on unmarried people
(Ledigensteuer), and—unavoidable in view of increased
unemployment—an increase in unemployment contributions from
3.5% to 4.5%.
The transition from covertly to openly presidential rule in July
1930 was characterized by a certain ineluctability. Four months
before, the president had rejected rule by parliamentary majority;
now, the July crisis represented the fulfilment of the law that had
guided the new chancellor into office on 30 March. Brüning was
unable to accommodate the Social Democrats without alienating
the right wing of the governing coalition, which would have been
at cross-purposes with the logic of his appointment. The Social
Democrats, for their part, could not accept the non-graduated civic
tax the government had insisted upon up to the dissolution of
parliament without causing massive harm to their supporters’ sense
of justice and handing the Communists a cheap victory. The main
protagonists of the July crisis had so little room to manoeuvre that
a parliamentary solution was all but impossible.
The parliamentary elections were preceded by attempts to
concentrate the forces of the bourgeois parties. The success was not
great. The DDP allied itself with the People’s National Reich
Association, the political arm of the conservative and—for the
period, relatively moderately—anti-Semitic Young German Order
(Jungdeutscher Orden), to form the German State Party (Deutscher
Staatspartei). This deeply upset many of the DDP’s Jewish
supporters, and not only them. The industrial interests promoted
the Conservative People’s Party (Konservative Volkspartei), formed
at the end of July by the volkskonservativ forces around Treviranus
834
and the anti-Hugenberg faction around Count Westarp. On the left,
everything remained as it had been, organizationally speaking. The
only innovation was the vehement nationalism put on display by
the KPD. In its 24 August ‘programatic declaration on the national
and social liberation of the German people’, the Communist party
said that the leaders of the SPD were
835
BVP’s 3% was down from 3.1% two years previously. Much greater
were the losses of the SPD, which still remained Germany’s largest
party. It fell from 29.6% to 24.5%. The newly founded
Conservative People’s Party, together with the ‘Guelf’ German-
Hanover Party (Deutsch-Hannoversche Partei), achieved all of 1.3%
of the vote.
The National Socialists were the main beneficiaries of the
increased electoral participation. Yet the non-voters in previous
elections were not the main source of ‘brown’ success. Most of the
NSDAP constituency had previously voted for other parties. A
methodical calculation has revealed that probably one in three
DNVP voters, one in four DVP or DDP voters, one in seven non-
voters, and one in ten SPD voters chose Hitler’s party in the 1930
elections. The conservative and liberal ‘camps’ thus contributed far
more to the rise of National Socialism than the Social Democrats.
Several other facts can also be considered certain. Protestants were
twice as susceptible to the NSDAP as Catholics. Farmers,
government officials, the self-employed, retirees, and pensioners
were, proportionally speaking, more strongly represented in the
NSDAP constituency than in the general working population. The
opposite was true of workers and employees. The unemployed
(whose official number was somewhat more than 3 million in
September 1930) contributed little to National Socialist success;
they were far more likely to vote for the party of Ernst Thälmann
than that of Adolf Hitler.
National Socialism’s appeal to the middle classes was so obvious
that the Social Democratic sociologist Theodor Geiger, writing in
the autumn of the same year, interpreted the NSDAP success as the
836
expression of a ‘panic in the Mittelstand’. Though accurate, this
diagnosis was not the whole story. The National Socialists won
over not only bourgeois and rural voters who were dissatisfied with
their previous parties or who had not voted before; they also
succeeded in penetrating the working class. It was to their
advantage that a significant number of German workers had never
developed the kind of proletarian class consciousness that led
others to vote for one of the ‘Marxist’ parties.
Thus the NSDAP of 1930 was, in a purely sociological sense, a
‘people’s party’, far more so than all other parties, which drew their
main support from a particular social or confessional ‘milieu’. It
was true that, by 1930, these social and confessional ‘milieus’ were
not nearly as segregated from each other as during the imperial
era. Gramophone, film, and radio had begun to prepare the way for
a new mass culture that would ride roughshod over the boundaries
between social groups. But the ‘old’ parties barely recognized the
challenge that lay in these developments. The National Socialists,
on the other hand, exploited the new media of mass
communication to great effect, responding to a widespread need for
a sense of community beyond the confines of estate, class, and
confession—a need that was especially vigorous in the younger
generation, but which had lain politically dormant until this
moment. As backwards-looking as were many of its promises to its
voters, the NSDAP owed its success mainly to its ability to adapt to
the conditions of the era of mass culture and, in this sense, to
demonstrate its ‘modernity’.
The National Socialist answer to the need for community was no
different in 1930 than it had been in the years before: extreme
837
nationalism. Nationalism would envelop in one giant embrace
everything that divided Germans from each other. The rhetorics of
anti-Semitism and nationalism often went hand-in-hand. If the
former was less prominent in 1930 than in previous electoral
campaigns, the main reason was because the National Socialists
were out for the votes of the working classes, who were generally
unresponsive to anti-Jewish agitation. The concept of ‘socialism’,
which tended to irritate many bourgeois voters, especially the older
among them, was assiduously reinterpreted by the National
Socialists. Hitler’s version of socialism did not entail the abolition
of private property, but rather an equality of opportunity in society
and an economic philosophy derived from a principle anchored in
the 1920 party platform: Gemeinnutz vor Eigennutz, ‘the common
good before the good of the individual’ or ‘service before self’.
The republican parties, as nationalist as they believed
themselves to be, could not attempt to outdo the nationalism of the
National Socialists. But the pledge of allegiance to the democratic
republic, the answer of Weimar’s staunch defenders to the
challenge from the radical right, was only able to mobilize a
minority. Even within the Weimar coalition, which still held
together in Prussia, republican pathos could achieve but little.
Opinions were too divided on what, precisely, was worth saving
about Weimar. In any case, the most obvious thing to do in light of
the political polarization reflected in the electoral results of 14
September was for each party to return to the sources of its own
strength. For the Social Democrats, this meant binding their
supporters more firmly to socialism as a cultural movement and a
lifestyle. For the Centre, it meant a return to the Catholic part of
838
political Catholicism.
Bourgeois liberalism was a far less clearly delineated and less
tightly ‘networked’ milieu than the social democratic movement or
political Catholicism. Its political convictions were less firmly
entrenched. Both liberal parties, the DVP no less than the DDP,
responded to the rightward migration of their voters by re-
orienting themselves in that direction. Indeed, within a few short
months of Stresemann’s death, there was very little left of the DVP
that could be called ‘liberal’. In January 1930 it joined a rightist
government in Thuringia in which the National Socialists, with
Wilhelm Frick, occupied the ministry of the interior and public
education (until April 1931). For the German Democratic party, the
alliance with the anti-Semitic People’s National Reich Association
in the 14 September elections had not paid off. Middle-class
tradesmen continued to abandon the party, and, presumably, no
small number of disillusioned Jews went over to the Social
Democrats. A few short weeks after the elections, the newly
founded German State Party split. On 7 October, led by their High
Master, Artur Mahraun, the Young Germans announced their
departure from the party, citing ‘insurmountable philosophic
differences’. Henceforth, the only reminder of the July 1930 fusion
was the new party name, which no longer contained the word
‘democratic’.36
The downfall of German liberalism prompted not only many
Jewish voters to look for a new political home, but also one of the
most famous German writers. In October 1930 Thomas Mann
delivered his ‘German Address’ in Beethoven Hall in Berlin.
Continually interrupted by hostile calls from the audience, Mann
839
exhorted the German bourgeois to take their political place at the
side of the Social Democrats. This party, he said, stood for the
peaceful rebuilding of Europe and had provided the most
dependable support for the policies of Gustav Stresemann.
840
the spring only to allow them a share in the power of the Reich
half a year later, and the governmental right wing thought the
same. Conversely, there was massive resistance within the SPD to
any kind of cooperation with Brüning and the forces backing him.
Max Seydewitz, a Reichstag delegate from Saxony, spoke for the
party left wing when he wrote in the 15 September issue of the
journal Klassenkampf that the intentions of the Centre Party
chancellor were no less fascist than the methods proposed by the
National Socialists, and it thus made no sense ‘for the Social
Democratic movement to differentiate between Brüning’s and
Hitler’s fascism in its struggle for democracy and against fascism’.
For the governmental camp, a widening to the right was just as
impossible as a widening to the left. A coalition with the NSDAP
was unthinkable for the bourgeois centre parties, and the army and
industrial interests also did not consider the National Socialists fit
for government in autumn 1930. Even Hitler’s spectacular
performance in the trial for high treason of the Reichswehr officers
from Ulm, Scheringer, Ludin, and Wendt, was as yet unable to
change this fact. Summonsed by the defence of the three young
National Socialists before the federal court in Leipzig on 25
September 1930, Hitler maintained under oath that the NSDAP
would take power only by legal means. However, after the
presiding judge enquired about his comment that heads would roll
in the sand after a National Socialist victory, Hitler then added that
a national court would be created through the normal legislative
process with the task of condemning the criminals of November
1918. Their execution would be accomplished by legal means, that
is.
841
Since neither the National Socialists nor the Social Democrats
were acceptable, the bourgeois minority cabinet under Brüning was
forced into toleration agreements. Political reasons militated
against seeking NSDAP support, which Hitler strictly rejected in
any case. Consequently, there was no realistic alternative to an
arrangement with the Social Democrats.
The SPD leaders shared this assessment. From their post-election
point of view, there were three main reasons to work with the
Brüning administration. First, it was the only way to avoid a
government even further to the right, dependent on the National
Socialists. Secondly, the Weimar coalition in Prussia under Otto
Braun would be in great danger if the Social Democrats brought
down the national government under the Centre Party chancellor.
The loss of government power in Germany’s largest state would
also have meant giving up control of the Prussian police, the state’s
most powerful force in dealing with National Socialists and
Communists. Thirdly, there was a wide field of substantive
agreement between the Social Democrats and the government
camp, based on the shared realization that the consequences of the
questionable ‘debt economy’ after 1924 could be overcome only by
means of a rigorous cost-cutting policy. The reform consensus did
not preclude dissent over the distribution of the social costs of
austerity, but it was not undone by this continuing antagonism.
The foundation of the cooperation policy was laid at the end of
September in confidential interviews between Brüning and the
Social Democratic Party leadership, with Rudolf Hilferding and
Hermann Pünder, the state secretary of the federal chancellery,
playing active roles. On 3 October the SPD fraction passed a
842
resolution justifying its support of the government. In view of the
election outcome, the party considered its foremost task to be the
preservation of democracy, the safeguarding of the constitution,
and the defence of parliamentary governance. Furthermore, the
SPD was fighting for democracy in order to defend the social
welfare programmes and raise the living standards of the working
class. ‘While safeguarding the existential interests of the laboring
masses, the Social Democratic fraction will stand up for the defence
of the parliamentary foundation and for a resolution of the most
pressing fiscal concerns.’
The Weimar Republic had experienced many fiery debates in the
Reichstag, but none were as tumultuous as those of 17 and 18
October 1930. In protest against a ban on uniforms in Prussia, the
107 National Socialists entered the main hall of the parliament
dressed in the brown shirts of the SA. Gregor Strasser, the leader of
the party’s national organization (Reichsorganisationsleiter),
announced that the National Socialists, ‘anti-parliamentarians on
principle’, were practically compelled to become ‘the defenders of
the Weimar constitution’ against the current dictatorial plans of the
bourgeoisie. ‘We are now for the democracy of Weimar, we are for
the law in defence of the republic, as long as it suits us. And we
will demand and maintain every position of power on the basis of
this constitution for as long as we wish.’
The Communist speaker, the delegate Wilhelm Pieck, did not
trouble himself with drawing a cloak of legality over the subversive
plans of his party. There was, he declared, only one way to bring
down the accursed system of capitalist exploitation and slavery:
843
Revolution and thus the destruction of capitalism and the removal
of all those who support this system. That is the task the
Communist party has set itself, and the day will come when the
labouring masses, when the unemployed, under the leadership of
the Communist Party, will drive out this parliament of businessmen
and fascists. Then the German Soviets will come together in its
stead and erect the Dictatorship of the Proletariat, setting up a free
socialist Soviet Germany in place of this rotten bourgeois society
and this republic of starvation.
844
4.5 to 6.5 per cent, a 6 per cent cut in the salaries of government
employees, and new measures for the protection of agriculture,
including higher customs duties for wheat and barley. On 7
December the Reichstag adjourned until 3 February 1931.
Vorwärts welcomed its departure. Three months after the new
elections, as the 13 December issue of the SPD party organ
announced, everybody was probably of one mind that ‘this
Reichstag is a failure and that we can be glad if we hear and see
nothing of it.’ The party’s floor leader in the Prussian parliament,
Ernst Heilmann, who was also a delegate in the Reichstag, wrote
that, in reality, a Reichstag with 107 National Socialists and 77
Communists could not function effectively. ‘A people that elects
such a parliament is effectively renouncing self-governance. And its
legislative privilege is automatically replaced by Art. 48.’ In a
broadcast speech on 17 December, Otto Braun argued that if the
Reichstag, partly owing to its infiltration by anti-parliamentary
groups, was unwilling and unable to accomplish the tasks conferred
upon it by the constitution, ‘then, but only then, the political SOS-
signal must be given, and the safety valve of the constitution must
be opened long enough to deal with the acute crisis the parliament
was unable, or unwilling, to master.’ Vorwärts published Braun’s
speech under the title ‘An Education in Democracy’.37
When the Social Democrats convened in Leipzig on 31 May
1931 for their first party congress after their departure from power,
the policy of cooperation drew a great deal of criticism.
Nonetheless, the negative views were outweighed by the support
for the main argument of the policy’s defenders: ‘National
Socialism has been held back from governmental power by us,’
845
declared Wilhelm Sollmann, deputy floor leader of the fraction,
846
On 5 June 1931, the day the SPD congress in Leipzig came to an
end, President Hindenburg decreed a new emergency measure, one
that had been anticipated for some time. The harshness of its social
impact was worse than all expectations. Unemployment payments
were reduced by an average of 10–12 per cent. The salaries of state
officials and company employees were cut by between 4 and 8 per
cent. Invalids and disabled veterans received lower pensions. The
Social Democrats joined in the general cry of outrage. Their
demand that the Reichstag—or at least the budgetary committee—
be convoked met with unqualified rejection by Brüning, who
believed the country to be on the brink of insolvency, if not of civil
war. The only prospect he held out to the Social Democrats was
that some of the harsher social consequences might be mitigated in
the provisions for the measure’s enforcement. In order to force the
SPD to back down, he threatened to terminate the Prussian
coalition. This worked. On 16 June the Social Democratic fraction
withdrew its motion to convoke the budgetary committee.
The party left wing revolted. On 1 July it published an
‘admonition’ against the continuation of the cooperation policy.
This was to become the seed of a new organization, the Socialist
Workers’ Party of Germany (Sozialistische Arbeiterpartei
Deutschlands), founded at the beginning of October. But the
majority of the SPD was not prepared to break with Brüning and to
give up power in Prussia. In the July issue of Die Gesellschaft,
Rudolf Hilferding spoke of the ‘tragic predicament’ of his party,
resulting from the collision between the economic crisis and the
situation of political emergency created by the elections on 14
September 1930:
847
The Reichstag is a parliament against parliamentary rule, its
existence a threat to democracy, to the working classes, to foreign
policy… Defending democracy against a majority that rejects it,
and doing so with the political tools of a democratic constitution
that presupposes a functional parliament—it is almost like squaring
the circle, the problem presented to the Social Democrats to solve.
It is truly a situation that has never occurred before.
It was not until well into 1931 that most Germans began to
realize that the nadir of the economic crisis was still to come and
that the world was in the middle of a Great Depression. On 20 June
there was still what appeared to be a ray of hope. The American
president, Herbert Hoover, proposed a year-long international ‘debt
holiday’ from the payment of compulsory state debts, including the
German reparations, and on 6 July, after the USA had overcome
French resistance, the Hoover Moratorium really did take effect.
Shortly afterward, however, Germany was shaken by a bank crisis
that further undermined confidence in capitalism and the market
economy. The buying and selling of foreign exchange was
rigorously cut back. The drastic increase of the minimum lending
rate and the rate for loans on security had a fatal effect on the
prostrate economy. Tax revenues were used to rehabilitate the
banks, amounting to a partial nationalization of the banking
system.
In September 1931 the Brüning government was faced with a
serious foreign policy crisis when the project of an Austro-German
customs union ended in a debacle. Julius Curtius, Stresemann’s
successor in the foreign ministry, was the main responsible party,
848
but in a larger sense Brüning too was at fault. The customs union,
which Vienna had also been pushing since 1930, was designed to
strengthen German influence in central and south-eastern Europe,
prepare the way for eventual political union between the two
German-speaking countries, and, above all, demonstrate renewed
German self-confidence both domestically and internationally—an
agenda in full agreement with Brüning’s emphatically nationalist
foreign policy.
When, on 18 March 1931, the cabinet committed itself to the
customs union, the Wilhelmstrasse foresaw an unfriendly reaction
from France. But it was not prepared for the intensity and the
sharpness with which Paris responded to the initiative. Austria,
which was in dire need of international and especially French
assistance, was so impressed that it began to withdraw from the
project in May. On 3 September Curtius, along with his colleague
in Vienna, Johannes Schober, announced to the European
committee of the Council of the League of Nations in Geneva that
the two countries would no longer pursue the project. Two days
later, in response to a Council request motivated by Great Britain,
the International Court of Justice in The Hague decided by eight
votes against seven that the customs union violated the 1922
Geneva Protocol on the economic and financial reconstruction of
Austria.
After the humiliating failure of his Austria policy, Curtius’s
position as foreign minister became untenable. On 3 October he
asked the chancellor to apply to the president for his resignation.
By this point, however, it was no longer simply a matter of a
change of leadership in the foreign office. In September Schleicher,
849
followed by Hindenburg, had exhorted the chancellor to perform a
decisive turn to the right. Brüning sought to accommodate these
demands by reshuffling his cabinet on 9 October. He himself took
over the foreign office. Defence Minister Groener also temporarily
assumed the interior ministry, succeeding the ‘leftist’ Centre
politician and former chancellor, Joseph Wirth. Curt Joël, the arch-
conservative state secretary, advanced to minister of justice.
The DVP, Curtius’s party, was no longer represented in the
Brüning cabinet. Its heavy-industry wing had demanded on 3
October that it go over to the opposition and bring forward a
parliamentary motion of no confidence in the government. This the
party leadership and Reichstag fraction agreed to do one week
later. Thus, one year after the beginning of the cooperation policy,
the chancellor was presented the bill for his occasional concessions
to the Social Democrats, on whom he depended in parliament. The
right wing of the business camp broke with Brüning because his
policies were not right enough for the industrial right.38
On 11 October the conservative portion of the Ruhr industrial
interests had the chance to go one step further and publicly join
ranks with the ‘nationalist opposition’. On that day, the parties and
associations of the staunch right held a military parade in Bad
Harzburg. But apart from Ernst Brandi, one of the coal mine
directors of the United Steel Works Corporation, no prominent
heavy industrialist took part in this meeting, which had been
initiated by Hugenberg. Even Brüning’s harshest critics in the
business world evidently still had reservations about openly joining
the radical right.
850
The ‘Harzburg front’ was made up of the NSDAP, DNVP,
Stahlhelm, National Land League, and Pan-German League, along
with numerous members of former ruling houses, the former chief
of the Army High Command, General von Seeckt (from 1930 a DVP
delegate in the Reichstag), and Hjalmar Schacht, former president
of the Reichsbank, who had resigned his office in March 1930 in
protest against the Young Plan. Schacht’s attacks on the bank
succeeded in starting a hectic debate that lasted several days.
Hitler, who had been officially received by Hindenburg for the first
time on the previous day, caused a stir by leaving the parade
before the Stahlhelm formations could follow his SA—a consciously
provocative gesture intended to demonstrate his independence
from the ‘old’ right.
The events in Harzburg made it easier for the Social Democrats
to come to terms with the more rightist second cabinet under
Brüning. The massive attacks against it by the ‘fascist reaction’
were by themselves almost sufficient to make the government seem
tolerable to the SPD. Schacht’s elucidations on monetary policy
inspired the headline ‘The Harzburg Inflation Front’ in the 12
October issue of Vorwärts. On this point there was full agreement
with Brüning, who was equally opposed to any experimentation
with the currency. In the short parliamentary session that began
two days after the Harzburg assembly, the chancellor could once
again count on the support of the SPD. On 16 October the votes of
the Social Democrats were decisive in rejecting all motions of no
confidence.
851
HINDENBURG’S RE-ELECTION AND THE FALL OF
BRÜNING
852
have to be immediately adjusted to the shattered conditions of the
world economy.
That meant nothing more nor less than a total revision of the
Young Plan. But the reparations conference planned in Lausanne at
the beginning of 1932 in response to the report did not interest
Brüning in the slightest, since, in his view, it would lead only to a
new moratorium and a reduction in the amount Germany owed,
that is, to a provisional half-measure, not to the desired complete
and final end to the reparations payments. Consequently, the
German government sought to postpone the conference, and was
successful. Set to begin on 25 January 1932, the meeting was
called off on 20 January. The domestic price for this foreign policy
decision was momentous. The drastic deflationary course continued
unabated, and social destitution and political radicalization spread.
At the beginning of 1932 all of German domestic political life
was focused on the presidential election planned for the spring of
the year. Brüning would have preferred a quick parliamentary
solution to the problem: Hindenburg’s reelection by the Reichstag
with the two-thirds majority necessary to revise the constitution.
But the German Nationalists and National Socialists refused to go
along. For his part the incumbent, who had celebrated his eighty-
fourth birthday in October 1931, did not want to run in a popular
election unless he had sufficient backing from his friends on the
right, those who had urged him to run in 1925.
That he would have this backing was not at all certain at the
beginning of 1932. Among those involved in the ‘Hindenburg
committee’, whose 1 February declaration called for the re-election
of the aged field marshal, were the poet Gerhart Hauptmann, the
853
painter Max Liebermann, the director of the National Association of
German Industry, Carl Duisberg, the High Master of the Young
German Order, Artur Mahraun, and two former defence ministers,
Otto Gessler and Gustav Noske. But none of the leaders of the
‘nationalist associations’ or big agriculture had signed the
document. Since the Stahlhelm, which counted Hindenburg as an
honorary member, did not want to vote for him, the Kyffhäuser
Union, of which he was honorary president, also hesitated to
endorse him openly. Only on 14 February did the organization’s
central executive come out in support of the president, and one day
later Hindenburg finally announced that, in recognition of his
‘responsibility for the fate of our Fatherland’, he would be available
for possible re-election.
Hindenburg’s announcement prompted the parties of the
moderate right and centre to publicly side with him. The Harzburg
front disintegrated. The Stahlhelm and the German Nationalists
resisted the National Socialists’ pretensions to leadership and, on
22 February, nominated their own candidate, the Stahlhelm deputy
national leader, Theodor Duesterberg. That same day Joseph
Goebbels, district leader (Gauleiter) of the NSDAP in Berlin,
declared in the Sportpalast that ‘Hitler will become our president.’
Four days later, the leader of the National Socialists had himself
appointed to an official post, that of Regierungsrat, in the Brunswick
legation in Berlin. With this step Adolf Hitler, born in Austria,
stateless since 1925, obtained the last thing that stood in the way
of his candidacy for the Reich presidency, German citizenship.
The far left had had a candidate ever since 12 January: Ernst
Thälmann, nominated by the central committee of the KPD as the
854
‘red workers’ candidate’ for the succession to Hindenburg. The
Comintern and the Communist party leadership expected Thälmann
to draw a large number of votes from the Social Democratic
workers if the SPD opted to endorse the incumbent. This
calculation was not entirely chimerical. It was true that, since the
beginning of the cooperation policy in October 1930, the members
and supporters of the SPD had put up with a great deal that ran
diametrically counter to the traditional ideas of the party.
Nonetheless, an election recommendation for the staunch
monarchist Hindenburg would have been beyond the pale for many
Social Democrats.
The National Socialists also sharply attacked the SPD for its
expected endorsement of Hindenburg. On 23 February, the first day
of a short parliamentary session characterized by constant tumult,
Goebbels caused a great stir by remarking that Hindenburg was
‘praised by the Berlin asphalt press, praised by the party of the
deserters’. He meant the Social Democrats, on whose behalf the
delegate Kurt Schumacher, a seriously wounded volunteer of 1914,
then hurled back the wildly applauded rejoinder that the whole
National Socialist agitation was ‘a continual appeal to the inner
bastard in people’, and that if anything about National Socialism
deserved recognition, it was the fact ‘that, for the first time in
German politics, it has successfully accomplished the total
mobilization of human stupidity’.
On 26 February, the last day of the session, the SPD made its
decision for Hindenburg officially known. The election statement
by the party leadership declared that on 13 March, the day of the
first presidential ballot, the German people would stand before the
855
question of whether Hindenburg should remain or be replaced by
Hitler.
856
Duesterberg, who managed 6.8%.
To assure the victory of the incumbent, 173,000 more votes
would have been enough. Unlike in 1925, Hindenburg did
especially well in all those areas where Social Democrats had their
strongholds and where the Catholic proportion of the population
was greater than the average. In Evangelical-rural areas, in
contrast, where he had won decisively seven years before, his
results were far below the national average. Outside of Bavaria,
Hindenburg lost among his erstwhile loyal voters and won among
his former adversaries.
Although Hitler won 5 million more votes than his party
achieved during the last parliamentary election in September 1930,
his chances of defeating Hindenburg in the now mandatory second
ballot were small. The Communists decided to run Thälmann again,
despite the fact that it was now even more clear than before that it
was a choice between Hitler and Hindenburg. Faithful to Stalin’s
maxim of November 1931 that the ‘main blow in the working class’
was to be aimed at the Social Democrats, the KPD emphasized the
main purpose of the ‘battle candidacy of Comrade Thälmann’: it
was a matter of ‘clearly exposing the character of the SPD as the
moderate wing of fascism and the twin brother of Hitler-fascism’.
Duesterberg did not run in the second ballot. The Stahlhelm
recommended abstention, and the German Nationalists opted
against ‘active participation’ in the vote.
The second ballot was held on 10 April 1932. By evening it was
obvious that the incumbent had a clear mandate for a second term
in office. Hindenburg obtained 53%, Hitler 36.8%, and Thälmann
10.2% of the vote.
857
Hindenburg’s triumph was, in great part, the result of the SPD’s
policy of cooperation. If the Social Democrats and their supporters
had not had the chance to get used to a ‘policy of the lesser of two
evils’ since the autumn of 1930, it would have been virtually
impossible to convince them in spring 1932 that they had to vote
for a dyed-in-the-wool monarchist in order to prevent a National
Socialist dictatorship. But precisely this was the choice. Nobody but
the field marshal could count on the backing of a portion of the
traditional right in addition to what support remained for the
Weimar coalition, thus relegating Hitler to second place. The Social
Democrats knew as well as anybody that Hindenburg was no
democrat. So far, however, the second president of the Reich had
shown himself to be a man of law and justice, one who respected
even the constitution he did not love. As things stood, nothing
more could be redeemed in the presidential elections of 1932. But
compared to what was yet again avoided on 10 April, the
proclamation of the ‘Third Reich’, it was a great deal.
For the victor, however, the election outcome was tainted.
Hindenburg was deeply upset that he owed his victory not to the
right, but primarily to the Social Democrats and Catholics. He took
his resentment out on the man who had been his most active
campaigner, Heinrich Brüning, finding a convenient excuse to
upbraid the chancellor in the ‘emergency measure to secure the
authority of the state’ of April 13, which banned Hitler’s private
armies, the SA (Sturmabteilungen or Storm Troopers) and the SS
(Schutzstaffeln or Security Force). The ban had been pushed by the
interior ministers of the most important states—Prussia, Bavaria,
Württemberg, Baden, Hesse, and Saxony—and was based on
858
information the Prussian police had secured in house searches in
mid-March concerning the secret military policy of the National
Socialists.
This information had prompted even General von Schleicher,
head of the ministerial office in the defence ministry, to support
the ban of the SA and SS at first. Even before the second ballot of
the presidential election, however, he had reversed his position.
Using his former regiment comrade Oskar von Hindenburg, son of
the field marshal, as a go-between, Schleicher succeeded in
persuading the president that the ban was not politically
opportune, since it would necessarily lead to a new conflict with
the right. Hindenburg did reluctantly sign the measure, but two
days later he went behind Groener’s, the defence minister’s, back to
obtain information from General von Hammerstein, head of the
army command, concerning the activities of the Reichsbanner
Schwarz-Rot-Gold, a republican paramilitary organization founded
in 1924 and dominated by Social Democrats. In Hindenburg’s view,
this information was incriminating and thus justified banning the
organization.
That did not happen. Groener, who was also temporary interior
minister, determined the information provided by the defence
ministry to be of no importance and came to an agreement with the
leader of the Reichsbanner, Karl Höltermann, on a tactical
measure: the elite units of the Reichsbanner, the so-called Schufos
(short for Schutzformationen), were sent on leave. The emergency
measure of 13 April remained in effect, but Groener had gained
three influential adversaries in the course of the conflict: Kurt von
Schleicher, who had been his Kardinal in politicis up to that point,
859
and both Hindenburgs, the father and the son.
Two weeks after the second presidential ballot, most Germans
were called to the ballot box yet again. On 24 April elections to
state parliaments took place in Prussia, Bavaria, Württemberg,
Anhalt, and in the Free and Hanse City of Hamburg. The NSDAP
achieved massive gains in all five Länder, and in all of them but
Bavaria—where the BVP was able to maintain an advantage of two
seats—it was now the strongest party. In Prussia, the Weimar
coalition lost the majority, but without the rightist parties, NSDAP,
DNVP, and DVP, being able to put one together on their own
account. On 12 April, in its last session before the election, the
previous parliament—that is, the governing majority—had
provided for this eventuality, voting to change election procedure.
Hitherto, a second ballot for the prime minister was to be
conducted as a run-off between the two most viable candidates; a
relative majority was sufficient. After the change, an absolute
majority of votes was needed for the second and all subsequent
ballots.
The effect was the same as a constructive vote of no confidence.
The parliament could only replace the incumbent head of
government by voting for a successor with a majority. The Vorwärts
made surprisingly clear what the Social Democratic initiators of the
amendment were aiming to accomplish: if, after 24 April, the
‘national opposition’ and the Communists should constitute a
negative majority, it would be up to the KPD to decide whether it
would bring the right to power by restoring the old procedure.
There had been cooperation between the extreme right and the
extreme left in Prussia in the summer of 1931 when the
860
Communists, on instructions from Stalin, joined the parties and the
organizations of the ‘national opposition’ in their campaign for a
referendum to dissolve the Prussian parliament. The negative
majority that did not yet exist on 9 August 1931, the day of the
referendum, was a reality by 24 April 1932. Still, it was unlikely
that the Communists would assume the responsibility for the
election of a prime minister from or close to the National Socialist
Party.
On 25 April the executive committee of the Comintern, together
with representatives of the KPD and the ‘Revolutionary Trade
Union Opposition’, published a manifesto that adopted a new tone
with regard to the Social Democrats. The manifesto ‘To the German
Workers’ announced the Communists’ willingness ‘to fight together
with any organization in which workers are organized and that
really wishes to do battle against the reduction of wages and
supports’. While this proclamation, too, contained the typical sharp
attacks against the leaders of the SPD and the Free Trade Unions,
the ‘unitary battle front’ against the ‘capitalist thieves and the
increasingly shameless fascist bands’, an alliance it considered long
overdue, was no longer expressly limited to a ‘unitary front from
below’, that is, directed against the leaders of the reformist
organizations. The May manifesto of the KPD central committee
expressed the new message in one short sentence: ‘National
Socialist participation in government would be a dangerous step
along the way to open, bloody dictatorship.’
The crisis in the Prussian parliament could only have been
solved in the form of a ‘black–brown coalition’. The NSDAP and
Centre held talks about such an alliance, as they had done before in
861
Hesse after parliamentary elections on 15 November 1931. But
since Brüning was unwilling to give Hitler’s party the key positions
of prime minister and interior minister, the chances for an
agreement were small. On 24 May the newly elected parliament
convened for its first session. The Braun government announced its
resignation on the same day, but remained in office in a caretaker
capacity, since there was no majority for a restoration of the old
procedure.40
On 9 May 1932 began a four-day Reichstag session. It would
form the immediate prelude to Brüning’s dismissal. Events were set
in motion by Groener, who, on 10 May, delivered a speech that
was drowned out by a flood of scornful heckling from the National
Socialist delegates and that, as Goebbels recorded in his journal,
proved to be the minister’s ‘funeral dirge’. Brüning tried to limit the
damage. His speech the next day emphasized the approaching
reparations conference in Lausanne and urged parliament and
public not to lose composure ‘in the last hundred metres before the
finish line’.
On 12 May the government, with the help of the SPD, once
again won all votes. By this point, however, the Reichswehr
leadership, headed by Schleicher, had decided not only to break
with Groener, but also to bring down Brüning. Groener himself
declared on 12 May that he intended to step down as defence
minister and to concentrate solely on the interior ministry (which
he had been directing in a temporary capacity). This would have
required Hindenburg’s assent. But the president, at the instigation
of Schleicher, insisted that Groener leave the government entirely.
Before leaving to spend his Whitsun holidays at Neudeck, his estate
862
in East Prussia, Hindenburg ordered Brüning to make no changes in
personnel while he was away.
The ban on the SA was one but not the only reason for
Schleicher’s break with Brüning. By April 1932 at the latest, the
politically active general had become convinced that the crisis of
the German state could only be solved in cooperation with National
Socialists. On 28 April and 7 May he conducted secret talks with
Hitler. The second of these (if not also the first) dealt with the
conditions under which the NSDAP would be willing to tolerate a
cabinet reformed or reshuffled towards the ‘right’. By 7 May
Schleicher knew Hitler’s price: the dissolution of the Reichstag,
new elections, and the lifting of the ban on the SA and SS.
Hindenburg knew about these interviews. What further weakened
Brüning’s position in the eyes of both Schleicher and Hindenburg
was the fact that the chancellor and the foreign minister had
returned on 30 April from the disarmament conference in Geneva
with almost completely empty hands. The press had good reason to
think that the chancellor would not be able to maintain himself in
office for very long after Groener had announced his resignation as
defence minister.
While Hindenburg was vacationing in Neudeck, another power
elite long desirous of Brüning’s fall got involved in the action—the
Junkers. The National Land League, firmly in the hands of the
‘national opposition’ by the autumn of 1930, was the only
significant economic interest group that had come out in support of
Hitler before the second presidential ballot. It was the government
itself that, on 21 May, provided the most important agrarian
umbrella organization with the watchword it needed to launch a
863
large-scale campaign against Brüning. The national commissioner
for ‘eastern assistance’ (Osthilfe), Hans Schlange-Schöningen,
presented the parliament-approved draft of a settlement decree that
held out the possibility of obtaining excessively indebted estates for
the state by ‘private contract’ (freihändig) or via compulsory
auction, and using them for farming settlements.
Immediately after the draft was made public, Count Kalckreuth,
president of the National Land League, and Ernst Brandes,
president of the German Agricultural Conference (Deutscher
Landwirtschaftstag), along with several of the Land League’s
administrative offices, lodged complaints with Hindenburg. The
message was uniform: the right to compulsory auction, as Baron
von Gayl, director of the East Prussian Landgesellschaft, put it,
represented a further ‘descent into state socialism’ and weakened
the ‘power of resistance of those circles that have traditionally
upheld the national animosity towards the Poles’. Men like Elard
von Oldenburg-Januschau, who owned estates in the vicinity of the
president’s, attempted to persuade him in the same direction in
personal conversations.
The pressure soon had the desired effect. On 25 May
Hindenburg informed Schlange-Schöningen through State Secretary
Meissner that he could not approve the measure in its current form.
A resolution of the German Nationalist fraction two days later
called the settlement measure ‘unqualified Bolshevism’. By 27 May
at the latest, Hindenburg was unable to retreat from his position;
the German Nationalists were standing in the way of the rightward
shift he sought to effect.
After his return to Berlin, on 29 May, the president received
864
Brüning in order to tell him that he awaited the resignation of the
government. In the late morning of 30 May, the chancellor
informed the cabinet of his conversation with Hindenburg. Shortly
before noon, Brüning submitted his resignation to the head of state.
The interview lasted only a few minutes. At noon, Hindenburg
received a parade of the Skagerrak Guard, an honorary naval unit,
before the entrance to the presidential palace.41
Of all the chancellors of the Weimar Republic, Heinrich Brüning
is historically the most contested. Partisan favour and hatred have
distorted the evaluation of his character, making it difficult to
assess. In the view of some, he systematically undermined the
foundations of German democracy and thereby inadvertently
prepared the way for Hitler. Others see him as the representative of
a conservative alternative both to failed parliamentary governance
and to National Socialist dictatorship. According to the second
interpretation, Brüning’s policies were historically necessary to a
great extent; the path to catastrophe did not begin until his fall
from power.
The truth is that the parliamentary system had already failed by
the time Brüning became chancellor on 30 March 1930. After the
collapse of the Grand Coalition, the shift to an openly presidential
system was only a matter of time. Brüning became the executor of
policies, the guidelines of which were largely determined by the
president and his circle. Economically he represented the broad-
based, supra-party reform consensus, which basically amounted to
a deflationary policy, until well into the second half of 1931. There
was, moreover, an ‘objective’ obstacle standing in the way of an
alternative, anti-cyclical economic policy until the end of 1931 and
865
the beginning of 1932, the unsolved reparations problem. Brüning
could only have changed his economic course after it became clear
that a return to the Young Plan was impossible. But he did not wish
to do so, believing that a reparations compromise would have had
a deleterious effect on national prestige, and because of his far-
reaching foreign policy objectives. And it is quite likely that he
even had little choice about the latter, since a more elastic foreign
policy would certainly have met with Hindenburg’s veto.
The president’s position was so strong that the question of
Brüning’s own longrange goals is only of limited interest. In exile
and in his memoirs (published in 1970, shortly after his death), the
chancellor of the years 1930–2 maintained that he worked
assiduously for the restoration of the monarchy, seeking in this way
to erect a barrier against a dictatorship of the National Socialists.
There can be no doubt about Brüning’s feelings for the pre-war
imperial order. In a cabinet meeting on 24 February 1932, he made
it clear that he had had nothing to do with the events of 9
November 1918, having been on that day among the troops ‘who
formed the head of the Winterfeldt group to put down the
revolution’. But there is no evidence from his term in office that
Brüning was actually working for the restoration of the monarchy.
His retrospective explanation is obviously a self-stylization, an
attempt to monumentalize himself as a conservative statesman of
farreaching vision.
In reality, Brüning was the half-willing, half-unwilling executor
of a politics that cannot be adequately described as ‘conservative’.
The real centre of power in the late republic was Hindenburg and
his camarilla. These men aspired to set up an authoritarian state in
866
which the will of the people would find only very limited
expression. Brüning, in contrast, was content to restrict the rights
of the Reichstag, especially with regard to state expenditures (a
practical reform policy he successfully implemented in February
1931). He believed the National Socialists could be domesticated,
but made their participation in government subject to conditions
they could not accept without radically changing their political
identity. Like the head of the Centre Party, the prelate Ludwig
Kaas, Brüning endorsed a shift to the right in German politics, but
wanted to adhere strictly to the constitution. When, in spring 1932,
Hindenburg and his circle decided to abandon all consideration for
the toleration policy of the Social Democrats and to accommodate
the National Socialists to a greater extent than Brüning considered
acceptable, the chancellor was forced to go.
The fall of Brüning was of profound historical significance. On
30 May 1932 the first, moderate, parliamentarily tolerated phase of
the presidential system came to an end, and a second,
authoritarian, openly anti-parliamentary phase began. The leaders
of the Reichswehr and the great east Elbian landowners who had
brought about the change in government wanted to enlist the
National Socialists as junior partners, so to speak—not in order to
let them rule, but to transform them into supporters of their own
regime. The fulfilment of Hitler’s conditions for accepting the
rightist cabinet necessitated the dissolution of the Reichstag, whose
legislative period was not scheduled to come to an end until
September 1934. If new elections had taken place at that time,
Germany would have looked very different than it did in the
summer of 1932. The popularity of the extremist parties would
867
have receded in the course of the expected economic recovery. The
new, anti-parliamentary presidential system of Hindenburg and the
Prussian power elite destroyed this opportunity. They brought the
crisis of state to a dramatic head and Germany into a situation
virtually impossible to master by constitutional means.42
Franz von Papen (born in 1879), whom Kurt von Schleicher chose
as Brüning’s successor, was a former officer of the general staff, a
landowner in Westphalia, an enthusiastic equestrian, the principal
shareholder of the Centre Party newspaper Germania, and a
member of the board of directors in several agricultural interest
groups. Until the elections on 24 April 1932 he was a delegate in
the Prussian parliament, one of the Centre Party’s backbenchers
furthest to the right. In Schleicher’s calculation, Papen would bind
the Centre Party to the new government just as Brüning had done
before him. On 31 May, however, immediately after the new
chancellor was authorized by Hindenburg to form his cabinet, the
Centre leader, Monsignor Kaas, made it clear to Papen that the
party would look upon any attempt on his part to take up Brüning’s
succession as treason. But when Hindenburg appealed to his sense
of patriotic duty, Papen decided for the chancellorship and left the
Centre Party.
The ‘strong man’ of the new cabinet was General Kurt von
Schleicher, who was 50 years of age at the time. A native of
Brandenburg, Schleicher took over the defence ministry, entering
868
the full limelight of public life for the first time. The new minister
of the interior was Baron Wilhelm von Gayl, who had been the
director of the East Prussian Landgesellschaft. The minister of
agriculture, Baron Magnus von Braun, was also from East Prussia.
Like Gayl, he had belonged to the DNVP before entering the
cabinet. The foreign ministry fell to the man who had been the
ambassador in London up to this time, Baron Konstantin von
Neurath, who was close to the German National People’s Party and
was a native of Württemberg. The new finance minister was a
staunch monarchist, Count Lutz Schwerin von Krosigk, who from
1929 had been the assistant head of the department he now led.
Among the ministers were one count, three barons, two further
members of the nobility, and only three bourgeois—a state of
affairs that prompted Vorwärts to the famous headline of 1 June,
‘The Cabinet of Barons’.
On 4 June the president dissolved the Reichstag, thus fulfiling
one of the conditions Hitler had placed on his toleration of the new
government. New elections were set for 31 July. On 14 June
Hindenburg signed the Papen cabinet’s first emergency measure.
Based on preliminary work of the Brüning government, it
decreased unemployment payments by an average of 23 per cent
and shortened the period of coverage from twenty to six months.
Thereafter, practically all social welfare claims and obligations
were ended, replaced by a system on a level far below what was
generally considered the ‘subsistence level’. Two days later, the
government fulfilled a promise Schleicher had made to Hitler on 4
June: the ban on the SA and the SS was lifted and the wearing of
uniforms once again generally permitted.
869
On the same day, 16 June, the reparations conference
(originally planned for January, but postponed at Brüning’s
entreaty) began in Lausanne. Papen could now reap the rewards of
his predecessor’s firmness. According to the agreement, which the
new chancellor signed on 9 July, Germany’s final payment would
be no more than 3 billion marks, which was to be paid only after
the expiration of a three-year period and over a longer period of
time in the form of national debenture bonds—provided the
economic equilibrium had been fully restored in the interim.
Though the treaty’s ratification by the parliaments in Paris and
London still depended on whether the United States was prepared
to accept an inter-Allied debt settlement, it was highly unlikely that
Germany would ever have to pay reparations again, apart from the
final sum, which was little more than symbolic.
This foreign policy coup, which only the Social Democrats and
the liberal press recognized as such, did nothing to calm the
domestic political scene. The electoral campaign in the summer of
1932 was the bloodiest in German history. The Communists and
National Socialists were responsible for most of the violence.
Immediately following the lifting of the SA ban, many areas of the
Reich experienced murderous clashes between political opponents.
They were especially frequent in the industrial districts of the
Rhine and Ruhr. In Prussia, 3 men died in political riots in the first
half of June, 2 National Socialists and 1 Communist. In the second
half of the month, after the SA and uniform ban was lifted, the
number of politically motivated deaths rose to 17, 12 National
Socialists and 5 Communists. Among the 86 deaths in July, 38 were
National Socialists, 30 Communists. Sundays were particularly
870
sanguinary. On 10 July, for example, 17 people died in Germany,
10 were fatally wounded, and 181 seriously wounded.
Although the escalation of violence and terror on the streets was
clearly connected to the lifting of the SA ban, the ‘cabinet of
barons’, in holding the Prussian police responsible, laid the blame
on the shoulders of the custodial government of Germany’s largest
state. In a ministers’ conference on 11 July, Interior Minister von
Gayl accused Carl Severing—after 1930 once again interior
minister of Prussia—of giving his police force orders to combat the
National Socialist movement, despite the fact that it continued to
grow stronger. Gayl came to the conclusion that a commissioner
from the Reich should be sent to Prussia, and that it would be best
if the chancellor assumed this task and appointed sub-
commissioners. The next day, the government set the deadline for
20 July. But Severing wasted no time in thwarting its plans. A
decree on the same day, 12 July, making it easier to ban outdoor
assemblies and requiring the police to proceed with utmost rigour
against illegal bearing of arms, temporarily undermined the
cabinet’s manoeuvre against Prussia.
What saved the government’s original timetable was the Bloody
Sunday of Altona on 17 July 1932. An unusual combination of bad
political, administrative, and police decisions was partly
responsible for nineteen civilian deaths in the course of a
demonstrative march by the SA through strongly ‘red’ districts of
the—at that time still Prussian—town. Most of the victims were
killed by bullets from the police. Perhaps Severing could have
prevented a federal strike against Prussia by immediately declaring
a state of emergency in Altona. But the Prussian interior minister
871
refrained from a demonstration of strength. The Reich government
acted instead. On 18 July, without consulting the Länder, it
promulgated a general ban on outdoor assemblies and ordered
three members of the Prussian cabinet—Welfare Minister Hirtsiefer
of the Centre Party representing Otto Braun, who had been relieved
of his duties, Interior Minister Severing, and Otto Klepper, the
independent finance minister—to appear at the Reich chancellery
on 20 July at ten o’clock in the morning.
The content of Papen’s interview with the three ministers
became known as the Preussenschlag, the ‘strike against Prussia’.
Using Article 48, the president appointed the chancellor federal
commissioner of Prussia, empowering him to dismiss the members
of the Prussian state ministry, take personal charge of the prime
minister’s affairs, and assign the leadership of the Prussian
ministries to other persons as commissioners of the Reich.
Thereupon Papen announced that, by virtue of his new authority,
he was dismissing Braun and Severing from their offices and
appointing Franz Bracht, the mayor of Essen, to be interior
minister.
Since the actions of the Reich violated both the national and the
Prussian constitutions, the custodial Prussian government
responded with a suit in the federal court. However, the people, or
the masses of workers, did not call upon the Prussian government,
the SPD, the Free Trade Unions, and the Reichsbanner Schwarz-
Rot-Gold to fight against the attack from the Reich government.
Instead, the Social Democrats responded by saying that the ‘cabinet
of barons’ would have to answer for their acts in the parliamentary
elections on 31 July. Young Reichsbanner activists were
872
particularly outraged at the lack of resistance, interpreting it as a
surrender in the face of violence, a judgement that has been echoed
in German historiography.
But there were good, indeed compelling reasons for the Social
Democrats to behave as they did. The ‘people’ had declared on 24
April that they no longer supported the Prussian government. The
outcome of the Prussian parliamentary elections had dealt a heavy
blow to the SPD’s belief in its democratic legitimacy. With
unemployment as high as it was—officially 5.5 million in June
1932, but in reality somewhat higher—a general strike was out of
the question. The situation was very different than in the spring of
1920, during the Kapp–Lüttwitz putsch. At that time, Germany was
experiencing near-full employment. Moreover, the strikers knew
themselves to be in agreement with the legitimate authority of the
state. The Preussenschlag, on the other hand, was ordered by the
newly elected president of the Reich. It would not have been
realistic to expect a large number of officials and policemen to rise
up against him.
Another important difference was that, in the summer of 1932,
the working class was more divided than ever. On 14 July, in
response to signs of creeping ‘social democratization’ within its
own ranks, the KPD had ended the more flexible tactic of the
unitary front initiated in April, warning against ‘any neglect of our
battle against the social-fascist leaders, any obscuring of the
principal antagonism between ourselves and the SPD’. Common
cause with the Communists for the restoration of the Braun
government was simply inconceivable. When, therefore, the
Communists asked the Social Democrats and Free Trade Unions on
873
20 July whether they were prepared for a general strike, it was
nothing more than a rhetorical question. In any case, the
Reichsbanner was neither militarily nor psychologically equipped
to take up armed resistance against the army. In this respect, the
paramilitary organization of the republicans was also inferior to
those of the right, the SA, SS, and Stahlhelm, which would
certainly have taken active part in a battle against the ‘Marxists’. A
civil war would have been the democratic left’s to lose in the
summer of 1932, probably with enormous casualties.
The reasons went back far beyond the year 1932. The SPD’s
passivity in the face of the Preussenschlag was not least a
consequence of its twenty-month policy of cooperation and long-
standing leadership in the Prussian government. It was objectively
impossible to be a governmental party—formally in Prussia,
informally on the national level—while preparing for civil war. On
20 July 1932 the Social Democrats lost what little power they had
left, power they had managed to maintain for so long because they
had staked everything on one card in the autumn of 1930: the fight
against National Socialism on the basis of the constitution and in
alliance with the moderate forces within the bourgeoisie.
A sharp observer in the ranks of the Social Democratic left,
Arkadij Gurland, wrote in June 1932—that is, before the
Preussenschlag—that the toleration policy had been based on the
assumption ‘that the main threat to democracy lies in the danger of
civil war. Accordingly, its practical goal was less the preservation
of democracy than the preservation of legality, less the prevention
of unparliamentary rule than the prevention of civil war.’ The SPD
adhered to this postulate on 20 July 1932, too. In doing so, it stuck
874
to the principle it had followed on its path to power in November
1918, during the founding of the first German republic.
It was in keeping with this priority of the ‘prevention of civil
war’ that the Social Democratic interior minister of Prussia saw the
Papen government’s takeover of the Prussian police as the ‘lesser
evil’, compared to a National Socialist ascendancy in Prussia. Even
before 20 July, Severing had made comments to the arch-
conservative interior minister of the Reich that the latter could
interpret as encouragement to appoint a federal commissioner for
the Prussian police. And Gayl, too, was no stranger to thoughts of a
preventive strike against the extreme right. While it was not the
main cause, the ‘nationalization’ (Verreichlichung) of the Prussian
police as a precaution against a National Socialist grab for the
executive power of Germany’s largest state was certainly a
recognizable secondary reason for the Reich’s strike against
Prussia.
The dismissal of the Braun government brought an unusual
chapter in the history of Prussia to an end. After 1918 the
Hohenzollern state had turned into the most dependable buttress of
the republic among all the German Länder. Though old Prussia had
not disappeared, the political scene was controlled by the three
Weimar coalition parties until spring 1932. Immediately after the
Preussenschlag the great purification began. State secretaries and
department heads, provincial, district government, and police
presidents, as well as other officials belonging to the coalition
parties were sent into temporary retirement and replaced by
conservatives, often from the German National People’s Party. Only
one of the four Social Democratic Oberpräsidenten remained, Gustav
875
Noske in Hanover. In the opinion of the Reich government, the
former defence minister stood so far to the right of his party that
he could hold onto his post, which he had occupied since July
1920.43
The Social Democrats’ desire to make the Papen government pay
for the Preussenschlag in the Reichstag election on 31 July 1932 did
not come to fruition. The outcome at the polls was a triumph for
Hitler, at least at first glance. With electoral participation at 84.1%,
the highest since 1920, 37.4% of the vote went to the NSDAP. This
represented an increase of 19.1% over the previous election on 14
September 1930. The number of National Socialist seats in the
Reichstag grew from 107 to 230. The Communists’ gains were
much less significant; they climbed from 13.1% to 14.5%. The two
Catholic parties also grew, the Centre from 11.8% to 12.5%, the
BVP from 3% to 3.2%. All other parties were losers. The SPD fell
from 24.5% to 21.6%, the DNVP from 7% to 5.9%, the DVP from
4.5% to 1.2%, and the German State Party from 3.8% to 1%. The
remaining parties together totalled 2.5% of the vote.
The National Socialists had succeeded in making themselves the
heirs to the parties of the liberal centre and the moderate right, as
well as of the splinter parties. They also mobilized many first-time
and traditional non-voters. The north and the east of Germany
were much more strongly ‘brown’ than the south and the west. But
also in Hesse, Franconia, the Palatinate, and the north of
Württemberg, the NSDAP overtook all other parties. With 51% of
its vote falling to the party of Hitler, Schleswig-Holstein was the
National Socialist ‘frontrunner’ among the thirty-five electoral
districts in the country.
876
As in 1930, the Catholic milieu and, to a lesser extent, the
fragmented ‘Marxist’ constituency proved relatively immune to
National Socialist rhetoric. Among the bourgeois-Protestant groups,
only the conservatives managed to preserve a modicum of
independence from the NSDAP. The 5.9% of the vote that went to
the DNVP, whose power was still concentrated in east Elbia, was
the nucleus of the monarchist camp, formerly much larger. Political
liberalism had been nearly wiped away. The National Socialists had
become the great protest movement against the ‘system’, and
whoever did not have strong ideological or political reservations
against it was susceptible to its call. The fact that Hitler’s party
made very contradictory promises was little noticed by its voters.
What counted was the hope that a ‘national revolution’ would
bring about positive changes for Germany and the Germans.
There were, however, no signs of a parliamentary majority after
the election. Although the National Socialists were by far the
strongest party in the Reichstag, they had actually improved very
little on their performance in the second ballot of the presidential
race on 10 April and in the state parliamentary elections on 24
April. Even with the help of the DNVP and the smaller rightist
parties, they were far from a majority. A black–brown coalition was
theoretically possible, but the experiences in Hesse and Prussia
made it very doubtful that such an alliance would take shape.
The National Socialists were bitterly disappointed that political
ascendancy was not yet in sight even after their great electoral
success. They gave vent to their disappointment in a wave of
bloody attacks against political enemies at the beginning of August.
The SA struck primarily where they were especially strong, in the
877
east. On 9 August the Reich government found it necessary to
proclaim a new emergency measure against political terror
extending the death penalty to cases of politically motivated
homicide and setting up special tribunals in the most affected
districts.
The decree went into effect at midnight on 10 August. One and
a half hours later, a crime of unusual brutality, even for the
increasingly brutal times, was committed in the town of Potempa
in the district of Gleiwitz. Inebriated members of the SA attacked
an unemployed supporter of the Communist party, shot at him and
kicked him to death before the eyes of his mother. The police
succeeded in apprehending most of the suspects within two days.
Given the new legal situation, it was probable that the special court
in Beuthen would pronounce death sentences—unless the National
Socialists came to power in the interim.
For several days in the first part of August, Hitler seemed very
close to doing just that. On 6 August he met with the defence
minister for a long confidential interview in the vicinity of Berlin.
The National Socialist leader succeeded in convincing Schleicher
that he, Hitler, should take over the leadership of the national
government, his party assuming the offices of Prussian prime
minister; the departments of the interior, education, and
agriculture in Prussia and on the national level, each in personal
union; the ministry of justice, and a new ministry of aviation. With
his fundamental assent to Hitler’s demands, the ‘strong man’ of the
cabinet performed a dramatic about-face. Apparently, at the
beginning of August 1932, Schleicher considered the Reichswehr a
sufficient counterweight to a National Socialist monopoly of
878
political power.
Hindenburg, who was vacationing in Neudeck at this time, saw
things very differently. He brusquely rejected Schleicher’s proposal.
His reaction was no different after his return to Berlin, when, on 10
August, Papen suggested appointing Hitler chancellor at the head
of a majority government inclusive of the Centre Party. This was
the occasion of Hindenburg’s much quoted comment that it was a
bit much to expect him to make the ‘Bohemian private’ into the
chancellor of the German Reich.
The cabinet, too, was very divided over the question of putting
governmental power in Hitler’s hands, as became clear in the late
afternoon of 10 August 1932. In favour were—indirectly—Justice
Minister Gürtner and—much more openly—Finance Minister
Schwerin von Krosigk. The latter made the crude comment that
civil war could best be avoided by ‘turning the poacher into the
forest warden’. The staunchest resistance came from the interior
minister, Gayl, who was even prepared to wage a ‘battle to the
death’ with the NSDAP, referred to a ‘revolution from above’, and
openly advocated a solution at odds with the constitution: the
dissolution of the Reichstag, postponement of new elections beyond
the constitutionally mandated deadline of sixty days, and the
imposition of a new voting law.
On the next day, 11 August, the government’s traditional
celebration of the constitution took place, with the president in
attendance. For the first time in the history of the Weimar
Republic, the main speaker held an oration against the constitution
of 1919. Interior Minister von Gayl commenced with the
observation that the Weimar constitution did not unite the German
879
people, but divided them. He called for constitutional reform in an
authoritarian sense, the main aspects of which were an elevation of
the voting age, extra votes for breadwinners and mothers,
detachment of governmental power from the parliament, and the
creation of a profession-based, corporative first chamber as a
counterweight to the Reichstag.
During the next two days, Hitler was scheduled to negotiate first
with the chancellor, then with the president. In order to emphasize
his claim to power, the National Socialist leader had marshalled
strong forces of SA all around Berlin. The interview with the
chancellor, set for 12 August, he postponed at short notice until the
next day. In the late morning of 13 August Hitler, accompanied by
SA chief of staff Ernst Röhm, called on Schleicher, then, together
with Wilhelm Frick, the NSDAP floor leader, on Papen. From the
defence minister and chancellor Hitler learned that Hindenburg
had thus far proved unwilling to grant him the chancellorship.
Without the president’s express authorization, Papen offered Hitler
the post of vice chancellor in his administration and even gave him
his word that he would resign in Hitler’s favour after a period of
cooperation, during which Hindenburg could become better
acquainted with him. But Hitler rejected the offer and continued to
insist on the chancellorship.
The interview with the president of the Reich took place in the
afternoon of 13 August 1932. Papen and Meissner were also
present, as well as Röhm and Frick on the side of the National
Socialists. This meeting was Hitler’s greatest political defeat since
the failed Munich putsch on 8 and 9 November 1923. According to
Meissner’s protocol, Hindenburg answered Hitler’s demand for the
880
chancellorship with a ‘clear, definite no’.
881
government. On 22 August the special court at Beuthen handed
down its judgments in the Potempa case. On the basis of the
emergency measure of 9 August, five National Socialist defendants
were condemned to death, four for joint politically motivated
homicide, one for incitement to politically motivated homicide.
Hitler immediately assured the perpetrators via telegram that the
battle for their liberty was, ‘from this moment on, a matter of our
honour, and the battle against a government under which this was
possible, a matter of our duty’. Two days later, Hitler announced in
the Völkischer Beobachter that Herr von Papen had ‘written his
name into German history in the blood of national warriors’.
Goebbels tried to outdo his leader. In the Angriff, of which he was
the editor, he identified the Jews as those responsible for the
Beuthen sentences. ‘Never forget, comrades! Say it a hundred times
a day to yourselves, so that it pursues you even into your dreams—
the Jews are responsible! And they will not escape the judgement
they deserve.’
The cabinet knew full well that the execution of the sentences
could lead to open civil war. On the other hand, there was the
danger that a pardon for the perpetrators would be interpreted as a
capitulation to the National Socialists. But this seemed the lesser
risk. Hindenburg stated on 30 August that he personally supported
a pardon not for political, but for legal reasons. The deed was
committed only one and a half hours after the emergency measure
against political terror had gone into effect, and it could not be
assumed that the offenders had known about the increase in the
severity of the penalties. This was the rationalization (which could
be found even in liberal newspapers like the Frankfurter Zeitung)
882
used by the acting Prussian government under Papen on 2
September when it commuted the sentences of the Potempa
murderers to life in prison.44
Hindenburg gave his vote in the Potempa case on 30 August at
Neudeck, where Papen, Gayl, and Schleicher had joined him for a
meeting. The main subject of discussion was the internal political
situation after 13 August. Papen assumed that the parliament
would soon have to be dissolved again, since it had produced no
majority willing to cooperate with the president and since a
conceivable black–brown coalition could only lead to an ‘illusory’
or ‘negative’ majority. After the Reichstag was dissolved again,
there was the question of whether new elections were to be held
within the constitutionally mandated deadline of sixty days. It was
true that a delay was a formal violation of Article 25.
Gayl, who had been the first to propose postponing elections (in
the cabinet meeting on 10 August), seconded the chancellor. At the
end of the meeting, Hindenburg made the announcement the three
visitors had been waiting for:
The Herr President made a statement to the effect that he, in order
to avert detriment to the German people, could, in the state of
883
national emergency following the dissolution of the parliament,
answer to his conscience for interpreting the provisions of Article
25 to the effect that new elections be postponed to a later date,
given the extraordinary situation in the country.
884
aura of legitimacy to a constitutional revision from ‘above’,
provided that the president succeed in establishing the new
constitution by virtue of his authority. The arguments of Johannes
Heckel, a colleague of Schmitt, were far more sophisticated in this
respect. In an essay published in the Archiv des öffentlichen Rechts in
October 1932, Heckel explained that Germany had entered a state
of constitutional paralysis after the parliamentary elections. Since
two openly anti-constitutional parties, the NSDAP and KPD, held an
absolute majority of seats, the Reichstag was no longer capable of
exercising its function as an organ of constitutional government. It
was not to be expected that new elections would alter this
situation. In such a state of acute constitutional crisis, Heckel
argued, the president could invoke his duty ‘to carry out the
overarching political purpose of the constitution despite the
abnormal situation, and adapting to it’. To be sure, he was not
permitted to use the postponement of new elections to the purpose
that Gayl, for one, had in mind: the setting up of a new,
authoritarian constitution. In exercising provisional dictatorial
power, the president could only be a ‘dictator ad tuendam
constitutionem’, not a ‘dictator ad constituendam constitutionem’—that
is, a dictator for the preservation of the existing constitution, not
for the promulgation of a new one.45
The day of the emergency meeting at Neudeck, 30 August 1932,
was also the day of the constituent session of the newly elected
Reichstag. In the conclusion to her address, the president by
seniority, the Communist Clara Zetkin (born on 5 July 1857),
expressed her hope that she would live long enough to give the
opening address as Alterspräsidentin of the first council congress of
885
Soviet Germany. Then a strong majority elected the National
Socialist Herman Goering as Reichstag president. The Centre Party
also voted for him, pointing to the parliamentary tradition that the
presidency fell to the strongest fraction.
The second Reichstag session was held on 12 September. The
only item on the agenda was the chancellor’s inaugural speech.
Right at the beginning, however, the Communist delegate Ernst
Torgler made a motion to change the agenda, proposing that his
party’s motions to revoke two new emergency measures (one for
the stimulation of the economy of 4 September and a related
decree for the increase and preservation of job opportunities of 5
September) and the motions for a vote of no confidence in the
government be dealt with first. It would have taken only one
delegate to prevent this change of procedure, but—to the surprise
of all—no one objected, not even the German Nationalists. The
NSDAP, in order to consult with Hitler, requested a half-hour
pause, during which the Centre urged the National Socialists to
reject the KPD’s motions. But Hitler opted to vote for them. This
brought negotiations between the two parties—in which Hitler was
personally involved, having met with Brüning on 29 August—to an
end for the time being.
Papen was taken completely by surprise by this turn of events.
Anticipating neither the initiative of the KPD nor the absence of
resistance, he had appeared in the Reichstag without the
dissolution order Hindenburg had signed in Neudeck on 30 August
(the document was not dated). He did not get hold of the ‘red file’
until the pause, brandishing it upon his returned to the main hall.
Goering, the president of the parliament, ignored Papen’s two
886
requests to speak, as well as the file the chancellor finally placed
on his desk. Instead, he called for a vote on the two KPD motions,
the results of which he announced long after the cabinet had left
the hall. Of 560 ballot papers, one was invalid, 512 delegates voted
in favour, 42 against, and 5 abstained. The negative votes came
from the DNVP and DVP. The members of several smaller groups,
the German State Party, the Christian-Social People’s Service
(Christlich-Sozialer Volksdienst), the German Farmers’ Party
(Deutsche Bauernpartei), and the Economic Party, were absent from
the vote. All other parties voted for the Communist motions.
The vote was invalid; from the moment the chancellor lay the
order on the desk of the Reichstag president, the assembly was
dissolved. But the political effect could no longer be undone. More
than four-fifths of the delegates had voted no confidence in the
Papen government, and the chancellor had only his own neglect to
blame for the debacle.
When the cabinet met two days later to discuss what was to be
done, Papen no longer had the confidence to undertake the trial of
strength for which he had gained the support of Hindenburg on 30
August. Only Gayl and Schleicher argued for the indefinite
postponement of new elections, the latter pointing out that the
legal scholars Carl Schmitt, Erwin Jacobi, and Carl Bilfinger, all
defenders of the national government in the case of ‘Prussia vs. the
Reich’, had agreed that a ‘true state of emergency’ existed in this
case. Papen and the other ministers believed that the moment had
not yet come to depart from the constitution. On 17 September the
cabinet decided to propose to the president 6 November, the last
possible date, for new parliamentary elections. Hindenburg signed
887
off on the proposal three days later.
The chancellor did not abandon the idea of constitutional
reform. In his inaugural speech, which he gave by radio on the
evening of 12 September, Papen addressed the subject in a manner
similar to Gayl at the constitution ceremony. The system of formal
democracy, which in the judgement of history and in the view of
the German nation had come to ruin, had to be replaced by a new
order, a ‘truly non-partisan national leadership’ supported by the
power and authority of the popularly elected president of the
Reich. The voting age was to be raised and the organ of popular
representation organically linked with the self-governing bodies,
evidently in the form of a first chamber based on professional
corporations. The Prussian and national governments were to be
‘organically’ joined, bringing independent and antagonistic courses
to an end—here Papen retrospectively presented the Preussenschlag
as a breakthrough to the long-discussed ‘reform of the Reich’. The
German people had the chancellor’s assurance that they themselves
would be able to decide concerning the new constitution, which
the government would present to them after thorough scrutiny.
Papen concluded his address with the call: ‘With Hindenburg and
for Germany!’
The last part of this speech was an outline of the ‘new state’ the
cabinet had been aspiring to establish since the summer of 1932. In
a pamphlet with an introduction by the chancellor, and which
therefore acquired official status, the publicist Walther Schotte
described in greater detail how the ‘new state’ would differ from
the parliamentary democracy of Weimar. The ‘new state’ was an
authoritarian presidential state with professional-corporative
888
elements. The will of the people found expression primarily
through a one-time elective act, the plebiscitary legitimation of the
head of state. The president, not the parliament, embodied the
general will and was the centre of power. The Reichstag, de-
radicalized through a new voting law taking age, marital status,
and number of children into consideration, was to share the
legislative function with an upper house. In this first chamber,
appointed by the president, the professional corporations would
cooperate harmoniously. At least, this was how the architects of the
‘new state’ saw things. Had a majority of Germans voted for a
constitutional plan oriented on this scheme, they would have been
electing to divest themselves of the greater part of their political
power.
Papen, Gayl, and their publicist friends could depend on the fact
that by 1932, parliamentary democracy had very few committed
supporters left in Germany. In a much-read book from 1927, the
jungkonservativ writer Edgar Jung—who would become Papen’s
speech-writer in the spring of 1933—denounced the parliamentary
system of the west as the ‘rule of the inferior’. The idea of an
authoritarian, ostensibly non-partisan presidential state—indeed, a
state that would render political parties obsolete—was the common
denominator of the reform plans that rightist groups like the ‘Ring
movement’ around Heinrich von Gleichen-Russwurm, founder of
the Herrenklub, and the circle around the newspaper Die Tat and its
publisher, Hans Zehrer, had been developing for some time. In
other respects, however, the ‘conservative revolution’ was not a
monolithic bloc. The ‘Tat circle’, which was closely connected to
Schleicher, emphasized the role of the masses far more than did the
889
intellectuals of the Herrenklub around Papen. It was true that the
role played by the defence minister in the discussions about new
links between ‘right’ and ‘left’ often had more to do with the
changing projects of his advisers than with any concrete plans of
his own. Still, by September 1932 it was clear that Schleicher was
increasingly sceptical about the distance the chancellor—who had
been selected by Schleicher himself—was putting between himself
and the Volk.
At no other time during his term in office was the fundamentally
backward-looking nature of Papen’s politics so in evidence than on
12 October 1932. Prompted, it seems, by Edgar Jung, a Protestant
who admired Catholicism, the chancellor, speaking at a meeting of
the Bavarian Industrialists’ Association in Munich, invoked the
‘invisible current of power of the sacrum imperium, the
indestructible idea of the holy German Empire’. During the crisis of
the German state, the myth of the Reich gained in charisma what
the republic lost. But the imperial idea also served to justify the
German claim to be something different and more than a nation
state in the western sense, shaped by the ideas of 1789. ‘Only a
Europe led by Germany can be a Europe at peace,’ declared
Wilhelm Stapel in 1932, publisher of the jungkonservativ journal
Deutsches Volkstum. ‘The Reich is becoming a watchword, in both
domestic and foreign policy,’ observed the Catholic publicist
Waldemar Gurian, a critic of the new political romanticism, also in
1932. ‘For the Reich, and against Versailles and parliamentary
democracy… We can call the Reich the German image of
humanity, placed over against western humanitarianism and yet
differing from eastern apocalypticism by virtue of its intimate
890
connection with European history.’
The idea of the empire, the Reichsidee, experienced a supra-
confessional renaissance in the early 1930s. It was usually
accompanied by an assertion of the grossdeutsch idea and also,
frequently, by a trans-national view of the German Volk. Both
Protestant and Catholic imperial ideologues considered the
opposition between kleindeutsch and grossdeutsch obsolete anyway,
now that the Habsburg empire no longer existed, and they saw
themselves in agreement with current German historiography on
this point. One could, in order to give a ‘positive’ answer to the
west and the Weimar Republic, invoke the idea of a supra-national
German empire as a force for order in central Europe, or the
Prussia of Frederick the Great, or even both myths together. Most
authors of the ‘conservative revolution’, as well as well-known
German historians, did just that. The mystical grand narrative of
the sacrum imperium, on the other hand, belonged primarily to the
Catholic right of which Papen was a member. It was a credo that
gave many people—and not just his political enemies—cause to
doubt the chancellor’s grasp on reality.46
That reality caught up with Papen on 25 October at the latest,
when the federal court in Leipzig handed down its judgment in the
matter of the Preussenschlag. The presidential decree of 20 July
1932 was declared constitutional, to the extent that it appointed
the chancellor federal commissioner of Prussia, authorizing him to
temporarily divest Prussian ministers of their powers of office and
to exercise them himself. But this empowerment was, as the text
continued, ‘not to be extended such as to strip the Prussian state
ministry and its members of the power to represent Prussia in the
891
Reich Council or vis-À-vis the Prussian Parliament, the State
Council, or other countries’.
The Leipzig decision did not abolish the ‘dualism’ between
Prussian and the Reich. It declared both sides, the accuser and the
plaintiff, partially in the right. In consequence, the authority of the
Prussia state was divided between the custodial Braun
administration and the provisional government installed by the
Reich. The latter retained the actual executive authority, whereas
the former’s most important right was to represent Prussia in the
Reich Council. Though the Braun cabinet regained no real power, it
could count as a success the fact that no dereliction of duty could
be proven. The Reich government, while continuing to exercise the
administrative authority of Germany’s largest state, including its
police, had to accept the judgment that it had acted
unconstitutionally on 20 July in dismissing the Prussian
government. This verdict also pertained to the president, in whose
name the measure had been pronounced.
However one looked at it, the Leipzig decision was a defeat for
the Reich government—which was not the same thing as a victory
for the former Prussian cabinet. In Vorwärts appeared a
commentary from ‘a special source’ (possibly Hermann Heller, the
counsel for the prosecution) hitting the nail on the head. The
Leipzig decision was not a legal, but rather a political decision,
wrote the author of the article.
The court avoided the serious conflict with the Reich government
that would have resulted if it had recognized the claim of the
Prussian government to the full extent… Its judgment is the
892
opposite of a Solomonic decision; it has sliced the contested child
neatly in two and given each of the two mothers a half… How that
is intended to work in practice, and how it will work, the gods
know.
893
than in July and sinking from 21.6% to 20.4%. The winners were
the German Nationalists and the Communists. Hugenberg’s party
gained over 900,000 more votes, an increase from 5.9% to 8.9%.
The KPD, with some 600,000 more votes, climbed from 14.5% to
16.9% and from 89 to the magic number of 100 seats. The other
parties experienced only small changes. But the decrease in
electoral participation was conspicuous. From 84.1% in July, it fell
to 80.6%
The outcome of the vote was primarily a manifestation of
political frustration. For most Germans, the 6 November elections
represented the fifth trip to the polls in 1932, including the two
presidential ballots and the five provincial elections on 24 April.
The NSDAP, whom the politicization of the traditionally non-voting
population had earlier benefited the most, now suffered the worst
from the decrease in participation, for ‘non-political’ voters were
the soonest to feel that their vote had had little effect on practical
politics.
Also in evidence was a modicum of greater confidence in the
Papen cabinet, reflected in the comparatively good performances of
the DNVP and DVP. The government and the parties supporting it
benefited from the first signs of economic recovery, which could be
interpreted as the result of the active economic policy Papen had
initiated in September. In addition, the political and social
radicalism of the National Socialists had had a sobering effect on
many in Germany. The cooperation with the Communists during
the Berlin transport strike shocked not only the affluent circles of
the capital, but scared away many bourgeois voters throughout the
country. Nonetheless, the government had no cause for triumph.
894
Nearly nine-tenths of the population had voted for parties in
opposition to the ‘cabinet of barons.’
The National Socialists’ losses filled their political adversaries
with satisfaction, the Social Democrats most of all. As Otto Wels,
the party chair, put it in an executive meeting on 10 November, the
SPD had fought five battles under the watchword ‘Defeat Hitler!’ in
the course of 1932, ‘and after the fifth he was defeated’. However,
the election outcome also had a very disturbing side for the Social
Democrats, their own losses and the Communists’ gains. The gap
between the two workers’ parties had shrunk by half, from 7.1% in
July to 3.5%. If this trend continued into new elections at the
beginning of 1933, at the height of mass unemployment, a
dramatic crisis was almost certain to result. ‘We are at the final
spurt with the Communists,’ observed Karl Böchel, a Chemnitz
district chairman and ‘leftist’, at the same meeting.
895
won the electoral victory.
Ever larger masses of workers are going over to the camp of the
revolution. Great revolutionary struggles are imminent. The present
wave of economic struggles will lead to ever greater strike
movements in whole branches and districts of industry, to the
political mass strike and political general strike under the
leadership of the Communist Party, to the battle for the proletarian
dictatorship.
896
to power’.48
On 8 November, speaking before the foreign press club,
Chancellor von Papen made his first official announcement
concerning the election outcome. He spoke of a ‘welcome increase
in understanding for the work of the government’, then expressed
his hope that a true national concentration would be possible, now
that the election was over. ‘Questions of personnel, as I have
always emphasized, will play no role in this matter.’ In a cabinet
meeting on the next day, Gayl sharply criticized the chancellor’s
statement, calling it a ‘sign of weakness’. The interior minister
recommended negotiations with the parties in order to sound out
the potential for a toleration agreement. ‘If this aim cannot be
achieved, then the consequences are clear. The possibility of
another dissolution of the parliament arises, and with it a state of
constitutional emergency. Then, for a certain time, dictatorship will
be unavoidable.’
The debate then returned to the government’s emergency plan
of 30 August. But none of the ministers supported Gayl. Schleicher
managed to find agreement for his suggestion to postpone
constitutional reform and take up negotiations with the parties, so
as to put them in the wrong. He himself, he said, was prepared to
speak with Hitler, although he, the defence minister, was
‘absolutely convinced that the National Socialists will not
participate in the government’. Hindenburg, for his part, wanted to
maintain the presidential cabinet and not switch chancellors, as he
explained to Papen the following day.
The president’s position was much more lucid than that of the
897
chancellor. On the one hand, Papen was willing to follow
Hindenburg loyally. On the other hand, his own desire was to
reach an understanding with the National Socialists, and he did not
personally wish to exclude the possibility of giving the
chancellorship to Hitler. His comments to the foreign press were
meant in this way and no other, and if the information sent on 13
November by Wilhelm Keppler, the leader of a circle of pro-
National Socialist industrialists, to the Cologne banker Kurt von
Schröder was true, the chancellor had twice expressed his support
for a Hitler government to Ewald Hecker, chairman of the board of
the Ilseder Steelworks, in the days before the press conference. For
his part, Papen learned from Hecker that a large circle of leading
personalities in industry, banking, and agriculture was preparing to
send a message to Hindenburg demanding that Hitler be made
chancellor.
This letter reached the president on 19 November. The key
sentence was the following:
898
businessmen, bankers, and landowners. The letter was also signed
by Count Eberhard von Kalckreuth, acting president of the National
Land League, and Fritz Thyssen, who had long been a National
Socialist supporter. Along with August Rosterg, an owner of potash
mines, Thyssen was the only major industrialist actively engaged
on Hitler’s behalf in this way. Albert Vögler, general director of the
United Steel Works, who did not himself sign the letter, told
Schröder on 21 November that two other heavy industrialists, Paul
Reusch, chairman of the board of Good Hope Steel and Iron Works,
and Fritz Springorum, general director of Hoesch, ‘basically share
the view expressed in the letter and consider it a real solution to
the present crisis’. They did not wish to sign, however, since they
both feared that such a political statement would make the
antagonisms within the industry of the Ruhr district all too clear.
Clearly, then, the letter to Hindenburg did not represent the
position of big industry as a whole. But it was also evident that,
after 6 November, the Papen government could no longer count on
the same, virtually unified support of the ‘business community’ as
in September and October. The election outcome was disappointing
for industry, which had given the greater part of its ‘political’ funds
to the two parties friendly to the government, the DVP and DNVP.
Particularly alarming were the KPD’s gains, for which Hitler held
the Papen cabinet responsible, and not entirely without cause.
Finance Minister von Krosigk expressed a broadly held opinion
when, in a cabinet meeting on 9 November, he justified his support
of NSDAP participation in government by saying that otherwise a
large number of National Socialists, ‘including the nationalist
youth’, would migrate into the Communist camp. Similar fears led
899
industrialists to change their opinion in Hitler’s favour in the wake
of the Reichstag elections. An observer at a meeting of the heavy
industrialist Langnam association in Düsseldorf at the end of
November gained the—certainly exaggerated—impression ‘that
nearly all of industry wants Hitler to be appointed, regardless of
the circumstances’.49
On 17 November the cabinet took stock of the chancellor’s
efforts to negotiate with the parties. The results were negative. Two
parties, the SPD and NSDAP, had refused to agree to an interview.
The Centre and BVP, on the other hand, demanded the resignation
of the cabinet and the inclusion of the National Socialists in the
government (which, given the distribution of seats in the new
Reichstag, was no longer possible in the shape of a ‘purely’ black–
brown coalition). The head of the BVP, Fritz Schäffer, even
approved of giving the chancellorship to Hitler. Papen himself
came to the conclusion that a ‘national concentration’ was not
possible under his own chancellorship, and recommended
proffering the resignation of the entire cabinet. The ministers
agreed. Hindenburg accepted the government’s resignation later
that same day, but asked them to remain in office for the time
being.
The next day, Hindenburg himself took up negotiations with
selected party leaders. The most important exchanges were those
with Hitler on 19 and 21 November. They led to no workable
result. Since Hugenberg rejected a Hitler chancellorship out of
hand, an NSDAP majority was not possible, and Hindenburg, too,
was unwilling to entrust Hitler with the leadership of a presidential
cabinet. He seemed unimpressed with the latter’s warning that
900
there would be ‘18 million Marxists, including perhaps 14 to 15
million Communists’ in Germany if his movement fell apart, and
equally unavailing was Hitler’s prophecy that a continuation of the
authoritarian government threatened to unleash a new revolution
and Bolshevist chaos in the coming months.
On 24 November Hindenburg, through the agency of State
Secretary Meissner, sent Hitler a letter—communicated to the press
at the same time—containing essentially the same message as his
decision on 13 August: the president believed
After his talks with Hitler, Hindenburg was convinced that there
was no longer any alternative to proclaiming a state of national
emergency. The chancellor and cabinet, however, were not nearly
so ready for battle. When, in a private conversation with Papen on
26 November, Krosigk spoke of the bloody battles that would occur
between the army and the ‘national’ youth if the chancellor
declared a state of emergency, the latter was impressed. In an
901
interview with Hindenburg on the same day (with the defence
minister also attending), the chancellor requested that the
formation of the new government not be entrusted to him. When
Hindenburg insisted Papen remain chancellor, the defence minister,
according to his own account, advised that ‘the atmosphere be
tested beforehand’, ‘given the general opposition to Papen, which is
making itself felt even in the industry of the Ruhr’. The president
had no objection, set no date for Schleicher, and gave him a
completely free hand in his choice of interlocutors.
The most important talks the defence minister held in the
following days were those with the heads of the ADGB and the
Social Democratic party fraction on 28 November. Schleicher
gained union leader Theodor Leipart’s support for a ‘ceasefire into
the next year’ by promising to abolish the measure of 5 September
allowing employers to undercut contract wages. The interview with
Rudolf Breitscheid went very differently. The critical point was
reached when the minister asked what the SPD would do if new
elections were postponed until the spring of 1933. Schleicher’s
question, ‘whether the Social Democrats would then immediately
go onto the barricades’, elicited the following answer from
Breitscheid:
902
Perceptive observers were not surprised by the difference
between the ADGB and the SPD. The Social Democrats were
worried that the mere appearance of a new ‘cooperation’ policy
would both lay the party open to unparalleled attacks from the
Communists and split it from within. The Free Trade Unions, on
the other hand, had for some time been underscoring their
independence from the SPD and emphasizing their nationalist
attitude. Leipart’s speech on the ‘cultural tasks of the unions’ at the
national school of his organization in Bernau on 14 October was
programatic in this regard. According to one of its key sentences,
‘No social stratum can escape the national development.’ The
unions had organized the workers in order to ‘awaken the feeling
of community within them and to cultivate the community spirit’.
They performed ‘service for the people’ and waged their ‘social
struggle in the interest of the nation’. Though socialists, they were
not lacking in ‘religious feeling, and what is more, they knew the
soldierly spirit of falling into line and of sacrifice for the good of
the whole’.
Ernst Jünger’s book Der Arbeiter. Herrschaft und Gestalt* had
come out a short time before this speech. Leipart did not quote it
directly, but it was obvious that his speech-writer—Lothar
Erdmann, editor of the union monthly Die Arbeit—borrowed central
ideas from the right’s most notable writer. The worker as soldier of
labour, who, in contrast to the liberal civilian, served the nation as
a whole, had little to do with the class-conscious proletarian, but
much in common with Jünger’s Gestalt.
It was no accident that, of all newspapers, the Tägliche
Rundschau printed Leipart’s speech. Taken over (with financial
903
assistance from the defence ministry) by the ‘Tat circle’ around
Hans Zehrer in 1932, the Tägliche Rundschau was thereafter
regarded—though not always justly—as Schleicher’s mouthpiece.
Even more spectacular was the applause the Social Democrat
Leipart received from Gregor Strasser, a National Socialist. On 20
October the leader of the NSDAP national party organization
declared in the Sportpalast in Berlin that Leipart’s speech included
sentences ‘that, if sincerely intended, open up wide prospects for
the future’. After the report in the Tägliche Rundschau Strasser, ‘in
unqualified recognition of the sine qua non of their professional
existence’, exhorted the unions ‘to draw the necessary conclusions
from the declaration of their director and openly demonstrate their
political neutrality by separating themselves from the party of
“Heilmann and Hilferding”, from the SPD, directed as it is by a
caste of internationally inclined intellectuals’.
The Bernau speech and the resonance it found seemed to prove
that the ‘crossfront’ from Leipart to Strasser, propagated by the ‘Tat
circle’ and promoted by Schleicher, was more than just an
expression of political wishful thinking. Nonetheless, events at the
end of November 1932 revealed that the obstacles to such an axis
were greater than Schleicher and Zehrer had expected. The
inflexible attitude of the SPD reduced the value of Leipart’s
willingness to work with Schleicher. And on the right side of the
political spectrum, Strasser failed to win Hitler over to an
arrangement with the defence minister. The NSDAP leader rejected
the offer of vice-chancellorship in the new cabinet on 30
November. To be sure, even if he had accepted, the left flank would
not have held together. National Socialists in ministerial positions
904
would have been confronted with resolute Social Democratic
opposition—and also, all differences between the ADGB and SPD
notwithstanding, with the hostility of the Free Trade Unions. The
polarization Schleicher sought to contain would have been even
stronger than before.
The alternative to an arrangement with the NSDAP as a whole—
even if only a ‘ceasefire’—was, for Schleicher, a pact with the
forces he imagined to be backing Strasser. If Strasser acted on his
own and brought a large number of National Socialists into the
government camp, it would open up entirely new prospects.
Indeed, it would have a veritably revolutionary effect on the
internal political situation of the country. For the time being,
however, this too was pure speculation. It was true that the NSDAP
national leader knew the party’s dismal financial situation better
than anyone else, and if any National Socialist apart from Hitler
had a broad following among the ‘old guard’, it was he. But
Strasser had always avoided any trial of strength with Hitler, and
this made a split in the NSDAP improbable.
In sum, Schleicher’s investigations did not lead to any kind of
breakthrough. There was no sign of a parliamentary majority for a
cabinet under his leadership. Nonetheless, he knew that large
portions of the working class considered him a lesser evil than
Papen. ‘Papen means war! The President does not have the power
to declare war on his own people!’ Thus declared Vorwärts on 29
November, as rumours about a Papen ‘battle cabinet’ were
circulating in Berlin and causing ‘a tremendous stir among the
workers’, according to the Social Democratic party organ.
Schleicher was not facing a declaration of war by the Social
905
Democrats, but only their opposition; Vorwärts hastened to assure
readers that opposition was a ‘normal function of political life’. The
relationship between the defence minister and the Reichsbanner
could almost be called one of mutual trust. He had even better
relations with the Christian-national and liberal unions than with
the Free Trade Unions. Among the Centre and other moderate
parties, Schleicher had never inspired the same aversion as Papen,
and from the leading industrial associations he had no opposition
to fear.
In a word, Schleicher enjoyed considerably greater backing in
society and political life than Papen. If Germany was facing a long
period of non-parliamentary rule under a state of emergency, this
fact could be crucially important. Schleicher’s attitude towards the
necessity of declaring a state of emergency did not, at the end of
November 1932, differ from that of the incumbent chancellor. But
he did see the risks of military dictatorship—however veiled—more
realistically than Papen, and this was the reason he wanted to do
everything possible to preclude a civil war.51
For his part, Papen was not out to put himself at the head of a
dictatorship. Only when there was no other choice was he willing
to do what Hindenburg asked of him. The latter, however, after
hearing on 1 December what Schleicher had to report concerning
the talks of the previous few days (Oskar von Hindenburg, Papen,
and Meissner were also present), was in no way persuaded that a
Schleicher chancellorship would improve the situation, and insisted
Papen form the new government. Papen gave in, but requested that
the president ‘place all presidential rights at his disposal for the
conflict with the Reichstag, certain to come’. After detailed legal-
906
constitutional elucidations from Meissner, Hindenburg agreed ‘to
take, in the event of conflict with the Reichstag, all necessary
measures to protect Germany from any harm that might arise from
a violation of the duties of the Reichstag’.
Weary of the back-and-forth in the cabinet, the president was
determined to cut the Gordian knot. It was in keeping with his
military way of thinking to not postpone a battle that, sooner or
later, was inevitable, but to fight it out quickly. Schleicher, the
‘desk general’, saw things more clearly. The kind of dictatorship
Hindenburg had chosen was the most dangerous, since there was
no popular support for it. To command the army to march against
the overwhelming majority of the people would undermine its
morale and risk its very existence. Finding himself unable to
countenance such a course, the defence minister rebelled against
the president.
Schleicher’s opportunity lay in the fact that he knew the
majority of the cabinet to be on his side—as well as in the outcome
of a military ‘war game’, known as Planspiel Ott, which he had
conducted at the end of November. In an improvised ministerial
meeting on the morning of 2 December, Lieutenant Colonel Eugen
Ott, at Schleicher’s behest, presented the lessons of the exercise.
The Reichswehr, it was concluded, could not win a two-front war
against Communists and National Socialists, and certainly not if it
had to ward off a Polish attack (part of the war-game scenario) on
the eastern border of Germany at the same time. The cabinet was
deeply impressed. When Papen reported on the meeting to the
president, the latter gave up his resistance to a Schleicher
chancellorship. ‘I am grown too old to assume responsibility for a
907
civil war at the end of my life.’ Such, according to Papen’s records,
were the words Hindenburg used to justify the reversal of a
position he had held the day before.
Schleicher took a great risk with Planspiel Ott. If, as chancellor,
he opted to declare a state of national emergency, the president
could use the results of the exercise against him. Schleicher might
hope that he, unlike Papen, would be able to keep the unions from
calling a general strike, thus eliminating one of the scenario’s
central assumptions. But he also faced the prospect that the belief
would become firmly ensconced that the Reichswehr, police, and
Technical Emergency Corps would not be able to defend the
country in the event that the elections were postponed.
Whether or not the defence minister was thinking of these risks
on 2 December, his main focus was the immediate effect of the
demonstration, and in this he was successful. On 3 December 1932
Hindenburg appointed Schleicher chancellor. His predecessor,
however, still enjoying the president’s special confidence, remained
in his official residence in the Wilhelmstrasse, to which Schleicher
assented. Thus Papen preserved what was perhaps even more
important than a government office, the privilege of immediate
access to the president of the Reich.52
908
general promised to not to engage in experiments with the
constitution and committed himself to easing social tensions. It was
true that as defence minister (a post he kept as chancellor),
Schleicher had been something of an enigma to observers from all
parts of the political spectrum. His attitude toward the National
Socialists was nothing less than contradictory and volatile. And yet,
in the political centre and on the moderate left, he was credited
with far more tactical skill than Papen, the ‘gentleman rider’, who
notoriously underestimated currents of political resistance—one of
the reasons for his failure. From industry-friendly newspapers like
the Deutsche Allgemeine Zeitung and the Rheinisch-Westfälische
Zeitung, the new chancellor heard primarily one message: that he
would have no choice but to seek out the active support of the
National Socialists.
Schleicher cleared his first hurdle without much effort: no
petitions for a vote of no confidence in the government were
brought forward during the short parliamentary session that began
on 6 December. With his assent, the Reichstag lifted the part of the
emergency decree of 4 September empowering the government to
institute the wage measure from 5 September. The cabinet also
agreed to the passage of an amnesty law. Furthermore, in response
to an NSDAP petition, the parliament amended Article 51 of the
constitution, which provided that the chancellor would represent
the president if the latter was unable to exercise the functions of
his office or in the event of an early termination of his presidency.
Hindenburg had celebrated his eighty-fifth birthday on 2 October
1932. If he died or took seriously ill during Schleicher’s
chancellorship, the powers of the president, chancellor, and
909
defence minister of the Reich would be unified in the hands of the
general. In order to preclude this, the National Socialists proposed
that the Reichstag president assume the representation of the Reich
presidency. They found the agreement of most of the bourgeois
parties and the Social Democrats, for whom the prospect of a
further increase in Schleicher’s power was equally threatening. The
proposal passed with the necessary two-thirds majority on 9
December. This was the last day of the session, the Reichstag
adjourning for an indefinite period of time.
That same day, the Tägliche Rundschau published a sensational
announcement: Gregor Strasser had resigned from all party offices.
The leader of the NSDAP national organization, whom Schleicher
had offered the vice chancellorship five days previously, was
reacting to Hitler’s refusal to tolerate the Schleicher government.
The newspaper, which was considered the chancellor’s mouthpiece,
interpreted Strasser’s decision as a declaration of war against
Hitler. But Strasser had no such intention, despite the fact that he
was, up to that point, the second most powerful man in the party.
On 9 December he departed for Munich, where his family resided,
then spent two weeks vacationing in the southern Tyrol. Hitler very
quickly succeeded in securing the allegiance of the party’s district
leaders, inspectors, and parliamentary delegates. The differences of
opinion between the Führer and the former Reichsorganisationsleiter
were swept under the rug as well as possible. ‘Palace revolution
failed,’ Goebbels noted in his journal. ‘Strasser is isolated. Dead
man!’
Schleicher saw things differently. He continued to place his bets
on Strasser. In a meeting of group and district commanders, which
910
took place from 13 to 15 December, the chancellor announced that
it was still his policy to work towards ‘a cooperation of the Nazis
under Strasser under the messianic blessing of Hitler’. By January,
he said, the question of a firm parliamentary majority would be
answered. As soon as the Reichstag was convened, the National
Socialists would be asked if they wished to cooperate. If they said
no, it would mean a fierce battle, including the dissolution of the
Reichstag and the Prussian parliament. In order to win such a
battle, the right would have to be on the side of the government.
Therefore, nobody was to be surprised if repeated attempts were
made to bring the National Socialists on board and confront them
with responsibility. A destruction of the NSDAP was not in the
interest of the state.
Thus, even in the middle of December, Schleicher still believed
he could reach an understanding with both Strasser and Hitler. He
would fight only after one last attempt to involve the National
Socialists in the government had been made and rejected. In both
cases, it was necessary to avoid as much as possible any
confrontation with the unionized working class. In this regard, the
chancellor had cause for optimism. On 8 December Heinrich
Imbusch, head of the Association of Christian Trade Unions, had
spoken extremely positively to Hindenburg about the new
chancellor and his cabinet. Similar comments from Leipart had
appeared in the Parisian Excelsior three days before. While this was
a far cry from the ‘crossfront’, it was clear that by mid-December
the government was no longer politically isolated.
When the chancellor announced his governing platform in a
radio address on 15 December, he did so with great self-confidence.
911
His views concerning a military dictatorship were generally well
known, he said, but he would state them again: ‘It is uncomfortable
to sit on the point of the bayonet. In the long run, that is, no
government can rule without broad popular backing.’ He greeted
the Reichstag with ‘a strong dose of healthy mistrust’, but it was
imperative that his government be accorded the opportunity to
carry out its programme, and this consisted in a single point: ‘Job
creation!’ He, Schleicher, was neither a devotee of capitalism nor of
socialism and had nothing against being seen as a ‘social general’.
The chancellor stressed the close relationship between job
creation, farm settlement, and border security in the east, and he
endorsed the idea of general conscription for a militia. In his
conclusion, Schleicher distanced himself clearly from Papen, whom
he had called his ‘friend’ and a ‘fearless and irreproachable knight’
at the beginning of his address. To those who thought that an
authoritarian government could get along without popular support,
Schleicher replied
that will and courage do not alone suffice for those who would
govern; it is also necessary to understand the sentiments of the
people and recognize the psychological moment. For that reason,
my government will take the best Moltke-proverb for the guiding
principle of its endeavours: ‘Look before you leap.’53
The ‘fearless and irreproachable knight’ did not agree that the
general was the better chancellor. He wanted to return to the
centre of power, and to that end he got together with another of
Schleicher’s adversaries. On 4 January 1933 Papen and Hitler met
912
at the house of the banker Kurt von Schröder in Cologne. What was
intended as a secret interview soon made headlines in both the
German and the international press. The purpose of the talks was to
find common ground between the National Socialist leader and the
Reich president and to bring about the end of the Schleicher
government. Before Papen could mediate between Hitler and
Hindenburg, however, he first had to clear up his personal
relationship with Hitler, which had been very tense since 13
August. After this was accomplished, the two men agreed to a sort
of ‘duumvirate’, whereby the question of the real leader of the
government was to remain unanswered for the time being.
It is certain that Hitler reiterated his demand for the
chancellorship at this meeting. Papen, according to what we know
about his position in August and November 1932, probably did not
insist on the leadership role in a future ‘cabinet of national
concentration’. But he would not have failed to mention
Hindenburg’s continuing reservations against a Hitler
chancellorship. In the further course of the interview, then, Hitler
apparently no longer absolutely rejected the possibility of a
temporary alternative to his chancellorship. On 10 January, after a
conversation with his Führer the day before, Goebbels wrote in his
journal:
913
The assumption that Hitler did not categorically reject a
nominal Papen chancellorship as a temporary solution is also
supported by comments the former head of government made
privately in the days following the meeting. When, on 7 January,
Papen met with the leading industrialists Krupp, Reusch,
Springorum, and Vögler in Dortmund in order to report on the
talks, he communicated the impression that Hitler would probably
be content to play the role of ‘junior partner’ in a cabinet
controlled by conservative forces. This kind of arrangement would
have suited the heavy industrial right wing perfectly, and by
pushing for it, Papen could be sure of the support of some of the
most important industrialists. Krupp, however, was not one of
them, nor was the National Association of German Industry he
directed. Despite its criticism of Schleicher, this organization saw
no reason to trade the current government for a cabinet that, it was
feared, would stir up even more political unrest among the people.
On 9 January Papen held talks first with Schleicher, who put on
a brave front, and then with Hindenburg. The president gained
from Papen the impression that Hitler was no longer insisting on
the entire authority of the government and was prepared to
participate in a coalition. He therefore asked the former chancellor
to keep up the dialogue with Hitler on this basis, and under the
strictest confidence. Hindenburg now set his sights on the
reconstitution of a Papen cabinet, a goal he justified to Meissner by
telling him that Hitler would not support or tolerate the current
government under Schleicher.54
Two days later the National Land League—which had actively
participated in Brüning’s ouster eight months before—mobilized
914
against the Schleicher administration. After a meeting with the
chancellor, Agricultural Minister von Braun, and Economics
Minister Warmbold, at which the president presided, the cabinet
learned of a decision the League had communicated to the press a
few hours before. It was tantamount to a declaration of war. ‘The
current administration has suffered the immiseration of German
agriculture, especially the peasant economy of breed improvement,
to take on dimensions that would not have seemed possible even
under a Marxist government,’ the League’s statement declared. ‘The
pillaging of agriculture for the sake of the almighty pocketbook
interests of the internationally-minded export industry and its
servants continues apace… For this reason the actions of the
government hitherto also do not satisfy the instructions the
president has also repeatedly issued.’
On 13 January the leader of the NSDAP Agrarian-Political
Apparatus, Richard Walther Darré, launched another attack. In an
open letter to Schleicher, he demanded from the government a
‘decisive change of course towards the domestic market’, though he
recognized that such a move was not to be expected from the
current administration. As Count Kalckreuth, president of the
National Land League, had done at his meeting with the Reich
president and chancellor on 11 January, Darré invoked the
‘disturbing spread of Bolshevism among the German people’. His
letter concluded with an allusion to the export-friendly policy of
Bismarck’s successor: ‘German agriculture’s tale of woe began with
“General” von Caprivi. Let us hope to God that “General” von
Schleicher will be the last representative of this unhappy and anti-
agricultural era.’
915
Schleicher could afford to ignore Darré’s letter, but not the
broadside from the National Land League. In the evening of 11
January the chancellor had an official announcement made
accusing the organization of demagogic and factually incorrect
attacks against the government. This reproach was followed by
sanctions: the cabinet broke off all relations with the League. But
the president did not join the boycott. On 17 January he addressed
a letter to the leadership of Germany’s largest agricultural interest
group expressing his hope that the measure he had just signed for
more effective stays of execution would help ease tensions in the
agricultural community.
By 11 January, when he settled accounts with the National Land
League, Schleicher had also been wanting for over a week to clarify
things between his government and the Reichstag. On 4 January he
let State Secretary Planck of the parliamentary advisory committee
know that the cabinet was ready at any time to appear before the
delegates and elucidate its platform. Thereafter he expected the
situation to be straightened out, and was not prepared to accept
that petitions for votes of no confidence be deferred. But the
advisory committee decided (with the NSDAP abstaining) not to
convene the Reichstag until 24 January, instead of 10 January, as
the SPD and KPD wanted. Rudolf Breitscheid observed that the
behaviour of the National Socialists was ‘practically tantamount to
a toleration of the Schleicher government’.
The National Socialists, unlike Schleicher, were playing for time.
Their first aim was to patch things up in the wake of the 6
November parliamentary elections and after the local elections in
Thuringia on 4 December, which had also been very disappointing
916
for the NSDAP. The parliamentary elections in Lippe-Detmold, the
second smallest German state, on 15 January offered an
opportunity. In the first part of the month, the party flooded the
northern German state with a wave of demonstrations and rallies.
Hitler himself spoke at sixteen large events. The effort paid off. The
NSDAP gained some 6,000 votes more than on 6 November,
climbing from 34.7% to 39.6% of the vote. Party propaganda
interpreted this outcome as proof the National Socialists were once
again on the rise. In reality, the success was comparatively modest.
In comparison with the Reichstag elections on 31 July 1932, the
NSDAP had lost 3,500 votes, and on the national level it could
never have organized an electoral campaign as intensive as in a
small state with 160,000 inhabitants.
Nonetheless, the psychological effect was what counted at the
moment. The party of Hitler seemed to be on the advance again,
and this strengthened Hitler’s hand vis-À-vis the bourgeois right,
especially the German Nationalists, from whom he had taken the
most votes. From now on, it was once again out of the question
that he would give up the post of chancellor in a ‘national’
government. Against Strasser, too, he could now take the decisive
step. On 16 January, at a meeting of district leaders in Weimar,
Hitler went on the offensive against the former head of the national
party organization. The result was unambiguous: Strasser found no
more defenders, and Hitler’s position within the party was stronger
than ever.55
On the day after the Lippe elections, the cabinet met for the first
time in the new year in order to discuss the political situation.
Schleicher showed himself determined to force a rapid decision
917
about the future of his cabinet and, with it, the future of German
politics. Though he still talked about the possibility of getting the
National Socialists to cooperate in some way, he concentrated
entirely on the consequences of their refusal. If the Reichstag set
the petitions for a vote of no confidence at the top of its agenda for
the coming session, Schleicher would send them the written order
of dissolution. There were, he said, strong reasons against holding
new elections within the constitutionally mandated period of sixty
days. The business community rejected them, and this attitude was
widespread among workers, too. ‘With things as they are, he
believes the idea of postponing the new elections until the autumn
very worthy of consideration.’
Unlike on the last day of Franz von Papen’s chancellorship,
there was virtually no objection to the promulgation of a state of
emergency on 16 January 1933. Interior Minister Bracht, who had
succeeded Gayl on 3 December, emphasized that ‘one thing, at
least, has been achieved: a united front against the government no
longer exists.’ He proposed to set the new Reichstag elections for
22 October or 12 November 1933. Finance Minister Count
Schwerin von Krosigk, too, who in November and at the beginning
of December 1932 had been one of the staunchest opponents of
violating the constitution, now supported the delay of elections
without qualification.
The possible legal and political consequences of the delay were
never discussed. Nor was an alternative plan of action that had
originated in the Wehrmacht department of the defence ministry
and been inserted into the protocol of the cabinet session: the ‘non-
recognition of a vote of no confidence and the confirmation of the
918
government by the president’. According to its author, ignoring a
vote of no confidence from a ‘negative’ majority represented,
‘relatively speaking, a lesser conflict with the constitution’ than
postponing new elections or forcing the Reichstag to adjourn.
Article 54, which made the chancellor and ministers subject to the
confidence of the parliament, did not prevent a government that
had been voted out from remaining in office in a custodial
capacity. It also did not establish a time limit for the life of an
administration after its parliamentary ‘death’.
In his 1928 book Verfassungslehre Carl Schmitt, one of Germany’s
most prominent constitutional scholars, had called a vote of no
confidence from a parliamentary majority itself incapable of
forming a government an ‘act of pure obstruction’. In such a case
there was no obligation to resign, ‘at least not when the dissolution
of the Reichstag is ordered at the same time’. In the December
1932 edition of Die Gesellschaft, the Social Democratic jurist Ernst
Fraenkel came to the conclusion—as Carl Schmitt and Johannes
Heckel, whom he cited, had already done—that the Reichstag, as
the central organ of the Weimar constitution, would be unable to
fulfil its obligations as long as Communists and National Socialists
had a majority. Fraenkel suggested a referendum to amend the
constitution as a means of preventing the ‘delinquent parliament’
from bringing the machinery of state to a halt and giving the
enemies of the constitution the longed-for pretext for a coup d’état.
According to his proposal, a parliamentary vote of no confidence in
the chancellor or a minister would legally compel resignation only
‘if the representative body combines the vote of no confidence with
a positive suggestion to the President naming the person who is to
919
be appointed minister in place of the fallen governmental official’.
Around the end of 1932 and beginning of 1933, political
pragmatists were advising the chancellor to the effect that a
constructive vote of no confidence be introduced in a de facto
sense, without explicit emendation of the constitution (on 1
December 1932 Franz Sperr, the Bavarian envoy to the Reich, and
on 19 and 26 January 1933 Wilhelm Simpfendörfer, Reichstag
delegate and head of the Christian Social People’s Service, a party
originating in the Pietist movement in Württemberg). But
Schleicher did not respond to any of these suggestions, which
sought to resolve the crisis below the threshold of an openly
declared state of emergency. He believed that delaying new
elections would cause a smaller loss of governmental authority
than a vote of no confidence. He could hold out to the president
the fact that his government was less isolated than that of his
predecessor and that the conditions of Planspiel Ott did not
therefore obtain. But it was very unclear whether Hindenburg
would accord Schleicher what he had promised Papen—the
dissolution of the Reichstag and the non-constitutional delay of
new elections. Since both those things were uncertain, the 16
January plan for a national state of emergency rested on very
shaky foundations.56
A few short days after the cabinet meeting, the press was full of
speculations about the imminent proclamation of a state of
emergency. At a meeting of SPD party officials in Berlin-
Friedrichshain, Rudolf Breitscheid revealed what Schleicher had
told him on 28 November concerning a possible delay of new
elections. The Social Democratic floor leader also quoted the
920
response he had given to the defence minister at the time: ‘Such a
provocation will doubtless cause the greatest convulsions.’
Another revelation on the same day drew even greater attention.
Joseph Ersing, Centre delegate and secretary of the Christian Trade
Unions, reported to the parliamentary budget committee on what
became known as the ‘Osthilfe scandal’, the misuse of public funds
for the rehabilitation of deeply indebted manorial estates,
especially in East Prussia. If the groups behind the National Land
League, which had continually received immense sums from the
entire German people, adopted the kind of language they had
recently used to the Reichstag, then, Ersing said, the Reichstag
would have to look into the matter. And if the federal funds were
not used to cover debt, but for the purchase of luxury automobiles
and racehorses and trips to the Riviera, the government would
have to demand the repayment of the funds. The circles of the
great landowners, Ersing said, were seeking to prevent further
parliamentary deliberation over the issues of eastern assistance.
This was the reason such intense efforts were being made behind
the scenes to dissolve the Reichstag.
One of the reasons Ersing’s disclosures caused such a stir was
because, shortly before, the name of a personal friend of the
president had appeared in press reports on the issue of eastern
assistance. Elard von Oldenburg-Januschau had allegedly done
extremely well in the distribution of public funds. The public was
also informed in detail about how Hindenburg came into the
possession of Neudeck. The ownership of this estate, which he had
received from the German business community in 1927 on the
occasion of his eightieth birthday, had been registered in the name
921
of his son Oskar (a son ‘not provided for in the constitution’, as
people sneered at the time) in order to spare the latter the payment
of the inheritance tax. Although this manipulation was not actually
illegal, it damaged the reputation of the head of state.
One day after Ersing’s sensational speech, the advisory
committee decided to postpone the convocation of the plenum
from 24 to 31 January. The cause of the delay was the National
Socialists, who had every reason to avoid a plenary session at the
moment. There were to be no disturbances in the political
negotiations Hitler had resumed shortly after the Lippe elections.
On 17 January he had an interview with Hugenberg. It failed to
produce concrete results, however, since the DNVP leader refused
to hand over to the National Socialists control of the Prussian
interior ministry and, with it, the police. The next day, Hitler met
Papen in the Dahlem villa of Joachim von Ribbentrop, a politically
active champagne salesman who had recently joined the NSDAP.
Pointing to his electoral successes, the National Socialist leader
pressed his demand for the chancellorship much more firmly than
on 4 January in Cologne. Nonetheless, whatever the former
chancellor thought of the idea, Hindenburg still rejected a Hitler
chancellorship.
In the meantime, the political pressure on Schleicher continued
to grow. On 21 January the German Nationalist parliamentary
fraction declared open opposition to the cabinet. In a statement
immediately communicated to the chancellor, but not published
until 24 January, the DNVP stated that the policy of deferment and
hesitation cast doubts upon all attempts to improve the situation.
The main target of criticism was the government’s economic policy,
922
which was ‘straying ever further into socialist-internationalist
ideas’. ‘It is a particular danger when antagonisms between big and
small are permitted to arise, especially in agriculture, and with
them the threat of Bolshevism in the countryside.’ The claim that
Schleicher was pursuing such a policy was as demagogic and
nonsensical as the accusation the German Nationalists had levelled
against Brüning in the Reichstag in May 1932, describing his
settlement measure as ‘pure Bolshevism’. Nevertheless, their
language once again seemed calculated to impress Hindenburg and
to repeat the rhetorical success of the previous spring.
On 22 January, one day after the DNVP went on the attack
against Schleicher, Hitler and Papen met again at Ribbentrop’s
house. The presence of Meissner and Oskar von Hindenburg, as
well as Goering and Frick for the National Socialists, lent particular
weight to the occasion. Hitler assured Papen that bourgeois
ministers could be well represented in a presidential government
under his leadership, as long as they were not beholden to their
parties. Goering told Meissner something similar. Papen’s words
indicate that he was prepared to be satisfied with the office of vice
chancellor in a Hitler cabinet. The most important part of the
meeting was a lengthy conversation Hitler held privately with the
son of the president of the Reich. On their drive back to the
Wilhelmstrasse, Oskar von Hindenburg told Meissner that he
thought Hitler’s views made a good deal of sense.
Hindenburg had already been informed about the Dahlem
meeting by the time he received the chancellor for an interview the
next day, on 23 January. Schleicher reported on the cabinet’s
emergency plans and was rebuffed. He would still have to consider
923
the dissolution of the parliament, said Hindenburg, but he could
not at the present time justify delaying new elections:
924
the parliamentary fraction registered ‘the strongest possible protest
against the planned proclamation of a so-called national state of
emergency’. The realization of this plan would be tantamount to a
coup d’état, and that would create an extra-legal situation ‘against
which every resistance is permitted and required’. The next day,
Monsignor Kaas wrote a letter to the chancellor warning him in the
name of the Centre Party against an ‘emergency delay of the date
of new elections’ and reminding him of their last conversation on
16 January, in which Kaas had vehemently criticized ‘the
fundamental tendency of Carl Schmitt and his followers, which is
to relativize the whole legal framework of the state’. ‘The
postponement of elections would be an undeniable violation of the
constitution, with all the legal and political consequences that
would necessarily arise from it… The illegality from above will
give an incalculable impetus to the illegality from below.’ At
Brüning’s suggestion, a copy of the letter was sent to the president
of the Reich.
At the end of January 1933 the Centre and Social Democratic
parties were behaving as if Schleicher represented a greater threat
to the Weimar Republic than Hitler. For these, the two largest
democratic parties in the country, the violation of a single article of
the Weimar constitution represented the greatest possible danger,
not the abolition of the constitution itself. The Centre had been
saying publicly for a long time that it considered a Hitler
chancellorship—provided it was backed by a parliamentary
majority and swore fealty to the constitution—to be a
democratically correct solution to the crisis, if not the only
legitimate one. The SPD had not, as yet, endorsed this point of
925
view. On 25 January, however, the delegate Siegfried Aufhäuser,
head of the a group of employees’ associations (the
Arbeitsgemeinschaft freier Angestelltenverbände) linked with the Free
Trade Unions, demanded that the parliament ‘convene and take
action to express the lack of confidence of the entire population in
the current governmental authority’. This could only be understood
as an attempt to gain National Socialist support for the fight
against Schleicher. The campaign against the delay of new
elections does seem to indicate that the Social Democrats also
believed that a Hitler government, coming to power by legal
means, was a lesser evil than a provisional Schleicher
dictatorship.57
On 27 January Berlin was full of rumors about another kind of
dictatorship, a Papen-led ‘battle cabinet’. It was true that
Hindenburg still wanted to appoint Papen as Schleicher’s successor,
not Hitler. But he was counting on the cooperation of the National
Socialists and sufficient backing for this move in the Reichstag. The
German Nationalists, on the other hand, were propagating the idea
of an anti-parliamentary battle cabinet. Their party leader,
Hugenberg, got into such a row with Hitler on 27 January over the
vexed question of which party would obtain the Prussian interior
ministry that Hitler cancelled a meeting with Papen planned for
that day. When the NSDAP declared publicly that it would fight
with all possible intensity against a dictatorial government headed
by the former chancellor, Papen was impressed. Speaking to
Ribbentrop that evening, he declared himself in favour of a Hitler
chancellorship in stronger language than he had used previously.
Ribbentrop considered it ‘the turning point of the whole issue’.
926
One of those who believed a presidential government led by
Papen or Hugenberg to be the greatest evil was the incumbent
chancellor himself. Such a cabinet, Schleicher announced in a
ministers’ conference on the morning of 28 January, could ‘soon
provoke a national and presidential crisis, since the sentiments of
the broad masses would oppose it in the strongest possible way’.
The difficulties would perhaps not be so great if the president
decided to appoint Hitler chancellor. According to his knowledge,
however, Hindenburg was not prepared to do that. Schleicher did
not believe his own government had any further chance of survival.
He was right. When, shortly after noon, he repeated his request for
an order to dissolve the Reichstag (there was no further mention of
delaying elections), the president tersely turned him down.
Thereupon Schleicher announced the resignation of his cabinet.
An even greater public stir than Schleicher’s dismissal was
caused by the official announcement that the Reich president had
authorized former chancellor Franz von Papen ‘to clarify the
political situation through negotiations with the parties and to
determine the available recourses’. Greatly alarmed, the acting
leaders of the National Association of German Industry and the
German Industry and Trade Association, Ludwig Kastl and Eduard
Hamm, warned State Secretary Meissner about the harm the
political crisis threatened to cause to the German economy. Labour
unions of all political stripes sent a telegram to the president
informing him that the whole of the German working class would
look upon the ‘appointment of a socially reactionary and labour-
hostile government’ as a provocation. The Centre Party and the
Bavarian prime minister, Heinrich Held, expressed themselves
927
similarly.
928
the evening of 28 January, he was able to inform Hindenburg that
reliable conservative politicians would determine the character of
Hitler’s cabinet, the president was impressed. For the first time, he
was prepared to give up his reservations against a Hitler
chancellorship.
The German Nationalists caused Papen the greatest difficulties.
Hugenberg was under pressure from politicians like Ewald von
Kleist-Schmenzin and Otto Schmidt-Hannover, who advocated an
authoritarian government and demanded that Hitler be fought
energetically. The party head himself had strong objections to the
National Socialists’ demand for new elections. In his view, what
made participation in a Hitler–Papen cabinet a possibility was the
fact that Hindenburg agreed to one of Hugenberg’s major demands,
his appointment as minister of economics and agriculture, both in
the Reich government and in Prussia.
Hitler, for his part, had to accept that Papen would be the
federal commissioner for Prussia, not Hitler. In exchange, Goering
was given the office of deputy commissioner, responsible for the
Prussian interior ministry and therefore in control of the police
forces of the largest German state. Goering also received the posts
of federal minister without portfolio and federal commissioner for
air transport. The federal ministry of the interior was granted to
Wilhelm Frick. Thus only three members of the NSDAP sat in the
cabinet. In numerical terms the conservatives—including Labour
Minister Franz Seldte, the leader of the Stahlhelm—were clearly
dominant.
One minister was appointed by Hindenburg himself. The district
commander for East Prussia, General von Blomberg (who on 29
929
January was in Geneva as technical adviser to the German
delegation at the disarmament conference) became Schleicher’s
successor as defence minister. Rumours that the Potsdam garrison
was planning a coup (false, as it turned out) caused Hindenburg to
swear Blomberg into office on the morning of 30 January,
immediately after his arrival in Berlin. This was a violation of the
constitution. The president could only appoint ministers when
asked to do so by the chancellor, and the chancellor was not yet in
office.
It long remained unclear whether the president would grant the
National Socialist demand for the dissolution of the Reichstag and
new elections, a demand Hitler justified by claiming that there was
no majority in the current parliament for the enabling law he
considered absolutely necessary. On 29 January Papen seems to
have convinced Hindenburg to agree to the move if the Centre and
the Bavarian People’s Party proved unwilling to support the new
government in any way. Hitler did not find it difficult to announce
negotiations with the two Catholic parties. Finally, after
Hugenberg, too, had given in on the question of new elections,
Hitler and the members of his cabinet took their oaths of office on
the Weimar constitution in the late afternoon of 30 January 1933.
Hindenburg concluded the brief ceremony with the words: ‘And
now, gentlemen, forward with God!’58
While in the Wilhelmstrasse the die was being cast for
Germany’s fate, the executive of the SPD was meeting with
representatives of the party fraction and the ADGB in the nearby
Reichstag building. When they received the news of the
appointment of Hitler’s cabinet, the leaders of the party and
930
fraction reacted by issuing an appeal cautioning against
‘undisciplined behaviour by individual organizations and groups
acting on their own’ and calling ‘cool-headedness, resolve’ the
demands of the hour. In a meeting of the party executive on the
next day, Rudolf Breitscheid, representing Otto Wels, who was ill,
emphatically rejected extra-parliamentary action:
931
But a unitary proletarian front had even less chance of success
on 30 January 1933 than on 20 July 1932. With over 6 million
Germans officially registered as unemployed, a longer general
strike was out of the question, and a limited general strike would
have been interpreted by the new government more as a sign of
weakness than as a demonstration of strength. Moreover, it was
extremely improbable that the Communists would have heeded any
call to bring the strike to an end. The Communist rhetoric of a
common defensive action lacked the key ingredient: credibility. For
years, the KPD had been waging war against the Social Democrats
as the ‘main social buttress of the bourgeoisie’ and as ‘social
fascists’. As late as 26 January, the Rote Fahne had rejected the
Vorwärts proposal that the SPD and KPD should conclude a ‘non-
aggression pact’ as an ‘infamous disparagement of anti-fascist
Berlin’. The Social Democrats and the Free Trade Unions could not
but expect that the Communists would immediately go over to the
kind of revolutionary violence the National Socialists were waiting
for in order to lend their terror a semblance of legitimacy. A civil
war could end only with the bloody defeat of the workers’
organizations. A divided left had no chance against the forces that
rightist groups, the police, and the army could call in.
On the evening of 30 January 1933 the streets belonged to
Hitler’s ‘brown battalions’, not only in Berlin, but in many places in
Germany. On the next day, in accordance with his promise to
Papen, the new chancellor began his negotiations with the Centre.
But they were mere pretence. Hitler’s true aim was to show that no
governance was possible with the parliament elected on 6
November 1932. For its part, the Centre was still interested in a
932
bona fide coalition with the NSDAP and much less indignant about
Hitler’s chancellorship than about the ‘reactionary’ character of his
cabinet. Nonetheless, Kaas had to reject Hitler’s demand that the
Reichstag be adjourned for one year, and in doing so, he gave the
chancellor the excuse he needed to declare the negotiations a
failure and to bring about the first major decision of his cabinet,
the request to Hindenburg for the dissolution of the Reichstag. The
order was given by the president on 1 February, as well as another
setting new elections for 5 March 1933. Until that date, the Hitler
cabinet would be able—indeed compelled—to govern by means of
the emergency powers of Article.59
933
year before, considering the declarations from the political centre
and the Social Democrats. Nothing compelled the president to
make Hitler chancellor. It was true that, despite the NSDAP losses
in the 6 November elections, Hitler was still the leader of the
strongest party. But he did not have a majority in the Reichstag.
Hindenburg had opposed a Hitler chancellorship until January
1933, seeking to prevent a National Socialist Party dictatorship. He
changed his position after his closest political advisers put pressure
on him, and also because he believed that the predominance of
conservative ministers in Hitler’s cabinet would reduce, if not
eliminate, the risk of dictatorship. The pressure came—directly—
from the large-scale agricultural interests in the east and—
indirectly, via Papen—from the right wing of heavy industry. It
came, as well, from nearly everybody who had access to
Hindenburg. To withstand this pressure was now beyond the
capacity of the old man. In January 1933 the power centre around
Hindenburg decided to take its chances with Hitler, and
Hindenburg the person was only one part of this centre of power.
Thus 30 January 1933 was neither the inevitable result of prior
political developments nor the product of chance. Hitler’s mass
support made his appointment possible, but it was the will of
Hindenburg and of the political milieu he embodied that made him
chancellor. The power of the ‘old elite’ demanding a ‘government
of national concentration’ under Hitler was, no less than the
popularity of his party, a social fact with a long history. The
erosion of confidence in the democratic state was a part of this
history. If the ‘belief in legitimacy’ (Legitimitätsglaube), which Max
Weber considered the most important non-material resource of
934
rulership, was weak in Weimar from the very beginning, the
reasons lay both in the nature of this beginning—the republic’s
birth out of Germany’s defeat in the First World War—as well as in
events, personalities, and phenomena that lay far back in German
history. If the collapse of the first German Republic can be traced
back to a single root cause, it lies in the long historical deferment
of the question of liberty in the nineteenth century—or, to put it
another way, in the non-simultaneity of Germany’s political
modernization: the early democratization of suffrage and the late
democratization of the system of government. Hitler was, after
1930, the main beneficiary of this contradiction and built the
foundations of his success upon it.
In his plan to destroy Weimar democracy, Hitler availed himself
of all the possibilities the Weimar constitution had to offer. The
tactics of legality he imposed on his party were far more successful
than the revolutionary violence he had professed ten years
previously—a credo to which the KPD, the other totalitarian party,
continued to subscribe. In openly advocating civil war, the
Communists gave the National Socialists—who had the largest civil
war army of all—the opportunity to present themselves as
guardians of the constitution, as a factor of order, standing ready
with the police and, if necessary, the army to strike down a coup
attempt from the left. At the same time, Hitler could himself
threaten the rulers of the country with revolutionary violence and
civil war if they broke the law or changed it to the detriment of the
National Socialists, as in the case of the emergency measure against
political terror of 9 August 1932.
Hitler’s conditional promise of legality, which contained an
935
implicit threat, fulfilled its purpose. The traditional right’s fear of
the revolutionary character of National Socialism ultimately gave
way to the belief that the leader of the ‘national’ masses would
provide the urgently needed popular basis for an authoritarian
politics. The illusions of the authoritarians were flanked by the
illusions of the democrats. In order to preserve the state under the
rule of law, its defenders would have had to violate—even if only
by accepting a ‘negative’ vote of no confidence—the letter of a
constitution that was, ultimately, neutral with regard to its own
validity. What stopped them was the predominance of a
‘functionalist’ view of legality, an attitude Carl Schmitt had
lambasted in the summer of 1932, saying that ‘it will remain
neutral to the point of suicide.’ Ernst Fraenkel had the same thing
in mind when, at the end of 1932, he castigated the widespread
‘constitutional fetishism’ of the period. Weimar had fallen into the
legality trap, which the fathers and mothers of the constitution
themselves had set.60
936
Looking ahead
During the night of 17–18 December 1941, nearly nine years after
he had come to power, Hitler, sitting in his main headquarters in
the ‘Wolf’s Den’ near Rastenburg, attempted to put the events of
the day in historical perspective. ‘At the time of the accession to
power,’ he said,
937
1919. It was more than one state among other states. As the
political right of the early 1930s saw it, the Reich was a ‘Greater
Germany’ by its very nature. Bismarck’s ‘Little Germany’, according
to the dominant opinion of the time, may have been the only viable
answer to the German question of those days; ever since the
downfall of the Habsburg monarchy, however, the nation state of
1871 was thought to be unperfected and no longer the last word of
history. The end of the multinational Austrian empire had removed
the strongest argument against the Greater Germany solution and
in favour of a Little Germany. After 1918, there were two republics
that considered themselves German, whose unification was
prevented not by the right to national self-determination, but by
the will of the victorious powers.
In 1920, shortly before the fiftieth anniversary of the founding
of the Reich, the historian Hermann Oncken pronounced the
following judgement:
938
idea, along with its supplement in the narrower and wider
federation, must automatically be absorbed into the idea of Greater
Germany.
939
road to the realization of a Greater Germany, not as its refutation.
The complete and not merely commercial Anschluss of Austria was
itself considered only an intermediate stop along the road to a
German hegemony in central Europe—a vision that had before
inspired the liberals in the Frankfurt Paulskirche and liberal
imperialists like Friedrich Naumann during the time of the Austro-
German alliance in the First World War.
A part of this project was the economic-political attachment to
Germany of Zwischeneuropa (‘Europe-in-Between’), a term
appearing in the title of a 1932 book by the jungkonservativ
publicist Giselher Wirsing, who would later become a Sturmbann
leader in the SS and, later still, editor-in-chief of the Evangelical
weekly Christ und Welt. During the economic crisis, voices calling
for a great central European economic area controlled by Germany
became louder. For intellectuals of the ‘conservative revolution’,
the myth of the old, supra-national Reich served to lend an aura of
historical legitimacy to the German claims to ascendancy.3
An active German ethnic politics was narrowly linked to the
cultivation of Reich ideology. Ethnic Germans outside the borders
of the Reich—in Poland, Czechoslovakia, and in the rest of eastern
central and southeastern Europe—were actively supported by
organizations like the League for Germans Abroad (Verein für das
Deutschtum im Ausland) and by official agencies, led by the foreign
office. This support was both defensive and offensive. The
defensive side was the opposition to efforts aimed at the
assimilation—that is, denationalization—of the German minorities
in the new national states. The offensive side was the attempt to
exploit Germans abroad for the purposes of German hegemonial
940
politics. This second aspect was a great deal more pronounced in
the period of presidential cabinets beginning with Brüning than
during the Stresemann era. The material furtherance of Germans
living abroad was accompanied by the scholarly investigation of
ethnic German communities outside of Germany. After 1931 these
studies were coordinated by research instututes like the
Volksdeutschen Forschungsgemeinschaften. The idea of the Volk
became, along with the idea of the Reich, a central part of the
jungkonservativ drive to intellectually supersede the Weimar state.4
In many ways, therefore, the ground was prepared for Hitler’s
rise to power. The National Socialist leader pursued incomparably
more radical goals than the academics in the circles of the
‘conservative revolution’. But since there was a wide field of
agreement between them and the National Socialists, they could
become a kind of intellectual reserve army for Hitler’s intellectually
challenged movement.
The mythology of the Reich was the most important bridge
between Hitler and large numbers of educated Germans. Hitler’s
intuitive grasp of the opportunities implicit in this concept was one
of the conditions for his success. The Reichsidee was the memory of
the greatness of the German Middle Ages and of the burden
Germany had taken upon itself in those days, the defence of the
whole of the Christian west against the threat from the heathen
east. In the heads and hearts of its devotees, the idea of the Reich
had survived the humiliations to which Germany had been
subjected from the west, by France, for centuries: the Peace of
Westphalia, the conquests of Louis XIV and Napoleon, the Treaty of
Versailles. ‘The Reich’ stood for a European order determined by
941
Germany, and as such was the German answer to the revolutions
and ideas of 1789 and 1917. ‘The Reich’ was the earthly reflection
of the Eternal, and as such formed the ultimate ground of a
particular German mission. They, the Germans, must lead Europe,
for only they had a universal calling, one elevating them far above
the other nations and their nation states.
Like the Volksgemeinschaft, the ‘community of the Volk’, the
‘Reich’ was a vision well suited to counter and transcend the
fragmentation of Germany into parties, classes, and confessions. In
Hitler’s opinion the Weimar Republic, unable to bring forth a
unitary political will, had failed in this task. He, the Führer,
claimed to embody the will of the nation. He knew the deep
longing for a saviour of Germany, who would extinguish the
disgrace of Versailles and all other humiliations and overcome the
internal divisions. He saw himself as this saviour. He was
convinced that the rescue of Germany demanded the suppression of
the Jews, and he allowed no doubts on this point among his
faithful. At no time did he distance himself from the anti-Jewish
excesses of his SA, which were especially frequent after the
Reichstag elections of September 1930 and July 1932. For Hitler
himself, however, anti-Semitism was not the most important aspect
of National Socialist agitation in the final years of the Weimar
Republic. He had long before won over the radical anti-Jewish
minority. A ‘moderate’ anti-Semite was not offensive to the
majority of Germans, but raw violence was. Hitler took this into
account in his campaign speeches.
In the last years before 1933, Hitler no longer referred to the
great war he was always resolved to wage. He spoke of ‘work and
942
bread’, of the reconciliation between the bourgeois citizen and the
worker, between nationalism and socialism, of the end of class
struggle and civil war, of the community of the German people. He
promised to drive the ‘thirty parties’ from the country, and invoked
the ‘new German Reich of greatness, of might and strength, of
power and of glory and of social justice’.
What Hitler did not say was that his German Reich would no
longer be what it had been during the time of the monarchy and
the republic—a constitutional state under the rule of law. But
nobody who had read Mein Kampf or listened to his speeches could
doubt his resolve to break radically with everything even remotely
connected with liberalism and enlightenment. When, on 30
January 1933, Hindenburg appointed him chancellor, Hitler had
the chance to suit his actions to his words and to shape Germany in
his own image.5
943
Notes
944
Mierow, ed. Austin P. Evans and Charles Knapp (New York:
Columbia University Press, 1928).
4. Karl Ferdinand Werner, ‘Les nations et le sentiment national
dans l’Europe médiévale’, Revue Historique, 244 (1970), 285–304;
id., ‘Mittelalter’ in the article ‘Volk, Nation, Nationalismus’, in
Geschichtliche Grundbegriffe. Historisches Lexikon zur
politischsozialen Sprache in Deutschland, ed. Otto Brunner, Werner
Conze, Reinhart Koselleck, vii (Stuttgart: E. Klett, 1992), 171–
281 (esp. 199 ff.); the article ‘Reich’, ibid., v. 423–508; Joachim
Ehlers, ‘Schriftkultur, Ethnogenese und Nationsbildung in
ottonischer Zeit’, Frühmittelalterliche Studien, 23 (1989), 302–17;
Johannes Fried, Der Weg in die Geschichte. Die Ursprünge
Deutschlands bis 1024 (Berlin: Propyläen Verlag, 1994), esp. 9 ff.,
853 ff.; Peter Moraw, ‘Vom deutschen Zusammenhalt in älterer
Zeit’, in Identität und Geschichte, ed. Matthias Werner (Weimar:
Böhlau, 1997), 27–59; Percy Ernst Schramm, Kaiser, Rom und
Renovatio. Studien und Texte zur Geschichte des römischen
Erneuerungsgedankens vom Ende des karolingischen Reiches bis zum
Investiturstreit, 2 vols. (Berlin: B. G. Teubner, 1929), i. 12 ff.; Gerd
Tellenbach, Libertas. Kirche und Weltordnung im Zeitalter des
Investiturstreites (Stuttgart: W. Kohlkammer, 1936), 16 ff., tr. as
Church, State, and Christian Society at the Time of the Investiture
Contest (Oxford: Blackwell, 1940).
5. Karl Ferdinand Werner, ‘Das hochmittelalterliche Imperium im
politischen Bewusstsein Frankreichs (10.–12. Jahrhundert)’, HZ
200 (1965), 2–60 (esp. 50 ff.); Heinz Löwe, ‘Kaisertum und
Abendland in ottonischer und frühsalischer Zeit’, HZ 196 (1963),
945
529–62.
6. The Letters of John of Salisbury, ed. W. J. Millor and H. E. Butler
(Latin-English), i (London, 1955), 206 (letter 124, to Master
Ralph of Sarre, June/July 1160); a German translation can be
found in Das Reich des Mittelalters, 3rd edn, ed. Joachim
Leuschner (Stuttgart 1972), 20–1; Horst Fuhrmann, ‘“Wer hat die
Deutschen zu Richtern über die Völker bestellt?” Die Deutschen
als Ärgernis im Mittelalter’, GWU 46 (1995), 625–41.
7. Rauh, Bild (note 2), 365 ff.; Friedrich Heer, Die Tragödie des
Heiligen Reiches (Stuttgart: W. Kohlhammer, 1952), 118 ff., 141
ff., 240 ff.; Gottfried Koch, Auf dem Weg zum Sacrum Imperium.
Studien zur ideologischen Herrschaftsbegründung der deutschen
Zentralgewalt im 11. u. 12. Jahrhundert (Berlin: Akademie Verlag,
1972), 149 ff.; Timothy Reuter, ‘The Medieval German
“Sonderweg”? The Empire and its Rulers in the High Middle
Ages’, in King and Kingship in Medieval Europe, ed. Anne Duggan
(London: 1993), 179–211; Karl Langosch, Geistliche Spiele.
Lateinische Dramen mit deutschen Versen (Darmstadt:
Wissenschaftliche Buchgesellschaft, 1957), 179–239 (Das Spiel
vom deutschen Kaiser und vom Antichrist [Ludus de Antichristo]).
8. Karl Jordan, ‘Investiturstreit und frühe Stauferzeit (1056–1197)’,
Frühzeit und Mittelalter, 9th edn, ed. Herbert Grundmann (=
Bruno Gebhardt, Handbuch der deutschen Geschichte, i) (Stuttgart:
Union Verlag, 1970), 223–425 (391 ff).
9. Hartmut Boockmann, Stauferzeit und spätes Mittelalter.
Deutschland 1125–1517 (Berlin: Siedler, 1987), 165 ff.
10. Geoffrey Barraclough, Die mittelalterlichen Grundlagen des
946
modernen Deutschland (Weimar: Herman Böhlaus Nachfolger,
1955), 195 ff., orig. The Origins of Modern Germany (Oxford:
Blackwell, 1946).
11. Alexander von Roes, Schriften, ed. Herbert Grundmann and
Hermann Heimpel (Monumenta Germaniae Historica [MGH].
Staatsschriften des späteren Mittelalters, i. Die Schriften des
Alexander von Roes und des Engelbert von Admont, 1. Stück:
Alexander von Roes) (Stuttgart: A. Hiersemann, 1958), 91–148
(Memoriale de prerogativa Romani Imperii, 126–7). A German
translation can be found in Alexander von Roes, Schriften, ed.
Herbert Grundmann and Hermann Heimpel, Deutsches
Mittelalter. Kritische Studientexte der MGH, 4 (Weimar, 1949),
18–67 (49); Hermann Heimpel, ‘Alexander von Roes und das
deutsche Selbstbewusstsein des 13. Jahrhunderts’, Deutsches
Mittelalter (Leipzig, 1941), 79–110.
12. Leuschner, Reich (note 6), 38–9.
13. Eugen Rosenstock-Huessy, Die europäischen Revolutionen und der
Charakter der Nationen, 1st edn (1931), 3rd edn (Stuttgart: W.
Kohlhammer, 1961), 131 ff.
14. Ibid. 239.
15. Marsilius von Padua, Der Verteidiger des Friedens (Defensor Pacis),
ed. Horst Kusch, based on the trans by Walter Kunzmann (Latin–
German) (Berlin: Rütten & Loening, 1958), part 2, 1078–85, tr.
as The Defender of Peace, tr. and introd. Alan Gewirth (New York,
Harper & Row, 1967); Wilhelm von Ockham, Texte zur politischen
Theorie. Exzerpte aus dem Dialogus, selected, trans., and ed.
Jürgen Miethke (Latin–German) (Stuttgart, 1995), 226–309;
947
William of Ockham, On the Power of Emperors and Popes, ed. and
trans. Annabel S. Brett (Durham: University of Durham; Sterling,
VA: Thoemmes Press, 1998).
16. Alois Dempf, Sacrum Imperium. Geschichts- und Staatsphilosophie
des Mittelalters und der politischen Renaissance (Munich:
Oldenbourg, 1929), 544; Ernst Troeltsch, Die Soziallehren der
christlichen Kirchen und Gruppen, in Gesammelte Schriften, i
(Tübingen: J. C. B. Mohr, 1912), 420, 432 ff., 794 ff., 858 ff., tr.
as The Social Teaching of the Christian Churches (Chicago:
University of Chicago Press, 1931, repr. 1981).
17. Eberhard Isenmann, ‘Kaiser, Reich und deutsche Nation am
Ausgang des 15. Jahrhunderts’, Ansätze und Diskontinuität
deutscher Nationsbildung im Mittelalter, ed. Joachim Ehlers
(Sigmaringen: Thorbecke, 1989), 145–246 (esp. 155 ff.); Alfred
Schröcker, Die Deutsche Nation. Beobachtungen zur politischen
Propaganda des ausgehenden 15. Jahrhunderts (Lübeck: Matthiesen
Verlag, 1974), esp. 116 ff.; Ulrich Nonn, ‘Heiliges Römisches
Reich Deutscher Nation. Zum Nationen-Begriff im 15.
Jahrhundert’, Zeitschrift für Historische Forschung, 9 (1982), 129–
42.
18. Heinz Angermeier, Die Reichsreform 1410–1555. Die
Staatsproblematik in Deutschland zwischen Mittelalter und
Gegenwart (Munich: C. H. Beck, 1984); Georg Schmidt, Geschichte
des Alten Reiches. Staat und Nation in der Frühen Neuzeit 1495–
1806 (Munich: C. H. Beck, 1999), 33 ff. (the ideas of the
‘komplementärer Reichs-Staat’, the ‘Verstaatung des Alten Reiches’
around 1500 and of the Old Reich as state of the German
948
nation); Wolfgang Reinhard, Geschichte der Staatsgewalt. Eine
vergleichende Verfassungsgeschichte Europas von den Anfängen bis
zur Gegenwart (Munich: C. H. Beck, 1999), 52 ff.
19. Max Wehrli, ‘Der Nationalgedanke im deutschen und
schweizerischen Humanismus’, Nationalismus in Germanistik und
Dichtung. Dokumentation des Germanistentages in München vom
17.–22. Oktober 1966, ed. Benno von Wiese and Rudolf Henss
(Berlin: E. Schmidt, 1967), 126–43 (quote: 131); Goez, Translatio
(note 2), 248–57; Bernd Schönemann, ‘Frühe Neuzeit und 19.
Jahrhundert’, in the article ‘Volk, Nation, Nationalismus’ (note
4), 281–431 (esp. 288 ff.); Herfried Münkler et al.,
Nationenbildung. Die Nationalisierung Europas im Diskurs
humanistischer Intellektueller. Italien und Deutschland (Berlin:
Akademie Verlag, 1998); id., ‘Nation als politische Idee im
frühneuzeitlichen Europa’, Nation und Literatur im Europa der
Frühen Neuzeit, ed. Klaus Garber (Tübingen: Niemeyer, 1989),
56–86; Wolfgang Hardtwig, ‘Ulrich von Hutten. Zum Verhältnis
von Individuum, Stand und Nation in der Reformationszeit’, in
Nationalismus und Bürgerkultur in Deutschland 1500–1914
(Göttingen: Vandenhoeck & Ruprecht, 1994), 15–33; Frank L.
Borchardt, German Antiquity in Renaissance Myth (Baltimore:
Johns Hopkins University Press, 1971); Ludwig Krapf,
Germanenmythos und Reichsideologie. Frühhumanistische
Rezeptionsweisen der taciteischen ‘Germania’ (Tübingen: Niemeyer,
1979).
20. Martin Luther, An den christlichen Adel deutscher Nation von des
christlichen Standes Besserung (1520), in D. Martin Luthers Werke,
949
complete critical edition, vi (Weimar, 1888), 381–469 (462–5),
tr. as To the Christian Nobility of the German Nation concerning the
Reform of the Christian Estate (1520), in Luther’s Works (St Louis
and Philadephia: Concordia Publishing House and Fortress Press,
1958–), xliv. 123–217 (207–10). For a modern German
translation, see Martin Luther, An den christlichen Adel deutscher
Nation und andere Schriften, 2nd edn (Stuttgart, 1964), 98–103.
21. Georg Wilhelm Friedrich Hegel, Vorlesungen über die Philosophie
der Geschichte, in Sämtliche Werke, xi, 3rd edn (Stuttgart:
Fromman, 1949), 519, 521 ff., 524, tr. as The Philosophy of
History, trans. J. Sibree (New York: Wiley, 1944), 414–17.
22. Karl Marx, ‘Zur Kritik der Hegelschen Rechtsphilosophie.
Einleitung’, Karl Marx and Friedrich Engels, Werke (= MEW;
Berlin, 1956–), i. 385–6, tr. as ‘Contribution to the Critique of
Hegel’s Philosphy of Law’, Karl Marx/Friedrich Engels, Collected
Works (= MECW; International Publishers: New York, 1975–
2004), iii. 175–87 (182).
23. Friedrich Nietzsche, Der Antichrist, in Werke (complete critical
edition, ed. Giorgio Colli and Mazzino Montinari), pt 6 (Berlin,
1969), iii. 248–9, tr. in Complete Works, ed. Oscar Levy and
Robert Guppy (New York: Russell & Russell, 1964), xvi. 228–9.
24. Marx, ‘Kritik’ (note 22), 391; Friedrich Engels, Zum ‘Bauernkrieg’
(1884), MEW, xxi. 402, tr. as; ‘On The Peasants’ War’, MECW,
xxvi. 554–5; Marx, ‘Critique’ (note 22), 187.
25. Peter Blickle, Die Revolution von 1525, 2nd edn (Munich:
Oldenbourg, 1981), tr. as The Revolution of 1525: The German
Peasants’ War from a New Perspective (Baltimore: Johns Hopkins
950
University Press, 1981), id., Gemeindereformation. Die Menschen
des 16. Jahrhunderts auf dem Weg zum Heil (Munich: Oldenbourg,
1985), tr. as Communal Reformation: The Quest for Salvation in
Sixteenth-Century Germany (Atlantic Highlands, NJ: Humanities
Press, 1992); Richard von Dülmen, Reformation als Revolution.
Soziale Bewegung und religiöser Radikalismus in der deutschen
Reformation (Munich: Deutscher Taschenbuch Verlag, 1977).
26. Rosenstock-Huessy, Revolutionen (note 13), 234–5.
27. Heinz Schilling, Aufbruch und Krise. Deutschland 1517–1648, 2nd
edn (Berlin: Siedler, 1994), 184 ff.
28. Troeltsch, Soziallehren (note 16), 518.
29. Ibid. 519, 684.
30. Franz Borkenau, ‘Luther: Ost oder West’, in Zwei Abhandlungen
zur deutschen Geschichte (Frankfurt: V. Klostermann, 1947), 45–
75 (59).
31. Ibid. 74; Helmuth Plessner, Die verspätete Nation. Über die
politische Verführbarkeit bürgerlichen Geistes (1st edn under the
title Das Schicksal deutschen Geistes im Ausgang seiner bürgerlichen
Epoche (Zurich, 1935)) (Stuttgart: W. Kohlhammer, 1959), 58 ff.
32. Martin Luther, Von den Juden und ihren Lügen (1543), in Werke
(note 20), liii (Weimar, 1920), 413–552 (esp. 522 ff.; quote:
479), tr. as On the Jews and their Lies (1543), in Works (note 20),
xlvii. 121–306 (esp. 267 ff.; quote: 213–14). Klaus Deppermann,
‘Judenhass und Judenfreundschaft im frühen Protestantismus’,
Die Juden als Minderheit in der Geschichte, ed. Bernd Martin and
Ernst Schulin (Munich: Deutscher Taschenbuch Verlag, 1981),
110–30; Heiko A. Oberman, Wurzeln des Antisemitismus.
951
Christenangst und Judenplage im Zeitalter von Humanismus und
Reformation, 2nd edn (Berlin: Severin und Siedler, 1983), 56 ff.,
125 ff., tr. as The Roots of anti-Semitism in the Age of Renaissance
and Reformation (Philadelphia: Fortress Press, 1984); Joshua
Trachtenberg, The Devil and the Jews: The Medieval Conception of
the Jew and its Relation to Modern Antisemitism (New Haven: Yale
University Press, 1943).
33. Leopold von Ranke, Deutsche Geschichte im Zeitalter der
Reformation, in Leopold von Ranke’s Werke (complete edition of
the Deutsche Akademie) (Munich: Duncker & Humblot, 1925), i.
4, tr. as History of the Reformation in Germany (New York: F.
Unger, 1905, repr. of 1966); Rosenstock-Huessy, Revolutionen
(note 13), 224; Paul Joachimsen, Vom deutschen Volk zum
deutschen Staat. Eine Geschichte des deutschen Nationalbewusstseins
(1916), 3rd edn, ed. and updated Joachim Leuschner (Göttingen:
Vandenhoeck & Ruprecht, 1956), 25 ff.; Heinrich Lutz, ‘Die
deutsche Nation zu Beginn der Neuzeit. Fragen nach dem
Gelingen und Scheitern deutscher Einheit im 16. Jahrhundert’,
HZ 234 (1982), 529–59.
34. Schilling, Aufbruch (note 27), 243.
35. Max Weber, ‘Die protestantische Ethik und der Geist des
Kapitalismus’, Archiv für Sozialwissenschaft und Sozialpolitik, 20
(1905), 1–54; 21 (1905), 1–110, tr. as The Protestant Ethic and the
Spirit of Capitalism (New York: Scribner, 1958); Alfred Müller-
Armack, Genealogie der Wirtschaftsstile. Die geistesgeschichtlichen
Ursprünge der Staatsund Wirtschaftsformen bis zum Ausgang des 18.
Jahrhunderts (Stuttgart: W. Kohlhammer, 1941); Herbert Lüthy,
952
‘Nochmals: “Calvinismus und Kapitalismus”. Über die Irrwege
einer sozialhistorischen Diskussion’ (1961), repr. in Gesellschaft in
der industriellen Revolution, ed. Rudolf Braun et al. (Cologne:
Kiepenheuer & Witsch, 1973), 18–36; Hartmut Lehmann,
‘Asketischer Protestantismus und ökonomischer Rationalismus:
Die Weber-These nach zwei Generationen’, in Max Webers Sicht
des okzidentalen Christentums, ed. Wolfgang Schluchter (Frankfurt:
Suhrkamp, 1988), 529–53.
36. Schilling, Aufbruch (note 27), 282 ff., 412 ff.; id., ‘Nationale
Identität und Konfession in der europäischen Neuzeit’, in
Nationale und kulturelle Identität. Studien zur Entwicklung des
kollektiven Bewusstseins in der Neuzeit, i, ed. Bernhard Giesen
(Frankfurt: Suhrkamp, 1991), 192–252.
37. Johannes Burkhardt, Der Dreissigjährige Krieg (Frankfurt:
Suhrkamp, 1992), 30 ff.; Georg Schmidt, Der Dreissigjährige Krieg
(Munich: C. H. Beck, 1995); Günter Barudio, Der Teutsche Krieg
1618–1648 (Frankfurt: Fischer, 1988); Adam Wandruszka,
Reichspatriotismus und Reichspolitik zur Zeit des Prager Friedens von
1635. Eine Studie zur Geschichte des deutschen Nationalbewusstseins
(Cologne: H. Böhlaus Nachfolger, 1955).
38. Christoph Dipper, Deutsche Geschichte 1648–1784 (Frankfurt:
Suhrkamp, 1991), 263 ff.; Schilling, Aufbruch (note 27), 396, 436
ff.
39. Burkhardt, Dreissigjähriger Krieg (note 37), 90 ff.
40. Samuel von Pufendorf, Die Verfassung des Deutschen Reiches
(Latin–German), ed. and trans. Horst Denzer (Frankfurt: Insel,
1994), 198–9; Norbert Elias, Über den Prozess der Zivilisation.
953
Soziogenetische und psychogenetische Untersuchungen, 2 vols., 1st
edn (1939), 18th edn (Frankfurt: Suhrkamp, 1993), ii. 129 ff., tr.
as The Civilizing Process (Oxford: Blackwell, 1994); Rudolf
Vierhaus, Staaten und Stände. Vom Westfälischen zum
Hubertusburger Frieden 1648–1763 (Berlin: Propyläen Verlag,
1984), 22 ff.
41. Johann Gustav Droysen, Geschichte der preussischen Politik, ii, pt
2 (Leipzig: Veit, 1870), 436.
42. Otto Hintze, ‘Kalvinismus und Staatsräson in Brandenburg zu
Beginn des 17. Jahrhunderts’, in id., Geist und Epochen der
preussischen Geschichte. Gesammelte Abhandlungen, ed. Fritz
Hartung (Leipzig: Koehler & Amelang, 1943), 289–346 (in the
order of the quotes: 289, 324–5, 345, 315–16, 302–3).
43. Müller-Armack, Genealogie (note 35), 147.
44. Hintze, Kalvinismus (note 42), 300, 315.
45. Troeltsch, Soziallehren (note 16), 599–600.
46. Otto Hintze, ‘Die Hohenzollern und der Adel’, in Geist (note 42),
38–63; Hans Rosenberg, Bureaucracy, Aristocracy, and Autocracy:
The Prussian Experience (Cambridge, MA: Harvard University
Press, 1958).
47. Otto Büsch, Militärsystem und Sozialleben im alten Preussen 1713–
1807. Die Anfänge der sozialen Militarisierung der preussisch-
deutschen Gesellschaft (Berlin: De Gruyter, 1952), 164–5, tr. as
Military System and Social Life in Old Regime Prussia, 1713–1807:
The Beginning of the Social Militarization of Prusso-German Society
(Atlantic Highlands, NJ: Humanities Press, 1997).
48. Heinz Schilling, Höfe und Allianzen. Deutschland 1648–1763, 2nd
954
edn (Berlin: Siedler, 1994), 392 ff.; Carl Hinrichs, Preussentum
und Pietismus. Der Pietismus in Brandenburg-Preussen als religiös-
soziale Reformbewegung (Göttingen: Vandenhoeck & Ruprecht,
1971); Richard L. Gawthrop, Pietism and the Making of Eighteenth-
Century Prussia (Cambridge: Cambridge University Press, 1993);
Gerhard Kaiser, Pietismus und Patriotismus im literarischen
Deutschland. Ein Beitrag zum Problem der Säkularisation
(Wiesbaden: F. Steiner, 1961), 39 ff.; Koppel S. Pinson, Pietism as
a Factor in the Rise of German Nationalism (New York: Octagon,
1934); Hartmut Lehmann, ‘Pietism and Nationalism: The
Relationship between Protestant Revivalism and National
Renewal in Nineteenth-Century Germany’, Church History, 51
(1982), 39–53; Geschichte des Pietismus, ed. Martin Brecht et al.
(Göttingen: Vandenhoeck & Ruprecht, 1993–).
49. Sibylle Badstübner-Gröger et al., Hugenotten in Berlin (Berlin:
Nicolai, 1988), 14 ff.
50. Schilling, Aufbruch (note 27), 415 ff.
51. Ibid. 303 ff.
52. Theodor Schieder, Friedrich der Grosse. Ein Königtum der
Widersprüche (Berlin: Propyläen Verlag, 1983), tr. as Frederick the
Great (London, New York: Longman, 2000); Ingrid Mittenzwei,
Friedrich II. von Preussen (Berlin: Deutscher Verlag, 1979); Rudolf
Augstein, Preussens Friedrich und die Deutschen (Frankfurt: S.
Fischer, 1968).
53. Rudolf von Thadden, Fragen an Preussen. Zur Geschichte eines
aufgehobenen Staates (Munich: C. H. Beck, 1981), 58–9, 167;
Hans Joachim Schoeps, Preussen. Geschichte eines Staates (Berlin:
955
Propyläen Verlag, 1966), 98; Schilling, Höfe (note 48), 431.
54. Reinhart Koselleck, Preussen zwischen Reform und Revolution.
Allgemeines Landrecht, Verwaltung und soziale Bewegung von 1791
bis 1848 (Stuttgart: Klett, 1967), 24–5.
55. Alexis de Tocqueville, L’Ancien Régime et la Révolution, in Œuvres
complètes (Paris: Gallimard, 1952), ii, pt 1. 269, tr. as The Old
Régime and the Revolution (Chicago: University of Chicago Press,
1955), i. 262.
56. Schilling, Höfe (note 48), 320 ff.; Horst Möller, Fürstenstaat oder
Bürgernation. Deutschland 1763–1815 (Berlin: Siedler, 1989), 308
ff.
57. Johann Wolfgang von Goethe, Aus meinem Leben. Dichtung und
Wahrheit, pt 1, bk II, in Werke, Weimar edition (Munich:
Deutscher Taschenbuch Verlag, 1987), xxvi. 71, tr. as From My
Life (New York: Suhrkamp Publishers, 1987); Theodor Schieder,
‘Friedrich der Grosse—eine Integrationsfigur des deutschen
Nationalbewusstseins im 18. Jahrhundert?’, in Nationalismus in
vorindustrieller Zeit, ed. Otto Dann (Munich: Oldenbourg, 1986),
113–27.
58. Thomas Abbt, Vom Tode für das Vaterland, in Vermischte Werke
(repr. Hildesheim: G. Olms, 1978), i. 17; Christoph Prignitz,
Vaterlandsliebe und Freiheit. Deutscher Patriotismus von 1750 bis
1850 (Wiesbaden: Steiner, 1981), 17; Benjamin W. Redekop,
‘Thomas Abbt and the Formation of an Enlightened German
“Public”’, Journal of the History of Ideas, 58 (1997), 81–103.
59. Harm Klueting, ‘“Bürokratischer Patriotismus”. Aspekte des
Patriotismus im theresianisch-josephinischen Österreich’,
956
Patriotismus, ed. Günter Birtsch, Aufklärung 4/2 (1991), 37–52.
60. Christoph Martin Wieland, ‘National-Poesie’ (1773), in
Sämmtliche Werke (Leipzig: G. J. Göschen, 1839–40), xxxv. 327–
35 (327); Prignitz, Vaterlandsliebe (note 58), 7 (Wieland); Rudolf
Vierhaus, ‘“Patriotismus”-Begriff und Realität einer
moralischpolitischen Haltung’, in Deutsche patriotische und
gemeinnützige Gesellschaften, ed. id. (Munich: Kraus International
Publ., 1980), 9–29 (14: Goethe, Möser, Nicolai); Horst Möller,
Aufklärung in Preussen. Der Verleger, Publizist und
Geschichtsschreiber Friedrich Nicolai (Berlin: Colloquium Verlag,
1974), 565 ff.
61. Pufendorf, Verfassung (note 40), 46–7; Hermann Conrad,
Deutsche Rechtsgeschichte, ii (Karlsruhe: C. F. Müller, 1966), 113
ff.; Jürgen Luh, Unheiliges Römisches Reich. Der konfessionelle
Gegensatz 1648 bis 1806 (Potsdam: Verlag für Berlin-
Brandenburg, 1995); Reinhard Stauber, ‘Nationalismus vor dem
Nationalismus? Eine Bestandsaufnahme der Forschung zu
“Nation” und “Nationalismus” in der Frühen Neuzeit’, GWU 47
(1996), 139–65; Manfred Jacobs, ‘Die Entwicklung des deutschen
Nationalgedankens von der Reformation bis zum deutschen
Idealismus’, in Volk— Nation—Vaterland. Der deutsche
Protestantismus und der Nationalismus, ed. Horst Zillessen
(Gütersloh: G. Mohn, 1970), 51–110.
62. Pufendorf, Verfassung (note 40), 46 ff.; Christoph Martin
Wieland, ‘Über Deutschen Patriotismus. Betrachtungen, Fragen
und Zweifel’ (May 1793), in Werke (note 60), 245–59 (252);
Prignitz, Vaterlandsliebe (note 58), 7 ff.; Gertrude Lübbe-Wolff,
957
‘Die Bedeutung der Lehre von den vier Weltreichen für das
Staatsrecht des römischdeutschen Reichs’, Der Staat, 23 (1984),
387–91; Notker Hammerstein, ‘“Imperium Romanum cum
omnibus suis qualitatibus ad Germanos est translatum”. Das
vierte Weltreich in der Lehre der Reichsjuristen’, Zeitschrift für
historische Forschung, suppl. 3 (1987), 187–202.
63. Reinhard Rürup, Johann Jacob Moser. Pietismus und Reform
(Wiesbaden: Steiner, 1965), esp. 151–2; Notker Hammerstein,
‘Das politische Denken Friedrich Carl von Mosers’, HZ 212
(1971), 316–38; Michael Stolleis, ‘Reichspublizistik und
Reichspatriotismus vom 16. zum 18. Jahrhundert’, Aufklärung, 4
(1989), 7–23; Karl Otmar Freiherr von Aretin,
‘Reichspatriotismus’, in Patriotismus, ed. Birtsch (note 59), 25–36;
id., Das Alte Reich 1648–1806, 3 vols. (Stuttgart: Klett Cotta,
1993–); John G. Gagliardo, Reich and Nation: The Holy Roman
Empire as Idea and Reality, 1763–1806 (Bloomington: Indiana
University Press, 1980), esp. 49 ff. The quotes: Friedrich Carl von
Moser, Von dem Deutschen Nationalgeist (1766; repr. Selb: Notos,
1976), 19, 21, 40 (emphases in the original).
64. Wolfgang Hardtwig, ‘Vom Elitebewusstsein zur
Massenbewegung. Frühformen des Nationalismus in Deutschland
1500–1840’, in Nationalismus (note 19), 34–54 (quote from
Gottsched: 44; from Hardtwig: 46); Hagen Schulze, Staat und
Nation in der europäischen Geschichte (Munich: C. H. Beck, 1994),
108 ff., 142 ff.; Otto Dann, Nation und Nationalismus in
Deutschland 1770–1990 (Munich: C. H. Beck, 1990), 28 ff.;
Bernhard Giesen and Kay Junge, ‘Vom Patriotismus zum
958
Nationalismus. Zur Evolution der “Deutschen Kulturnation”’, in
Identität, ed. Giesen (note 36), 255–303; Helga Schultz, ‘Mythos
und Aufklärung. Frühformen des Nationalismus in Deutschland’,
HZ 261 (1996), 31–67; Dichter und ihre Nation, ed. Helmut
Scheuer (Frankfurt: Suhrkamp, 1993); Winfried Woesler, ‘Die
Idee der deutschen Nationalliteratur in der zweiten Hälfte des
18. Jahrhunderts’, Nation, ed. Garber (note 19), 716–33; Hans
Peter Herrmann et al., Machtphantasie Deutschland. Nationalismus,
Männlichkeit und Fremdenhass im Vaterlandsdiskurs deutscher
Schriftsteller des 18. Jahrhunderts (Frankfurt: Suhrkamp, 1996);
Jörg Echternkamp, Aufstieg des deutschen Nationalismus (1770–
1840) (Frankfurt: Campus, 1998), 41 ff.
65. Conrad Wiedemann, ‘Deutsche Klassik und nationale Identität.
Eine Revision der Sonderwegs-Frage’, Klassik im Vergleich.
Normativität und Historizität europäischer Klassiken, ed. Wilhelm
Vosskamp (Stuttgart: Metzler, 1993), 541–69 (547–8); Franz
Josef Worstbrock, ‘Translatio artium. Über Herkunft und
Entwicklung einer kulturhistorischen Theorie’, Archiv für
Kulturgeschichte, 47 (1965), 1–22; Manfred Landfester, ‘Griechen
und Deutsche: Der Mythos einer “Wahlverwandtschaft”’, in
Mythos und Nation. Studien zur Entwicklung des kollektiven
Bewusstseins in der Neuzeit, iii. ed. Helmut Berding (Frankfurt:
Suhrkamp, 1996), 198–219.
66. Gotthold Ephraim Lessing, Hamburgische Dramaturgie, parts 101–
4, in Werke, iv (Munich: C. Hanser, 1973), 698; Goethe, Werke
(note 57), v. 218.
67. Johann Gottfried Herder, Briefe zur Beförderung der Humanität, in
959
Werke in zehn Bänden, vii (Frankfurt: Deutscher Klassiker Verlag,
1991), 687 (letter 115 [1797]); id., Ideen zur Philosophie der
Geschichte der Menschheit, ibid. vi (Frankfurt: Deutscher Klassiker
Verlag, 1989), 630, 639 (pt III, 15, I, II); id., Philosophical
Writings (Cambridge: Cambridge University Press, 2002); id.,
Reflections on the Philosophy of the History of Mankind (Chicago:
University of Chicago Press, 1968).
68. Immanuel Kant, Reflexionen zur Anthropologie, in Gesammelte
Schriften (Akademie edn), xv (Berlin: G. Reimer, 1913), 590–1;
Schönemann, Frühe Neuzeit (note 19), 320.
69. Johann Wolfgang von Goethe, Faust. Der Tragödie erster und
zweiter Teil, pt 1 (Auerbach’s Keller in Leipzig), in Werke (note
57), xiv. 99, tr. as Faust, pt 1, trans. Randall Jarrell (New York:
Farrar, Straus & Giroux, 1976); Rudolf Schlögl, ‘Die patriotisch-
gemeinnützigen Gesellschaften. Organisation, Sozialstruktur,
Tätigkeitsfelder’, in Aufklärungsgesellschaften, ed. Helmut
Reinalter (Frankfurt: P. Lang, 1993), 61–81.
70. Herder, Briefe (note 67), 305 letter 57 [1795]; supplement:
‘Haben wir noch das Publikum und Vaterland der Alten?’); Otto
Dann, ‘Herder und die Deutsche Bewegung’, in Gerhard Sauder,
Johann Gottfried Herder 1744–1803 (Hamburg: F. Meiner, 1987),
308–40; Harro Segeberg, ‘Germany’, in Nationalism in the Age of
the French Revolution, ed. Otto Dann and John Dinwiddy
(London: Hambledon Press, 1988), 137–56.
71. Wilhelm Dilthey, ‘Die dichterische und philosophische
Bewegung in Deutschland 1770–1800’, in Die geistige Welt.
Einleitung in die Philosophie des Lebens. Gesammelte Schriften, v,
960
6th edn (Stuttgart: Teubner, 1974), 12–27 (13).
72. Dann, ‘Herder’ (note 70), 324 ff.; Wilhelm Abel, Massenarmut
und Hungerkrisen im vorindustriellen Deutschland (Göttingen:
Vandenhoeck & Ruprecht, 1972), esp. 191 ff., 252 ff.; R. R.
Palmer, The Age of Democratic Revolution: A Political History of
Europe and America, 1760–1800, 2 vols. (Princeton: Princeton
University Press, 1964); Eric J. Hobsbawm, The Age of Revolution:
Europe 1789–1848 (London: Weidenfeld and Nicolson, 1962), tr.
as Europäische Revolutionen 1789–1848 (Zurich: Kindler, 1973);
Reinhart Koselleck, Kritik und Krise. Eine Studie zur Pathogenese
der bürgerlichen Welt (Freiburg: K. Alber, 1959); id.,
‘Introduction’, Geschichtliche Grundbegriffe (note 4), i. pp. xii–
xxvii (for ‘Sattelzeit’, see from about 1750); Jürgen Habermas,
Strukturwandel der Öffentlichkeit. Untersuchungen zu einer Kategorie
der bürgerlichen Gesellschaft (Neuwied: Luchterhand, 1961), esp.
104 ff., tr. as The Structural Transformation of the Public Sphere: An
Inquiry into a Category of Bourgeois Society (Cambridge, MA: MIT
Press, 1989).
961
Philosophy of Right (Cambridge, MA: MIT Press, 1982).
2. Rudolf Vierhaus, ‘Montesquieu in Deutschland. Zur Geschichte
seiner Wirkung als politischer Schriftsteller im 18. Jahrhundert’,
in Deutschland im 18. Jahrhundert. Politische Verfassung, soziales
Gefüge, geistige Bewegungen (Göttingen: Vandenhoeck & Ruprecht,
1987), 9–32; id., ‘Politisches Bewusstsein in Deutschland vor
1789’, ibid. 183–201; id., ‘“Sie und nicht wir.” Deutsche Urteile
über den Ausbruch der Französischen Revolution’, ibid. 202–15.
3. Christoph Martin Wieland, ‘Kosmopolitische Adresse an die
Französische Nationalversammlung, von Eleutherius Philoceltes’
(Oct. 1789), in Sämtliche Werke, xxxi (Leipzig: G. J. Göschen,
1857), 30–58 (58); id., ‘Unparteiische Betrachtungen über die
Staatsrevolution in Frankreich’ (May 1790), ibid. 69–101 (73);
id., ‘Zufällige Gedanken über die Abschaffung des Erbadels in
Frankreich’ (July 1790), ibid. 102–25 (123–4); id., ‘Worte zur
rechten Zeit und an die politischen und moralischen
Gewalthaber. Nachträge’ (Jan. 1793), ibid. 319–25 (320);
Irmtraut Sahmland, Christoph Martin Wieland und die deutsche
Nation. Zwischen Patriotismus, Kosmopolitismus und Griechentum
(Tübingen: M. Niemeyer, 1990).
4. Johann Gottfried Herder, Briefe die Fortschritte der Humanität
betreffend (1792), in Werke in zehn Bänden, vii (Frankfurt:
Deutscher Klassiker Verlag, 1991), 774 (letter 13), 779 (letter
16), 780 (letter 17; emphases in original), tr. in Philosophical
Writings (Cambridge: Cambridge University Press, 2002); Hans-
Wolf Jäger, ‘Herder und die Französische Revolution’, in Johann
Gottfried Herder 1744–1803, ed. Gerhard Sauder (Hamburg: F.
962
Meiner, 1987), 299–307.
5. Georg Forster, ‘Darstellung der Revolution in Mainz’ (1793), in
Werke. Sämtliche Schriften, Tagebücher, Briefe, x.
Revolutionsschriften 1792/93 (Berlin: Akademie Verlag, 1990),
505–91 (556); Volker Mehnert, Protestantismus und radikale
Spätaufklärung. Die Beurteilung Luthers und der Reformation durch
aufgeklärte deutsche Schriftsteller zur Zeit der Französischen
Revolution (Munich: Minerva, 1982), 52–3 (Forster), 55, 117
(Rebmann); Walter Grab, Norddeutsche Jakobiner. Demokratische
Bestrebungen zur Zeit der ersten französischen Republik (Hamburg:
Europäische Verlagsanstalt, 1967); Heinrich Scheel, Süddeutsche
Jakobiner. Klassenkämpfe und republikanische Bestrebungen zur Zeit
der Französischen Revolution (Berlin: Akademie Verlag, 1962);
Franz Dumont, Die Mainzer Republik 1792/93. Studien zur
Revolutionierung in Rheinhessen und der Pfalz (Alzey: Verlag der
Rheinhessischen Druckwerkstätte, 1993); Wolfang Reinbold,
Mythenbildungen und Nationalismus. ‘Deutsche Jakobiner’ zwischen
Revolution und Reaktion (1789–1800) (Bern: Lang, 1999); Jacques
Droz, L’Allemagne et la Révolution Française (Paris: Presses
Universitaries de France, 1949); Volker Press, ‘Warum gab es
keine deutsche Revolution? Deutschland und das revolutionäre
Frankreich’, in Revolution und Krieg. Zur Dynamik historischen
Wandels seit dem 18. Jahrhundert, ed. Dieter Langewiesche
(Paderborn: Schöningh, 1989), 67–85; Ernst Wolfgang Becker,
Zeit der Revolution!—Revolution der Zeit? Zeiterfahrungen in
Deutschland in der Ära der Revolutionen 1789–1848/49
(Göttingen: Vandenhoeck & Ruprecht, 1999), 37 ff.
963
6. Friedrich Meinecke, Das Zeitalter der deutschen Erhebung (1795–
1815) (1906), new (6th) edn (Göttingen: Vandenhoeck &
Ruprecht, 1957), 46 (Struensee), tr. as The Age of German
Liberation, 1795–1815 (Berkeley: University of California Press,
1977); Max Braubach, ‘Von der Französischen Revolution bis
zum Wiener Kongress’, in Von der Französischen Revolution bis
zum Ersten Weltkrieg (= Bruno Gebhardt, Handbuch der deutschen
Geschichte, iii), 9th edn (Stuttgart: Union Deutsche
Verlagsgesellschaft, 1970), 2–96 (35); Reinhart Koselleck,
Preussen zwischen Reform und Revolution. Allgemeines Landrecht,
Verwaltung und soziale Bewegung von 1791 bis 1848 (Stuttgart:
Klett, 1967); Hanna Schissler, Preussische Agrargesellschaft im
Wandel. Wirtschaftliche, gesellschaftliche und politische
Transformationsprozesse von 1763 bis 1847 (Göttingen:
Vandenhoeck & Ruprecht, 1978).
7. Immanuel Kant, Zum ewigen Frieden, in Gesammelte Schriften
(Akademie edn; Berlin: G. Reimer, 1900–), viii. 351 ff., 378; id.,
Der Streit der Fakultäten, ibid., vii. 88; id., Die Metaphysik der
Sitten/Rechtslehre, ibid., vi. 338–42 (§§ 51 and 52; emphases in
original); tr. as Perpetual Peace, ed. and introd. Lewis White Beck
(New York: Liberal Arts Press, 1957); id., The Conflict of the
Faculties, trans. and introd. Mary J. Gregor (New York: Abaris
Books, 1979); id., The Philosophy of Law: An Exposition of the
Fundamental Principles of Jurisprudence as the Science of Right,
trans. W. Hastie (Clifton, NJ: A. M. Kelley, 1974); Karl
Vorländer, ‘Kants Stellung zur französischen Revolution’,
Philosophische Abhandlungen. Festschrift für Hermann Cohen
(Berlin: B. Cassirer, 1912), 247–69; Peter Burg, Kant und die
964
Französische Revolution (Berlin: Duncker und Humblot, 1974);
Zwi Batscha, ‘Bürgerliche Republik und bürgerliche Revolution
bei Immanuel Kant’, in Studien zur politischen Theorie des
deutschen Frühliberalismus (Frankfurt: Suhrkamp, 1981), 43–65;
Volker Gerhardt, Immanuel Kants Entwurf ‘Zum ewigen Frieden’.
Eine Theorie der Politik (Darmstadt: Wissenschaftliche
Buchgesellschaft, 1995), esp. 88–9; Leonard Krieger, The German
Idea of Freedom: History of a Political Tradition (Boston: Beacon
Press, 1957), 86 ff.; Vierhaus, ‘Montesquieu’ (note 2), 9 ff.; the
article ‘Demokratie’, in Geschichtliche Grundbegriffe. Historisches
Lexikon zur politisch-sozialen Sprache in Deutschland, ed. Otto
Brunner, Reinhart Koselleck, Werner Conze, i (Stuttgart: Klett,
1972), 820–99 (esp. 822 ff.); Hajo Holborn, ‘Der deutsche
Idealismus in sozialgeschichtlicher Beleuchtung’, Moderne
deutsche Sozialgeschichte, ed. Hans-Ulrich Wehler (Cologne:
Kiepenheur und Witsch, 1966), 85–108.
8. Rudolf Stadelmann, ‘Deutschland und die westeuropäischen
Revolutionen’, in Deutschland und Westeuropa (Laupheim: Schloss
Laupheim-Württemberg, U. Steiner, 1948), 11–33 (28).
9. Edmund Burke/Friedrich Gentz, Über die Französische Revolution.
Betrachtungen und Abhandlungen, ed. and with an appendix by
Hermann Klenner (Berlin: Akademie Verlag, 1990); Friedrich
Meinecke, Weltbürgertum und Nationalstaat. Studien zur Genesis des
deutschen Nationalstaates (1907), in Sämtliche Werke (Munich:
Oldenbourg, 1957–68), v. 113 ff., tr. as Cosmopolitanism and the
National State (Princeton: Princeton University Press, 1970); Fritz
Valjavec, Die Entstehung der politischen Strömungen in Deutschland
965
1770–1815 (Munich: Oldenbourg, 1951), esp. 244 ff.; Frieda
Braune, Edmund Burke in Deutschland. Ein Beitrag zur Geschichte
des historischpolitischen Denkens (Heidelberg: C. Winter, 1917);
Golo Mann, Friedrich von Gentz. Geschichte eines europäischen
Staatsmannes (Zurich: Europa Verlag, 1947), 36 ff.
10. Christoph Martin Wieland, ‘Patriotischer Beitrag zu
Deutschlands höchstem Flor, veranlasst durch einen im Jahr
1780 gedruckten Beitrag dieses Namens’, in Werke (note 3), xxxi.
349–68 (364, 365, 367); id., ‘Betrachtungen über die
gegenwärtige Lage des Vaterlandes’, 208–59 (237–8); Vierhaus,
‘Montesquieu’ (note 2), 9 ff.
11. Heinrich August Winkler, ‘Der Nationalismus und seine
Funktionen’, in Nationalismus, ed. id., (Königstein: Verlagsgruppe
Athenäum, Hain, Scriptor, Hanstein, 1978), 5–46; Herfried
Münkler, ‘Nation als politische Idee im frühneuzeitlichen
Europa’, in Nation und Literatur im Europa der Frühen Neuzeit, ed.
Klaus Garber (Tübingen: M. Niemeyer, 1989), 56–86;
Nationalismus in vorindustrieller Zeit, ed. Otto Dann (Munich:
Oldenbourg, 1986).
12. Thomas Abbt, Vom Tode für das Vaterland, in Vermischte Werke, i
(repr. Hildesheim: G. Olms, 1978), 51, 62, 70, 72. In the last
sentence I (H.A.W.) have corrected the word Stimme (‘voice’),
which is obviously an error, to Summe (‘sum’).
13. Karl Otmar Freiherr v. Aretin, Heiliges Römisches Reich 1776–
1806. Reichsverfassung und Staatssouveränität, 2 vols. (Wiesbaden:
F. Steiner, 1967), i. 318 ff.; John S. Gagliardo, Reich and Nation:
The Holy Roman Empire as Idea and Reality, 1763–1806
966
(Bloomington: Indiana University Press, 1980), 165 ff.; Horst
Möller, Fürstenstaat oder Bürgernation. Deutschland 1763–1815
(Berlin: Siedler, 1989), 532 ff.; Georg Schmidt, Geschichte des
Alten Reiches (Munich: C. H. Beck, 1999).
14. Friedrich Schiller, ‘Deutsche Grösse’, in Schillers Werke.
Nationalausgabe, ii, pt I (Weimar: Hermann Böhlaus Nachfolger,
1983), 431–7; on the question of date see ibid., ii, pt II B
(Weimar: Hermann Böhlaus Nachfolger, 1993), 257–9; Meinecke,
Weltbürgertum (note 9), 54 ff.; Jörg Echternkamp, Der Aufstieg des
deutschen Nationalismus (1770–1840) (Frankfurt: Campus, 1998),
163 ff.
15. Georg Wilhelm Friedrich Hegel, Die Verfassung Deutschlands
(1802), in Politische Schriften (Frankfurt: Suhrkamp, 1966), 23–
138 (23, 62, 106, 122, 132, 136); Hermann Heller, Hegel und der
nationale Machtstaatsgedanke in Deutschland (Berlin: B. G.
Teubner, 1921), 12 ff.
16. Arnold Berney, ‘Reichstradition und Nationalstaatsgedanke
(1789–1815)’, HZ 140 (1929), 57–86; Gerhard Schuck,
Rheinbundpatriotismus und politische Öffentlichkeit zwischen
Aufklärung und Frühliberalismus. Kontinuitätsdenken und
Diskontinuitätserfahrung in den Staatsrechts- und
Verfassungsdebatten der Rheinbundpublizistik (Stuttgart: F. Steiner,
1994); Georg Schmidt, ‘Der Rheinbund und die deutsche
Nationalbewegung’, in Die Entstehung der Nationalbewegung in
Europa 1750–1849, ed. Heiner Timmermann (Berlin: Duncker &
Humblot, 1993), 29–44.
17. Elisabeth Fehrenbach, Traditionelle Gesellschaft und revolutionäres
967
Recht. Die Einführung des Code Napoléon in den Rheinbundstaaten
(Göttingen: Vandenhoeck & Ruprecht, 1974) (quote: 12);
Eberhard Weis, ‘Ergebnisse eines Vergleichs der
grundherrschaftlichen Strukturen Deutschlands und Frankreichs
vom Hochmittelalter bis zum Ausgang des 18. Jahrhunderts’, in
Deutschland und Frankreich um 1800. Aufklärung—Revolution
—Reform (Munich: C. H. Beck, 1990), 67–81; Ernst Rudolf
Huber, Deutsche Verfassungsgeschichte seit 1789, i. Reform und
Restauration 1789 bis 1830, 3rd edn (Stuttgart: Kohlhammer,
1990), 314 ff.
18. Helmut Berding, Napoleonische Herrschafts- und
Gesellschaftspolitik im Königreich Westfalen 1807–1813 (Göttingen:
Vandenhoeck & Ruprecht, 1973), esp. 15; Napoleonische
Herrschaft und Modernisierung, ed. id., GG 6/4 (1989).
19. Lorenz von Stein, Geschichte der sozialen Bewegung in Frankreich,
3 vols., i. Der Begriff der Gesellschaft und die soziale Geschichte der
Französischen Revolution bis zum Jahre 1850 (1850; repr.
Darmstadt: Wissenschaftliche Buchgesellschaft, 1959), 426 (427,
429–30; emphases in original); Heinrich August Winkler,
‘Gesellschaftsform und Aussenpolitik. Eine Theorie Lorenz von
Steins in zeitgeschichtlicher Perspektive’, in Liberalismus und
Antiliberalismus. Studien zur politischen Sozialgeschichte des 19. und
20. Jahrhunderts (Göttingen: Vandenhoeck & Ruprecht, 1979),
235–51.
20. Thomas Nipperdey, Deutsche Geschichte 1800–1866. Bürgerwelt
und starker Staat (Munich: C. H. Beck, 1983), 25; Braubach, ‘Von
der Französischen Revolution’ (note 6), 45.
968
21. Friedrich Gentz, Fragmente aus der neuesten Geschichte (1806), in
Ausgewählte Schriften, iv (Stuttgart: L. F. Rieger, 1837), 33–4
(‘Teutschland’ and ‘teutsch’ in the original); Mann, Gentz (note
9), 167 ff.; ‘Hinrich C. Seeba, Zeitgeist und deutscher Geist. Zur
Nationalisierung der Epochentendenz um 1800’, Deutsche
Vierteljahrschrift für Literaturwissenschaft und Geistesgeschichte, 61
(1987), Sonderheft, 185–215.
22. Georg Wilhelm Friedrich Hegel, Grundlinien der Philosophie des
Rechts oder Naturrecht und Staatswissenschaft im Grundrisse, in
Sämtliche Werke (note 1), vii (Stuttgart: Fromman, 1952), 241 ff.
(§ 257 ff.; quote: 241); Hans-Ulrich Wehler, Deutsche
Gesellschaftsgeschichte, i. Vom Feudalismus des Alten Reiches bis zur
Defensiven Modernisierung der Reformära 1700–1815 (Munich: C.
H. Beck, 1987), 397 ff. (quote: 446); Nipperdey, Geschichte (note
20), 331 ff.; Koselleck, Preussen (note 6), 163 ff.; Schissler,
Agrargesellschaft (note 6), 105 ff.
23. Johann Gottlieb Fichte, Beitrag zur Berichtigung der Urtheile des
Publicums über die Französische Revolution (1793), in Fichtes
Werke, 8 vols. (Berlin: De Gruyter, 1845–6, repr. Berlin, 1971),
vi. 39–256 (esp. 152 ff.); id., Reden an die deutsche Nation (1808),
ibid. 259–499 (quotes in order: 278, 355, 346, 380–1, 384–5,
456, 470–1, 496), tr. as Addresses to the German Nation (New
York: Harper & Row, 1968).
24. Fichte, Reden, 467–8, 386–7, 353–4, 387–8, 385–6; id., Aus dem
Entwurfe zu einer politischen Schrift im Frühlinge 1813, ibid. in
Fichtes Werke, vi. 554, 565 (emphases in original); Meinecke,
Weltbürgertum (note 9), 84 ff.; Eugene Newton Anderson,
969
Nationalism and the Cultural Crisis in Prussia 1806–1815 (New
York: Octagon, 1939), 16–63; Becker, Zeit (note 5), 108 ff.;
Huber, Verfassungsgeschichte (note 17), i. 478–9 (Kalisch).
25. Stein, Geschichte (note 19), i. 481, 428 (emphases in original).
26. Georg Wilhelm Friedrich Hegel, Phänomenologie des Geistes
(1807), in Werke (note 1), ii. 132, tr. as Phenomenology of Spirit
(Oxford: Clarendon Press, 1977).
27. O. C. C. Höpffner [Friedrich Ludwig Jahn], ‘Über die
Beförderung des Patriotismus im Preussischen Reiche’ (1800), in
Friedrich Ludwig Jahns Werke (Hof: G. A. Grau, 1884–7), i. 3–22;
Friedrich Ludwig Jahn, Deutsches Volkstum (1810), ibid. 143–380
(146 ff., 159–63, 203; emphases in original); Rainer Wiegels and
Winfried Woesler, Arminius und die Varusschlacht. Geschichte
—Mythos—Literatur (Paderborn: Schöningh, 1995); Andreas
Dörner, Politischer Mythos und symbolische Politik. Sinnstiftung
durch symbolische Formen am Beispiel des Hermannmythos
(Opladen: Westdeutscher Verlag, 1995).
28. Jahn, Volkstum (note 27), 285, 307 ff., 235–45 (emphases in
original); Hans Kohn, Wege und Irrwege. Vom Geist des deutschen
Bürgertums (Düsseldorf: Droste, 1962), 73 ff., orig. The Mind of
Germany: The Education of a Nation (New York: Scribner, 1960).
29. Ernst Moritz Arndt, Geist der Zeit (1806–17) (Leipzig: Max Hesse,
n.d. [1908]), pt 2 (1809), 85; id., ‘Gebet’, in Arndts Werke
(Berlin: Deutsches Verlagshaus Bong, n.d. [1912]), pt 1, 74; id.,
‘Katechismus für den deutschen Kriegs- und Wehrmann’ (1813),
pt 10, 131–62 (142–3, 147); id., ‘An die Preussen’ (1813), ibid.
163–70 (169); id., ‘Was bedeutet Landsturm und Landwehr?’
970
(1813), ibid. 171–86 (176); id., ‘Über Volkshass und den
Gebrauch einer fremden Sprache’ (1813), in Ernst Moritz Arndt’s
Schriften für und an seine lieben Deutschen. Zum ersten Mal
gesammelt und durch Neues vermehrt (Leipzig: Weidmann, 1845),
353–73 (358, 361, 367–8; all emphases in the original), tr. as
Arndt’s Spirit of the Times (London: Printed by and for W. M.
Thiselton and the author, 1808).
30. Arndt’s Schriften, 372; id., ‘Der Rhein, Teutschlands Strom, aber
nicht Teutschlands Graenze’ (1813), in Werke (note 29), pt 11,
37–82 (42); id., ‘Das preussische Volk und Heer im Jahre 1813’
(1814), ibid. 7–36 (29, 32); id., ‘Bemerkungen über Teutschlands
Lage im November 1814’, id., Blick aus der Zeit auf die Zeit
(Germanien, n.d. [1814]), 1–79; id., ‘Noch eine Vermahnung an
die politischen teutschen Philister’, ibid. 257–81; id., ‘Über
Preussens Rheinische Mark und über Bundesfestungen’ (1815), in
Werke (note 29), pt 11, 143–99 (164, 187, 190); Günther Ott,
Ernst Moritz Arndt. Religion, Christentum und Kirche in der
Entwicklung des deutschen Publizisten und Patrioten (Bonn:
Presseverband der Evangelischen Kirche im Rheinland, 1966),
197 ff.; Alexander Scharff, Der Gedanke der preussischen
Vorherrschaft in den Anfängen der deutschen Einheitsbewegung
(Bonn: K. Schroeder, 1929), 39–40 (on Arndt’s position in 1814);
Klaus Vondung, Die Apokalypse in Deutschland (Munich:
Deutscher Taschenbuch Verlag, 1988), 152 ff., tr. as The
Apocalypse in Germany (Columbia, MO: University of Missouri
Press, 2000).
31. Arndt, Geist der Zeit (note 29), pt 2 (1809), 134–5; pt 3 (1813),
971
156; id., ‘Über künftige ständische Verfassungen in Teutschland’
(1814), in Werke (note 29), pt 11, 83–130 (106, 121); Anderson,
Nationalism (note 24), 64–103; Michael Jeismann, Das Vaterland
der Feinde. Studien zum nationalen Feindbegriff und
Selbstverständnis in Deutschland und Frankreich 1792–1918
(Stuttgart: Klett-Cotta, 1992), 27 ff.
32. Fichte, Reden (note 23), 464–5; Arndt, ‘Preussens Rheinische
Mark’ (note 30), 191; Otto Kallscheuer and Claus Leggewie,
‘Deutsche Kulturnation versus französische Staatsnation? Eine
ideengeschichtliche Stichprobe’, Nationales Bewusstsein und
kollektive Identität. Studien zur Entwicklung des kollektiven
Bewusstseins in der Neuzeit, ii ed. Helmut Berding (Frankfurt:
Suhrkamp, 1994), 112–62; Bernhard Giesen, Die Intellektuellen
und die Nation. Eine deutsche Achsenzeit (Frankfurt: Suhrkamp,
1993), 27 ff.; Conrad Wiedemann, ‘Zwischen Nationalgeist und
Kosmopolitismus. Über die Schwierigkeiten der deutschen
Klassiker, einen Nationalhelden zu finden’, in Patriotismus, ed.
Günter Birtsch, Aufklärung, 4/2 (1991), 75–101.
33. Ernst Moritz Arndt, ‘Bemerkungen’ (note 30), 59 ff.; id., ‘Die
Schweitzer, Holländer und Elsässer’, in Blick (note 30), 80–112;
id., ‘Noch etwas über die Juden’, ibid. 180–201 (181, 199, 188);
Fichte, Beitrag (note 23), 150–1; Jacob Katz, Vom Vorurteil zur
Vernichtung. Der Antisemitismus 1700–1933 (Munich: C. H. Beck
1989), 61, orig. From Prejudice to Destruction: Anti-Semitism,
1700–1933 (Cambridge, MA: Harvard University Press, 1980),
Hans Rothfels, ‘Grundsätzliches zum Problem der Nationalität’,
in Zeitgeschichtliche Betrachtungen. Vorträge und Aufsätze
972
(Göttingen: Vandenhoeck & Ruprecht, 1959), 89–111.
34. Otto W. Johnston, Der deutsche Nationalmythos. Ursprung und
politisches Programm (Stuttgart: Metzler, 1990), orig. The Myth of
a Nation: Literature and Politics in Prussia under Napoleon (New
York: Camden House, 1989); Eric J. Hobsbawm, Nationen und
Nationalismus. Mythos und Realität seit 1780 (Frankfurt: Campus,
1991), orig. Nations and Nationalism since 1780: Programme,
Myth, Reality (Cambridge: Cambridge University Press, 1990);
The Invention of Tradition, ed. id. and Terence Ranger
(Cambridge: Cambridge University Press, 1983); Benedict
Anderson, Die Erfindung der Nation (Frankfurt: Campus, 1988),
orig. Imagined Communities: Reflections on the Origin and Spread of
Nationalism (London: Verso, 1983); Anthony D. Smith, ‘The
Nation: Invented, Imagined, Reconstructed?’, Millennium: Journal
of International Studies, 20 (1991), 353–68.
35. Meinecke, Weltbürgertum (note 9), 58 ff., 113 ff.; Kohn, Wege
(note 28), 53 ff.; Carl Schmitt, Politische Romantik (1919), 3rd
edn (Berlin: Duncker & Humblot, 1968), tr. as Political
Romanticism (Cambridge, MA: MIT Press, 1986); Hermann
Kurzke, Romantik und Konservatismus. Das ‘politische’ Werk
Friedrich von Hardenbergs (Novalis) (Munich: W. Fink, 1983).
36. Ernst Moritz Arndt, Was ist des Deutschen Vaterland? (1813), in
Werke (note 29), pt 1, 126–7; id., Geist der Zeit (note 29), pt III,
177; Heinz Angermeier, ‘Deutschland zwischen Reichstradition
und Nationalstaat. Verfassungspolitische Konzeptionen und
nationales Denken zwischen 1801 und 1815’, in Das alte Reich in
der deutschen Geschichte. Studien über Kontinuitäten und Zäsuren
973
(Munich: Oldenbourg, 1991), 449–521; Meinecke, Weltbürgertum
(note 9), 142 ff.; Scharff, Gedanke (note 30), 8 ff.; the article
‘Ideologie’, Geschichtliche Grundbegriffe (note 7), iii (Stuttgart:
Klett, 1982), 131–69; ‘Humboldts “Denkschrift über die deutsche
Verfassung” vom Dezember 1813’, in Wilhelm von Humboldt. Eine
Auswahl aus seinen Schriften, ed. Siegfried A. Kaehler (Berlin: R.
Hobbing, 1922), 88–103 (quote: 92–3).
37. Leopold Zscharnack, ‘Die Pflege des religiösen Patriotismus
durch die evangelische Geistlichkeit 1806–1815’, in Harnack-
Ehrung. Beiträge zur Kirchengeschichte. Festschrift für Adolf von
Harnack (Leipzig: J. C. Hinrichs, 1921), 394–423 (on the
appearance of the concept of ‘religious patriotism’, see 400);
Gerhard Graf, Gottesbild und Politik. Eine Studie zur Frömmigkeit in
Preussen während der Befreiungskriege 1813–1815 (Göttingen:
Vandenhoeck & Ruprecht, 1993); Artie J. Hoover, The Gospel of
Nationalism: German Patriotic Preaching from Napoleon to Versailles
(Stuttgart: F. Steiner, 1986), 6 ff.; Georg Schmidt, ‘Von der
Nationaleinheit zum Nationalstaat. Der gedankliche
Kontinuitätsbruch in Deutschland zu Beginn des 19.
Jahrhunderts’, Die evangelische Diaspora. Jahrbuch des Gustav-
Adolf-Werks, 63 (1994), 59–75; Jerry Dawson, Friedrich
Schleiermacher: The Evolution of a Nationalist (Austin: University
of Texas Press, 1966); Otto Dann, ‘Schleiermacher und die
nationale Bewegung’, in Internationaler Schleiermacher-Kongress
Berlin 1984, 2 vols. (Berlin: De Gruyter, 1985), ii. 1107–20;
Rudolf von Thadden, ‘Schleiermacher und Preussen’, ibid. 1099–
1106; Kurt Nowak, Schleiermacher und die Frühromantik. Eine
literaturgeschichtliche Schule zum romantischen Religionsverständnis
974
und Menschenbild am Ende des 18. Jahrhunderts in Deutschland
(Weimar: Hermann Böhlaus Nachfolger, 1986), 92 ff.; Karin
Hagemann, ‘Nation, Krieg und Geschlechterordnung. Zum
kulturellen und politischen Diskurs in der Zeit der
antinapoleonischen Erhebung Preussens 1806–1815’, GG 22
(1996), 562–91; Hasko A. Zimmer, Auf dem Altar des Vaterlands.
Religion und Patriotismus in der deutschen Kriegslyrik des 19.
Jahrhunderts (Frankfurt: Thesen Verlag, 1971); Reinhard
Wittram, ‘Kirche und Nationalismus in der Geschichte des
deutschen Protestantismus im 19. Jahrhundert’, in Das Nationale
als europäisches Problem. Beiträge zur Geschichte des
Nationalitätsprinzips vornehmlich im 19. Jahrhundert (Göttingen:
Vandenhoeck & Ruprecht, 1954), 109–48; Lothar Gall, ‘Die
Germania als Symbol nationaler Einheit im 19. und 20.
Jahrhundert’, in Nachrichten der Akademie der Wissenschaften in
Göttingen. I. Philologisch-Historische Klasse, Jg. 1993, 2 (Göttingen:
Vandenhoeck & Ruprecht, 1993); Gerhard Brunn, ‘Germania und
die Entstehung des deutschen Nationalstaates. Zum
Zusammenhang von Symbolen u. Wir-Gefühl’, in Politik der
Symbole, Symbole der Politik, ed. Rüdiger Voigt (Opladen: Leske
+ Budrich, 1989), 101–22.
38. Anthony D. Smith, Theories of Nationalism, 3rd edn (London:
Duckworth, 1983); Hagen Schulze, Staat und Nation in der
europäischen Geschichte (Munich: C. H. Beck, 1994), 150 ff. tr. as
States, Nations, and Nationalism: From the Middle Ages to the
Present (Oxford: Blackwell, 1996); Miroslav Hroch, ‘Das
Erwachen kleiner Nationen als Problem der komparativen
Forschung’, in Nationalismus (note 11), ed. Winkler, 155–72;
975
Jürgen Wilke, ‘Der nationale Aufbruch der Befreiungskriege als
Kommunikationsereignis’ Volk—Nation—Vaterland, ed. Ulrich
Herrmann (Hamburg: F. Meiner, 1996), 353–66.
39. Dieter Grimm, Deutsche Verfassungsgeschichte 1776–1866
(Frankfurt: Suhrkamp, 1988), 68 ff., 113 ff.; Karsten Ruppert,
Bürgertum und staatliche Macht in Deutschland zwischen
Französischer und deutscher Revolution (Berlin: Duncker &
Humblot, 1998); Huber, Verfassungsgeschichte (note 17), i. 314 ff.,
475 ff., 583 ff.
40. Huber, Verfassungsgeschichte, 704 ff.; Wehler,
Gesellschaftsgeschichte (note 22), ii. Von der Reformära bis zur
industriellen und politischen ‘Deutschen Doppelrevolution’ 1815–
1848/49 (Munich: C. H. Beck, 1987), 333 ff.; Franz Schnabel,
Deutsche Geschichte im neunzehnten Jahrhundert, ii. Monarchie und
Volkssouveränität, 2nd edn (Freiburg: Herder, 1949), 218 ff.; 175
Jahre Wartburgfest. 18. Oktober 1817–18. Oktober 1992. Studien
zur politischen Bedeutung und zum Zeithintergrund der
Wartburgfeier, ed. Karl Malettke (Heidelberg: Winter, 1992);
Wolfgang Hardtwig, ‘Studentische Mentalität—Politische
Jugendbewegung—Nationalismus. Die Anfänge der deutschen
Burschenschaft’, in Nationalismus und Bürgerkultur in Deutschland
1500–1914 (Göttingen: Vandenhoeck & Ruprecht, 1994), 108–
48; Peter Brandt, ‘Das studentische Wartburgfest vom 18./19.
Oktober 1817’, in Öffentliche Festkultur. Politische Feste in
Deutschland von der Aufklärung bis zum Ersten Weltkrieg, ed. Dieter
Düding et al. (Hamburg: Rowohlt, 1988), 89–112. Quote from
Görres in ‘Wiedererneuerung des Vertrags von Chaumont’,
976
Rheinischer Merkur, 225, 19 Apr. 1815. Facsimile edition in
Joseph von Görres, Gesammelte Schriften, ix, 11 (Cologne: Gilde-
Verlag, 1928).
41. Dieter Düding, Organisierter gesellschaftlicher Nationalismus in
Deutschland (1808–1847). Bedeutung und Funktion der Turner- und
Sängervereine für die deutsche Nationalbewegung (Munich:
Oldenbourg, 1984); Karin Luys, Die Anfänge der deutschen
Nationalbewegung von 1815 bis 1819 (Münster: Nodus
Publikationen, 1992); Wolfgang v. Groote, Die Entstehung des
Nationalbewusstseins in Nordwestdeutschland 1790–1830
(Göttingen: Musterschmidt-Verlag, 1955); Otto Dann,
‘Nationalismus und sozialer Wandel in Deutschland 1806–1850’,
in Nationalismus und sozialer Wandel, ed. id. (Hamburg: Hoffmann
und Campe, 1978), 77–128; Wehler, Gesellschaftsgeschichte, ii
(note 40), 394 ff.; Koselleck, Preussen (note 6), 284 ff.; Friedrich
Meinecke, 1848. Eine Säkularbetrachtung (Berlin: L. Blanvalet,
1948), 9; Theodor Schieder, ‘Partikularismus und
Nationalbewusstsein im Denken des deutschen Vormärz’, Staat
und Gesellschaft im deutschen Vormärz 1815–1848, ed. Werner
Conze (Stuttgart: Klett, 1962), 9–38; Werner Conze, ‘Das
Spannungsfeld von Staat und Gesellschaft im Vormärz’, ibid.
207–69; Volker Sellin, ‘Nationalbewusstsein und Partikularismus
in Deutschland im 19. Jahrhundert’, in Kultur und Gedächtnis, ed.
Jan Assmann and Tonio Hölscher (Frankfurt: Suhrkamp, 1988),
241–64; Dieter Langewiesche, ‘Reich, Nation und Staat in der
jüngeren deutschen Geschichte’, HZ 254 (1992), 341–81.
42. Lothar Gall, Benjamin Constant. Seine politische Ideenwelt und der
977
deutsche Vormärz (Wiesbaden: F. Steiner, 1963); Liberalismus im
19. Jahrhundert, ed. Dieter Langewiesche (Göttingen:
Vandenhoeck & Ruprecht, 1988); Wolfram Siemann, Vom
Staatenbund zum Nationalstaat. Deutschland 1806–1871 (Munich:
C. H. Beck, 1995), esp. 29 ff.; Christoph Hauser, Anfänge
bürgerlicher Organisation. Philhellenismus und Frühliberalismus in
Südwestdeutschland (Göttingen: Vandenhoeck & Ruprecht, 1990).
43. Katz, Vorurteil (note 33), 95 ff.; Helmut Berding, Moderner
Antisemitismus in Deutschland (Frankfurt: Suhrkamp, 1988), 42 ff.
(Marwitz: 46); Reinhard Rürup, Emanzipation und Antisemitismus.
Studien zur ‘Judenfrage’ der bürgerlichen Gesellschaft (Göttingen:
Vandenhoeck & Ruprecht, 1975), esp. 37 ff. (Rotteck: 61);
Wolfgang Altgeld, Katholizismus, Protestantismus, Judentum. Über
religiös begründete Gegensätze und nationalreligiöse Ideen in der
Geschichte des deutschen Nationalismus (Mainz: Matthias-
Grünewald-Verlag, 1992), 47 ff.; Eleonore Sterling, Judenhass.
Die Anfänge des politischen Antisemitismus in Deutschland (1815–
1850), 2nd edn (Frankfurt: Europäische Verlagsanstalt, 1969), 77
ff.; Rainer Erb and Werner Bergmann, Die Nachtseite der
Judenemanzipation. Der Widerstand gegen die Integration der Juden
in Deutschland 1780–1860 (Berlin: Metropol, 1989), 15 ff., 217
ff.; Stefan Rohrbacher, Gewalt im Biedermeier. Antijüdische
Ausschreitungen in Vormärz und Revolution (1815–1848/49)
(Frankfurt: Campus, 1993), 94 ff.; Rüdiger von Treskow,
Erlauchter Vertheidiger der Menschenrechte! Die Korrespondenz Karl
von Rottecks, 2 vols. (Freiburg: Ploetz, 1990), i. 160 ff.
44. Ludwig Börne, ‘Der ewige Jude’, in Ludwig Börnes gesammelte
978
Schriften. Vollständige Ausgabe in sechs Bänden (Leipzig: M. Hesse,
n.d.), iii. 139–71 (141, 171).
45. Katz, Vorurteil (note 33), 107 ff.; Ernst Schulin, ‘Weltbürgertum
und deutscher Volksgeist. Die romantische Nationalisierung im
frühen neunzehnten Jahrhundert’, in Deutschland in Europa. Ein
historischer Rückblick, ed. Bernd Martin (Munich: Deutscher
Taschenbuch Verlag, 1992), 105–21; Thomas Nipperdey, ‘Auf der
Suche nach Identität. Romantischer Nationalismus’, in
Nachdenken über die deutsche Geschichte (Munich: C. H. Beck,
1986), 110–25.
979
Polenfreundschaft, ed. Peter Ehlen (Munich: J. Berchmanns,
1982).
3. Hans-Ulrich Wehler, Deutsche Gesellschaftsgeschichte, ii. Von der
Reformära bis zur industriellen und politischen ‘Deutschen
Doppelrevolution’ 1815–1845/49 (Munich: C. H. Beck, 1987), 345
ff.; Wolfram Siemann, Vom Staatenbund zum Nationalstaat 1806–
1871 (Munich: C. H. Beck, 1995), 343 ff.; Elisabeth Fehrenbach,
Verfassungsstaat und Nationsbildung 1815–1871 (Munich:
Oldenbourg, 1992), 9 ff.; Dieter Langewiesche, Europa zwischen
Restauration und Revolution 1815–1849 (Munich: Oldenbourg,
1985), 65 ff.; Hartwig Brandt, ‘Die Julirevolution (1830) und die
Rezeption der “principes de 1789” in Deutschland’, in Revolution
und Gegenrevolution 1789–1830. Zur geistigen Auseinandersetzung
in Frankreich und Deutschland, ed. Roger Dufraisse (Munich:
Oldenbourg, 1991), 225–35; Hans-Gerhard Husung, Protest und
Repression im Vormärz. Norddeutschland zwischen Restauration und
Revolution (Göttingen: Vandenhoeck & Ruprecht, 1983), 43 ff.;
Die Französische Julirevolution von 1830 und Europa, ed. Manfred
Kossok and Werner Loch (Berlin: Akademie-Verlag, 1985), esp.
177 ff.; Kurt Holzapfel, ‘Der Einfluss der Julirevolution von
1830/31 auf Deutschland’, in Demokratische und soziale
Protestbewegungen in Mitteleuropa 1815–1848/49, ed. Helmut
Reinalter (Frankfurt: Suhrkamp, 1986), 105–40; Ernst Wolfgang
Becker, Zeit der Revolution!—Revolution der Zeit? Zeiterfahrungen
in Deutschland in der Ära der Revolutionen 1789–1848/49
(Göttingen: Vandenhoeck & Ruprecht, 1999), 147 ff.
4. Lothar Gall, ‘Gründung und politische Entwicklung des
980
Grossherzogtums bis 1848’, in Badische Geschichte. Vom
Grossherzogtum bis zur Gegenwart, ed. Josef Becker et al.
(Stuttgart: Theiss, 1979), 11–36 (Welcker motion: 32); Manfred
Meyer, Freiheit und Macht. Studien zum Nationalismus
süddeutscher, insbesondere badischer Liberaler 1830–1848
(Frankfurt: P. Land, 1994), 102 ff., 170 ff.; Reiner Schöttle,
Politische Theorien des süddeutschen Liberalismus im Vormärz.
Studien zu Rotteck, Welcker, Pfizer, Murhard (Baden-Baden:
Nomos, 1994), 147 ff., 183 ff.; Friedrich Meinecke, Weltbürgertum
und Nationalstaat (1907), in Werke, v (Munich: Oldenbourg,
1962), 281 ff., tr. as Cosmopolitanism and the National State
(Princeton: Princeton University Press, 1970).
5. Ernst Rudolf Huber, Deutsche Verfassungsgeschichte seit 1789, ii.
Der Kampf um Einheit und Freiheit 1830 bis 1850, 3rd edn
(Stuttgart: W. Kohlhammer, 1988), 133 ff.; Heinrich von
Treitschke, Deutsche Geschichte im Neunzehnten Jahrhundert, pt 4,
Bis zum Tode König Friedrich Wilhelms IV., 5th edn (Leipzig: S.
Hirzel, 1907), 261 ff., tr. as History of Germany in the Nineteenth
Century (London, Jarrold & Sons, 1915–19), vii; Hans Kohn, Wege
und Irrwege. Vom Geist des deutschen Bürgertums (Düsseldorf:
Droste, 1962), 125–6, orig. The Mind of Germany: The Education
of a Nation (New York: Scribner, 1960); Cornelia Foerster, Der
Press- und Vaterlandsverein von 1832/33. Sozialstruktur und
Organisationsformen der bürgerlichen Bewegung in der Zeit des
Hambacher Festes (Trier: Verlag Trierer Historischer Forschungen,
1982); Wolfgang Schieder, ‘Der rheinpfälzische Liberalismus von
1832 als politische Protestbewegung’, in Vom Staat des Ancien
Régime zum modernen Parteienstaat. Festschrift für Theodor
981
Schieder, ed. Helmut Berding et al. (Munich: Oldenbourg, 1978),
169–95; Liberalismus in der Gesellschaft des deutschen Vormärz, ed.
id., GG, Sonderheft 9 (Göttingen: Vandenhoeck & Ruprecht,
1983); Hans Fenske, ‘Politischer und sozialer Protest in
Süddeutschland nach 1830’, in Protestbewegungen (note 3), ed.
Reinalter, 143–201; Jörg Echternkamp, Der Aufstieg des deutschen
Nationalismus (1770–1840) (Frankfurt: Campus, 1998), 420 ff.
Quotes von Siebenpfeiffer and Wirth in J. G. A. Wirth, Das
Nationalfest der Deutschen zu Hambach, 1 (Neustadt: In
Commission bei Philipp Christmann, 1832, repr. Vaduz: Topos,
1977), 38–9, 46 (emphases in original). Quote from Rotteck in:
Meyer, Freiheit (note 4), 149.
6. Huber, Verfassungsgeschichte (note 5), ii. 92 ff., 125 ff., 151 ff.;
Wehler, Gesellschaftsgeschichte (note 3), ii. 366 ff. (quote from
Metternich: 367); Treitschke, Geschichte (note 5), pt 4, 657
(quote: 666).
7. Huber, Verfassungsgeschichte (note 5), ii. 177 ff.; Lothar Gall,
‘Liberalismus und “bürgerliche Gesellschaft”. Zu Charakter und
Entwicklung der liberalen Bewegung in Deutschland’, in
Liberalismus, ed. tr., 3rd edn (Königstein: Verlagsgruppe
Athenäum, Hain, Scriptor, Hanstein, 1985), 162–86; Wolfram
Fischer, ‘Das Verhältnis von Staat und Wirtschaft in Deutschland
am Beginn der Industrialisierung’, in Industrielle Revolution.
Wirtschaftliche Aspekte, ed. Rudolf Braun et al. (Cologne:
Kiepenheuer & Witsch, 1972), 287–304; Staat und Gesellschaft im
deutschen Vormärz 1815–1848, ed. Werner Conze (Stuttgart:
Klett, 1962); James J. Sheehan, Der Ausklang des alten Reiches.
982
Deutschland seit dem Ende des Siebenjährigen Krieges bis zur
gescheiterten Revolution 1763 bis 1850 (Berlin: Propyläen, 1994),
546 ff.; id., Der deutsche Liberalismus. Von den Anfängen im 18.
Jahrhundert bis zum Ersten Weltkrieg 1770–1914 (Munich: C. H.
Beck, 1983), 26 ff., orig. German Liberalism in the Nineteenth
Century (Chicago: University of Chicago Press, 1978).
8. Wolfram Fischer, ‘Der Deutsche Zollverein. Fallstudie einer
Zollunion’, in Wirtschaft und Gesellschaft im Zeitalter der
Industrialisierung, ed. id. (Göttingen: Vandenhoeck & Ruprecht,
1972), 110–28; Hans-Werner Hahn, Geschichte des Deutschen
Zollvereins (Göttingen: Vandenhoeck & Ruprecht, 1984); Thomas
Nipperdey, Deutsche Geschichte 1800–1866. Bürgerwelt und starker
Staat (Munich: C. H. Beck, 1983), 358 ff.; Wehler,
Gesellschaftsgeschichte (note 3), ii. 125 ff.
9. Treitschke, Geschichte (note 5), pt 5, Bis zur Märzrevolution, 4th
edn (Leipzig 1899), 3 ff. (quote: 57–8), tr. as History of Germany
(note 5), vii; Nipperdey, Geschichte (note 8), 396 ff.; David E.
Barclay, Anarchie und guter Wille. Friedrich Wilhelm IV. und die
preussische Monarchie (Berlin: Siedler, 1995), 85 ff.; orig.
Frederick William IV and the Prussian Monarchy, 1840–1861
(Oxford: Oxford University Press, 1995); Walter Bussmann,
Zwischen Preussen und Deutschland. Friedrich Wilhelm IV. Eine
Biographie (Berlin: Siedler, 1990), 101 ff.; Frank-Lothar Kroll,
Friedrich Wilhelm IV. und das Staatsdenken der deutschen Romantik
(Berlin: Colloquium-Verlag, 1990); Dirk Blasius, ‘Friedrich
Wilhelm IV. Persönlichkeit und Amt’, HZ 263 (1996), 589–607.
10. Irmline Veit-Brause, ‘Die deutsch-französische Krise von 1840.
983
Studien zur deutschen Einheitsbewegung’, Ph.D. thesis, Cologne,
1967 (Rheinlieder: 125 ff.); Ute Schneider, Politische Festkultur im
19. Jahrhundert. Die Rheinprovinz von der französischen Zeit bis
zum Ende des Ersten Weltkrieges (1806–1918) (Essen: Klartext,
1995), 79 ff.; Dietmar Klenke, Der singende ‘deutsche Mann’.
Gesangvereine und deutsches Nationalbewusstsein von Napoleon bis
Hitler (Munich: Waxmann, 1998), 82 ff.; Meyer, Freiheit (note 4),
205 ff.; Echternkamp, Aufstieg (note 5), 464 ff.; Deutscher
Liberalismus im Vormärz. Heinrich von Gagern, Briefe und Reden
1815–1848, ed. Paul Wentzcke and Wolfgang Klötzer (Göttingen:
Musterschmidt, 1959), 99–101 (letter to Hans Christoph von
Gagern, 13 July 1832), 261–3 (letter to Friedrich von Gagern, 4
Jan. 1843; quote: 263).
11. Karl Biedermann, ‘Die Fortschritte des nationalen Prinzips in
Deutschland (1842)’, in Vormärz und Revolution 1840–1849, ed.
Hans Fenske, 2nd edn (Darmstadt: Wissenschaftliche
Buchgesellschaft, 1991), 54–64 (58, 60); Veit-Brause, Krise (note
10), 219 (Pfizer), 244 (Steinacker), 251–2; Meyer, Freiheit (note
4), 193 ff.; Hans Fenske, ‘Ungeduldige Zuschauer. Die Deutschen
und die europäische Expansion 1815–1880’, in Imperialistische
Kontinuität und nationale Ungeduld im 19. Jahrhundert, ed.
Wolfgang Reinhard (Frankfurt: Fischer Taschenbuch Verlag,
1991), 87–123. For a summary of Paul Pfizer’s national-political
ideas from the early 1840s see id., Das Vaterland. Aus der Schrift:
Gedanken über Recht, Staat und Kirche (1842) (Stuttgart:
Hallberger’sche Verlagshandlung, 1845) (on the connection
between Austria and Germany see esp. 197). Quote from
Friedrich List in id., ‘Die politisch-ökonomische Nationaleinheit
984
der Deutschen’, in Schriften, Reden, Briefe, vii (Berlin: R. Hobbing,
1931), 441–502 (444). The quasi-biblical passage is taken from
Matthew 16: 26 (‘For what is a man profited, if he shall gain the
whole world, and lose his own soul?’)
12. Friedrich Julius Stahl, Der christliche Staat. Vortrag über
Kirchenzucht (1847), 2nd edn (Berlin: L Oehmigke, 1858);
Bussmann, Zwischen Preussen (note 9), 172 ff.; Barclay, Anarchie
(note 9), 85 ff.; Wehler, Gesellschaftsgeschichte (note 3), ii. 407,
571; Huber Verfassungsgeschichte (note 5), ii. 185 ff.; Heinrich
Lutz, Zwischen Habsburg und Preussen. Deutschland 1815–1866
(Berlin: Siedler, 1985), 211 ff.; Meyer, Freiheit (note 4), 197;
Richard Hinton Thomas, Liberalism, Nationalism, and the German
Intellectuals 1822–1847 (Cambridge, MA: W. Heffner, 1951), 81
ff., 114–15; Georg Bollenbeck, Bildung und Kultur. Glanz und
Elend eines deutschen Deutungsmusters (Frankfurt: Insel, 1994),
186 ff.; Thomas Nipperdey, ‘Der Kölner Dom als
Nationaldenkmal’, in Nachdenken über die deutsche Geschichte, ed.
id. (Munich: C. H. Beck, 1986), 156–71; id., ‘Nationalidee und
Nationaldenkmal im 19. Jahrhundert’, Gesellschaft, Kultur,
Theorie, ed. id. (Göttingen: Vandenhoeck & Ruprecht, 1976),
133–73.
13. Gustav Mayer, ‘Die Anfänge des politischen Radikalismus im
vormärzlichen Preussen’ (1913), in Radikalismus, Sozialismus und
bürgerliche Demokratie (Frankfurt: Suhrkamp, 1969), 7–107; Hans
Rosenberg, ‘Theologischer Rationalismus und vormärzlicher
Vulgärliberalismus’ (1930), in Politische Denkströmungen im
deutschen Vormärz (Göttingen: Vandenhoeck & Ruprecht, 1972),
985
18–50; Jörg Echternkamp, ‘Religiosität und Nationskonzeption.
Zum Verhältnis von Theologischem Rationalismus und
Liberalnationalismus im deutschen Vormärz’, Jahrbuch zur
Liberalismusforschung, 6 (1994), 137–51; Jörn Brederlow,
‘Lichtfreunde’ und ‘Freie Gemeinden’. Religiöser Protest und
Freiheitsbewegung im Vormärz und in der Revolution von 1848/49
(Munich: Oldenbourg, 1976); Friedrich Wilhelm Graf, Die
Politisierung des religiösen Bewusstseins. Die bürgerlichen
Religionsparteien im deutschen Vormärz: Das Beispiel des
Deutschkatholizismus (Stuttgart: Frommann-Holzboog, 1978);
Stephan Walter, Demokratisches Denken zwischen Hegel und Marx.
Die politische Philosophie Arnold Ruges (Düsseldorf: Droste, 1995),
101 ff.; Karl Löwith, Von Hegel zu Nietzsche. Der revolutionäre
Bruch im Denken des neunzehnten Jahrhunderts. Marx und
Kierkegaard, 3rd edn (Stuttgart: W. Kohlhammer, 1953), 78 ff., tr.
as From Hegel to Nietzsche: The Revolution in Nineteenth-Century
Thought (New York: Holt, Rinehart and Winston, 1964);
Wolfgang Schieder, ‘Kirche und Revolution. Sozialgeschichtliche
Aspekte der Trierer Wallfahrt von 1844’, Archiv für
Sozialgeschichte, 14 (1974), 419–54; Kurt Nowak, Geschichte des
Christentums in Deutschland. Religion, Politik und Gesellschaft vom
Ende der Aufklärung bis zur Mitte des 20. Jahrhunderts (Munich: C.
H. Beck, 1995), 64 ff.; Manfred Botzenhart, Reform, Restauration,
Krise. Deutschland 1789–1847 (Frankfurt: Suhrkamp, 1985), 120
ff.; Wehler, Gesellschaftsgeschichte (note 3), ii. 413 ff.
14. Johann Jacoby, Vier Fragen, beantwortet von einem Ostpreussen
(1841; Munich, 1910); Arnold Ruge, Preussen und die Reaction.
Zur Geschichte unserer Zeit (Leipzig: Otto Wigand, 1838), 69 ff.;
986
id., ‘Selbstkritik des Liberalismus’ (1841), in Gesammelte
Schriften, pt 13 (Mannheim, 1846), 76–116 (116); Mayer,
‘Anfänge’ (note 13), 9; Friedrich Zunkel, Der rheinisch-westfälische
Unternehmer 1834–1879. Ein Beitrag zur Geschichte des deutschen
Bürgertums im 19. Jahrhundert (Cologne: Westdeutscher Verlag,
1962); Elisabeth Fehrenbach, ‘Rheinischer Liberalismus und
gesellschaftliche Verfassung’, in Liberalismus (note 5), ed.
Schieder, 272–94; Paul Nolte, Gemeindebürgertum und
Liberalismus in Baden 1800–1850. Tradition—Radikalismus—
Republik (Göttingen: Vandenhoeck & Ruprecht, 1994); Walter,
Denken (note 13), 112 ff., 145 ff., 269 ff.
15. Karl Marx, ‘Zur Kritik der Hegelschen Rechtsphilosophie
Einleitung’, MEW i. 378–91 (quotes: 379, 385, 391; emphases in
original); Karl Marx/Friedrich Engels, Manifest der
Kommunistischen Partei, ibid., iv. 459–93 (493); Karl Marx,
‘Contribution to the Critique of Hegel’s Philosphy of Law’, MECW
iii. 175–87 (quotes: 176, 182, 187), tr. as Manifesto of the
Communist Party, MECW, vi. 477–519 (519); Heinrich August
Winkler, ‘Zum Verhältnis von bürgerlicher und proletarischer
Revolution bei Marx und Engels’, in Revolution, Staat, Faschismus.
Zur Revision des Historischen Materialismus (Göttingen:
Vandenhoeck & Ruprecht, 1978), 8–34.
16. Wehler, Gesellschaftsgeschichte (note 3), ii. 241 ff., 585 ff.;
Wolfgang Hardtwig, Vormärz. Der monarchische Staat und das
Bürgertum (Munich: Deutscher Taschenbuch Verlag, 1985), 88 ff.;
Jürgen Kocka, Weder Stand noch Klasse. Unterschichten um 1800
(Bonn: J. H. W. Dietz, 1990); id., Arbeitsverhältnisse und
987
Arbeiterexistenzen. Grundlagen der Klassenbildung im 19.
Jahrhundert (Bonn: J. H. W. Dietz, 1990), esp. 373 ff.; Wolfgang
Schieder, Anfänge der deutschen Arbeiterbewegung. Die
Auslandsvereine im Jahrzehnt nach der Juli-Revolution von 1830
(Stuttgart: Klett, 1963); Werner Conze, ‘Vom “Pöbel” zum
“Proletariat”, Sozialgeschichtliche Voraussetzungen für den
Sozialismus in Deutschland’, in Moderne deutsche Sozialgeschichte,
ed. Hans-Ulrich Wehler (Cologne: Kiepenheuer und Witsch,
1966), 111–36; Frolinde Balser, Sozial-Demokratie 1848/1863.
Die erste deutsche Arbeiterorganisation ‘Allgemeine Deutsche
Arbeiterverbrüderung’ nach der Revolution, 2 vols., 2nd edn
(Stuttgart: Klett, 1966).
17. Heinrich von Sybel, Die politischen Parteien der Rheinprovinz, in
ihrem Verhältniss zur preussischen Verfassung (Düsseldorf: Verlag
von Julius Buddeus, 1847), 63, 81–2; Huber,
Verfassungsgeschichte (note 5), ii. 492 ff. (Frederick William IV
quote: 495); Nipperdey, Geschichte (note 8), 399 ff.; Wehler,
Gesellschaftsgeschichte (note 3), ii., 677 ff.; Michael Neumüller,
Liberalismus und Revolution. Das Problem der Revolution in der
deutschen liberalen Geschichtsschreibung des 19. Jahrhunderts
(Düsseldorf: Pädagogischer Verlag Schwann, 1963); 1848/49 in
Europa und der Mythos der Französischen Revolution, ed. Irmtraud
Götz von Olenhusen (Göttingen: Vandenhoeck & Ruprecht,
1998).
18. Peter J. Katzenstein, Disjoined Partners: Austria and Germany since
1815 (Berkeley: University of California Press, 1976), 35 ff.;
Nipperdey, Deutsche Geschichte (note 8), 337 ff.; Lutz, Zwischen
988
Habsburg (note 12), 169 ff.; Langewiesche, Europa (note 3), 116
ff.; Huber, Verfassungsgeschichte (note 5), ii. 451 ff.
19. Huber Verfassungsgeschichte, ii. 448–51 (here also quotes from
the Offenburg and Heppenheim demands), 502 ff., 663 ff.; Ulrike
von Hirschhausen, Liberalismus und Nation. Die Deutsche Zeitung
1847–1850 (Düsseldorf: Droste, 1998); Veit Valentin, Geschichte der
deutschen Revolution 1848–1849, 2 vols., i. Bis zum Zusammentritt
des Frankfurter Parlaments (1931; Cologne: Kiepenheuer &
Witsch, 1970), 161 ff.; Treitschke, Geschichte, pt 5, (note 9), 564
ff.; Reimer Hansen, ‘Was bedeutet up ewig ungedeelt? Das Ripener
Privileg von 1460 im deutsch–dänischen Nationalkonflikt des 19.
Jahrhunderts’, Grenzfriedenshefte 1996/4, 215–32. Quote from
Engels: id., ‘Die Bewegungen von 1847’, MEW, iv. 494–503
(502–3), tr. as ‘The Movements of 1847’, MECW, vi. 520–9
(529).
20. Marx, ‘Kritik’ (note 15), 391 (emphases in original), tr. ‘Critique’
(note 15), 182 (emphases in original); Roger Price, The Second
French Republic: A Social History (London: Batsford, 1972);
Gilbert Ziebura, Frankreich 1789–1870. Entstehung einer
bürgerlichen Gesellschaftsformation (Frankfurt: Campus, 1979),
142 ff.
21. Valentin, Geschichte (note 19), i. 338 ff., 410 ff. (quotes: 426,
451); ii. Bis zum Ende der Volksbewegung (1932; Cologne:
Kiepenheuer & Witsch, 1970), 75 ff.; Rudolf Stadelmann, Soziale
und politische Geschichte der Revolution von 1848, (1948), 2nd
edn, Darmstadt: Wissenschaftliche Buchgesellschaft, 1962), 22
ff., tr. as Social and Political History of the German 1848 Revolution
989
(Athens, OH: Ohio University Press, 1975); Wehler,
Gesellschaftsgeschichte (note 3), ii. 642 ff., 703 ff.; Huber,
Verfassungsgeschichte (note 5), ii. 502 ff., 547 ff., 571 ff.; Die
Revolutionen von 1848/49. Erfahrung— Verarbeitung—Deutung, ed.
Christian Jansen/Thomas Mergel (Göttingen: Vandenhoeck &
Ruprecht, 1998); Revolution in Deutschland und Europa 1848/49,
ed. Wolfgang Hardtwig (Göttingen: Vandenhoeck & Ruprecht,
1998); Wolfgang J. Mommsen, 1848. Die ungewollte Revolution
(Frankfurt: Fischer, 1998); Europa 1848. Revolution und Reform,
ed. Dieter Dowe et al. (Bonn 1998); Dieter Hein, Die Revolution
von 1848/49 (Munich: C. H. Beck, 1998); 1848. Revolution in
Europa. Verlauf, politische Programme, Folgen und Wirkungen, ed.
Heiner Timmermann (Berlin: Duncker & Humblot, 1999); 1848.
Revolution in Deutschland, ed. Christof Dipper and Ulrich Speck
(Frankfurt: Insel, 1998); Helmut Berding, Moderner
Antisemitismus in Deutschland (Frankfurt: Suhrkamp, 1988), 74 ff.;
Stefan Rohrbacher, Gewalt im Biedermeier. Antijüdische
Ausschreitungen in Vormärz und Revolution (1815–1848/49)
(Frankfurt: Campus, 1993), 181 ff.
22. Verhandlungen des deutschen Parlaments. Officielle Ausgabe. Mit
einer geschichtlichen Einleitung über die Entstehung der Vertretung
des ganzen deutschen Volkes, 2 Lieferungen, ed. F. S. Jucho
(Frankfurt: J. D. Sauerländer, 1848), i. 172 (Poland resolution);
Dokumente zur deutschen Verfassungsgeschichte, i. Deutsche
Verfassungsdokumente 1803–1850, 3rd edn, ed. Ernst Rudolf
Huber (Stuttgart: Kohlhammer, 1978), 448–9 (Frederick William
IV’s proclamation on 21 Mar. 1848); Huber, Verfassungsgeschichte
(note 5), ii. 571 ff.; Valentin, Geschichte (note 19), i. 466 ff., 520
990
ff.; Meinecke, Weltbürgertum (note 4), 301 ff.; Franz X. Vollmer,
‘Die 48er Revolution in Baden’, in Becker et al., Badische
Geschichte (note 4), 37–64; Rüdiger Hachtmann, Berlin 1848. Eine
Politikund Gesellschaftsgeschichte der Revolution (Bonn: J. H. W.
Dietz, 1997), 157 ff.; Hans Christof Kraus, Ernst Ludwig Gerlach.
Politisches Denken und Handeln eines preussischen Altkonservativen,
2 vols. (Göttingen: Vandenhoeck & Ruprecht, 1994), i. 395 ff.;
Wolfram Siemann, Die deutsche Revolution von 1848/49
(Frankfurt: Suhrkamp, 1985), 146 ff., tr. as The German
Revolution of 1848–49 (New York: St Martin’s Press, 1998);
Reinhard Wittram, Die Nationalitätenfrage in Europa und die
Erschütterung des europäischen Staatensystems (1848–1917)
(Stuttgart: Klett, 1954), 2–3 (PalackÝ); Hans Rothfels, ‘Das erste
Scheitern des Nationalstaates in Ost-Mittel-Europa 1848/49’, in
Zeitgeschichtliche Betrachtungen (Göttingen: Vandenhoeck &
Ruprecht, 1959), 40–53; 1848/49. Revolutionen in
Ostmitteleuropa, ed. Rudolf Jaworski/Robert Luft (Munich:
Oldenbourg, 1996).
23. Stenographischer Bericht über die Verhandlungen der deutschen
constituierenden Nationalversammlung zu Frankfurt am Main. Hg.
auf Beschluss der Nationalversammlung durch die Redactions-
Commission und in deren Auftrag von Prof. Franz Wigard, 9 vols.
(Leipzig: Breitkopf & Härtel, 1848–9), iii. 1182 (Dahlmann, 5
Sept. 1848; emphases in original), 2048 (Venedey, 14 Sept.
1848); Gustav Rümelin, Aus der Paulskirche. Berichte aus dem
Schwäbischen Merkur aus den Jahren 1848 und 1849 (Stuttgart: G.
J. Göschen, 1892), 85 ff.; Valentin, Geschichte, ii. (note 21), 1 ff.,
95 ff., 149 ff.; Huber, Verfassungsgeschichte (note 5), ii. 587 ff.,
991
682 ff.; Wehler, Gesellschaftsgeschichte (note 3), ii. 706 ff.;
Nipperdey, Geschichte (note 8), 617 ff.; Manfred Botzenhardt,
Deutscher Parlamentarismus in der Revolutionszeit 1848–1850
(Düsseldorf: Droste, 1977), 184 ff.; Werner Boldt, Die Anfänge des
deutschen Parteiwesens. Fraktionen, politische Vereine und Parteien
in der Revolution 1848 (Paderborn: Schöningh, 1971), 18 ff.;
Frank Eyck, Deutschlands grosse Hoffnung. Die Frankfurter
Nationalversammlung (Munich: P. List, 1973), 77 ff., orig. The
Frankfurt Parliament 1848–1849 (London: St Martin’s Press,
1968); Vollmer, ‘Revolution’ (note 22), 53 ff.; Franz Mehring,
Geschichte der deutschen Sozialdemokratie (1897–8), 2 vols., pt 1,
Von der Julirevolution bis zum preussischen Verfassungsstreite 1830
bis 1863 (Berlin: Dietz, 1960), 429 ff.; Balser, Sozial-Demokratie
(note 16); Jürgen Bergmann, Wirtschaftskrise und Revolution,
Handwerker und Arbeiter 1848/49 (Stuttgart: Klett-Cotta, 1986);
Thomas Mergel, Zwischen Klasse und Konfession. Katholisches
Bürgertum im Rheinland 1794–1914 (Göttingen: Vandenhoeck &
Ruprecht, 1994), 117 ff. The battles named by Venedey:
Fehrbellin, 1675, Prussian victory under the Great Elector over
Sweden; Rossbach, 1757, Prussian victory under Frederick the
Great in the Seven Years War over the French and the army of
the Reich; Katzbach, 1813, Prussian victory under Blücher over
the French.
24. Verhandlungen der Versammlung zur Vereinbarung der Preussischen
Staats-Verfassung, 3 vols. (Berlin: Verlag der Deckerschen
Geheimen Ober-Hofbuchdruckerei, 1848), i. 417 (Waldeck), iii.
154–5 (Temme); Manfred Botzenhardt, ‘Das preussische
Parlament und die deutsche Nationalversammlung im Jahre
992
1848’, in Regierung, Bürokratie und Parlament in Preussen und
Deutschland von 1848 bis zur Gegenwart, ed. Gerhard A. Ritter
(Düsseldorf: Droste, 1983), 14–40; Peter Borowsky, ‘Was ist
Deutschland? Wer ist deutsch? Die Debatte zur nationalen
Identität 1848 in der deutschen Nationalversammlung zu
Frankfurt und der preussischen Nationalversammlung zu Berlin’,
in Vom schwierigen Zusammenwachsen der Deutschen. Nationale
Identität und Nationalismus im 19. und 20. Jahrhundert, ed. Bernd
Jürgen Wendt (Frankfurt: P. Lang, 1992), 81–95; Meinecke,
Weltbürgertum (note 4), 327 ff.
25. Karl Marx, ‘Die revolutionäre Bewegung’, MEW, vi. 148–50
(150; emphasis in original); Friedrich Engels, ‘Der magyarische
Kampf’, ibid. 165–76 (172, 176); id., ‘Der demokratische
Panslawismus’, ibid., 271–86 (286); Karl Marx, ‘The
Revolutionary Movement’, MECW, viii. 213–15 (215; emphases
in original); Friedrich Engels, ‘The Magyar Struggle’, ibid. 227–
38 (234, 238); id., ‘Democratic Pan-Slavism’, ibid., 362–78 (377–
8); Valentin, Geschichte (note 21), ii. 183 ff.; Huber,
Verfassungsgeschichte (note 5), ii. 710 ff.; Lutz, Zwischen Habsburg
(note 12), 292 ff.
26. Joseph Hansen, Gustav von Mevissen. Ein rheinisches Lebensbild
1815–1899, ii. (Berlin: G. Reimer, 1906), 448 (letter to Georg
Mallinckrodt); Stenographischer Bericht (note 23), vi. 4096–7
(Dahlmann, 15 Dec. 1848); Hachtmann, Berlin (note 22), 688 ff.;
Wehler, Gesellschaftsgeschichte (note 3), ii. 753 ff.; Nipperdey,
Geschichte (note 8), 647 ff.; Valentin, Geschichte (note 21), ii. 227
ff.; Huber, Verfassungsgeschichte (note 5), ii. 737 ff.; Meinecke,
993
Weltbürgertum (note 4), 349 ff.; Kraus, Gerlach (note 22), i. 430
ff.; Konrad Canis, ‘Ideologie und politische Taktik der junkerlich-
militaristischen Reaktion bei der Vorbereitung und Durchführung
des Staatstreiches in Preussen im Herbst 1848’, Jahrbuch für
Geschichte, 7 (1972), 461–503; Blasius, ‘Friedrich Wilhelm IV’
(note 9), 600 ff. Quote from Jacoby: Verhandlungen (note 24), iii.
325; Waldecks speech on 31 Oct. 1848: ibid., 292–3.
27. Stenographischer Bericht (note 23), i. 183 (declaration in defense
of the nationalities from of 31 May 1848), 214–15. (Arndt, 5
June 1848); ii. 1145 (Jordan, 24 July 1848), 1156 (Vogt, 12
Aug. 1848); iv. 2779–80. (Arneth, 20 Oct. 1848), 2855
(Mühlfeld, 24 Oct. 1848), 2859 (Vincke, 24 Oct. 1848), 2869
(Reichensperger, 24 Oct. 1848), 2876 (Uhland, 26 Oct. 1848),
2881 (Deym, 26 Oct. 1848), 2898 ff. (Gagern 26 Oct. 1848);
Dokumente (note 22), i. ed. Huber, 395 (Art. XIII, § 188 of the
imperial constitution in defense of the nationalities); id.,
Verfassungsgeschichte (note 5), ii. 792 ff.; Valentin, Geschichte
(note 21), ii. 215 ff.; Günter Wollstein, Das ‘Grossdeutschland’ der
Paulskirche. Nationale Ziele in der bürgerlichen Revolution 1848/49
(Düsseldorf: Droste, 1977), 266 ff.; Rudolf Lill, ‘Grossdeutsch und
kleindeutsch im Spannungsfeld der Konfessionen’, in Probleme des
Konfessionalismus in Deutschland seit 1800, ed. Anton Rauscher
(Paderborn: Schöningh, 1984), 29–47.
28. Stenographischer Bericht (note 23), vi. 4626 ff. (Beseler, 13 Jan.
1849), 5807 ff. (Radowitz, 17 Mar. 1849), 5823 (Vogt, 17 Mar.
1849), 5839 ff. (Mohl, 17 Mar. 1849); Huber,
Verfassungsgeschichte (note 5), ii. 814 ff.; Siemann, Revolution
994
(note 22), 192 ff.; Valentin, Geschichte (note 22), ii. 305 ff.; Jörg-
Detlef Kühne, Die Reichsverfassung der Paulskirche. Vorbild und
Verwirklichung im späteren deutschen Rechtsleben, 2nd edn
(Neuwied: Luchterhand, 1998).
29. Valentin, Geschichte (note 21), ii. 381 ff., 448 ff.; Huber,
Verfassungsgeschichte (note 5), ii. 854 ff.; Hachtmann, Berlin (note
22), 798 ff.; Barclay, Anarchie (note 9), 272 ff.; Bussmann,
Zwischen Preussen (note 9), 284 ff.; Christoph Klessmann, ‘Zur
Sozialgeschichte der Reichsverfassungskampagne von 1849’, HZ
218 (1974), 283–337; Friedrich Engels, Die deutsche
Reichsverfassungskampagne (1850), MEW, vii. 109–97 (196), tr. as
The Campaign for the German Imperial Constitution, MECW, x.
147–239 (237); Fürst Otto von Bismarck, Die gesammelten Werke
(Friedrichsruh edition, Berlin: O. Stollberg, 1924–); x. 103 ff. The
‘mud and clay’ quote is from a letter of Frederick William IV to
the Prussian envoy in London, Christian Karl Josias von Bunsen,
on 13 Dec. 1848; see Leopold von Ranke, Aus dem Briefwechsel
Friedrich Wilhelms IV. mit Bunsen (Leipzig: Duncker & Humblot,
1873), 233–4.
30. Rümelin, Paulskirche (note 23), 14, 240–1; Stenographischer
Bericht (note 23), ii. 1101 (Ruge), vi. 4596 (Beckerath);
Nipperdey, Geschichte (note 8), 663 ff.; Wehler,
Gesellschaftsgeschichte (note 3), ii. 759 ff.; Siemann, Revolution
(note 22), 223 ff.; Becker, Zeit (note 3), 294 ff.; Die deutsche
Revolution von 1848, ed. Dieter Langewiesche (Darmstadt:
Wissenschaftliche Buchgesellschaft, 1983); Hans Rothfels, 1848.
Betrachtungen im Abstand von hundert Jahren (Darmstadt:
995
Wissenschaftliche Buchgesellschaft, 1972, orig. ‘1848: One
Hundred Years after’, Journal of Modern History, 20 (1948);
Wolfgang Schieder, ‘1848/49: Die ungewollte Revolution’, in
Wendepunkte deutscher Geschichte 1948–1990, ed. Carola Stern
and Heinrich August Winkler (Frankfurt: Fischer Taschenbuch,
1994), 17–42; Lothar Gall, ‘Die Germania als Symbol nationaler
Einheit im 19. und 20. Jahrhundert’, in Nachrichten der Akademie
der Wissenschaften in Göttingen. I. Philologisch-Historische Klasse,
Jg. 1993, 2 (Göttingen: Vandenhoeck & Ruprecht, 1993), 10–11.
996
Staatsfinanzen und Politik 1848–1860, 2 vols. (Göttingen:
Vandenhoeck & Ruprecht, 1978); Ernst Rudolf Huber, Deutsche
Verfassungsgeschichte seit 1789, iii. Bismarck und das Reich, 3rd
edn (Stuttgart: W. Kohlhammer, 1988), 35 ff.; Dieter Grimm,
Deutsche Verfassungsgeschichte 1776–1866 (Frankfurt: Suhrkamp,
1988), 214 ff.; Hans Boldt, ‘Die preussische Verfassung vom 31.
Januar 1850. Probleme ihrer Interpretation’, in Preussen im
Rückblick, ed. Hans-Jürgen Puhle and Hans-Ulrich Wehler, GG,
Sonderheft 6 (Göttingen: Vandenhoeck & Ruprecht, 1980), 224–
46; Günther Grünthal, Parlamentarismus in Preussen 1848/49–
1857/58. Preussischer Konstitutionalismus, Parlament und Regierung
in der Reaktionsära (Düsseldorf: Droste, 1982); Hartwin
Spenkuch, Das Preussische Herrenhaus. Adel und Bürgertum in der
Ersten Kammer des Landtags 1854–1918 (Düsseldorf: Droste,
1998), 47 ff.; Wilhelm Füssl, Professor in der Politik: Friedrich
Julius Stahl (1802–1861). Das monarchische Prinzip und seine
Umsetzung in die parlamentarische Praxis (Göttingen: Vandenhoeck
& Ruprecht, 1988), 44 ff., 108 ff. (quotes from Stahls ‘Entwurf
für eine conservative Partei’ of Feb./Mar. 1849: 183 ff.); Gerhard
Masur, Friedrich Julius Stahl. Geschichte seines Lebens. Aufstieg und
Entfaltung, 1802–1840 (Berlin: E. S. Mittler, 1930), 200 ff.; Horst
Zillessen, Protestantismus und politische Form. Eine Untersuchung
zum protestantischen Verfassungsverständnis (Gütersloh: G. Mohn,
1971), 67 ff.; Hans-Christof Kraus, Ernst Ludwig von Gerlach.
Politisches Denken und Handeln eines preussischen Altkonservativen,
2 vols. (Göttingen: Vandenhoeck & Ruprecht, 1994), ii. 586 ff.;
David E. Barclay, Anarchie und guter Wille. Friedrich Wilhelm IV.
und die preussische Monarchie (Berlin: Siedler, 1995), 308 ff., orig.
997
Frederick William IV and the Prussian Monarchy, 1840–1861
(Oxford: Oxford University Press, 1995).
3. Helmut Böhme, Deutschlands Weg zur Grossmacht. Studien zum
Verhältnis von Wirtschaft und Staat während der
Reichsgründungszeit 1848–1881 (Cologne: Kiepenheuer & Witsch,
1966), 19 ff.; Richard Löwenthal, ‘Internationale Konstellation
und innerstaatlicher Systemwandel’, HZ 212 (1971), 41–58;
Anselm Doering-Manteuffel, Vom Wiener Kongress zur Pariser
Konferenz. England, die deutsche Frage und das Mächtesystem 1815–
1856 (Göttingen: Vandenhoeck & Ruprecht, 1991), 187 ff.;
Winfried Baumgart, Der Friede von Paris 1856. Studien zum
Verhältnis von Kriegführung, Politik und Friedensbewahrung
(Munich: R. Oldenbourg, 1972), tr. as The Peace of Paris, 1856:
Studies in War, Diplomacy, and Peacemaking (Santa Barbara, CA:
ABC-Clio, 1981); Werner E. Mosse, The Rise and Fall of the
Crimean System 1855–1871: The Story of a Peace Settlement
(London: Macmillan, 1963); Hans Joachim Schoeps, Das andere
Preussen. Konservative Gestalten und Probleme im Zeitalter Friedrich
Wilhelms IV., 2nd edn (Honnef: Peters, 1957), 124 ff.; Ernst
Engelberg, Bismarck. Urpreusse und Reichsgründer (Berlin: Siedler,
1985), 417 ff.; Lothar Gall, Bismarck. Der weisse Revolutionär
(Berlin: Propyläen, 1980), 167 ff., tr. as Bismarck, the White
Revolutionary (London: Allen & Unwin, 1986). For the quotes
from Bismarck’s letters to Leopold von Gerlach, see Fürst Otto
von Bismarck, Die gesammelten Werke, Friedrichsruh edition
(Berlin: O. Stollberg, 1924–), xiv/i: 464 ff. (2 May 1857), 470 ff.
(30 May 1857; emphasis in original).
998
4. Ludwig August von Rochau, Grundsätze der Realpolitik.
Angewendet auf die staatlichen Zustände Deutschlands (1853), ed.
and introd. Hans-Ulrich Wehler (Frankfurt: Ullstein, 1972), 25–6,
67, 169, 173; Siegfried A. Kaehler, ‘Realpolitik zur Zeit des
Krimkrieges. Eine Säkularbetrachtung’, HZ 174 (1952), 417–78;
Hans Rothfels, ‘Zeitgeschichtliche Betrachtungen zum Problem
der Realpolitik’, in Zeitgeschichtliche Betrachtungen (Göttingen:
Vandenhoeck & Ruprecht, 1959), 179–98; Karl-Georg Faber,
‘Realpolitik als Ideologie. Die Bedeutung des Jahres 1866 für das
politische Denken in Deutschland’, HZ 203 (1966), 1–45.
5. Rochau, Grundsätze (note 4), 126 ff., 146 ff., 150–1.
6. Lorenz von Stein, Geschichte der sozialen Bewegung in Frankreich
von 1789 bis auf unsere Tage, 3 vols., iii. Das Königtum, die
Republik und die Souveränität der französischen Gesellschaft seit der
Februarrevolution 1848 (1850; new edn Darmstadt:
Wissenschaftliche Buchgesellschaft, 1959), 406, 408.
7. Ibid. 7, 37–41; id., Zur preussischen Verfassungsfrage (1852; 2nd
edn Darmstadt: Wissenschaftliche Buchgesellschaft, 1961) 4, 6,
12, 34 ff. (emphases in original); Dirk Blasius, ‘Lorenz von Stein’,
in Deutsche Historiker, i., ed. Hans-Ulrich Wehler (Göttingen:
Vandenhoeck & Ruprecht, 1971), 25–38; Hermann Beck, The
Origins of the Authoritarian Welfare State in Prussia: Conservatives,
Bureaucracy, and the Social Question, 1815–1870 (Ann Arbor:
University of Michigan Press, 1995), esp. 71 ff.
8. Karl Marx, Die Klassenkämpfe in Frankreich (1850), MEW, vii. 9–
107 (89–90); id., ‘Die moralisierende Kritik und die kritisierende
Moral’ (1847), MEW, iv. 331–59 (339); id., ‘Sieg der
999
Kontrerevolution in Wien’ (7 Nov. 1848), MEW, v. 455–7 (457;
all emphases in the original); id., letter to Joseph Weydemeyer of
5 Mar. 1852, MEW, xxviii. 503–9 (508); Marx, The Class Struggles
in France, MECW, x. 45–146 (127); id., ‘Moralising Criticism and
Critical Morality’, MECW, vi. 312–40 (319); id., ‘Victory of the
Counter-Revolution in Vienna’, MECW, vii. 503–6 (505–6); letter
to Joseph Weydemeyer of 5 Mar. 1852, MECW, xxxix. 60–6 (62–
5).
9. Rochau, Grundsätze (note 4), 184; Karl Marx, Der achtzehnte
Brumaire des Louis Bonaparte, MEW (note 8), viii. 111–207 (in the
order of the quotes: 118, 204, 154); Karl Marx/Friedrich Engels,
Manifest der Kommunistischen Partei, MEW, iv. 459–93 (464); ‘Die
politischen Parteien in England—Die Lage in Europa’ (24 June
1858), MEW, xii. 503–6 (505; emphases in original); id., ‘Die
französische Abrüstung’ (12 Aug. 1859), ibid. 447–9 (448);
Engels’s letter to Marx of 15 Nov. 1857, ibid. xxix. 208–12 (211–
12); Karl Marx, ‘Das neue Ministerium’ (9 Nov. 1858), xii. 636–9;
Friedrich Engels, ‘Europa im Jahre 1858’ (23 Dec. 1858), ibid.,
654–8 (658); Karl Marx, The Eighteenth Brumaire of Louis
Bonaparte, MECW, xi. 99–197 (in the order of the quotes: 106–7,
194, 143); Karl Marx/Friedrich Engels, Communist Party
Manifesto MECW, vi. 477–519 (486); Marx, ‘Political Parties in
England—Situation in Europe’, MECW, xv. 566–9 (568); id., ‘The
French Disarmament’ MECW, xvi. 442–4 (443); Engels’s letter to
Marx of 15 Nov. 1857; MECW, xl. 200–4 (203); Marx, ‘The New
Ministry’, MECW, xvi. 101–5; Engels, ‘Europe in 1858’, ibid.
120–4 (124); Heinrich August Winkler, ‘Zum Verhältnis von
bürgerlicher und proletarischer Revolution bei Marx und Engels’,
1000
in Revolution, Staat, Faschismus. Zur Revision des Historischen
Materialismus (Göttingen: Vandenhoeck & Ruprecht, 1978), 8–34;
id., ‘Primat der Ökonomie? Zur Rolle der Staatsgewalt bei Marx
und Engels’, ibid. 35–64. Marx’s ‘translation’ of ‘Hic Rhodus, hic
salta!’ is from Georg Wilhelm Friedrich Hegel, Grundlinien der
Philosophie des Rechts oder Naturrecht und Staatswissenschaft im
Grundrisse (Vorrede), in Sämtliche Werke, 3rd edn, vii (Stuttgart:
Fromman, 1952), p. xxii.
10. Felix Salomon, Die deutschen Parteiprogramme (Leipzig: B. G.
Teubner, 1907), i. 41–2; Ludolf Parisius, Deutschlands politische
Parteien und das Ministerium Bismarck (Berlin: J. Guttentag,
1878), 15 ff.; Huber, Verfassungsgeschichte, iii. (note 2), 269 ff.;
Heinrich August Winkler, Preussischer Liberalismus und deutscher
Nationalstaat. Studien zur Geschichte der Deutschen Fortschrittspartei
1861–1866 (Tübingen: Mohr, 1964), 3 ff.
11. Bismarck und der Staat. Ausgewählte Dokumente, introd. Hans
Rothfels (Stuttgart: Kohlhammer, n.d. [1953]), 111–15 (113–14;
Bismarck’s letter to Alvensleben of 5 May 1859); Ferdinand
Lassalle, Der italienische Krieg und die italienische Aufgabe
Preussens. Eine Stimme aus der Demokratie (1859), in Gesamtwerke,
ed. Erich Blum (Leipzig: Pfau, n.d. [1899–1909]), ii. 369–442
(391–2, 435–8); Friedrich Engels, Po und Rhein (April 1859),
MEW, xiii. 225–68 (227, 268), tr. as Po and Rhine, MECW, xvi.
213–55 (215, 255). Hans Rosenberg, Die nationalpolitische
Publizistik Deutschlands vom Eintritt der Neuen Ära in Preussen bis
zum Ausbruch des deutschen Krieges. Eine kritische Bibliographie, 2
vols. (Munich: R. Oldenbourg, 1935), i. 20–158; Hermann
1001
Oncken, Lassalle. Zwischen Marx und Bismarck (1904: 5th edn
(Stuttgart: Kohlhammer, 1966), 131 ff.; Arnold Oskar Meyer,
Bismarck. Der Mensch und der Staatsmann (Stuttgart: K. F.
Koehler, 1949), 123 ff.; Erich Portner, Die Einigung Italiens im
Urteil liberaler deutscher Zeitgenossen (Bonn: L. Röhrscheid, 1959);
Huber, Verfassungsgeschichte, iii (note 2), 254 ff.; Wehler,
Gesellschaftsgeschichte, iii. (note 1), 228 ff. On Rochau see Ludwig
August von Rochau, ‘Die Frage von Krieg und Frieden’, introd. to
the 2nd edn of id., Grundsätze (note 4), 192–203.
12. Peter Katzenstein, Disjoined Partners: Austria and Germany since
1815 (Berkeley: University of California Press, 1976), 66 ff.;
Heinrich Lutz, Zwischen Habsburg und Preussen, Deutschland
1815–1866 (Berlin: Siedler, 1985), 403 ff.; Thomas Nipperdey,
Deutsche Geschichte 1800–1866. Bürgerwelt und starker Staat
(Munich: C. H. Beck, 1983), 704 ff.; Huber, Verfassungsgeschichte,
iii (note 2), 138 ff., 378 ff., 384 ff.; Georg Franz, Liberalismus. Die
deutsch-liberale Bewegung in der habsburgischen Monarchie
(Munich: G. D. W. Callwey, 1955), 264 ff.; Andreas Biefang,
Politisches Bürgertum in Deutschland 1857–1868. Nationale
Organisationen und Eliten (Düsseldorf: Droste, 1994), 17 ff., 185
ff., 301 ff.; Shlomo Na’aman, Der Deutsche Nationalverein. Die
politische Konstituierung des deutschen Bürgertums 1859–1867
(Düsseldorf: Droste, 1987), 41 ff.; Rainer Noltenius, ‘Schiller als
Führer und Heiland. Das Schillerfest 1859 als nationaler Traum
von der Geburt des zweiten deutschen Kaiserreiches’, Öffentliche
Festkultur. Politische Feste in Deutschland von der Aufklärung bis
zum Ersten Weltkrieg, ed. Dieter Düding et al. (Hamburg:
Rowohlt, 1988), 237–58; Dieter Düding, ‘Nationale
1002
Oppositionsfeste der Turner, Sänger und Schützen im 19.
Jahrhundert’, ibid. 166–90; id., ‘Die deutsche Nationalbewegung
des 19. Jahrhunderts als Vereinsbewegung. Anmerkungen zu
ihrer Struktur und Phänomenologie zwischen
Befreiungskriegszeitalter und Reichsgründung’, GWU 42 (1991),
601–24; Hans-Thorwald Michaelis, Unter schwarz-rot-goldenem
Banner und dem Signum des Doppeladlers. Gescheiterte
Volksbewaffnungs- und Vereinigungsbestrebungen in der Deutschen
Nationalbewegung und im Deutschen Schützenbund 1859–1869
(Frankfurt: P. Lang, 1993), 229 ff.; Dietmar Klenke,
‘Nationalkriegerisches Gemeinschaftsideal als politische Religion.
Zum Vereinsnationalismus der Sänger, Schützen und Turner am
Vorabend der Einigungskriege’, HZ 260 (1995), 395–448; id., Der
singende ‘deutsche Mann’, Gesangvereine und deutsches
Nationalbewusstsein von Napoleon bis Hitler (Münster: Waxmann,
1998); Svenja Goltermann, Körper der Nation. Habitusforschung
und die Politik des Turnens 1860–1890 (Göttingen: Vandenhoeck
& Ruprecht, 1998); Stefan Illig, Zwischen Körperertüchtigung und
nationaler Bewegung. Turnvereine in Bayern 1860–1890 (Cologne:
SH-Verlag, 1998); Dieter Langewiesche, ‘Nation, Nationalismus
und Nationalstaat: Forschungsstand und
Forschungsperspektiven’, NPL 40 (1995), 190–236.
13. Deutsche Parteiprogramme, ed. Wilhelm Mommsen (Munich: Isar,
1960), 132–5 (platform of the German Progressive Party);
Bismarck, Werke (note 3), x. 139–40 (budget commission, 30
Sept. 1862), xv. 179–80, 194 ff. (Erinnerung und Gedanke);
Heinrich von Sybel, Die Begründung des Deutschen Reiches durch
Wilhelm I., 3 vols. (new edn Meersburg: F. M. Hendel, 1930), i.
1003
508 ff., tr. as The Founding of the German Empire by William I
(New York: Greenwood Press, 1968); Thomas Nipperdey, Die
Organisation der deutschen Parteien vor 1918 (Düsseldorf: Droste,
1961), 196 ff.; Fritz Löwenthal, Der preussische Verfassungsstreit
1862–1866 (Munich: Duncker & Humblot, 1914), 120 ff.; Rolf
Helfert, Der preussische Liberalismus und die Heeresreform von 1860
(Bonn: Holos, 1989), 67 ff.; Winkler, Liberalismus (note 10), 10
ff.; Huber, Verfassungsgeschichte, iii (note 2), 275 ff.
14. National-Zeitung [NZ], 2 Oct. 1862, Morgenblatt [Mbl], on
Bismarck’s ‘napoleonic ideas’; (Löwenstein);NZ, 23 Oct. 1862,
Abendblatt [Abl]; ‘Politische Correspondenz’, Preussische
Jahrbücher, 10 (1862), 402–18 (412); [Karl Twesten], ‘Lehre und
Schriften Auguste Comtes’, ibid. 4 (1859), 279–307 (306);
Stenographischer Bericht über die Verhandlungen des preussischen
Abgeordnetenhauses [Sten. Ber., LT], 1866 (1st session), 1, 79 ff.
(Twesten, 22 Feb. 1866); Otto Westphal, Weltund Staatsauffassung
des deutschen Liberalismus. Eine Untersuchung über die Preussischen
Jahrbücher und den konstitutionellen Liberalismus in Deutschland,
1858 bis 1863 (Munich: Oldenbourg, 1919), 297 ff.; Winkler,
Liberalismus (note 10), 16 ff.
15. Ludolf Parisius, Leopold Freiherr von Hoverbeck, 2 vols. (Berlin: J.
Guttentag, 1897–), ii. 53–5 (letters to Karl Witt of 27 June and
30 July 1865); Ferdinand Lassalle, Über Verfassungswesen (speech
on 16 Apr. 1862) (repr. Darmstadt: Wissenschaftliche
Buchgesellschaft, 1958), 56–7; Oncken, Lassalle (note 11), 178
ff., 296 ff.; Shlomo Na’aman, Demokratische und soziale Impulse in
der Frühgeschichte der deutschen Arbeiterbewegung der Jahre
1004
1862/63 (Wiesbaden: F. Steiner, 1969); id., Lassalle (Hanover,
1970), 527 ff.; Gustav Mayer, ‘Lassalle und Bismarck’, in id.,
Arbeiterbewegung und Obrigkeitsstaat, ed. Hans-Ulrich Wehler
(Bonn: Verlag Neue Gesellschaft, 1972), 93–118; Wolfgang
Schieder, ‘Das Scheitern des bürgerlichen in Radikalismus und
die sozialistische Parteibildung in Deutschland’, in
Sozialdemokratie zwischen Klassenbewegung und Volkspartei, ed.
Hans Mommsen (Frankfurt: Athenäum-Fischer Taschenbuch-
Verlag, 1974), 17–34; Rita Aldenhoff, Schulze-Delitzsch. Ein
Beitrag zur Geschichte des Liberalismus zwischen Revolution und
Reichsgründung (Baden-Baden: Nomos, 1984), 161 ff.
16. Heinrich Bernhard Oppenheim, ‘Ein Wort über politische und
staatsbürgerliche Pflichterfüllung’, Deutsche Jahrbücher, 13
(1864), 112–28 (118–19, 122; emphasis in original); Wilhelm
Löwe (von Calbe), ‘Preussens Beruf in der deutschen Sache’, ibid.
2 (1861), 169–90 (179); NZ, 17 Apr. 1862, Mbl.; Parisius,
Hoverbeck (note 15), i. 164–5. (Forckenbeck’s letter of 21 Aug.
1859); Winkler, Liberalismus (note 10), 26 ff.; Ernst Rudolf
Huber, Deutsche Verfassungsgeschichte seit 1789, i. Reform und
Restauration 1789–1830, 3rd edn (Stuttgart: Kohlhammer, 1990),
610 ff.; Dokumente zur deutschen Verfassungsgeschichte, i. Deutsche
Verfassungsdokumente 1803–1850, ed. id., 3rd edn (Stuttgart:
Kohlhammer, 1978), 119–20 (Bundeskriegsverfassung of 9 Apr.
1821).
17. Der Deutsche Nationalverein 1859–1867. Vorstands- und
Ausschussprotokolle, ed. Andreas Biefang (Düsseldorf: Droste,
1995), 246–8 (resolution of the committee of 25 May 1863);
1005
Sybel, Begründung (note 13), i. 581 ff.; Huber,
Verfassungsgeschichte, iii (note 2), 393 ff., 421 ff.; Biefang,
Bürgertum (note 12), 221 ff.; Willy Real, Der deutsche
Reformverein. Grossdeutsche Stimmen und Kräfte zwischen
Villafranca und Königgrätz (Lübeck: Matthiesen, 1966); Hans
Rosenberg, ‘Honoratiorenpolitiker und grossdeutsche
Sammlungsbestrebungen im Reichsgründungszeitalter’, in
Machteliten und Wirtschaftskonjunkturen. Studien zur neueren
deutschen Sozial- und Wirtschaftsgeschichte (Göttingen:
Vandenhoeck & Ruprecht, 1978), 198–254. On the controversy
between Sybel und Ficker see Universalstaat oder Nationalstaat.
Macht und Ende des Ersten Deutschen Reiches. Die Streitschriften von
Heinrich von Sybel und Julius Ficker zur deutschen Kaiserpolitik des
Mittelalters, 2nd edn, ed. Friedrich Schneider (Innsbruck:
Universitätsverlag Wagner, 1943).
18. Sten. Ber., LT, 1862 (2nd session), VI, enclosure 78 (report of the
unified commissions for finances and customs and trade and
industry in the commercial treaty with France); Victor Böhmert,
‘Deutschlands wirtschaftliche Neugestaltung’, Preussische
Jahrbücher, 18 (1866), 269–304 (270); Böhme, Weg (note 3), 91
ff.; Huber, Verfassungsgeschichte, iii (note 2), 615 ff.; Winkler,
Liberalismus (note 10), 71 ff.; Eugen Franz, Der
Entscheidungskampf um die wirtschaftspolitische Führung
Deutschlands (1856 bis 1867) (Munich: Verlag der Kommission,
1933); Volker Hentschel, Die deutschen Freihändler und der
volkswirtschaftliche Kongress 1858–1885 (Stuttgart: Klett, 1975),
61 ff.
1006
19. Lawrence D. Steefel, The Schleswig-Holstein Question (Cambridge,
MA: Harvard University Press, 1932), 3 ff.; Otto Brandt,
Geschichte Schleswig-Holsteins. Ein Grundriss, 4th edn (Kiel:
Mühlau, 1949), 168 ff.; Joachim Daebel, ‘Die Schleswig-Holstein-
Bewegung in Deutschland 1863/64’, Ph.D. thesis, Cologne, 1969,
45 ff.; Theodor Schieder, ‘Vom Deutschen Bund zum Deutschen
Reich’, Von der Französischen Revolution bis zum Ersten Weltkrieg
(= Bruno Gebhardt, Handbuch der Deutschen Geschichte, iii), 9th
edn, ed. Karl Erich Born et al. (Stuttgart: Union-Verlag, 1970),
99–220 (183 ff.); Ernst Rudolf Huber, Deutsche
Verfassungsgeschichte seit 1789, ii. Der Kampf um Einheit und
Freiheit 1830 bis 1850, 3rd edn (Stuttgart: Kohlhammer, 1988),
933 ff.; id., Verfassungsgeschichte, iii (note 2), 449 ff. Quote from
Lerchenfeld in Real, Reformverein (note 17), 185.
20. Sten. Ber., LT, 1863/64, I, 274 (Twesten, 2 Dec. 1863), 497–8
(Waldeck, 18 Dec. 1863); NZ, 29 May 1864, Mbl.; 11 Aug. 1864,
Mbl.; 12 Aug. 1864, Mbl.; Volkszeitung [VZ], 8. 16. 1864;
Winkler, Liberalismus (note 10), 50 ff.; Real, Reformverein (note
17), 170 ff.; Gall, Bismarck (note 3), 312 ff.; Huber,
Verfassungsgeschichte, iii (note 2), 482 ff.; Schieder, ‘Vom
Deutschen Bund’ (note 19), 188 ff.; Rudolf Stadelmann, Das Jahr
1865 und das Problem von Bismarcks deutscher Politik (Munich:
Oldenbourg, 1933); Andreas Kaernbach, Bismarcks Konzepte zur
Reform des Deutschen Bundes. Zur Kontinuität der Politik Bismarcks
und Preussens in der deutschen Frage (Göttingen: Vandenhoeck &
Ruprecht, 1991), 98 ff.; Eberhard Kolb, ‘Grosspreussen oder
Kleindeutschland? Zu Bismarcks deutscher Politik im
Reichsgründungszeitalter’, in Umbrüche deutscher Geschichte
1007
1866/71, 1918/19, 1929/33 (Munich: Oldenbourg, 1993), 11–
33.
21. Theodor Mommsen, Die Annexion Schleswig-Holsteins. Ein
Sendschreiben an die Wahlmänner der Stadt Halle und des
Saalkreises (Berlin: Weidmannsche, 1865), 17, 23; Sten. Ber., LT,
1865, III, 2117 (Michaelis, 13 June 1865); NZ, 1 June 1865,
Mbl.; 2 Aug. 1865, Mbl.; 3 Aug. 1865, Mbl.; VZ, 6 Aug. 1865; 16
Aug. 1865; 25 Aug. 1865, Mbl.; VZ, 12 Oct. 1865 (Harkorts
declaration); Schulthess’ Europäischer Geschichtskalender, 6 (1865)
(Nördlingen: C. H. Beck, 1866), 120–1 (declaration of the
Delegate Congress of 1 Oct. 1865); Nationalverein (note 17), 366–
7 (resolution of the general assembly of 29 Oct. 1865); Die
Sturmjahre der preussisch-deutschen Einigung 1859–1870. Politische
Briefe aus dem Nachlass liberaler Parteiführer, sel. and ed. Julius
Heyderhoff (= Deutscher Liberalismus im Zeitalter Bismarcks. Eine
politische Briefsammlung, i) (1925; repr. Osnabrück: Biblio-Verlag,
1967), 253–5 (Mommsen’s letter of 28 Sept. 1865), 255–7
(Twesten’s letter of 28 Sept. 1865); Huber, Verfassungsgeschichte,
iii (note 2), 506 ff.; Winkler, Liberalismus (note 10), 60 ff.;
Andreas Biefang, ‘National-preussisch oder deutsch-national? Die
Deutsche Fortschrittspartei in Preussen 1861–1867’, GG 23
(1997), 360–83; Alfred Heuss, Theodor Mommsen und das 19.
Jahrhundert (Kiel: F. Hirt, 1956), 177 ff.
22. Nationalverein (note 17), 212–13 (Miquel, Oct. 1865); Na’aman,
Nationalverein (note 12), 186 ff.; Adolf Rapp, Die Württemberger
und die nationale Frage 1863–1871 (Stuttgart: Kohlhammer,
1910), 98 (Beobachter, 17 Apr. 1864), 84 (Beobachter, 10 Feb.
1008
1864), 99–100 (Beobachter, 23–9 Apr. 1864); Dieter
Langewiesche, Liberalismus und Demokratie in Württemberg
zwischen Revolution und Reichsgründung (Düsseldorf: Droste,
1974), 309 ff.; Gerlinde Runge, Die Volkspartei in Württemberg von
1864 bis 1871. Die Erben der 48er Revolution im Kampf gegen die
preussisch-kleindeutsche Lösung der nationalen Frage (Stuttgart:
Kohlhammer, 1970), 25 ff.
23. Gustav Mayer, ‘Die Trennung der proletarischen von der
bürgerlichen Demokratie in Deutschland, 1863–1870’, in
Radikalismus, Sozialismus und bürgerliche Demokratie, ed. Hans-
Ulrich Wehler (Frankfurt: Suhrkamp, 1969), 108–78; Rolf Weber,
Kleinbürgerliche Demokraten in der deutschen Einheitsbewegung
1863–1866 (Berlin: Rütten & Loening, 1962), 81 ff.; Schieder,
‘Scheitern’ (note 15), 17 ff.
24. Lothar Gall, Der Liberalismus als regierende Partei. Das
Grossherzogtum Baden zwischen Restauration und Reichsgründung
(Wiesbaden: F. Steiner, 1968), 169 ff.; id., ‘Die parteiund
sozialgeschichtliche Problematik des badischen Kulturkampfes’,
Zeitschrift für die Geschichte des Oberrheins, 113 (1965), 151–96;
Josef Becker, Liberaler Staat und Kirche in der Ära von
Reichsgründung und Kulturkampf. Geschichte und Strukturen ihres
Verhältnisses in Baden 1860–1876 (Mainz: Matthias-Grünewald-
Verlag, 1973), 35 ff.; Böhme, Weg (note 3), 197 ff.; Kaernbach,
Konzepte (note 20), 204 ff.
25. Bismarck, Werke (note 3), v. 421 (edict to the Prussian envoy in
Munich, Heinrich VII Prince Reuss, of 24 Mar. 1866); NZ 5 Apr.
1866, Mbl.; 12 Apr. 1866, Abl.; 18 Apr. 1866, Abl. (Twesten);
1009
Kaernbach, Konzepte (note 20), 211 ff.; Huber,
Verfassungsgeschichte, iii (note 2), 510 ff.; Lutz, Zwischen Habsburg
(note 12), 552 ff.; Sybel, Begründung (note 13), ii. 390 ff.;
Heinrich Ritter von Srbik, Deutsche Einheit. Idee und Wirklichkeit
vom Heiligen Reich bis Königgrätz, 4 vols. (Munich: F. Bruckmann,
1935), iv. 320 ff.; Theodor Schieder, Die kleindeutsche Partei in
Bayern in den Kämpfen um die nationale Einheit 1863–1871
(Munich: C. H. Beck, 1936), 93 ff.; Rapp, Württemberger (note
22), 162 ff.; Gall, Liberalismus (note 24), 345 ff.; Winkler,
Liberalismus (note 10), 83 ff.
26. Franz, Liberalismus (note 12), 261–2 (Neue Freie Presse, 19 June
1866); NZ 30 June 1866, Mbl.; 1 July 1866, Mbl. (Duncker);
Rapp, Württemberger (note 22), 134 (manifesto of the People’s
Party of 2 Apr. 1866), 157 (Upper Swabia), 159 (Beobachter, 24
June 1866); Winkler, Liberalismus (note 10), 88 ff.; Gall,
Liberalismus (note 24), 352 ff.; Huber, Verfassungsgeschichte, iii
(note 2), 555 ff.; Sybel, Begründung (note 13), ii. 591 ff.; Die
nationalpolitische Publizistik von 1866 bis 1871. Eine kritische
Bibliographie, 2 vols., ed. Karl-Georg Faber (Düsseldorf: Droste,
1963), i. 13 ff.
27. Denkwürdigkeiten des Fürsten Chlodwig zu Hohenlohe-
Schillingsfürst, 2 vols. (Stuttgart: Deutsche Verlagsanstalt, 1907),
i. 168–9 (13 July 1866), tr. as Memoirs of Prince Chlodwig of
Hohenlohe-Schillingsfuerst, 2 vols., ed. George W. Chrystal (New
York: AMS Press Edition, 1970); Christa Stache, Bürgerlicher
Liberalismus und katholischer Konservatismus in Bayern 1867–1871.
Kulturkämpferische Auseinandersetzungen vor dem Hintergrund von
1010
nationaler Einigung und wirtschaftlich-sozialem Wandel (Frankfurt:
Lang, 1981), 42–3 (declaration of the Progressive party of 28
Aug. 1866); Schieder, Partei (note 25), 129 ff.; Gall, Liberalismus
(note 24), 376 ff.; Langewiesche, Liberalismus (note 22), 324 ff.;
Rapp, Württemberger (note 22), 184 ff.; Wehler,
Gesellschaftsgeschichte, iii (note 1), 283 ff. (civil war argument).
28. NZ 31 July 1866, Mbl.; 25 July 1866, Mbl. (emphases in
original); [Edmund Jörg], ‘Das deutsche Volk zwischen heute
und morgen’, Historisch-politische Blätter für das katholische
Deutschland [HPB], 58 (1866/II), 313–28 (324, 328); Adolf M.
Birke, Bischof Ketteler und der deutsche Liberalismus. Eine
Untersuchung über das Verhältnis des liberalen Katholizismus zum
bürgerlichen Liberalismus in der Reichsgründungszeit (Mainz:
Matthias-Grünewald-Verlag, 1971), 74 (Ketteler’s to Emperor
Franz Joseph of 28 Aug. 1866); Wilhelm Emmanuel von Ketteler,
Deutschland nach dem Kriege von 1866 (Mainz: F. Kirchheim,
1867), 82–4; Faber, ‘Realpolitik’ (note 4), 1 ff.
29. [Ernst Ludwig v. Gerlach], ‘Krieg und Bundesreform’, Neue
Preussische Zeitung (Kreuz-Zeitung), 8 May 1866; Kraus, Gerlach
(note 2), ii. 804 (quote from the pamphlet ‘Die Annexionen und
der Norddeutsche Bund’; emphases in original); Winkler,
Liberalismus (note 10), 99 ff.; Rapp, Württemberger (note 22), 190
(Österlen); August Bebel, Aus meinem Leben, (1911; repr. Bonn: J.
H. W. Dietz, 1997), 127–8 (platform of the Saxon People’s party),
tr. as My Life (New York: H. Fertig, 1973); Weber, Demokraten
(note 23), 237 ff.; Mayer, ‘Trennung’ (note 23), 129 ff.; Werner
Conze and Dieter Groh, Die Arbeiterbewegung in der nationalen
1011
Bewegung. Die deutsche Sozialdemokratie vor, während und nach der
Reichsgründung (Stuttgart, 1966); Michael Stürmer, Das ruhelose
Reich. Deutschland 1866–1918 (Berlin: Klett, 1983), 143 ff. Quote
from Jacoby in Sten. Ber., LT, 1866/67, 1, 72; on Klopp see
Lorenz Matzinger, Onno Klopp 1822–1903. Leben und Werk
(Aurich: Ostfriesische Landschaft, 1993), 78 ff.
30. MEW, xxxi. 240–1 (Engels to Marx, 25 July 1866); xxxvi. 238–9
(Engels to Bebel, 18 Nov. 1884; all emphases in the original);
MECW, xl. 297–8 (Engels to Marx, 25 July 1866); xlvii. 220–3
(Engels to Bebel, 18 Nov. 1884; all emphases in the original);
Bismarck, Werke (note 3), vi. 120 (Bismarck to Manteuffel, 11
Aug. 1866), viii. 459 (conversation with the writer Paul Lindau
and the bank director Löwenfeld on 8 Dec. 1882; here the quote
from the conversation with Napoleon III, probably Oct. 1864 or
1865); Hans Joachim Schoeps, Der Weg ins Deutsche Kaiserreich
(Berlin: Propyläen, 1970), 147–8 (Bluntschli, Hess); Sten. Ber.,
LT, 1866/67, III, 1299–300 (Twesten), II, 833 (Unruh); Jacob
Burckhardt, Über das Studium der Geschichte. Der Text der
‘Weltgeschichtlichen Betrachtungen’, ed. Peter Ganz (Munich: C. H.
Beck, 1982), 373 (emphases in original); Winkler, Liberalismus
(note 10), 106 ff.; id., ‘Primat’ (note 9), 52–3; Henry A. Kissinger,
‘Der weisse Revolutionär: Reflexionen über Bismarck’, in Das
Bismarck-Problem in der Geschichtsschreibung nach 1945, ed.
Lothar Gall (Cologne: Kiepenheuer & Witsch, 1971), 392–428;
Gustav Adolf Rein, Die Revolution in der Politik Bismarcks
(Göttingen: Musterschmidt, 1957), 325 ff.; Meyer, Bismarck (note
11), 328 ff.; Ernst Engelberg, ‘Über die Revolution von oben’,
Zeitschrift für Geschichtswissenschaft, 22 (1974), 1183–1212;
1012
Dieter Langewiesche, ‘“Revolution von oben”? Kriege und
Nationalstaatsgründung in Deutschland’, in Revolution und Krieg.
Zur Dynamik historischen Wandels seit dem 18. Jahrhundert, ed. id.
(Paderborn: Schöningh, 1989), 117–34.
31. Parisius, Parteien (note 10), 77; Huber, Verfassungsgeschichte, iii
(note 2), 352–3; Winkler, Liberalismus (note 10), 91 ff.; id.,
‘Bürgerliche Emanzipation und nationale Einigung. Zur
Entstehung des Nationalliberalismus in Preussen’, in Liberalismus
und Antiliberalismus. Studien zur politischen Sozialgeschichte des 19.
und 20. Jahrhunderts (Göttingen: Vandenhoeck & Ruprecht,
1979), 24–35.
32. Huber, Verfassungsgeschichte, iii (note 2), 305 ff. (argument about
the legitimacy of Bismarck’s policies, 1862–6); Hans Boldt,
‘Verfassungskonflikt und Verfassungshistorie. Eine
Auseinandersetzung mit Ernst Rudolf Huber’, in Der Staat, suppl.
1, Probleme des Konstitutionalismus im 19. Jahrhundert (Berlin:
Duncker & Humblot, 1975), 75–102; Ernst Wolfgang
Böckenförde, ‘Der Verfassungstyp der deutschen
konstitutionellen Monarchie im 19. Jahrhundert’, in Moderne
deutsche Verfassungsgeschichte (1815–1914), ed. id. (Cologne:
Kiepenheuer & Witsch, 1972), 146–70; Rainer Wahl, ‘Der
preussische Verfassungskonflikt und das konstitutionelle System
des Kaiserreichs’, ibid. 171–94. For Bismarck’s conversation with
Twesten on 1 Oct. 1862 see Bismarck, Werke (note 3), vii. 59–60.
33. Sten. Ber., LT, 1866/67, I, 151 (Waldeck, 1 Sept. 1866), 182 ff.
(Lasker, 3 Sept. 1866), 161 (Michaelis, 1 Sept. 1866), 198–9
(Twesten, 3 Sept. 1866), 72 (Virchow), 23 Aug. 1866), 252–3
1013
(Harkort, 7 Sept. 1866); NZ 19 Aug. 1866, Mbl.; 7 Nov. 1866,
Mbl. (Twesten); 4 Dec. 1866, Mbl. (Bamberger); Hermann
Baumgarten, ‘Der deutsche Liberalismus. Eine Selbstkritik’,
Preussische Jahrbücher, 18 (1866), 455–515 (470–1), 575–628
(596–7, 624–5, 627; emphases in original); repr. ed. Adolf M.
Birke (Berlin: Ullstein, 1974); Parisius, Parteien (note 10), 78 ff.
(announcement of 24 Oct. 1866); Burckhardt, Studium (note 30),
378; Carl Schmitt, Verfassungslehre (Berlin 1928), 31–2; id.,
Staatsgefüge und Zusammenbruch des zweiten Reiches. Der Sieg des
Bürgers über den Soldaten (Hamburg: Hanseatische Verlagsanstalt,
1934), 7 ff.; Winkler, Liberalismus (note 10), 113 ff.; Biefang,
‘National-preussisch’ (note 21), 360 ff.
34. Heinrich von Treitschke, ‘Bundesstaat und Einheitsstaat’ (1864),
in Historische und politische Aufsätze, 6th edn (Leipzig: Hirzel,
1903), 77–241; Gerhard Ritter, Die preussischen Konservativen und
Bismarcks deutsche Politik 1858 bis 1876 (Heidelberg: C. Winter,
1913), 147 ff.; Klaus-Erich Pollmann, Parlamentarismus im
Norddeutschen Bund, 1867–1870 (Düsseldorf: Droste, 1985), 93 ff.
(fraction strength: 171); Andreas Biefang, ‘Modernität wider
Willen. Bemerkungen zur Entstehung des demokratischen
Wahlrechts des Kaiserreichs’, in Gestaltungskraft des Politischen.
Festschrift f. Eberhard Kolb, ed. Wolfram Pyta and Ludwig Richter
(Berlin: Duncker & Humblot, 1998), 239–59; Otto Becker,
Bismarcks Ringen um Deutschlands Gestaltung (Heidelberg: Quelle
& Meyer, 1958), 211 ff.; Wolfgang J. Mommsen, ‘Das deutsche
Kaiserreich als System umgangener Entscheidungen’, in Der
autoritäre Nationalstaat. Verfassung, Gesellschaft und Kultur im
deutschen Kaiserreich (Frankfurt: Fischer Taschenbuch, 1990), 11–
1014
38, tr. as Imperial Germany 1867–1918: Politics, Culture, and
Society in an Authoritarian State (London: Edward Arnold; New
York: St Martin’s Press, 1995); id., ‘Die Verfassung des Deutschen
Reiches von 1871 als dilatorischer Herrschaftskompromiss’, ibid.
39–65; Huber, Verfassungsgeschichte, iii (note 2), 629 ff., 643 ff.
(659); Böhme, Weg (note 3), 236 ff.; Schieder, ‘Vom Deutschen
Bund’ (note 19), 198 ff.
35. Hans Georg Aschoff, Welfische Bewegung und politischer
Katholizismus 1866 bis 1918. Die Deutsch-Hannoversche Partei und
das Zentrum in der Provinz Hannover während des Kaiserreiches
(Düsseldorf: Droste, 1987), 19 ff.; Schieder, ‘Scheitern’ (note 15),
31 ff.; Mayer, ‘Trennung’ (note 23), 138 ff.; Conze and Groh,
Arbeiterbewegung (note 29), 78 ff.; Runge, Volkspartei (note 22),
115 ff.; Rapp, Württemberger (note 22), 289 ff.; Gall, Liberalismus
(note 24), 376 ff.; Becker, Staat (note 24), 201 ff.; Stache,
Liberalismus (note 27), 45 ff.; Schieder, Partei (note 25), 193 ff.;
Friedrich Hartmannsgruber, Die Bayerische Patriotenpartei 1868–
1887 (Munich: C. H. Beck, 1986), 33 ff.; Rolf Wilhelm, Das
Verhältnis der süddeutschen Staaten zum Norddeutschen Bund
(1867–1870) (Husum: Matthiesen, 1978), 97 ff.; Lothar Gall,
‘Bismarcks Süddeutschlandpolitik 1866–1870’, in Europa vor dem
Krieg von 1870. Mächtekonstellation—Konfliktfelder—
Kriegsausbruch, ed. Eberhard Kolb (Munich: Oldenbourg, 1987),
23–32.
36. Pollmann, Parlamentarismus (note 34), 259 ff., 433 ff.; Wilhelm,
Verhältnis (note 35), 103 ff. (quote from the NZ of 22 Feb. 1870:
103–4), 183–4 (Bismarck’s conversation with Spitzemberg of 27
1015
Mar. 1870); Meyer, Bismarck (note 11), 379–80; Gall, Bismarck
(note 3), 466 ff.
37. ‘Die confessionelle Leidenschaft im Ruine Deutschlands’, HPB 58
(1866/II), 781–96 (783–4: quote from the Protestantischen
Kirchenzeitung); Sten. Ber., LT, 1873, I, 631 (Virchow, 17 Jan.
1873); Georg Franz, Kulturkampf. Staat und katholische Kirche in
Mitteleuropa von der Säkularisation bis zum Abschluss des
preussischen Kulturkampfes (Munich: D. W. Callwey, 1954), 61 ff.
(89: Antonelli quote); Adam Wandruszka, Schicksalsjahr 1866
(Graz: Verlag Styria, 1966), 13 ff.; Karl Heinrich Höfele,
‘Königgrätz und die Deutschen von 1866’, GWU 17 (1966), 393–
416; Rudolf Lill, ‘Italien im Zeitalter des Risorgimento (1815–
1870)’, in Europa von der Französischen Revolution zu den
nationalstaatlichen Bewegungen des 19. Jahrhunderts (= Handbuch
der europäischen Geschichte, ed. Theodor Schieder, v) (Stuttgart:
Klett-Cotta, 1981), 827–85 (879 ff.); id., ‘Katholizismus und
Nation bis zur Reichsgründung’, in Katholizismus, nationaler
Gedanke und Europa seit 1800, ed. Albrecht Langner (Paderborn:
Schöningh, 1985), 51–64; Thomas Mergel, Zwischen Klasse und
Konfession. Katholisches Bürgertum im Rheinland 1794–1848
(Göttingen: Vandenhoeck & Ruprecht, 1994), 263 ff.; Ernst
Rudolf Huber, Deutsche Verfassungsgeschichte seit 1789, iv.
Struktur und Krisen des Kaiserreichs, 2nd edn (Stuttgart:
Kohlhammer, 1969), 645 ff.; Becker, Staat (note 24), 201 ff.;
Stache, Liberalismus (note 27), 162 ff. On the German Protestant
Association, founded in 1863, see Claudia Lepp, Protestantisch-
liberaler Aufbruch in die Moderne. Der Deutsche Protestantenverein
in der Zeit der Reichsgründung (Gütersloh: Chr. Kaiser, 1996), esp.
1016
298 ff.
38. Bismarck, Werke (note 3), xiii 468 (address to the delegation
from the University of Jenaon, 30 July 1892); Gall, Bismarck
(note 3), 417 ff. (431: Gramont); Otto Pflanze, Bismarck. Der
Reichsgründer (Munich: C. H. Beck, 1997), 449 ff., orig. Bismarck
and the Development of Germany: The Period of Unification, 1815–
1871 (Princeton: Princeton University Press, 1963); Huber,
Verfassungsgeschichte, iii (note 2), 702 ff.; Jochen Dittrich,
‘Ursache und Ausbruch des deutsch-französischen Krieges’, in
Reichsgründung 1870/71. Tatsachen, Kontroversen, Interpretationen,
ed. Theodor Schieder and Ernst Deuerlein (Stuttgart: Seewald,
1970), 64–94; Eberhard Kolb, Der Kriegsausbruch 1870
(Göttingen: Vandenhoeck & Ruprecht, 1970), 143 ff. (argument
about France’s sole responsibility for the outbreak of the war);
Europa (note 35), ed. id.; Josef Becker, ‘Zum Problem der
Bismarckschen Politik in der spanischen Thronfrage 1870’, HZ
212 (1971), 529–607; id., ‘Von Bismarcks “spanischer Diversion”
zur “Emser Legende” des Reichsgründers’, in Johannes Burkhardt
et al., Lange und kurze Wege in den Ersten Weltkrieg. Vier
Augsburger Beiträge zur Kriegsursachenforschung (Munich: Ernst
Vögel, 1996), 87–113 (argument about Bismarck’s war plans
from the beginning of the Spanish succession question)
39. MEW, xvii. 3–7 (5: ‘Erste Adresse des Generalrats’); Conze and
Groh, Arbeiterbewegung (note 29), 86 ff.; Rapp, Württemberger
(note 22), 364 (Reutlinger Neue Bürgerzeitung), 383 (Mayer), 396
(flags); Runge, Volkspartei (note 22), 161 ff.; Hans Fenske, Der
Weg zur Reichsgründung 1850–1870 (= Quellen zum politischen
1017
Denken der Deutschen im 19. u. 20. Jahrhundert. Freiherr vom Stein-
Gedächtnisausgabe, v) (Darmstadt: Wissenschaftliche
Buchgesellschaft, 1977), 419–21 (Jörg, 19 July 1870), 421–4
(Sepp, 19 July 1870); Huber, Verfassungsgeschichte, iii (note 2),
724 (vote in Bavarian parliament); Hartmannsgruber,
Patriotenpartei (note 35), 362 ff.; Lothar Gall, ‘Zur Frage der
Annexion von Elsass und Lothringen 1870’, HZ 206 (1968), 265–
326; id., ‘Das Problem Elsass-Lothringen’, in Reichsgründung (note
38), ed. Schieder and Deuerlein, 366–85; Rudolf Buchner, ‘Die
deutsche patriotische Dichtung vom Kriegsbeginn 1870 über
Frankreich und die elsässische Frage’, ibid. 327–36; Heinrich von
Treitschke, ‘Was fordern wir von Frankreich?’ (manuscript
completed on 30 Aug. 1870), Preussische Jahrbücher, 26 (1870),
367–409 (371, 380, 406); Eberhard Kolb, Der Weg aus dem Krieg.
Bismarcks Politik im Krieg und die Friedensanbahnung 1870/71
(Munich: Oldenbourg, 1989), 113 ff.
40. Paul Piechowski, Die Kriegspredigt von 1870/71 (Leipzig: Scholl,
1917), esp. 78 ff.; Hasko Zimmer, Auf dem Altar des Vaterlandes.
Religion und Patriotismus in der deutschen Kriegslyrik des 19.
Jahrhunderts (Frankfurt: Thesen Verlag, 1971), 71 ff.; Michael
Jeismann, Das Vaterland der Feinde. Studien zum nationalen
Feindbegriff und Selbstverständnis in Deutschland und Frankreich
1792–1918 (Stuttgart: Klett-Cotta, 1992), 161 ff., 246
(Rittershaus), 267 (Leipzig theatre prologue), 276 (Vossische
Zeitung, 24 July 1870); Marianne und Germania 1789–1889.
Frankreich und Deutschland. Zwei Welten—Eine Revue, ed. Marie-
Louise von Plessen (Berlin: Argon, 1996), 405–48; Kolb, Weg
(note 39), 1 ff., 195 ff.; Klaus Hildebrand, No intervention. Die Pax
1018
Britannica und Preussen 1865/66–1869/70. Eine Untersuchung zur
englischen Weltpolitik im 19. Jahrhundert (Munich: Oldenbourg,
1997); Ernst Moritz Arndt, ‘Ein Wort über die Feier der Leipziger
Schlacht’ (1814), in Werke (Berlin: Deutsches Verlagshaus Bong,
n.d.), 131–42 (133); Theodor Schieder, Das deutsche Kaiserreich
von 1871 als Nationalstaat (Cologne: Westdeutscher Verlag,
1961), 72 ff., 125–53 (Bodelschwingh’s speech on 27 June 1871:
135–45); Hartmut Lehmann, ‘Friedrich von Bodelschwingh und
das Sedanfest’, HZ 202 (1966), 542–73; Claudia Lepp,
‘Protestanten feiern ihre Nation—Die kulturprotestantischen
Ursprünge des Sedantages’, Historisches Jahrbuch, 118 (1998),
201–22; MEW xvii. 268–70 (269: Marx and Engel’s letter to the
committee of the Social Democratic Workers’ party, written
between 22 and 30 Aug. 1870), 271–9 (275: ‘Zweite Adresse des
Generalrats’, 9 Sept. 1870; all emphases in the original), tr.
MECW, xxii. 260–2 (Marx and Engels’s letter to the committee of
the Social Democratic Workers’ party), 263–70 (267: ‘Second
Address of the General Council’, 9 Sept. 1870); Stenographischer
Bericht über die Verhandlungen des Reichstags des Norddeutschen
Bundes, 1870, II. ausserordentliche Session, 187–8 (Lasker’s
motion); Neue Preussische Zeitung (Kreuz-Zeitung), 10 Feb. 1870;
Ritter, Konservative (note 34), 355; Hartmannsgruber,
Patriotenpartei (note 35), 368 (Bayerisches Vaterland, 13 Jan.
1871); Sturmjahre (note 21), 494 (Sybel to Baumgarten, 27 Jan.
1871); Elisabeth Fehrenbach, Wandlungen des deutschen
Kaisergedankens 1871–1918 (Munich: Oldenbourg, 1969), 14 ff.;
Schoeps, Weg (note 30), 189 ff.; Huber, Verfassungsgeschichte, iii
(note 2), 724 ff.
1019
CHAPTER 5. THE TRANSFORMATION OF
NATIONALISM 1871–1890
1. Bismarck im Urteil der Zeitgenossen und der Nachwelt, 2nd edn, ed.
Walter Bussmann (Stuttgart: Klett, 1956), 28 (Disraeli); Hansard’s
Parliamentary Debates, Third Series, cciv (London: Cornelius
Buck, 1871), 81–2; Ludwig Dehio, ‘Deutschland und die Epoche
der Weltkriege’ (1951), in Deutschland und die Weltpolitik im 20.
Jahrhundert (Munich: Oldenbourg, 1955), 9–36 (15); Eberhard
Kolb, ‘Der Pariser Commune-Aufstand und die Beendigung des
deutsch-französischen Krieges’, in Umbrüche deutscher Geschichte.
1866/71. 1918/19. 1929/33. Ausgewählte Aufsätze (Munich:
Oldenbourg, 1993), 163–88; id., ‘Kriegsniederlage und
Revolution: Pariser Commune 1871’, ibid. 189–206.
2. Günter Brakelmann, ‘Der Krieg von 1870/71 und die
Reichsgründung im Urteil des Protestantismus’, in Kirche
zwischen Krieg und Frieden. Studien zur Geschichte des deutschen
Protestantismus, ed. Wolfgang Huber and Johannes Schwerdtfeger
(Stuttgart: Klett, 1976), 293–320 (303: Neue Evangelische
Kirchenzeitung [NEKZ], 7 Jan. 1871; 304: NEKZ 18 Mar. 1871;
306: quotes on the French as ‘satanic’ people and ‘Antichrist’;
307–8: quotes from the Evangelischen Kirchenzeitung, 1871);
Walter Frank, Hofprediger Adolf Stoecker und die christlichsoziale
Bewegung (Berlin: R. Hobbing, 1928), 32–3 (Stoecker, 27 Jan.
1871); Fürst Otto von Bismarck, Die gesammelten Werke
(Friedrichsruh edition) (Berlin: O. Stollberg, 1924–), xi. 256–7
(speech on 6 Mar. 1872), xiii. 475 (speech on 31 July 1892 in
Jena); Theodor Schieder, Das deutsche Kaiserreich von 1871 als
1020
Nationalstaat (Cologne: Westdeutscher Verlag, 1961), 175
(Bennigsen, 26 Jan. 1881); Elisabeth Fehrenbach, Wandlungen
des deutschen Kaisergedankens 1871–1918 (Munich: Oldenbourg,
1969), 14 ff.; Helmuth Walser Smith, German Nationalism and
Religious Conflict: Culture, Ideology, Politics, 1870–1914
(Princeton: Princeton University Press, 1995); Horst Dippel,
‘1871 versus 1789: German Historians and the Ideological
Foundations of the Deutsche Reich’, History of European Ideas, 16
(1993), 829–37; Dieter Langewiesche, ‘Deutschland und
Österreich: Nationswerdung und Staatsbildung in Mitteleuropa
im 19. Jahrhundert’, GWU 42 (1991), 754–66; Adam
Wandruszka, ‘Grossdeutsche und kleindeutsche Ideologie 1840–
1871’, in Deutschland und Österreich. Ein bilaterales Geschichtsbuch,
ed. Robert A. Kann and Friedrich Prinz (Vienna: Jugend und
Volk, 1980), 110–42; Karl Heinrich Höfele, ‘Sendungsglaube und
Epochenbewusstsein in Deutschland 1870/71’, Zeitschrift für
Religionsund Geistesgeschichte, 15 (1963), 265–76.
3. Bismarck, Werke (note 2), xv. Erinnerung und Gedanke, 199;
Manfred Hanisch, ‘Nationalisierung der Dynastien oder
Monarchisierung der Nation? Zum Verhältnis von Monarchie und
Nation in Deutschland im 19. Jahrhundert’, in Bürgertum, Adel
und Monarchie. Wandel der Lebensformen im Zeitalter des
bürgerlichen Nationalismus, ed. Adolf M. Birke and Lothar
Kettenacker (Munich: K. G. Saur, 1989), 71–91; id., Für Fürst und
Vaterland. Legitimitätsstiftung in Bayern zwischen Revolution 1848
und deutscher Einheit (Munich: Oldenbourg, 1991), 20 ff.;
Siegfried Weichlein, ‘Sachsen zwischen Landesbewusstsein und
Nationsbildung 1866–1871’, in Sachsen im Kaiserreich—Politik,
1021
Wirtschaft und Gesellschaft im Umbruch, ed. Simone Lässig and
Karl Heinrich Pohl (Dresden: Böhlau, 1997), 241–70; Gerhard
Ritter, Die preussischen Konservativen und Bismarcks deutsche
Politik 1858–1876 (Heidelberg: C. Winter, 1913), 359 (Kleist-
Retzow); Hans Booms, Die Deutschkonservative Partei. Preussischer
Charakter, Reichsauffassung, Nationalbegriff (Düsseldorf: Droste,
1954), 5 ff.; James N. Retallack, Notables of the Right. The
Conservative Party and Political Mobilization in Germany, 1876–
1918 (London: Unwin Hyman, 1988), 13 ff.; Oliver Janz, Bürger
besonderer Art. Evangelische Pfarrer in Preussen 1850–1914 (Berlin:
Walter de Gruyter, 1994), esp. 58 ff.
4. Christoph Weber, ‘Eine starke, enggeschlossene Phalanx’. Der
politische Katholizismus und die erste deutsche Reichstagswahl 1871
(Essen: Klartext, 1992), 54 ff.; (Edmund Jörg,) ‘Das grosse
Neujahr’, HPB 67 (1871), 1–15 (6); Wilhelm Emmanuel von
Ketteler, Die Katholiken im Deutschen Reiche. Entwurf zu einem
politischen Programm (Mainz: F, Kirchheim, 1873), 5; Adolf M.
Birke, Bischof Ketteler und der deutsche Liberalismus. Eine
Untersuchung über das Verhältnis des liberalen Katholizismus zum
bürgerlichen Liberalismus in der Reichsgründungszeit (Mainz:
Matthias-Grünewald-Verlag, 1971), 78 ff.; Rudolf Lill, ‘Die
deutschen Katholiken und Bismarcks Reichsgründung’, in
Reichsgründung 1870/71. Tatsachen, Kontroversen, Interpretationen,
ed. Theodor Schieder and Ernst Deuerlein (Stuttgart: Seewald,
1970), 345–65; Schieder, Kaiserreich (note 2), 125 ff. (152:
Ketteler, 19 Aug. 1874); Fehrenbach, Wandlungen (note 2), 20 ff.
5. Brakelmann, ‘Krieg’ (note 2), 318 (Wichern); Stenographischer
1022
Bericht über die Verhandlungen des Deutschen Reichstags [Sten.
Ber.], xx. 920–1 (Bebel, 25 May 1871); August Bebel, Aus meinem
Leben (1910–14; Bonn: J. H. W. Dietz, 1997), 299 ff., tr. as My
Life (New York: H. Fertig, 1973); Karl Marx, ‘Der Bürgerkrieg in
Frankreich’ (1871), MEW, xvii. 313–65; Friedrich Engels,
‘Einleitung zu Karl Marx Bürgerkrieg in Frankreich’ (1891), MEW,
xx. 188–99 (199), tr. as Marx, ‘The Civil War in France’, MECW,
xx. 307–57; Engels, ‘Introduction to Karl Marx’s The Civil War in
France’, MECW, xxvii. 179–91 (190–1); Dieter Groh and Peter
Brandt, ‘Vaterlandslose Gesellen’. Sozialdemokratie und Nation
1860–1990 (Munich: C. H. Beck, 1992), 26 ff. On the ‘inner
Düppel’ see Lothar Gall, Bismarck. Der weisse Revolutionär (Berlin:
Propyläen, 1980), 331.
6. Franz Mehring, Geschichte der deutschen Sozialdemokratie, 1st edn
(1897–8), 2 vols., ii. Von Lassalles ‘Offenem Antwortschreiben’ bis
zum Erfurter Programm 1863 bis 1891 (Berlin: J. H. W. Dietz,
1960), 378, 434–5 (Jacoby); Ernest Renan, Was ist eine Nation?
Und andere politische Schriften, ed. Walter Euchner (Vienna:
1995), 57 (Paris address ‘What is a nation?’, 11 Mar. 1882), 59
(the Franco-German war [Sept. 1870]), 118 (Strauss to Renan,
29 Sept. 1870), 131–2 (Renan to Strauss, 15 Sept. 1871); id.,
Qu’est-ce qu’une nation?/What is a Nation? (Toronto: Tapir Press,
1996); Jörg, ‘Neujahr’ (note 4), 9 (the German Reich as ‘pure’
but ‘incomplete nation-state’); Helmuth Plessner, Die verspätete
Nation. Über die politische Verführbarkeit bürgerlichen Geistes (1st
edn under the title Das Schicksal deutschen Geistes im Ausgang
seiner bürgerlichen Epoche (Zurich, 1935)) (Stuttgart:
Kohlhammer, 1959), 39.
1023
7. Sten. Ber., xxiv. 356 (Bismarck, 14 May 1872); Sten. Ber., LT,
1873, I, 631 (Virchow, 17 Jan. 1873); Bismarck, Werke (note 2),
xv. 333 (Erinnerung und Gedanke); Der Kulturkampf in Italien und
in den deutschsprachigen Ländern, ed. Rudolf Lill and Francesco
Traniello (Berlin: Duncker & Humblot, 1993); Winfried Becker,
‘Der Kulturkampf als europäisches und deutsches Phänomen’,
Historisches Jahrbuch, 101 (1981), 422–46; id., ‘Liberale
Kulturkampf-Positionen und politischer Katholizismus’, in
Innenpolitische Probleme des Bismarck-Reiches, ed. Otto Pflanze
(Munich: Oldenbourg, 1983), 47–72; Winfried Grohs, Die Liberale
Reichspartei. 1871–1874. Liberale Katholiken und föderalistische
Protestanten im ersten Deutschen Reichstag (Frankfurt: Peter Lang,
1990), 26 ff.; Weber, Phalanx (note 4), 42 ff.; Ritter, Konservative
(note 3), 361 ff.; Claudia Lepp, Protestantisch-liberaler Aufbruch in
die Moderne. Der Deutsche Protestantenverein in der Zeit der
Reichsgründung und des Kulturkampfes (Gütersloh: Chr. Kaiser,
1996), 319 ff.; Ernst Rudolf Huber, Deutsche Verfassungsgeschichte
seit 1789, iv. Strukturen und Krisen des Kaiserreichs, 2nd edn
(Stuttgart: Kohlhammer, 1969), 672 ff.
8. NZ 21 Oct. 1876, Mbl.; NZ 2 Sept. 1877, Mbl.; Thomas Mergel,
Zwischen Klasse und Konfession. Katholisches Bürgertum im
Rheinland 1794–1914 (Göttingen: Vandenhoeck & Ruprecht,
1994), 235 ff.; Gottfried Korff, ‘Kulturkampf und
Volksfrömmigkeit’, in Volksreligiosität in der modernen
Sozialgeschichte, ed. Wolfgang Schieder, GG, Sonderheft 11
(Göttingen: Vandenhoeck & Ruprecht, 1986), 137–51; Margaret
L. Anderson and Kenneth Barkin, ‘The Myth of the Puttkamer
Purge and the Reality of the Kulturkampf’, Journal of Modern
1024
History, 54 (1982), 647–86; Margaret L. Anderson, ‘The
Kulturkampf and the Course of German History’, Central
European History, 19 (1986), 82–115; ead., Windthorst.
Zentrumspolitiker und Gegenspieler Bismarcks (Düsseldorf: Droste,
1988), 130 ff., orig. Windthorst: A Political Biography (Oxford:
Clarendon Press, 1981); 130 ff.; Jonathan Sperber, Popular
Catholicism in Nineteenth-Century Germany (Princeton: Princeton
University Press, 1984), 207 ff.; David Blackbourn, ‘Progress and
Piety: Liberalism, Catholicism, and the State in Imperial
Germany’, History Workshops Journal, 26 (1988), 57–78; id.,
Wenn ihr sie wieder seht, fragt, wer sie sei. Marienerscheinungen in
Marpingen—Aufstieg und Niedergang des deutschen Lourdes
(Reinbek: Rowohlt, 1997), orig. Marpingen: Apparitions of the
Virgin Mary in Bismarckian Germany (Oxford: Clarendon Press,
1993); Gustav Schmidt, ‘Die Nationalliberalen—eine
regierungsfähige Partei? Zur Problematik der inneren
Reichsgründung 1870–1878’, in Die deutschen Parteien vor 1918,
ed. Gerhard A. Ritter (Cologne: Kiepenheuer & Witsch, 1973),
208–23; Hans-Ulrich Wehler, Deutsche Gesellschaftsgeschichte, iii.
Von der ‘Deutschen Doppelrevolution’ bis zum Beginn des Ersten
Weltkrieges 1849–1914 (Munich: C. H. Beck, 1995), 892 ff.
9. Heinrich Heine, ‘Gedanken und Einfälle’, in Sämtliche Werke in
zwölf Bänden (Leipzig: Hesse & Becker, n.d.), xii. Vermischte
Schriften, 3–50 (9); Richard Wagner, ‘Das Judentum in der Musik’
(1850), in Gesammelte Schriften und Dichtungen, v, 2nd edn
(Leipzig: E. W. Fritsch, 1888), 66–85 (68–9); tr. as Judaism in
Music (London: W. Reeves, 1910); id., ‘Was ist deutsch?’ (1865
and 1878), ibid. 9 (Berlin 1913), 36–53 (50–1); Sten. Ber., LT,
1025
1873, I, 937–51 (Lasker, 7 Feb. 1873); (Franz Perrot), ‘Die Ära
Bleichröder-Delbrück-Camphausen und die neudeutsche
Wirtschaftspolitik’, Neue Preussische Zeitung (Kreuz-Zeitung), 29
June 1873; Otto Glagau, Deutsches Handwerk und historisches
Bürgertum (Osnabrück: B. Wehberg, 1879), 80; id., Der Börsenund
Gründungs-Schwindel in Berlin, 2 vols. (Leipzig: P. Frohberg,
1876–7); ‘Milliarden-Noth und Krach-Segen’, HPB 74 (1874),
963–76 (976: cosmopolitan powers of finance); [Edmund Jörg],
‘Vor fünfundzwanzig Jahren’, HPB, 79 (1877), 1–17 (10: The
Kulturkampf as a tool of the stock market); id., ‘Das “Gründer”-
Unwesen mit Staatshülfe’, ibid. 237–52 (239: ‘Jewified
liberalism’); Mergel, Klasse (note 8), 258–9 (Catholic Press);
Birke, Ketteler (note 4), 92 (Ketteler, 1873); Gordon A. Craig,
Deutsche Geschichte 1866–1945. Vom Norddeutschen Bund bis zum
Ende des Dritten Reiches (Munich: C. H. Beck, 1980), 85 (Freytag,
Raabe, Dahn), orig. Germany, 1866–1945 (Oxford: Oxford
University Press, 1978), George L. Mosse, ‘The Image of the Jew
in German Political Culture: Felix Dahn and Gustav Freytag’,
Publications of the Leo Baeck Institute of Jews from Germany.
Year Book 2 (1957), 218–27; id., Germans and Jews: The Right, the
Left and the Search for a ‘Third Force’ in Pre-Nazi Germany
(London: Orbach & Chambers, 1971), 34 ff.; Hans Rosenberg,
Grosse Depression und Bismarckzeit. Wirtschaftsablauf, Gesellschaft
und Politik in Mitteleuropa (Berlin: de Gruyter, 1967), 22 ff. (29);
James F. Harris, The People Speak! Anti-Semitism and Emancipation
in Nineteenth-Century Bavaria (Ann Arbor: University of Michigan
Press, 1994), esp. 209 ff.; Olaf Blaschke, Katholizismus und
Antisemitismus im deutschen Kaiserreich (Göttingen: Vandenhoeck
1026
& Ruprecht, 1997), 42 ff.; Fritz Stern, Gold und Eisen. Bismarck
und sein Bankier Bleichröder (Reinbek: Rowohlt, 1988), 680 ff.
(Germania article: 682 ff.) orig. Gold and Iron: Bismarck,
Bleichröder, and the Building of the German Empire, 2nd edn (New
York: Vintage Books, 1979); Erik Lindner, Patriotismus deutscher
Juden von der napoleonischen Ära bis zum Kaiserreich. Zwischen
korporativem Loyalismus und individueller deutsch-jüdischer Identität
(Frankfurt: Lang, 1997), 267 ff.; A. Sartorius von Waltershausen,
Deutsche Wirtschaftsgeschichte 1815–1914, 2nd edn (Jena: G.
Fischer, 1923), 275 ff. (statistics of the economic crisis).
10. Carl Wilmanns, Die ‘goldene’ Internationale und die Nothwendigkeit
einer socialen Reformpartei (Berlin: Niendorf, 1876); Constantin
Frantz, Der National-Liberalismus und die Judenherrschaft (Munich:
Literarische Institut von M. Huttler, 1874), 49 (Fichte), 64
(quote); id., Literarisch-politische Aufsätze (Munich: Literarische
Institut von M. Huttler, 1876), p. xvii (foreword; all emphases in
the original); Wilhelm Marr, Der Sieg des Judenthums über das
Germanenthum. Vom nicht confessionellen Standpunkt aus betrachtet
(Bern: Rudolph Costenoble, 1879), 3, 46; Helmut Berding,
Moderner Antisemitismus in Deutschland (Frankfurt: Suhrkamp,
1988), 86 ff.; Reinhard Rürup, Emanzipation und Antisemitismus.
Studien zur ‘Judenfrage’ der bürgerlichen Gesellschaft (Göttingen:
Vandenhoeck & Ruprecht, 1975), 74 ff., 101 ff. (Marr, Frantz,
Dühring); Jacob Katz, Vom Vorurteil bis zur Vernichtung. Der
Antisemitismus 1700–1933 (Munich: C. H. Beck, 1989), 253 ff.
(also on the reception of Rohling), orig. From Prejudice to
Destruction: Anti-Semitism, 1700–1933 (Cambridge, MA: Harvard
University Press, 1980); Paul W. Massing, Vorgeschichte des
1027
politischen Antisemitismus (Frankfurt: Europäische Verlagsanstalt,
1959), 5 ff., orig. Rehearsal for Destruction: A Study of Political
Anti-Semitism in Imperial Germany (New York: Harper, 1949);
Peter G. J. Pulzer, Die Entstehung des politischen Antisemitismus in
Deutschland und Österreich 1867–1914 (Gütersloh: S. Mohn,
1966), orig. The Rise of Political Anti-Semitism in Germany and
Austria (New York: Wiley, 1964); John Weiss, Der lange Weg zum
Holocaust. Die Geschichte der Judenfeindschaft in Deutschland und
Österreich (Hamburg, Hoffmann & Campe, 1997), 118 ff., orig.
Ideology of Death: Why the Holocaust Happened in Germany
(Chicago: I. R. Dee, 1996); Rainer Erb and Werner Bergmann, Die
Nachtseite der Judenemanzipation. Der Widerstand gegen die
Integration der Juden in Deutschland 1780–1860 (Berlin: Metropol,
1989), 97 ff.; Shulamit Volkov, The Rise of Popular Antimodernism
in Germany: The Urban Master Artisans 1873–1896 (Princeton:
Princeton University Press, 1978), 215 ff.; ead., ‘Antisemitismus
als kultureller Code’, in, Jüdisches Leben und Antisemitismus im 19.
u. 20. Jahrhundert (Munich: C. H. Beck, 1990), 13–36; Blaschke,
Katholizismus (note 9), 48 ff., 74 ff. (on Rohling); Thomas
Nipperdey, Deutsche Geschichte 1866–1918, i. Arbeitswelt und
Bürgergeist (Munich: C. H. Beck, 1990), 396 ff.; ii. Machtstaat vor
der Demokratie (Munich: C. H. Beck, 1992), 289 ff.; Wehler,
Gesellschaftsgeschichte, iii (note 8), 924 ff. For Fichte’s views on
the Jews, see above p. 60.
11. Adolf Stoecker, ‘Unsere Forderungen an das moderne Judentum’
(19 Sept. 1879), in Christlich-Sozial. Reden und Aufsätze, 2nd edn
(Berlin: Buchhandlung der Berliner Stadtmission, 1890), 359–69
(359–60, 367 ff.); id., ‘Notwehr gegen das moderne Judentum’
1028
(26 Sept. 1879), ibid., 369–82 (381); id., ‘Die Selbstverteidigung
des modernen Judentums in dem Geisteskampf der Gegenwart’
(5 Jan. 1880), ibid. 382–89 (385: on Treitschke); Frank, Stoecker
(note 2), 88 ff.; Heinrich von Treitschke, ‘Unsere Aussichten’
(Nov. 1879), in Der Berliner Antisemitismusstreit, ed. Walter
Boehlich (Frankfurt: Insel, 1965), 5–12 (6 ff., 11); Manuel Joël,
‘Offener Brief an Heinrich von Treitschke’ (1879), 13–25 (23: on
Lasker’s defeat); Hermann Cohen, ‘Ein Bekenntnis in der
Judenfrage’ (Jan. 1880), ibid. 124–49; Ludwig Bamberger,
‘Deutschthum und Judenthum’ (1880), ibid. 149–79 (165 ff.);
‘Erklärung’ (by Forckenbeck et al., 12 Nov. 1880), ibid. 202–4;
Theodor Mommsen, ‘Ein Brief an die Nationalzeitung’ (19 Nov.
1880), ibid. 208–9; id., ‘Auch ein Wort über unser Judenthum’
(1880), ibid. 210–25 (208, 220, 224); Hans-Michael Bernhardt,
‘“Die Juden sind unser Unglück!” Strukturen eines Feindbildes im
deutschen Kaiserreich’, in Feindbilder in der deutschen Geschichte.
Studien zur Vorurteilsgeschichte im 19. u. 20. Jahrhundert, ed.
Christoph Jahr et al. (Berlin: Metropol, 1994), 25–54; John C. G.
Röhl, Wilhelm II. Die Jugend des Kaisers 1859–1888 (Munich: C.
H. Beck, 1993), 414–15 (Crown Prince Frederick on anti-
Semitism); Katz, Vorurteil (note 10), 217 ff.; Günter Brakelmann,
Protestantismus und Politik. Werk und Wirken Adolf Stoeckers
(Hamburg: Christians, 1982); Grit Koch, Adolf Stoecker 1835–
1909. Ein Leben zwischen Politik und Kirche (Erlangen: Palm &
Enke, 1993).
12. Heinrich Heine, Zur Geschichte der Religion und Philosophie in
Deutschland (1834), in Werke (note 9), viii. Über Deutschland, I,
3–122 (120–1); Eugen Dühring, Die Judenfrage als Frage der
1029
Racenschädlichkeit für Existenz, Sitte und Cultur der Völker (1st edn
under the title Die Judenfrage als Racen-, Sitten- und Culturfrage
(1880)), 3rd edn (Karlsruhe: H. Reuther, 1886), 159; Paul de
Lagarde, ‘Juden und Indogermanen’ (1887), in Ausgewählte
Schriften (Munich: Lehmann, 1924), 195–216 (209); Bamberger,
‘Deutschthum’ (note 11), 176–7; Frank, Stoecker (note 2), 118
(anti-Semite petition); Katz, Vorurteil (note 10), 275 ff.; Berding,
Antisemitismus (note 10), 99 ff.; Gerd-Klaus Kaltenbrunner, ‘Vom
Konkurrenten des Karl Marx zum Vorläufer Hitlers’, in Propheten
des Nationalismus, ed. Karl Schwedhelm (Munich: List, 1969), 36–
55; Fritz Stern, Kulturpessimismus als politische Gefahr. Eine
Analyse nationaler Ideologie in Deutschland (Bern: Scherz, 1963),
orig. The Politics of Cultural Despair: A Study in the Rise of the
Germanic Ideology (Berkeley: University of California Press,
1961); Kurt Wawrzinek, Die Entstehung der deutschen
Antisemitenparteien (1873–1890) (Berlin: E. Ebering, 1927);
Stefan Scheil, Die Entwicklung des politischen Antisemitismus in
Deutschland zwischen 1881 und 1912. Eine wahlgeschichtliche
Untersuchung (Berlin: Duncker & Humblot, 1999); Norbert
Kampe, Studenten und ‘Judenfrage’ im Deutschen Kaiserreich. Die
Entstehung einer akademischen Trägerschicht des Antisemitismus
(Göttingen: Vandenhoeck & Ruprecht, 1988), 23 ff.; Plessner,
Nation (note 6). For the quote from the Antisemitische
Correspondenz, see the article ‘Antisemitische Parteien 1879–
1914’, in Lexikon zur Parteiengeschichte. Die bürgerlichen und
kleinbürgerlichen Parteien und Verbände in Deutschland (1789–
1945), ed. Dieter Fricke et al., 4 vols. (Leipzig: VEB
Bibliographisches Institut Leipzig, 1983–), i. 77–88.
1030
13. NZ 4 Dec. 1866, Mbl. (Bamberger); Hermann Oncken, Rudolf von
Bennigsen. Ein deutscher liberaler Politiker, 2 vols., ii. Von 1867 bis
1902 (Stuttgart: Deutsche Verlags-Anstalt, 1910), 297 ff.; Huber,
Verfassungsgeschichte, iv (note 7), 129 ff., 351 ff.; Karl Erich Born,
‘Von der Reichsgründung bis zum 1. Weltkrieg’, in Von der
Französischen Revolution bis zum Ersten Weltkrieg (= Bruno
Gebhardt, Handbuch der deutschen Geschichte, ed. Herbert
Grundmann, iii), 9th edn (Stuttgart: Union-Verlag, 1970), 261 ff.,
296 ff.; Hans Gerhard Benzig, Bismarcks Kampf um die
Kreisordnung von 1871 (Hamburg: Kovac, 1996), 173 ff.; Hartwin
Spenkuch, Das Preussische Herrenhaus. Adel und Bürgertum in der
Ersten Kammer des Landtags 1854–1918 (Düsseldorf: Droste,
1998), 93 ff.; Schmidt, ‘Nationalliberale’ (note 8), 208 ff.; Booms,
Deutschkonservative Partei (note 3), 17 ff.; Ritter, Konservative
(note 3), 361 ff.; Gall, Bismarck (note 5), 359 (Grand Duke
Frederick I’s letter to Heinrich Gelzer, 3 Apr. 1878); Lexikon
(note 12), iv, ed. Fricke et al., 358–67 (the article ‘Vereinigung
der Steuer- und Wirtschaftsreformer 1876–1928’), 509–43 (the
article ‘Zentralverband Deutscher Industrieller 1876–1919’).
14. Norddeutsche Allgemeine Zeitung, 31 July 1878; ‘Auf den Grund’,
Neue Preussische Zeitung (Kreuz-Zeitung), 30 July 1878; ‘Unsere
Hauptanklagen gegen den Liberalismus’, part III, ibid. 15 Aug.
1878, part VI, 21 Aug. 1878 (emphases in original); Otto Pflanze,
Bismarck. Der Reichskanzler (Munich: C. H. Beck, 1998), 118 ff.,
orig. Bismarck and the Development of Germany: The Period of
Unification, 1815–1871 (Princeton: Princeton University Press,
1990); Wolfgang Pack, Das parlamentarische Ringen um das
Sozialistengesetz 1878/90 (Düsseldorf: Droste, 1961); Vernon L.
1031
Lidtke, The Outlawed Party: Social Democracy in Germany, 1878–
1890 (Princeton: Princeton University Press, 1966); Lucian
Hölscher, Weltgericht oder Revolution. Protestantische und
sozialistische Zukunftsvorstellungen im deutschen Kaiserreich
(Stuttgart: Klett-Cotta, 1989), 221 ff.; Gerhard A. Ritter and
Klaus Tenfelde, Arbeiter im Deutschen Kaiserreich 1871 bis 1914
(Bonn: J. H. W. Dietz, 1992), esp. 679 ff.; Mehring, Geschichte, ii
(note 6), 492 ff. (673–4: statistics on the effects of the anti-
socialist law); Wehler, Gesellschaftsgeschichte, iii (note 8), 902 ff.
15. Bismarck und der Staat. Ausgewählte Dokumente, introd. Hans
Rothfels (Stuttgart: Kohlhammer, n.d. [1953]), 329 ff. (329:
Bismarck’s letter to Itzenplitz, 17 Nov. 1871); Wolfgang Saile,
Hermann Wagener und sein Verhältnis zu Bismarck. Ein Beitrag zur
Geschichte des konservativen Sozialismus (Tübingen: Mohr, 1958),
49 ff. (on L. v. Stein’s influence on the Prussian conservatives);
Helmut Böhme, Deutschlands Weg zur Grossmacht. Studien zum
Verhältnis von Wirtschaft und Staat während der
Reichsgründungszeit 1848–1881 (Cologne: Kiepenheuer & Witsch,
1966), 530 ff.; Karl W. Hardach, Die Bedeutung wirtschaftlicher
Faktoren bei der Wiedereinführung der Eisen- und Getreidezölle in
Deutschland 1879 (Berlin: Duncker & Humblot, 1967); Anderson,
Windthorst (note 8), 226 ff.; Hans Rosenberg, ‘Die
Pseudodemokratisierung der Rittergutsbesitzerklasse’, in
Machteliten und Wirtschaftskonjunkturen. Studien zur neueren
deutschen Sozial- und Wirtschaftsgeschichte (Göttingen:
Vandenhoeck & Ruprecht, 1978), 83–101; id., ‘Zur sozialen
Funktion der Agrarpolitik im Zweiten Reich’, ibid. 102–17; id.,
Depression (note 9), 169 ff.; Wehler, Gesellschaftsgeschichte, iii
1032
(note 8), 934 ff.
16. Ludwig Bamberger, ‘National’ (September 1888), in Politische
Schriften (Berlin 1897), v. 203–37 (217); NZ 14 Nov. 1878, Abl.
(quote from the Volkswirtschaftliche Correspondenz); NZ 13 Nov.
1878, Mbl. (on the ‘Jewification’ of Posen); E. v. d. Brüggen, ‘Die
Kolonisation in unserem Osten und die Herstellung des
Erbzinses’, Preussische Jahrbücher, 44 (1879), 32–51 (35);
‘Sozialismus und Deportation’, Grenzboten, 37/2 (1878), 41–50
(46); ‘Die Julitage des deutschen Liberalismus’, ibid. 38/3
(1879), 124–6 (124); ‘Die natürliche Gruppierung deutscher
Parteien’, ibid. 200–4 (203–4); Im Neuen Reich 1871–1890.
Politische Briefe aus dem Nachlass liberaler Parteiführer, sel. and ed.
Paul Wentzcke (= Deutscher Liberalismus im Zeitalter Bismarcks.
Eine politische Briefsammlung, ii) (1925; repr. Osnabrück: Biblio-
Verlag, 1967), 307–11 (Lasker’s letter to his constituents, March
1880; quotes: 309), 356 (statement upon leaving the party, 30
Aug. 1880); Heinrich August Winkler, ‘Vom linken zum rechten
Nationalismus. Der deutsche Liberalismus in der Krise von
1878/79’, in Liberalismus und Antiliberalismus. Studien zur
politischen Sozialgeschichte des 19. u. 20. Jahrhunderts (Göttingen:
Vandenhoeck & Ruprecht, 1979), 36–51; id., ‘Der Nationalismus
und seine Funktionen’, ibid. 52–80.
17. Huber, Verfassungsgeschichte, iv (note 7), 767 ff.; Nipperdey,
Geschichte, ii (note 10), 428 ff.; Wehler, Gesellschaftsgeschichte, iii
(note 8), 897 ff.; Winfried Becker, ‘Die Deutsche Zentrumspartei
im Bismarckreich’, in Die Minderheit als Mitte. Die Deutsche
Zentrumspartei in der Innenpolitik des Reiches 1871–1933, ed. id.
1033
(Paderborn: Schöningh, 1986), 9–46; David Blackbourn, ‘Die
Zentrumspartei und die deutschen Katholiken während des
Kulturkampfes und danach’, in Probleme (note 7), ed. Pflanze,
73–94; Anderson, Windthorst (note 8), 278 ff.
18. Deutsche Parteiprogramme, ed. Wilhelm Mommsen (Munich: Isar,
1960), 157 (platform of the German Liberal Party from 5 Mar.
1884), 158–160 (Heidelberg declaration of the National Liberals
from 23 Mar. 1884); Bismarck (note 15), ed. Rothfels, 359
(Bismarck’s statement to Busch, 26 June 1881); id., Theodor
Lohmann und die Kampfjahre der staatlichen Sozialpolitik (1871–
1905) (Berlin: C. S. Mittler und Sohn, 1927), 48 ff.; Gerhard A.
Ritter, Der Sozialstaat. Entstehung und Entwicklung im
internationalen Vergleich (Munich: Oldenbourg, 1989), 60 ff.;
Wehler, Gesellschaftsgeschichte, iii (note 8), 936 ff.; Wolther von
Kieseritzky, Liberalismus und Sozialstaat. Liberale Politik zwischen
Machtstaat und Arbeiterbewegung in Deutschland (1878–1893)
(Cologne: Böhlau, 2001); Dan S. White, The Splintered Party.
National Liberalism in Hessen and the Reich, 1867–1918
(Cambridge, MA: Harvard University Press, 1976), 84 ff.
19. Hans-Ulrich Wehler, Bismarck und der Imperialismus, 1st edn
(Cologne: Kiepenheuer & Witsch, 1969), 112 ff. (164: Miquel,
166: Hohenlohe-Langenburg); id., ‘Von den “Reichsfeinden” zur
“Reichskristallnacht”: Polenpolitik im Deutschen Kaiserreich
1871–1918’, id., Krisenherde des Kaiserreichs 1871–1918
(Göttingen: Vandenhoeck & Ruprecht, 1970), 181–199 (193 f.);
id., ‘Unfähig zur Verfassungsreform: Das “Reichsland” Elsass-
Lothringen von 1870 bis 1918’, ibid. 17–63; id.,
1034
Gesellschaftsgeschichte, iii (note 8), 961 ff.; Schieder, Kaiserreich
(note 2), 95 ff.; Christoph Klessmann, Polnische Bergarbeiter im
Ruhrgebiet 1870–1945. Soziale Integration und nationale Subkultur
einer Minderheit in der deutschen Industriegesellschaft (Göttingen:
Vandenhoeck & Ruprecht, 1978), 23 ff.; Hermann Hiery,
Reichstagswahlen im Reichsland. Ein Beitrag zur Landesgeschichte
von Elsass-Lothringen und zur Wahlgeschichte des Deutschen Reiches
1871–1918 (Düsseldorf: Droste, 1986), 60 ff.; Nationale
Minderheiten und staatliche Minderheitenpolitik in Deutschland im
19. Jahrhundert, ed. Hans Henning Hahn and Peter Kunze (Berlin:
Akademie Verlag, 1999).
20. Die Grosse Politik der Europäischen Mächte 1871–1914. Sammlung
der Diplomatischen Akten des Auswärtigen Amtes, ii. Der Berliner
Kongress und seine Vorgeschichte (Berlin: Deutsche
Verlagsgesellschaft für Politik und Geschichte, 1922), 153 f.
(Kissinger Diktat, 15 June 1877, also on the ‘cauchemar des
coalitions’); Klaus Hildebrand, Das vergangene Reich. Deutsche
Aussenpolitik von Bismarck zu Hitler (Stuttgart: Deutsche Verlags-
Anstalt, 1995), 34 ff.; Andreas Hillgruber, Bismarcks Aussenpolitik
(Freiburg: Rombach, 1972), 175 ff.; George W. F. Hallgarten,
Imperialismus vor 1914. Die soziologischen Grundlagen der
Aussenpolitik europäischer Grossmächte vor dem Ersten Weltkrieg, 2
vols. (Munich: C. H. Beck, 1963), i. 160 ff.; Hans-Ulrich Wehler,
‘Bismarcks späte Russlandpolitik 1879–1890’, in Krisenherde
(note 19), 163–81; id., Gesellschaftsgeschichte, iii (note 8), 970 ff.;
Gall, Bismarck (note 5), 634 (Herbert von Bismarck to Bill von
Bismarck, 19 June 1887), 636 (Bismarck to Bronsart von
Schellendorf, 30 Dec. 1887), 642 ff. (‘system of expedients’);
1035
Horst Müller-Link, Industrialisierung und Aussenpolitik. Preussen–
Deutschland und das Zarenreich von 1860 (Göttingen:
Vandenhoeck & Ruprecht, 1977), 191 ff.
21. Theodor Fontane, Der Stechlin, Roman, id., Werke, Schriften,
Briefe, ed. Walter Keitel and Helmuth Nürnberger, v, pt 1
(Munich: Hanser, 1980), 307; Röhl, Wilhelm II. (note 11), 711 ff.;
id., ‘Staatsstreichspläne oder Staatsstreichbereitschaft? Bismarcks
Politik in der Entlassungskrise’, HZ 203 (1966), 610–24; Huber,
Verfassungsgeschichte, iii (note 7), 202 ff. (theory on the coup
plan as a means of ‘concealing the intention to resolve the crisis
in a manner in keeping with the constitution’); Anderson,
Windthorst (note 8), 401 ff.; Wehler, Gesellschaftsgeschichte, iii
(note 5), 684 ff.; Gall, Bismarck (note 5), 684 ff.; Erich Eyck,
Bismarck. Leben und Werk, 3 vols., iii (Erlenbach: E. Rentsch,
1944), 500 ff.
22. Sten. Ber., cii. 725, 733 (Bismarck, 6 Feb. 1888); liiv. 28–31, 33–
4 (Bismarck on Lasker, 13 Mar. 1884); liv/2. 2249 (Hänel, 10
July 1879); Fontane, Werke (note 21), iii, pt 4, 674 (letter to
Georg Friedlaender, 7 Jan. 1889); Lothar Wickert, Theodor
Mommsen, 4 vols., iv (Frankfurt: Klostermann, 1980), 93
(Mommsen to Brentano, 3 Jan. 1902); Friedrich Nietzsche,
Unzeitgemässe Betrachtungen. David Strauss. Der Bekenner und der
Schriftsteller, in Werke (critical edition, ed. Giorgio Colli and
Mazzino Montinari), i, pt 3 (Berlin: de Gruyter, 1972), 153–238
(155–6; emphasis in original), tr. as Thoughts out of Season, in
Complete Works, ed. Oscar Levy and Robert Guppy (New York:
Russell & Russell, 1964), iv, pt 1/4 Jacob Burckhardt, Briefe, v,
1036
ed. Max Burckhardt (Basel: B. Schwabe, 1949–86), 184 (letter to
Friedrich von Preen, 31 Dec. 1872); Emil Du Bois-Reymond, ‘Der
deutsche Krieg’ (3 Aug. 1870), id., Reden, i. Literatur, Philosophie,
Zeitgeschichte (Leipzig: Veit, 1886), 65–92 (92); NZ 5 Dec. 1878,
Mbl.; Louis L. Snyder, ‘Bismarck and the Lasker Resolution 1884’,
Review of Politics, 79 (1967), 41–64; Craig, Geschichte (note 9),
156 ff.; Eyck, Bismarck, iii (note 21), 543 ff.; Hans Kohn, Wege
und Irrwege. Vom Geist des deutschen Bürgertums (Düsseldorf:
Droste, 1962, 179 ff., orig. The Mind of Germany: The Education of
a Nation (New York: Scribner, 1960); Wehler,
Gesellschaftsgeschichte, iii (note 8), 327 ff.; Huber,
Verfassungsgeschichte, iv (note 7), 190 ff. (Geffcken affair); Hiery,
Reichstagswahlen (note 19), 200 ff.; Peter Steinbach, ‘Politisierung
und Nationalisierung der Region im 19. Jahrhundert.
Regionalspezifische Politikrezeption im Spiegel historischer
Wahlforschung’, in Probleme politischer Partizipation im
Modernisierungsprozess, ed. id. (Stuttgart: Klett-Cotta, 1982), 321–
49, 241 ff.; Bismarck und der deutsche Nationalmythos, ed. Lothar
Machtan (Bremen: Temmen, 1994); Arno Borst, ‘Barbarossas
Erwachen—Zur Geschichte der deutschen Identität’, in Identität,
Poetik und Hermeneutik, VIII, 2nd edn, ed. Odo Marquard and
Karlheinz Stierle (Munich: W. Fink, 1996), 17–60. On the crisis
provoked by ‘Is War in Sight? see Otto Pflanze, Bismarck. Der
Reichsgründer (note 14), 775 ff.
1037
CHAPTER 6. WORLD POLICY AND WORLD WAR
1890–1918
1038
1849–1914 (Munich: C. H. Beck, 1995), 718 ff.; Thomas
Nipperdey, Deutsche Geschichte 1866–1918, ii. Machtstaat vor der
Demokratie (Munich: C. H. Beck, 1992), 595 ff. On Michaelis see
above, p. 151.
2. The article ‘Bund der Landwirte 1893–1920’, in Lexikon zur
Parteiengeschichte. Die bürgerlichen und kleinbürgerlichen Parteien
und Verbände in Deutschland (1789–1945), 4 vols., iv, ed. Dieter
Fricke et al. (Leipzig: VEB Bibliographisches Institut Leipzig,
1983), 241–70 (Dec. 1892 manifesto: 243; membership: 242);
Hans-Jürgen Puhle, Agrarische Interessenpolitik und preussischer
Konservatismus im Wilhelminischen Reich 1893–1914. Ein Beitrag
zur Analyse des Nationalismus in Deutschland am Beispiel des
Bundes der Landwirte und der Deutsch-Konservativen Partei, 2nd
edn (Bonn: Verlag Neue Gesellschaft, 1975), 28 ff.; id., Von der
Agrarkrise zum Präfaschismus. Thesen zum Stellenwert der
agrarischen Interessenverbände in der deutschen Politik am Ende des
19. Jahrhunderts (Wiesbaden: F. Steiner, 1972). Quote from
Caprivi in Sten. Ber., cxviii. 3307 (10 Dec. 1891).
3. John C. G. Röhl, Deutschland ohne Bismarck. Die Regierungskrise
im Zweiten Kaiserreich 1890–1900 (Tübingen: Rainer Wunderlich
Verlag, 1969), 57 ff., orig. Germany without Bismarck: The Crisis of
Government in the Second Reich, 1890–1900 (London: Batsford,
1967); Hans-Jörg v. Berlepsch, ‘Neuer Kurs’ im Kaiserreich? Die
Arbeiterpolitik des Freiherrn von Berlepsch 1890–96 (Bonn: Verlag
Neue Gesellschaft, 1987); Karl Erich Born, Staat und Sozialpolitik
seit Bismarcks Sturz 1890–1914 (Wiesbaden: F. Steiner, 1957);
Gerhard A. Ritter, Die Arbeiterbewegung im Wilhelminischen Reich.
1039
Die Sozialdemokratische Partei und die Freien Gewerkschaften 1890–
1900, 2nd edn (Berlin: Colloquium Verlag, 1963), 15 ff.; Ludwig
Elm, Zwischen Fortschritt und Reaktion. Geschichte der liberalen
Bourgeoisie 1893–1918 (Berlin: Akademie-Verlag 1968); Ernst
Rudolf Huber, Deutsche Verfassungsgeschichte seit 1789, iv.
Struktur und Krisen des Kaiserreichs, 2nd edn (Stuttgart:
Kohlhammer, 1969), 247 ff., 554 ff., 950 ff., 1075 ff., 1220 ff.;
Wehler, Gesellschaftsgeschichte, iii (note 1), 1000 ff.
4. Hans Herzfeld, Johannes von Miquel. Sein Anteil am Ausbau des
Deutschen Reiches bis zur Jahrhundertwende, 2 vols. (Detmold:
Meyersche Hofbuchhandlung, Verlag Dr. Catharina Staercke,
1938), i. 394 (Miquel, 26 Jan. 1879); Heinrich August Winkler,
‘Der rückversicherte Mittelstand: Die Interessenverbände von
Handwerk und Kleinhandel im deutschen Kaiserreich’, in
Liberalismus und Antiliberalismus. Studien zur politischen
Sozialgeschichte des 19. und 20. Jahrhunderts (Göttingen:
Vandenhoeck & Ruprecht, 1979), 83–98; Dirk Stegmann, Die
Erben Bismarcks. Parteien und Verbände in der Spätphase des
Wilhelminischen Deutschland. Sammlungspolitik 1897–1918
(Cologne: Kiepenheuer & Witsch, 1970), 59 ff. (66: William II, 18
June 1897; manifesto of the free traders of 1898); Hartmut
Kaelble, Industrielle Interessenpolitik in der Wilhelminischen
Gesellschaft. Centralverband Deutscher Industrieller 1895–1914
(Berlin: de Gruyter, 1967), 51 ff.; Hans-Peter Ullmann, Der Bund
der Industriellen. Organisation, Einfluss und Politik klein- und
mittelbetrieblicher Industrieller im Deutschen Kaiserreich 1895–1914
(Göttingen: Vandenhoeck & Ruprecht, 1976), 165 ff.; Röhl,
Deutschland (note 3), 224 ff.
1040
5. Sten. Ber., cxlix. 5149 (Marschall, 18 Mar. 1897); Eckart Kehr,
‘Englandhass und Weltpolitik’ (1928), in Der Primat der
Innenpolitik. Gesammelte Aufsätze zur preussischdeutschen
Sozialgeschichte im 19. u. 20. Jahrhundert, ed. and introd. Hans-
Ulrich Wehler (Berlin: de Gruyter, 1965), 149–75 (quotes from
Kehr: 164, Tirpitz, 21 Dec. 1895: 165); id., Economic Interest,
Militarism, and Foreign Policy: Essays on German History (Berkeley:
University of California Press, 1977); id., Schlachtflottenbau und
Parteipolitik 1894–1901 (Berlin: E. Ebering, 1930; repr. Vaduz:
Klaus Reprint, 1965), esp. 276 ff.; Volker Berghahn, Der Tirpitz-
Plan. Genesis und Verfall einer innenpolitischen Krisenstrategie unter
Wilhelm II. (Düsseldorf: Droste, 1971); Kenneth D. Barkin, The
Controversy over German Industrialization 1890–1902 (Chicago:
University of Chigaco Press, 1970), 211 ff.; Konrad Schilling,
Beiträge zu einer Geschichte des radikalen Nationalismus in der
Wilhelminischen Ära 1890–1909 (Cologne: Gouder & Hansen,
1968); Geoff Eley, Reshaping the German Right: Radical
Nationalism and Political Change after Bismarck (New Haven: Yale
University Press, 1980), 68 ff.; id., ‘Die Umformierung der
Rechten: Der radikale Nationalismus und der Deutsche
Flottenverein 1898–1908’, in Wilhelminismus, Nationalismus,
Faschismus. Zur historischen Kontinuität in Deutschland (Münster:
Westfälisches Dampfboot, 1991), 144–73; Huber,
Verfassungsgeschichte, iv (note 3), 565 ff.; Wehler,
Gesellschaftsgeschichte, iii (note 1), 1129 ff.; the article ‘Deutscher
Flottenverein 1898–1934’, in Lexikon (note 2), ii, ed. Fricke et
al., 67–89 (membership: 68).
6. Sten. Ber., clix. 60 (Bülow, 6 Dec. 1897); Konrad Canis, Von
1041
Bismarck zur Weltpolitik. Deutsche Aussenpolitik 1890 bis 1902
(Berlin: Akademie-Verlag, 1997), 223 ff.; Gustav Schmidt, Der
europäische Imperialismus (Munich: Oldenbourg, 1985), 90 ff.;
Klaus Hildebrand, Das vergangene Reich. Deutsche Aussenpolitik
von Bismarck bis Hitler (Stuttgart: Deutsche Verlags-Anstalt,
1995), 149 ff.
7. The article ‘Alldeutscher Verband 1891–1939’, in Lexikon (note
2), i, ed. Fricke et al., 13–47 (membership: 13, quotes from the
statutes: 16, of the year 1894: 19); the article ‘Deutscher
Ostmarkenverein 1894–1934’, ibid., ii. 225–50 (membership:
225–6, quote from the statutes: 228); Roger Chickering, We Men
Who Feel Most German: A Cultural Study of the Pan-German League
1886–1914 (London: Allen & Unwin, 1984), 44 ff.; Mildred S.
Wertheimer, The Pan-German League 1890–1914 (New York:
Octagon, 1924); Hans-Ulrich Wehler, ‘Von den “Reichsfeinden”
zur “Reichskristallnacht”: Polenpolitik im Deutschen Kaiserreich
1871–1918’, in Krisenherde des Kaiserreichs 1871–1918. Studien
zur deutschen Sozial- und Verfassungsgeschichte (Göttingen:
Vandenhoeck & Ruprecht, 1970), 181–200; id., ‘Deutsch-
polnische Beziehungen im 19. und 20. Jahrhundert’, ibid. 201–
18; Martin Broszat, 200 Jahre deutsche Polenpolitik (Munich:
Ehrenwirth, 1963), 25 ff.
8. Thomas Rohkrämer, Der Militarismus der «kleinen Leute». Die
Kriegervereine im Deutschen Kaiserreich 1871–1914 (Munich:
Oldenbourg, 1990), esp. 34 ff.; Jakob Vogel, Nationen im
Gleichschritt. Der Kult der ‘Nation in Waffen’ in Deutschland und
Frankreich, 1871–1914 (Göttingen: Vendenhoek & Ruprecht,
1042
1997); Robert von Friedeburg, ‘Klassen-, Geschlechter- oder
Nationalidentität? Handwerker und Tagelöhner in den
Kriegervereinen der neupreussischen Provinz Hessen-Nassau
1890–1914’, Militär und Gesellschaft im 19. und 20. Jahrhundert,
ed. Ute Frevert (Stuttgart: Klett-Cotta, 1997), 229–44; the article
‘Kyffhäuser-Bund der Deutschen Landeskriegerverbände
1899/1900–1943’, in Lexikon (note 2), iii, ed. Fricke et al., 325–
44 (membership: 326–7).
9. Thomas Nipperdey, ‘Nationalidee und Nationaldenkmal in
Deutschland im 19. Jahrhundert’, in Gesellschaft, Kultur, Theorie.
Gesammelte Aufsätze zur neueren Geschichte (Göttingen:
Vandenhoeck & Ruprecht, 1976), 133–73 (143–4); Das
Kyffhäuser-Denkmal 1896–1996. Ein nationales Monument im
europäischen Kontext, ed. Gunther Mai (Cologne: Böhlau, 1997);
Wolfgang Hardtwig, ‘Bürgertum, Staatssymbolik und
Staatsbewusstsein im Deutschen Kaiserreich 1871–1914’, in
Nationalismus und Bürgerkultur in Deutschland 1500–1914.
Ausgewählte Aufsätze (Göttingen: Vandenhoeck & Ruprecht,
1994), 191–218 (208); Charlotte Tacke, Denkmal im sozialen
Raum. Nationale Symbole in Deutschland und Frankreich im 19.
Jahrhundert (Göttingen: Vandenhoeck & Ruprecht, 1995), 80 ff.;
ead., ‘Die 1900Jahrfeier der Schlacht im Teutoburger Wald 1909.
Von der “klassenlosen Bürgergesellschaft” zur “klassenlosen
Volksgemeinschaft”?’, in Bürgerliche Feste. Symbolische Formen
politischen Handelns im 19. Jahrhundert, ed. Manfred Hettling and
Paul Nolte (Göttingen: Vandenhoeck & Ruprecht, 1993), 192–
230; Reinhard Alings, Monument und Nation. Das Bild vom
Nationalstaat im Medium Denkmal—zum Verhältnis von Nation und
1043
Staat im deutschen Kaiserreich 1871–1918 (Berlin: de Gruyter,
1996); Michael S. Cullen, Der Reichstag. Die Geschichte eines
Monumentes, 2nd edn (Stuttgart: Parkland Verlag, 1990), 219
(William II, 25 Apr. 1893 in Rome), 242 (Vossische Zeitung, 6
Dec. 1894), 246 (William II to Eulenburg, 8 Dec. 1894); Fritz
Schellack, ‘Sedan- und Kaisergeburtstagsfeste’, in Öffentliche
Festkultur. Politische Feste in Deutschland von der Aufklärung bis
zum Ersten Weltkrieg, ed. Dieter Düding et al. (Hamburg:
Rowohlt, 1988), 278–97; Beatrix Bouvier, ‘Die Märzfeiern der
sozialdemokratischen Arbeiter: Gedenktage des Proletariats—
Gedenktage der Revolution. Zur Geschichte des 18. März’, ibid.
334–51; Georg L. Mosse, Die Nationalisierung der Massen.
Politische Symbolik und Massenbewegungen in Deutschland von den
Napoleonischen Kriegen bis zum Dritten Reich (Frankfurt: Ullstein,
1976), 62 ff., orig. The Nationalization of the Masses: Political
Symbolism and Mass Movements in Germany from the Napoleonic
Wars through the Third Reich (New York: H. Fertig, 1975).
10. Kurt Wawrzinek, Die Entstehung der deutschen Antisemitenparteien,
1873–1890 (Berlin: E. Ebering, 1927); Richard S. Levy, The
Downfall of the Anti-Semitic Parties in Imperial Gemany (New
Haven: Yale University Press, 1975); Stefan Scheil, Die
Entwicklung des politischen Antisemitismus in Deutschland zwischen
1881 und 1912: eine wahlgeschichtliche Untersuchung (Berlin:
Duncker & Humblot, 1999), 72 ff.; Handbuch zur ‘Völkischen
Bewegung’ 1871–1918, ed. Uwe Puschner et al. (Munich: K. G.
Saur, 1996); the article ‘Antisemitische Parteien 1879–1894’,
Lexikon (note 2), i, ed. Fricke et al., 77–88; ‘Deutschsoziale Partei
1900–1914’, ibid. ii. 534–7; ‘Deutschsoziale Reformpartei 1894–
1044
1900’, ibid. 540–9; Hans Booms, Die Deutschkonservative Partei.
Preussischer Charakter, Reichsauffassung, Nationalbegriff
(Düsseldorf: Droste, 1954), esp. 97 ff.; James N. Retallack,
Notables of the Right: The Conservative Party and Political
Mobilization in Germany, 1876–1918 (London: Unwin Hyman,
1988), 91 ff.; Deutsche Parteiprogramme, ed. Wilhelm Mommsen
(Munich: Isar Verlag, 1960), 78–80 (‘Tivoli platform’); Handbuch
der Deutsch-Konservativen Partei, 4th edn (Berlin: Reimar
Hobbing, 1911), 4–9 (the article ‘Antisemitismus’); Uwe Mai,
‘“Wie es der Jude treibt”. Das Feindbild der antisemitischen
Bewegung am Beispiel der Agitation Hermann Ahlwardts’,
Feindbilder in der deutschen Geschichte. Studien zur
Vorurteilsgeschichte im 19. u. 20. Jahrhundert, ed. Christoph Jahr
et al. (Berlin: Metropol, 1994), 55–80; Puhle, Interessenpolitik
(note 2), 130 (quote from 1895); Iris Hamel, Völkischer Verband
und nationale Gewerkschaft. Der Deutschnationale
Handlungsgehilfen-Verband 1893–1933 (Frankfurt: Europäische
Verlagsanstalt, 1967), 14 ff. (quote from the statutes: 53); Jürgen
Kocka, Unternehmensverwaltung und Angestelltenschaft am Beispiel
Siemens 1847–1914. Zum Verhältnis von Kapitalismus und
Bürokratie in der deutschen Industrialisierung (Stuttgart: Klett,
1969), 148 ff., 536 ff.; Gerhard A. Ritter with Merith Niehuss,
Wahlgeschichtliches Arbeitsbuch. Materialien zur Statistik des
Kaiserreiches 1871–1918 (Munich: C. H. Beck, 1980), 38 ff.;
Nipperdey, Geschichte (note 1), 289 ff.
11. Parteiprogramme (note 10), ed. Mommsen, 166–8 (Grundlinien des
Nationalsozialen Kreises, 25 Nov. 1896); Friedrich Naumann,
‘Demokratie und Kaisertum’, in Werke. Politische Schriften, ii.
1045
Schriften zur Verfassungspolitik (Cologne: Westdeutscher Verlag,
1964), 266–7; Werner Conze, ‘Friedrich Naumann. Grundlagen
und Ansatz seiner Politik in der national-sozialen Zeit’, in
Schicksalswege deutscher Vergangenheit. Festschrift für Siegfried A.
Kaehler, ed. Walther Hubatsch (Düsseldorf: Droste, 1950), 355–
86; Richard Nürnberger, ‘Imperialismus, Sozialismus und
Christentum bei Friedrich Naumann’, HZ 170 (1950), 525–48;
Theodor Heuss, Friedrich Naumann. Der Mann, das Werk, die Zeit
(Stuttgart: R. Wunderlich, 1949); Dieter Düding, Der
Nationalsoziale Verein 1896–1903. Der gescheiterte Versuch einer
parteipolitischen Synthese von Nationalismus, Sozialismus und
Liberalismus (Munich: Oldenbourg, 1972); James J. Sheehan,
‘Deutscher Liberalismus im postliberalen Zeitalter 1890–1914’,
GG 4 (1978), 29–98; Konstanze Wegner, Theodor Barth und die
Freisinnige Vereinigung. Studien zur Geschichte des Linksliberalismus
im Wilhelminischen Deutschland 1893–1910 (Tübingen: Mohr,
1968); Elisabeth Fehrenbach, Wandlungen des deutschen
Kaisergedankens 1871–1918 (Munich: Oldenbourg, 1969), 200 ff.;
Mommsen, Weber (note 1), 73 ff., 186 ff.
12. Ernst Deuerlein, ‘Die Bekehrung des Zentrums zur nationalen
Idee’, Hochland, 62 (1970), 432–49 (Kopp’s letter to Bachem, 4
Feb. 1900: 446–7; emphasis in original); Rudolf Morsey, ‘Die
deutschen Katholiken und der Nationalstaat zwischen
Kulturkampf und erstem Weltkrieg’, in Die deutschen Parteien vor
1918, ed. Gerhard A. Ritter (Cologne: Kiepenheuer & Witsch,
1973), 270–98 (quote from Bebel: 283); Wilfried Loth, Katholiken
im Kaiserreich. Der politische Katholizismus in der Krise des
wilhelminischen Deutschland (Düsseldorf: Droste, 1984), 38 ff.;
1046
David Blackbourn, Class, Religion and Local Politics in Wilhelmine
Germany: The Centre Party in Württemberg before 1914 (New
Haven: Yale University Press, 1980), 23 ff.; Olaf Blaschke,
Katholizismus und Antisemitismus im deutschen Kaiserreich
(Göttingen: Vandenhoeck & Ruprecht, 1997), esp. 119 ff. (on the
journalism of the 1890s: 125 ff.; on the debate in the Prussian
House of Representatives in November 1880: 238 ff.); Uwe
Mazura, Zentrumspartei und Judenfrage 1870/71–1933.
Verfassungsstaat und Minderheitenschutz (Mainz: Matthias-
Grünewald-Verlag, 1994); Winfried Becker, ‘Die Deutsche
Zentrumspartei im Bismarckreich’, in Die Minderheit als Mitte. Die
Deutsche Zentrumspartei in der Innenpolitik des Reiches 1871–1933,
ed. id. (Paderborn: Schöningh, 1986), 9–46; Norbert
Schlossmacher, ‘Der Antiultramontanismus im Wilhelminischen
Deutschland’, in Deutscher Katholizismus im Umbruch zur Moderne,
ed. Wilfried Loth (Stuttgart: Kohlhammer, 1991), 164–98; M.
Rainer Lepsius, ‘Parteiensystem und Sozialstruktur. Zum Problem
der Demokratisierung der deutschen Gesellschaft’, in Demokratie
in Deutschland. Soziologisch-historische Konstellationsanalysen
(Göttingen: Vandenhoeck & Ruprecht, 1993), 25–50. For quote
from Theodor Mommsen’s letter to Lujo Brentano of 3 Jan. 1902
see Lothar Wickert, Theodor Mommsen, 4 vols., iv (Frankfurt:
Klostermann, 1980), 93. On the Evangelischer Bund see Gangolf
Hübinger, Kulturprotestantismus und Politik. Zum Verhältnis von
Liberalismus und Protestantismus im wilhelminischen Deutschland
(Tübingen: Mohr, 1994), 52–3 (membership: 53).
13. Protokoll über die Verhandlungen des Parteitags der
Sozialdemokratischen Partei Deutschlands, abgehalten zu Erfurt vom
1047
14. bis 20. Oktober 1891 (Berlin 1891), 169 (Bebel);
Parteiprogramme (note 10), ed. Mommsen, 344–9 (Engels’s
critique of the draft of the Erfurt platform), 349–53 (Erfurt
platform); Karl Kautsky, Der Weg zur Macht. Politische
Betrachtungen über das Hineinwachsen in die Revolution, 2nd edn
(Berlin: Buchhandlung Vorwärts, 1909), 44–52 (article from the
Neue Zeit, Dec. 1893; quotes: 44–6), tr. as The Road to Power:
Political Reflections on Growing into the Revolution (Atlantic
Highlands, NJ: Humanities Press, 1996); Dieter Grosser, Vom
monarchischen Konstitutionalismus zur parlamentarischen
Demokratie. Die Verfassungspolitik der deutschen Parteien im letzten
Jahrzehnt des Kaiserreichs (The Hague: Nijhoff, 1970), 33–4
(Kautsky to Mehring, 5 July 1893); Ingrid Gilcher-Holtey, Das
Mandat des Intellektuellen. Karl Kautsky und die Sozialdemokratie
(Berlin: Siedler, 1986), 59 ff.; Erich Matthias, ‘Kautsky und der
Kautskyanismus. Die Funktion der Ideologie in der deutschen
Sozialdemokratie vor dem ersten Weltkrieg’, Marxismus-Studien,
2nd series (Tübingen: Mohr, 1957), 151–97; Lucian Hölscher,
Weltgericht oder Revolution. Protestantische und sozialistische
Zukunftsvorstellungen im deutschen Kaiserreich (Stuttgart: Klett-
Cotta, 1989), 231 ff., 307 ff.
14. Eduard Bernstein, Die Voraussetzungen des Sozialismus und die
Aufgaben der Sozialdemokratie (Stuttgart: J. H. W. Dietz,
Nachfolger, 1909), pp. v–vii (to the Stuttgart party congress), pp.
viii–ix (to Vorwärts, 20 Oct. 1898), 124, 129, 165, 169, 179, 183,
187 (emphasis in original), tr. as The Preconditions of Socialism
(Cambridge: Cambridge University Press, 1993); Susanne Miller,
Das Problem der Freiheit im Sozialismus. Freiheit, Staat und
1048
Revolution in der Programmatik der Sozialdemokratie von Lassalle
bis zum Revisionismus-Streit (Frankfurt: Europäische
Verlagsanstalt, 1964), 227 ff.; Peter Gay, Das Dilemma des
Demokratischen Sozialismus. Eduard Bernsteins Auseinandersetzung
mit Marx (Nuremberg: Nest, 1954), orig. The Dilemma of
Democratic Socialism: Eduard Bernstein’s Challenge to Marx (New
York: Columbia University Press, 1952); Helmut Hirsch, Der
‘Fabier’ Eduard Bernstein. Zur Entwicklungsgeschichte des
evolutionären Sozialismus (Berlin: Dietz, 1977); Francis Ludwig
Carsten, Eduard Bernstein 1850–1932. Eine politische Biographie
(Munich: C. H. Beck, 1993), 81 ff.; id., August Bebel und die
Organisation der Massen (Berlin: Siedler, 1991), 179 ff.
15. Rosa Luxemburg, Sozialreform oder Revolution?, in Politische
Schriften, 3 vols., ed. Ossip K. Flechtheim (Frankfurt: Europäische
Verlagsanstalt, 1966–), i. 47–133 (54, 90, 113–14, 119, 123,
130; emphases in original), tr. as Reform or Revolution? (New
York: Pathfinder Press, 1973); Karl Kautsky, Bernstein und das
Sozialdemokratische Programm. Eine Antikritik (Stuttgart: J. H. W
Dietz, Nachfolger, 1899; repr. Bonn, 1976, 2nd edn), 43, 183,
191, 193, 195; Friedrich Engels, Herrn Eugen Dührings Umwälzung
der Wissenschaft, MEW, xx. 1–303, tr. as Anti-Dühring: Herr Eugen
Dühring’s Revolution in Science, MECW, xxv. 1–312; August Bebel,
Die Frau und der Sozialismus (1878; repr. Bonn: Dietz, 1994), tr.
as Woman under Socialism (New York: Schocken Books, 1904,
repr. 1971); Peter Nettl, Rosa Luxemburg (Cologne, 1967), 54 ff.
(orig. New York: Oxford University Press, 1965).
16. Protokoll über die Verhandlungen des Parteitages der
1049
Sozialdemokratischen Partei Deutschland, abgehalten zu Hannover
vom 9. bis 14. Oktober 1899 (Berlin: Expedition der
Buchhandlung Vorwärts, 1899), 243–4 (Bebel’s resolution);
Parteiprogramme (note 10), ed. Mommsen, 332–44 (Vollmar’s
speech of 1 June 1891, quotes: 336; emphases in original); Hans
Georg Lehmann, Die Agrarfrage in der Theorie und Praxis der
deutschen und internationalen Sozialdemokratie. Vom Marxismus
zum Revisionismus und Bolschewismus (Tübingen: Mohr, 1970);
Gerhard A. Ritter and Klaus Tenfelde, Arbeiter im Deutschen
Kaiserreich 1871 bis 1914 (Bonn: J. H. W. Dietz, 1992), esp. 111
ff.; Ritter, Arbeiterbewegung (note 3), 128 ff.; Hans Gerd Henke,
Der ‘Jude’ als Kollektivsymbol in der deutschen Sozialdemokratie
1890–1914 (Mainz: Decaton, 1994); Robert S. Wistrich, Socialism
and the Jews: The Dilemmas of Assimilation in Germany and Austria-
Hungary (London: Associated University Press, 1982); Rosemarie
Leuschen-Seppel, Sozialdemokratie und Antisemitismus im
Kaiserreich. Die Auseinandersetzung der Partei mit konservativen und
völkischen Strömungen des Antisemitismus 1971–1914 (Bonn:
Verlag Neue Gesellschaft, 1978). Quote about anti-Semitism in
Ludwig Knorr, Sozialdemokratischer Katechismus für das arbeitende
Volk, 4th edn (Nuremberg: Wörlein & Co., 1894), 30–1, quoted
from Carsten, Bernstein (note 14), 56.
17. Julius Braunthal, Geschichte der Internationale, 2 vols. (Hanover:
Dietz Nachfolger, 1961), i. 263 ff., tr. as History of the
International (New York: Praeger, 1967); Georges Lefranc, Le
mouvement socialiste sous la troisième république, i. 1875–1919,
2nd edn (Paris: Payot, 1977), 105 ff., 196 ff.; Carsten, Bebel (note
14), 137 ff.; Huber, Verfassungsgeschichte, iv (note 3), 106 ff.
1050
18. Hannelore Horn, Der Kampf um den Bau des Mittellandkanals
(Cologne: Westdeutscher Verlag, 1964); Puhle, Interessenpolitik
(note 2), 240 ff.; Theodor Eschenburg, Das Kaiserreich am
Scheidewege. Bassermann, Bülow und der Block (Berlin: Verlag für
Kulturpolitik, 1929); Peter Christian Witt, Die Finanzpolitik des
Deutschen Reiches von 1903 bis 1913. Eine Studie zur Innenpolitik
des Wilhelminischen Deutschland (Lübeck: Matthiesen, 1970), 152
ff.; Gustav Schmidt, Der europäische Imperialismus (Munich:
Oldenbourg, 1985), 95 ff.; Hildebrand, Reich (note 6), 213 ff.;
Wehler, Gesellschaftsgeschichte, iii (note 1), 1008 ff.; Huber,
Verfassungsgeschichte, iv (note 3), 287 ff. Quotes from Bernhard
Fürst von Bülow, Denkwürdigkeiten, 4 vols., ii. Von der Marokko-
Krise bis zum Abschied (Berlin: Ullstein, 1930), 351 ff. (Daily
Telegraph interview), 356–7 (on the mood in Germany), tr. as
Memoirs of Prince von Bülow, 4 vols., ii. From the Morocco Crisis to
Resignation, 1903–1909 (New York: AMS Press, 1931); Sten. Ber.,
ccxxxiii. 5935 (Bülow, 10 Nov. 1908); Dokumente zur Deutschen
Verfassungsgeschichte, iii. Deutsche Verfassungsdokumente (1900–
1918), 3rd edn, ed. Ernst Rudolf Huber (Stuttgart: Kohlhammer,
1990), 28 (statement on 17 Nov. 1908).
19. Huber, Verfassungsgeschichte, iv (note 3), 318 (‘threshold’ quote);
Manfred Rauh, Die Parlamentarisierung des Deutschen Reiches
(Düsseldorf: Droste, 1977) (theory on the parliamentarization of
the German Reich prior to 1914); Thomas Kühne,
Dreiklassenwahlrecht und Wahlkultur in Preussen 1867–1914
(Düsseldorf: Droste, 1994); Barbara Greven-Aschoff, Die
bürgerliche Frauenbewegung in Deutschland 1894–1933 (Göttingen:
Vandenhoeck & Ruprecht, 1981), 87 ff.; Angelika Schaser,
1051
‘Bürgerliche Frauen auf dem Weg in die linksliberalen Parteien
(1908–1933)’, HZ263 (1996), 641–80; ‘Heraus mit dem
Frauenwahlrecht’. Die Kämpfe der Frauen in Deutschland und
England um die politische Gleichberechtigung, ed. Christel Wickert
(Pfaffenweiler: Centaurus-Verlagsgesellschaft, 1990). Quotes
from the platform of the Progressive People’s Party in
Parteiprogramme (note 10), ed. Mommsen, 173–6; for the election
numbers see Arbeitsbuch (note 10), ed. Ritter, 140, 146.
20. MEW, xxxiv. 276 (Engels’s letter to Johann Philipp Becker of 10
Feb. 1882; here the ‘avant-garde’ quote), xxxvi. 305–7 (Engels’s
letter to Vera Zasulich of 23 Apr. 1885; here the remaining
quotes; emphasis in original); Friedrich Engels, ‘Einleitung zu
“Die Klassenkämpfe in Frankreich 1848 bis 1850” von Karl Marx’
(1895 edition), MEW, vii., 511–27 (quote: 523); MECW, xlvi.
196–8 (Engels’s letter to Johann Philipp Becker of 10 Feb. 1882;
here the ‘avant-garde’ quote), xlvii. 279–81 (Engels’s letter to
Vera Zasulich of 23 Apr. 1885; here the remaining quotes;
emphasis in original); Engels, ‘Introduction to Karl Marx’s The
Class Struggles in France, 1848 to 1850’, MECW, xxvii. 506–25;
Rosa Luxemburg, Massenstreik, Partei und Gewerkschaften (Sept.
1906), in Politische Schriften, (note 15), 135–228 (173, 178, 203;
emphases in original), tr. as The Mass Strike: The Political Party
and the Trade Unions, and the Junius Pamphlet (New York: Harper
& Row, 1971); Protokoll über die Verhandlungen des Parteitages der
Sozialdemokratischen Partei Deutschlands. Abgehalten zu Bremen
vom 18. bis 24. September 1904 (Berlin: Expedition der
Buchhandlung Vorwärts, 1904), 193–4 (Bernstein); Protokoll über
die Verhandlungen des Parteitages der Sozialdemokratischen Partei
1052
Deutschlands. Abgehalten zu Jena vom 17. bis 23. September 1905
(Berlin: Expedition der Buchhandlung Vorwärts 1905), 142–3
(Bebel’s resolution); Protokoll über die Verhandlungen des
Parteitages der Sozialdemokratischen Partei Deutschlands.
Abgehalten zu Mannheim vom 23. bis 29. September 1906 (Berlin:
Expedition der Buchhandlung Vorwärts, 1906), 137 (resolution
of the labor union congress in Cologne), 473 (Mannheim
resolution); Peter Lösche, Der Bolschewismus im Urteil der
deutschen Sozialdemokratie 1903–1920 (Berlin: Colloquium
Verlag, 1967), 23 ff. (on Luxemburg’s critique of Lenin, 1904);
Carsten, Bernstein (note 14), 115 ff.; Guenther Roth, The Social
Democrats in Imperial Germany: A Study in the Working Class
Isolation and National Integration (Totowa, NJ: Bedminster Press,
1963); Dieter Groh, Negative Integration und revolutionärer
Attentismus. Die deutsche Sozialdemokratie am Vorabend des 1.
Weltkriegs (Frankfurt: Propyläen, 1973); Carl E. Schorske, Die
grosse Spaltung. Die deutsche Sozialdemokratie von 1905 bis 1917
(Berlin: Olle & Wolter, 1981), esp. 123 ff., orig. German Social
Democracy, 1905–1917: The Development of the Great Schism
(Cambridge, MA: Harvard University Press, 1955); Klaus
Schönhoven, ‘Expansion und Konzentration. Studien zur
Entwicklung der Freien Gewerkschaften im Wilhelminischen
Kaiserreich 1890 bis 1918’, in Klaus Tenfelde et al., Geschichte
der deutschen Gewerkschaften von den Anfängen bis 1945, ed.
Ulrich Borsdorf (Cologne: Bund-Verlag, 1988), 167–278, esp. 236
ff. (figures: 237). On the ‘translatio revolutionis’ in Marx see 93–4.
21. Julius Bachem, ‘Wir müssen aus dem Turm heraus!’, Historisch-
politische Blätter für das katholische Deutschland, 137 (1906), 376–
1053
86 (384 ff.; emphasis in original); Parteiprogramme (note 10), ed.
Mommsen, 245–6. (Berlin declaration in 1909); Karl Bachem,
Vorgeschichte, Geschichte und Politik der Deutschen Zentrumspartei,
vii (Cologne: J. P. Bachem, 1930), 156 ff.; Loth, Katholiken (note
12), 81 ff.
22. Theodor Mommsen, ‘Was uns noch retten kann’, Die Nation, 20
(1902), 163–4; Peter Gilg, Die Erneuerung des demokratischen
Denkens im Wilhelminischen Deutschland. Eine ideengeschichtliche
Studie zur Wende vom 19. zum 20. Jahrhundert (Wiesbaden:
Steiner, 1965), 218 ff.; Siegfried Mielke, Der Hansa-Bund für
Gewerbe, Handel und Industrie 1909–1914. Der gescheiterte Versuch
einer antifeudalen Sammlungspolitik (Göttingen: Vandenhoeck &
Ruprecht, 1976), 29 ff.; Elm, Fortschritt (note 3); Wegner, Barth
(note 11), 134 ff.; Donald Warren, The Red Kingdom of Saxony:
Lobbying Grounds for Gustav Stresemann 1901–1909 (The Hague:
M. Nijhoff, 1964); Dieter Langewiesche, Liberalismus in
Deutschland (Frankfurt: Suhrkamp, 1988), 200 ff., tr. as Liberalism
in Germany (Princeton: Princeton University Press, 1999); the
article ‘Demokratische Vereinigung 1908–1918’, in Lexikon (note
2), i, ed. Fricke et al., 496–503 (membership and ballot records:
496); the article ‘Fortschrittliche Volkspartei 1910–1918’, ibid. ii.
599–609.
23. Sten. Ber., cclix. 898 (Oldenburg, 29 Jan. 1910), cclxviii. 7718
(Hertling, 9 Nov. 1911), 7721–2 (von Heydebrand und der Lasa
[Lase], 9 Nov. 1911), 7730 (Bebel, 9 Nov. 1911), 7737 ff.
(Bassermann, 9 Nov. 1911); Konrad H. Jarausch, The Enigmatic
Chancellor: Bethmann Hollweg and the Hybris of Imperial Germany
1054
(New Haven: Yale University Press, 1973); Hans-Günther
Zmarzlik, Bethmann Hollweg als Reichskanzler 1909–1914
(Düsseldorf: Droste, 1957), esp. 24 ff.; Stig Förster, Der doppelte
Militarismus. Die deutsche Heeresrüstungspolitik zwischen Status-quo-
Sicherung und Aggression 1890–1913 (Stuttgart: F. Steiner, 1985),
208 ff.; Klaus Wernecke, Der Wille zur Weltgeltung. Aussenpolitik
und Öffentlichkeit im Kaiserreich am Vorabend des Ersten
Weltkrieges (Düsseldorf: Droste, 1970), 26 ff. (Bassermann’s letter
to Kiderlen-Waechter, 24 July 1911: 30), 102 ff. (echo in the
press); Thomas Meyer, ‘Endlich eine Tat, eine befreiende Tat… ‘
Alfred Kiderlen-Waechters ‘Panthersprung nach Agadir’ unter dem
Druck der öffentlichen Meinung (Husum: Matthiesen, 1996), 141
ff.; Ralf Forsbach, Alfred von Kiderlen-Wächter (1852–1912). Ein
Diplomatenleben im Kaiserreich, 2 vols. (Göttingen: Vandenhoeck
& Ruprecht, 1997), ii. 411 ff.; Emily Oncken, Panthersprung nach
Agadir. Die deutsche Politik während der Zweiten Marokkokrise von
1911 (Düsseldorf: Droste, 1981), esp. 219 ff.; Flucht in den Krieg?
Die Aussenpolitik des kaiserlichen Deutschland, ed. Gregor
Schöllgen (Darmstadt: Wissenschaftliche Buchgesellschaft, 1991);
Bereit zum Krieg. Kriegsmentalität im wilhelminischen Deutschland
1890–1914. Beiträge zur historischen Friedensforschung, ed. Jost
Dülffer and Karl Holl (Göttingen: Vandenhoeck & Ruprecht,
1986); Wolfgang J. Mommsen, Bürgerstolz und Weltmachtstreben.
Deutschland unter Wilhelm II. 1890 bis 1918 (Berlin: Propyläen,
1995), 450 ff.; Volker Ullrich, Die nervöse Grossmacht 1871–1918.
Aufstieg und Untergang des deutschen Kaiserreiches (Frankfurt: S,
Fischer, 1997), 223 ff.; Joachim Radkau, Das Zeitalter der
Nervosität. Deutschland zwischen Bismarck und Hitler (Munich:
1055
Hanser, 1998), 263 ff.; Hildebrand, Reich (note 6), 260 ff.;
Schmidt, Imperialismus (note 6), 100 ff.; Chickering, We Men
(note 7), 262 ff.; the article ‘Alldeutscher Verband’, Lexikon (note
2), i, ed. Fricke et al., 26–7.
24. Friedrich von Bernhardi, Deutschland und der nächste Krieg
(Stuttgart: J. G. Cotta, 1912), pp. v, 9, 12–13, 34, 73, 89, 110 ff.,
123–4, 275, 293–4, 333 (emphasis in original), tr. as Germany
and the Next War (New York: Longmans, Green, and Co., 1914);
Weber, ‘Nationalstaat’ (note 1), 569; Sten. Ber., cclxviii. 7728
(Bebel, 9 Nov. 1911); Marilyn Shevin Coetzee, The German Army
League: Popular Nationalism in Wilhelmine Gemany (New York:
Oxford University Press, 1990); Förster, Militarismus (note 23),
144 (theory about ‘militarism from below’); Eley, Reshaping (note
5), 160 ff. (theory about ‘populist’ nationalism); Wernecke, Wille
(note 23), 174 ff.; Hans Günther Zmarzlik, ‘Der
Sozialdarwinismus in Deutschland als geschichtliches Problem’,
VfZ (1963), 246–73; Klaus Saul, Staat, Industrie, Arbeiterbewegung
im Kaiserreich. Zur Innen- und Aussenpolitik des Wilhelminischen
Deutschland 1903–1914 (Düsseldorf: Droste, 1974), 115 ff.; the
article ‘Reichsverband gegen die Sozialdemokratie 1904–1918’,
in Lexikon (note 2), iv, ed. Fricke et al., 63–77 (membership: 63);
the article ‘Deutscher Wehrverein 1912–1935’, ibid. ii. 330–41
(membership: 330). For a pre-1914 international comparison of
imperialism and nationalism see Georg W. F. Hallgarten,
Imperialismus vor 1914. Die soziologischen Grundlagen der
Aussenpolitik europäischer Grossmächte vor dem Ersten Weltkrieg, 2
vols. (Munich: C. H. Beck, 1963), ii. 209 ff.; Zara S. Steiner,
Britain and the Origins of the First World War (London: St Martin’s
1056
Press, 1977); Nationalist and Racialist Movements in Britain and
Germany before 1914, ed. Paul Kennedy and Anthony Nicholls
(Oxford: Macmillan in Association with St Antony’s College,
1981); Markus Ingenlath, Mentale Aufrüstung.
Militarisierungstendenzen in Frankreich und Deutschland vor dem
Ersten Weltkrieg (Frankfurt; Campus, 1998); Gerd Krumeich,
Aufrüstung und Innenpolitik in Frankreich vor dem Ersten Weltkrieg.
Die Einführung der dreijährigen Dienstpflicht 1913–1914
(Wiesbaden: F. Steiner, 1980), tr. as Armaments and Politics in
France on the Eve of the First World War: The Introduction of Three-
Year Conscription, 1913–1914 (Leamington Spa, Warwicks.: Berg,
1984). Quote from Ernst Moritz Arndt from ‘Letzter Zug an Gott’
(1844), in Sämmtliche Werke (Leipzig, n.d., v 128–30.
25. ‘Daniel Frymann’ [= Heinrich Class], Wenn ich der Kaiser wär’.
Politische Wahrheiten und Notwendigkeiten, 5th edn (Leipzig:
Dieterich, 1914), 30 ff., 67, 74, 103, 111, 135, 149 ff., 179, 227,
254, 256, 259–60; Paul de Lagarde, Deutsche Schriften, 3rd edn
(Munich: J. F. Lehmann, 1937); Rembrandt als Erzieher. Von einem
Deutschen (Julius Langbehn), (Leipzig: C. L. Hirschfeld, 1890; 3rd
edn, 1891), 3; Houston Stewart Chamberlain, Die Grundlagen des
Neunzehnten Jahrhunderts, 2 vols., unabridged popular edition
(Munich: F. Bruckmann, n.d.); orig. Foundations of the Nineteenth
Century (New York: H. Fertig, 1910); Fritz Stern,
Kulturpessimismus als politische Gefahr (Bern: Scherz, 1963) esp.
25 ff., 127 ff., orig. The Politics of Cultural Despair: A Study in the
Rise of the Germanic Ideology (Berkeley: University of California
Press, 1961); Lamar Cecil, ‘Wilhelm II. und die Juden im
Wilhelminischen Deutschland 1890–1914’, in Die Juden im
1057
wilhelminischen Deutschland 1890–1914, ed. Werner E. Mosse
(Tübingen: Mohr, 1976), 313–47; Norbert Kampe, Studenten und
Judenfrage im Deutschen Kaiserreich. Die Entstehung einer
akademischen Trägerschicht des Antisemitismus (Göttingen:
Vandenhoeck & Ruprecht, 1988); Notker Hammerstein,
Antisemitismus und deutsche Universitäten 1871–1933 (Frankfurt:
Campus, 1995), 64 ff.; Konrad Jarausch, Deutsche Studenten
1800–1970 (Frankfurt: Suhrkamp, 1984), 82 ff.; Michael Peters,
Der Alldeutsche Verband am Vorabend des Ersten Weltkrieges
(1908–1914) (Frankfurt: P. Lang, 1992), 165 ff.
26. Werner Jochmann, ‘Antisemitismus im Deutschen Kaiserreich’,
in Gesellschaftskrise und Judenfeindschaft in Deutschland 1870–
1945 (Hamburg: Christians, 1988), 30–98 (on Gebsattel esp. 87
ff.); Hartmut Pogge von Strandmann, ‘Staatsstreichpläne,
Alldeutsche und Bethmann Hollweg’, in id. and Immanuel Geiss,
Die Erforderlichkeit des Unmöglichen. Deutschland am Vorabend des
ersten Weltkriegs (Frankfurt: Europäische Verlagsanstalt, 1965),
7–45 (38: Wilhelm II); Walter Laqueur, Die deutsche
Jugendbewegung. Eine historische Studie, 2nd edn (Cologne:
Wissenschaft & Politik, 1983), 89 ff.; Werner E. Mosse, ‘Die
Juden in Wirtschaft und Gesellschaft’, in Juden (note 25), ed. id.,
57–113; Peter Pulzer, ‘Die jüdische Beteiligung an der Politik’,
ibid. 143–239 (figures for Berlin: 189); id., ‘Rechtliche
Gleichstellung und öffentliches Leben’, in Deutsch-jüdische
Geschichte in der Neuzeit, 4 vols., iii. Umstrittene Integration 1871–
1918, ed. Michael A. Meyer (Munich: C. H. Beck, 1997), 151–92;
id., ‘Die Wiederkehr des alten Hasses’, ibid. 193–248; id., ‘Die
Reaktion auf den Antisemitismus’, ibid. 249–77; Jacob Toury, Die
1058
politischen Orientierungen der Juden in Deutschland. Von Jena bis
Weimar (Tübingen: Mohr, 1966), 202 ff.; Hadassa Ben-Itto, Die
Protokolle der Weisen von Zion. Anatomie einer Fälschung (Berlin:
Aufbau-Verlag, 1998); Michael Bönisch, ‘Die “Hammer”-
Bewegung’, in Handbuch (note 10), 341–65; the article
‘Reichshammerbund 1910/12–1920’, in Lexikon (note 2), ed.
Fricke et al., iii. 681–83; the article ‘Verein zur Abwehr des
Antisemitismus (Abwehrverein) 1890–1933’, ibid. iv. 375–8
(membership: 375); the article ‘Zionistische Vereinigung für
Deutschland 1897/98–1938/39’, ibid. iv. 636–41; the article
‘Reichsdeutscher Mittelstandsverband 1911–1920’, ibid. iii. 657–
62 (membership: 657); Puhle, Interessenpolitik (note 2), 162 ff.;
Stegmann, Erben (note 4), 352 ff.; Heinrich August Winkler,
Mittelstand, Demokratie und Nationalsozialismus. Die politische
Entwicklung von Handwerk und Kleinhandel in der Weimarer
Republik (Cologne: Kiepenheuer & Witsch, 1972), 52–3 (quotes
on the ‘Cartel of the Productive Estates’).
27. Schulthess’ Europäischer Geschichtskalender, NS 29 (1913)
(Munich: C. H. Beck, 1915), 331–2 (Thieme, 18 Oct. 1913); Sten.
Ber., cclxxxix. 4552 (Müller-Meiningen, Apr. 1913); Nipperdey,
‘Nationalidee’ (note 9), 153 (Arndt), 163 ff. (dedication,
monument); Hardtwig, ‘Bürgertum’ (note 9), 218; Stefan-Ludwig
Hoffmann, ‘Sakraler Monumentalismus um 1900. Das Leipziger
Völkerschlachtdenkmal’, in Der politische Totenkult.
Kriegerdenkmäler in der Moderne, ed. Reinhart Koselleck and
Michael Jeismann (Munich: Fink, 1994), 249–80; id., ‘Mythos
und Geschichte. Leipziger Gedenkfeiern der Völkerschlacht im
19. und frühen 20. Jahrhundert’, in Nation und Emotion.
1059
Deutschland und Frankreich im Vergleich. 19. und 20. Jahrhundert,
ed. Étienne François et al. (Göttingen: Vandenhoeck & Ruprecht,
1995), 111–32; Steffen Poser, ‘Die Jahrhundertfeier der
Völkerschlacht und die Einweihung des Völkerschlachtdenkmals
zu Leipzig 1913’, Feste und Feiern. Zum Wandel städtischer
Festkultur in Leipzig, ed. Katrin Keller (Leipzig: Edition Leipzig,
1994), 196–213; Wolfram Siemann, ‘Krieg und Frieden in
historischen Gedenkfeiern des Jahres 1913’, Festkultur (note 9),
ed. Düding et al., 298–320 (299: Rathenau); Ute Schneider,
Politische Festkultur im 19. Jahrhundert. Die Rheinprovinz von der
französischen Zeit bis zum Ende des Ersten Weltkrieges (1806–1918)
(Essen: Klartext, 1995), 319 ff.; Dietrich Schäfer, ‘Rede zur
Erinnerung an die Erhebung der deutschen Nation im Jahre
1813’, in Aufsätze, Vorträge und Reden, ii (Jena: G. Fischer, 1913),
438–59 (459); Die Wandervogelzeit. Quellenschriften zur deutschen
Jugendbewegung 1896–1919, ii, ed. Werner Kindt (Düsseldorf:
Diederichs, 1968), 501–5 (Wyneken, 12 Oct. 1913: 501); Rüdiger
vom Bruch, ‘Krieg und Frieden. Zur Frage der Militarisierung
deutscher Hochschullehrer und Universitäten im späten
Kaiserreich’, in Bereit zum Krieg. Kriegsmentalität im
wilhelminischen Deutschland 1890–1914, ed. Jost Dülffer and Karl
Holl (Göttingen: Vandenhoeck & Ruprecht, 1986), 74–98; George
L. Mosse, Nationalisierung (note 9), 83 ff.
28. Witt, Finanzpolitik (note 18), 356 ff.; Groh, Integration (note 20),
434 ff.; Wernecke, Wille (note 23), 180 ff. (208 ff.: on the ‘racial
war’; 210: quotes from Germania of 8 Mar. 1912); Krumeich,
Aufrüstung (note 24), 1 ff., 130 ff.; Mommsen, Bürgerstolz (note
23), 482 ff.; Schmidt, Imperialismus (note 6), 101 ff., 197 ff.;
1060
Gregor Schöllgen, Das Zeitalter des Imperialismus (Munich:
Oldenbourg, 1986), 68 ff.; Der Kaiser… Aufzeichnungen des Chefs
des Marinekabinetts Admiral Georg Alexander von Müller über die
Ära Wilhelms II. (Göttingen: Musterschmidt, 1965), 124–5
(conversation on 8 Dec 1912); Fritz Fischer, Krieg der Illusionen.
Die deutsche Politik von 1911 bis 1914 (Düsseldorf: Droste, 1969),
232 ff. (conversation on 8 Dec 1912), 270–1 (William II to
Ballin, 15 Dec 1912; emphases in original), tr. as War of Illusions:
German Policies from 1911 to 1914 (New York: Norton, 1975).
29. Sten. Ber., ccxc 5763 (Delbrück, 25 June 1913); Dieter
Gosewinkel, ‘Die Staatsangehörigkeit als Institution des
Nationalstaats. Zur Entstehung des Reichsund
Staatsangehörigkeitsgesetzes von 1913’, in Offene Staatlichkeit.
Festschrift für Ernst-Wolfgang Böckenförde, ed. Rolf Grawert et al.
(Berlin: Duncker & Humblot, 1995), 359–78; Roger Brubaker,
Staats-Bürger: Deutschland und Frankreich im historischen Vergleich
(Hamburg: Junius, 1994), origi. Citizenship and Nationahood in
France and Germany (Cambridge, MA: Harvard University Press,
1992); Mommsen, Bürgerstolz (note 23), 434 ff.; Huber,
Verfassungsgeschichte, iv (note 3), 581 ff.; Hans-Ulrich Wehler,
‘Symbol des halbabsolutistischen Herrschaftssystems: Der Fall
Zabern von 1913/14 als Verfassungskrise des Wilhelminischen
Kaiserreichs’, in Krisenherde (note 7), 65–84; David Schoenbaum,
Zabern 1913: Consensus Politics in Imperial Germany (London:
Allen & Unwin, 1982).
30. Kurt Riezler, Tagebücher—Aufsätze—Dokumente, ed. and introd.
Karl Dietrich Erdmann (Göttingen: Vandenhoeck & Ruprecht,
1061
1972), 185 (Bethmann Hollweg, 14 July 1914); Wolfgang
Steglich, Die Friedenspolitik der Mittelmächte 1917/18, i
(Wiesbaden: F. Steiner, 1964), 418 (Bethmann Hollweg to
Conrad Haussmann, 1917); Julikrise und Kriegsausbruch 1914, ed.
and introd. Immanuel Geiss, 2 vols. (Hanover: Verlag für
Literatur und Zeitgeschehen, 1963–4), ii. 184–5 (Wilhelm II to
State Secretary von Jagow, Foreign Office, 28 July 1914), 380–1
(Bethmann Hollweg to ambassador von Tschirschky in Vienna,
30 July 1914), 439–40 (Moltke, Berchtold, 31 July 1914), tr. as
July 1914: The Outbreak of the First World War. Selected Documents
(New York: Scribner, 1967); George F. Kennan, Bismarcks
europäisches System in der Auflösung. Die französisch-russische
Annäherung 1875 bis 1890 (Frankfurt: Propyläen, 1981), 12, orig.
The Decline of Bismarck’s European Order: Franco-Russian Relations,
1875–1890 (Princeton: Princeton University Press, 1979), 3;
Fischer, Krieg (note 28), 663 ff. (724: Müller’s journal, 1. Aug.
1914); Wolfgang J. Mommsen, ‘Innenpolitische
Bestimmungsfaktoren der deutschen Aussenpolitik vor 1914’, in
Der autoritäre Nationalstaat. Verfassung, Gesellschaft und Kultur im
deutschen Kaiserreich (Frankfurt: Fischer Taschenbuch, 1990),
316–57; id., ‘Der Topos vom unvermeidlichen Krieg:
Aussenpolitik und öffentliche Meinung im Deutschen Reich im
letzten Jahrzehnt vor 1914’, ibid. 380–406; id., Imperial Germany
1867–1918: Politics, Culture, and Society in an Authoritarian State
(London: Edward Arnold, 1995); id., Bürgerstolz (note 23), 535
ff.; Dieter Groh, ‘“Je eher, desto besser!” Innenpolitische
Faktoren für die Präventivkriegsbereitschaft des Deutschen
Reiches 1913/14’, Politische Vierteljahresschrift [PVS], 13 (1972),
1062
501–2; Krumeich, Aufrüstung (note 24), 243 ff.; Dietrich Geyer,
Der russische Imperialismus. Studien über den Zusammenhang von
innerer u. auswärtiger Politik 1860–1914 (Göttingen: Vandenhoeck
& Ruprecht, 1977), 220 ff., tr. as Russian Imperialism: The
Interaction of Domestic and Foreign Policy, 1860–1914 (New
Haven: Yale University Press, 1987); Der Erste Weltkrieg. Wirkung,
Wahrnehmung, Analyse, ed. Wolfgang Michalka (Munich: Piper,
1994); James Joll, The Origins of the First World War (London:
Longman, 1984); Der Erste Weltkrieg. Ursachen, Entstehung und
Kriegsziele, ed. Wolfgang Schieder (Cologne: Kiepenheuer &
Witsch, 1969); Niall Ferguson, Der falsche Krieg. Der Erste
Weltkrieg und das 20. Jahrhundert (Stuttgart: Deutscher
Taschebuch Verlag, 1999), 188 ff., orig. The Pity of War (London:
Allen Lane, 1998). On the ‘duel mentality’ see Ute Frevert,
Ehrenmänner. Das Duell in der bürgerlichen Gesellschaft (Munich:
Deutscher Taschenbuch Verlag, 1991), esp. 233 ff., tr. as Men of
Honour: A Social and Cultural History of the Duel (Cambridge:
Polity Press, 1995); Norbert Elias, ‘Die satisfaktionsfähige
Gesellschaft’, in Studien über die deutschen Machtkämpfe und
Habitusentwicklung im 19. und 20. Jahrhundert (Frankfurt:
Suhrkamp, 1989), 61–158.
31. Braunthal, Geschichte (note 17), i. 340 ff. (Stuttgart congress,
1907; Bebel: 342–3), 349 ff. (Basel congress, 1912); Sten. Ber.,
cccvi. 1–2 (Wilhelm II, 4 Aug 1914), 7 (Bethmann Hollweg, 4
Aug 1914), 8–9. (Haase, 4 Aug. 1914); Die Reichstagsfraktion der
deutschen Sozialdemokratie 1898 bis 1918, 2 vols., ed. Erich
Matthias and Eberhard Pikart, part 2 (Düsseldorf: Droste, 1966),
3–4 (session of 3–4 Apr. 1914); Das Kriegstagebuch des
1063
Reichstagsabgeordneten Eduard David 1914 bis 1918, ed. Susanne
Miller with Erich Matthias (Düsseldorf: Droste, 1966), 4–13
(entries from 1 Aug. to 4 Aug. 1914); Wilhelm Dittmann,
Erinnerungen, ed. and introd. Jürgen Rojahn, 3 vols. (Frankfurt:
Campus, 1995), ii. 241 ff.; Wolfgang Kruse, Krieg und nationale
Integration. Eine Neuinterpretation des sozialdemokratischen
Burgfriedensschlusses 1914/15 (Essen: Klartext, 1993), 42 ff. (51:
party leadership’s communication to the press on 19 July 1914);
Groh, Integration (note 16), 634; Susanne Miller, Burgfrieden und
Klassenkampf. Die deutsche Sozialdemokratie im Ersten Weltkrieg
(Düsseldorf: Droste, 1974), 31 ff. (on the fraction meeting on 3
Aug. 1914: 61 ff.); Gottfried Schramm, ‘1914: Sozialdemokraten
am Scheideweg’, in Wendepunkte deutscher Geschichte 1848–1990,
ed. Carola Stern and Heinrich August Winkler (Frankfurt: Fischer
Taschenbuch, 1994), 71–97; Jürgen Rojahn, ‘Arbeiterbewegung
und Kriegsbegeisterung: Die deutsche Sozialdemokratie 1870–
1914’, in Kriegsbegeisterung und mentale Kriegsvorbereitung.
Interdisziplinäre Studien, ed. Marcel van der Linden and Gottfried
Mergner (Berlin: Duncker & Humblot, 1991), 57–72; Thomas
Raithel, Das ‘Wunder’ der inneren Einheit. Studien zur deutschen
und französischen Öffentlichkeit des Ersten Weltkrieges (Bonn:
Bouvier, 1996); Christian Geinitz, Kriegsfurcht und
Kampfbereitschaft. Das Augusterlebnis in Freiburg. Eine Studie zum
Kriegsbeginn 1914 (Essen: Klartext, 1998); Benjamin Ziemann,
Front und Heimat. Ländliche Kriegserfahrungen im südlichen Bayern
1914–1923 (Essen: Klartext, 1997); Ernst Rudolf Huber, Deutsche
Verfassungsgeschichte seit 1789, v. Weltkrieg, Revolution und
Reichserneuerung 1914–1919 (Stuttgart: Kohlhammer, 1978), 27
1064
ff. Lenin quote from W. I. Lenin, ‘Der Krieg und die russische
Sozialdemokratie’, in Werke (Berlin: Dietz, 1950), xxi. 11–21
(20), tr. as ‘The War and Russian Social Democracy’, in Collected
Works (New York: International Publishers, 1927), 25–34.
32. Ernst von Dryander, Erinnerungen aus meinem Leben (Bielefeld:
Velhagen & Klasing, 1922), 276; Arlie J. Hoover, The Gospel of
Nationalism: German Patriotic Preaching from Napoleon to Versailles
(Stuttgart: F. Steiner, 1986), 53 (Lehmann); Wilhelm Pressel, Die
Kriegspredigt 1914–1918 in der evangelischen Kirche Deutschlands
(Göttingen: Vandenhoeck & Ruprecht, 1967), 204 (Dibelius);
Karl Hammer, Deutsche Kriegstheologie (1870–1918) (Munich:
Kösel-Verlag, 1971), 242 (Rade), 266 (Poertner), 241–2
(Faulhaber); Otto Seeber, ‘Kriegstheologie und Kriegspredigten in
der Evangelischen Kirche Deutschlands im Ersten und Zweiten
Weltkrieg’, in Kriegsbegeisterung (note 31), ed. von der Linden and
Mergner, 233–58; Gunter Brakelmann, ‘Konfessionalismus und
Nationalismus’, Bochumer Beiträge zur Nationalismusdebatte, ed.
Bernd Faulenbach et al. (Essen: Klartext, 1997), 36–50; Die
protestantischen Kirchen Europas im Ersten Weltkrieg. Ein Quellen-
und Arbeitsbuch, ed. Gerhard Besier (Göttingen: Vandenhoeck &
Ruprecht, 1984); Heinrich Missalla, ‘Gott mit uns’: Die deutsche
katholische Kriegspredigt 1914–1918 (Munich: KöselVerlag, 1968),
85, 89 (Peters); Richard van Dülmen, ‘Der deutsche
Katholizismus und der Erste Weltkrieg’, Francia, 2 (1974), 347–
76. The Bible quote ‘If God is for us… ’ from Romans 8: 31; ‘Give
to Caesar what is Caesar’s… ’ from Matthew 22: 21. On Arndt’s
‘German God’ see above, p. 57.
1065
33. Aufrufe und Reden deutscher Professoren im Ersten Weltkrieg, ed.
and introd. Klaus Böhme (Stuttgart: Reclam, 1975), 47–9 (‘Aufruf
an die Kulturwelt’); Jürgen von Ungern-Sternberg and Wolfgang
von Ungern-Sternberg, Der Aufruf ‘An die Kulturwelt’. Das Manifest
der 93 und die Anfänge der Kriegspropaganda im Ersten Weltkrieg
(Stuttgart: F. Steiner, 1996); Johann Plenge, Der Krieg und die
Volkswirtschaft, 2nd edn (Münster: Borgemeyer & Co., 1915),
173–4; id., 1789 und 1914. Die symbolischen Jahre in der
Geschichte des politischen Geistes (Berlin: J. Springer, 1916), 82;
Rudolf Kjellén, Die Ideen von 1914—Eine weltgeschichtliche
Perspektive (Leipzig: S. Hirzel, 1915), esp. 146; Max Scheler, Der
Genius des Krieges und der deutsche Krieg (Leipzig: Verlag der
Weissen Bücher, 1915), 54, 73–4, 100, 150; Paul Lensch, Die
deutsche Sozialdemokratie in ihrer grossen Krisis (Hamburg: Auer &
Co., 1916), 7 ff. (9: ‘world war means world revolution!’); id.,
Drei Jahre Weltrevolution (Berlin: S. Fischer, 1917), esp. 44 ff., 86
ff., 172 ff.; Werner Sombart, Händler und Helden. Patriotische
Besinnungen (Munich: Duncker & Humblot, 1915), 84–5, 116,
121, 142–3; Thomas Mann, ‘Gedanken im Kriege’ (1914), in
Gesammelte Werke in dreizehn Bänden (Frankfurt: Fischer
Taschenbuch Verlag, 1990), xiii. 527–45; id., ‘Friedrich und die
grosse Koalition’, ibid. ii. 76–135; id., Betrachtungen eines
Unpolitischen (1918), ibid. xii. 1–589 (1,587); id., ‘Leiden und
Grösse Richard Wagners’ (lecture in Munich, 10 Feb. 1933), ibid.
ix. 363–426 (419); id., Reflections of a Nonpolitical Man (New
York: F. Ungar, 1983); id., Freud, Goethe, Wagner (New York: A.
A. Knopf, 1937); Johannes Burkhardt, ‘Kriegsgrund Geschichte?
1870, 1813, 1756—historische Argumente und Orientierungen
1066
bei Ausbruch des Ersten Weltkrieges’, in id. et al., Lange und
kurze Wege in den Ersten Weltkrieg. Vier Augsburger Beiträge zur
Kriegsursachenforschung (Munich: Ernst Vögel, 1996), 9–86; Klaus
von See, Die Ideen von 1789 und die Ideen von 1914. Völkisches
Denken in Deutschland zwischen Französischer Revolution und
Erstem Weltkrieg (Frankfurt: Athenaion, 1975), 108 ff.; Reinhard
Rürup, ‘Der “Geist von 1914” in Deutschland. Kriegsbegeisterung
und Ideologisierung des Krieges im Ersten Weltkrieg’, in
Ansichten vom Krieg. Vergleichende Studien zum Ersten Weltkrieg in
Literatur und Gesellschaft, ed. Bernd Hüppauf (Königstein: Forum
Academicum, 1984), 1–30; Wolfgang Kruse, ‘Die
Kriegsbegeisterung im Deutschen Reich zu Beginn des Ersten
Weltkrieges’, in Kriegsbegeisterung (note 31), ed. van der Linden
and Mergner, 73–87; Christoph Jahr, ‘“Das Krämervolk der eitlen
Briten.” Das deutsche Englandbild im Ersten Weltkrieg’, in
Feindbilder (note 10), ed. id. et al., 115–42; Robert Sigel, Die
Lensch-Cunow-Haenisch-Gruppe. Eine Studie zum rechten Flügel der
SPD im Ersten Weltkrieg (Berlin: Duncker & Humblot, 1976); Hans
Kohn, Wege und Irrwege. Vom Geist des deutschen Bürgertums
(Düsseldorf: Droste, 1962), 308 ff., orig. The Mind of Germany:
The Education of a Nation (New York: Scribner, 1960); Ludwig
Dehio, ‘Gedanken über die deutsche Sendung 1900–1918’, in
Deutschland und die Weltpolitik im 20. Jahrhundert (Munich:
Oldenbourg, 1955), 71–106; Hermann Lübbe, Politische
Philosophie in Deutschland. Studien zu ihrer Geschichte (Stuttgart: B.
Schwabe 1963), 173 ff.; Bernd Faulenbach, Ideologie des
deutschen Weges. Die deutsche Geschichte in der Historiographie
zwischen Kaiserreich und Nationalsozialismus (Munich: C. H. Beck,
1067
1980), 122 ff.; Helmut Fries, Die grosse Katharsis. Der Erste
Weltkrieg in der Sicht deutscher Dichter und Gelehrter, 2 vols.
(Konstanz: Verlag am Hockgraben, 1994); Wolfgang J.
Mommsen, ‘Der Geist von 1914. Das Programm eines politischen
“Sonderwegs” der Deutschen’, in Nationalstaat (note 30), 407–21;
Kultur und Krieg. Die Rolle der Intellektuellen, Künstler und
Schriftsteller im Ersten Weltkrieg, ed. id. (Munich: Oldenbourg,
1996); Klaus Vondung, ‘Deutsche Apokalypse 1914’, in Das
wilhelminische Bildungsbürgertum. Zur Sozialgeschichte seiner Ideen,
ed. id. (Göttingen: Vandenhoeck & Ruprecht, 1976), 153–71; id.,
Die Apokalypse in Deutschland (Munich: Deutscher Taschenbuch
Verlag, 1988), 189 ff., orig. The Apocalypse in Germany
(Columbia, MO: University of Missouri Press, 2000). On the
distribution of the concept of the ‘Volksgemeinschaft’ see Gunther
Mai, ‘“Verteidigungskrieg”. Staatliche Selbstbehauptung,
nationale Solidarität und soziale Befreiung in Deutschland in der
Zeit des Ersten Weltkrieges (1900–1925)’, in Weltkrieg (note 30),
ed. Michalka, 583–607. Quote from Heinrich von Treitschke
from id., Politik. Vorlesungen, gehalten an der Universität zu Berlin,
2 vols. (Leipzig: S. Hirzel, 1899–1900), ii. 362.
34. Heinrich Class, Wider den Strom. Vom Werden und Wachsen der
nationalen Opposition im alten Reich (Leipzig: K. F. Koehler, 1932),
341 ff.; Ursachen und Folgen. Vom deutschen Zusammenbruch 1918
und 1945 bis zur staatlichen Neuordnung Deutschlands in der
Gegenwart, ed. Herbert Michaelis and Ernst Schraepler (Berlin:
Dokumenten-Verlag, 1958-II), i. 351–68 (petition of the six
economic organizations to the chancellor on 20 May 1915 (358:
‘Germanization’); confidential memorandum from German
1068
university teachers and officials to the chancellor on 20 June
1915; message from the SPD party and fraction leadership to the
chancellor on 25 June 1915); Hans Delbrück, ‘Der
Kanzlerwechsel.—Die Friedensresolution.—Lloyd Georges
Antwort’, Preussische Jahrbücher, 169 (1917), 302–19 (306–7:
Wolff’s memorandum); Fritz Fischer, Griff nach der Weltmacht.
Die Kriegszielpolitik des kaiserlichen Deutschland 1914/18, 3rd edn
(Düsseldorf: Droste, 1964), 123 ff. (Bethmann Hollweg), 120 ff.
(Class, Erzberger, Thyssen), tr. as Germany’s Aims in the First
World War (New York: W. W. Norton, 1967); Klaus Schwabe,
‘Ursprung und Verbreitung des alldeutschen Annexionismus in
der deutschen Professorenschaft im Ersten Weltkrieg’, VfZ 14
(1966), 105–38 (quote: 131, numbers and names: 127, 132); id.,
Wissenschaft und Kriegsmoral. Die deutschen Hochschullehrer und
die politischen Grundfragen des Ersten Weltkrieges (Göttingen:
Musterschmidt Verlag, 1969), 19 ff.; Annelise Thimme, Hans
Delbrück als Kritiker der Wilhelminischen Epoche (Düsseldorf:
Droste, 1955), 116 ff.; Heinz Hagenlücke, Deutsche
Vaterlandspartei. Die nationale Rechte am Ende des Kaiserreiches
(Düsseldorf: Droste, 1997), 49 ff.; Friedrich Naumann,
Mitteleuropa (Berlin: G. Reimer, 1915), 31, 40–2, 101; Egmont
Zechlin, Die deutsche Politik und die Juden im Ersten Weltkrieg
(Göttingen: Vandenhoeck & Ruprecht, 1969), 516 ff. (518–19:
Liebig, 525: Erzberger, 550: Oppenheimer); Werner Jochmann,
‘Die Ausbreitung des Antisemitismus in Deutschland 1914–1923’,
in Gesellschaftskrise (note 26), 99–170; George L. Mosse, ‘The
Jews and the German War Experience 1914–1918’, Leo Baeck
Memorial Lecture 21 (New York: Leo Baeck Institute, 1977);
1069
Jürgen Kocka, Klassengesellschaft im Krieg. Deutsche
Sozialgeschichte 1914–1918, 2nd edn (Göttingen: Vandenhoeck &
Ruprecht, 1978), esp. 96 ff., tr. as Facing Total War: German
Society, 1914–1918 (Cambridge, MA: Harvard University Press,
1984); Peter Graf Kielmansegg, Deutschland und der Erste
Weltkrieg, 2nd edn (Stuttgart: Klett-Cotta, 1980), 129 ff., 385 ff.;
Jarausch, Chancellor (note 23), 349 (‘policy of the diagonal’).
35. Miller, Burgfrieden (note 31), 75 ff., 156 ff., 283 ff.; Braunthal,
Geschichte (note 17), ii. 50 ff. (64–5: Kienthal resolutions);
Huber, Verfassungsgeschichte, v (note 31), 101 ff., 154 ff.; Hartwin
Spenkuch, Das Preussische Herrenhaus. Adel und Bürgertum in der
Ersten Kammer des Landtags 1854–1918 (Düsseldorf: Droste,
1998), 124 ff.; Fischer, Griff (note 34), 512 ff.; Gerald D.
Feldman, Armee, Industrie und Arbeiterschaft in Deutschland 1914–
1918 (Berlin: J. H. W. Dietz, 1985), 169 ff., 243 ff.; orig. Army,
Industry, and Labor in Germany, 1914–1918 (Princeton: Princeton
University Press, 1966); Arthur Rosenberg, Entstehung der
Weimarer Republik (1928; repr. Frankfurt: Europäische
Verlagsanstalt, 1961), 160 ff.; Klaus Epstein, Matthias Erzberger
und das Dilemma der deutschen Demokratie (Berlin: Verlag
Annedore Leber, 1962), 186 ff., orig. Matthias Erzberger and the
Dilemma of German Democracy (Princeton: Princeton University
Press, 1959).
36. Der Interfraktionelle Ausschuss 1917/18, 2 parts, ed. Erich
Matthias (Düsseldorf: Droste, 1959), part 1, 3–118 (July crisis),
213–602 (from Michaelis to Hertling); Max Weber, ‘Die Lehren
der deutschen Kanzlerkrisis’, in Gesammelte politische Schriften,
1070
ed. Johannes Winckelmann, 2nd edn (Tübingen: Mohr, 1958),
211–16 (213–14, 216); Georg Michaelis, Für Volk und Staat
(Berlin: Furche-Verlag, 1922), 321; Sten. Ber., cccx. 3572
(Michaelis, 19 July 1917); Parteiprogramme (note 10), ed.
Mommsen, 417–20 (founding manifesto of the German
Fatherland Party, 2 Sept. 1917), 420–1 (manifesto of the People’s
Federation for Liberty and Fatherland, 14 Nov. 1917);
Hagenlücke, Vaterlandspartei (note 34), 143–4 (foundation, social
structure), 180 (membership), 192 ff. (war goals), 362 ff.
(People’s Federation); the article ‘Volksbund für Freiheit und
Vaterland, 1917–1920’, in Lexikon, iv (note 2), ed. Fricke et al.,
414–19 (membership: 414); Stegmann, Erben (note 4), 497 ff.;
Fischer, Griff (note 34), 425 ff.; Ursula Ratz, Zwischen
Arbeitsgemeinschaft und Koalition. Bürgerliche Sozialreformer und
Gewerkschaften im Ersten Weltkrieg (Munich: K. G. Saur, 1994),
307 ff.; Günter Brakelmann, Der deutsche Protestantismus im
Epochenjahr 1917 (Witten: Luther-Verlag, 1974); Huber,
Verfassungsgeschichte, v (note 31), 372 ff. (argument about the
‘breakthrough to parliamentary government’ in November 1917).
37. Sten. Ber., cccvi. 6–7 (Bethmann Hollweg, 4 Aug. 1914); Karl
Kautsky, ‘Die Diktatur des Proletariats’ (1918), in Die Diktatur des
Proletariats/W. I. Lenin, Die proletarische Revolution und der
Renegat Kautsky/Karl Kautsky, Terrorismus und Kommunismus
(Berlin: Dietz, 1990), 7–87 (33, 36–7 39; emphasis in original);
Clara Zetkin, Mit Entschiedenheit für das Werk der Bolschewiki! Aus
einem Brief an eine Konferenz des Reichsausschusses und der
Frauenkonferenz der USPD (early summer 1919), in Ausgewählte
Reden und Schriften, 3 vols., ii (Berlin: Dietz, 1960), 8–40 (26);
1071
Rosa Luxemburg, ‘Die russische Revolution’ (1918), in Schriften
(note 20), iii. 106–41 (128 ff., quote: 134), tr. as ‘The Russian
Revolution’ and ‘Leninism or Marxism?’ (Ann Arbor: University of
Michigan Press, 1961); Lenins Rückkehr nach Russland. Die
deutschen Akten, ed. Werner Hahlweg (Leiden: E. J. Brill, 1957);
Winfried B. Scharlau and Zbynlk A. Zeman, Freibeuter der
Revolution. Parvus-Helphand. Eine politische Biographie (Cologne:
Verlag Wissenschaft und Politik, 1964); Deutschland und die
Russische Revolution, ed. Helmut Neubauer (Stuttgart:
Kohlhammer, 1968); Lösche, Bolschewismus (note 20), 160 ff.;
Jürgen Zarusky, Die deutschen Sozialdemokraten und das
sowjetische Modell. Ideologische Auseinandersetzung und
aussenpolitische Konzeption (Munich: Oldenbourg, 1992), 39 ff.;
Uli Schöler, ‘Despotischer Sozialismus’ oder ‘Staatssklaverei’. Die
theoretische Verarbeitung der sowjetrussischen Entwicklung in der
Sozialdemokratie Deutschlands und Österreichs 1917–1919, 2 vols.,
i (Münster: Lit., 1990), 84 ff.; Deutschland und die Russische
Revolution, ed. Gerd Koenen and Lew Kopelew (Munich: W. Fink,
1998); Manfred Hildermeier, Die Russische Revolution 1905–1921
(Frankfurt: Suhrkamp, 1989), 229 ff.; Dietrich Geyer, Die
Russische Revolution. Historische Probleme und Perspektiven, 2nd
edn (Göttingen: Vandenhoeck und Ruprecht, 1977), 107 ff., tr. as
The Russian Revolution (New York: St Martin’s Press, 1987);
Helmut Altrichter, Russland 1917. Ein Land auf der Suche nach
sich selbst (Paderborn: Schöningh, 1997). On Bismarck in 1866
see Otto Pflanze, Bismarck. Der Reichskanzler (Munich: C. H.
Beck, 1998), 313 ff., orig. Bismarck and the Development of
Germany: The Period of Unification, 1815–1871 (Princeton:
1072
Princeton University Press, 1990).
38. Winfried Baumgart, Deutsche Ostpolitik 1918. Von Brest-Litowsk
bis zum Ende des Ersten Weltkrieges (Vienna: Oldenbourg, 1966);
Georg von Rauch, ‘Sowjetrussland von der Oktoberrevolution bis
zum Sturz Chruschtschows 1917–1964’, Europa im Zeitalter der
Weltmächte (= Handbuch der europäischen Geschichte, ed. Theodor
Schieder, vii/1), ed. Theodor Schieder (Stuttgart: Klett-Cotta,
1979), 481–521 (numbers: 485); Fischer, Griff (note 34), 627 ff.;
Hildebrand, Reich (note 6), 363 ff.; Huber, Verfassungsgeschichte,
v (note 31), 432 ff. (January strike), 447 ff. (Reichstag debate,
25–6 Feb. 1918); Miller, Burgfrieden (note 31), 358 ff.;
Rosenberg, Entstehung (note 35), 183 ff. (187, 189); Pressel,
Kriegspredigt (note 32), 305–6 (Doehring, 3 Feb. 1918); Sten. Ber.,
cccxi. 4, 171 (Heydebrand, 26 Feb. 1918); Dittmann,
Erinnerungen (note 31), ii. 527 ff.
39. Arno J. Mayer, Political Origins of the New Diplomacy, 1917–1918,
2nd edn (New York: Vintage Books, 1970), 329 ff.; Klaus
Schwabe, Deutsche Revolution und Wilson-Friede. Die amerikanische
und die deutsche Friedensstrategie zwischen Ideologie und
Machtpolitik 1918/19 (Düsseldorf: Droste, 1971), 17 ff.;
Hagenlücke, Vaterlandspartei (note 34), 195 ff.; Huber,
Verfassungsgeschichte, v (note 31), 428 ff., 500–1; Kielmansegg,
Deutschland (note 34), 629 ff.; Erich Ludendorff, Meine
Kriegserinnerungen 1914–1918 (Berlin: E. S. Mittler und Sohn,
1919), 547 ff. (547, 551); id., Ludendorff’s Own Story, August
1914November 1918 (New York: Harper, 1919); Karl-Ludwig Ay,
Die Entstehung einer Revolution. Die Volksstimmung in Bayern
1073
während des Ersten Weltkrieges (Berlin: Duncker & Humblot,
1968), 101 (Brettreich); Ernst Troeltsch, Spektator-Briefe. Aufsätze
über die deutsche Revolution und die Weltpolitik 1918/22
(Tübingen: Mohr, 1924), 10; Reichstagsfraktion (note 31), 458
(Geck); Wilhelm Deist, ‘Der militärische Zusammenbruch des
Kaiserreiches. Zur Realität der “Dolchstosslegende”’, in Das
Unrechtsregime. Internationale Forschungen über den
Nationalsozialismus, I, ed. Ursula Büttner (Hamburg: Christians,
1986), 101–29; id., ‘Verdeckter Militärstreik im Kriegsjahr
1918?’, in Der Krieg des kleinen Mannes. Eine Militärgeschichte von
unten, ed. Wolfram Wette (Munich: Piper, 1992), 146–68;
Wolfgang Kruse, ‘Krieg und Klassenheer. Zur Revolutionierung
der deutschen Armee im Ersten Weltkrieg’, GG 22 (1996), 530–
61; Christoph Jahr, Gewöhnliche Soldaten. Desertion und Deserteure
im deutschen und britischen Heer 1914–1918 (Göttingen:
Vandenhoeck & Ruprecht, 1998). Wilson first made the
statement ‘The world must be made safe for democracy’ for the
first time on 22 Jan. 1917 in a speech to the American Congress.
August Heckscher, Woodrow Wilson (New York: Scribner, 1991),
440.
40. Reichstagsfraktion (note 31), 417–460 (session on 23 Sept. 1918;
442: Ebert); Interfraktioneller Ausschuss (note 36), part 2, 668–9
(memorandum of 21 Sept. 1918), 710 ff. (28 Sept. 1918);
Albrecht von Thaer, Generalstabsdienst an der Front und in der
OHL. Aus Briefen und Tagebuchaufzeichnungen 1915–1919, ed.
Siegfried A. Kaehler (Göttingen: Vandenhoeck & Ruprecht,
1958), 234–5 (Ludendorff, 1 Oct. 1918); Friedrich von Payer,
Von Bethmann Hollweg bis Ebert. Erinnerungen und Bilder
1074
(Frankfurt: Frankfurter Societäts-Druckerei, 1923), 82; Karl
Kautsky, Terrorismus und Kommunismus. Ein Beitrag zur
Naturgeschichte der Revolution (1919), in Diktatur (note 37), 177–
347 (233 ff.); W. I. Lenin, ‘Das sozialistische Vaterland in
Gefahr!’ (appeal from the council of the people’s commissars, 21
Feb. 1918), in Werke (note 31), xxvii. 15–16, tr. as ‘The Socialist
Fatherland is in Danger!’, in Collected Works (note 31), xxvii. 30–
3; Stegmann, Erben (note 4), 515 (Class, 3 Oct. 1918); Jochmann,
‘Ausbreitung’ (note 34), 118–19, 388 (Gebsattel/Class, 15 Oct.
1918), 120–1 (Class, 19–20 Oct. 1918); Huber,
Verfassungsgeschichte, v (note 31), 521. Quote from Kleist from
Germania an ihre Kinder, referring to Napoleon: ‘Strike him dead!
On Judgement Day | your reasons won’t be questioned’, Heinrich
von Kleist, Sämtliche Werke und Briefe, i, 2nd edn (Munich: C.
Hanser, 1994), 27.
41. Huber, Verfassungsgeschichte, v (note 31), 551 ff. (continuance of
the war ‘with the utmost determination’: 577); Troeltsch,
Spektator-Briefe (note 39), 21; Wolfgang Sauer, ‘Das Scheitern der
parlamentarischen Monarchie’, in Vom Kaiserreich zur Weimarer
Republik, ed. Eberhard Kolb (Cologne: Kiepenheuer & Witsch,
1972), 77–99 (84); Die Weizsäcker-Papiere 1900–1932, ed.
Leonidas E. Hill (Berlin: Propyläen, 1982), 309 (journal entry, 28
Oct. 1918); Leonidas E. Hill, ‘Signal zur Konterrevolution? Der
Plan zum Vorstoss der deutschen Hochseeflotte am 30. Oktober
1918’, VfZ 36 (1988), 114–29; Wilhelm Deist, ‘Die Politik der
Seekriegsleitung und die Rebellion der Flotte Ende Oktober
1918’, VfZ 14 (1966), 325–43; Gerhard Paul Gross, Die
Seekriegsführung der Kaiserlichen Marine im Jahre 1918 (Frankfurt:
1075
P. Lang, 1989), 390 ff.
42. Dirk Dähnhardt, Revolution in Kiel. Der Übergang vom Kaiserreich
zur Weimarer Republik 1918/19 (Neumünster: K. Wachholtz,
1978); Die deutsche Revolution 1918–1919. Dokumente, ed.
Gerhard A. Ritter and Susanne Miller, 2nd edn (Hamburg:
Hoffmann und Kampe, 1975), 41–64 (58–9: Brunswick, Munich;
62 ff.: Cologne), 68 ff. (Groener, 8 Nov. 1918); Prinz Max von
Baden, Erinnerungen und Dokumente, ed. Golo Mann and Andreas
Burckhardt (new edn, Stuttgart: Klett, 1968), 584 (Vorwärts, 5
Nov. 1918), 567 (Ebert, 7 Nov. 918), 588 (Prussian ministry of
war, 8 Nov. 1918; emphases in original); Heinrich August
Winkler, Von der Revolution zur Stabilisierung. Arbeiter und
Arbeiterbewegung in der Weimarer Republik 1918–1924, 2nd edn
(Berlin: J. H. W. Dietz, 1985), 27 ff.
43. Prinz Max, Erinnerungen (note 42), 579–80; Reichstagsfraktion
(note 31), ii. 513–14 (ultimatum of 7 Nov. 1918); Die Regierung
des Prinzen Max von Baden, ed. Erich Matthias and Rudolf Morsey
(Düsseldorf: Droste, 1962), 579 (7 Nov. 1918), 583–612 (talks
over the SPD demands, 8. Nov. 1918), 609–10 (women’s
suffrage: 8 Nov. 1918); Schulthess’ Europäischer
Geschichtskalender, ns 34th year, 1918, part 1 (Munich: C. H.
Beck, 1922), 422–31 (SPD announcements and report in the
B[erliner] Z[eitung] am Mittag); Reinhold Patemann, Der Kampf
um die preussische Wahlreform im Ersten Weltkrieg (Düsseldorf:
Droste, 1964), 202 ff.; Huber, Verfassungsgeschichte, v (note 31),
600 ff.; Winkler, Revolution (note 42), 40 ff.
44. Prinz Max, Erinnerungen (note 42), 596 ff.; Die Regierung der
1076
Volksbeauftragten 1918/19, 2 vols., ed. Susanne Miller and
Heinrich Potthoff (Düsseldorf: Droste, 1969), i. 3–8 (9 Nov.
1918); Revolution (note 42), ed. Ritter and Miller, 64 ff. (9 Nov.
1918, 79–80: Ebert’s declarations on 9 Nov. 1918, 86–92:
formation of the government and assembly in the Zirkus Busch,
9–10 Nov. 1918); Philipp Scheidemann, Der Zusammenbruch
(Berlin: Verlag für Sozialwissenschaft, 1921), 174–6 (2 Oct.
1918); id., Memoiren eines Sozialdemokraten, 2 vols. (Dresden: C.
Reissner 1928), ii. 311 ff.; id., The Making of New Germany: The
Memoirs of Philipp Scheidemann (New York: D. Appleton and Co.,
1929); Manfred Jessen-Klingenberg, ‘Die Ausrufung der Republik
durch Philipp Scheidemann am 9. 11. 1918’, GWU 19 (1968),
649–56; Georg Kotowski, Friedrich Ebert. Eine politische
Biographie, i Der Aufstieg eines deutschen Arbeiterführers 1871 bis
1917 (Wiesbaden: F. Steiner, 1963); Winkler, Revolution (note
42), 45 ff.
45. Richard Müller, Vom Kaiserreich zur Republik, 2 vols., ii Die
Novemberrevolution (Vienna: Malik-Verlag, 1925), 17 (Berliner
Tageblatt, 10 Nov. 1918); Troeltsch, Spektator-Briefe (note 39),
24; Pressel, Kriegspredigt (note 32), 308–9 (Doehring, 27 Oct.
1918); Martin Greschat, Der deutsche Protestantismus im
Revolutionsjahr 1918–1919 (Witten: Luther-Verlag, 1974); Max
Weber, Wirtschaft und Gesellschaft, study edn, ed. Johannes
Winckelmann, pt 1 (Cologne: Kiepenheuer & Witsch, 1964), 27,
197 (emphases in original), tr. as Economy and Society: An Outline
of Interpretive Sociology (New York: Bedminster Press, 1968);
Heinrich August Winkler, ‘Vom Kaiserreich zur Republik. Der
historische Ort der Revolution von 1918/19’, in Streitfragen der
1077
deutschen Geschichte. Essays zum 19. und 20. Jahrhundert (Munich:
C. H. Beck, 1997), 52–70.
1. Modris Eksteins, Tanz über Gräben. Die Geburt der Moderne und
der Erste Weltkrieg (Reinbek b. Hamburg: Rowohlt, 1990); esp.
213 ff.; orig. Rites of Spring: The Great War and the Birth of the
Modern Age (Boston: Houghton Mifflin, 1989), Kriegserlebnis. Der
Erste Weltkrieg in der literarischen Gestaltung und symbolischen
Deutung der Nationen, ed. Klaus Vondung (Göttingen:
Vandenhoeck & Ruprecht, 1980); Kriegserfahrungen. Studien zur
Sozial- und Mentalitätsgeschichte des Ersten Weltkriegs, ed. Gerhard
Hirschfeld et al. (Essen: Klartext, 1997); Keiner fühlt sich hier
mehr als Mensch… Erlebnis und Wirkung des Ersten Weltkriegs, ed.
id. et al. (Essen: Klartext, 1993); Eine Welt von Feinden. Der Grosse
Krieg 1914–1918, ed. Wolfgang Kruse (Frankfurt: Fischer
Taschenbuch Verlag, 1997); Krieg im Frieden. Die umkämpfte
Erinnerung an den Ersten Weltkrieg, ed. Bernd Ulrich and Benjamin
Ziemann (Frankfurt: Fischer Taschenbuch Verlag, 1997).
2. Susanne Miller, ‘Das Ringen um “die einzige grossdeutsche
Republik”. Die Sozialdemokratie in Österreich und im Deutschen
Reich zur Anschlussfrage 1918/19’, AfS 11 (1971), 1–67; Alfred
D. Low, Die Anschlussbewegung in Österreich und Deutschland,
1918–1919, und die Pariser Friedenskonferenz (Vienna: W.
Braumüller, 1975), 7 ff.; Peter Borowsky, ‘Die “bolschewistische
Gefahr” und die Ostpolitik der Volksbeauftragten in der
1078
Revolution 1918/19’, in Industrielle Gesellschaft und politisches
System. Beiträge zur politischen Sozialgeschichte. Festschrift für Fritz
Fischer, ed. Dirk Stegmann et al. (Bonn: Verlag Neue Gesellschaft,
1978), 389–403; Henning Köhler, Novemberrevolution und
Frankreich. Die französische Deutschland-Politik 1918–1919
(Düsseldorf: Droste, 1980); Eberhard Kolb, ‘Internationale
Rahmenbedingungen einer demokratischen Neuordnung in
Deutschland 1918/19’, in Umbrüche deutscher Geschichte
1866/71. 1918/19. 1929/33 (Munich: Oldenbourg, 1993), 261–
87; Richard Löwenthal, ‘Die deutsche Sozialdemokratie in
Weimar und heute. Zur Problematik der “versäumten”
demokratischen Revolution’, in Gesellschaftswandel und
Kulturkrise. Zukunftsprobleme der westlichen Demokratien
(Frankfurt: Fischer Taschenbuch Verlag, 1979), 197–211; id.,
Social Change and Cultural Crisis (New York: Columbia University
Press, 1984); Eduard Bernstein, Die deutsche Revolution von
1918/19. Geschichte der Entstehung und ersten Arbeitsperiode der
deutschen Republik, (1921), ed. and introd. Heinrich August
Winkler, annotated Teresa Löwe (Bonn: J. H. W. Dietz, 1998), 65
(Liebknecht), 237–8.
3. Susanne Miller, Die Bürde der Macht. Die deutsche
Sozialdemokratie 1918–1920 (Düsseldorf: Droste, 1978), 104 ff.
(Hilferding: 107); Wolfgang Elben, Das Problem der Kontinuität in
der deutschen Revolution. Die Politik der Staatssekretäre und der
militärischen Führung vom November 1918 bis Februar 1919
(Düsseldorf: Droste, 1965); Wolfgang Runge, Politik und
Beamtentum im Parteienstaat. Die Demokratisierung der politischen
Beamten in Preussen zwischen 1918 und 1933 (Stuttgart: Klett,
1079
1965); Ulrich Kluge, Soldatenräte und Revolution. Studien zur
Militärpolitik in Deutschland 1918/19 (Göttingen: Vandenhoeck &
Ruprecht, 1975), 206 ff.; Eberhard Kolb, Die Arbeiterräte in der
deutschen Innenpolitik 1918–1919 (Düsseldorf: Droste, 1962), 359
ff.; Gerald D. Feldman and Irmgard Steinisch, Industrie und
Gewerkschaften 1918–1924. Die überforderte
Zentralarbeitsgemeinschaft (Stuttgart: Deutsche Verlags-Anstalt,
1985); Jens Flemming, Landwirtschaftliche Interessen und
Demokratie. Ländliche Gesellschaft, Agrarverbände und Staat 1890–
1925 (Bonn: Verlag Neue Gesellschaft, 1978), 252 ff.; Stephanie
Merkenich, Grüne Front gegen Weimar. Reichslandbund und
agrarischer Lobbyismus 1918–1933 (Düsseldorf: Droste, 1998);
Heinrich August Winkler, Von der Revolution zur Stabilisierung.
Arbeiter und Arbeiterbewegung in der Weimarer Republik, 1918–
1924, 2nd edn (Berlin: J. H. W. Dietz, 1985), 68 ff. Quote from
Ebert in Sten. Ber., cccxxvi. 2–3 (6 Feb. 1919); quote from Rosa
Luxemburg in ead., ‘Die Nationalversammlung’, in Gesammelte
Werke, iv (Berlin: Dietz, 1974), 407–10 (408).
4. Allgemeiner Kongress der Arbeiter- und Soldatenräte Deutschlands.
Vom 16. bis 21. Dezember 1918 im Abgeordnetenhaus zu Berlin
(Berlin: Zentralrat der Sozialistischen Republik Deutschlands,
1919), 209–24 (Cohen-Reuss), 226–36 (Däumig), 282, 288, 230
(motions and ballots on the council system and the date for
elections), 127–43, 180–91 (the military question), 252, 288–
300, 309 (Central Council); Winkler, Revolution (note 3), 100 ff.
(121–2: Liebknecht); Miller, Bürde (note 3), 112 ff.; Kolb,
Arbeiterräte (note 3), 197 ff.; Kluge, Soldatenräte (note 3), 250 ff.;
Arthur Rosenberg, Geschichte der Weimarer Republik, (1935; repr.
1080
Frankfurt: Europäische Verlagsanstalt, 1961), 50 ff. (52); Hagen
Schulze, Freikorps und Republik 1918–1920 (Boppard: H. Boldt,
1969), 22 ff.; Robert G. L. Waite, Vanguard of Nazism: The Free
Corps Movement in Postwar Germany 1918–1923 (Cambridge, MA:
Harvard University Press, 1952; repr. New York: W. W. Norton,
1969), 13 ff.; Wolfram Wette, Gustav Noske. Eine politische
Biographie (Düsseldorf: Droste, 1987), 281 ff.
5. Deutsche Parteiprogramme, ed. Wilhelm Mommsen (Munich: Isar
Verlag, 1960), 481–6 (manifesto and platform of the German
Center Party, 30 Dec. 1918), 519–31 (DVP platform, 18 Dec.
1918); Die deutsche Revolution 1918–1919. Dokumente, ed.
Gerhard A. Ritter and Susanne Miller, 2nd edn (Hamburg:
Hoffmann & Campe, 1975), 296–8 (founding manifesto of the
DNVP, 24 Nov. 1918), 300 (DNVP election statement, 22 Dec.
1918), 311–13 (founding manifesto of the DDP, 16 Nov. 1918);
Nationalliberalismus in der Weimarer Republik. Die Führungsgremien
der Deutschen Volkspartei 1918–1933, 1 vol. in 2 pts, ed. Eberhard
Kolb and Ludwig Richter (Düsseldorf: Droste, 1999), pt 1, 5 ff.;
Rudolf Morsey, Die Deutsche Zentrumspartei 1917–1923
(Düsseldorf: Droste, 1966), 110 ff.; Lothar Albertin, Liberalismus
und Demokratie am Anfang der Weimarer Republik. Eine
vergleichende Analyse der Deutschen Demokratischen Partei und der
Deutschen Volkspartei (Düsseldorf: Droste, 1972); Wolfgang
Hartenstein, Die Anfänge der Deutschen Volkspartei 1918–1920
(Düsseldorf: Droste, 1962); Larry Eugene Jones, German
Liberalism and the Dissolution of the Weimar Party System, 1918–
1933 (Chapel Hill: University of North Carolina Press, 1988);
Werner Liebe, Die Deutschnationale Volkspartei 1918–1924
1081
(Düsseldorf: Droste, 1956); Annelise Thimme, Flucht in den
Mythos. Die Deutschnationale Volkspartei und die Niederlage von
1918 (Göttingen: Vandenhoeck & Ruprecht, 1969); Christian F.
Trippe, Konservative Verfassungspolitik 1918–1923. Die DNVP als
Opposition in Reich und Ländern (Düsseldorf: Droste, 1995), 23 ff.;
Ernst Rudolf Huber, Deutsche Verfassungsgeschichte seit 1789, v.
Weltkrieg, Revolution und Reichserneuerung 1914–1919 (Stuttgart:
Kohhammer, 1978), 953 ff.; Heinrich August Winkler, Weimar
1918–1933. Die Geschichte der ersten deutschen Demokratie, 3rd
edn (Munich: C. H. Beck, 1998), 62 ff.
6. Gerhard A. Ritter, ‘Kontinuität und Umformung des deutschen
Rätesystems 1918–1920’, in Arbeiterbewegung, Parteien und
Parlamentarismus (Göttingen: Vandenhoeck & Ruprecht, 1976),
116–57; Winkler, Revolution (note 3), 135 ff.; id., ‘Der
überforderte Liberalismus. Zum Ort der Revolution von 1848/49
in der deutschen Geschichte’, in Revolution in Deutschland und
Europa 1848/49, ed. Wolfgang Hardtwig (Göttingen:
Vandenhoeck & Ruprecht, 1998), 185–206 (esp. 201 ff.); Dieter
Langewiesche, 1848 und 1918—zwei deutsche Revolutionen (Bonn:
Friedrich-Ebert-Stiftung, 1998).
7. Bernstein, Revolution (note 2), 198 (emphases in original); Die
Regierung der Volksbeauftragten, introd. Erich Matthias, ed.
Susanne Miller with Heinrich Potthoff (Düsseldorf: Droste, 1969),
ii. 225, 228 (Ebert and Scheidemann in the session on 14 Jan.
1919); Peter von Oertzen, Betriebsräte in der Novemberrevolution
(Düsseldorf: Droste, 1963), 109 ff.; Allan Mitchell, Revolution in
Bayern 1918/19. Die Eisner-Regierung und die Räterepublik
1082
(Munich: C. H. Beck, 1967), 236 ff. orig. Revolution in Bavaria,
1918–1919: The Eisner Regime and the Soviet Republic (Princeton:
Princeton University Press, 1965); David Clay Large, Hitlers
München. Aufstieg und Fall der Hauptstadt der Bewegung (Munich:
C. H. Beck, 1998), 118 ff., orig. Where Ghosts Walked: Munich’s
Road to the Third Reich (New York: W. W. Norton, 1997); Trude
Maurer, Ostjuden in Deutschland 1918–1933 (Hamburg:
Christians, 1986), 148 ff.; Winkler, Weimar (note 5), 69 ff. (76:
Noske’s declaration on 9 Mar. 1919; 79: Munich declaration on
6–7 Apr. 1919).
8. Die Deutsche Nationalversammlung im Jahre 1919 in ihrer Arbeit für
den Aufbau des neuen deutschen Volksstaats, ed. Eduard Heilfron,
iv (Berlin: Norddeutsche Buchdruckerei und Verlagsanstalt,
1919), 2646 (Scheidemann), 2650 (Hirsch), 2716 (Fehrenbach);
Protokoll über die Verhandlungen des Parteitags der
Sozialdemokratischen Partei Deutschlands, abgehalten in Weimar
vom 10. bis 15. Juni 1919. Bericht über die 2. Frauenkonferenz,
abgehalten in Weimar am 15. und 16. Juni 1919 (Berlin: J. H. W.
Dietz, 1919; repr. Glashütten: D. Auvermann, 1973), 242–7
(Bernstein), 281 (Scheidemann), 277–8 (Bernstein, quote here);
Jürgen C. Hess, ‘Das ganze Deutschland soll es sein!’ Demokratischer
Nationalismus in der Weimarer Republik, am Beispiel der Deutschen
Demokratischen Partei (Stuttgart: Klett-Cotta, 1978), 76 ff.; Ulrich
Heinemann, Die verdrängte Niederlage. Politische Öffentlichkeit und
Kriegsschuldfrage in der Weimarer Republik (Göttingen:
Vandenhoeck & Ruprecht, 1983); Hagen Schulze, Weimar.
Deutschland 1917–1933 (Berlin: Severin und Siedler, 1982), 189
ff.; Klaus Hildebrand, Das vergangene Reich. Deutsche Aussenpolitik
1083
von Bismarck bis Hitler 1871–1945 (Stuttgart: Deutsche Verlags-
Anstalt, 1995), 383 ff.; Winkler, Weimar (note 5), 87 ff.; Huber,
Verfassungsgeschichte, v (note 5), 1152 ff.; id., Deutsche
Verfassungsgeschichte seit 1789, vii. Ausbau, Schutz und Untergang
der Weimarer Republik (Stuttgart: Kohlhammer, 1984), 37–8
(Hindenburg, 18 Nov. 1919). On the Austria protocol of 22 Sept.
1919 see Dokumente zur deutschen Verfassungsgeschichte, iv.
Deutsche Verfassungsdokumente 1918–1933, ed. id., 3rd edn
(Stuttgart: Kohlhammer, 1991), 180. For the text of the peace
conditions see Treaty of Peace between the Allied and Associated
Powers and Germany and Protocol Signed at Versailles, June 28,
1919 (Paris: Imprimerie nationale, 1919); the official German
translation is Die Friedensbedingungen der Alliierten und
Assoziierten Regierungen (Berlin: R. Hobbing, 1919).
9. Hugo Preuss, ‘Volksstaat oder verkehrter Obrigkeitsstaat?’, in
Staat, Recht und Freiheit. Aus 40 Jahren deutscher Politik und
Geschichte (Hildesheim: G. Olms, 1964), 365–8; Die SPD-Fraktion
in der Nationalversammlung 1919–1920, introd. Heinrich Potthoff,
ed. Heinrich Potthoff and Hermann Weber (Düsseldorf: Droste,
1986), 43 (Molkenbuhr); Sten. Ber., cccxxvi. 374 (Fischer, 28
Feb. 1919); ibid., cccxxix. 219 (David); Carl Schmitt,
Verfassungslehre (Berlin: Duncker & Humblot, 1928; repr. 1957),
111–12 (‘provisional dictatorship of the president’); Ludwig
Richter, Kirche und Schule in den Beratungen der Weimarer
Nationalversammlung (Düsseldorf: Droste, 1996); Reinhard Rürup,
‘Kontinuität und Grundlagen der Weimarer Verfassung’, in Vom
Kaiserreich zur Weimarer Republik, ed. Eberhard Kolb (Cologne:
Kiepenheuer & Witsch, 1972), 218–43; Heinrich Potthoff, ‘Das
1084
Weimarer Verfassungswerk und die deutsche Linke,’ AfS 12
(1972), 433–83; Reinhard Schiffers, Elemente direkter Demokratie
im Weimarer Regierungssystem (Düsseldorf: Droste, 1971), 117 ff.;
Dieter Grimm, Die Bedeutung der Weimarer Verfassung in der
deutschen Verfassungsgeschichte (Heidelberg: Stiftung
Reichspräsident-Friedrich-Ebert-Gedenkstätte, 1990); Karl
Dietrich Bracher, Die Auflösung der Weimarer Republik. Eine Studie
zum Problem des Machtverfalls in der Demokratie, 4th edn
(Villingen: Ring-Verlag, 1964), 21 ff.; Hagen Schulze, ‘Das
Scheitern der Weimarer Republik als Problem der Forschung’, in
Weimar. Selbstpreisgabe einer Demokratie. Eine Bilanz heute, ed.
Karl Dietrich Erdmann and Hagen Schulze (Düsseldorf: Droste,
1980), 23–41 (on the ‘presidential reserve constitution’: 30);
Hans Mommsen, Die verspielte Freiheit. Der Weg der Republik von
Weimar in den Untergang von 1918 bis 1933 (Berlin: Propyläen,
1989), esp. 70; id., The Rise and Fall of Weimar Democracy
(Chapel Hill: University of North Carolina Press, 1996); Huber,
Verfassungsgeschichte, v (note 5), 1178 ff.; Winkler, Weimar (note
5), 99 ff.
10. Klaus Epstein, Matthias Erzberger und das Dilemma der deutschen
Demokratie (Berlin: Verlag Annedore Leber, 1962), 369 ff., orig.
Matthias Erzberger and the Dilemma of German Democracy
(Princeton: Princeton University Press, 1959); Carl Ludwig
Holtfrerich, Die deutsche Inflation 1914–1923. Ursachen und
Wirkungen in internationaler Perspektive (Berlin: de Gruyter,
1981), 115 ff., tr. as The German Inflation, 1914–1923: Causes and
Effects in International Perspective (Berlin, New York: De Gruyter,
1986); Gerald D. Feldman, The Great Disorder: Politics, Economics,
1085
and Society in the German Inflation, 1914–1923 (Oxford: Oxford
University Press, 1993), 25 ff.
11. Johannes Erger, Der Kapp-Lüttwitz-Putsch. Ein Beitrag zur
deutschen Innenpolitik 1919/20 (Düsseldorf: Droste, 1967), 15 ff.;
Heinrich Potthoff, Gewerkschaften und Politik zwischen Revolution
und Inflation (Düsseldorf: Droste, 1979), 267 ff.; Miller, Bürde
(note 3), 377 ff.; Winkler, Revolution (note 3), 295 ff.; id., Weimar
(note 5), 118 ff. (declarations from 13 to 15 Mar. 1920: 122–4,
Ehrhardt Brigade and trade union demands of 18 Mar.: 127);
Gerald D. Feldman, Eberhard Kolb, and Reinhard Rürup, ‘Die
Massenbewegungen der Arbeiterschaft in Deutschland am Ende
des Ersten Weltkriegs’, PVS 18 (1978), 353–439; Wolfgang J.
Mommsen, ‘Die deutsche Revolution von 1918–1920. Politische
Revolution und soziale Protestbewegung’, GG 44 (1978), 362–91;
Huber, Verfassungsgeschichte, vii (note 8), 44 ff.; Wette, Noske
(note 4), 627 ff. On the DVP declaration of 13 Mar. 1920 see
Nationalliberalismus (note 5), pt 1, 247–50.
12. Ernst Laubach, Die Politik der Kabinette Wirth 1921/22 (Lübeck:
Matthiesen, 1968), 9 ff.; Heinrich Küppers, Joseph Wirth.
Parlamentarier, Minister und Kanzler der Weimarer Republik
(Stuttgart: Steiner, 1997), 104 ff.; Ulrike Hörster-Philipps, Joseph
Wirth 1879–1956. Eine politische Biographie (Paderborn:
Schöningh, 1998), 98 ff.; Gerhard Schulz, Zwischen Demokratie
und Diktatur. Verfassungspolitik und Reichsreform in der Weimarer
Republik, i. Die Periode der Konsolidierung und der Revision des
Bismarckschen Reichsaufbaus 1919–1930 (Berlin: de Gruyter,
1963), 320 ff.; Huber, Verfassungsgeschichte, vii (note 8), 25 ff.,
1086
169 ff.; Winkler, Weimar (note 5), 154 ff.; Norbert Elias, ‘Die
Zersetzung des staatlichen Gewaltmonopols in der Weimarer
Republik’, in Studien über die Deutschen. Machtkämpfe und
Habitusentwicklung im 19. u. 20. Jahrhundert (Frankfurt:
Suhrkamp, 1989), 282–94; id., ‘Kriegsbejahende Literatur in der
Weimarer Republik (Ernst Jünger)’, ibid. 274–81; James M.
Diehl, Paramilitary Politics in Weimar Germany (Bloomington:
University of Indiana Press, 1977); Bernd Weisbrod, ‘Gewalt in
der Politik. Zur politischen Kultur in Deutschland zwischen den
beiden Weltkriegen’, GWU 43 (1992), 391–405; Rolf Geissler,
Dekadenz und Heroismus. Zeitroman und völkisch-
nationalsozialistische Literaturkritik (Stuttgart: Deutsche Verlags-
Anstalt, 1964). On Max Weber’s analysis see id., Wirtschaft und
Gesellschaft, study edition, ed. Johannes Winckelmann, pt 1
(Cologne 1964), 39 (pt 1, ch. 1, § 17), 197 (pt 1, ch. 3, § 13), tr.
as Economy and Society: An Outline of Interpretive Sociology (New
York: Bedminster Press, 1968).
13. Akten der Reichskanzlei. Weimarer Republik [AdR]. Die Kabinette
Wirth I und II. 10. Mai 1921 bis 26. Oktober 1921, 26. Oktober
1921 bis 22. November 1922, 2 vols., i. Mai 1921 bis März 1922,
ed. Ingrid Schulze-Bidlingmaier (Boppard: H. Boldt, 1973), 7–13
(Schmidt’s memorandum of 19 May 1921), 88–90 (cabinet
session on 24, June 1921); Ludwig Thoma, Sämtliche Beiträge aus
dem ‘Miesbacher Anzeiger’ 1920/21, ed. and comm. Wilhelm
Volkert (Munich: Piper, 1989), 278, 286, 341; Epstein, Erzberger
(note 10), 428 ff. (Oletzkoer Zeitung and Kreuz-Zeitung: 433);
Gotthard Jasper, Der Schutz der Republik. Studien zur staatlichen
Sicherung der Demokratie in der Weimarer Republik 1922–1930
1087
(Tübingen: Mohr, 1963), 34 ff. (36: Berliner Lokalanzeiger);
Schulz, Demokratie (note 12), 364 ff.; Laubach, Politik (note 12),
263 ff.; Huber, Verfassungsgeschichte, vii (note 8), 206 ff.;
Winkler, Weimar (note 5), 160 ff.
14. Sigrid Koch-Baumgarten, Aufstand der Avantgarde. Die Märzaktion
der KPD 1921 (Frankfurt: Campus, 1986); Martin Walsdorff,
Westorientierung und Ostpolitik. Stresemanns Russlandpolitik in der
Locarno-Ära (Bremen: Schünemann, 1972), 31 (Wirth quotes);
Francis L. Carsten, Reichswehr und Politik 1918–1933 (Cologne:
Kiepenheuer & Witsch, 1964), 78–9 (Seeckt quote), tr. as The
Reichswehr and Politics, 1918–1933 (Oxford: Clarendon Press,
1966); Theodor Schieder, Die Probleme des Rapallo-Vertrags. Eine
Studie über die deutsch-russischen Beziehungen 1922–1926
(Cologne: Westdeutscher Verlag, 1956); Hermann Graml, ‘Die
Rapallopolitik im Urteil der westdeutschen Forschung’, VfZ 18
(1970), 366–91; Manfred Zeidler, Reichswehr und Rote Armee
1920–1933 (Munich: Oldenbourg, 1993); Peter Krüger, Die
Aussenpolitik der Republik von Weimar (Darmstadt:
Wissenschaftliche Buchgesellschaft, 1985), 166 ff.; Hildebrand,
Reich (note 8), 422 ff.; Küppers, Wirth (note 12), 154 ff.; Winkler,
Revolution (note 3), 459 ff. (Breitscheid quote: 464); id., Weimar
(note 5), 166 ff.
15. Harry Graf Kessler, Walther Rathenau. Sein Leben und Werk
(Wiesbaden: Rheinische Verlags-Anstalt, 1928), tr. as Walther
Rathenau: His Life and Work (New York: H. Fertig, 1969); Ernst
Schulin, Walther Rathenau. Repräsentant, Kritiker und Opfer seiner
Zeit (Göttingen: Musterschmidt, 1979); Martin Sabrow, Der
1088
Rathenaumord. Rekonstruktion einer Verschwörung gegen die
Republik von Weimar (Munich: Oldenbourg, 1994); Winkler,
Weimar (note 5), 174 ff.; Huber, Verfassungsgeschichte, vii (note
8), 249 ff.; Laubach, Politik (note 12), 263 ff.; Jasper, Schutz
(note 13), 56 ff. (196 ff.: quotes on the judiciary); Christoph
Gusy, Weimar—die wehrlose Republik? Verfassungsschutzrecht und
Verfassungsschutz in der Weimarer Republik (Tübingen: Mohr,
1991), 134 ff.; Uwe Lohalm, Völkischer Radikalismus. Die
Geschichte des Deutschvölkischen Schutz- und Trutzbundes 1919–
1923 (Hamburg: Leibnitz-Verlag, 1970); the article
‘Deutschvölkischer Schutz- und Trutzbund 1919–1922’, in
Lexikon zur Parteiengeschichte. Die bürgerlichen und
kleinbürgerlichen Parteien und Verbände in Deutschland (1789–
1945), 4 vols., ed. Dieter Fricke et al. (Leipzig: VEB
Bibliographisches Institut Leipzig, 1983-), ii. 562–8
(membership: 562); Bernd Hüppauf, ‘Schlachtenmythen und die
Konstruktion des “Neuen Menschen”, Keiner fühlt sich (note 1),
ed. Hirschfeld et al., 43–84 (on the Langemarck myth). Quote
from Wirth in Sten. Ber., ccclvi. 8058.
16. Heinrich August Winkler, ‘Die deutsche Gesellschaft der
Weimarer Republik und der Antisemitismus’, in Die Juden als
Minderheit in der Geschichte, ed. Bernd Martin and Ernst Schulin
(Munich: Deutscher Taschenbuch Verlag, 1981), 271–89
(numbers: 274–5); Werner Jochmann, ‘Die Ausbreitung des
Antisemitismus in Deutschland 1914–1923’, in Gesellschaftskrise
und Judenfeindschaft in Deutschland 1870–1945 (Hamburg:
Christians, 1988), 99–170; Donald L. Niewyk, The Jews in Weimar
Germany (Manchester: Manchester University Press, 1980); Dirk
1089
Walter, Antisemitische Kriminalität und Gewalt. Judenfeindschaft in
der Weimarer Republik (Bonn: Dietz, 1999), 52 ff.; Maurer,
Ostjuden (note 7); Ulrich Herbert, Best. Biographische Studien über
Radikalismus, Weltanschauung und Vernunft 1903–1989, 2nd edn
(Bonn: Dietz, 1996), 51 ff. (on the battle of the German
Hochschulring against the ‘foreign people’ of the Jews: 63);
Michael H. Kater, Studentenschaft und Rechtsradikalismus in
Deutschland 1918–1933. Eine sozialgeschichtliche Studie zur
Bildungskrise in der Weimarer Republik (Hamburg: Hoffmann und
Campe, 1975); Konrad H. Jarausch, Deutsche Studenten 1800–
1970 (Frankfurt: Suhrkamp, 1984), 117 ff.; Jonathan R. C.
Wright, ‘Über den Parteien’. Die politische Haltung der evangelischen
Kirchenführer 1918–1933 (Göttingen: Vandenhoeck & Ruprecht,
1977), 66, 84, 66 (political position of the Evangelical church in
general), 84 (Rathenau’s murder), orig. ‘Above Parties’: The
Political Attitudes of the German Protestant Church Leadership
1918–1933 (London: Oxford University Press, 1974); Kurt
Nowak, Evangelische Kirche und Weimarer Republik. Zum
politischen Weg des deutschen Protestantismus zwischen 1918 und
1922 (Weimar: H. Böhlaus Nachfolger, 1981), 117 ff. (quotes on
Rathenau’s murder: 118); id., Geschichte des Christentums in
Deutschland. Religion, Politik und Gesellschaft vom Ende der
Aufklärung bis zur Mitte des 20. Jahrhunderts (Munich: C. H. Beck,
1995), 205 ff.; Heinrich Lutz, Demokratie im Zwielicht. Der Weg
der deutschen Katholiken aus dem Kaiserreich in die Republik 1914–
1925 (Munich: Kösel-Verlag, 1963); Heinz Hürten, Deutsche
Katholiken 1918–1945 (Paderborn: Schöningh, 1992); Morsey,
Zentrumspartei (note 5), 401 ff. On the Munich Catholic congress
1090
see Schulthess’ Europäischer Geschichtskalender, NS 38, 1922
(Munich: C. H. Beck, 1927), pt 1, 106–8. Quote from the
Würzburg constitution of the German Students’ Association from
Ernst Rudolf Huber, Deutsche Verfassungsgeschichte seit 1789, vi.
Die Weimarer Reichsverfassung (Stuttgart: Kohlhammer, 1981),
1011.
17. Liebe, Deutschnationale Volkspartei (note 5), 62 ff. (quote from
Henning: 159); Large, München (note 7), 162 ff.; Winkler,
Revolution (note 3), 434 ff. (Görlitz party congress), 468 ff.
(USPD split in 1920), 486 ff. (reunion of MSPD and USPD); id.,
Weimar (note 5), 178 ff.; Parteiprogramme (note 5), ed.
Mommsen, 453–8 (Görlitz platform of the SPD), 461–9
(Heidelberg platform); Laubach, Politik (note 12), 296 ff.
(commission of experts and reparations note); Huber,
Verfassungsgeschichte, vii (note 8), 258 ff.; Gerald D. Feldman,
Hugo Stinnes. Biographie eines Industriellen 1870–1924 (Munich: C.
H. Beck, 1998), 741 ff.; Peter Wulf, Hugo Stinnes. Wirtschaft und
Politik 1918–1924 (Stuttgart: Klett-Cotta. 1979), 317 ff.; Alfred
Kastning, Die deutsche Sozialdemokratie zwischen Koalition und
Opposition 1919–1923 (Paderborn: Schöningh, 1970), 110 ff.
18. Jacques Bariéty, Les relations franco-allemandes après la première
guerre mondiale (Paris: Éditions Pedone, 1977), 91 ff.; Die
Ruhrkrise 1923. Wendepunkt der internationalen Beziehungen nach
dem Ersten Weltkrieg, ed. Klaus Schwabe (Paderborn: Schöningh,
1984); Hermann J. Rupieper, The Cuno Government and
Reparations 1922–1923: Politics and Economics (The Hague:
Nijhoff, 1979), 13 ff.; Winkler, Weimar (note 5), 186 ff; Laubach,
1091
Politik (note 12), 262, 263 (‘first bread, then reparations’);
Dokumente und Materialien zur Geschichte der deutschen
Arbeiterbewegung [DuM], vii/2 (Berlin: Dietz, 1966), 210–13
(KPD declaration of 22 Jan. 1923); Hitler, Sämtliche
Aufzeichnungen 1905–1924, ed. Eberhard Jäckel with Axel Kuhn
(Stuttgart: Deutsche Verlags-Anstalt, 1980), 785–6 (declaration
on 12 Jan 1923).
19. Protokoll der Konferenz der Erweiterten Exekutive der
Kommunistischen Internationale. Moskau, 12.–23. Juni 1923
(Hamburg: C. Hoym Nachfolger L. Cahnbley, 1923; repr. Milan:
Feltrinelli, 1967), 240–5; Otto-Ernst Schüddekopf, Linke Leute von
rechts. Die nationalrevolutionären Minderheiten und der
Kommunismus in der Weimarer Republik (Stuttgart: Kohlhammer
1960), 139 ff.; Louis Dupeux, ‘Nationalbolschewismus’ in
Deutschland 1919–1933. Kommunistische Strategie und konservative
Dynamik (Munich: C. H. Beck, 1985), 178 ff., French orig.
National bolchevisme: stratégie communiste et dynamique
conservatrice (Paris: H. Champion, 1979); Marie-Luise Goldbach,
Karl Radek und die deutsch-sowjetischen Beziehungen 1918–1923
(Bonn: Verlag Neue Gesellschaft, 1973), 121 ff.; Werner T.
Angress, Die Kampfzeit der KPD 1921–1923 (Düsseldorf; Droste,
1973), 374 ff., orig. Stillborn Revolution: The Communist Bid for
Power in Germany, 1921–1923 (Princeton: Princeton University
Press, 1963); Rosenberg, Geschichte (note 4), 136 ff.; Kastning,
Sozialdemokratie (note 17), 114 ff.; Günter Arns, ‘Die Linke in der
SPD-Reichstagsfraktion im Herbst 1923’, VfZ 22 (1974), 191–
203; Winkler, Revolution (note 3), 561 ff.; id., Weimar (note 5),
190 ff. (on inflation figures: 193, on the KPD: 200, SPD demands:
1092
203).
20. AdR. Die Kabinette Stresemann I u. II. 13. August bis 6. Oktober
1923, 6. Oktober bis 30. November 1923, 2 vols., ed. Karl Dietrich
Erdmann and Martin Vogt (Boppard: Boldt, 1978), ii. 1215–16
(Seeckt to Wiedfeldt); Angress, Kampfzeit (note 19), 426 ff.;
Winkler, Revolution (note 3), 619 ff.; id., Weimar (note 5), 209 ff.
(Völkischer Beobachter: 211, Severing and Löbe 31 Oct. 1923:
228–9, on Hilferding’s theories concerning the ‘political wage’
and ‘organized capitalism’: 328–9); Huber, Verfassungsgeschichte,
vii (note 8), 330 ff.; Maurer, Ostjuden (note 7), 405 ff.; Reiner
Pommerin, ‘Die Ausweisung von “Ostjuden” aus Bayern 1923.
Ein Beitrag zum Krisenjahr der Weimarer Republik’, VfZ 34
(1986), 311–40; Gerald D. Feldman, ‘Bayern und Sachsen in der
Hyperinflation 1922/23’, HZ 238 (1984), 569–609; Heinz
Hürten, Reichswehr und Ausnahmezustand. Ein Beitrag zur
Verfassungsproblematik der Weimarer Republik in ihrem ersten
Jahrfünft (Opladen: Westdeutscher Verlag, 1977), 33–4; Der
Hitler-Putsch. Bayerische Dokumente zum 8./9. November 1923, ed.
Ernst Deuerlein (Stuttgart: Deutsche Verlags-Anstalt, 1968);
Harold J. Gordon, Jr., Hitler-Putsch 1923. Machtkampf in Bayern
1923–24 (Frankfurt: Bernard und Graefe, 1971), orig. Hitler and
the Beer Hall Putsch (Princeton: Princeton University Press, 1972);
Hanns-Hubert Hofmann, Der Hitlerputsch. Krisenjahre deutscher
Geschichte 1920–1924 (Munich: Nymphenburger
Verlagshandlung, 1961).
21. AdR, Kabinette Stresemann (note 20), ii. 1059 (Adenauer, 13 Nov.
1923); Gustav Stresemann, Vermächtnis. Der Nachlass in drei
1093
Bänden, i (Berlin: Ullstein, 1932), 245 (Stresemann, 23 Nov.
1923), tr. as Gustav Stresemann: His Diaries, Letters, and Papers
(New York: AMS Press, 1935). Karl Dietrich Erdmann, Adenauer
in der Rheinlandpolitik nach dem Ersten Weltkrieg (Stuttgart: Klett,
1966), 71 ff.; Henning Köhler, Adenauer. Eine politische Biographie
(Berlin: Propyläen, 1994), 154 ff.; Hans Peter Schwarz, Adenauer.
Der Aufstieg: 1876–1952 (Stuttgart: Deutsche Verlags-Anstalt
1986), 258 ff., tr. as Konrad Adenauer: A German Politician and
Statesman in a Period of War, Revolution, and Reconstruction
(Providence, RI: Berghahn Books, 1995); Das Krisenjahr 1923.
Militär und Innenpolitik 1922–1924, ed. Heinz Hürten (Düsseldorf:
Droste, 1980); Krüger, Aussenpolitik (note 14), 263 ff.; Bariéty,
Relations (note 18), 263 ff.; Huber, Verfassungsgeschichte, vii (note
8), 420 ff.; Winkler, Weimar (note 5), 236 ff. (enabling act of 8
Dec. 1923: 247, worker numbers: 245–6). Quote from Franz
Eulenburg from id., ‘Die sozialen Wirkungen der
Währungsverhältnisse’, Jahrbücher zur Nationalökonomie u.
Statistik, 122 (1924), 748–94 (789).
22. Deutscher Geschichtskalender 40 (1924), i. (internal events), ed.
Friedrich Purlitz (Leipzig: Felix Meiner, n.d.), 296–9 (verdict in
the Hitler case); Der Hitler-Prozess 1924. Wortlaut der
Hauptverhandlung vor dem Volksgericht München I, ed. and comm.
Lothar Gruchmann and Reinhard Weber with Otto Gritschneder
(Munich: K. G. Saur, 1997–); Bernd Steger, ‘Der Hitlerprozess u.
Bayerns Verhältnis zum Reich 1923/24’, VfZ 25 (1977), 441–66;
Otto Gritschneder, Bewährungsfrist für den Terroristen Adolf H. Der
Hitler-Putsch und die bayerische Justiz (Munich: C. H. Beck, 1990);
Werner Link, Die amerikanische Stabilisierungspolitik in Deutschland
1094
1921–1932 (Düsseldorf: Droste, 1970), 201 ff.; Eckhard Wandel,
Die Bedeutung der Vereinigten Staaten von Amerika für das deutsche
Reparationsproblem 1924–1929 (Tübingen: Mohr, 1971); Adam B.
Ulam, Expansion and Coexistence: The History of Soviet Foreign
Policy, 1917–1967, 3rd edn (New York: Praeger, 1969), 154 ff.;
Winkler, Revolution (note 3), 725 ff.
23. Sozialdemokratischer Parteitag 1924. Protokoll mit dem Bericht der
Frauenkonferenz (Berlin: J. H. W. Dietz, 1924; repr. Glashütten:
D. Auvermann, 1974), 83 (Müller; emphasis in original), 99
(Dissmann), 139 (ballot), 204 (Müller’s motion); Peter Haungs,
Reichspräsident und parlamentarische Kabinettsregierung. Eine Studie
zum Regierungssystem der Weimarer Republik in den Jahren 1924
bis 1926 (Cologne: Westdeutscher Verlag, 1968), 74 ff.; Liebe,
Deutschnationale Volkspartei (note 5), 76 ff.; Krüger, Aussenpolitik
(note 14), 237 ff.; Huber, Verfassungsgeschichte, vii (note 8), 495
ff.; Winkler, Weimar (note 5), 261 ff. (May elections: 261–2,
economic figures: 268, December elections: 271–2).
24. Adolf von Harnack, ‘Brief an Friedrich Ebert’, Berliner Tageblatt,
27 Dec. 1924; Erich Eyck, Geschichte der Weimarer Republik, 2
vols., 4th edn (Erlenbach-Zurich: E. Rentsch Verlag, 1962), i. 436
ff.; Wolfgang Birkenfeld, ‘Der Rufmord am Reichspräsidenten. Zu
Grenzformen des politischen Kampfes gegen die frühe Weimarer
Republik’, AfS 5 (1965), 453–500; Günter Arns, ‘Friedrich Ebert
als Reichspräsident’, in Beiträge zur Geschichte der Weimarer
Republik, ed. Theodor Schieder, HZ, suppl 1 (Munich:
Oldenbourg, 1981), 1–3; Waldemar Besson, Friedrich Ebert.
Verdienst und Grenze (Göttingen: Musterschmidt, 1963); Hans
1095
Mommsen, ‘Friedrich Ebert als Reichspräsident’, in
Arbeiterbewegung und Nationale Frage (Göttingen: Vandenhoeck &
Ruprecht, 1979), 296–317; Winkler, Weimar (note 5), 276 ff.
(quotes about Ebert: 277; the emphasis in the Vorwärts quote of
28 Feb. 1925 is in the original).
25. DuM viii. 130–3 (KPD declaration of 11 Apr. 1925);
‘Hindenburg, President by the Grace of Thälmann’ (‘Hindenburg
von Thälmanns Gnaden’) Vorwärts, 169, 27 Apr. 1925; ‘Der
Präsident der Minderheit’, ibid. 197, 27 Apr. 1925; ‘Es lebe die
Republik!’, Frankfurter Zeitung, 309, 27 Apr. 1925; Ernst Feder,
‘Der Retter’, Berliner Tageblatt, 198, 28 Apr. 1925; Heinrich
Mann, ‘Geistige Führer zur Reichspräsidentenwahl’, Deutsche
Einheit 7 (1925), 633–5; Harry Graf Kessler, Tagebücher 1918–
1937 (Frankfurt: Insel, 1961), 441–2, tr. as Berlin in Lights: The
Diaries of Count Harry Kessler, 1918–1937 (London: Weidenfeld
and Nicolson, 1971); Noel D. Cary, ‘The Making of the Reich
President, 1925: German Conservatism and the Nomination of
Paul von Hindenburg’, CEH 23 (1990), 179–204; Peter Fritzsche,
Rehearsals for Fascism: Populism and Political Mobilization in
Weimar Germany (New York: Oxford University Press, 1990), 154
ff.; John Zeender, ‘The German Catholics and the Presidential
Elections of 1925’, JMH 35 (1963), 366–81; Karl Holl,
‘Konfessionalität. Konfessionalismus und demokratische
Republik. Zu einigen Aspekten der Reichspräsidentenwahl von
1925’, VfZ 17 (1969), 254–75; Ulrich von Hehl, Wilhelm Marx
1863–1946. Eine politische Biographie (Mainz: Matthias-
Grünewald-Verlag, 1987), 335 ff.; Andreas Dorpalen, Hindenburg
in der Geschichte der Weimarer Republik (Berlin: A. Leber, 1966),
1096
68 ff., orig. Hindenburg and the Weimar Republic (Princeton:
Princeton University Press, 1964); John W. Wheeler-Bennett, Der
hölzerne Titan. Paul von Hindenburg (Tübingen: Mohr, 1969), orig.
Hindenburg: The Wooden Titan (London: St Martin’s Press, 1967);
Nowak, Kirche (note 16), 160 ff.; Heinrich August Winkler, Der
Schein der Normalität. Arbeiter und Arbeiterbewegung in der
Weimarer Republik 1924–1930, 2nd edn (Berlin: J. H. W. Dietz,
1987), 239 ff.; in Weimar (note 5), 278 ff.
26. Kurt Tucholsky, ‘Berlin und die Provinz’, in Gesammelte Werke,
ii. 1925–1928 (Reinbek: Rowohlt, 1960), 1072–75; Martin
Heidegger, Sein und Zeit (1927), 8th edn (Tübingen: Niemeyer,
1957), 127, tr. as Being and Time (New York: Harper, 1962); Carl
Schmitt, Die geistesgeschichtliche Lage des heutigen
Parlamentarismus (1923), 2nd edn (Berlin: Duncker & Humblot,
1926), 8, tr. as The Crisis of Parliamentary Democracy (Cambridge,
MA: MIT Press, 1985); id., Verfassungslehre (note 9), 350–1; Hans
Freyer, Revolution von rechts (Jena: E. Diederichs, 1931); Oswald
Spengler, Preussentum und Sozialismus (Munich: Beck, 1920), 97–
8 (emphases in original); Adolf Hitler, ‘Warum musste ein 8.
November kommen?’ (April 1924), in Aufzeichnungen (note 18),
1216–27 (1226; emphasis in original); Armin Mohler, Die
Konservative Revolution in Deutschland 1918–1932. Grundriss ihrer
Weltanschauungen (Stuttgart: F. Vorwerk, 1950); Rolf-Peter
Sieferle, Die Konservative Revolution. Fünf biographische Skizzen
(Frankfurt: Fischer Taschenbuch, 1995); Stefan Breuer, Anatomie
der Konservativen Revolution (Darmstadt: Wissenschaftliche
Buchgesellschaft, 1993); Raimund von dem Bussche,
Konservatismus in der Weimarer Republik. Die Politisierung des
1097
Unpolitischen (Heidelberg: C. Winter, 1998); Kurt Sontheimer,
Antidemokratisches Denken in der Weimarer Republik. Die
politischen Ideen des deutschen Nationalismus zwischen 1918 und
1933 (Munich: Nymphenburger Verlagshandlung, 1962);
Christian Graf v. Krockow, Die Entscheidung. Eine Untersuchung
über Ernst Jünger, Carl Schmitt und Martin Heidegger (Stuttgart: F.
Enke, 1958); Detlef Felken, Oswald Spengler. Konservativer Denker
zwischen Kaiserreich und Diktatur (Munich: C. H. Beck, 1988), 25
ff.; Peter Gay, Die Republik der Aussenseiter. Geist und Kultur in der
Weimarer Zeit: 1918–1933 (Frankfurt: S. Fischer, 1970); 23
(quote), orig. Weimar Culture: The Outsider as Insider (New York:
Harper & Row, 1968); Walter Laqueur, Weimar. Die Kultur der
Republik (Frankfurt: Ullstein, 1976), orig. Weimar: A Cultural
History, 1918–1933 (London: Weidenfeld & Nicolson, 1974);
Detlev J. K. Peukert, Die Weimarer Republik. Krisenjahre der
Klassischen Moderne (Frankfurt: Suhrkamp, 1987); Friedhelm
Kröll, Das Bauhaus 1919–1933 (Düsseldorf: Bertelsmann-
Universitätsverlag, 1974).
27. Theodore S. Hamerow, Die Attentäter. Der 20. Juli—von der
Kollaboration zum Widerstand (Munich: C. H. Beck, 1999), 85
(Dibelius); orig. On the Road to the Wolf’s Lair: German Resistance
to Hitler (Cambridge, MA: Belknap Press of Harvard University
Press, 1997); Der Grosse Herder, 4th edn, i (Freiburg: Herder,
1926), 725; Victor Klemperer, Leben sammeln, nicht fragen wozu
und warum. Tagebücher 1925–1932 (Berlin: Aufbau-Verlag,
1996), 281 (entry for 11 July 1926 on Hiddensee); Kurt
Tucholsky, ‘Gefühlskritik’, in Werke (note 26), i. 827–8; id.,
‘Feldfrüchte,’ ibid., ii. 508–9.; Thomas Mann, ‘Von deutscher
1098
Republik’ (1922), in Gesammelte Werke in dreizehn Bänden, (=
Reden u. Aufsätze, ii) (Frankfurt: Fischer Taschenbuch Verlag,
1965), 9–52; Kampf um München als Kulturzentrum. Sechs Vorträge
von Thomas Mann, Heinrich Mann, Leo Weismantel, Walter
Courvoisier and Paul Renner. Mit einem Vorwort von Thomas Mann
(Munich: R. Pflaum, 1926), 9; Friedrich Meinecke, ‘Republik,
Bürgertum und Jugend’ (1925), in Werke, ii. Politische Schriften u.
Reden (Darmstadt: Siegfried-Toech-Mittler, 1958), 369–83 (376);
id., ‘Die deutschen Universitäten und der heutige Staat’ (1926),
ibid. 402–13 (410, 413); Istvan Déak, Weimar Germany’s Left-
Wing Intellectuals: A Political History of the ‘Weltbühne’ and its
Circle (Berkeley: University of California Press, 1968); Winkler,
Gesellschaft (note 16), 171 ff.; id., Weimar (note 5), 285 ff.
28. Akten zur Deutschen Auswärtigen Politik 1918–1945. Aus dem
Archiv des Auswärtigen Amts. Serie B: 1925–1933, ii/1. Dezember
1925 bis Juni 1926. Deutschlands Beziehungen zur Sowjet-Union, zu
Polen, Danzig und den Baltischen Staaten (Göttingen: Vandenhoeck
& Ruprecht, 1967), 363–5 (Stresemann, 19 Apr. 1926);
Documents on German Foreign Policy, 1918–1945, from the
Archives of the German Foreign Ministry (Washington: U.S. Govt.
Print. Off., 1949–); ‘Der Sieg des Friedens’, Vorwärts, 250, Oct. 17
1925; Jon Jacobson, Locarno Diplomacy. Germany and the West
1925–1929 (Princeton: Princeton University Press, 1972); Klaus
Megerle, Deutsche Aussenpolitik 1925. Ansatz zu aktivem
Revisionismus (Bern: Herbert Lang, 1974); Jürgen Spenz, Die
diplomatische Vorgeschichte des Beitritts Deutschlands zum
Völkerbund 1924–1926. Ein Beitrag zur Aussenpolitik der Weimarer
Republik (Göttingen: Musterschmidt, 1960), 33 ff.; Helmut
1099
Lippelt, ‘“Politische Sanierung”. Zur deutschen Politik gegenüber
Polen 1925/26’, VfZ 19 (1972), 323–73; Krüger, Aussenpolitik
(note 14), 269 ff.
29. Henry A. Turner, Jr., Stresemann—Republikaner aus Vernunft
(Berlin: Leber, 1968), 217 ff., orig. Streseman and the Politics of
the Weimar Republic (Princeton: Princeton University Press,
1963); Klaus E. Rieseberg, ‘Die SPD in der “Locarno-Krise”
Oktober/November 1925’, VfZ 30 (1982), 130–61; Ulrich
Schüren, Der Volksentscheid zur Fürstenenteignung 1926. Die
Vermögensauseinandersetzungen mit den depossedierten Landesherren
als Problem der deutschen Innenpolitik unter besonderer
Berücksichtigung der Verhältnisse in Preussen (Düsseldorf: Droste,
1978); Michael Stürmer, Koalition und Opposition in der Weimarer
Republik 1924–1928 (Düsseldorf: Droste, 1967), 132 ff.; Carsten,
Reichswehr (note 14), 276 ff.; Huber, Verfassungsgeschichte, vii
(note 8), 576 ff.; Winkler, Schein (note 25), 246 ff.; id., Weimar
(note 5), 306 ff. For Scheidemann’s speech on 16 Dec. 1926, see
Sten. Ber., cccxci. 8576–86.
30. Huber, Verfassungsgeschichte, vi (note 16), 1013 ff. (‘Keudell
case’); Stürmer, Koalition (note 29), 213 ff.; Haungs,
Reichspräsident (note 23), 208 ff.; Ludwig Preller, Sozialpolitik in
der Weimarer Republik, 2nd edn (Düsseldorf: Droste, 1978), 350
ff.; Peter Lewek, Arbeitslosigkeit und Arbeitslosenversicherung in der
Weimarer Republik 1918–1927 (Stuttgart: F. Steiner, 1992), 287
ff.; Gerhard A. Ritter, Der Sozialstaat. Entstehung und Entwicklung
im internationalen Vergleich (Munich: Oldenbourg, 1989), 110 ff.;
Wolfram Fischer, Deutsche Wirtschaftspolitik 1918–1945, 3rd edn
1100
(Opladen: C. W. Leske, 1968), 43–4; Ellen L. Evans, The German
Center Party 1870–1933: A Study in Political Catholicism
(Carbondale: Southern Illinois University Press, 1981), 217 ff.;
Karsten Ruppert, Im Dienst am Staat von Weimar. Das Zentrum als
regierende Partei in der Weimarer Demokratie 1923–1930
(Düsseldorf: Droste, 1992), 287 ff.; Günter Grünthal,
Reichsschulgesetz und Zentrumspartei in der Weimarer Republik
(Düsseldorf: Droste, 1968), 196 ff.; Wolfgang Wacker, Der Bau
des Panzerschiffes «A» und der Reichstag (Tübingen: Mohr, 1959),
33 ff.; Wolfgang Horn, Führerideologie und Parteiorganisation in der
NSDAP (1919–1933) (Düsseldorf: Droste, 1972), 209 ff. On the
reinterpretation of the NSDAP platform see the article
‘Nationalsozialistische Deutsche Arbeiterpartei (NSDAP) 1919–
1945’, in Lexikon (note 15), iii, ed. Fricke et al., 460–523 (481).
31. Wacker, Bau (note 30), 90 ff. (Vossische Zeitung, 137–8); Rudolf
Heberle, Landbevölkerung und Nationalsozialismus. Eine
soziologische Untersuchung der politischen Willensbildung in
Schleswig-Holstein 1918–1932 (Stuttgart: Deutsche Verlags-
Anstalt, 1963), 48 ff.; Martin Vogt, Die Entstehung des Youngplans
dargestellt vom Reichsarchiv 1931–1933 (Boppard: H. Boldt,
1970); Hermann Weber, Die Wandlung des deutschen
Kommunismus. Die Stalinisierung der KPD in der Weimarer Republik,
2 vols. (Frankfurt: Europäische Verlagsanstalt, 1969), i. 195 ff.;
Siegfried Bahne, ‘“Sozialfaschismus” in Deutschland. Zur
Geschichte eines politischen Begriffs’, International Review of
Social History, 10 (1965), 211–45; Andreas Wirsching, Vom
Weltkrieg zum Bürgerkrieg? Extremismus in Deutschland und
Frankreich 1918–1933/39. Berlin und Paris im Vergleich (Munich:
1101
Oldenbourg, 1999), 361 ff.; Thomas Weingartner, Stalin und der
Aufstieg Hitlers. Die Deutschlandpolitik der Sowjetunion und der
Kommunistischen Internationale 1929–1934 (Berlin: de Gruyter,
1970), 70 ff.; Thomas Kurz, ‘Blutmai’. Sozialdemokraten und
Kommunisten im Brennpunkt der Berliner Ereignisse von 1929
(Bonn: Dietz, 1988); Léon Schirmann, Blutmai Berlin 1929.
Dichtungen und Wahrheit (Berlin: Dietz, 1992); Winkler, Schein
(note 25), 521 ff.; id., Weimar (note 5), 334 ff. (election results,
formation of the government, battleship), 346 (Paris
negotiations), 349 ff. (KPD; quotes from the Wedding party
congress and unemployment figures: 351–2), 352 ff. (social
policy, Stresemann’s death).
32. Otmar Jung, ‘Plebiszitärer Durchbruch 1929? Zur Bedeutung
von Volksbegehren und Volksentscheid gegen den Young-Plan
für die NSDAP’, GG 15 (1989), 489–510; id., Direkte Demokratie
in der Weimarer Republik. Die Fälle ‘Aufwertung,’
‘Fürstenenteignung’, ‘Panzerkreuzerverbot’ und ‘Young-Plan’
(Frankfurt: Campus, 1989), 109 ff.; Friedrich Freiherr Hiller von
Gaertringen, ‘Die Deutschnationale Volkspartei’, in Das Ende der
Parteien 1933, ed. Erich Matthias and Rudolf Morsey (Düsseldorf:
Droste, 1960), 543–652 (544 ff.); John A. Leopold, Alfred
Hugenberg: The Radical Nationalist Campaign against the Weimar
Republic (New Haven: Yale University Press, 1981), 55 ff.;
Gerhard Stoltenberg, Die politischen Stimmungen im schleswig-
holsteinischen Landvolk 1918–1933 (Düsseldorf: Droste, 1962),
125 ff.; Anselm Faust, Der Nationalsozialistische Deutsche
Studentenbund. Studenten und Nationalsozialismus in der Weimarer
Republik, 2 vols. (Düsseldorf: Droste, 1973); Kater,
1102
Studentenschaft (note 16), 147 ff., 218–19, 288 (student
numbers); Bracher, Auflösung (note 9), 147–8 (AStAelections);
Fischer, Wirtschaftspolitik (note 30), 43 ff. (economic figures);
Preller, Sozialpolitik (note 30), 166–7 (unemployment figures);
Ilse Maurer, Reichsfinanzen und Grosse Koalition. Zur Geschichte
des Reichskabinett Müller (1928–1930) (Bern: Lang, 1973), 101
ff.; Rosemarie Leuschen-Seppel, Zwischen Staatsverantwortung und
Klasseninteresse. Die Wirtschafts- und Finanzpolitik der SPD zur Zeit
der Weimarer Republik unter besonderer Berücksichtigung der
Mittelphase 1924–1928/29 (Bonn: Verlag Neue Gesellschaft,
1981), 217 ff.; Winkler, Weimar (note 5), 354 ff. (‘liberty law’:
355, election returns: 356, Hilferding: 360).
33. Aufstieg oder Niedergang? Deutsche Wirtschafts- und Finanzreform
1929. Eine Denkschrift des Präsidiums des Reichsverbandes der
Deutschen Industrie (Berlin, 1929), 45–6.; Michael Grübler, Die
Spitzenverbände der Wirtschaft und das erste Kabinett Brüning. Vom
Ende der Grossen Koalition 1929/30 bis zum Vorabend der
Bankenkrise 1931 (Düsseldorf: Droste, 1982), 49 ff.; Erasmus
Jonas, Die Volkskonservativen 1928–1933. Entwicklung, Struktur,
Standort und staatspolitische Zielsetzung (Düsseldorf: Droste, 1965),
186–8 (Hindenburg-Westarp talks, 18 Mar. 1929); Andreas
Rödder, ‘Dichtung und Wahrheit. Der Quellenwert von Heinrich
Brünings Memoiren und seine Kanzlerschaft’, HZ 265 (1997), 77–
116; Politik und Wirtschaft in der Krise 1930–1932. Quellen zur Ära
Brüning, introd. Gerhard Schulz, ed. Ilse Maurer and Udo Wengst
with Jürgen Heideking, 2 vols. (Düsseldorf: Droste, 1980), i. 15–
18 (Hindenburg–Westarp talks, 15 Jan. 1930), 61–2
(Hindenburg–Brüning interview, 1. Mar. 1930), 87–8 (letter of
1103
the delegate Gilsa [DVP] to the general director of the Good
Hope Steel and Iron Works, Paul Reusch, 18 Mar. 1930), 94–5
(Meissner to Schleicher, 19 Jan. 1930; all emphases in the
original); Johannes Hürter, Wilhelm Groener. Reichswehrminister
am Ende der Weimarer Republik (1928–1932) (Munich:
Oldenbourg, 1993), 240 ff.; Winkler, Weimar (note 5), 364 ff.
(Young ballot: 368).
34. ‘Eine unheilvolle Entscheidung’, Frankfurter Zeitung, 232–4, 28
Mar. 1930; Rudolf Hilferding, ‘Der Austritt aus der Regierung’,
Die Gesellschaft, 7 (1930/I), 385–92 (386); Winkler, Schein (note
25), 797 ff.; id., Weimar (note 5), 369 ff.
35. Sten. Ber., cdxxviii. 6401 (Breitscheid, 16 July 1930); Der
deutsche Kommunismus. Dokumente, ed. Hermann Weber
(Cologne: Kiepenheuer & Witsch, 1963), 58–65 (‘programmatic
declaration’); Larry E. Jones, ‘Sammlung oder Zersplitterung? Die
Bestrebungen zur Bildung einer neuen Mittelpartei in der
Endphase der Weimarer Republik 1930–1933’, VfZ 25 (1977),
265–304; id., Liberalism (note 5), 374 ff.; Huber,
Verfassungsgeschichte, vii (note 8), 749 ff.; Heinrich August
Winkler, Der Weg in die Katastrophe. Arbeiter und Arbeiterbewegung
in der Weimarer Republik 1930–1933, 2nd edn (Berlin: J. H. W.
Dietz Nachfolger, 1990), 123 ff.; id., Weimar (note 5), 375 ff.
36. Theodor Geiger, ‘Die Panik im Mittelstand’, Die Arbeit 7 (1930),
637–54; Jürgen W. Falter, Hitlers Wähler (Munich: C. H. Beck,
1991), 98 ff. (electoral currents); Thomas Childers, The Nazi
Voter: The Social Foundations of Fascism in Germany, 1919–1933
(Chapel Hill: University of North Carolina Press, 1983), 119 ff.;
1104
Richard F. Hamilton, Who Voted for Hitler? (Princeton: Princeton
University Press, 1982), 309 ff.; Jerzy Holzer, Parteien und
Massen. Die politische Krise in Deutschland 1928–1930 (Wiesbaden:
Steiner, 1975), 64 ff.; M. Rainer Lepsius, Extremer Nationalismus.
Strukturbedingungen vor der nationalsozialistischen Machtergreifung
(Stuttgart: Kohlhammer, 1964); Siegfried Weichlein, Sozialmilieus
und politische Kultur in der Weimarer Republik. Lebenswelt,
Vereinskultur, Politik in Hessen (Göttingen: Vandenhoeck &
Ruprecht, 1996); Politische Teilkulturen zwischen Integration und
Polarisierung. Zur politischen Kultur in der Weimarer Republik, ed.
Detlef Lehnert and Klaus Megerle (Opladen: Westdeutscher
Verlag, 1990); Heinrich August Winkler, Mittelstand, Demokratie
und Nationalsozialismus. Die politische Entwicklung von Handwerk
und Kleinhandel in der Weimarer Republik (Cologne: Kiepenheuer
& Witsch, 1972), 157 ff.; id., Weg (note 35), 189 ff.; id., Weimar
(note 5), 388 ff. The NSDAP platform in Parteiprogramme (note
5), ed. Mommsen, 547–50.
37. Thomas Mann, ‘Deutsche Ansprache. Ein Appell an die
Vernunft’, in Werke (note 27), xi (= Reden und Aufsätze, iii),
870–90 (889–90); ‘Braun zur politischen Lage’, Vorwärts, 433, 16
Sept. 1930; Max Seydewitz, ‘Der Sieg der Verzweiflung’,
Klassenkampf, 4/18 (1930), 545–50 (15 Sept.); ‘Für Republik und
Arbeiterrecht. Entschliessung der sozialdemokratischen
Reichstagsfraktion,’ Vorwärts, 465, 4 Oct. 1930; Sten. Ber., cdxliv.
64 (Strasser, 17 Oct. 1930), 72 (Pieck, 17 Oct. 1930); ‘Ferien
vom Reichstag’, Vorwärts, 583, 13. Dec. 1930; ‘E. H. [Ernst
Heilmann], Frick und Flick’, Das Freie Wort 2/49 (1930), 1–4 (7
Dec.); ‘Erziehung zur Demokratie’, Vorwärts, 591, 18 Dec. 1930;
1105
Huber, Verfassungsgeschichte, vii (note 8), 685 ff. (Reichswehr
case in Ulm); Winkler, Weg (note 35), 207 ff.; id., Weimar (note
5), 391 ff.
38. Sozialdemokratischer Parteitag in Leipzig 1931 vom 31. März bis 5.
Juni im Volkshaus. Protokoll (Berlin: J. H. W. Dietz, 1931), 114
(Sollmann); ‘Mahnruf an die Partei,’ Klassenkampf, 5/13 (1931–
2), 384–5 (7 Jan.); Rudolf Hilferding, ‘In Krisennot’, Die
Gesellschaft 8 (1931/II), 1–8 (1); Karl-Erich Born, Die deutsche
Bankenkrise 1931. Finanzen und Politik (Munich: Piper, 1967), 64
ff.; Harold James, The Reichsbank and Public Finance in Germany
1924–1933: A Study of the Politics of Economics during the Great
Depression (Frankfurt: F. Knapp, 1985), 173 ff.; Gerhard Schulz,
Von Brüning zu Hitler. Der Wandel des politischen Systems in
Deutschland 1930–1933 (Berlin: de Gruyter, 1992), 384 ff.;
Andreas Rödder, Stresemanns Erbe: Julius Curtius und die deutsche
Aussenpolitik 1929–1931 (Paderborn: Schöningh, 1996), 186 ff.;
Winkler, Weg (note 35), 288 ff.; id., Weimar (note 5), 408 ff.
39. ‘Es geht ums Ganze’, Vorwärts, 478, 12 Oct. 1931; ‘Die
Harzburger Inflationsfront’, ibid. 479, 13 Oct. 1931; ‘Schlagt
Hitler!’, ibid. 97, 27 Feb. 1932; Das Deutsche Reich von 1918 bis
heute. Jg. 1932, ed. Cuno Horkenbach (Berlin: Verlag für Presse,
Wirtschaft und Politik, 1933), 43–57 (Hindenburg commitee,
Kyffhäuser Union, Goebbels on the presidential election); Sten.
Ber., cdxlvi. 2250 (Goebbels), 2254 (Schumacher); Schulthess’
Europäischer Geschichtskalender, 73 (1932) (Munich: Beck, 1933),
58–9 (Brüning’s speech); Knut Borchardt, ‘Das Gewicht der
Inflationsangst in den wirtschaftspolitischen
1106
Entscheidungsprozessen während der Weltwirtschaftskrise’, in
Die Nachwirkungen der Inflation auf die deutsche Geschichte 1924–
1933, ed. Gerald D. Feldman (Munich: Oldenbourg, 1985), 233–
60; Thilo Vogelsang, Reichswehr, Staat und NSDAP. Beiträge zur
deutschen Geschichte 1930–1933 (Stuttgart: Deutsche Verlags-
Anstalt, 1962), 147 ff.; Volker R. Berghahn, Der Stahlhelm. Bund
der Frontsoldaten 1918–1935 (Düsseldorf: Droste, 1966), 195 ff.;
id., ‘Die Harzburger Front und die Kandidatur Hindenburgs für
die Präsidentschaftswahlen 1932’, VfZ 13 (1965), 64–82; Rudolf
Morsey, ‘Hitler als braunschweigischer Regierungsrat’, VfZ 8
(1960), 419–48; Hagen Schulze, Otto Braun oder Preussens
demokratische Sendung. Eine Biographie (Frankfurt: Propyläen,
1977), 719–20; Huber, Verfassungsgeschichte, vii (note 8), 925 ff.;
Winkler, Weg (note 35), 511 ff.; id., Weimar (note 5), 444 ff.
(Thälmann: 445).
40. Ernst Thälmann, ‘Letzter Appell’, Rote Fahne, 75, 8 Apr. 1932;
‘An alle deutschen Arbeiter’, ibid. 89, 26 Apr. 1932; ‘Kampfmai
gegen Hunger, Krieg, Faschismus,’ ibid. 93, 30 Apr. 1932;
Horkenbach, 1932 (note 39), 76–85 (Stahlhelm and DNVP
positions); ‘Landtagsschluss!’, Der Abend. Spätausgabe des
Vorwärts, 171, 12 Apr 1932; Jürgen W. Falter, ‘The Two
Hindenburg Elections of 1925 und 1932: A Total Reversal of
Voter Coalitions’, CEH 23 (1990), 225–41; Horst Möller,
Parlamentarismus in Preussen 1919–1932 (Düsseldorf: Droste,
1985), 386 ff.; Dietrich Orlow, Weimar Prussia 1925–1933: The
Illusion of Strength (Pittsburgh: University of Pittsburg Press,
1991), 68–9; Hans-Peter Ehni, Bollwerk Preussen? Preussen-
Regierung, Reich-Länder-Problem und Sozialdemokratie 1928–1932
1107
(Bonn: Verlag Neue Gesellschaft, 1975), 244 ff.; Richard
Breitman, German Socialism and Weimar Democracy (Chapel Hill:
University of North Carolina Press, 1981), 178 ff.; Siegfried
Bahne, Die KPD und das Ende von Weimar. Das Scheitern einer
Politik 1928–1932 (Frankfurt: Campus, 1976), 23 ff.; Die
Generallinie. Rundschreiben des Zentralkomitees der KPD an die
Bezirke 1929–1933, introd. Hermann Weber with Johann
Wachtler (Düsseldorf: Droste, 1981), pp. xlvi ff.; Peter Longerich,
Die braunen Bataillone. Geschichte der SA (Munich: C. H. Beck,
1989), 153 ff.; Weingartner, Stalin (note 31), 119 ff.; Winkler,
Weg (note 35), 385 ff. (1931Prussian referendum), 491 ff. (Stalin,
Nov 1931), 528 ff. (second presidential ballot), 545 ff. (Landtag
elections on 24 Apr. 1932); id., Weimar (note 5), 449 ff. (SA
ban); Huber, Verfassungsgeschichte, vii (note 8), 938 ff.
41. Sten. Ber., cdxlvi. 2545–50 (Groener,. 10 May 1932), 2593–602
(Brüning, 11 May 1932); Die Tagebücher von Joseph Goebbels.
Sämtliche Fragmente, ed. Elke Fröhlich, part 1, Aufzeichnungen
1924–1941, ii. 1 Jan. 1931–31 Dec. 1936 (Munich: K. G. Saur,
1987), 166–7; Politik (note 33), ii. 1486–99 (settlement measure;
letters to Hindenburg; 1486–7: Gayl’s letter of 24 May 1932);
AdR. Die Kabinette Brüning I und II. 30. März bis 10. Oktober 1931,
10. Oktober 1931 bis 1. Juni 1932, 3 vols., ed. Tilman Koops
(Boppard: H. Boldt, 1982–1990), iii. 2578–9 (resolution of the
German Nationalist fraction); Heinrich Brüning, Memoiren 1918–
1934 (Stuttgart 1970), 597 ff.; Werner Conze, ‘Zum Sturz
Brünings’, VfZ 1 (1953), 261–88; Heinrich Muth, ‘Agrarpolitik
und Parteipolitik im Frühjahr 1931’, in Staat, Wirtschaft und
Politik in der Weimarer Republik. Festschrift für Heinrich Brüning,
1108
ed. Ferdinand A. Hermens and Theodor Schieder (Berlin:
Duncker & Humblot, 1967), 317–60; Udo Wengst, ‘Schlange-
Schöningen, Ostsiedlung und die Demission der Regierung
Brüning’, GWU 30 (1979), 538–51; Merkenich, Grüne Front (note
3), 310 ff.; Schulz, Brüning (note 38), 859 ff.; Huber,
Verfassungsgeschichte, vii (note 8), 956 ff.; Winkler, Weg (note
35), 560 ff.; id., Weimar (note 5), 461 ff.
42. Sten. Ber., cdxlvi. 2331 (Brüning, 24 Feb. 1932); Knut Borchardt,
‘Zwangslagen und Handlungsspielräume in der grossen
Weltwirtschaftskrise der frühen dreissiger Jahre: Zur Revision
des überlieferten Geschichtsbildes’, in Wachstum, Krisen,
Handlungsspielräume der Wirtschaftspolitik. Studien zur
Wirtschaftsgeschichte des 19. u. 20. Jahrhunderts (Göttingen:
Vandenhoeck & Ruprecht, 1982), 165–82; id., Perspectives on
Modern German Economic History and Policy (Cambridge:
Cambridge University Press, 1991); Carl-Ludwig Holtfrerich,
‘Alternativen zu Brünings Wirtschaftspolitik in der
Weltwirtschaftskrise?’, HZ 235 (1982), 605–31; Harold James,
‘Gab es eine Alternative zur Wirtschaftspolitik Brünings?’, VSWG
70 (1983), 523–41; Gottfried Plumpe, ‘Wirtschaftspolitik in der
Weltwirtschaftskrise. Realität u. Alternativen’, GG 11 (1985),
326–57; Peter-Christian Witt, ‘Finanzpolitik als Verfassungs- und
Gesellschaftspolitik des Deutschen Reiches 1930–1932’, GG 8
(1982), 386–414; Economic Crisis and Political Collapse: The
Weimar Republic 1924–1933, ed. Jürgen Baron von Kruedener
(New York: Berg, 1990); Josef Becker, ‘Heinrich Brüning und das
Scheitern der “konservativen Alternative,”’ APZ 22 (1980), 3–17;
Udo Wengst, ‘Heinrich Brüning und die “konservative
1109
Alternative”. Kritische Anmerkungen zu neuen Thesen über die
Endphase der Weimarer Republik’, ibid. 50, 19–26; Josef Becker,
‘Geschichtsschreibung im historischen Optativ? Zum Problem der
Alternativen im Prozess der Auflösung einer Republik wider
Willen’, ibid. 27–36; Rudolf Morsey, Zur Entstehung, Authentizität
und Kritik von Brünings ‘Memoiren 1918–1934’ (Opladen:
Westdeutscher Verlag, 1974); William L. Patch, Heinrich Brüning
and the Dissolution of the Weimar Republic (Cambridge: Cambridge
University Press, 1999); Rödder, ‘Dichtung’ (note 33), 77–8;
Winkler, Weimar (note 5), 472 ff.
43. ‘Das Kabinett der Barone’, Der Abend. Spätausgabe des Vorwärts,
254, 1 June 1932; Generallinie (note 40), 526–34 (circular from
the KPD secretariat, 14 July 1932); A.G. [Arkadij Gurland],
‘Tolerierungsscherben—und was weiter?’, Marxistische Tribüne,
2/12 (1932), 351–6 (352–3; (15 June; emphases in original);
Joachim Petzold, Franz von Papen. Ein deutsches Verhängnis
(Berlin: Buchverlag Union, 1995); Johann Wilhelm Brügel and
Norbert Frei, ‘Berliner Tagebuch 1932–1934. Aufzeichnungen
des tschechoslowakischen Diplomaten Camill Hoffmann’, VfZ 36
(1988), 131–83 (148–9: on the anti-National Socialist aspect of
the Preussenschlag); Ludwig Dierske, ‘War eine Abwehr des
“Preussenschlags” vom 20. 7. 1932 möglich?’, Zeitschrift für
Politik, 17 (1970), 197–245; Thomas Alexander, Carl Severing,
Sozialdemokrat aus Westfalen mit preussischen Tugenden (Bielefeld:
Wetfalen Verlag, 1992), 189 ff.; Léon Schirmann, Altonaer
Blutsonntag, 17. Juli 1932. Dichtungen und Wahrheit (Hamburg:
Ergebnisse, 1994); Huber, Verfassungsgeschichte, vii (note 8), 977
ff.; Winkler, Weg (note 35), 611 ff. (on contacts between Gayl
1110
and Severing: 630–1); id., Weimar (note 5), 477 ff. (figures on
the violence: 490). For the unemployment figures, see Preller,
Sozialpolitik (note 30), 166–7.
44. AdR. Das Kabinett von Papen. 1. Juni 1932 bis 3. Dezember 1932,
2 vols., ed. Karl-Heinz Minuth (Boppard: H. Boldt, 1989), i. 377–
86 (ministers’ meeting on 10 Aug. 1932), 386–91 (negotiations
on 13 Aug. 1932), 474–9 (Neudeck meeting, 30 Aug. 1932),
491–500 (meeting of the acting Prussian government, 2 Sept.
1932); Walther Hubatsch, Hindenburg und der Staat. Aus den
Papieren des Generalfeldmarschalls und Reichspräsidenten von 1878
bis 1934 (Berlin: Musterschmidt, 1966), 335–8: Meissner’s notes,
11 Aug. 1932); Horkenbach 1932 (note 39), 284 (Gayl, 11 Aug.
1932); Hitler, Reden, Schriften, Anordnungen. Februar 1925 bis
Januar 1933, v. Von der Reichspräsidentenwahl bis zur
Machtergreifung April 1932–Januar 1933, part 1, April 1932–
September 1932, ed. and comm. Klaus A. Lankheit (Munich: K. G.
Saur, 1996), 300–3 (talks on 13 Aug. 1932), 317 (telegram of 22
Aug. 1932), 318–20 (declaration on 23 Aug. 1932); Preussen
contra Reich vor dem Staatsgerichtshof in Leipzig vom 10. bis 14. u.
vom 17. Oktober 1932 (Berlin: J. H. W. Dietz, 1933), 44 (quote
from the article ‘Die Juden sind schuld’ by Goebbels);
‘Begnadigung zu Zuchthaus?’, Frankfurter Zeitung, 629/30, 24
Aug. 1932; Falter, Wähler (note 36), 34 ff.; Richard Bessel, ‘The
Potempa Murder’, CEH 10 (1977), 241–54; id., Political Violence
and the Rise of Nazism: The Storm-Troopers in Eastern Germany
1925–1934 (New Haven: Yale University Press, 1984), 157–8;
Paul Kluke, ‘Der Fall Potempa’, VfZ 5 (1957), 279–97; Thilo
Vogelsang, ‘Zur Politik Schleichers gegenüber der NSDAP 1932’,
1111
ibid. 6 (1958), 86–118 (esp. 89–90); id., Reichswehr (note 39),
256 ff.; Huber, Verfassungsgeschichte, vii (note 8), 1048 ff.;
Winkler, Weg (note 35), 681 ff.; id., Weimar (note 5), 505 ff.
45. AdR, Kabinett Papen (note 44), ii. 474–9 (Neudeck meeeting, 30
Aug. 1932; emphasis in original); Carl Schmitt, Legalität und
Legitimität (Berlin: Duncker & Humblot, 1932), esp. 88 ff., tr. as
Legality and Legitimacy (Durham, NC: Duke University Press,
2004); id., Der Hüter der Verfassung (Tübingen: Mohr, 1931), 132
ff.; Johannes Heckel, ‘Diktatur, Notverordnungsrecht,
Verfassungsnotstand mit besonderer Rücksicht auf das
Budgetrecht’, Archiv des öffentlichen Rechts, ns 22 (1932), 257–
338 (260, 310–11); Eberhard Kolb and Wolfram Pyta, ‘Die
Staatsnotstandsplanung unter den Regierungen Papen und
Schleicher’, in Die deutsche Staatskrise 1930–1933.
Handlungsspielräume und Alternativen, ed. Heinrich August
Winkler (Munich: Oldenbourg, 1992), 153–79; Dieter Grimm,
‘Verfassungserfüllung—Verfassungsbewahrung—
Verfassungsauflösung. Positionen der Staatsrechtslehre in der
Staatskrise der Weimarer Republik’, ibid. 181–97; Heinrich
Muth, ‘Carl Schmitt in der deutschen Innenpolitik des Sommers
1932’, Beiträge (note 24), ed. Schieder, 75–147; Joseph W.
Benderski, Carl Schmitt: Theorist for the Reich (Princeton:
Princeton University Press, 1983), 172 ff.; Paul Noack, Carl
Schmitt. Eine Biographie (Berlin: Propyläen, 1993), 137 ff.; Ernst
Rudolf Huber, ‘Carl Schmitt in der Reichskrise der Weimarer
Endzeit’, in Complexio Oppositorum. Über Carl Schmitt, ed. Helmut
Quaritsch (Berlin: Duncker & Humblot, 1988), 33–50; id.,
Verfassungsgeschichte, vii (note 8), 1073 ff.; Winkler, Weimar
1112
(note 5), 518 ff.
46. Sten. Ber., cdliv., 1–3 (Zetkin), 13–14 (Torgler), 14–16 (Göring),
17–21 (ballot); AdR, Kabinett v. Papen (note 44), ii. 546–61
(Papen’s radio address, 12 Sept. 1932), 576–85 (ministers’
meeting, 14 Sept. 1932), 593–600 (ministers’ meeting, 17 Sept.
1932), 754–64 (Papen’s Munich speech, 12 Oct. 1932); Walther
Schotte, Der neue Staat (Berlin: Neufeld & Henius, 1932); Edgar J.
Jung, Die Herrschaft der Minderwertigen, ihr Zerfall und ihre
Ablösung durch ein Neues Reich, 2nd edn (Berlin: Verlag Deutsche
Rundschau, 1930), tr. as Die Herrschaft der Minderwertigen The
Rule of the Inferior (Lewiston, NY: E. Mellen Press, 1995);
Wilhelm Stapel, Der christliche Staatsmann. Eine Theologie des
Nationalismus, 2nd edn (Hamburg: Hanseatische Verlagsanstalt,
1932), 255; Walter Gerhart [Waldemar Gurian], Um des Reiches
Zukunft. Nationale Wiedergeburt oder politische Reaktion?
(Freiburg: Herder & Co., 1932), 121, 123; Klaus Breuning, Die
Vision des Reiches. Deutscher Katholizismus zwischen Demokratie
und Diktatur (1929–1934) (Munich: Hueber, 1969); Bernd
Faulenbach, Ideologie des deutschen Weges. Die deutsche Geschichte
in der Historiographie zwischen Kaiserreich und Nationalsozialismus
(Munich: C. H. Beck, 1980), 35 ff.; Willi Oberkrome,
Volksgeschichte. Methodische Innovation und völkische
Ideologisierung in der deutschen Geschichtswissenschaft 1918–1945
(Göttingen: Vandenhoeck & Ruprecht, 1993), 22 ff.; Hans
Mommsen, ‘Regierung ohne Parteien. Konservative Pläne zum
Verfassungsumbau am Ende der Weimarer Republik’, in
Staatskrise (note 45), ed. Winkler, 1–18; Klaus Fritzsche, Politische
Romantik und Gegenrevolution. Das Beispiel des ‘Tat-Kreises’
1113
(Frankfurt: Suhrkamp, 1976); Ebbo Demant, Von Schleicher zu
Springer. Hans Zehrer als politischer Publizist (Mainz: Hase und
Koehler, 1971), 84 ff.; Sontheimer, Denken (note 26), 180 ff.
47. Preussen (note 44), 492–517; ‘Was bedeutet das Urteil?’,
Vorwärts, 504, 25 Oct. 1932 (emphases in original); Goebbels,
Tagebücher (note 41), 270; Schulthess 1932 (note 39), 194–6
(Papen’s address); Ehni, Bollwerk (note 40), 271 ff.; Schulze,
Braun (note 39), 763 ff.; Schulz, Brüning (note 38), 1000 ff.;
Klaus Rainer Röhl, Nähe zum Gegner. Kommunisten und
Nationalsozialisten im Berliner BVG-Streik von 1932 (Frankfurt:
Campus, 1994); Huber, Verfassungsgeschichte, vii (note 8), 1128
ff.; Winkler, Weg (note 35), 765 ff.; id., Weimar (note 5), 529 ff.
48. Anpassung oder Widerstand? Aus den Akten des Parteivorstands der
deutschen Sozialdemokratie 1932/33, ed. Hagen Schulze (Bonn:
Neue Gesellschaft, 1975), 55 (Böchel), 71 (Wels); ‘Wahlsieg der
KPD im Feuer der Streikkämpfe’, Inprekorr, 12 (1932), 94 3025–7
(11 Jan.); ‘Die Prawda zu den Ereignissen der Reichstagswahlen
in Deutschland’, ibid. 3027–8; Stefan Altevogt,
‘Bürgerkriegsangst am Ende der Weimarer Republik, Juni 1931
bis Januar 1933’, master’s thesis, Humboldt University, Berlin,
1996, 80 (Vossische Zeitung, 8 Nov. 1932); Bernd Sösemann, Das
Ende der Weimarer Republik in der Kritik demokratischer Publizisten.
Theodor Wolff, Ernst Feder, Julius Elbau, Leopold Schwarzschild
(Berlin: Colloquium-Verlag, 1976), 162 ff.; Goebbels, Tagebücher
(note 41), 276–7. (entry for 10 Nov. 1932).
49. Horkenbach 1932 (note 39), 374 (Papen’s announcement of 8
Nov. 1932); Schulthess 1932 (note 39), 198 (Papen with
1114
Hindenburg, 10 Nov. 1932); AdR, Kabinett Papen (note 44), ii.
901–7 (ministers’ meeting, 9 Nov. 1932), 937–8 (Keppler to
Schröder, 13 Nov. 1932); Eberhard Czichon, Wer verhalf Hitler
zur Macht? Zum Anteil der deutschen Industrie an der Zerstörung der
Weimarer Republik (Cologne: Pahl-Rugenstein, 1967), 64–72 (on
the letter to Hindenburg; for the text itself, see 69–71), 73 (D
Scholz on the meeting of the Langnam Association); Henry A.
Turner, Jr., Die Grossunternehmer und der Aufstieg Hitlers (Berlin:
Siedler, 1985), 358 ff., orig. German Big Business and the Rise of
Hitler (Oxford: Oxford University Press, 1985); Reinhard Neebe,
Grossindustrie, Staat und NSDAP 1930–1933. Paul Silverberg und
der Reichsverband der Deutschen Industrie in der Krise der Weimarer
Republik (Göttingen: Vandenhoeck & Ruprecht, 1981), 167–8;
Heinrich Muth, ‘Das “Kölner Gespräch” am 4. Januar 1933’,
GWU 37 (1986), 463–80, 529–41; Petzold, Papen (note 43), 119
ff.; Winkler, Weimar (note 5), 538 ff.
50. AdR, Kabinett Papen (note 44), ii. 956–63 (ministers’ meeting, 17
Nov. 1932), 984–6 (Hindenburg–Hitler talks, 19 Nov. 1932),
988–92 (Hindenburg–Hitler talks, 21 Nov. 1932), 998–1000
(Meissner to Hitler, 24 Nov. 1932), 1012–22 (ministers’ meeting,
25 Nov. 1932), 1025–7 (notes by retired State Secretary Hans
Schäffer on the Hindenburg–Papen–Schleicher talks, 26 Nov.
1932, with extracts from the journal of Schwerin von Krosigk);
Die Gewerkschaften in der Endphase in der Republik 1930–1933, ed.
Peter Zahn with Detlev Brunner (= Quellen zur Geschichte der
deutschen Gewerkschaftsbewegung im 20. Jahrhundert, iv) (Cologne:
Bund-Verlag, 1988), 766–70 (Schleicher–Leipart–Eggert talks, 28
Nov. 1932); Archiv der sozialen Demokratie, Bonn, ADGB-
1115
Restakten, NB 112: Verhandlungen mit der Reichsregierung
(Breitscheid’s notes on his talk with Schleicher, 28 Nov. 1932;
here also Schleicher’s report on his talk with Hindenburg on 26
Nov); Richard Breitman, ‘German Socialism and General
Schleicher,’ CEH 9 (1976), 352–88 (esp. 367 ff.); Winkler, Weg
(note 35), 793 ff.; id., Weimar (note 5), 543 ff.
51. Theodor Leipart, Die Kulturaufgaben der Gewerkschaften. Vortrag
in der Aula der Bundesschule in Bernau am 14. Oktober 1932
(Berlin: Verlagsgesellschaft des Allgemeinen Deutschen
Gewerkschaftsbundes, 1932), 3, 16–20; Ernst Jünger, Der
Arbeiter. Herrschaft und Gestalt (Hamburg: Hanseatische
Verlagsanstalt, 1932); ‘Papen nicht!’, Vorwärts, 562, 29 Nov.
1932; ‘Sturm in den Betrieben’, ibid.; ‘Alarmierende Gerüchte’,
ibid. (all emphases in the original); Peter Zahn, ‘Gewerkschaften
in der Krise. Zur Politik des ADGB in der Ära der
Präsidialkabinette 1930 bis 1933’, in Solidarität und
Menschenwürde. Etappen der deutschen Gewerkschaftsgeschichte von
den Anfängen bis zur Gegenwart, ed. Erich Matthias and Klaus
Schönhoven (Bonn: Verlag Neue Gesellschaft, 1984), 233–53 (the
effect of the Bernau speech: esp. 251–2); Axel Schildt,
Militärdiktatur mit Massenbasis? Die Querfrontkonzeption der
Reichswehrführung um General von Schleicher am Ende der
Weimarer Republik (Frankfurt: Campus, 1981), 109 ff.; Peter
Hayes, ‘“A Question Mark with Epaulettes”? Kurt von Schleicher
and Weimar Politics’, JMH 52 (1980), 35–65; Winkler, Weg (note
35), 746 ff.; id., Weimar (note 5), 550 ff.
52. AdR, Kabinett Papen (note 44), ii. 1035–6 (ministers’ conference,
1116
2 Dec. 1932), 1036–8 (Krosigk’s journal notes for 2 Dec. 1932),
1039–40 (ministers’ conference, 2 Dec. 1932); Franz von Papen,
Der Wahrheit eine Gasse (Munich: P. List, 1952), 243–52
(Hindenburg quote: 250); id., Vom Scheitern einer Demokratie
1930–1933 (Mainz: Hase & Koehler, 1968), 308–14: Wolfram
Pyta, ‘Vorbereitungen für den militärischen Ausnahmezustand
unter Papen/Schleicher’, MGM 51 (1992), 385–428; Vogelsang,
Reichswehr (note 39), 332 ff., 482 ff.; Huber,
Verfassungsgeschichte, vii (note 8), 1154 ff.; Winkler, Weimar
(note 5), 553 ff.
53. Goebbels, Tagebücher (note 41), 299 (entry for 10 Dec. 1932);
Thilo Vogelsang, ‘Neue Dokumente zur Geschichte der
Reichswehr 1930–1933’, VfZ 2 (1954), 397–436 (Schleicher’s
speech: 426–8); AdR. Das Kabinett von Schleicher. 3. Dezember
1932 bis30. Januar 1933, ed. Anton Golecki (Boppard: H. Boldt,
1986), 101–17 (Schleicher’s inaugural address); Udo
Kissenkötter, Gregor Strasser und die NSDAP (Stuttgart: Deutsche
Verlags-Anstalt, 1978), 170 ff.; Huber, Verfassungsgeschichte, vii
(note 8), 1162 ff.; Winkler, Weg (note 35), 810 ff.; id., Weimar
(note 5), 557 ff. (557–8: press reaction to Schleicher’s
appointment, 559–60: Reichstag session, 562: Imbusch, Leipart).
54. Schulthess’ Europäischer Geschichtskalender, 74 (1933) (Munich:
C. H. Beck, 1934), 5–6 (Cologne talks), 7–8 (Schleicher–Papen
meeting, 9 Jan. 1933); Otto Meissner, Staatssekretär unter Ebert–
Hindenburg–Hitler (Hamburg: Hoffmann und Kampe, 1951), 261–
2.; Kissenkötter, Strasser (note 53), 191–2; Muth, ‘Gespräch’ (note
49), 529 ff.; Goebbels, Tagebücher (note 41), 331–2; Hans Otto
1117
Meissner and Harry Wilde, Die Machtergreifung. Ein Bericht über
die Technik des nationalsozialistischen Staatsstreichs (Stuttgart:
Cotta, 1958), 148 ff.; Henry A. Turner, Jr., Hitlers Weg zur Macht.
Der Januar 1933 (Munich: Luchterhand, 1996), 44 ff., orig.
Hitler’s Thirty Days to Power: January 1933 (Reading, MA:
Addison-Wesley, 1996); id., Grossunternehmer (note 49), 378 ff.;
Neebe, Grossindustrie (note 49), 171 ff.; Czichon, Wer verhalf
(note 49), 77 ff.; Petzold, Papen (note 43), 134 ff.; Winkler,
Weimar (note 5), 567 ff.
55. Schulthess 1933 (note 54), 5 (advisory committee, 4 Jan. 1933),
11–14 (conflict between government and National Land League);
AdR, Kabinett Schleicher (note 53), 206–20 (National Land
League, Darré emphases in original); ‘Reichstag erst am 24.
Januar’, Vorwärts, 7, 1 May 1933 (Breitscheid); Horst Gies,
‘NSDAP und landwirtschaftliche Organisation in der Endphase
der Weimarer Republik’, VfZ 15 (1967), 341–76; Bert Hoppe,
‘Von Schleicher zu Hitler. Dokumente zum Konflikt zwischen
dem Reichslandbund und der Regierung Schleicher in den letzten
Wochen der Weimarer Republik,’ ibid. 45 (1997), 629–57;
Merkenich, Grüne Front (note 3), 315 ff.; Dieter Gessner,
Agrarverbände in der Weimarer Republik. Wirtschaftliche und soziale
Voraussetzungen agrarkonservativer Politik vor 1933 (Düsseldorf:
Droste, 1976), 242 ff.; Jutta Ciolek-Kümper, Wahlkampf in Lippe.
Die Wahlkampfpropaganda der NSDAP zur Landtagswahl am 15.
Januar 1933 (Munich: Verlag Dokumantation, 1976).
56. AdR, Kabinett Schleicher (note 53), 230–43 (ministers’
conference, 16 Jan. 1933), 297–300 (Simpfendörfer to
1118
Schleicher, 24. Jan. 1933); Vogelsang, Reichswehr (note 39),
482–4 (Sperr); Schmitt, Verfassungslehre (note 9), 345; Ernst
Fraenkel, ‘Verfassungsreform und Sozialdemokratie’, Die
Gesellschaft, 9 (1932/II), 484–500 (492, 494); Wolfram Pyta,
‘Verfassungsumbau, Staatsnotstand und Querfront: Schleichers
Versuche zur Fernhaltung Hitlers von der Reichskanzlerschaft
August 1932 bis Januar 1933’, in Gestaltungskraft des Politischen.
Festschrift f. Eberhard Kolb, ed. id. and Ludwig Richter (Berlin:
Duncker & Humblot, 1998), 173–97; id., ‘Konstitutionelle
Demokratie statt monarchischer Restauration. Die
verfassungspolitische Konzeption Schleichers in der Weimarer
Staatskrise’, VfZ 47 (1999), 417–41; Huber, Verfassungsgeschichte,
vii (note 8), 1227 ff.; Winkler, Weg (note 35), 802 ff., 835 ff.; id.,
Weimar (note 5), 574 ff.
57. ‘Warnung an Schleicher. Breitscheid über seine Pläne,’ Vorwärts,
33, 20 Jan. 1933; ‘Staatsstreich-Pläne’, ibid. 41, 25 Jan. 1933;
Siegfried Aufhäuser, ‘Reichstag arbeite!’, ibid.; Schulthess 1933
(note 54), 21–4 (Osthilfe scandal), 21 (Hitler’s talks with
Hugenberg and Papen), 25–6 (DNVP statement); AdR, Kabinett
Schleicher (note 53), 282–3 (DNVP communiqué to Schleicher),
284–5 (Hindenburg’s reception of Schleicher), 304–5 (Kaas’
letter to Schleicher); Die Deutschnationalen und die Zerstörung der
Weimarer Republik. Aus dem Tagebuch von Reinhold Quaatz 1928–
1933, ed. Hermann Weiss and Paul Hoser (Munich: Oldenbourg,
1989), 223–7; Joachim von Ribbentrop, Zwischen London und
Moskau. Erinnerungen und letzte Aufzeichnungen (Leoni: Druffel-
Verlag, 1953), 37 ff., tr. as The Ribbentrop Memoirs (London:
Weidenfeld and Nicolson, 1954); Wolfgang Wessling,
1119
‘Hindenburg, Neudeck und die deutsche Wirtschaft’, VSWG 64
(1977), 41–73; Noack, Schmitt (note 45), 155 ff.; Huber,
Verfassungsgeschichte, vii (note 8), 1240 ff.; Winkler, Weg (note
35), 837; id., Weimar (note 5), 578 ff.
58. AdR, Kabinett Schleicher (note 53), 306–10 (ministers’
conference, 28 Jan. 1933), 310–11 (Hindenburg’s reception of
Schleicher, 28 Jan. 1933), 313 (Kastl’s and Hamm’s letter to
Meissner, 28 Jan. 1933), 314 (letter of the trade union
associations to Hindenburg, 28 Jan. 1933); Schulthess 1933 (note
54), 28–30 (Schleicher’s resignation, Papen’s exploratory talks);
‘Schleicher zurückgetreten’, Vorwärts, 48, 28 Jan. 1933; ‘Das rote
Berlin marschiert!’, ibid. 49, 29 Jan. 1933 (all emphases in the
original); Ribbentrop, Zwischen London (note 57), 40–1;
Vogelsang, Reichswehr (note 39), 382 ff.; Huber,
Verfassungsgeschichte, vii (note 8), 1251 ff.; Winkler, Weg (note
35), 849 ff.; id., Weimar (note 5), 584 ff.
59. Anpassung (note 48), 131–36 (session of the SPD party executive
on 30 Jan. 1933), 145–6 (Breitscheid in the party committee, 31
Jan. 1933); ‘Nichtangriffspakt!’, Vorwärts, 42, 25 Jan. 1933;
‘Arbeitendes Volk! Republikaner!’, ibid. 51, 31 Jan. 1933;
‘SPD-“Nichtangriffspakt” gegen die Werktätigen!’, Rote Fahne, 22,
26 Jan. 1933; Die Antifaschistische Aktion. Dokumentation u.
Chronik Mai 1932 bis Januar 1933, ed. and introd. Heinz Karl and
Erika Kücklich (Berlin: Dietz, 1965), 354–6; AdR. Die Regierung
Hitler, pt 1, 1933/34, i. 30. Januar 1933 bis 31. August 1935, ed.
Karl-Heinz Minuth (Boppard: H. Boldt, 1983), 5–10 (ministers’
conference on 31 Jan. and 1 Feb. 1933); Schulthess 1933 (note
1120
54), 32–7 (Hitler’ talks with the Center); Die Protokolle der
Reichstagsfraktion und des Fraktionsvorstands der Deutschen
Zentrumspartei 1926–1933, ed. Rudolf Morsey (Mainz: Matthias-
Grünewald-Verlag, 1969), 611–15 (sessions of the fraction and
fraction leadership on 31 Jan and 1 Feb. 1933); Rudolf Morsey,
‘Hitlers Verhandlungen mit der Zentrumsführung am 31. Januar
1933. Dokumentation’, VfZ 9 (1961), 182–94; id., ‘Die Deutsche
Zentrumspartei’, in Matthias and Morsey, Ende (note 32), 281–
453 (esp. 339 ff.); Winkler, Weg (note 35), 305 ff. (on the
Comintern’s March 1931 argument about the SPD as the ‘main
social pillar of the bourgeoisie’), 858 ff.; id., Weimar (note 5),
593 ff.
60. Hitler, Reden, v, pt 1 (note 44), 335 (4 Sept. 1932), 353 (15
Sept. 1932); Schmitt, Legalität (note 45), 50; Fraenkel,
‘Verfassungsreform’ (note 56), 491; Weber, Wirtschaft (note 12),
i. 23 (part 1, ch. 1, § 5), 157 (part 1, ch. 3, § 1); Huber,
Verfassungsgeschichte, vii (note 8), 1264 ff.; Winkler, Weg (note
35), 861 ff.; id., Weimar (note 5), 595 ff.
LOOKING AHEAD
1121
62, 64; emphasis in original); Bernd Faulenbach, Ideologie des
deutschen Weges. Die deutsche Geschichte in der Historiographie
zwischen Kaiserreich und Nationalsozialismus (Munich: C. H. Beck,
1980), 67 ff.; Stanley Suval, ‘Overcoming “Kleindeutschland”:
The Politics of Historical Mythmaking in the Weimar Republic’,
CEH 2 (1969), 312–30; the article ‘Österreichisch-deutscher
Volksbund 1920–1933’, in Lexikon zur Parteiengeschichte. Die
bürgerlichen und kleinbürgerlichen Parteien und Verbände in
Deutschland (1789–1945), ed. Dieter Fricke et al., 4 vols.
(Leipzig: VEB Bibliographisches Institut Leipzig, 1983–), iii. 3,
566–68; Karl Rohe, Das Reichsbanner Schwarz Rot Gold. Ein
Beitrag zur Geschichte und Struktur der politischen Kampfverbände
zur Zeit der Weimarer Republik (Düsseldorf: Droste, 1966), 227 ff.;
Heinrich August Winkler, Der Schein der Normalität. Arbeiter und
Arbeiterbewegung in der Weimarer Republik 1924–1930, 2nd edn
(Berlin: J. H. W. Dietz, 1987), 378 ff. (Schützinger quote: 382).
3. Giselher Wirsing, Zwischeneuropa und die deutsche Zukunft (Jena:
E. Diederichs, 1932); Reinhard Frommelt, Paneuropa oder
Mitteleuropa. Einigungsbestrebungen im Kalkül deutscher Wirtschaft
und Politik 1925–1933 (Stuttgart: Deutsche Verlags-Anstalt,
1977); Alan S. Milward, ‘Der deutsche Handel und der
Welthandel 1925–1939’, in Industrielles System und politische
Entwicklung in der Weimarer Republik, ed. Hans Mommsen et al.
(Düsseldorf: Droste, 1974), 472–84; Dörte Doering, ‘Deutsch-
österreichische Aussenhandelsverflechtung während der
Weltwirtschaftskrise’, ibid. 514–30. On the projected German-
Austrian customs union see above p.444; on the jungkonservativ
ideology of the Reich, see p.466.
1122
4. The article ‘Verein für das Deutschtum im Ausland 1881–1945’,
in Lexikon (note 2), iv, ed. Fricke et al., 282–97; Norbert
Krekeler, Revisionsanspruch und geheime Ostpolitik der Weimarer
Republik. Die Subventionierung der deutschen Minderheit in Polen
1919 –1933 (Stuttgart: Deutsche Verlags-Anstalt, 1973); Michael
Fahlbusch, Wissenschaft im Dienst der nationalsozialistischen
Politik? Die ‘Volksdeutschen Forschungsgemeinschaften’ von 1931–
1945 (Baden-Baden: Nomos, 1999), 65 ff.
5. Hitler, Reden, Schriften, Anordnungen. Februar 1925 bis Januar
1933, v. Von der Reichspräsidentenwahl bis zur Machtergreifung
April 1932–Januar 1933, part 1, April 1932–September 1932, ed.
and comm. Klaus A. Lankheit (Munich: K. G. Saur, 1996), 31
(‘Reich’ quote from the Elbing speech on 5 Apr. 1932); Frank-
Lothar Kroll, Utopie als Ideologie. Geschichtsdenken und politisches
Handeln im Dritten Reich (Paderborn: Schöningh, 1998), 65 ff.;
Herfried Münkler, ‘Das Reich als politische Macht und politischer
Mythos’, in Reich—Nation—Europa. Modelle politischer Ordnung
(Weinheim: Beltz Athenaeum, 1996), 11–59; Jean F. Neurohr,
Der Mythos vom Dritten Reich. Zur Geistesgeschichte des
Nationalsozialismus (Stuttgart: Cotta, 1957); Klaus Schreiner,
‘“Wann kommt der Retter Deutschlands?” Formen und
Funktionen von politischem Messianismus in der Weimarer
Republik’, Saeculum, 49 (1998), 107–60; Dirk Walter,
Antisemitische Kriminalität und Gewalt. Judenfeindschaft in der
Weimarer Republik (Bonn: Dietz, 1999), 209 ff.
1123
Index
1124
‘conservative revolutionaries’ and 415–16
First World War 309, 328
Goebbels 462
in higher education 209, 383–4
militant 380–1
National Socialists 493–4
nationalism and 286–90, 438–9
pan-European 70
party-based 253–5, 257–8, 265, 352, 384–5
peasant uprising 90–1
Restoration era 66, 69–70
rise of ‘modern’ 204–13
social democracy and 265
Antichrist 5, 6, 12, 13, 15, 193
Antonelli, Cardinal 179
Arco-Valley, Count Anton 355
Arminius (‘Hermann’), leader of the Cheruscans 56, 62, 251
Army 139, 246
anti-Semitism 288
conscription 477
German revolution 333–7
Hamburg points 346–7
increased in size 293
Jews in the 309
Kapp putsch 366–8, 369, 370–1
1125
military service 242
Prussian 25, 136–7, 138, 139, 142, 174–5, 182, 241
Reichswehr 395, 397, 401, 420, 454, 475
revolts during First World War 311, 324
‘stab in the back’ 360–1
Versailles Treaty terms 358, see also OHL
Army League 284
Arndt, Ernst Moritz 55, 57–60, 61, 67, 77, 106, 184, 186, 283,
290, 292, 302
Arneth, Alfred von 107
Arnim, Heinrich von 95
Arnim-Heinrichsdorff, Count 111
Arnim-Suckow, Harry von 181
Aryan race 211, 383
Ascher, Samuel 66
Association against Anti-Semitism 288–9
Association against Jewish Arrogance 287
Association of Tax and Economic Reformers 215
Auerswald, General Hans von 99
Auerswald, Rudolf von 103
Aufhäuser, Siegfried 484
Augsburg, Religious Peace of (1555) 17–18, 20
Augsburg councils 356, 414
Augustenburg, Duke Christian August von 145, 146
Augustenburg, dukes of 95, 149, 150
1126
Austerlitz, Battle of (1805) 44
Austria 4, 21–2, 27, 32, 59, 62, 96–7
absolutism 29–30, 119
Catholic doctrine of infallibility 181
counter-revolution 91–2, 102, 104, 105
customs union 444
defeat in war with Prussia 155–9
German Confederation 64–5
German Customs Union 77
German unification 105–8
Hesse-Kassel crisis 112–13
and Hungary 160–1
imposed constitutions 91, 108–9, 119
Napoleonic wars 44, 55, 63
Peace of Paris 122
psychological alienation from Germany 87–8
reforms 100
as a republic 340
Austria (Cont.)
Schleswig-Holstein 146–53, 155
Teplitz Accord 66
territorial expansion 30–1
trade policy 144–5
Versailles Treaty 357–8
war against revolutionary France 43–4
1127
war for the unification of Italy 132–4
Austria-Hungary (1867–1918) 160–1, 187, 194, 236, 248, 491
anti-Semitism 210
arms race 293
Balkans 269
First World War 296–8, 306, 312
peace negotiations with Allies 326
and Russia 229, 230
stock market crash 204
Austro-German customs union 444, 492
1128
Schleswig-Holstein question 146, 157
trade agreements 144
Baden, Prince Max von 326, 328, 333
Bakunin, Mikhail 111
Balkan wars 269, 293–6
Ballin, Albert 294
Bamberger, Ludwig 105, 133, 172, 202, 210, 211–12, 213, 221,
223
banking 86, 204, 206–7, 387, 431, 444, 445
bankruptcies 204
Barth, Emil 335, 347
Barth, Theodor 256, 277, 288
Basel, Treaty of (1795) 43
Bassermann, Ernst 278, 280, 281
Bassermann, Friedrich 88, 90
Battenberg, Alexander von, prince of Bulgaria 230
Battle of the Nations, Liepzig (1813) 64, 66, 186, 290
Bauer, Gustav 317, 327, 359, 367, 368
Bauer, Max 313
Bauhaus movement 412
Baumgarten, Hermann 170–1, 191
Bavaria 19, 20, 31, 51, 62, 112, 330
abdication 91
annexation 44, 46, 47
anti-Semitism 356–7
1129
conflict with Catholic Church 181
conflict with Weimar constitution 381–2
coup d’etat 370
customs parliament 176
deportation of Jews from 394
disarmament 374–5
founding of the Reich 188, 190
Franco-Prussian war 184
German Confederation 65
Kulturkampf 200, 202
liberalism 73, 161–2
Munich council republics 355–7
Munich putsch 396–7
North German Confederation 177–8, 179
particularism 182, 190, 392
political allegiance 155
press censorship 377
and Prussia 156
Schleswig-Holstein question 149
Bavarian Patriots’ Party 162, 178, 196
Bavarian Peasants’ League 370
Bebel, August 154, 164, 165, 176, 177, 183, 197–8, 246, 257,
258–9, 263, 265, 274–5, 281, 283–4, 299
Becher, Johannes R. 416
Becker, Carl Heinrich 421
1130
Becker, Nikolaus 78
Beckerath, Hermann von 83
Behrenhorst, Georg Heinrich von 28
Belcredi, Count 160
Belgium 60, 71, 159, 297, 306, 307, 324, 388, 389, 404
Bell, Johannes 359
Benedetti, Count 182
Benedict XV, Pope 313
Bennigsen, Rudolf von 135, 173, 193, 214, 216, 217
Berchtold, Count 298
Berlin:
alienation between Munich and 417
anti-Semitism 209–10
Bloody May Day 427
Christmas battles 347
Huguenot immigrants 26
January uprisings 348–50
Jewish citizens 289
Kapp putsch 367–8
March uprisings 355
mass strikes 322, 323
National Socialist municipal representation 430
new Reichstag building 251–2
public transport strike in 467, 468
revolution (1918) 332–3, 335–6
1131
revolutionary agitation (1848) 92–3, 103
University of 50
Berlin, Treaty of (1850) 112
Berlin, Treaty of (1926) 418
Bernhardi, Friedrich von 282–3, 285, 286
Bernstein, Eduard 259, 260–2, 263, 264, 273–4, 310, 340–1,
354, 360, 385
Beseler, Georg 108, 110
Bethmann-Hollweg, Moritz August von 121
Bethmann Hollweg, Theobald von 278, 288, 290, 294, 295–6,
319
First World War 297–8, 301, 306, 307–8, 311–13
Beust, Friedrich von 155
Biedermann, Karl 79
Bilfinger, Carl 464
Bismarck, Herbert von 231
Bismarck, Otto von 133, 143, 319, 491
and Catholic Church 181, 223–4
and colonialism 226–7, 247
consequences of 235–8
constitution and 138–9, 156–7
and Crimean War 121, 122
defence of Olmütz treaty 113, 114
on ‘Evangelical empire’ 193
founding of the Reich 188–9, 188–91
1132
Franco-Prussian war 181–8
goal to weaken Polish nobility 228
as idol 283, 284
indemnity bill 167–73
Kulturkampf 201–3, 223–4, 236
memorial celebrations 304
and National Liberals 214–15, 219–20
North German Confederation 173–9
opponents of 164
Reinsurance Treaty 229–31, 247
resignation 235
‘revolution from above’ 167, 200, 221
Schleswig-Holstein question 150
secret meetings with Lassalle 141
and social democracy 216–17
social policy 225–6
trade policy 145, 214–20, 231
war with Austria 159, 166–7
and William II 233–4
Bismarck associations 250–1
‘Blanquism’ 272, 273, 327
Bleichröder, Gerson 206
Bloch, Joseph 264
Blomberg, General von 484
Blum, Robert 101, 102, 143
1133
Bluntschli, Johann Caspar 166
Böchel, Karl 468–9
Bodelschwingh, Friedrich von 186, 197
Boeckel, Otto 253
Boer War (1899) 269
Bohemia 10, 21, 29, 64, 96, 106, 157
Bolshevism 273, 319–21, 324, 340, 341, 351, 353, 443, 479, 483
Bonaparte family 47, 48
Bonapartism 123, 130, 165, 185–6, 193
Book of Revelation 5, 57
Borchardt, Julian 310
Borkenau, Franz 14, 15
Born, Stephan 86, 100
Börne, Ludwig 70, 75
Bos, Jean Baptiste du 33
Bosnia 269, 296
Boulanger, General 229, 231
Bourbon monarchy 55, 59, 181
bourgeoisie 7, 95
anti-Semitism 286–7
appeal of National Socialism to 438
aristocratization of 240
and Bonapartism 165
culture and 33, 63
currency devaluation 400
1134
dominating German National Union 135
First World War 309
grossdeutsch idea 492
Jewish 289, 351
Marx on 129
during Morocco crisis 281
patriotic monuments 250, 290–1
political immaturity of 141–2, 171
Prussian 24, 25, 50, 68–9, 141–2
radical nationalism 250–1
Rhine province 83
social democracy and 277–8, 303
Thirty Years War 19–20
unification 80, 167
women’s movements 272
Bouvines, battle of (1214) 7, 8
Boxer rebellion (1900) 247
Bracht, Franz 456, 457, 480
Bracke, Wilhelm 177
Brandenburg 10, 19, 22–7, 62, 331, 333
Brandes, Ernst 452
Brandi, Ernst 445
Brandler, Heinrich 394, 395
Brant, Sebastian 11
Brater, Karl 143
1135
Braun, Baron Magnus von 455, 479
Braun, Otto 322, 369, 387, 408, 440, 441, 442, 451, 456, 458
Brauns, Labour Minister 392, 393
Bray, Count 178
Brecht, Bertolt 416
Breitscheid, Rudolf 277, 278, 380, 385, 436, 472, 479, 487
Brentano, Lujo 237, 317
Brest-Litovsk, Treaty of (1918) 321, 360
Brettreich, Ritter von 325
Briand, Aristide 419
Britain 6, 7, 230
admired by Germans 40, 304
Balkans crisis 294
Calvinism 14
Congress of Vienna 65
constitutionalism 8
Counter-Reformation 18
Crimean War 121
economic competition 215
First World War 297, 298
Franco-Prussian war 187
German Weltpolitik and 268–9
Germany’s ideological warfare with 304–5
and Hanover 64
imperialism 247
1136
Industrial Revolution 35
London ultimatum (1921) 373–4, 397, 402
monarchy 10
and Morocco crisis 279, 281
Napoleonic wars 44
national pride 42
naval power 245
overproduction 86
papacy and 9
Reformation 15
Schleswig-Holstein crisis 96, 98
socialist realignment 402–3
Bruck, Karl Ludwig von 120
Brüning, Heinrich 422, 433, 434, 435–6, 440–54
Brunswick 72, 157, 330
Büchner, Ludwig 154
Bukharin, Nikolai 427
Bulgaria 230, 293, 298, 326
Bülow, Count Bernhard von 246, 247, 257, 266–70
Burckhardt, Jacob 166, 172–3, 238
Burgundy, Free Margravate of 6, 21
Burke, Edmund 40
Busch, Moritz 220, 225
Büsch, Otto 25
BVP (Bavarian People’s Party) 350, 370, 404, 409, 436, 437,
1137
450, 459, 471, 486
Byzantium see Constantinople
1138
mixed marriages 81
nationalism 61, 302–3
political agenda 100–1
Prussian victory at Königgrätz 158, 179–80
war with revolutionary France 44, see also Centre Party; papacy
Catholic League 19
Catholic Pius Association 100
Cavour, Count Camillo di 132
censorship 27, 66–7, 83, 84, 87, 92, 143
Central Association of German Industrialists 215, 219
Central Cooperative Union 387, 399
Centre Party 196, 201, 203, 217, 219, 235, 241, 245, 256–8, 278,
484
Brest-Litovsk treaty 322
Cuno administration 391
Grand Coalition 428, 433–4
and Hitler’s chancellorship 488
integralist campaign 275–7
July crisis 436
KPD’s motion 464
Morocco crisis 280
opening-up membership 350–1
peace resolution 313, 315, 319
political realignments 386, 439
Reichstag elections 372, 403–4, 424, 437, 458–9, 471
1139
SDP/DDP alliance 354
Upper Silesia 377
Versailles Treaty 359
and von Bülow 267, 270, 271
war reparations 374
Weimar constitution 363–4, 484
Chamberlain, Houston Stewart 287
Charlemagne, King of the Franks 4, 11
Charles, Emperor of Austria-Hungary 312
Charles I, King of England 139
Charles V, Holy Roman Emperor 13, 16, 21
Charles VI, Holy Roman Emperor 22
Chicherin, Georgy 379, 380
China 247
choral societies 67
Christian IX, King of Denmark 145–6
Christian National Farmers’ and Rural People’s Party 424, 436
Christian Social Party 352
Christian-Social People’s Service 464, 481
Christian Social Workers’ Party 208
Christian Trade Unions 275–6, 482
Christian VIII, King of Denmark 89
Christianity 61, 208–10, 211, 213, see also Catholic Church;
Evangelicalism; Protestants
Class, Heinrich 279, 285–7, 288, 306, 328, 426
1140
Clemenceau, Georges 324
Cleves 24
coal-mining industry 355
Cohen, Hermann 210
Cohen-Reuss, Max 346
Cologne 10, 81, 99–100, 330
Colonial Association 284
colonialism 8, 21, 22, 222, 226–7, 246–7, 256, 257, 267, 268
Comintern 378, 385, 391–2, 394, 426–7, 427, 447, 450
Committee of Fifty (1848) 94, 96
communism 128, 177, 266, 310, 347–50, 370–1, 378, 382, 389,
391–2, 403, 417, 450, see also KPD
Communist League 86, 99–100
Comte, Auguste 140, 166
Concordat of Worms (1122) 9
Congress of German Economists (1858) 144
Conrad, duke of Masovia 22
conservative nationalism 221–3
Conservative People’s Party 437
‘conservative revolution’ 413–15, 466, 492
conservatives 122, 123, 173, 178, 189, 215, 217, 219, 267, 268,
352
Constant, Benjamin 68
Constantinople 4, 7, 21
constitutions 124–5
1141
Alsace-Lorraine 279, 295
Austrian 91, 108–9, 134
German Confederation 65, 143
German National Assembly 97
German Reich 189–90, 201
March governments 90
North German Confederation 175
Prussian National Assembly 104
Three Kings’ Alliance 112
Weimar Republic 354, 361–5, 381–2, 386, 413, 463, 481, 484,
489
Consul organization 380–1
Corday, Charlotte 377
Corradini, Enrico 304
Council of People’s Commissars 335–7, 336, 340–7, 354
Counter-Reformation 18–21
Crimean War (1853–56) 121–2
Crispien, Artur 385
Croatia 21, 92, 102
culture 32–5, 53, 194, 199, 305–6
‘bourgeoisification’ of 250–1
festivals 135
folk 55, 56–7
Jews 415
mass 438
1142
militarism and 303
Weimar 412
Cuno, Wilhelm 388, 391
Cunow, Heinrich 304
Curtius, Julius 424, 444–5
customs unions 77, 79, 89, 121, 144, 154, 176, 194, 444, 492
Czechoslovakia 88, 92, 96–7, 102, 418, 492
Czernin, Count 312
1143
SDP alliance 354
Upper Silesia 377–8
Versailles Treaty 359
war reparations 374
Weimar constitution 362–4
Dehio, Ludwig 192
Deimling, General von 295
Delbrück, Clemens von 295
Delbrück, Hans 307, 317, 407
Delcassé, foreign minister 268
demagoguery 66–7
democracy:
Arndt’s commitment to 59
and dictatorship debate 260–6
economic 355
inhibitive effect on revolution 341–2
Pan-German League’s opposition to 328
parliamentary 110, 271–8, 338, 340–1, 361–5, 413, 422, 443,
453, 465
rejected by Fatherland Party 316
Schleswig-Holstein question 153–4, see also suffrage
Democratic People’s Party 153–4, 184
Democratic Students Association 417
Democratic Union 277–8
democrats 81–2, 88, 89, 95, 109, 125, 131, 135, 137, 142, 143,
1144
144, 158, 164–5, 177
Dempf, Alois 9
Denmark 64, 89, 201, 228, 236
Schleswig-Holstein 95–6, 98, 112, 145–53
Deym, Count Friedrich von 107
Dibelius, Otto 302, 415
Dilthey, Wilhelm 35
Dimitrijevic, Dragutin 296
Disraeli, Benjamin 192
Dissmann, Robert 404
Dittmann, Wilhelm 310, 322, 335, 342, 347, 385, 395
DNVP (German National People’s Party) 352, 363, 372, 384,
405, 464
anti-Semitism 385, 392
in government for the first time 406
Harzburg front 445
and Hindenburg 409
July crisis 436
Luther administration 420–1
opposition to modernism 412
Rapallo Treaty 380
Reichstag elections 372, 403, 404, 424, 437, 459, 468
split in 429
Versailles treaty 359
war reparations 374
1145
Doehring, Bruno 323, 337
Dohm, Christian Wilhelm 70
Döllinger, Ignaz von 180
Dresden 157
Dreyfus, Captain Alfred 265
Droste-Vischering, Archbishop von 81
Droysen, Johann Gustav 22, 110, 132
Dryander, Ernst von 302
Du Bois Reymond, Emil 238
Duesterberg, Theodore 447, 448
Dühring, Eugen 209
Duisberg, Carl 447
Duncker, Franz 149, 158, 164
DVP (German People’s Party) 271, 351, 352–3, 405, 410, 464
Cuno administration 391
Grand Coalition 424, 428, 433
July crisis 436
Kapp putsch 368
political realignments 386, 439
Reichstag elections 403, 424, 437, 459, 468
Upper Silesia 377
war reparations 374, 387
Weimar constitution 363–4, 381
East Elbian duchies 24, 220, 233, 240, 244, 246, 267, 277, 337,
352, 411–12, 454
1146
Eastern Marches Association 248, 249, 284
Ebert, Friedrich 318, 322, 325–6, 331, 332, 333–4, 343, 345,
354, 364, 367, 372, 382–3, 391, 396, 399, 406–8
Eckhart 9
economic liberalism 204, 215–20
Economic Party 406, 436, 464
economy 247
Calvinism 26
east-west divide 144
German Reich 214–20, 282
municipal 277
stock market crashes 204, 431
United States 431
Weimar Republic 365–6, 375–6, 386–8, 391, 397, 399–400,
402, 405–6, 423, 425–6, 428, 430–2, 443–4, 483
Edict of Nantes (1598) 18
Edict of Worms 16–17
education 26, 50, 178, 200, 202, 209, 241, 364, 422, see also
universities
Ehrhardt, Lieutenant Commander 371, 374
Eichhorn, Emil 348, 349
Einstein, Albert 307, 412
Eisner, Kurt 330, 355, 356
Elbau, Julius 469
Elbe duchies 146, 147, 149, 155, 157
1147
elections see federal; parliamentary; presidential; Reichstag;
suffrage
Elisabeth, Empress of Russia 28
emigration 227, 384
employees organisations 254
employment legislation 233, 241, 267, 311, 363, 387, 405, 477
working hours 393, 399
Engels, Friedrich 85, 86, 89–90, 165, 177, 183, 187, 218, 261,
263
on Bonapartism 165
on the counter-revolution 102–3
on direct suffrage 274
and imperial constitution 111
prediction of Russian revolution 272
proletariat power 159–60
on Prussia’s political reawakening 130–1
on the Reformation 12
war for the unification of Italy 133–4
English Civil War 139, 341
Enlightenment 27, 53, 63
Epp, Franz Xaver Ritter von 376
Erdmann, Lothar 473
Ernst August, King of Hanover 75
Ersing, Joseph 482
Erzberger, Matthias 267, 276, 307, 309, 312, 313, 327, 332, 358,
1148
365–6, 376–7
Escherich, Georg 370, 375
Eulenburg, Count 143
Eulenburg, Count Botho zu 242
Eulenburg, Count Philipp von 233
Eulenburg, Franz 400
Evangelicalism 13, 15, 16, 18–19, 20
anti-republicanism 382, 411
anti-revolutionary credo 101
German nationalism 62
kleindeutsch advocates 106, 199
nationalism 302
political parties 254–6, 352
victory over the French 186–7, 193, 197–8
1149
122, 123, 145, 156, 157
federal elections 217, 224, 227, 242, 243, 256, 284–5, 287
federalism 174, 177, 195
Federation of Agriculturalists 240–1, 246, 253, 254, 287
Fehrenbach, Elisabeth 46
Fehrenbach, Konstantin 358, 372, 374
Fellisch, Alfred 395
Ferdinand I, Emperor of Austria 91, 92, 102, 108
Ferdinand I, King of Spain 21
Ferdinand II, King of Spain 19
feudalism 24, 28
Fichte, Johann Gottlieb 61, 135, 283
German nationalism 52–4, 55, 59
on Jews 60, 207, 384
Marx and 85
Ficker, Julius 143
film industry 415, 438
First World War (1914–18):
armistice 332, 339
Brest-Litovsk treaty 321–2
final year 321–9
German war aims 306–9
July crisis 314–19
political left 309–13, 319–24
road to war 296–301
1150
war propaganda 302–6
Fischbeck, Otto 327
Fischer, Herman 380
Fischer, Richard 362
Flanders 306, 383
folk culture 55, 56–7
Fontane, Theodor 232–3, 237
Forckenbeck, Max von 137, 142, 175, 210, 214, 223
Forster, Johann Georg 38
Forstner, Lieutenant von 295
Fraenkel, Ernst 481, 490
France 6, 223, 229, 231, 410
anti-Semitism 70, 210
Arndt’s ‘bloody hatred of’ 58
Austro-Prussian war 155–6, 158–9
Baghdad railway 268
Bourbon restoration 55, 59
Congress of Vienna 65
Counter-Reformation 18
Crimean War 121–2
as cultural model 33
First World War 298–9, 306
Franco-Prussian War 181–8
French Revolution 36–42, 52, 59, 60, 61. 85, 193, 207, 341
Germany’s war reparations 397–8
1151
imperialism 247
increased military service 293
and Italy 248
Jahn’s demonization of 57
July Revolution 70–1
liberalism concept 68
Locarno accords 416
medieval Reich 8
monarchy 10
and Morocco 268, 279–80, 281
papacy and 9
Paris Commune 192
protectionism 221
Rapallo Treaty 380
Revolution of 1848 90, 104
revolutionary and Napoleonic wars 43–8, 58–9, 63
Rhine Crisis 78–9
Rhineland occupation 426
Ruhr occupation 388–9, 404
Schmalkaldic War 17
socialism 125, 265–6, 403
territorial encroachment 32
Thirty Years War 19, 20–1
trade agreements 144, 176
Versailles Treaty 357
1152
war for the unification of Italy 132
Francis Ferdinand, Crown Prince 296
Francis I, King of France 13, 16, 30
Francis II, Holy Roman Emperor 45–6
Francis Joseph, Emperor of Austria 108, 134, 143, 312
Francke, August Hermann 26
Franckenstein, Georg von 219
Franco-Prussian War (1870–1) 181–91
Frank, Ludwig 309
Frankfurt 159
Frankfurt uprising (1848) 99
Franks 4, 5, 11
Frantz, Constantin 207
Franz, Georg 180
fraternities 66, 67, 75
Frederick, Emperor of German Reich 210, 215, 232
Frederick, King of Württemberg 47
Frederick I, Holy Roman Emperor 5
Frederick I, King of Prussia 6–7, 22, 56
Frederick II, Holy Roman Emperor 7, 8
Frederick II, King of Prussia 22, 23, 25, 27–8, 30, 43, 62, 283,
305
Frederick III, Holy Roman Emperor 10
Frederick VII, King of Denmark 145
Frederick VIII, Duke of Schleswig-Holstein 146, 147
1153
Frederick William, Count of Brandenburg 103
Frederick William, Elector of Hesse-Kassel 112
Frederick William, Great Elector 22
Frederick William I, King of Prussia 25, 26
Frederick William III, King of Prussia 51, 54, 67, 81
Frederick William IV, King of Prussia 77–8, 81, 84, 86, 92, 93,
103, 110, 111–12, 119
Free Conservative German Empire Party 189, 195, 215, 217,
224, 226, 227, 246, 248–9, 271, 278, 352
free trade 51, 69, 76, 77, 80, 100, 144, 154, 178, 214, 215, 220,
221, 222, 225, 244
Free Trade Unions 264, 274, 275, 343, 367, 368–9, 375, 390,
393, 399, 451, 472, 474, 484
freedom of the press 72, 88, 90, 260, 377
French language 53, 57
French Revolution (1789) 36–42, 52, 59, 60, 61, 85, 193, 207,
341
Freud, Sigmund 412
Freyer, Hans 414
Freytag, Gustav 205
Frick, Wilhelm 439, 461, 483, 486
Friedberg, Robert 318, 327
Friedlaender, George 237
Friedland, Battle of (1807) 49
Fritsch, Theodor 209, 287–8, 289
1154
Fröbel, Julius 143
Frölich, August 394, 396
Gagern, Heinrich von 79, 88, 90, 91, 94, 97, 103, 105, 108, 109,
133, 231, 308
Galicia 87
Gall, Lothar 231
Gambetta, Léon 186
Gareis, Carl 376
Gastein Convention (1865) 150, 151, 155, 157
Gay, Peter 412
Gayl, Baron Wilhelm von 452, 455, 456, 460, 462–3, 464, 465,
469
Gebsattel, Konstantin Baron von 288, 328
Geffcken, Heinrich 237
Geiger, Theodor 438
General German Workers’ Association 141, 154, 165
Gentz, Friedrich 40, 48–9, 52
Gerlach, Ernst Ludwig von 103, 120, 133, 163–4
Gerlach, Hellmut von 277
Gerlach, Leopold von 103, 122, 123, 133
German Combat League 393
German Confederation (1806–66) 64–5, 89, 94–6, 98, 101, 108,
143, 151, 155, 157, 159, 161, 162–3
German Conservative Party 195, 215, 221, 224, 278, 352
1155
anti-Semitism 384
fall of Bülow 270, 271
federal elections 195
imperial citizenship 295
on mass strikes 323
Morocco crisis 280
Tivoli platform 253, 255, 384
German Customs Union 77, 79–80, 89, 121, 144, 176, 194
German Delegate Conference (Frankfurt) 143–4, 151
German Democratic Party see DDP
German Ethnic Freedom Party (later National Socialist
Freedom Party) 385, 401, 403, 406
German Ethnic Party 352
German Farmers’ party 464
German Liberal Party 225
German National Assembly (1848) 96, 97, 98–9, 100, 101–2,
104, 105, 106–8, 110, 163
German National League of Commercial Employees 254, 287
German National People’s Party see DNVP
German National Protective and Defensive League 380, 382
German National Union 135, 137, 142, 143, 144, 146, 151, 153,
173
German Nationalist Cooperative Union (formed from DNVP)
429, 468, 480, 483, 485, 486
German New Guinea 227
1156
German Party (Wurttemberg) 161
German People’s Partysee DVP
German Progressive Party 137–8, 140, 141, 142–3, 148–9, 152,
156, 225, 239
German Protestant Association 186, 202
German Reform Party 209
German Reform Union 143, 144, 146
German Reich (1871–1918) 188–338
abdication of the emperor 333–4
anti-socialist legislation 216–17, 233, 242, 243
Bülow era 266–70, 278
chancellorship crisis 313, 318
citizenship of 294–6
colonialism 226–7, 246–7, 267
Council of People’s Commissars’ 335–6
economy 214–20, 282
first parliamentary government 326–8
First World War 296–32
‘Germanization of the soil’ policy 227–8, 307
Kulturkampf 200–4, 206, 223–4, 236, 256
military policy 214, 229, 256
modern anti-Semitism 204–13
monument to founding of 290–1
Morocco crisis 268, 278–84
naval policy 244–7, 252, 256, 268–9
1157
‘parliamentary rule’ question 271–8, 313–18, 325–6
party system 252–66
presidential elections 285
revolution 330–8
Sammlungspolitik 243–4, 246
shift to conservative nationalism 221–3
social policy 225–6
trade treaties 240–1
Weltpolitik 246–8, 268–9, 282–3
‘German societies’ 33
German State Party 437, 439, 459, 464
German Students’ Associations 420–1
Germania 62, 115
Germanic Order 376
Germany (before 1871) 156
campaign for the imperial constitution 111
democratic movement 153–4
fear of revolutionary socialism 125
Hesse-Kassel crisis 112–13
impact of July Revolution on 71–7
liberalism and 115
March governments 90–1
philhellenism 69
Preparliament (1848) 94
Restoration era 64–70
1158
Rhine Crisis 78–9
Schleswig-Holstein question 98, 146
socio-economic change 118
war for the unification of Italy 133–4
Gessler, Otto 369, 376, 377, 392, 394, 406, 447
Gierke, Otto von 307
Giesberts, Johann 317
Gilbert, Parker 425, 431
Gilbert, Paul 422
Glagau, Otto 206
Gleichen-Russwurm, Heinrich von 465
Gneisenau, Field Marshal Neidhardt von 50
Gneist, Rudolf von 288
Gobineau, Count Joseph Arthur de 211
Goebbels, Joseph 447, 451, 462, 467, 469, 476, 478
Goering, Herman 463–4, 483, 486
Goerres Society for the Cultivation of Science in Catholic
Germany 276
Goethe, Johann Wolfgang von 30, 31, 34, 53, 354
Görres, Joseph 65–6
Göttingen Seven (university professors) 75–6, 77
Gradnauer, Georg 376
Graefe, Albrecht 384, 385
Gramont, Antoine Agénor, duc de 182, 183
Graupe, labour minister 395
1159
Greece 33, 68–9, 293
Greek-Russian Orthodoxy 14
Gregory VII, Pope 5–6, 8
Grey, Sir Edward 279, 294, 298
Grimm, Jacob 75, 77, 81
Grimm, Wilhelm 75, 77
Gröber, Adolf 327
Groener, General Wilhelm 329, 343, 358, 359, 388, 423, 432,
436, 445, 449–50, 451, 452
Gropius, Walter 412
grossdeutsch solution 104–10, 133, 143, 153, 161, 167, 192, 340,
466, 491–2
Guelfs 7, 163, 175, 177, 189, 201, 227, 245
Guesde, Jules 266
Gurian, Waldemar 466
Gurland, Arkadij 458
Gutzkow, Karl 75
gymnastic movement 55, 57, 66, 67, 135, 202, 287
1160
Hammer League 287, 288
Hammerstain, General von 449
Hänel, Albert 236
Hanover 64, 72, 91, 111–12, 121, 157
Hanse League for Commerce, Trade and Industry 277
Hansemann, David 83, 88, 94
Hardenburg, Karl August von 50, 51, 62
Hardtwig, Wolfgang 33, 251
Harkort, Friedrich 83, 152, 172
Harnack, Adolf von 307, 407
Hauenstein, Heinz 389
Hauptmann, Gerhart 417, 447
Haussmann, Conrad 297, 327, 330
Heckel, Johannes 463, 481
Hecker, Ewald 470
Hecker, Friedrich 88, 94–5
Heeringen, chief of the admiralty 294
Hegel, Georg Wilhelm Friedrich 35, 55
on the French Revolution 36
on Prussian self-image 51
on the Reformation 12
on statelessness of Germany 45
Heidegger, Martin 413
Heilmann, Ernst 442
Heine, Heinrich 75, 211
1161
Heine, Wolfgang 264, 343, 369
Heinrich I 5
Heinze, Karl Rudolf 395
Held, Heinrich 408, 485
Helfferich, Karl 366
Hellpach, Willy 408
Henning, Wilhelm 384–5
Henry I, King 56
Henry II, King of France 17
Henry IV, Emperor 5–6
Henry IV, King of Navarre 18
Henry the Lion 7
Henry VI, Emperor 7
Henry VIII, King of England 15
Hep-Hep Riots (1819) 69
Heppenheim circle (1847) 88–9
Herder, Johann Gottfried 34, 38, 70
Herero rebellion (1906) 267
Hergt, Oskar 385
Hermann (Arminius), leader of the Cheruscans 56, 62, 251
Hermann monument, Teutoburg Forest 251
Herriot, Édouard 403
Hertling, Georg Baron von 276, 280, 318, 319, 326
Herzegovina 269
Herzl, Theodor 289
1162
Hess, Heinrich von 166
Hesse 72, 90, 91, 159
Hesse-Darmstadt, landgraviate of 65, 79, 159, 172, 190
Hesse-Kassel, principality of 19, 112–13
Heuss, Theodor 317
Heydt, August von der 138
Hilferding, Rudolf 310, 342, 385, 387, 388, 391, 392–3, 397,
424–5, 431, 433, 434–5, 441, 443–4
Hindenburg, Oskar von 449, 482, 483
Hindenburg, Paul von Benneckendorff und von 191, 304, 311,
314, 326, 331, 343, 452, 470, 476
appointment of Schleicher as chancellor 475
austerity measures 443
camarilla 453
dissolution of Reichstag 422, 462–3
and Hitler’s chancellorship 488–9, 494
and Hitler’s leadership negotiations 460–1, 471–2, 478, 483–4
July crisis 436
Neudeck estate 482
Potempa case 462
presidential elections 409–11, 446–9
unconstitutional ministerial appointment 486
Young plan 432–4
Hintze, Otto 22–3, 24
Hintze, Paul von 327
1163
Hirsch, Paul von 348, 358, 368
Hirtsiefer, Welfare Minister 457
Hitler, Adolf 357, 382, 385, 394, 426, 430, 438, 439, 443, 451,
454, 469, 471
accession to power 491
chancellorship 487–90, 494
as defence witness 440
on French occupation of the Ruhr 389
and German Combat League 393
Harzburg front 445
imprisonment 401
leadership negotiations 460–1, 471–2, 473, 478, 482–3
Munich putsch 396–7
obtaining German citizenship 447
Potempa case 462
presidential elections 447–9
Reich mythology 493
secret talks with Schleicher 451–2
secret talks with von Papen 477–8
speeches before 1933 494
and Strasser 476–7, 480
support from industry and banking 470
uncontested Führer 423, see also National Socialists
Hödel, Max 216
Hoffmann, Adolph 350
1164
Hoffmann, General Max 322
Hoffmann, Johannes 356, 370
Hohenlohe-Ingelfingen, Prince Adolf von 138, 144
Hohenlohe-Langenburg, Hermann von 226
Hohenlohe-Schillingsfürst, Prince Chlodwig zu 161–2, 178, 242
Hohenstaufen dynasty 5–7, 61
Hohenzollern dynasty 22, 59, 64, 93, 160, 181–2, 238, 271, 304,
329, see also Prussia
Hohenzollern-Sigmaringen, Prince Karl Anton von 130
Hohenzollern-Sigmaringen, Prince Leopold von 181
Hölder, Julius 153
Hölderlin, Friedrich 36
Hollmann, Admiral 245
Holstein 64, 150, 155, 157
Holstein, Friedrich von 235
Höltermann, Karl 450
‘Holy Coat,’ Trier (1844) 82
Holy Roman Empire (843–1806) 4–11, 5, 6, 22, 64
confessional schism 4, 12–19
dissolution of 45–6, 54
imperial patriotism 31–2
loss of prestige 31, 32, 35
myth of 466, 492–4
Napoleonic wars 43–5
Thirty Years War (1618) 19–21
1165
war against revolutionary France 43
Hoover, President Herbert 444
Hötzendorf, Conrad von 298
Hoverbeck, Leopold von 137, 140–1
Huber, Ernst Rudolf 174
Hugenberg, Alfred 285, 426, 429, 432, 436, 445, 468, 482, 485,
486
Hughes, Charles 398
Huguenots 26
human rights 55, 59–60
humanism 11, 33, 60
Humboldt, Wilhelm von 62
Hungarian revolution (1849) 122
Hungary 21, 22, 29, 64, 87–8, 92, 102, 134, 160
hyperinflation 386–7, 389
1166
Speyer (1526) 16
imperial estates (Holy Roman Empire) 16, 19, 20–1, 23, 27, 35,
40–1, 44, 45, 56
Imperial German Middle Class Association 289–90
Imperial League against Social Democracy 284
imperialism 196, 199, 306–9, 466
Industrial Revolution 35, 51, 86, 118
industrialization 68, 80, 126, 204, 208, 240, 241
industry and industrialists 215, 219, 220, 277
Brüning administration 445
Cartel of the Productive Estates 290
commercial treaties 281
First World War 306
Hindenburg programme 311
hyperinflation 386–7
against social democracy 284
stock market crash in 1929 432
support for Hitler 470–1, 478
Weimar Republic 422
inflation 365–6, 367, 373, 375, 386, 389, 399–400
Innocent III, Pope 9, 11
International Workers’ Associations 177, 266, 310–11
Interparty Committee (1917) 313, 318, 325–6
Irene, empress-widow 4
Italy 42, 64, 132–4, 229
1167
annexation of Papal States 201
fascism 390, 415
and France 248
Napoleon III’s downfall 187
protectionism 221
Risorgimento 180
Tripolitan war 281–2
war with Austria 157
Itzenplitz, Heinrich Friedrich August von 218
1168
deportations of 394
‘Eastern’ 384
emancipation of 51, 91, 178, 207
Fichte on 61
Luther’s emnity towards 15–16
middle class 289, 351
Jews (Cont.)
in the military 309
Palestine 306
Restoration era 69
as scapegoats 383
Joël, Curt 445
Johann, Archduke 91, 97
John of Salisbury, bishop of Chartres 6
John XII, Pope 5
Jolly, Julius 177
Jordan, Wilhelm 107
Jörg, Edmund 162, 176, 178, 184, 196, 206
Joseph II, King of Austria 29–30
Jülich, Duchy of 24
Jung, Edgar 465, 466
Jünger, Ernst 473
Junkers 24, 50, 120, 140, 232–3, 239–40, 246, 271, 277, 283,
325, 343–4, 366, 452
Jura 58
1169
Jutland 146
1170
Keudell, Interior Minister von 420–1, 422, 429
Kiderlen-Waechter, Alfred von 279, 280, 281, 378
Killinger, Manfred 376
Kjellén, Rudolf 303
kleindeutsch solution 105–9, 115, 135, 143, 158, 167, 194, 455,
491
Kleist-Retzow, Hans von 195
Klepper, Otto 456
Klopp, Onno 163
Knilling, Eugen Ritter von 381
Köhler, Finance Minister 422
Königgrätz, battle of (1866) 158, 160, 179, 194, 492
Kopp, Cardinal Georg 256
Körner, Theodor 292
Koselleck, Reinhart 29
Kossuth, Lajos 92
Köstner, Adolf 382
Kotzebur, August von 66
KPD (Communist Party of Germany) 347–50, 377, 405, 409–10,
441–2
ban against 401
‘battleship A’ 423
Bloody May Day 427–8
direct action against National Socialists 487–8
Kapp putsch 367–8
1171
motion of no confidence 464
Munich councils 356
National-Bolshevist agitation 390
nationalism 437
political terror 456, 460, 467
presidential elections 447, 448–9
Reichstag elections 372, 406, 424, 437, 458, 468–9
Ruhr uprising 370–1
in Saxony 392, 394–5
secret weapons production 420
sensational cooperation with National Socialists 467, 468
and Social Democrats 418, 450–1, 457
war reparations 374
Kraus, Christian Jakob 83
Krause, Paul von 327
Krosigk, Finance Minister von 471
Kulturkampf 200–4, 206, 223–4, 236
Kun, Béla 356
Kurland 304, 321, 322
Kyffhäuser monument 250, 291
1172
land 10, 423
land reform 344
Landauer, Gustav 356
Länder 361, 365, 366, 382, 450, 458, 480
landowners 400, 482
Landsberg, Otto 335
Langbehn, Julius 287
Lange, Helen 272
Langnam association 426, 471
language(s) 5, 10, 33, 194, 199
commonality of 35, 53
foreign 53, 57
Luther’s Bible 16
official government 29, 228, 268
‘primal language’ 52
Romanticism and 81
as valid natural boundary 58–9
Lansing, Robert 329
Lasa, Ernst von Heydebrand und der 246, 280, 323
Lasker, Eduard 169, 170, 171, 179, 190, 202, 205, 211, 212, 217,
222–3, 237
Lassalle, Ferdinand 133, 141, 154, 177
Lassalleans 189, 190, 197
Latour, Count 102
Lauenburg, duchy of 64, 147, 150
1173
law 29, 46, 47, 66, 119–20, 141
League of German Women’s Associations 272
League of Nations 324, 357, 360, 391, 418–19, 444
League of Red Front Fighters 427–8
Legien, Carl 264, 275, 369
Lehmann, Max 307
Lehmann, Walter 302
Leiningen, Prince Karl von 97
Leipart, Theodor 390, 472, 473, 477
Lenin, Vladimir 273, 301, 310, 319–20, 327, 356, 402
Lensch, Paul 304
Leo III, Pope 4, 215
Leo XIII, Pope 223
Leopold I, Emperor 22
Leopold II, King of Austria 29
Lerchenfeld, Count 377, 381
Lessing, Gotthold Ephraim 34, 35
Levetzow, Albert von 251
Levien, Max 356
Leviné, Eugen 356
Liberal Association 242, 256
Liberal People’s Party 242, 271
Liberal Union 223, 224, 225, 271
liberals and liberalism 68–79
after Crimean War 122
1174
after Königgrätz (1866) 161–2, 166
anti-revolutionary credo of 87
in Baden 72, 76, 79, 154–5, 161
Bismarck on 139
Bülow era 268
and capitalism 204–6
and the Catholic Church 181, 200–1, 208
de-liberalization of 221–3
democrats and 81–2, 88, 95, 137, 154
downfall of German 437, 439
economic 204, 215–20
events of September 1848 101
founding of the Reich 200
German National Union 135, 137, 142, 143, 144, 145
moderate 114–15
National Socialism’s break with 494
nationalism and 220–3
political renaissance 277
Prussia 147, 150–1, 167–73
Rhenish 83–4
and socialism 141
war with Austria 157–8
Lichnowsky, Prince Friedrich 99
Lieber, Ernest 257
Liebermann, Max 253, 447
1175
Liebknecht, Karl 266, 309–10, 334, 335, 340, 345, 348, 349–50
Liebknecht, Wilhelm 99, 164, 177, 183, 197
Liegen, Carl 317, 343
Linsingen, General von 331, 332
List, Friedrich 80, 107
Lithuania 304, 322
Lloyd George, David 279, 324, 373, 379
Löbe, Paul 396, 411, 492
Locarno accords (1925) 416
Loebell, Friedrich Wilhelm von 409
Lohmann, Theodor 225
Lombardo-Venetia 92, 132, 134
London Protocol (1852) 145, 146
London Treaty (1840) 78
London Treaty (1871) 231
London ultimatum (1921) 373–4, 397, 402
Lorraine 58, 60, 80, 184–5, 187–8, 197
Lossow, General von 392, 396, 401
Louis IV, King of France 26
Louis XIV, King of France 21
Louis XVI, King of France 37, 139
Louis Philippe, King of France 70, 90
Löwe-Calbe, Wilhelm 142
Löwenstein, Rudolf 140
Ludendorff, Erich 311, 313, 314, 322, 324, 326, 329, 396, 401,
1176
408
Ludwig I, King of Bavaria 91
Ludwig II, King of Bavaria 190
Ludwig the Bavarian 9
Lunéville, Treaty of (1801) 44, 45
Luther, Hans 388, 391, 397, 399, 419
Luther, Martin 9, 11, 12–17, 38, 52, 53, 56, 57, 85
Lutheranism 18–19, 22, 24, 26, 82, 163–4, 337
Lüttwitz, General Walter von 366, 371
Lützow, Major von 66
Luxembourg, grand duchy of 59, 64, 176
Luxemburg, Rosa 262–3, 266, 273, 310, 331, 345, 348, 349–50
1177
manorial system 24, 39, 46, 50, 100, 120, 178, 213, 343–4, 482
Manteuffel, General Erwin von 165
Manteuffel, Otto von 113, 120, 121
Marat, Jean 377
Maria Theresa, Empress of Austria 22, 27
Marr, Wilhelm 207–8
Marschall, Baron von 245
Marsilius of Padua 9
Marwitz, Friedrich August Ludwig von der 51, 69, 207
Marx, Karl 50, 86, 90, 99, 133, 153, 165, 177, 218, 261
Bernstein’s critique of 261–2
class struggle 129–30, 259
on Franco-Prussian War 183, 187
Kautsky’s interpretation of 263–4
Paris Commune 198
on the Reformation 12, 15
revolutionary tradition in Germany 272, 273
theory of revolution 84–5
on a ‘world war’ 103
Marx, Wilhelm 399, 405, 408–9, 420, 421
Marxism 259–61, 325, 383–4, 403, 443
mass strikes 273, 274, 298, 311, 322–3, 332, 354–5, 367, 370–1,
390, 457, 467, 468
Mathy, Karl 88, 161
Matthias, Emperor 19
1178
Maximilian I, Emperor 10
Maximilian II, King of Bavaria 91
Mayer, Karl 183
Mazzini, Guiseppe 75, 88
Mecklenburg 157, 178
Mecklenburg, Duke John Albert of 316
Mehring, Franz 260, 266, 310, 320
Meinecke, Friedrich 67, 317, 407, 417
Meissner, Otto 432, 434, 452, 461, 471, 474, 478, 483, 485
Melanchthon, Philip 16
Metternich, Prince Clemens Lothar Wenzel 40, 65, 66, 75, 76,
87, 88, 91
Metz 188, 228
Mevissen, Gustav 83, 84, 104
Meyer, Eduard 307
Meyer, Gustav 83
Michaelis, Georg 314–15, 318, 319
Michaelis, Otto 151, 169, 171, 239
middle class see bourgeoisie
migrant labour 240
militarism 284, 292, 293, 305, 306–7
Millerand, Alexandre 265, 266
Miquel, Johannes von 153, 225, 226, 243–4, 246
Mitteleuropa project 308
modernism 208, 412
1179
Möhl, General Arnold Ritter von 370
Mohl, Moriz von 109
Moldavia 121
Moldenhauer, Paul 433
Molkenbuhr, Hermann 362
Moltke, Count Helmuth von 150, 186, 188, 251, 279, 294
Mommsen, Theodor 150, 152, 153, 210, 223, 237, 257, 277, 288
monarchy 10, 23, 37–8, 39, 40, 68, 119, 126–7, 337, 352, 400,
453
Montesquieu, Charles-Louis de Secondat 36, 39, 40
Montez, Lola 91
Moravia 29, 64, 96, 106
Morocco 268, 279–84
Moser, Friedrich Carl von 32
Moser, Johann Jacob 32
Möser, Justus 31
MSPD (Majority Social Democrats) 312, 313, 315, 318, 319,
320, 323, 326, 332, 377
Brest-Litovsk treaty 322
Constituent National Assembly 352
January uprisings 349–50
July crisis 359, 360
People’s Commissars 336, 343, 345, 347
political realignment 385–6
Reichstag elections 372
1180
revolution 332–3, 334, 335–6
war reparations 374
Weimar constitution 362–4
Muhammad Ali, viceroy of Egypt 78, 79
Mühlfeld, Eugen von 105
Mühsam, Erich 356
Müller, Adam 61
Müller, Admiral, chief of the navy cabinet 294, 299
Müller, August 315
Müller, Hermann 359, 369, 393, 404, 424–5, 428, 434
Müller, Richard 348
Müller-Armack, Alfred 23
Müller-Meiningen, Ernst 292–3
Munich:
alienation between Berlin and 417
council republics (1918) 355–7, 414
coup (1920) 370
German Ethnic Freedom Party 385
National Socialist putsch 396–7
Münzenberg, Willy 416
Mussolini, Benito 390
mysticism 9–10
Napoleon I, Emperor 44–9, 53, 54, 55, 59, 62, 66
Napoleon III, Emperor 104, 121, 125, 126, 130, 132, 134, 158–9,
172, 176, 180, 182–6
1181
Napoleonic Code 46, 47, 66
Nassau 159
nation states 4, 9, 10, 33
national anthem 382–3
National Assembly (1848) 96, 97, 98–9, 100, 101–2, 104, 106–8,
110, 163
National Assembly (Weimar Republic) 342, 345, 346, 348–9,
350, 352–3, 358–9, 361–5, 371–2
National Committee for the German Referendum 426, 429–30
national identity 30–1, 193–4
National Land League 445, 452, 470, 479, 482
National Liberal Party 171, 172, 173–4, 178–9, 224–5, 227, 242,
252, 351
abdication of the emperor 332
attacking Catholicism 197, 203
and Bismarck 214–15, 219–20
constitutional agenda 271
factions 222–3
federal elections 195
founding of the Reich 188, 189
‘Heidelberg declaration’ 225, 243
Jewish 205–6
Morocco crisis 280
Pan-German agenda 248–9
peace resolution 313, 315
1182
and social democracy 277–8
and von Bülow 267
National Party 79
National Social Association 254–6
National Socialist Freedom Party 406
National Socialist German Workers’ Party (NSDAP) 382, 385,
389, 392, 394, 401, 415, 423, 424, 426, 441, 454
Agrarian-Political Apparatus 479
anti-Semitism 493–4
dismal financial situation 474
gains in parliamentary elections 480
Harzburg front 445
Hitler as chancellor 487–90
media of mass communication 438
Munich putsch 396–7
national referendum 430, 436
opposition to modernism 412
political terror 455–6, 459–60, 461–2, 467
Reichstag elections 403, 406, 424, 437–8, 458–9, 468–9, 480
on the rise 430
SA and SS bans 449, 451–2, 455–6
sensational cooperation with KPD 467, 468
and Strasser 476–7, 480: see also Hitler, Adolf
nationalism 41–3, 136
and anti-Semitism 286–90, 438–9
1183
‘conservative revolutionaries’ and 415
Czech 88
German 48–63, 67, 73–4, 79, 161, 220–3, 238
Hungary 88
imperial 196, 199, 238
myth of war innocence 360
National Socialists 439
organisations and symbols 248–52
patriotic celebrations 290–3
pseudo-religious character of 57
radical 250–1, 285–7
religion and 302–3
war 302–3, 317
nationalization 344, 346, 351, 355, 365, 458
Naumann, Friedrich 254–6, 272, 308, 317, 492
Navy 135–6, 244–6, 252, 256, 268–9, 269
Balkans crisis 294
‘battleship A’ 423, 424–5
gunboat diplomacy 279, 280
mutinies 311, 318, 322–3, 329–30, 331, 347
Prussian 150–1
Versailles Treaty terms 358
war criminals 374
Navy League 245–6, 247, 256–7, 284
Nazi party see National Socialist German Workers’ Party
1184
Netherlands 14, 20, 21, 60, 64, 159
Neuhaus, Karl 406
Neurath, Baron Konstantin von 455, 486
Nicholas I, Tsar 112
Nicholas II, Tsar 298, 308, 327
Niekisch, Ernst 356, 414
Nietzsche, Friedrich Wilhelm 12, 237–8
Nikolai, Friedrich 31
Nipperdey, Thomas 250, 291
Nobiling, Dr Karl Eduard 216, 238
nobility 20, 25, 28, 46, 51, 61, 63, 67, 134, 228, 240, see also
princes
North German Confederation (1867–71) 159, 173–9, 181, 182,
183, 188, 190
Noske, Gustav 330, 349, 355, 358, 366, 367, 368, 369, 447, 458
Novalis 61
NSDAP see National Socialist German Workers’ Party
1185
Oppenheim, Heinrich Bernhard 141
Oppenheimer, Franz 309
Ott, Lieutenant Colonel Eugen 475
Otto, Bishop of Freising 5
Otto IV, Emperor 7, 8
Otto the Great 5
Ottoman Empire 78
1186
482–3, 484–5, 486–7
paramilitary organizations 374–5, 449–50, 457, 492, see also SA;
SS
Paris, siege of (1870) 186, 188
Paris Commune (1871) 192, 197, 198
parliamentary elections 104, 137–8, 143, 156, 168, 173, 233–4,
253, 267, 270, 457, 480
particularism 152, 164, 165, 182, 189, 190, 195–6
Pasic, Nikola 296
Passau, treaty of (1552) 17
Patow, Robert von 138
Patriotic Leagues 382, 393
patriotism 160
constitutional 41
cult of sacrifice 383
during Franco-Prussian war 182
German states 195
imperial 31–2, 43, 46
literary 32–5
love of the Fatherland 43, 53–4, 62
monuments 290–1
music 291–2, 382–3
nationalism and 42, 48–9
organisations and symbols 248–52
religious 62–3
1187
social 301, 304
war sermons 302–3
Paulskirche see National Assembly (1848) 16
Payer, Friedrich von 318, 327
peasantry 24, 29, 39, 46, 50–1, 86, 90, 100
The Peasants’ War (1524–5) 12
People’s Federation for Liberty and Fatherland 317–18
People’s National Reich Association 436, 437, 439
Perrot, Franz 206
Peters, Carl 227
Peters, Norbert 302
Peucker, General von 98
Pfau, Ludwig 153
Pfizer, Paul 73, 80, 91, 105, 107
Pfordten, Ludwig von der 155, 156
Pfuel, General Ernst von 103
Philip II, King of Spain 21
Philip of Anjou 21
Philip the Fair 8, 9
Pietist movement 9, 20, 26, 34–5, 62, 82, 481
Pillersdorf, Franz von 91
Pius VII, Pope 44
Pius IX, Pope 180, 201, 215
Pius X, Pope 276
Planck, Max 307
1188
Plenge, Johann 303
Plessner, Helmuth 199–200, 213
Poertner, Balthasar 302
Pöhner, Ernst 370
Poincaré, Raymond 380, 397–8, 403
Poland 60, 209, 340, 360, 418, 419, 492
anti-Semitism 384
First World War 304, 306, 308, 322, 324
Jews 384
Kulturkampf 201
partitions of 64, 96
Prussia’s dismembering of 28
revolution in 1830 72
Soviet Union and 378
Upper Silesia 373
Versailles Treaty 357
Poles 72, 227–8, 236, 240, 245, 249, 294–5
Portugal 18
Posadowsky-Wehner, Count 267
Posen, Grand Duchy of 64, 94, 96, 105, 107, 114, 222, 357
Potempa case (1932) 459–60, 461–2
Prague, Peace of (1635) 19
Preparliament (1848) 94, 96, 97
presidential elections 409–11, 446–9
press:
1189
anti-Semitic 309, 392
freedom of the 72, 88, 90, 143, 260
on Hindenburg as President 410
Jewish 289, 351, 415
Leipzig decision 467
political re-education 283
right-wing 376, 466
Social Democrat 300
war mongering 294
Press Association 73
Preuss, Hugo 240, 317, 354, 361–2
Prim, Marshal 181–2
princes 54, 62, 81
absolutism 20
alliances 151
critics of 59
German liberty 49
German Reich 190
mediatization 44, 46
medieval Reich 7
Reformation 12, 16, 17–18
regional reform 10–11
renewal of German Confederation 143
Rhenish Confederation 44–5, 46–7
unification and 63, see also nobility
1190
professors 16, 75–6, 77, 307, 417
Progressive People’s Party 189, 271, 272, 277, 292–3, 297, 313,
315, 322, 326, 351
proletariat:
appeal of National Socialism to 438
democratic movements 154, 165, 167
‘dictatorship of the’ 259–60, 262, 273, 320, 346, 414
divided 457
First World War 299–300, 309, 311, 322–3
KPD recruitment 390
political parties 177, 254–6, 258, 265–6
revolution 84–5, 128–30, 273
Ruhr uprising 370–1
Russian 273
self-help organizations 141
unemployment insurance 421–2
Workers’ and Soldiers’ Councils 336, 345–6
works committees 355, see also mass strikes
protectionism 80, 144, 176, 214, 218–20, 221, 222, 225, 231,
240, 244
Protestants 22, 53, 82, 179–80, 208, 213, 438
Prussia 4, 21, 32, 46, 441
anti-Semitism 204–13, 383
anti-socialist law 242–3
army reform budget 136–7, 138, 139
1191
bitterness against (1918) 325
constitutionalism 67–8, 119–21, 124, 127–8, 136–43, 167–73
counter-revolution 103–4
Crimean War 121–3
diplomatic relations with Vatican 224
east-west dichotomy 22–4
Franco-Prussian War 181–91
German Confederation 64–5, 96
German Union project 111–12
government efficiency 26–7
Kalish manifesto 54–5
Kulturkampf 200–4, 206, 223–4
landowners 100, 240–1
liberalism 115
militarism 28–9
modern nationalism 43
Napoleonic wars 49, 58–9, 63
National Assembly 97, 101–2, 103, 104–5
National Socialists 430
neutrality 44, 49, 134
New Era 130–2
parliamentary crisis 450–1
parliamentary rule 271
patriotism 30, 62
peasants liberation 50–1
1192
Preussenschlag 456–8, 465, 466–7
Punctuation of Olmütz 112–13
reaction to doctrine of infallibility 181
reform 39, 49–51, 62, 136–7, 139, 213–14
rejection of unification 110
revolutionary agitation in 92
and Russia 112
Schleswig-Holstein question 98, 112, 146–53, 155, 159, 162
Social Democrats 443
social militarization of 25–6
socio-economic modernization 68, 76–7, 86–7
students associations 421
Teplitz Accord 66, 67
trade policy 144–5, 176
victory in war with Austria 155–9, 162
Prussian Progressive Party 167–9, 169, 174
publishing industry 415
Pufendorf, Samuel 21, 31
Pünder, Hermann 441
Puttkamer, Robert von 232
1193
Rade, Martin 302
Radek, Karl 378, 389–90, 391
Radowitz, Joseph Maria von 109, 112
railways 86, 88, 204, 404–5
Ranke, Leopold von 16
Rapallo Treaty (1922) 378–80, 419
Rapp, Adolf 158
Rathenau, Walther 291–2, 306, 309, 375, 378–81, 382, 383,
384–5
Rauscher, Ulrich 367
Ravensburg 24
realpolitik 120, 123–5, 131, 154
Rebmann, Georg Friedrich 38, 39
Rechberg, Count 146
Red Ruhr Army 370–1
Reformation 9, 12–17, 53
Reformed Church 17, 19, 20, 22, 82
regicide 37–8
Rehdiger, Karl Niklas von 54
Reichensperger, Peter 176, 257
Reichsbank 387, 431, 445
Reichsbanner 449–50, 457, 474, 492
Reichstag elections 372, 403, 424, 458–9, 467–8, 470–1, 480–1
Reichswehr armed forces 395, 397, 401, 420, 454, 475
Reinsurance Treaty (1887) 229–31, 234–5
1194
religion 22
after Königgrätz (1866) 162–3
confessional schism 4, 12–19
democratic movement and 82
German Reich 194, 196
ideology and 160
imperial ideologues 466
and nationalism 302–3
political agenda 100–1
and state 200–4, 223–4
Weimar Republic 382, see also Catholic Church; Evangelicalism;
Protestants
Religious Peace of Augsburg (1555) 17–18, 20
religious tolerance 20, 22, 25, 29, 201
Remmele, Hermann 408
Renaissance 12
Renan, Ernest 198–9
republicanism 39, 42, 94–5, 439
Reusch, Paul 470, 478
Reuter, Ernst 367
Revolutionary Shop Stewards 311, 322, 332, 335, 346, 348
revolution(s) 40, 66, 122–3
during 1848 90–104, 114–17
and agricultural societies 341–2
American 42
1195
campaign for the imperial constitution 111
conservative 413–18, 492
French 36–42, 52, 59, 60, 61. 85, 193, 207, 341
‘from above’ 166–7, 200, 221, 327
German (1918–19) 330–7, 355–7, 370–1
Hungarian 122
July Revolution 70–1
Marx’s theory of 84–5
Paris Commune 192
reform or 260–6
Russian 272–3, 308, 319–21, 325, 327, 353
socialism 125–8
Rhenish Confederation 44–5, 46–7, 64, 73
Rhineland 83–4, 397, 400, 404, 426, 455
Ribbentrop, Joachim von 482, 483, 485
Richard Lionheart, King of England 7
Richter, Eugen 246
Rickert, Heinrich 288
Riezler, Kurt 297
Ripen, Treaty of (1460) 89
Rittershaus, Emil 185
Rochau, Ludwig August von 123–5, 128, 129, 133
Roeren, Hermann 275–6
Roes, Alexander von 8
Roggenbach, Franz von 154
1196
Röhm, Ernst 401, 461
Roman empire 4–5, 9, 10, 11, 31, 33, 52, 85
Romania 240, 293, 298
Romanticism 70, 81
Römer, Friedrich 88, 91
Roon, Albrecht von 136, 138, 150, 251
Rosenberg, Arthur 322, 348
Rosenberg, Hans 204
Rosenstock-Huessey, Eugen 8, 12, 16
Rössler, Konstantin 133
Rosterg, August 470
Rothardt, Erwin 406–7
Rothschilds 87
Rotteck, Carl von 69, 72, 74–5, 79
Rousseau, Jean-Jacques 36
Ruge, Arnold 84, 101, 107, 114, 133
Ruhr 355, 370–1, 373, 388, 400, 402, 404, 445, 455
Rümelin, Gustav 114–15
Ruprecht-Ransern, Alfred 241
Russia 110, 114, 240, 246, 293
anti-Semitism 210
army 293
and Austria 155
Brest-Litovsk treaty 321, 360
Congress of Vienna 65
1197
Crimean War defeat 121–2
Eastern treaty with Britain 268
First World War 298, 299, 300, 304, 305, 306–7, 321–2
Kalish manifesto 54–5
Napoleon III’s downfall 187
Napoleonic wars 44
protectionism 221
and Prussia 155
Reinsurance Treaty 229–31, 234–5, 247–8
revolution 272–3, 319–21, 325, 327, 353
Schleswig-Holstein question 96, 98, 112, see also Soviet Union
SA (Storm Troopers) 401, 441, 449, 451, 452, 455, 456, 457,
459–60, 461, 493
Salzburg 44
Sand, Karl Ludwig 66
Sardinia-Piedmont 92, 122, 132, 172
Sauer, Wolfgang 329
Saxony 64, 84, 157, 396
customs treaty 144
dukes of 10, 13, 17
founding of the Reich 188
KPD 392, 394–5
opposition 177
political allegiance 155
1198
revolution 72
Schleswig-Holstein 146, 147
Three Kings’ Alliance 111–12
trade tariffs 219
Saxony-Weimar 65
Scandinavia 60
Scapa Flow 358
Schacht, Hjalmar 431–2, 445
Schäfer, Dietrich 292, 307
Schäffer, Fritz 471
Scharnhorst, Gerhard 50
Scheidemann, Philipp 285, 317, 322, 327, 331, 332, 334, 335,
354, 358, 359, 360, 420
Scheler, Max 304, 305
Schellendorf, Bronsart von 231
Schelling, Friedrich Wilhelm Joseph von 36
Scheüch, Heinrich 331, 333, 336
Schiele, Martin 436
Schiffer, Vice Chancellor 368
Schiller, Friedrich 34, 45, 53, 135
Schilling, Heinz 17
Schlageter, Albert Leo 389
Schlange-Schöningen, Hans 429, 452
Schlegel, Friedrich 61
Schleicher, General Kurt von 432, 445, 449–50, 451–2, 455,
1199
460, 462–3, 464, 466, 469, 472–5
Schleiermacher, Friedrich 35, 63
Schleswig 228, 357
Schleswig-Holstein 89, 95, 98, 106, 145–53, 155–60, 430, 459
Schmalkaldic War (1546–7) 17
Schmerling, Anton von 99, 108, 134
Schmidt, Robert 375
Schmitt, Carl 173, 413–14, 463, 464, 481, 484, 490
Schmitz, Bruno 250
Schmoller, Gustav von 225, 307
Schneckenburger, Max 292
Schober, Johannes 444
Schön, Theodor von 78, 83
Schönerer, Georg Ritter von 254
Schotte, Walther 465
Schröder, Kurt von 470, 478
Schulz, Heinrich 376
Schulze-Delitzsch, Hermann 135, 137, 141, 149, 164
Schumacher, Kurt 447
Schützinger, Hermann 492
Schwander, Rudolf 315
Schwarzenberg, Prince Felix 102, 107–8, 113, 120
Schweitzer, Johann Baptist von 165
Schwerin von Krosigk, Count Lutz 455, 460, 480, 486
secularization 44, 52
1200
Sedan, Battle of (1870) 186, 197, 290
Seeberg, Reinhold 307
Seeckt, General Hans von 367, 369, 378, 392, 394, 396, 401,
445
Seghers, Anna 416
Seldte, Franz 426, 429, 486
Sepp, Johann Nepomuk 184
Serbia 230, 269, 293, 296, 297–8
Seven Years War (1756–63) 27, 30, 32, 46
Severing, Carl 369–70, 396, 425, 456, 458
Seydewitz, Max 440
Shaw, George Bernard 261
shooting clubs 135, 420
Shuvalov, Count 234
Siebenpfeiffer, Philipp Jacob 73, 74
Silesia 27, 29, 86, 358, 373, 377–8
Simpfendörfer, Wilhelm 481
Simson, Eduard 190
Simson, Heinrich 109, 110
Slavs 102, 103, 294
Slovenia 21, 97, 102
Smith, Adam 83
social democracy 86, 126, 223, 233, 242, 251, 259–66, 277–8,
283–4, 286
social democratic movements 99, 177, 183, 216–17, 258–9, 439
1201
Social Democratic Party of Germany (SPD) 165, 189, 190, 201,
203, 224, 227, 234, 242, 245, 257, 259, 267, 273, 274–5,
278, 492
abdication of the emperor 330–1
ban of USPD 331
‘battleship A’ 423, 424–5
Bavaria 370
and ‘cabinet of barons’ 457
capital gains tax 293
and Communists 418, 420, 427–8
Constituent National Assembly elections 352–3
Council of People’s Commissars 336, 340–7
Cuno administration 390–1
factionalism 309–10
First World War 297, 299–301, 303–4, 307, 309–13, 317, 320,
325–6
German revolution 331–6
Grand Coalition 424, 427, 428, 433, 435
and Hindenburg 447–8, 449
and Hitler 485–6, 487–8
imperial citizenship 295
increased representation 284–5
Social Democratic Party of Germany (Cont.)
Jewish intellectuals 289
Morocco crisis 280–1
1202
motion of no confidence in Stresemann 397–8
MSPD/USPD reunification 386
non-participation policy 404, 405
opposition to 284
parliamentary rule 271
Preussenschlag 457–8
Reichstag elections 372, 403, 404, 406, 424, 437, 459, 468
Schleicher talks 472–3
socialism and 439
Stresemann’s foreign policy and 419–20
toleration policy with Brüning administration 440–3, 445, 454,
458
Tucholsky’s mockery of 416
Upper Silesia 377
Versailles Treaty 359
war reparations 374
warning on militarism 292
Weimar Republic 354, 362–4, 484
withdrawal from Grand Coalition 395–6, see also MSPD; USPD
Social Democratic Workers’ Party 177, 187, 197–8
socialism 81, 85–6, 100, 125, 141, 177, 261–6, 402–3
capitalism or 414
First World War 303–4, 309–13, 310–11, 319–21, 359–60
Hitler’s version of 439
Jews and 207
1203
legislation against 216–17, 233, 242–3
revolutionary 125–8
women and 272
Socialist Workers’ Party of Germany 443
Solf, Wilhelm Heinrich 333
Sollmann, Wilhelm 442–3
Sombart, Werner 240, 304
Somme, battle of the (1916) 308
Sonnemann, Leopold 154
Sonnenberg, Liebermann von 254
Souchon, Admiral 330
Soviet Union 402
agitation for revolution in Germany 391–2, 394
Franco-German confrontation in 1923 390
Jews 384
KPD 469
power struggle 427
Rapallo Treaty 378–80, 419
Treaty of Berlin 418–19, see also Russia
Spahn, Peter 285, 315
Spain 18, 21, 181–2, 247
Spartacus League 310, 320, 335–6, 340, 345, 348
SPD see Social Democratic Party of Germany
Spengler, Oswald 414
Sperr, Franz 481
1204
Spitzemberg, Baron 179
Springorum, Fritz 470, 478
SS (Security Force) 449, 452, 455, 457
Stadelmann, Rudolf 40
Stahl, Julius 81, 119–20
Stahlhelm (League of Frontline Soldiers) 445, 447, 449, 457,
486
Stalin, Joseph 322, 391, 402, 423, 427, 450
Stapel, Wilhelm 466
Stauffenberg, Franz Schenk von 214, 224
Stavenhagen, Friedrich von 138
Stegerwald, Adam 317
Stein, Baron Karl 49–50, 51, 62
Stein, Lorenz von 47, 55, 71, 125–8, 162, 218, 225
Steinacker, Karl 79–80
Stewart-Wortley, Sir Edward James 269
Stinnes, Hugo 343, 386–7, 393
stock market crashes 204, 431
Stoecker, Adolf 254
Stosch, General von 245
Strasbourg 32, 229
Strasser, Gregor 423, 473–4, 476–7, 480
Strasser, Otto 423
Strauss, David Friedrich 198–9
Stresemann, Gustav 277, 313, 351, 368, 386, 387, 391, 393, 395,
1205
398–9, 409
Grand Coalition 424
League of Nations 418, 419
sudden death of 428–9
Stroecker, Adolf 208–9, 210
Strousberg, Bethel Henry 204
Struensee, Karl Gustav von 39, 52
Struve, Gustav von 88, 94, 99
students 67, 417, 420–1
anti-Semitism 209, 383–4, 421, 430
associations 66, 420–1
Austrian 91
Berlin 92
Jewish 286
National Socialists 430, see also universities
Stüve, Carl 91
Styria 64
submarine warfare 308–9, 313
Südekum, Albert 299, 368
suffrage 90, 91, 104, 109, 119, 137, 141, 156, 172, 253, 255,
259, 260, 326, 338
Engels on 274
female 332, 342, 353
Prussian 271–2, 278, 285, 312, 313, 331, 332
Swabia 158
1206
swastika symbol 287
Sweden 19, 20–1, 44
Switzerland 12, 14, 20, 42, 60, 89, 218
Sybel, Heinrich von 31, 87, 132, 138, 143, 191
Tacitus 11
tax boycotts 104–5
taxation 215–16, 241, 270, 293, 365, 399, 425, 433, 436–7, 441
Temme, Jodocus 102
Teplitz Accord (1819) 66, 67
Teutonic Order 22
Textor, Johann Wolfgang 30
Thälmann, Ernst 408, 409, 410, 423, 428, 438, 447, 448
Thieme, Clemens 290, 291
Thirty Years War (1618–48) 19–21, 26
Thoma, Ludwig 376
Three Emperors’ League (1881) 230
Three Kings’ Alliance (1849) 112
Thuringia 394, 396, 412, 430, 439, 480
Thyssen, August 307
Thyssen, Fritz 470
Tiedemann, Christoph 228
Tillessen, Heinrich 376
Tilsit, Treaty of (1807) 49
Tirpitz, Rear Admiral Alfred von 244–6, 247, 256, 294, 316,
1207
404, 409
Tocqueville, Alexis de 29
Togo and Cameroon 227
Toller, Ernst 356
Torgler, Ernst 464
trade 121, 144–5, 178, 214–15, 218–22, 225, 231, 240, 244, see
also free trade
trade unions 258, 264, 274–5, 300, 311, 322, 343, 390, 399, 472,
482, 484, 485
Kapp putsch 367–9
protests over Rathenau’s murder 381
Ruhr uprising 370–1
Transylvania 21
Treitschke, Heinrich von 76, 174, 184, 209–10, 213, 222, 252,
283, 305
Treviranus, Gottfried 429
trial by jury 90, 120
Trias 153, 156, 160
Trieste 64, 106
Trimborn, Karl 327
Triple Alliance (1902) 248
Tripoli 7, 281–2
Troeltsch, Ernst 13–14, 307, 317, 325, 329, 336
Trotsky, Leon 322, 391–2
Tucholsky, Kurt 411–12, 416
1208
Tunis 7
Turkey 281, 298
Turks 15, 21, 68–9, 78, 293
Twesten, Karl 138, 140, 147, 152, 153, 156–7, 166, 168, 169,
170, 171, 173, 277
Tyrol 44, 64, 97, 106
1209
Dawes Plan 402
First World War 309, 319, 324
Germany’s war reparations 398
Hoover Moratorium 444
protectionism 221
stock market crash 431
universities 10, 238, 243, 286
anti-Semitism in 209, 383–4
Karlsbad Decrees 66–7
Pietism 26
professors 16, 75–6, 77, 307, 417
Prussia 50
shift to the right 430, see also students
Unruh, Hans Viktor von 166, 169, 171
USPD (Independent German Social Democratic Party) 311,
320, 322–3, 331, 332, 333–4, 334–6, 340–1, 347, 352, 377
Berlin uprisings 348–9
political realignment 385–6
Rapallo Treaty 380
Reichstag elections 372, 403
war reparations 374
Weimar constitution 364
Utrecht, Peace of (1713) 22
1210
Venetia 159, 180
Venice 44
Verdun, battle of (1916) 308
Verdun, treaty of (843) 5
Versailles, Treaty of (1919) 357–61, 373, 491, 493
Victoria, Queen 269
Vienna:
Congress of (1815) 59, 63, 64, 71, 134
rebellions (1848) 91–2, 102
sieges of 21
stock market crash (1873) 204
ultimatum 108–9, see also Austria
Villafranca, Treaty of (1859) 134
Vincke, Georg von 105, 132
Virchow, Rudolf 169, 181, 200
Vogt, Karl 109–10
Volk 52, 56, 63, 251, 280, 282–3, 384, 414, 493
Vollmar, Georg von 264
Voltaire 33, 69
1211
Wallachia 121
Wallot, Paul 251, 252
Wandervogel 288
war criminals 373, 374
war guilt 360, 405, 420
War of the Spanish Succession (1700) 21
war reparations 358, 373–4, 379, 387, 388, 389, 391, 397–8,
402, 425–6, 444, 446, 455
Warmbold, Economics Minister 479
Warrburg, Festival of (1817) 66
warrior associations (Kriegervereine) 249–50
Webb, Sidney and Beatrice 261
Weber, Alfred 307, 351
Weber, Max 307, 317, 375, 489
on Calvinism and capitalism 22
on Germany after the First World War 337–8
on the July crisis (1917) 315
on Junkerdom 239, 240, 246, 283
on parliament 255–6
recommendations to Weimar Republic 362
sociology 412
Wedel, Count 295
Wehler, Hans-Ulrich 51, 228
Wehrenpfennig, Wilhelm 222
Weill, Kurt 416
1212
Weimar Republic (1919–33): accession to the League of Nations
418, 419
agricultural policy 421
anti-republican legislation 381–2
anti-Semitism in 415–16
Brüning administration 435–54
civil state of emergency 395–401
‘conservative revolution’ 413–18
constitution 354, 361–5, 381–2, 386, 413, 463, 481, 484, 489
currency reform 397
economic crisis 443–6
Erfüllungspolitik 374, 376, 380, 383
founding of 354
Germans living outside 492–3
Grand Coalition 390–3, 396, 424–35
Hitler as chancellor 487–90
Kapp putsch 366–70
myths created by war defeat 360–1, 383
Osthilfe scandal 482
Papen administration 454–75
Rapallo Treaty 378–80, 419
referendum against the Young plan 429–30
Ruhr uprising 370–1
Saxony crisis 394–5
Schleicher administration 475–87
1213
social policy 421–2, 428, 434–5, 436, 442, 455
treaty with Soviet Union 419
Upper Silesia 373, 377–8
Versailles Treaty terms 357–8, 373
war reparations 373–4, 379, 387, 388, 389, 391, 397–8, 402,
425–6, 444, 446, 455, see also economy
Weis, Eberhard 46
Weitling, Wilhelm 86
Welcker, Carl Theodore 72, 88, 109
Wels, Otto 332, 347, 367, 468, 487
Weltpolitik 246–8, 268–9, 282–3
West Pomerania 64
Westarp, Count 422, 429, 432, 437
Westphalia, Kingdom of 47, 51, 64
Westphalia, Peace of (1648) 20–1, 27, 31, 45
Wichern, Johann Hinrich 197–8
Wiedfeldt, Otto 396
Wieland, Christoph Martin 31, 37–8, 41
Wilamowitz-Moellendorff, Ulrich von 307
William, Crown Prince 288
William I, King of Prussia 130, 131, 136–7, 138, 139, 150, 159,
168, 176
assassination attempts 216, 238
death 232
as Emperor 191
1214
Franco-Prussian war 182
German Reich 189, 190
monument dedicated to 250
William II, Emperor of German Reich 242
abdication of 329, 330–1, 333, 337
anti-Semitism 287, 288
appointment of von Bülow 266–7
during the Balkans crisis 294
and Bismarck 233–5
Daily Telegraph affair 269–70
First World War 297–8
jubilee celebrations 291
Kyffhäuser monument 250
landing in Morocco 268
Naumann’s hopes for 255–6
new Reichstag 251–2, 253
William of Ockham 9
Wilmanns, Carl 207
Wilson, President Woodrow 324, 326, 329, 337, 338
Windischgrätz, Prince Alfred 102
Windthorst, Ludwig 176, 234, 257
Wirsing, Giselher 492
Wirth, Johann Georg August 73, 74
Wirth, Joseph 374, 376, 377–9, 380, 386, 387, 388, 445
Wissell, Rudolf 425, 434
1215
Wolff, Theodor 307, 336, 351
women’s rights 285
women’s suffrage 272
Workers’, Farmers’, and Soldiers’ Councils 363
Workers’ Brotherhood 86
working class see proletariat
Wrangel, General Friedrich von 96, 103, 166
Wulle, Reinhold 384, 385
Württemberg 44, 46, 47, 51, 110, 112
founding of the Reich 188, 190
German Confederation 65
German democratic movement 153–4
March demands 91
moderate liberals 161
North German Confederation 177
particularism 182
political allegiance 155
Prussian defeat of Austria 158
Wydenbrugk, Oskar von 146
Wyneken, Gustav 292
1216
Young Germany (secret society) 75
Young Hegelians 82, 84
Young Plan (1929) 425–6, 429, 432–4, 446
1217
* Ranke, History of the Romance and Germanic Peoples.
1218
* Anon., The Plays of the Antichrist.
1219
* Brant, The Ship of Fools.
1220
† that all of earth its vassal be, | as it should be by right and law.
1221
† Luther, To the Christian Nobility of the German Nation Concerning
the Reform of the Christian Estate.
1222
* Hegel, Lectures on the Philosophy of History.
1223
† Marx, Critique of the Hegelian Philosophy of Right. Introduction.
1224
† Nietzsche, The Antichrist.
1225
* Luther, On the Jews and Their Lies.
1226
* Ranke, German History in the Age of the Reformation.
1227
* Droysen, History of Prussian Politics.
1228
† Hintze, Calvinism and Reasons of State in Brandenburg at the
Beginning of the Seventeenth Century.
1229
* Müller-Armack, Genealogy of Economic Styles.
1230
* Tocqueville, The Old Regime and the Revolution.
1231
* [Translator’s note: fritzisch gesinnt, ‘Fritz-minded’.]
1232
† Goethe, Poetry and Truth.
1233
† Abbt, On Death for the Fatherland.
1234
* Moster, On the German National Spirit.
1235
* Lessing, Hamburg Dramaturgy.
1236
* To make yourself a nation—for this you hope, | Germans, in vain;
| make yourselves instead— you can do it! | into men the more
free.
1237
† Herder, Ideas on the Philosophy of Human History.
1238
* [Translator’s note: literally, ‘saddle-era’, from the secondary
meaning of German Sattel ‘bridge’, ‘mountain pass’.]
1239
* Kant, Perpetual Peace.
1240
† Kant, The Conflict of the Faculties.
1241
† Kant, Doctrine of Right of Metaphysics of Morality.
1242
* Montesquieu, Sprit of the Laws.
1243
* Schiller, German Greatness.
1244
† Hegel, The Constitution of Germany.
1245
* Stein, History of the Social Movement in France.
1246
* Anon., Germany in its Deepest Humiliation.
1247
* Hegel, Philosophy of Right.
1248
* Fichte, The Closed Commercial State.
1249
† Fichte, Characteristics of the Present Age.
1250
† Fichte, Addresses to the German Nation.
1251
* Fichte, Outline of a Political Writing in the Spring of 1813.
1252
* Hegel, Phenomenology of Spirit.
1253
† Jahn, ‘On the Promotion of Patriotism in the Prussian Empire’.
1254
* Jahn, German Folk Culture.
1255
* Arndt, ‘Prayer’.
1256
† Arndt, ‘Catechism for the German Soldier and Militiaman’.
1257
* Arndt, ‘To the Prussians’.
1258
† Arndt, ‘What do Landsturm and Landwehr mean?’.
1259
† Arndt, ‘On Ethnic Hatred’.
1260
¶ Arndt, ‘The Rhine, Germany’s River, but not Germany’s Border’.
1261
* Humboldt, ‘Memorandum on the German Constitution’.
1262
* Pfizer, Correspondence between Two Germans.
1263
* Treitschke, German History in the Nineteenth Century
1264
* They shall not have it, | the free German Rhine, | though like
greedy ravens, | they caw themselves hoarse for it.
1265
† Becker, ‘Song of the Rhine’.
1266
† Schneckenburger, ‘The Guard on the Rhine’.
1267
§ A call goes out like a thunder-echo, | like clash of swords and
crash of waves, | ‘To the Rhine, to the Rhine, the German Rhine!’ |
Who will the stream’s protector be?
1268
¶ Fallersleben, ‘Song of Germany’.
1269
* Meyer, ‘The Origins of Political Radicalism in Pre-March Prussia’.
1270
† Jacoby, Four Questions, Answered by an East Prussian.
1271
* Ruge, ‘Liberalism: A Self-Critique’.
1272
* Sybel, The Political Parties of the Rhenish Province in their Relation
to the Prussian Constitution.
1273
* [Translator’s note: the word Mittelstand, though often simply
translated as ‘middle class(es)’, is an economic term meaning small
and medium-sized tradesmen and business owners.]
1274
* [Translator’s note: the Neman is the Memel in German.]
1275
* [Translator’s note: the German National Assembly/Frankfurt
parliament is often called ‘the Paulskirche’.]
1276
* [Translator’s note: ‘the canister prince’ (he ordered canister shot
to be fired on the demonstrators during the uprising).]
1277
* Rochau, Principles of Realpolitik. Applied to the Political Conditions
in Germany.
1278
* Marx, Class Struggles in France.
1279
* Marx, The Eighteenth Brumaire of Louis Bonaparte.
1280
* Jacoby, ‘Election Appeal for the Prussian Democrats’.
1281
* Lassalle, The Italian War and the Mission of Prussia: A Democratic
Voice.
1282
† Engels, Po and Rhine.
1283
* Bismarck, Thoughts and Recollections.
1284
* ‘Luck and Power’.
1285
* Büchner, Force and Matter.
1286
* Rapp, The National Question in Württemberg.
1287
* Ketteler, Germany after the War of 1866.
1288
* [Translator’s note: Rundschauer: ‘one who looks around’,
‘observer’. The title of Gerlach’s column in the Kreuz-Zeitung was
the ‘Rundshau’.]
1289
* Baumgarten, ‘German Liberalism: A Self-Critique’.
1290
* Rittershaus, ‘Against Bonaparte’.
1291
† A united Germany! Ah, how long desired, | how often besought
in our dreams’ gloaming! – | And it is here! Now must the Frankish
sword | hammer us together with one blow! | But ever since the
Mother was dealt disgrace, | no border on our map we recognize; |
we only know one single cry of rage | and battle to the finish, to
the death! | One war cry only: Down with Bonaparte!
1292
* God has judged! Ours is the victory! | Full of laurels bloom the
graves of our dead. | He who hounded us into this holy war | Now
lies, outcast, before us on the ground. — | Four weeks in all; not
just Germany’s, they are | The greatest weeks in all of world
history!
1293
* [Translator’s note: ‘[An] empire of parity!’, i.e. representing
Evangelicals and Catholics equally.]
1294
* ‘Dynasties and Clans’.
1295
* Marx, ‘The Civil War in France’; ‘Address of the General Council
of the International Workers’ Association’.
1296
* Wagner, ‘Judaism in Music’.
1297
† Wagner, ‘What is German?’
1298
* Marr, The Victory of Jewry over Germandom: A Non-confessional
Point of View.
1299
* Rohling, The Talmud Jew.
1300
* Heine, History of Religion and Philosophy in Germany.
1301
* Dühring, The Jewish Question as a Racial, Moral, and Cultural
Qusetion.
1302
* Newspaer article, ‘Is War in Sight?’.
1303
* Nietzche, Untimely Meditations.
1304
* Weber ‘The Nation State and National Economic Policy’.
1305
* [Translator’s note: a Generalpächter leased fallow land from
landowners and rented it out at higher prices.]
1306
* Nipperdey, National Idea and National Monument in Germany in the
Nineteenth Century.
1307
* ‘Hail to Thee in Victor’s Laurels’.
1308
* Naumann, Democracy and Empire.
1309
* Bernstein, ‘Problems of Socialism’.
1310
† Bernstein, The Preconditions of Socialism and the Tasks of Social
Democracy.
1311
* Luxemburg, ‘Social Reform or Revolution’.
1312
* Bebel, Woman under Socialism.
1313
† Kautsky, Bernstein and the Social-Democratic Programme.
1314
* Bachem, ‘We Must Come Down from Our Tower’.
1315
† [Translator’s note: ‘splendid isolation’ is in English in the
original.]
1316
* Mommsen, ‘What Can Still Save Us’.
1317
* Let bright the sword flash | From your stellar citadel; | Hack the
whole miserable hash | asunder in one blow fell!
1318
* Class, If I Were the Emperor.
1319
† Class, The Emperor Book.
1320
* Lagarde, German Writings.
1321
† Langbehn, Rembrandt as Educator.
1322
† Chamberlain, The Foundations of the Nineteenth Century.
1323
* [Translator’s note: the word means ‘migratory bird.’]
1324
* Rathenau, 1813: A Contribution to the Centenary Celebrations.
1325
† Let’s away! When the second | of the two suns awakes, | it will
shine on the battle, | the fight of the glorious. || And if death
crouches | and waits in trenches, | will flash freedom and life | out
of fiery red.
1326
* Lehmann, The German God.
1327
* ‘Appeal to the Cultured World’.
1328
* Max Scheler, The Genius of War and the German War.
1329
* Sombart, Patriotic Meditations.
1330
† Mann, Reflections of a Non-political Man.
1331
* Mann, ‘The Sufferings and Greatness of Richard Wagner’.
1332
* [Translator’s note: Courland part of present-day Latvia, was a
separate province of the Russian empire from 1797 to 1918. It was
bounded by the Baltic Sea to the west, Lithuania to the south, and
the Drina River (now the Daugava) to the north.]
1333
* Rosenberg, The Birth of the German Republic, 1871–1918.
1334
* Ludendorff, My War Memories.
1335
* ‘Strike them dead! On Judgement Day | your reasons won’t be
questioned.’
1336
* [Translator’s note: the workers’ and soldiers’ councils are also
frequently referred to as ‘soviets’.]
1337
* Bernstein, The German Revolution: Its Origins, Its Course, and Its
Achievements.
1338
* [Translator note: Räterepublik is also frequently translated as
1339
* ‘Swastika on helmet, band of black-red-gold, | the Erhardt
1340
* Hitler, My Struggle.
1341
* Heidegger, Being and Time.
1342
† Schmitt, The Crisis of Parliamentary Democracy.
1343
† Schmitt, Constitutional Doctrine.
1344
* Spengler, The Decline of the West.
1345
† Spengler, Prussianism and Socialism.
1346
*Rise or Decline?
1347
* Schmitt, Legality and Legitimacy.
1348
* Jünger, The Worker: Mastery and Form.
1349
目录
Title Page 3
Copyright Page 4
Preface 7
Contents 8
Abbreviations 9
Introduction 12
1. Legacy of a Millennium 17
2. Hampered by Progress 1789–1830 79
3. Liberalism in Crisis 1830–1850 148
4. Unity before Liberty 1850–1871 236
5. The Transformation of Nationalism 1871–1890 375
6. World Policy and World War 1890–1918 463
7. The Impaired Republic 1918–1933 650
Notes 944
Index 1124
Footnotes 1218
1350