- GOVERNMENTAL INSTABILITY IN WEIMAR GERMANY, '
1919-1931 ' ‘

Dissertation for the Degree mm D
MICHIGAN STATE UNIVERSITY
LON S. FELKER
1975

'E'VW‘83NL I-I _4.‘. ..-.L.. - '

lhfi
Irv-r965 on y
I ‘21.. £4 AK i‘\1

       
   
   

Michigan Sta“
(Johnnshy

This is to certify that the
thesis entitled
Governmental Instability in
Woimar Germany, 1919-1931

presented by

Len S. Folkor

‘ has been accepted towards fulfillment
of the requirements for

Ph D degreein Political Science

 

 

I

Major professor

Date June 3, 1975

 

0-7639

 

 

 

 

ABSTRACT

.3y\ GOVERNMENTAL INSTABILITY IN WEIMAR

GERMANY, 1919-1931

By
Lon S. Felker

Governmental instability in Weimar Germany--measured

as the duration of the government--is studied in relation

to several variables: (1) ideological diversity in the

governing coalition; (2) post maldistribution; (3) the

number of parties in the parliament (Reichstag); (4) the

 

difference in the size of the coalition government from

minimum-winning size; and (5) radical party growth/decline.

The following hypotheses are tested:

H.l.

H.6.

That governmental duration declines with
increases in radical party electoral
support.

Governmental duration is negatively
related to the ideological diversity
in the governmental coalition.

That governmental duration is inversely
related to post maldistribution.

Increases in the total number of parties
represented in the national parliament
are negatively related to governmental
duration.

As a coalition government's size deviates
from minimum-winning size, its duration
declines.

That governmental instability is partially
produced by a combination of some or all
of the five variables.

Lon S. Felker

The data for this study were drawn from the coalition
governments of Germany from 1919 to 1931. The seventeen
governments of this period can be classified into four coali-
tion types: (1) the Weimar Coalition, consisting of the
Social Democratic Party; the Catholic Zentrum party, and
the German Democratic Party; (2) the Grand Coalition, made
up of the Social Democrats, Zentrum, Democrats, and German
Peoples Party; (3) the Bourgeois Coalition, which was the
Grand Coalition minus the Social Democrats, and (4) the
Right Coalition, consisting the Zentrum, the Bavarian
Peoples Party, the German Peoples Party, and the German
Nationalists. Five of the governments during the 1919-1931
period were of the Weimar type. Three Grand Coalitions arose,
while seven Bourgeois and only two Right coalitions were
formed.

Using correlation and regression to analyze the data,

a pattern of relationships emerged. Ideological diversity

in the government was found to have the strongest correlation
with duration. The number of parties variable also had a
strong relation with duration which was not dispelled by

systematic tests for spuriousness.

GOVERNMENTAL INSTABILITY IN WEIMAR

GERMANY, 1919—1931

BY

Lon S? Felker

A DISSERTATION

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

DOCTOR OF PHILOSOPHY

Department of Political Science

1975

ACKNOWLEDGMENTS

I would like to thank Professor Frank Pinner for the
many hours in excess of his teaching load allowance that
he spent in helping me with the present work. The other
Dissertation Committee members--Professors William Baugh
and Joseph Schlesinger--made valuable editorial comments.
The German Academic Exchange Service (DAAD) helped fund
the research for this study, and the Courtesy and assis-
tance of their staff in Bonn greatly aided me during my
stay in the Federal Republic. The staff of the Federal
Archive in Koblenz greatly facilitated my collection of
the documentary material used in this study. Professor
Karl Dietrich Bracher of the University of Bonn made many
useful suggestions as to where Weimar-related materials
could be found.

Finally, I would like to thank Professor Lois M.
Pelekoudas for her many useful editorial criticisms.

The responsibility for the finished product is, of course,

my own.

ii

TABLE OF CONTENTS
Page
LIST OF TABLES . . . . . . . . . . . . . . . . . . . vi

LIST OF FIGURES . . . . . . . . . . . . . . . . . . Vii

Chapter
1. GOVERNMENTAL STABILITY IN NEIMAR GERMANY:
AN INTRODUCTION . . . . . . . . . . . . . . 1
INTRODUCTION . . . . . . . . . . . . . . . 1
THEORIES AND HYPOTHESES . . . . . . . . . . 3
THE MEASURES . . . . . . . . . . . . . . . 22
SUMMARY . . . . . . . . . . . . . . . . . . 2?
NOTES . . . . . . . . . . . . . . . . . . . 28

2. THE SETTING OF THE STUDY: GOVERNMENTAL
STABILITY AND THE WEIMAR PARTY SYSTEM,

1919-1931 . . . . . . . . . . . . . . . . . 32
INTRODUCTION . . . . . . . . . . . . . . . 32
THE PARTY SYSTEM, 1919-1931 . . . . . . . . 35

The Confessional—Nonradical Parties:
Social Democrats and Zentrum . . . . . 36

Bourgeois—Protestant Parties: DDP, DVP,

emdDmm . ... .. ... ... .. . 42
Radical Parties: The KPD and NSDAP . . . M7
COALITION PATTERNS . . . . . . . . . . . . 58

The Weimar Coalition: SPD, DDP, and Zentrum 61

The Bourgeois Center Coalition: DDP,
Zentrum, and DVP . . . . . . . . . . . 63

iii

Chapter Page

The Grand Coalition: SPD, DDP, Zentrum,

and DVP . . . . . . . . . . . . . . . . . 64

The Coalition of the Right: Zentrum,
DVP, and DNVP . . . . . . . . . . . . . . 65
A MODEL OF THE CABINET FORMATION PROCESS . . 67
SUMMARY . . . . . . . . . . . . . . . . . . 69
NOTES . . . . . . . . . . . . . . . . . . . 7O

3. DEVELOPING A MODEL OF GOVERNMENTAL STABILITY
FOR WEIMAR GERMANY . . . . . . . . . . . . . 72

INTRODUCTION . . . . . . . . . . . . . . . . 72
DATA ANALYSIS . . . . . . . . . . . . . . . . 74

CORRELATIONAL ANALYSIS . . . . . . . . . . . 75

RECHECK OF THE MODEL . . . . . . . . . . . . 83
THE AMOUNT OF VARIANCE IN DURATION EXPLAINED

BY THE MODEL . . . . . . . . . . . . . . . 86
CONCLUSIONS . . . . . . . . . . . . . . . . . 87
NOTES . . . . . . . . . . . . . . . . . . . . 91

4. THE ROLE OF IDEOLOGICAIIDIVERSITY IN NATIONAL
CABINET FORMATION: 1919-1931 . . . . . . . . 92

INTRODUCTION . . . . . . . . . . . . . . . . 92
THE WEIMAR COALITION . . . . . . . . . . . . 93
THE BOURGEOIS COALITIONS . . . . . . . . . . 103
THE GRAND COALITIONS . . . . . . . . . . . . 119
THE RIGHT COALITIONS . . . . . . . . . . . . 134

SUMMARY . . . . . . . . . . . . . . . . . . . 142

NOTES . . . . . . . . . . . . . . . . . . . . 144
5. SUMMARY AND CONCLUSIONS . . . . . . . . . . . . 148
SUMMARY . . . . . . . . . . . . . . . . . . . 148

iv

Chapter Page
HYPOTHESES . . . . . . . . . . . . . . . . . 151
CONCLUSIONS . . . . . . . . . . . . . . . . 154
NOTES . . . . . . . . . . . . . . . . . . . 159

BIBLIOGRAPHY . . . . . . . . . . . . . . . . . . . 160

APPENDIX.....................l66

Table

2.

CIDNONU'l-P'

LIST OF TABLES

Simple Correlations among Six Variables . . .

Partial Correlations of Duration with Five Varia-
bles O O O O O O O O O O O O I O O O O O O 0

Equations and Path Coefficients for the Causal
Model in Figure 4 . . . . . . . . . . . . . .

Regression Analysis of Duration . . . . . . . .
Bourgeois Coalitions . . . . . . . . . . . . . .
First, Second, and Third Reichstag Elections . .
The Stresemann Cabinets . . . . . . . . . . .

The Right Coalitions . . . . . . . . . . . .

vi

Page
76

77

85
86
110
114

124

135

LIST OF FIGURES
Figure

l. The Coalition Alternatives within the Weimar
Party System: 1919-1932 . . . . . . .

2. Various Logical Causal Paths between the Six
Variables, Including Correlation Coefficients

3. Results of the Tests of Alternative Models of
Governmental Destabilization . . . . . . . .

4. Alternative Models of Governmental
Destabilization with Path Coefficients

vii

Page

60

79

82

84

Chapter I

GOVERNMENTAL STABILITY IN WEIMAR GERMANY:
AN INTRODUCTION

Introduction

 

The sources of governmental stability and instability
have long been sought by scholars. Plato discussed the

forces underlying stability in The Republic, and Aristotle,

 

in his Politics, was the first to employ empirical methods

in the study of stable and unstable regimes. Later theorists
of the natural law and social contract schools sought to
establish the basis of a stable political order.

In this study, our prime objective is the analysis of
stability in a limited number of governments taken from a
particular era and nation—state: Germany during the period
1919-1931. The Weimar Republic provides an interesting
setting for the study of governmental instability because
it contained almost every conceivable influence which might
contribute to instability: military defeat; demands by the
Allied powers for economic reparations; inflation; extreme
multipartyism as a consequence of a proportional system of
parliamentary representation imposed on a pluralistic society;
personality differences among the leaders of the major poli-
tical parties; and the development of radical parties at both

I

2
extremes of the party system. While this is not an exhaus—
tive list, it is suggestive of the multiplicity of forces
affecting any German cabinet during the period.

The question of how various episodic events affected
governmental duration during this period is an interesting
one. However, the primary focus of our study is the rela-
tionship between governmental stability and various poli—
tical variables, such as ideological diversity within the
government, conflict over rewards, coalition size diffe—
rences, and the growth or decline of radical parties. By
political variables then, we refer to those events related
to the competition for and exchange of power in a multi—
party parliamentary system, as well as those events in—
timately connected with such competition and exchange.
While such events as foreign military occupation, infla—
tion, and internal revolt are episodic, political events
in a parliamentary democracy in normal times are regular,
and hence predictable. This regularity facilitates the
measurement and analysis of such phenomena.

Our purpose in studying Weimar Germany is simply to

examine a set of political explanations of governmental

 

instability in Weimar Germany. In so doing, we must note
that these governments were often influenced not only by
domestic political interests, but by foreign ones as well.
We treat foreign and economic variables in a separate

chapter (Chapter IV), but we link them to the political

3

variables identified in our data analysis chapter (Chap—
ter III).

Prior to describing the methodological procedures of
our study, we review the various theories of governmental
stability/instability which have been offered. These in-
clude not only political theories, but historical, cultural,
economic, sociological, and formal-legal. Following a dis-
cussion of these theories, the hypotheses of the study are
presented. In the final section of this chapter, the mea-
sures of the variables employed and a description of the

statistical techniques used in this study are presented.

Theories and Hypotheses

 

Theories of governmental stability abound in the nor-
mative literature of classical political philosophy, and
when one turns to contemporary theory-oriented research,
the literature is equally rich and varied.

A considerable portion of the research into govern-
mental stability has centered on France (especially on the
Fourth Republic) and on post-war Italy.l Recently, the
coalition behavior school has made substantial contribu-
tions to the literature.2 However, before discussing
various political explanations of stability, we should note

the contributions made in other disciplines to the study

of this phenomenon.
Many explanations have been advanced as to why certain

parliamentary democracies are less stable than others. These

3

variables identified in our data analysis chapter (Chap-
ter III).

Prior to describing the methodological procedures of
our study, we review the various theories of governmental
stability/instability which have been offered. These in-
clude not only political theories, but historical, cultural,
economic, sociological, and formal—legal. Following a dis-
cussion of these theories, the hypotheses of the study are
presented. In the final section of this chapter, the mea-
sures of the variables employed and a description of the

statistical techniques used in this study are presented.

Theories and Hypotheses

 

Theories of governmental stability abound in the nor—
mative literature of classical political philosophy, and
when one turns to contemporary theory-oriented research,
the literature is equally rich and varied.

A considerable portion of the research into govern-
mental stability has centered on France (especially on the
Fourth Republic) and on post-war Italy.l Recently, the
coalition behavior school has made substantial contribu-
tions to the literature.2 However, before discussing
various political explanations of stability, we should note

the contributions made in other disciplines to the study

of this phenomenon.
Many explanations have been advanced as to why certain

parliamentary democracies are less stable than others. These

II,
explanations may be grouped under the following general
categories: (1) historical; (2) economic; (3) political,
g,g., parties and party systems; (4) constitutional
(formal-legal); (5) cultural; and (6) sociological.3

The historical explanation is generally couched in
terms of a nation’s institutional development. Political
stability is thus the residual of past experiences in coping
with political crises. The historical explanation can not
be faulted on its logic, but it can be attacked on grounds
of its relevance: as each nation‘s historical development
differs, no generalizable theory can be derived. Nor is
there any technique for the determination of the importance
of one set of events over another.4

The economic explanation is one of the most controver-
sial. Lipset and Cutright exemplify this school. Both
attempted to show that there exists a correlation between
economic conditions and political stability. Cutright
demonstrated, in his seventy-seven nation study, a clear
relationship between economic and social variables and
political development. However, critics of Cutright’s
methodology note that his scale of political development
is good only for certain levels of political development,
and not for others.5 Lipset examined the relationships
between internal stability and national wealth, industria-
lization, education, and urbanization. While certain of

Lipset’s measures for this concept correlated highly with

5
political stability, Cutright criticized the study for failing
to demonstrate the association between economic development
and democracy. This, Cutright asserted, was largely a failure
to distinguish among the democratic features of the countries
in Lipset‘s sample.6

Perhaps the soundest conclusion that one can reach about
the economic explanation is that it is a limited one. Once
the relation between economic development and political sta—
bility is demonstrated, then further variables must be isolated
in order to explain variations in stability at roughly identi—
cal levels of economic development. Therefore, an economic
explanation is useful in accounting for stability across
broad categories of nation-states (developed, underdeveloped,
etc.), but is less useful in explaining differences within
those categories.

Political explanations based on the party systems and
types of parties are encountered both in contemporary litera-
ture and in that of the late nineteenth century. Typical of
this school are the writings of A. Lawrence Lowell, who
advanced the theory that a national legislature must contain
"two parties, and two parties only . . . in order that the
parliamentary form of government should permanently produce
good results." His major contention was that coalition
ministries were short-lived compared to homogeneous ones:
the greater the number of parties in a government, the

greater the dissension within it. A corollary of Lowell‘s

6

theory was that the opposition should consist of only one
party, in order that it may, when called upon by the elec-
torate, give the country a united government.7

A more contemporary example of this school is found in
the writings of Giovanni Sartori, who describes what he
terms a pattern of "polarized pluralism” in multiparty
systems, such as those of Fourth Republic France, contem-
porary Italy, and Weimar Germany. The increase in the
number of parties beyond a certain limit, Sartori states,
leads to increase in ideological conflict. Nor will econo-
mic affluence lessen ideological tensions, for while inten—
sity may decrease, the ideological approach remains the same:

The temperature of ideology may cool, but
this . . . does not imply that a society

will lose the habit of perceiving politi—

cal problems in an unrealistic and dogma-

tic fashion; and it implies even less that
the party system will turn to a pragmatic
approach. Even less, because party fragmen-
tation is not merely a reflection of the
ideological cleavages in a society. The
other side of the coin is that a fragmented
party system invests a great deal of energy 8
in the effort to disintegrate basic consensus.

One of the distinctive features of a polarized multiparty
system, Sartori claims, is that parties-~more concerned about
securing a stable electorate of ideologically loyal voters--
cease to perform the brokerage function. Moreover, the sys—
tem encourages irresponsible opposition-—irresponsible, for
in no sense are the radical parties in the opposition a real

alternative government: they constitute alternative party

systems in themselves. In fact, they desire the end of

7

parliamentary rule and the establishment of a single-party
state.

The major theme of Sartori‘s theory is that the increase
in the number of parties leads to ideological diversity of
such a level that stability (although Sartori does not

express it as such) is jeopardized and ideological immobilisme

 

becomes the order of the day. His theory is a fine example

of the political explanation of governmental instability.
Somewhat linked to the political approach is the consti-

tutional (formal-legal) approach. Older than the political

approach discussed above, this approach seeks to explain

instability by the legal system and the constitutionally

established institutions of society. While this approach

has been criticized for ignoring the fact that stability

more often affects the durability of legal systems than vice

versa, a variant of this approach persists.9 This variant

is concerned with the relationship between electoral laws and

political stability. The works of Duverger and Hermens are

characteristic of this school, and Rae‘s The Political

 

Consequences of Electoral Laws is a more recent example.10

 

The basic tenet of this school is that multiparty systems
are caused by proportional representation. The contention
is that there is a strong correlation between electoral laws
and party systems, with direct and, in the case of multiparty
systems, negative consequences for political stability.

While this argument still has some adherents, it has

largely been discredited by the work of Grumm and others.11

8

These writers have shown conclusively that the introduction
of proportional representation does not lead to an automatic
increase in the number of parties. In Belgium and the
Scandinavian countries, proportional representation was
introduced after these nations had multiparty systems.
Nor has multi-partyism necessarily led to unstable govern-
ments, as the cases of Switzerland and Sweden testify.l2

Explanations for political instability have most recently
been offered by the political culture school. This school
looks for its explanation of instability (or stability) in
the political socialization and attitudinal patterns of
various countries. Usually, one cultural variable--religion,
child—rearing customs, traditional decision-making patterns,
etc.--is isolated as the source of some attitudinal orienta—
tion toward the political system.13

Harry Eckstein, for example, argues that political sta-
bility is due to a particular authority pattern within a
society. In a stable democratic system, such as Norway,
there is a "convergence" of attitudes within the society and
within the party system, making for a limited range of con-
flicts.14 While authority patterns are very probably linked
to political instability, it is difficult to devise a set of
universally accepted measures with which to test for their
importance to political stability.15

Sociological explanations of stability are fairly numerous,
but the general nature of such theories provides few simple

explanations as to why stability obtains in certain societies.16

9

Other sociological theories of a middle-range type offer more
in the way of concrete hypotheses concerning the relation
between social behavior and political stability.

William Kornhauser’s The Politics of Mass Society is

 

a particularly interesting example of middle-range theory of
this sort. It deals with inter-war Germany, the setting of
our study, as one example of the type of a society prone to
political instability as a consequence of widespread anomie,
which causes mass movements. Kornhauser states that the lack
of intermediary associational links between elite and mass
in some societies makes direct access between the two
necessary. In such a situation, there is a likelihood that
the masses will react in a violent manner, providing a large
pool of recruits for radical political movements, such as
Nazism and Communism in Weimar Germany.

One difficulty with Kornhauser‘s theory is that it is
a complex theory of social behavior that fails to discredit
a far simpler explanation of the same mass behavior.
Kornhauser asserts that individuals join mass movements
because they lack a moral code and the movement provides
clear guidelines to action and thought. However, it is
equally possible that an individual joins a mass movement
not for the psychological benefits it affords, but for
economic and social reasons. It may be in the individual’s
economic interest to belong to a political organization, or
he may receive strong pressure from his peer group to join.

In both cases, his motivations are not rooted in anomie.

10

An added problem with Kornhauser’s theory is the linkage
between the social and political systems. The mere presence
of large radical mass movements may be an indication of
societal instability, but the impact of such movements on
political stability must be more closely investigated. An
examination of the case histories of cabinets in the Weimar
Republic reveals that few fell because of the activities of
such movements.

While it would be presumptuous to dismiss any of the six
theoretical explanations discussed above, the present study
focuses only on political variables (of a parliamentary,
party or cabinet type). In searching for explanations of
instability within the parliament, party system, and cabinet,
our concern is with those regular political events of a
parliamentary system: electoral changes, changes in the
party control of cabinet posts, fluctuations in governmental
coalition size, and the degree of ideological diversity within
the governmental coalition.

Some of the variables were selected because of their
inclusion in previous studies of governmental instability.
Others--such as post maldistribution and minimum—winning
coalition size—-are of theoretical interest.17

One variable of empirical and theoretical interest to
the study of governmental stability is that of radical party
growth/decline. The growth of radical parties has been offered

as a contributing influence in the destabilization process.

11

The same holds for ideological diversity in the government
and disproportionate allocation of rewards (i;e., cabinet
posts) among the coalition parties. All of these variables
are exclusively political; they represent various dimensions
of the cabinet, party system, and parliamentary forces affecting
cabinet duration.

The objective of this study is not to account for all
variables contributing to political instability in Weimar
Germany, but only for the most important among a limited
number. Identifying the most important influences on stability
from a limited set of political variables and describing the
relations among the variables within the set will clarify
the process by which governmental instability occurred in
Weimar Germany.

The hypotheses of the study are, for the most part,
original. In those instances in which the hypothesis is
derived from a particular theory, it will be so noted. In
every instance, governmental stability (duration) is the
dependent variable.

One of the most persistent explanations of governmental
instability is the presence and growth of radical parties.
However, not all students of political stability have agreed
that the growth of radical parties in itself is necessarily
a cause of instability. Huntington, for example, states

that the real issue is the degree to which political participation

12

is institutionalized:

The stability of a modernizing political

system depends on the strength of its poli-

tical parties. A party, in turn, is strong

to the extent that it has institutionalized

mass support. Its strength reflects the

scope of that support and the level of

institutionalization.l
Sartori provides the more classic interpretation of the
effects of radical parties in a multiparty system:

An opposition is forced to be responsible if

it knows that it may be called to execute

what it has promised--to respond. But such

motivation is tenuous if the opposition knows

that at most it may only share some peripheral

governmental responsibility . . . And no such

motivation exists for the parties that oppose

the system. Indeed, for the extreme parties

irresponsible opposition is both natural and

rewarding.
The presence of sizable radical parties in a national parlia-
ment is thought to be a destabilizing element because such
parties provide ready support for any movement within the
non-radical opposition parties to bring down the government.
Indeed, radical parties frequently initiate such moves. Whe-
ther this is often an effective strategy, however, is open to
question.

The first hypothesis, then, is simply a statement of the

above:

H. 1: That governmental duration declines with
increases in radical party electoral support.

If, as has often been supposed, the combination of Communist
and Nazi delegations in the Reichstag had a direct and
debilitating effect on governmental stability, a correlation

of the combined growth rates of the two parties (relative to

13
the last electoral period) with governmental duration will
provide positive evidence.

The role of ideological diversity in the process of
governmental destabilization is one that no study of stabi-
lity should ignore. Previous research has pointed to the
significance of such dimensions of ideological stability as
fractionalization of the government in relation to duration
of the government. Taylor and Herman, for example, found
that a weak but significant relation existed between stabi-
lity and such variables as the number of parties in parlia-
ment and the fractionalization of government.20 Axelrod, in
his study of postwar Italian governments, found an inverse
relationship between conflict of interest within the cabinet
(using a measure similar to a variance measure of the distri-
bution of government parties) and their duration.21 Both of
these studies suggest that ideology is an important considera-
tion in any study of governmental stability.

Thus, our second hypothesis is concerned with the problem
of fragmented, ideologically diverse governments once they
are formed. The theory underlying the hypothesis is simple:
that such ideologically diverse and fragmented coalition
governments will endure less long than those containing less
ideological diversity. The hypothesis is thus stated:

H. 2: Governmental duration is negatively related

to the ideological diversity in the govern—
mental coalition.

14

Those who would proclaim the primacy of ideological
diversity in explaining instability must contend with another
school of thought that argues the importance of personal
ambition among the party leaders. MacRae, in his study of
the French Fourth Republic, found no evidence to support this
View; in fact, he found the opposite:

By considering the chronology of the rise and
fall of cabinets, we shall see that the parties
that last went into opposition to a cabinet
before it fell-—and these are the ones whom
Parisian political commentators considered
”responsible"--tended to be ideological

parties . . . . In each case, matters of prin-
ciple rather than the quest for posts in
succeeding cabinets were the main source of
disaffection.

Still, it is reasonable to suspect that the maldistribu-
tion, or disproportionateness, of ministerial posts within a
coalition government has some connection with instability.
One of the most recent works on post disproportionateness
concluded that "The percent ministries received by parties
in governing coalitions is, over all cases examined, directly
proportional to the percent seats they contribute to the coali-
tion."23 While having no direct bearing on governmental
duration, this finding of proportional post distribution
leads to speculation on the role disproportionateness may
play in the matter of stability. In those cabinets in which
the per cent of ministries received is not proportionate to

their resource contributions, what pattern of governmental

duration is to be found? Such speculation leads to the

15
formulation of a third hypothesis:

H. 3: That governmental duration is inversely
related to post maldistribution.

The argument that increases in the size of the party
system are in part responsible for instability has already
been encountered briefly in our discussion of Sartori's
theory. In that theory, the party system size was perceived
as only a function of the pluralism existing in certain
societies having multiparty systems. And, also as noted
previously, theories advancing the proposition that the
number of parties is related to stability are usually of
the constitutional—legal variety. That is, the increase in
party system size is seen as a product of the electoral
laws of that system. As Weimar Germany had a proportional
representation system from its inception, and as no other
political system having a different electoral arrangement is
examined, we can not investigate the relation of proportional
representation to governmental instability. Rather, we
concern ourselves with the number of parties in the system
as a crude indicator of the role of party-system fractiona-
lization in the destabilization process. The fourth hypothe-
sis is, therefore, formulated with an eye to such an inves-
tigation of fractionalization of the party system as it relates
to governmental duration:

H. 4: Increases in the total number of parties
'represented in the national parliament

are negatively related to governmental
duration.

16

A final two—variable relationship to be investigated in
this study involves the size of the governing coalition.
Coalition size stands in such close relation to stability
that it could well be considered a dimension of it. In this
study, however, size is perceived as an independent variable
in relation to the dependent variable, duration.

William Riker’s theory of political coalitions provided
the impetus for a number of studies of coalition size, many
of them dealing with multiparty systems. Riker presented the
principle that, in social situations similar to n-person,
zero—sum games with side payments, participants create coali-
tions just as large as they believe will ensure winning, and
no larger. Researchers have examined this principle in multi-
party parliamentary systems.2LL

Using the size principle as a point of reference, we note
that minority cabinets have always been characterized as weak,
and those of simple majority-or—larger size as strong. Accor-
ding to Riker, however, it is rational for the coalition
members to "pare down” the size of the coalition to a minimum
winning size in order that each member might receive the maxi-
mum rewards. If this strategy is pursued in a coalition
government, a new coalition results that is smaller than its
predecessor, but still of winning size. It would seem, then,
that coalitions meeting the minimum-winning-size criterion at
their inception would endure longer than those of larger-than-
winning size. In other words, the greater the deviation from

minimum-winning size, the shorter the duration of the governmental

1?
coalition.25 Formulated as a hypothesis, this proposition
might be stated as follows:
H. 5: As a coalition government’s size deviates
above minimum-winning size, its duration
declines.

It should be noted that deviations may be below minimum-
winning size as well as above it. We use the range of values
from fifty-one to fifty-five per cent of the parliamentary
strength as the basis for computing deviations below or
above minimum-winning size, the lower value of fifty-one
per cent being that used for less-than—winning sized coali-
tions, while the upper range value of fifty-five per cent is
used for larger—than-winning sized coalitions.

While the five hypotheses advanced above may seem to be
unrelated, there is a theory-~or, at least, the beginnings of
one-—under which the above variables are related in respect
to governmental stability.

A problem one immediately encounters when studying
governmental stability is the dearth of theories explaining
governmental maintenance. While many authors have dealt with
the phenomenon of governmental formation, few have considered
what accounts for the endurance of a coalition government.

Sven Groennings' article "Notes Toward Theories of Coali-
tion Behavior in Multiparty Systems: Formation and Mainte-
nance" is an attempt to deal with just this problem.26
Groennings attempts to dispel the notion that coalition

maintenance is simply repeated coalition formation. He states

that there is a need to identify the maintenance variables and

18

to theorize, stating hypotheses about the impact and rela—
tionships of these variables.

According to Groennings, the master determinants of
behavior within the coalition are similar to those at the
formation level: the party’s desire for reward and for
self-preservation. His model incorporates five elements,
each of which represents a category of variables. These
five categories are termed motivation and communications,
situation, compatibility, strategy, and coalition apparatus.27

The coalition apparatus refers to the structure through
which interaction of the other variables occurs. Included
in this category are such variables as positions (posts),
nature of leadership, decision-making model, programs, and
rules of the game. Groennings hypothesizes that if a party
with a record of dissent in a particular policy area gains
control of the ministry within that area, the stability of
the cabinet is threatened. Durable coalitions, Groennings
further hypothesizes, will find posts for the leaders of the
important party factions and provide for representation of
more than one party at the highest echelon. The decision-
making structure, also included under the coalition apparatus
category, tends to follow either the dissent model or the
unanimity model. The dissent model allows the member parties
to propose bills which will not be the basis for cabinet

questions. The unanimity model, according to Groennings,

19
is the one every coalition opts for because it minimizes the
opposition’s ability to exploit differences among the coali-
tion parties. The unanimity model includes a veto provision,
whereby any coalition party may eliminate any policy to which
it is strongly opposed. This leads to cooperative behavior
among the member parties (sub—coalitions and log-rolling),
which contributes to durability.28

Variables within the strategy category include goal
intensity, possible concomitants, size, position in spectrum,
and success estimate. Not all of these variables are easily
measured, while others are (size, for example). Groennings
states that the larger the party, the more it will insist on
payoffs proportionate to its resource contribution. As for
position in the spectrum, pivot parties (those at the peri—
phery of the coalition spread) are likely to be hard bar-
gainers, while "captive" parties tend to be weak bargainers
who appeal to standards of justice and fairness.

Turning to the compatibility category, we find five
variables: policy goals, stage of coalition development,
resources of the parties, reliability of parties, and number
of parties. Groennings suggests that policy confrontations
occur in stages, which leads him to further suggest that
there are stages to coalition maintenance. The first stage
is that of forming the coalitions, followed by a "honeymoon
period” in which mutual adjustment among the coalition parties
occurs. Eventually, common goals are exhausted, and the coa-

lition enters a period of confrontation: while minor

2O

conflicts do not break up a coalition, according to Groennings,
parties have distinct primary goals which must be accentua-
ted, especially as an election approaches. These two stages-—
the period of exhausting mutual goals and the electoral
period——are the stages in which crisis becomes a distinct
possibility.29

Situational variables include only external pressures
and constitutional variables (for example, elections).
Interest groups underwriting the coalition parties are often
identified as the major external pressure, but others which
might easily be of greater relevance to stability include
foreign military and economic penetration, for example.
Constitutional variables would include governments‘ frequent
resignations after a national election. Groennings states
that this is because the election removes the coalition
situation or changes the number of partners required to
constitute a minimum controlling coalition.3O

Coalition membership entails the altering of decision-
making and internal communications processes, in order that
the party may be effective in its new role. A motivational
problem is inherent in this change, for coalition membership
increases the number of mixed-motive situations confronting
the party leadership. Groennings formulates three hypotheses
concerning this problem:

(1) that amenability to compromise decreases as

communications are passed downward from cabinet
representatives to parliamentary parties to local

 

21

party organizations; (2) that the more cen—
tralized the party structure, the easier it is
for the party to remain in the coalition;

(3) the more centralized is each of the
coalition partners and the greater the disci-
pline in3iach, the more stable the coalition
Will be.

The five categories of variables developed by Groennings
comprise a number of dimensions of the stability problem.
Except for the statement that the first four categories
of variables are interrelated through the coalition
apparatus, his model of coalition maintenance makes no
statements concerning the interaction among the five cate—
gories. Groennings‘ major contribution is in isolating the
categories of variables possibly related to instability.

He does not operationalize or test his model.

This study seeks to determine how certain of the
variables categorized by Groennings are interrelated to one
another and to governmental stability. It is not a test
of Groennings’ theory, as the "theory" is little more than
a rough model of the process of coalition maintenance, or,
as we term it, stabilization. Moreover, not all the variables
listed by Groennings are measurable, given the limitations on
data collection in a historical setting. Nor is Groennings'
categorization exhaustive in all respects: he fails to
include any consideration of the role of radical parties.32

Groennings' model is presented in order to give the

reader an idea of the level of theorization in the area of

coalition stability and to provide a loose theoretical framework

22

for the study. We are merely following Groennings’ conclu-
sion:

One needs to penetrate more deeply to the
many reasons for the effective functioning
of some coalitions and the many possible
causes of failure of others. There is a
need to contemplate and test the impact
and relationships of these factors and
thereby to move beyond this rudimentary
and tentative sketch, which is presented
as a stimulus toward more accurate and
useful theory.33

The final hypothesis of our study is related to the task
of ascertaining the "impact and relationships" of certain of
those factors, namely, those variables described in the
preceding five hypotheses.

H. 6: Governmental instability is partially

produced by a combination of some or
all of the following: (1) radical
party growth; (2) ideological diversity
in the governmental coalition; (3) post
maldistribution; (4) increases in the
number of parties; and (5) deviations
from minimum—winning size.

Governmental duration-—as in the preceding hypotheses--

provides our measure of governmental stability.

The Measures

 

The choice of measures is always a matter of crucial
importance in any piece of research. In this section, we
describe the measures of our independent and dependent
variables.

The rate of radical party growth or decline (RP) is
measured as the difference between the total percentage of
radical party (Communist and Nazi) parliamentary representa-

tion following the most recent election and that of the

23
electoral period for the preceding cabinet. Radical parties
are those political parties identified as ideologically
opposed to the existing governmental system. The rate of
radical party change is thus measured in terms of changes
from each preceding electoral period.

The measure of ideological diversity in the governing
coalition (IDG) is an average of two measures: one of
governmental fractionalization and the other of ordinal
disagreement. The ordinal disagreement measure is derived
from one first proposed by Robert M. Leik.34 Taylor and
Herman adapted this measure for use in gauging ordinal
disagreement in party systems, governments, and parliamen-
tary oppositions.

The ordinal disagreement measure is based on the assump-
tion of a left-right ideological continuum.35 For each party,
i, a cumulative proportion, CPi, is calculated. This is done
by summing the seat proportions of all parties to the left
of party i together with i's proportion. In symbols:

CPi = l/h‘Z'rj, for all jtg i, in an ideological ordering,
where fl, f2, ? . . fN, are the number of seats held by
N parties, and N equals the total number of seats, with fl as
the party furthest to the left. If the cumulative proportion
of disagreement for each party is less than one-half, then
the proportion of ordinal disagreement is defined as equal
to CPi. Otherwise, it is defined as the remainder when the
seat proportion is subtracted from one. That is:

di = PCi, if PCi ( 3, otherwise ai = l — PCi.

-1-

24

Ordinal disagreement is defined as:

N
D: g a.,
1:1 1

The reader should be aware of some peculiarities of the above
measures. The index, D, is highly sensitive to the number
of parliamentary parties. Therefore, the existence of
splinter parties can have a remarkable effect on it. To
compensate partially for this effect (as the number of
splinter parties in the Weimar Republic was high), a
standardized set of criteria is applied in determining which
parties are to be included in the computation. These criteria
consist of the following:

The party is included if it (1) is a major party,

i.e., one that is considered as eligible for

coalition membership or else one that is con-

sidered to be an important opposition party;

(2) has obtained 5 per cent or more of the

popular vote in the election; (3) while not

conforming to either of the above criteria,

the party has participated in at least one

of the governments during the electoral period

in question.

The second measure of ideological diversity--fractiona-
lization--was originally devised by Rae as an indicator of
party system fragmentation.36 It may be described as
follows: given N as the number of political parties in
the government, with n being the total number of seats in

the parliament controlled by coalition parties, fractiona-

lization is one minus the sum of the square of all the seat

25

proportions for all parliamentary parties. That is,
N
2
FZl-Epi:
i=1

where pi(=fi/n) is the proportion of seats held by the ith
party.

As we are interested in governmental fractionalization,
that is, F defined for all of the parties in the cabinet,
F is calculated only for those parties, not for the entire
party system. F varies from a minimum value of zero, in
which case only one party forms the government, to a
maximum of one, in which case there is an infinite member
of parties.37

The above formula is applicable only when the f’s are
large. If this is not the case-—and in most of the govern-
ments of Weimar Germany it was not--then the following
modified formula is used:

N

F = l - mil—II 1g fi(fi-l)
with F again varying from a minimum of zero (given one party
in the government) to a maximum of one, in which case there
is an infinite number of parties. This formula is suitable
when a sizeable number of nominal groups are involved and
where frequencies are too small to permit the approximation
provided by the first formula.38

The same criteria for the inclusion of parties used in
the computation of ordinal disagreement are used in the calcu—

lation of governmental fractionalization. For each of the

26

seventeen coalitions in the population, these calculations
of ordinal disagreement and fractionalization are averaged
together to produce a combined measure of ideological
diversity for each government.

Post maldistribution (PM) is measured as the sum of the
difference of each coalition party's contribution (as the
percentage of parliamentary seats it controls) to the
coalition from its percentage share of the total number
of ministerial posts. The value of this variable can vary
below or above zero, zero being the absence of post mal—
distribution (PM) in a coalition government.

The party system growth/decline measure (NUM) is a
straight-forward representation of the growth rate of the
party system. It consists of the percentage increase or

decrease in the number of parties in parliament (Reichstag)

 

over the last parliamentary period. The same set of

criteria used in calculating ordinal disagreement is employed
in determining which parties are included in the computation
of the party system growth/decline measure.

Deviations below and above minimum—winning size are
measured in terms of the coalition's parliamentary strength
(as a percentage) after the adjusted minimum—winning size
(fifty-five per cent) has been subtracted. Thus a govern-
mental coalition controlling fifty-seven per cent of all
seats in the Reichstag would have a DMW value of +.O2.

The adjusted minimum—winning size is based on Riker's

\

39

modified minimum-winning size.

27

The most important variable is the dependent variable:
governmental stability/instability. In this study, we
employ duration-in-months as the measure of a government‘s
stability. Two reasons dictate the use of this measure:
(1) it is linear, thus lending itself to use in regression
analysis, and (2) it is a traditional measure of stability

. 40
in studies of parliamentary regimes.

Summary

In this chapter, we have presented the various theories
concerning governmental stability, together with a discussion
of the concept. In this study, we have selected duration as
the measure of governmental stability.

The data for our study are drawn from the seventeen
coalition governments of Weimar Germany during the period
1919—1931. While episodic events, such as inflation and
foreign military intervention, definitely affected the
course of domestic German politics in this period, our major
concern is with political variables and their relationships
with governmental stability. Our primary purpose is to
determine the importance of such variables in the process of

governmental instability.

28

NOTES

lDuncan MacRae, Jr., Parliament, Parties, and Society in
France, 1946-1958 (New York: St. Martin’s Press, 1967), and
Robert Axelrod, Conflict of Interest (Chicago: Markham
Publishing Company, 1970). Chapter 8 is an application of
Axelrod’s theory to post-war Italian governments.

 

 

 

2Michael Leiserson, Coalitions in Politics: A
Theoretical and Empirical Study (Ph.D. dissertation,
Yale University, 1966); Sven Groennings, E. W. Kelley,
and Michael Ieiserson, ed., The Study of Coalition
Behavior: Theoretical Perspectives and Cases from Four
Continents (New York: Holt, Rinehart, and Winston, 1970);
Abram DeSwaan, Coalition Theories and Cabinet Formations
(San Francisco: Jossey-Bass, 1973).

 

 

 

 

 

 

3These categories, as well as much of the following
discussion of them, were suggested by Robert J. Jackson and
Michael B. Stein, Issues in Comparative Politics (New York:
St. Martin’s Press,'l971), Chapter 3.

 

uSome idea of the difficulties in analyzing historical
events is suggested by Abram DeSwaan, ”An Empirical Model
of Coalition Formation as an N—Person Game of Policy Distance
Minimization,” in Sven Groennin s et al., eds., The Study
of Coalition Behavior, pp. 424- 25: ”The data that are
supplied by expert observers, historians, and autobiographers,
on the other hand, Show a multitude of issues, an almost
inextricable interplay between accidental circumstances and
personal ambitions, principles and sympathies, so as to
complicate analysis beyond what is feasible."

 

 

5Deane E. Neubauer, "Some Conditions of Democracy,"
American Political Science Review, 61 (December, 1967),
607-619 0

 

6Philips Cutright, "National Political Development: Its
Measurement and Social Correlates," in Nelson Polsby et al.,
Politics and Social Life (Boston: Houghton Mifflin Co., 1963).

 

7A. Lawrence Lowell, Governments and Parties in Continental

 

Europe, Vol. 1 (Cambridge: Harvard University Press, 1896),
pp. 73-74.

Giovanni Sartori, "European Political Parties: The
Case of Polarized Pluralism," in Joseph LaPalombara and Myron
Wiener, eds., Political Parties and Political Development
(Princeton: Princeton University Press, 1966), p. 159.

 

29

9Maurice Duverger, Political Parties (New York:
John Wiley and Sons, 1954). This work is a classic example
of systemic instability explained in terms of constitutional
and legal institutions, especially in its treatment of the
relationship between proportional representation and multi—
partyism.

10F. A. Hermans, Democracy or Anarchy? (South Bend:
Notre Dame University Press, 1940). Hermans argues that
proportional representation leads to multipartyism, which,
in turn, leads to governmental instability.

11John G. Grumm, "Theories of Electoral Systems,"
Midwest Journal of Political Science, 2 (1958), 357-376.

12Nils Stjernquist, "Sweden: Stability or Deadlock?"
in Robert A. Dahl, Political Oppositions in Western Democra-
cies (New Haven: Yale University Press, 1966), pp. 116—146.
Stjernquist concludes that the modern Swedish stable pattern
of government has its roots not only in the difficulty of
achieving institutional reform, but also in the manner in
which party organizations act as "built-in stabilizers."
He terms this pattern a "stable deadlock."

13The best multinational study of this sort is Almond
and Verba, The Civic Culture (Princeton: Princeton University
Press, 1963).
1“Harry Eckstein, Division and Cohesion in Democracy
(Princeton: Princeton University Press, 1966).

 

 

 

 

 

 

 

15But some tentative effort has been made in this
direction. See Harry Eckstein, "Authority Relations and
Governmental Performance: A Theoretical Framework,"
Comparative Political Studies, 2 (October, 1969), 269—325.

l6Ivo K. Feierabend et al., "Aggressive Behavior within
Politics, 1948—1962: A Cross—national Study " Journal of
Conflict Resolution, 10 (September, 1966), 249-27I. THIS
study presents a series of societal stability indicators,
which are then compared with the average durations of the
seventeen democracies in the sample.

 

 

 

17The post maldistribution variable was suggested by
a study of coalition payoffs in parliamentary systems.
See Eric C. Browne and Mark N. Franklin, ”Aspects of Coalition
Payoffs in European Parliamentary Democracies," American
Political Science Review, 67 (June, 1973), 453-469. The
difference from minimum-winning size variable was derived from
the size principle of William H. Riker, The Theory of Political
Coalitions (New Haven: Yale University Press, 1962), pp. 32—33.

 

 

 

 

30

18Samuel P. Huntington, Political Order in Changing
Societies (New Haven: Yale University Press, 1968).

19Sartori, op. cit., pp. 157-158.
20Michael Taylor and V. M. Herman, "Party Systems and

Governmental Stability,” American Political Science Review,

65 (March, 1971), 28-37.

 

21Axelrod, op. cit., pp. 177-178.
22

23
24

MacRae, op. cit., p. 8.
Browne and Franklin, op. cit., p. 467.
Riker, op. cit., pp. 32-33.

25This is an original hypothesis, in no sense attribu-
table to Riker.

26Sven Groennings, ”Notes Toward Theories of Coalition
Behavior in Multiparty Systems: Formation and Maintenance,"
in Sven Groennings et al., The Study of Coalition Behavior
(New York: Holt, Rinehart and Winston, 1970), pp. 445-465.

 

27Ibid., p. 460.
28Ibid., p. 462.
99Ibid., p. 464.
30Ibid., p. 463.
3lIbid., p. 462.
32At times, it seems that Groennings is formulating a
theory for moderately pluralistic systems, not those like
the Weimar Republic, which Sartori termed "polarized pluralistic."

33Ibid., p. 465.

3LLRobert M. Leik, "A Measure of Ordinal Consensus,"
Pacific Journal of Sociology, 9 (1968), 85—90. Herman and

Taylor, op. cit., pp. 33-34-

35The ideological continuum, or ordering of the Weimar
parties, is that found in Figure l, p. 60. This conforms with
the usual ordering of Weimar parties. See Sartori, op. cit.,
P- 155, and DeSwaan, op. cit., p. 163.

31
36
Douglas Rae, The Political Consequences of Electoral
Laws (New Haven: Yale University Press, 1967), pp. 53-58.
See also Dougals Rae and Michael Taylor, An Analysis of
Political Cleavages (New Haven: Yale University Press,
1970), Chapter 2.

 

 

37Taylor and Herman, op. cit., p. 30, provides the
following illustration of the measure as it is employed for
the party system:

"If the seats are shared between two parties
in the ratio 60:40, then F has the value of
0.48, and if the seats are shared between
three parties in the ratio 40:40:20, then
F=0.64."

38The computation of fractionalization is based on the
assumption of mixed and matched pairs in any hypothetical
group. The number of matched pairs in any group is
fi(fi-l)/2. If this is summed over all groups and divided by
the total number of pairs, a proportion of all the pairs which
are matched is obtained:

1'1
5‘3 fi(fi-l)
l=l

§N(N-l)

 

Since all pairs which are not matched are mixed, this quantity
is the complement of F, and we obtain

n
F21 " l 2 fi(fi-l).

NIN-I) i=1

See Rae and Taylor, op. cit., pp. 20-33, for a fuller explana-
tion of this measure.

39Riker, op. cit., 88-89.

quxelrod, op. cit., p. 167, states the argument for
duration as a measure of stability: "A coalition with low
conflict of interest can be expected to last longer once
formed than an average coalition, just because disputes
within such a coalition will be easier to receive. In a
sense, this is a theory of ’natural selection' and survival
of viable political coalitions."

Chapter 2

THE SETTING OF THE STUDY: GOVERNMENTAL
STABILITY AND THE WEIMAR PARTY SYSTEM, 1919-1931

Introduction

 

Weimar Germany is a well-known example of governmental
instability. Although more recent examples of nation states
experiencing governmental instability as a chronic, recurring
phenomenon are available, such as the Fourth Republic of
France and modern Italy, Weimar Germany presents a standing
challenge to any theory of governmental instability.

Any study of the Weimar Republic must come to terms
with the complexities of this system. The obvious features
on which students might concentrate are the electoral laws,
the electorate, interest groups, and parties. To these
might be added the international political climate (important
for Weimar Germany) and economic conditions, such as infla-
tion, unemployment, and declining GNP. While all of these
were contributing influences to the instability of Germany's
governments between 1919 and 1931, they are by no means the
only variables through which instability might be explained.
Our measures are based on readily perceivable changes
within the party system, parliament, and cabinet: party
electoral gains or losses; control of cabinet posts by

Parties; fluctuations in the size of the governing coalition,

32

33

and (somewhat more elusive in nature) the ideological diver-
sity among the governing parties.

The student of Weimar politics has a large and varied
literature on which to draw. Some of the early studies of
this period concerned themselves with the importance of
electoral laws, especially the role of proportional repre-
sentation in facilitating the growth of an unwieldy party
system.1 However, the major thrust of postwar research on
Weimar Germany has been directed at explaining systemic
instability through examination of the shifts in electoral
support for the various parties. This trend is reflected
in the work of R. Bendix, S. M. Lipset, K. D. Bracher,

K. O'Lessker, and, most recently, W. P. Shively.2

While electoral analysis may explain the ultimate
collapse of the Republic, it does not elucidate the funda-
mental causes for the systemic instability of the Weimar
political system. These causes are more likely to be found
in societal cleavages reflected in the party system, and
the interface between such party cleavages and coalition
formation.

Studies of coalition formation in Weimar Germany are
few in number, compared to electoral studies, and mostly
inaccessible to American political scientists. M. Stuermer's
work on coalition formation is restricted to the so-called
stable period of Weimar political history, 1924 to 1928.3

P. Haungs, another German scholar, studied the relationship

34
between the Reich President and the proceSs of cabinet
formation.4 Most recently, V. Rittberger's work has
contributed to the literature on cleavages in the party
system and their effects on coalition formation.5

Abram DeSwaan's Coalition Theories and Cabinet Formations

 

presents a systematic test of contemporary formal coalition
theories. Applying these theories to nine European parlia-
ments after 1918, including the Weimar Republic, DeSwaan
found that no theory considered could adequately explain
specific historical coalitions. His findings did indicate,
however, that closed minimal range theory (e.g., the theory
predicting that parliamentary coalitions will be closed and
of minimal range) explains the long—term coalition tendencies
operating throughout the period and the systems considered.
All of these studies deal exclusively with national level
cabinet formation. None of them deals specifically with
the question of governmental instability; they confine
themselves primarily to the cabinet formation process.

Any study of cabinet formation and governmental stability
in Weimar Germany Should begin with a description of the
most prominent actors, the major political parties. Through-
out the thirteen-year period, number and types of parties
changed to only a limited degree. By 1924, all the
major parties had made their appearance, and those of
a transitory nature had disappeared. In addition to the
major parties, many splinter and minor parties existed.

The proportional representation system encouraged the growth

35

of splinter parties, with the inclusion of a provision for
pooling "unused" votes in the districts at the national
level. Thus, a party having only miniscule strength in the
electoral districts would still obtain seats in the Reichstag
once its votes were tallied for the national constituency.
Few of these parties participated in the governing coalitions;
most were, in fact, legislative lobbies with party labels.

A description of the major party groups is presented
below. -In the following sections, we describe the basic
coalition patterns during the 1919-1931 period and conditions

affecting the cabinet formation process.

The Party System, 1919-1931

 

Walter Burnham, in his paper "Political Immunization
and Political Confessionalism: Some Comparative Inquiries,"
provides a scheme for classifying political parties in
Weimar Germany.7 Burnham dichotomizes the Weimar party sys-
tem into confessional and non-confessional parties, on the
basis of the presence or absence of strong organizational
networks. Burnham classifies the Social Democratic Party
(SPD), Communist Party (KPD), and Zentrum as confessional
parties. In these parties, he maintains, ties were social
(non-political) as well as political.

While the confessional/non-confessional dichotomy is
Usexful in discussing organizational and structural characteris-
tics; of political parties, in this study we are concerned

.Prilnaljly'with the identification of radical parties, or

Pa1?txies having as their goal the termination of the parliamentary

36
system and its replacement with a new regime. The social
base of a party becomes important in this context only when
it is broadened to increase the electoral strength of a
party.

In one respect, the Burnham typology is somewhat
misleading, for it places the KPD among the confessional
parties. The Communist Party in Weimar Germany never
developed auxiliary organizational structure to the extent
of the Social Democratic or Zentrum parties. While the KPD
attempted to establish such an organizational substructure,
it was limited by the dominant position of the SPD as
representative spokesman for proletarian interests. Only
when the worsening employment picture of the late 1920's
and early 1930's severed many of the ties between the SPD
and its membership did KPD voting increase. The KPD was a
radical sectarian party throughout the Weimar period, with
none of the auxiliary trappings of the other confessional

parties.

The Confessional-Nonradical Parties: Social Democrats and Zentrum

 

The SPD and Zentrum parties were the major supporters of
the Republic. Prior to 1918, both were excluded from the
government. Both entered the postwar period with little
experience in policy making. The SPD was, organizationally
speaking, the more impressive of the two. Possessing as it did
a remarkably well-developed party bureaucracy, it was rigidly

organized, highly disciplined, and supported by a large

37

dues-paying membership drawn from the ranks of organized
labor.

However, as Robert Michels, Carl Schorske, Richard Hunt,
and other students of the SPD have determined, the SPD
suffered from the effects of over-bureaucratization and
a widening gulf between the party leadership and the rank-
and-file. As Schorske notes concerning the immediate pre-
war period in Germany:

The same socio-economic situation which made the
union leaders conservative had the opposite effect
on the rank and file. The rising cost of living,
the intense and widely shared experience of strike
and lockout, and the unprecedented aggressiveness
of the employers generated in the workers a new
militancy and a receptiveness to radical political
ideas. In this tense social situation, German

Social Democracy received and reacted to the
challenging tidings of the revolution in Russia.

8
The SPD leaders' decision to support the German war
effort caused an open split in the ranks in 1915, with the
dissidents, led by Karl Liebknecht and Rosa Luxemburg,
leaving the party to form a separate and independent
socialist movement. Although by 1923 most of the dissidents
had been taken back into the fold, a residue of hardcore
irreconcilables joined the nascent German Communist Party
(KPD). The emergence of the Independent Socialist movement,
and subsequently the KPD, weakened the SPD's claim to
leadership of the German proletariat, forcing the party to
forego cooperation with the bourgeois parties in its efforts

to recapture working—class votes. At the national level,

this weakened the pro—system forces at a time critical to

38
the future course of the Republic.

Certain drastic modifications were made in the struc-
ture of the SPD following the First World War in order to
meet the exigencies of the new Weimar system. Hunt describes
these changes that took place after the old electoral
districts of imperial Germany were redrawn:

First, the basic unit of the party was changed
from the electoral district organization
(Sahlkreisverein), founded on the old imperial
constituencies, to an electoral region organi-
zation (Bezirksverband), based on the new elec-
toral regions offthe Republic. Proportional
representation required much larger electoral
divisions than the constituency system. Instead
of nearly 400 districts there were now only

35 regions. With certain exceptions the old
Social Democratic Land and regional organizations
were adapted to fit The Weimar electoral divisions,
and, in this shape, became the new regional
units. They inherited en block most of the
functions of the former district organizations:
they had the final word in the selection of
Reichstag candidates, ran the campaigns, and
served as the party's basic administrative units
for finances, congress elections, etc. The
prewar Land and regional organizations, as

such, were dissolved in consonance with the
centralizing features of the Weimar Constitution.
But where a Land had two or more regional organi-
zations they were permitted to meet together,
should any pressing state issues arise.9

 

 

The most significant result of these changes was further

to widen the gap between the party elite and the rank-and-
file. Hunt observes that the new party power centers--the
regional organizations--were too large and unwieldy for
any form of democratic decision-making.10 The practice of
electing party congress delegates from the regional organi-
zations provoked sustained and widespread discontent among

the lower echelons of the SPD membership.

39

Another weakness of the SPD in the Weimar period was
its lack of dynamic leadership. Various explanations have
been offered for this, the most convincing, perhaps, being
the nature of socialization within the party, with the
emphasis on time-serving as a criterion for leadership
rather than demonstrable political skill. Yet another
explanation might be the gradual embourgeoisement of the
party leadership, which had sapped much of the reformist
zeal of the party. Michels was the first to isolate the
source of this development of a middle-class ethos among
the leadership of a supposedly proletarian party:

The embourgeoisement of the party is an
unquestionable fact, but its causes will

be found in a process very different from
the entry into the organizations of the
fighting proletariat of a few hundred
members of the middle class. The chief

of these causes is the metamorphosis

which takes place in the leaders of
working class origin, with the resulting
embourgeoisement of the whole atmosphere in
which the political activities of the party
are carried out.11

It seems safe to conclude that the SPD during the Weimar
period suffered from major weaknesses of an ideological,
organizational, and functional nature. Its leadership was
challenged on the Left, while on the Right it was incapable
of transcending its working-class electoral base. It was
a party whose leadership lacked dynamism, and whose
instincts were defensive rather than offensive.

The Catholic Zentrum Party (Z) was a confessional party

in the original sense. This party arose out of the struggle

40
between the Prussian state and the Church over the role of
Church—supported schools. Like the SPD, the Zentrum had
been in the parliamentary opposition under the Imperial
regime. It served principally as an umbrella party for
German Catholics. Under the Weimar Republic, it ascended
to power with the SPD, but its identification of its role
was not always an easy task. Rudolf Morsey encapsulates
much of the dilemma of the Zentrum in the following passage:

From 1870 onward this party had represented
primarily the interests of the German
Catholics, and after the Revolution of

1918 it found it far from easy to adjust

to the Republic. One group split away in
1918 to form the Bavarian People's party,

and other Catholic voters defected to other
parties, so that the Center was unable to
maintain its pre-war election strength.
Although it had 60 per cent of the women's
votes, it received no more than 13 per cent
of the total vote, or, combined with the
Bavarian People's party, 17 per cent in

1924 and 15 per cent in 1932. Nevertheless,
from 1919 on its position as a stabilizing
element in the various Weimar coalitions

gave positive roof of this "natural—born
middle party's' capacity to govern. In the
words of its historian Carl Bachem it "embraced
all classes” and was therefore able to "harmo-
nize within its own ranks the inevitable con-
flicts of interest."12

The most important characteristic of the Zentrum's elec-
toral base was that it was geographically delimited to those
areas in which Catholics resided. The major strongholds
of the Zentrum were Bavaria (before the creation of the
Bavarian People's Party) and the Rhineland. The role of
the Zentrum in certain Laender was minimal or nonexistent,

as in Thuringia and Saxony, while in others (Prussia, Baden,

41
and Hesse, for example) it was a major actor in the state
governing coalitions.

Like the Social Democratic Party, the Zentrum offered
its membership a variety of social services, social activi-
ties, and various other selective benefits. These acted to
strengthen the members' ties to the party, with the Church
often acting as the mediating agent. Such incentives were
a necessity to a party so dependent upon a specific stratum
of the population for its support.

The Zentrum was situated at the ideological center of
the party system, providing the central pole of a multi—
polar system, according to Sartori's terminology. As such,
the Zentrum enjoyed a natural advantage: it was included
in every ruling coalition at the national level from 1919
to 1931. In view of the wide range of coalition partners
with which the Zentrum shared power, it is surprising to
find so little variance in the Zentrum's electoral strength
over the period from 1919 to 1932: the rate of voter
attrition was markedly lower for the Zentrum than for any
other middle—class party of the center or right.

However, as has already been emphasized, both the SPD
and the Zentrum were constrained in their electoral appeals
'by the very nature of their traditional voter alignments.
Neither party could transcend its identification as a
confessional party in the minds of the uncommitted electorate.
Both were threatened by schisms which forced them to devote

considerable energies to fence-mending rather than to widening

42
their electoral base and reaffirming the commitments of their
coalition partners. Such constraints had repercussions on
the stability of the regimes in which these two parties

participated.

Bourgeois-Protestant Parties: DDP, DVP, and DNVP

Concerning the bourgeois liberal and conservative parties
of the Weimar period, S. M. Lipset makes the following
observation:

As the Nazis grew, the liberal bourgeois center
parties, based on the less traditionalist ele-
ments of German society--primarily small business
and white—collar workers--complete1y collapsed.
Between 1928 and 1932 these parties lost almost

80 per cent of their vote, and their proportion

of the total vote dropped from a quarter to less
than 3 per cent. The only center party which
maintained its proportionate support was the
Catholic Center party whose support was reinforced
by religious allegiance. The Marxist parties, the
socialists and the Communists, lost about a tenth
of their percentage support, although their total
vote dropped only slightly. The proportionate
support of the conservatives dropped about

40 per cent, much less than that of the more
liberal middle-class parties.l3

The poor showing of the liberal parties toward the close
of the Weimar Republic was due to several characteristics
of these parties, but one of the major causes of their
decline was certainly the lack of unity in ideology or in
the realm of practical political compromise of these parties.
This inability to cooperate can be traced to the fact that
the Democratic and German Peoples’ parties had a common

origin: the National Liberal Party (NLP).

43

It would be more accurate to state that the Democratic
Party (DDP) was a union of two parties: the NLP and the
Progressive Peoples' Party (Fon). Following the end of the
war and the breakup of the NLP, the left wing of the NLP
and the Progressives united to form the DDP. The major
influences in the party were banks, liberal industry, and
various nationalist-minded intellectuals. The mass support
for the DDP came primarily from the bourgeoisie, various
circles of the white-collar clerical and civil service
class, and the intellectuals.lLL

The party initially hoped to attract all moderate forces
in the Republic so as to form a strong political movement
in support of the Constitution. Such a movement never
materialized. The party was relegated to a weak but some-
times crucial position in the Weimar Coalition of Social
Democrats, the Zentrum, and the DDP. By 1930, the party had
suffered severe electoral setbacks, and the DDP was ripe for
dissension. It appeared in the form of the State Party
(DStPL a faction of young DDP dissidents who formed a
right-wing splinter party. This defection spelled the
effective end of the DDP as a viable political force.

Long before the defection, however, the DDP suffered

serious setbacks. In the national Reichstag elections of

 

May, 1924, the DDP vote fell from its previous (June, 1920)
level of 8.4 per cent to 5.7 per cent. In an effort to
consolidate moderate support, the leader of the Democrats,

Koch-Weser, sounded out the other liberal party--the

44

People's Party (DVP)--about the possibility of a merger.
The DVP leader, Gustav Stresemann, refused the offer to
merge, despite the fact that his own party had suffered a
greater loss of support than the DDP (from 14.0 per cent
to 9.2 per cent of the total vote).15

This refusal was probably caused by the poor relations
that had characterized the liberal movement since the war
years. The first attempt to found a liberal party out of
the old NLP had foundered when Stresemann took the right
wing of the party out of the progressive camp to form the
DVP. The financial support for the DVP came largely from
heavy industry, the banks, large business concerns, and the
upper echelons of the civil service. Its electoral base

seems to have been the old bourgeoisie and the nouveau riche

 

class of merchants that was emerging in the urban sector.
It also drew the support of some peasant and petty bourgeois
voters.

Stresemann's party suffered from divisions within its
own ranks, given the strong conservative influence exerted
by some of its members, whose political sympathies were very
similar to those of the conservative German National Peoples'
Party (DNVP). Stresemann altered his political views to the
dimensions of the Republic, but certain members of the DVP
never developed any tolerance for the new democratic politics
of Weimar Germany. The influence exerted by the DNVP upon

the coalition of all moderate, pro-Republican parties (the

45
so—called Grand Coalition) was a negative one. Following
Stresemann's death in 1929, the pressure from the right
became irresistible.

If the Peoples' Party could be said to have had a
positive orientation toward the Republic-~at least under the
leadership of Stresemann-~the same could not be said for the
German National Peoples' Party (DNVP). Formerly the
Conservative Party under the Empire, the DNVP was a party
of the reactionary right. It was monarchist, anti—Semitic,
chauvinistic, and strongly opposed to the policies of the
Republican parties, particularly to the implementation
of the terms of the Versailles Treaty. Its support came
primarily from the landed aristocracy of Prussia, as well
as various ultra-nationalist industrialists, high-ranking
civil servants and military officers. One of its most power—
ful sources of electoral support was the veterans' associa-

tions, above all the Stahlhelm (Steelhelmet), an avowedly

 

rightist organization with a numerous following. Yet another
source of support for the DNVP was the orthodox Protestant
Church.

The stronghold of the DNVP was undoubtedly the eastern
provinces of Germany: East Prussia, Silesia, and the other
border areas. The Nationalists made few inroads in urban,
industrialized, or Catholic areas. The DNVP policies were
oriented toward the agrarian interests of the East, and the
party had little to offer--save symbolic opposition to the

Versailles Treaty--to other sectors of the electorate.

46
Until the rise of the Nazis in 1928, the DNVP was the
focus of all forces in opposition to the Republic. The
party reached a major turning point in 1924, when, despite
its opposition to the Dawes Plan in the preceding election
campaign, it became necessary for the party to vote the required
two-thirds majority necessary to pass the bill in the

Reichstag. This volte face was due to strong pressure from

 

industrial and agrarian interests within the party to accept
the Dawes Plan, as it promised major sources of foreign
credit for German industry and agriculture.

Under the leadership of Hergt, the DNVP joined a
rightist governing coalition on several occasions, but
reactionary forces within the party invariably succeeded
in forcing the DNVP delegation out of the government.

The Nationalists, as Lipset noted, lost support at the
polls toward the end of the Weimar Republic, but at a slower
rate than the other bourgeois Protestant parties. Schnaiberg
and Herberle contend that the DNVP was the most important
source of Nazi support prior to the final election in March,
1933.l6 Lipset has argued that the DNVP and Nazis were
competing for different sectors of the voting population, but
a recent unpublished study by Stefanie Cameron suggests that
the conservatives showed a decided trend toward losses
associated with Nazi gains: in studying the change in two
national elections in 1932, Cameron found that as the Nazi
tide shifted with the second election, the DNVP was able to

regain some of its lost voters in its strongest areas.

47
She suggests, furthermore, that the DNVP may have been the
recipient of votes that had previously gone to the DVP prior
to 1932.17

Outside of Prussia, the DNVP'S role in the politics of
the other states was constrained by the sectionalist appeal
of the party and the narrowness of its interest group
affiliations. In Bavaria, the party never controlled more
than 11 per cent of the Iandtag seats. In Saxony, it gained
20 per cent of the seats in the November, 1920, election,
but by May, 1929, this proportion had declined to 8 per cent.
In other Laender, the DNVP never received more than 13 per
cent of the Landtag seats.

The three parties discussed above by no means exhaust
the list of bourgeois Protestant parties. They were, however,
the major parliamentary representatives of the German Protes-
tant middle classes. Other parties, such as the Economic

Party (Wirtschaftspartei) or the Peasants' League (Bauernbund),

 

 

seem to have functioned more as parliamentary pressure groups
throughout most of the period than as office-seeking organi-
zations. In the case of the Economic Party, this impression
is borne out by the fact that the composition of the party's

Reichstag delegation was made up exclusively of businessmen

 

who were active in interest group associations for small
business.
Radical Parties: The KPD and NSDAP

The Communists and the Nazis shared some similarities.

First, both parties emerged after the inception of the

48
Republic; they were, in short, creations of the era rather
than artifacts of the previous system, as were most of the
Weimar parties. Second, both were thoroughly opposed to
the existing system, and both sought by parliamentary and
extra-parliamentary means to further the destruction of that
system. Not infrequently, they were in tacit alliance with
one another (as in the Berlin transportation strike of 1932),
having the overthrow of the Republic as their common goal.

As we noted earlier in discussing the SPD, the German
working-class movement was divided at the end of the First
World War. The Independent Socialists, having broken with
the majority SPD leadership over the so-called war credits
issue, formed their own political party. The Independence
suffered an irreparable loss in the deaths of their most
dynamic theorists and leaders, Karl Liebknecht and
Rosa Luxemburg. By 1922, most of the dissidents had grown
disillusioned with the independent role, and major ideologi—
cal differences had arisen as a result of the rise of Soviet
Russia as the center and leader of the Communist movement.

In September, 1922, most of the Independents returned to the
main SPD organization.

A small group, however, refused to take this step and
left the party to join the small German Communist Party (KPD).
In a critical essay on the early history of the German
Communist movement, Ossip K. Flechtheim observes that only

after the Halle party conference of 1920, when the left wing

49
of the Independents merged with the KPD, can one speak of a
mass Communist Party in Germany.

A premature uprising in central Germany resulted from
the strong influence exerted by the left wing of the party.
The March, 1921, disaster, Flechtheim observes, resulted
in a swing to the right. This policy, regardless of how
sensible it may have been after a major political reversal,
seems somewhat shortsighted when we consider that it was
maintained throughout the inflation of 1923, when there
was widespread discontent within the proletariat and lower
middle class.

By 1925, the KPD had come under the domination of Moscow,
and the left wing was purged from the party. In the middle
period of the Weimar Republic (1924 to 1929), the KPD
played a minor role. Hermann Weber has suggested certain
reasons for this:

There existed no prospect of attaining revolu-
tionary goals in the face of such economic
growth. As an active political role tailored
to the specifications of parliamentary govern—
ment was no the KPD's style, the party
stagnated.l

If the role of the KPD in the interim years was less
than active, the party certainly compensated for it in the
dosing years of the Republic, for, with a membership of
300,000 and electoral support of nearly 6 million, the

Communists could claim to be the third largest party in

Germany in November, 1932. But the processes of purge,

50
Stalinization, and stagnation had taken their toll: the
leadership in the Central Committee was comprised of those
German Communists completely subservient to the Russians.
The dogmatic attitude of the KPD leadership and their stead-
fast rejection of any cooperation with the SPD in the face
of rising Nazi strength and aggressiveness made any united
proletarian opposition to Hitler impossible. To the
Communists, the Social Democrats were the "social fascist"
enemy, with whom cooperation was taboo. That is not to say
that there were no calls for a "united front" from the KPD,
but these were for the purpose of dividing the SPD, and
nothing came of them.19
Although the KPD presented a definite challenge to the
SPD's claim to leadership of the German proletariat, the
party never won over the majority of the working class.
Only in periods of intense economic hardship did the German
workers defect in significant numbers from the SPD. If
the KPD had any permanent support from any specific sectors
of the German working class, it was from certain mining
communities, some organized urban laborers, and the unemployed.
Flechtheim concludes his essay with this observation:

The KPD was never strong enough organizationally

and politically to threaten the Weimar Republic.

It was always only just strong enough to serve as

a scapegoat for rightist and Nazi attacks on the

democratic order. By contrast, the NSDAP succeeded,

through a skillful united-front policy, in driving

back the left and dealing the Republic its death-

blow under the pretext of fighting Communism. This

is the most important lesson to be drawn from the

otherwise not very edifying history of the KPD.
It is a lesson worth pondering.

51

The Nazis could scarcely be termed a party in any
national sense in 1919. When the National Socialist Party
(NSDAP) was formed by Anton Drexler after the war, it was
more a workingman's anti-Semitic debating society than a
political party. In the reactionary climate of post-war
Munich, the party received much of its impetus. The
career of Adolf Hitler has been extensively described
elsewhere, as has the development of the NSDAP under his
leadership. One fact deserves emphasis, however° the Nazis
received national attention only with the abortive putsch
of 1923 and the subsequent trial and imprisonment of Hitler.
Prior to that, the party had been confined to Bavaria, one
of a myriad of right-wing organizations in a fairly solid
Catholic Zentrum area.

With such an origin, the electoral successes of the
NSDAP first occurred in an unlikely area-~northern Germany,
especially Schleswig-Holstein. It was predominantly in
agricultural districts, with considerable numbers of small
landowners, that the Nazis found their voting support

21 In the national elections of May, 1924, they

initially.
gained 6.6 per cent of the total vote, but dropped to 3.3
per cent in December of the same year. The underlying reason
for such fluctuation is fairly evident: measures taken to
stabilize the currency and the general economic situation had

not had their full impact in May; they had by the time of

the Reichstag elections in December, 1924.

 

52

After the ban on the Nazis, several front organizations
arose in various parts of Germany. A split within the move-
ment threatened its unity during this phase. While Hitler
recognized that only a policy of legal opposition would be
tolerated by the state and that no overtly anti—Church
policy could be pursued in Bavaria, other members, such as
Ludendorff and Gregor Strasser, were attempting to found a
nationalistic, anti-Marxist, anti-Church movement. In
February, 1925, Ludendorff surrendered the leadership of
the National Socialist Party, and Hitler announced the rebirth
of the movement. Divisions continued within the NSDAP, how-
ever, particularly between the faction which desired a
revolutionist, anti-capitalistic policy (Strasser, Roehm, and
the SA) and the Rightists, who favored a policy of legalistic
opposition until such time as the movement would be strong
enough to seize power by other methods. To that end, the
NSDAP joined an oppositional coalition with the DNVP under
Hugenberg (the Harzburg Front), whose contacts among German
business proved particularly helpful to the flagging finances
of the Nazis.

The period of relative economic prosperity from 1924 to
1929 did little to better the electoral situation for the
National Socialists. They were, however, active in the rural
areas among farmers and agricultural workers, groups whom
the new prosperity benefited little. With the sharp decline

in economic conditions after 1929, the Nazis bettered their

53
electoral fortunes, polling 18.3 per cent of the total vote

in the election of 1930. Despite repeated bans on their
activities in Prussia and elsewhere, the Nazis moved into
the urban areas and pressed the attack, first against the
Communists, then against the SPD. Street battles, assasi—
nations, and other forms of political violence became common.

It is doubtful that the Nazis achieved any following
from former Social Democratic supporters. In a detailed
ecological analysis of the electoral results in 1932,

Pratt found that the larger the city, the smaller the Nazi
vote. As most of the urban vote went to the SPD and KPD,
it seems likely that if there was any defection in the SPD
vote, it benefited the Communists, not the Nazis.

Of the seven largest “erman states, Thuringia came
earliest under Nazi influence. In the Iagdtag elections of
1929, the NSDAP polled 11.3 per cent of the vote. As a
member of the coalition government, the NSDAP controlled the

interior ministry. This led to a conflict with the Reich

 

government in March, 1930, when the minister of the interior,
Frick, appointed Nazis to positions in the state police and
instituted racist prayers in the public schools of Thuringia.
To combat the infiltration of the police by the Nazis, Wirth,
the minister of the interior for the Reich, refused to con-
tinue the subsidy for the Thuringian police.

Yet another Land to come heavily under the sway of the

National Socialists at an early stage was Saxony. As

54

Halperin reports:

The extent to which Hitlerism was profiting politi-

cally by the current depression was convincingly

demonstrated by the outcome of the Saxon elections

of June 22, 1930. The Nazis increased their

representation in the diet from five to fourteen

and thus became the second strongest party in

Saxony, being surpassed only by the Social Democrats,

who remained far ahead of all rivals with thirty-

two seats. Hitler's men polled 20,000 more votes

than the Communists, who gained only one additional

seat to raise their total to thirteen. The Nazis

scored at the expense of the Nationalist and

Peoples' parties, both of which lost heavily.

Another serious casualty was the Democratic

party.22
Where the Nazis did join coalition governments, their most likely
coalition partners were the DNVP and peasant splinter parties.
Prior to 1932, the Nazis exerted little influence at the Land
level. Their success at the national level came as a conse-
quence of a number of reversals for parliamentary democracy
at the national level, and later at the Prussian state level.

An explanation of the collapse of the Weimar Republic is

beyond the scope of this study. Certainly governmental
instability was a major cause of the collapse, but other
features of the system should also be taken into account:
the personalities of the political leadership, the emergency
powers of the President, and the widespread economic problems
of the period. All of these have been considered elsewhere;
for the purposes of this study, the governmental instability
of the national cabinets concerns us not as an explanation

for the systemic collapse, but as a phenomenon to be studied

in itself.

55

The rise of radical parties such as the KPD and the
Nazis is an important consideration in any explanation of
governmental instability. But it should not be viewed as
the sole explanation; radical parties of the type discussed
above seem to occur most frequently in those political systems
having multiparty pluralistic systems. The sequence of
evens leading to the rise of radical parties is of import,
for it has a bearing on the question of how expansion of the
party system affects governmental stability.

The dramatic loss of democratic support is one of the
most striking characteristics of the late Weimar Republic.
That it should occur in a period of high political polariza-
tion and mass mobilization of social forces within Germany
has led to speculation as to the connection between the two
phenomena: was the loss of democratic substance due to the
new voters, who came into the electorate at a time when older,
more experienced voters were growing dissatisfied with the
political performance of the center and liberal bourgeois
parties, or did these older voters take the lead in shifting
their support to the radical Right? Lipset and Shively
evidently favor a qualified version of the latter theory,
while others, such as O'Lessker, have sided with the former.23

The flight into radicalism of a considerable portion of
the electorate is always difficult to explain, and the nature
of the radical movement seldom reveals much about the motives

that compel individuals to change allegiances or become

56
mobilized for political causes. Certainly no revelations
will be forthcoming from this particular study regarding the
social background or underlying motives of radical voters.
A number of theories have been generated to explain the
phenomenon of radical voting. Some have focused on the
nature of the electoral system; others have singled out the
nature of voter attachments to particular types of parties
or the particularities of the independent voter in the
electorate. MacRae, for example, in his study of the French
"surge" movements in the Fourth Republic, observed the
following:

Even though proportional representation
permitted the easy reflection in the vote

of sudden changes of opinion expressed through
new parties, it did not alone account for the
peculiarities of these "surge" movements. In
contrast to the "independent" vote that might
strengthen an established party on matters of
general policy, the shifts in the vote went
either to new movements breaking completely
with the "system," or to expressions of very
specific interests; Poujadism in its career
from pressure group to party, combined both but
significantly failed to become absorbed in an
established party. Thus the Fourth Republic
did indeed know fluctuation of the vote; but
the fluctuation was not such to permit a systema-
tic alteration of government policy within a
stable set of "rules of the game." 4

"surge" move-

The contrast between the fate of the French
ments and the Nazi movement in Weimar Germany is marked.
MacRae concludes that the nature of social cleavages in
French society, coupled with the effects of a proportional

representation system and a party system that well reflected

those cleavages, accounts for the relatively uninfluential

role of Poujadism and other such movements.

57

The most interesting aspect of the Fourth Republic‘s
radical movements is that they failed to have any true
impact on governmental stability; instability resulted from
sources other than electoral polarization, expansion of the
party system, or other electoral influences. MacRae
concludes that a major source of cabinet instability in
France was the existence of multiple parties with irreconcilable
programs.25 The absence of stable majorities in the National
Assembly led to a situation of shifting majorities and
unstable governments. MacRae sees the Scandinavian-type
party system as more stable, but the adoption of such a
party system would require a less heated political climate
than that which existed in France or, for that matter, in
Weimar Germany.

The ideological barriers to coalition formation were
considerable in the Weimar Republic. A set of coalition
possibilities emerged out of the frequent shifting in the

Reichstag. This set was largely the model for those coali-

 

tion governments that resulted at the land level. In some
instances, certain types of coalition forms were attempted
at the state level prior to any negotiations for such a
coalition at the national level. This was the case with the
cooperation between the SPD and the DVP in Prussia, which
eventually led to the Grand Coalition in 1923.

In describing the Weimar party system we have not dwelt

upon the dynamics of this system, although we have alluded

58

to them from time to time. It should be noted that in the
course of the thirteen-year history of the Weimar Republic,
the party system expanded considerably, first to the left

and later to the right. Such changes in the number of
parties, as well as changed in the proportionality of

seats controlled by the various parties, seem to have affected
the electoral strategies and coalition tactics to the system's
major parties, e.g., those most likely to be members of

governments and those which remained in the opposition.

Coalition Patterns

 

Some idea of the span of coalitions across the party
system can be gathered from Figure 1. Figure 1 shows the
national level party system and coalition alternatives
within that system; it is not a complete representation of
the party system at any one time, although several periods
did find all of these parties represented in the national
parliament. Figure l is intended as a model of the typical
Weimar party system, suggestive of the ideological distances
separating the various parties and coalitions discussed in
the preceding sections. From this it is apparent that
Burnham's typology of the Weimar party system does not always
parallel the ideological dimensions of that system.

Figure 1 illustrates the ideological limits within
various coalition combinations. It is a cardinal assumption
of coalition politics that actors form coalitions with other
actors having similar issue preferences and ideological

orientations. In light of this, it is interesting to note

59
the gradual shift to the right in the spread of the coali-

tions along the unidimensional party scale over the thirteen—
year period of the Republic. The principal reason for this
spread--and the concomitant inclusion of more parties to the
right--was the gradual loss of electoral support for the
parties of the Weimar Coalition and the need to include
bourgeois parties in the ruling coalition in order to obtain
parliamentary support (or at least tolerance) for the govern-
ment.

From Figure 1 it is apparent that the Zentrum party
played a major role in the governing coalitions throughout
the period. It was the nucleus of the Weimar Coalition, as
well as of the many interim Zentrum-Bourgeois coalitions.

It had a pivotal role in the Grand Coalition and formed the

left flank of the Coalition of the Right.

6O

Coalition of the Right

 

I I

(KPD-USPD) uSPD-DDP-Zentrum—BVP-Dt. Hann.2-Wirt.3-DVPFDNVP-(NSDAP)

in opposi- _ j in opposi-
tion Wei ar Coalition ” tion until
1932

 

Center Bourgeois Coalition
(excluding splinter parties)

 

 

 

Grand Coalition
(SPD, DDP, Zentrum, DVP)

 

Figure 1

The Coalition Alternatives within
the Weimar Party System:

1919-1932

 

1The Independent Socialist Party (USPD) dissolved in
1922; its members entered either the KPD or the SPD.

2The German Hannovarian Party (Deutsche Hannovarische
Partei) was a splinter party with only sectionalist support.
It participated in Land, but not national level governmental
coalitions.

3The Economic Party (Wirtschaftspartei) was a splinter
party of the right, acting as a representative of small
business. Only in the Bruening Cabinet (1931-32) was it
considered for inclusion in a national government.

Note: These alternative coalitions are those which actually
arose during the 1919-1931 period. No attempt is
made to represent other, hypothetical coalitions.

61

Such versatility warrants some explanation. The Zentrum
had no competitor in representing Roman Catholic interests
in Germany (except for the BVP, which was confined to Bavaria).
Its umbrella party nature enabled it to find ministerial
candidates in its ranks acceptable to most potential coali-
tion partners. Its electoral strength fluctuated least
among the Weimar parties, providing a security not enjoyed by
the SPD or the bourgeois parties. Finally, the Zentrum's
financial support base--the German Church--was quite secure,
adding another source of security to the party leadership.

The role of the Zentrum was crucial in the coalition
formation process, but it varied in importance and scope
with the particular type of coalition in question. As
shown in Figure 1, the four patterns of coalition varied in
their scope. What follows is a description of the tensions

present in each of these four forms.

The Weimar Coalition: SPD, DDP, and Zentrum

The Weimar Coalition was the basis of the first parlia-
mentary regime in Germany following the collapse of the
Imperial regime. This strange combination of Socialism,
liberal business, and Catholicism functioned effectively for
one basic reason: all of these interests had been in the
opposition under the old Imperial system. They were
essentially "outsiders" whose presence in the Reichstag had
been so much parliamentary window-dressing for what was at

heart a non-democratic system of monarchical absolutism.

62

The Social Democrats were the leaders of the Weimar
Coalition. In the early days of the Republic, the SPD held
the Chancellorship and several other important offices.
Next in importance was the Zentrum, which roughly matched
the SPD in electoral strength. The Democrats were not a
well-established party, and their electoral strength never
approached that of their two confessional partners. The
DDP did, however, contain men of international stature, such
as Rathenau, whose diplomatic skills would prove useful in
the negotiations with the victorious Allied powers.

One tension point in this coalition was organized
labor. The SPD's trade union was in competition with the
Christian unions of the Zentrum. Here the confessional
nature of the two parties worked to their disadvantage, as
their affiliated organizations conflicted in their recruit-
ment of workers. But years of cooperation as oppositional
parties did much to limit the seriousness of such differences.
This fund of common experience explains the absence of
serious ideological disputes between the parties, given the
Marxian perspective of the SPD and the Catholicism of the
Zentrum. As for the DDP, the SPD leadership considered them
to be ”progressive" and hence acceptable as coalition part-
ners. This was in contrast to the SPD's opinion of the DVP,
which was held to be the party of unreconstructed capitalism,
monarchist in its political sympathies.

The greatest obstacle for the Weimar Coalition--and the

ultimate reason for its failure--was its inability to retain

63
the confidence of the electorate in the face of rising
inflation, Allied reparations demands, and increasing shortages
of foodstuffs. The radical left (the USPD and later the
Communists) exploited the SPD's cooperation with "bourgeois"
parties for its own electoral ends. This tactic forced the
SPD to withdraw from the Weimar Coalition in order to strengthen

its base support among the proletariat.

The Bourgeois Center Coalition: DDP, Zentrum, and DVP

The problem of minority government was undoubtedly the
major symptom of instability within the Republic. With
the withdrawal of the Social Democrats, rule by an interim
government would be necessary until such time as a party
commanding sufficient parliamentary strength could be induced
to join the coalition.

The most likely alternative coalition partner to the
SPD was the German Peoples Party (DVP). This party was
primarily the creation of one man, Gustav Stresemann. It
was formed out of what remained of the old National Liberal
Party following the demise of the Empire in 1918. Principally
a party of industry, the DVP, unlike the DDP, did not imme-
diately embrace the new republican system; Stresemann remained
a monarchist and thus was opposed to the Constitution until
well into the Weimar period, and his ultimate conversion to
republicanism was one of the head rather than of the heart.

Undoubtedly the greatest tension in a coalition of the

DDP, Zentrum, and DVP was that of the differences between

64

the parties at the peripheries of this coalition. The DDP

and DVP were in a sense in competition for the identical
electoral sector, the Protestant bourgeoisie. They both
sought the financial support of industry, another source of
contention between the two parties. Moreover, the personality
differences between Stresemann and the DDP leadership were
formidable. All of these factors combined to make the

Bourgeois Center Coalition a tenuous combination.

The Grand Coalition: SPD, DDP, Zentrum, and DVP

This coalition represents the often sought but seldom
achieved goal of the Weimar party leadership. The number of
instances in which this coalition obtained was few. Usually
it came about as the result of a major crisis within the
system, one in which ideological and personal differences
were set aside in order to present a united front in the
face of an external threat to national security. This would
at least explain the two Stresemann cabinets in which this
coalition came about. In the case of the second Mueller
cabinet (June 28, 1928 to March 27, 1930), the rationale is
less clear.

The sources of tension within the Grand Coalition are
obvious. The coalition span, if we apply only an ordinal
scale, was greater than in the previous coalitions considered.
Moreover, the ideological conflict implicit in a coalition
comprised of a Socialist party, two bourgeois parties with

industrial backing (especially the DVP, which the SPD

65
leadership viewed as the representative of the proletarian
class enemy), and a Catholic party with both industrial labor
and agrarian interests would seem to present a highly
unwieldy partnership.

In these circumstances, Sartori's (1966) argument con-
cerning the centrifugal effects of a polarized pluralist
system seems most relevant: the centrifugal drives created
by the presence of extremist parties at the poles of the
party spectrum would militate against a long duration for
the coalition. Both the SPD and the DVP would be under
considerable pressure from the neighboring non-coalition
parties, as in fact was the case. Such drives would augment
and exacerbate those tensions within the coalition to which
I have previously alluded.

In the face of such overwhelming tensions and stresses,
it would seem that the Grand Coalition was a reasonable
undertaking only when the survival of the system itself was
in question, a situation that called for submergence of party
differences in the national interest. M. Olson has aptly
described such a situation as one of "collective goods with

unlimited domain,’ and the complement of that term,

l

"collective goods with limited domain,’ might be applied to

those periods when normal politics, or the more common

26

pattern of party competition, was the state of affairs.

The Coalition of the Right: Zentrum, DVP, and DNVP

This last of the feasible coalition alternatives in the

Weimar Republic is perhaps the most interesting from a

66
theoretical perspective. If Axelrod's (1970) notion of
conflict of interest in coalitions is applied to this
coalition structure, it is apparent that in order for the
coalition to include the German Nationalist Party (DNVP), the
coalition span must pass over numerous small splinter parties
which are to be found between the DVP and the DNVP. This,
Axelrod would contend, increases the level of conflict
within the coalition, for conflict of interest is measured
in terms of distance spanned by the coalition spread
(assuming an ordinal scale). Clearly, the Coalition of the
Right would contain a high degree of conflict by virtue of
its spread across the party spectrum.

In View of this, it is significant that the German
Nationalist Party had been a formal member of a ruling coa-
lition on only two occasions prior to the first Bruening
cabinet. This suggests that the difficulties of bringing
in the DNVP as a coalition partner were readily apparent to
the center and bourgeois party leadership. It ignores,
however, the intransigence within the DNVP itself, and its
extremely anti-republican attitude.

The DNVP was primarily the party of the landed aristo-
cracy, and it favored a return to the monarchy of the ancien
regime. The party gained support from middle-class voters
who were disillusioned with the Weimar Coalition and its
inefficacy in solving Germany's post-war problems. However,
the DNVP’s paucity of constructive policies, coupled with the

competition of the Nazis for the right—wing extremist vote,

67
led to the party’s eventual electoral decline in the late
Weimar period.

As a potential coalition partner, the DNVP was often
considered as less than desirable, even by former proponents
of the monarchy like Stresemann. The strident militarism
of the DNVP, together with the widespread suspicion that
certain of its members had been involved in the Kapp Putsch
of May, 1920, frequently led to its exclusion from serious
consideration for coalition inclusion. Interestingly, however,
it seems to have appeared more attractive with the increase

of National Socialist support in the electorate.

A Mbdel of the Cabinet Formation Process

 

In the preceding, we have described the four types of
coalitions that were common in the Weimar Republic. We might
mention in passing that we have treated the so-called "non-
partisan" coalitions no differently from the other coalitions
during this period, as most of the cabinet members had party
affiliations and were therefore as partisan as any other
parliamentary-based coalition.

What follows is an attempt to formulate the rules that
governed the cOalition formation process under the Weimar
Republic. The formal rules are, of course, to be found in
the laws governing the nomination process for the selection
of the Chancellor. The President of the Republic normally

selected the leader of the party having the largest

68

representation in the parliament (Reichstag), and this

 

individual then undertook the task of forming a coalition
with other parties that would constitute a majority of the
parliamentary delegates.

The President‘s powers in this process should not be
underestimated, for the initial selection of the party
leader biased the final outcome. The nomination of the SPD
as the coalition leader normally meant that the DNVP would
be excluded. And, furthermore, the President deviated
frequently from the custom of selecting the leader of the

largest party in the Reichstag. This was particularly true

 

under the Presidency of Hindenburg, when decisions affecting
the leadership of the nation seem to have been made by a
small camarilla of the President's associates.

Certain conditions governed the process of cabinet
formation. 'These were the informal rules that a coalition
leader must observe in forming and maintaining the ruling
coalition. Among the most important of these were the
following:

1. Coalitions must, if at all possible, possess a
parliamentary majority.

2. Only major parties would be included in a ruling
coalition.

3. A ruling coalition should include only those parties
that respect the constitution.

4. A vote of no confidence would lead to the resignation

of the cabinet.

69

5. No policies should be produced that directly contra-

dict the interests of the army.

Summary

In exploring the setting of this study, the emphasis has
been on the party system and the various coalition patterns
within that system. The latter were clearly limited on both
the left and the right of the party system by radical parties
denying any commitment to the Constitution. These radical
parties exercised a definite influence on the potential
coalition parties, particularly as recipients of voter
dissidence. Such pressure increased the ideological conflict
across the party system, sometimes forcing moderate parties
to reconsider a commitment to a coalition in which an
interest antagonistic to its perceived electoral or financial
support base was represented. This inhibiting influence was
present throughout the history of the Republic.

Such an influence had obvious consequences for govern-
mental stability. While it is seldom possible to elicit
the real psychological motives for political decisions in a
historical system such as the Weimar Republic, some assessment
of these motives is attempted in Chapter III, using docu-
mentary evidence. This supplements the statistical analysis
of the data, in which the relation between the ideological
diversity within party system, government and opposition,
as well as the parliamentary strength of the parties described

above, and the stability of the governing coalitions are the

major concerns.

7O

NOTES

lFredrich A. Hermans, Europe under Democracy or Anarchy
(South Bend: University of Notre Dame Press, 1940).

2Reinhard Bendix, "Social Stratification and Political
Power," in R. Bendix and S. M. Lipset, eds., Class, Status
and Power (Glence: The Free Press, 1956); S. M} Lipset,
Political Man (Garden City, N.Y.: 1959); K. D. Bracher,
Die Aufloesung der Weimarer Republik: Eine Studie zum
Problem des Machtverfalls in der Demokratie, rev. ed.
(Stuttgart: Ring-Verlag, 1957);Kar1 OILessker, "Who
Voted for Hitler?" A New Look at the Class Basis of
Naziism," American Journal of Sociology, 74 (July, 1968),
63-69; W. P. Shively, "Party Identification, Party Choice
and Voting Stability: The Weimar Case," American Political
Science Review, 65 (December, 1972), 1203-1225.

 

 

 

 

 

 

 

 

3Michael Stuermer Koalition und Opposition in der
Weimarer Republik, 1924-1928(Duesse1dorf: Droste,l967).

"Peter Haungs, Reichspraesident und Parlamentarische
Kabinettsregierung: Eine Studie zum Regierungasystem der
Weimarer Republik in den Jahren 1924—1929‘(Cologne-OpIaden:
Westdeutscher Verlag,l968).

 

 

 

 

5Volker Rittberger, "Revolution and Pseudo-Democratization:
The Formation of the Weimar Republic," in Gabriel A. Almond,
Scott Flanagan and Robert MUndt, eds., Crisis, Choice, and
Change: Historical Studies of Political Develbpment (Boston:
Little, Brown and Company, 1973), pp. 285-391.

 

 

6Abram DeSwaan, Coalition Theories and Cabinet Forma-
tions (San Francisco: Jossey-Bass, Inc., 1973), p. 298.

 

7Walter Dean Burnham, "Political Immunization and
Political Confessionalism: Some Comparative Inquiries,” a
paper presented at the 1970 International Political Science
Association Convention.

8Carl Schorske, German Social Democracy, 1905—1917
(New York: Harper, 1955), p. 32.

 

9Richard N. Hunt, German Social Democracy, 1918-1933
(Chicago: Quadrangle Books, 1964), pp. 43-44.

1O

 

Ibid., p. 45.

11Robert Michels, Political Parties (New York: Dover,
1959), p. 270.

 

71

12Rudolf Morsey, "The Center Party between the Fronts,"
in Conway, ed., The Path to Dictatorship, 1918-1933 (New
York: Anchor, l966),p.69.

l3

 

Lipset, op. cit., p. 138.

1"Die Buergerlichen Partien in Deutschland, Vol. 1
(Leipzig: VEB Bibliographisches Institut, 1968), pp. 302-331.

 

15Alfred Milatz, waehler und wahlen in der Weimarer
Republik (Neuwied: Raiffeissendruecherei, 1968), pp. 91—93.

l6Schnaiberg, op. cit., pp. 732-735; Rudolf Herberle,
From Democracy to Na21sm (Baton Rouge: Louisiana State
University Press,l945), pp. 113—11 , 119.

 

 

 

Stefanie Cameron, "Nazi Ascendancy and the Weimar
Party System," a paper presented at the Midwest Regional
Meeting of the International Studies Association, Toronto,
Canada, 1972, p. 12.

18Hermann Weber, Die wandlung des Deutschen Kommunismus
(Frankfurt am Main: Europaeische Verlagsanstalt,l969),
p. 8 (our translation).

 

19Ossip K. Flechtheim, "The Role of the Communist Party,"
in Conway, ed., The Path to Dictatorship, pp. 89-112.

20

 

Ibid., pp. 109-110.

21Ibid., p. 112.

22S. William Halperin, Germany Tried Democracy (New York:
Norton, 1965), p. 427.

 

23Cameron, op. cit., on the basis of an aggregate analysis
of electoral districts concludes that higher turnout levels
did not significantly contribute to the rising Nazi vote:
"Instead we find that Nazi gains at this time show strong
correlation with losses of the conservative DNVP and to a
lesser extent the more liberal DVP,” p. 15.

24Duncan MacRae Jr., Parliament, Parties, and Society
in France, 1946-1958 (NewYork: St. Martin‘s Press, 1967),

 

 

251bid., pp. 326-327.

Mancur Olson, Jr., The Logic of Collective Action:
Public Goods and the Theory of Groups (Cambridge, Mass.:
Harvard University Press, 1965). The union of the SPD and
DVP during the first and second Stresemann cabinets exempli-
fies this type of behavior. Here Olson's "collective goods
with unlimited domain" was perceived by the parties as the
survival of the political system.

 

 

Chapter 3

DEVELOPING A MODEL OF GOVERNMENTAL STABILITY
FOR WEIMAR GERMANY
Introduction

The measures developed in Chapter 1 are representative
of the variables whose relationships with governmental
stability (duration) we are presently to examine.
Ideological diversity in the government (hereafter referred
to as IDG) measures the degree of diversity resulting from
both the fractionalization and ordinal disagreement among
the governmental coalition parties. It is a composite
measure, obtained by averaging the two measures of
fractionalization and ordinal disagreement for each of the
seventeen coalitions in the sample.

Post maldistribution (PM) measures the proportionality
of the distribution of cabinet posts within each of the coa-
lition governments. It is calculated by summing the differences
between each party's contribution of parliamentary support
to the governing coalition and its proportion of the posts.

The total number of major parties (selected according
to the criteria specified in Chapter 1) in the parliament is
a crude but expressive indicator of fractionalization of

the party system.
72

73

Another variable of theoretical interest is the difference
from minimum-winning size (DMW) of each of the governing
coalitions. As previously mentioned, this measure takes as
its inspiration Riker's size principle. Here we are interested
in the effect that divergences from minimum-winning size
have upon governmental duration.

Radical party growth (RP) measures the degree of change
in the electoral support for all radical parties during
each governing coalition‘s term in office. The predicted
direction of the relation between duration and radical party
growth is, of course, negative.

Finally, the duration (DUR) of the government (in months)
provides the measure of the dependent variable-~governmental
stability. Duration is a linear variable measured as the
number of months from its installation to its resignation.

In this chapter, we outline the process of governmental
destabilization in the Weimar Republic. In order to test
some popular conceptions about the causes of governmental
instability at the national level of the Weimar system, the
following six hypotheses are considered:

1. Governmental duration declines with increases in
radical party electoral support.

2. Governmental duration is negatively related to
ideological diversity in the governmental coalition.

3. Governmental duration is inversely related to post
maldistribution.

4. Increases in the number of parties in the national
legislature contribute to governmental instability.

74

5. As a coalition government's parliamentary support
deviates from minimum-winning size, its duration
declines.

6. Declining governmental duration is produced by a
combination of some or all of these five variables.

The Six hypotheses above constitute the core of the
study. Each, is, in its own right, a possible explanation
of governmental instability. Each is a political variable;
no attempt is made to measure the effects of economic,
societal, historical or other possible variables on govern-
mental duration. Our purpose here is to determine how much
of the Weimar instability can be explained in terms of the

regular electoral and party political variables.

Data Analysis

 

The strategy of data analysis is simple. We study first
the simple correlations among the five variables. After
ascertaining which relations exhibit significant correlation
coefficients, we conduct tests for spuriousness among these
relations.1 At this point, variables exhibiting scant
relationship to the other study variables are deleted in
order to construct a simpler model of the destabilization
process. After spuriousness is tested by means of partial
correlation analysis, path coefficients are calculated, and
the provisional model is then cross-checked with path
estimation techniques.2

The purpose of this stage of the analysis is to clarify

some of the relations that Groennings has identified in his

coalition maintenance model. The variables have been

75
selected on the basis of their representativeness of the
underlying conceptual dimension and the absence of multi-
collinearity among them. By limiting the number of variables,
we have, in effect, sacrificed the virtues of the over-
identified model for the compactness and economy of a smaller

model.

Correlational Analysis

 

Examination of the zero-order correlation coefficients
in Table 1 reveals several relations of interest to our study.
First, the dependent variable, governmental duration (DUR),
is significantly related to two of the five other variables:
ideological diversity in the government (IDG) and the number
of parties (NUM). One supposition that might be tentatively
made on the basis of the simple correlation coefficients is
that, while ideological diversity is negatively related to
duration, the number of parties is positively related. Of
course, further analysis is required in both instances
before such a supposition can be entertained.

Turning to the ideological diversity-duration relation,
we note that the -.415 coefficient is significant at the
.05 level. Performing a partial correlational analysis of
this relation and controlling for the number of parties
(NUM), we find that the coefficient diminishes only to
—.37. Another partial of the NUM-DUR relation with IDG as
the variable controlled produced a coefficient of .34. The

strong NUM¥DUR relation suggests that the expansion of the

party system occurred concomitant with periods of durable cabinets.

76
Table 2 presents the results of a systematic partial
correlational analysis of the six variables. The results
of this partial correlation point to two variables as having
a consistently strong relationship with duration: ideolo-
gical diversity in the government (IDG) and the number of
parties (NUM). In the case of IDG, the relation is a
negative one, with the post maldistribution (PM) variable
having the strongest impact (-.34 compared to —.41) on the
IDG-DUR correlation. The PM-DUR correlation of -.27 is
only significant at the 0.15 level of significance (with
sixteen degrees of freedom). The IDG and NUM variables also
limit the PM—DUR relation considerably. The relation
of the number of parties (NUM) with duration is strongly

affected by ideological diversity, as by a comparison

of the NUM-DUR coefficients in Tables 1 and 2.

 

 

 

Table 1
Simple Correlations among Six variables
N=l7
variable IDG . PM NUM DMW RP DUR
IDG 1.000.
PM .415 1.000
NUM -.220 -.163 1.000
DMW .477 .066 -.545 1.000
RP -.095 -.039 .209 .098 1.000
DUR -.415 -.270 .410 -.l72 .052 1.000

 

 

III A

77

 

 

 

Table 2
Partial Correlations of Duration with Five Variables
N217

Relation Variable Controlled

IDG PM NUM DMW RP
IDG-DUR --- -.346 -.365 -.384 -.413
PM-DUR -.118 --- -.227 -.264 -.269
NUM-DUR .359 .385 --- .383 .409
DMW-DUR .032 -.161 .067 --- -.179
RP-DUR .010 .042 .030 .020 ---

 

The number of parties-governmental duration relationship
is not in the direction predicted by H.4, but is rather
strongly positive. This result suggests that periods of
enduring coalitions were coincidental with those electoral
periods in which the number of parties in the Reichstag
increased.

Radical party growth/decline (RP) has the weakest
relation with duration of all five of the independent variables.
Only with number of parties (NUM) does it show a relation of
any size, and that (.209) correlation coefficient is not
significant at the .05 level. Partial correlation does not
alter this result.

The absence of any significant relation between radical
party growth and governmental instability is a somewhat

surprising finding. We can only note that theoretical

78
speculations about the destabilizing influences of radical
parties at the extremes of the multi-party system receive
little support from our statistical analysis. On the basis
of our findings of no significant relation at the .05
significance level we reject H.l.

Before conducting the correlation analysis, we were
working on the assumption that all or some (or none) of the
five variables might be significantly related to duration.

On the basis of the analysis, a clearer picture of relation-
ship between the variables and duration emerges. Two variables
in particular--ideological diversity and the number of parties--
have been shown to be related to governmental duration, while
others--post maldistribution (PM), difference from minimum
winning size (DMW), and radical party growth/decline (RP)--
exhibit little or no relationship with governmental duration.

If ideological diversity (IDG) is the strongest predictor
of declines in duration, it is possible that it is merely the
end result of a chain of relationships leading to such
declines. That is, number of parties, difference from
minimum-winning size, and post maldistribution may stand in
some undefined relation to ideological diversity, which, in
turn, is related to duration.3

In Figure 2, all possible causal paths to DUR (based on
the results of the simple correlations) are presented.

Figure 2 represents a non-recursive model, a series of one-way

79

relationships with one variable (number of parties) as the
initial variable in the chain."

First let us consider the direct relation between the
number of parties (NUM) variable and ideological disagree-
ment in the government. The zero—order correlation of -.22
was obtained for the NUM—IDG relation. When the difference
from minimum-winning size (DMW) variable is entered into the
equation, however, the coefficient becomes .05. Thus, as

weak as the relation between number of parties and duration

-.27

  
  
 

 

 

 

 

 

 

PM
-.16/
NUM -.22
~90 .549 4:.
.09
R: DMW
I .09
.05

 

Figure 2
various Logical Causal Paths between the
Six Variables, Including Correlation
Coefficients
is, it is even weaker when the DMW variable is controlled.
This suggests that the difference from minimum-winning size
(DMW) variable interprets the relationship between number of
parties (NUM) and ideological diversity in government (IDG).

The other alternative would be post maldistribution (PM)

80

as the intervening variable. Controlling for PM, we find
that the NUM-IDG correlation is -.l7. Any assumption of a
direct link between number of parties (NUM) and ideological
diversity (IDG) can be abandoned in the light of these
results, all of them not significant at the .05 level.
Parenthetically, we make note of DMW as the intervening
variable.

The next possibility to be considered is a three-step
linkage of number of parties (NUM), post maldistribution (PM),
and difference from minimum-winning size (DMW). First,
controlling for NUM, we calculate the partial correlation
for DMW-PM. The result is that the simple correlation of
.06 is reduced to -.02. Next, as an additional precautionary
check, we calculate a partial correlation for PM—DMW controlling
for IDG. This produces a coefficient of -.17. On the basis
of these results, any linkage between post maldistribution
(PM) and difference-from-minimum-winning size (DMW) of the
ruling coalition can be ruled out.

Continuing the partial correlation analysis, we find that
the relation between post maldistribution (PM) and ideological
diversity (IDG) does not noticeably diminish when either DMW
or NUM, or both, is controlled (.44; .40; .45). Introducing
a control for PM in the calculation of the DMW-IDG correla-
tion only increases the coefficient of this relation to .50.

The role of radical party growth/decline (RP) in the

governmental destabilization process does not appear great

81

on the basis of the simple correlations. Partial correla-
tional analysis provides even less evidence of radical party
growth/decline being a significant variable in the process.
The only variable in our model with which radical party
growth/decline is even moderately related is number of parties
(.21). Otherwise, RP is correlated weakly with DMW (.10) and
IDG (.10). The only logical causal paths in which radical
parties could play a role are (1) NUM-RP-DMW-IDG-DUR, and

(2) NUM-RP-IDG-DUR. In the first case, controlling for radi-
cal party growth in the NUM-DMW relation only reduces the
initial Simple correlation from -.59 to -.58. In the second
case, the NUM-IDG relationship is reduced from .22 to .20,

a rather insignificant change. Thus, radical party growth/
decline (RP) does not indirectly affect the duration of
governments, at least in terms of the range of independent
variables considered in this study.

Returning to the original model, we find that two
variables contribute to the destabilization process, as seen
from the preceding analysis. Post maldistribution (PM) is
related to ideological diversity (IDG), while the difference
from minimum-winning size (DMW) is also strongly related,
and in addition acts as an intervening variable in the
NUM-IDG relation. It is through ideological diversity (IDG)
that these variables are indirectly linked to governmental
duration (DUR).

The NUM-DUR relationship holds up very well under partial
correlation analysis. But as the direction of this relation

is positive, H.4 must be rejected.5

82

By interpreting Figure 3, we trace the process of
governmental destabilization at the national level in the
Weimar Republic. As the number of parties in the system
increased, governmental coalitions tended to decline below
minimum-winning size. The majority of the new parties
were splinter parties whose tacit support could be obtained
by the coalition leaders in return for policy rewards. Thus
the size of governmental coalitions might be reduced below
minimum-winning specifications without necessarily attenuating
a government's stability (duration).6 This accounts for the
strong direct relation (+.4l) of number of parties (NUM) and
duration (DUR). When coalition governments were formed of
greater—than-winning size they contained high ideological
diversity (IDG) which directly contributed to declining
duration (DUR). An additional factor contributing to
ideological diversity (and indirectly to destabilization)

was post maldistribution (PM).

 

.41
PM
\\\\‘ii\
. -.41
NUM IDC DUR
DMW
Figure 3

Results of the Tests of Alternative Models
of Governmental Destabilization

83
Recheck of the Model

 

Figure 4 presents the model of governmental destabili-
zation as derived from tests for spuriousness following the
Simon-Blalock technique. This procedure places heavy emphasis
on deleting arrows from the model by systematic tests for
spuriousness, $42,, by testing for zero values when a third
variable is controlled. Obviously, such a technique involves
extensive use of partial correlational analysis.

A partial correlation, however, is a measure of the
amount of variation explained by one variable after the
others have explained the maximum amount. It does not
measure the amount of change in the dependent variable
produced by a standardized change in one of the independent
variables when the others are controlled.7

Beta weights are adjusted partial slopes representing
standardized b's. In the recent literature on causal
analysis, these beta weights (or path coefficients, as they
are termed when used in recursive models) are used in the
estimation of the parameters of a causal model. Stokes,
for example, illustrates this procedure with a study of the
forces underlying a Congressman‘s voting behavior. He uses
the combined path coefficients of his model to estimate the

importance of combined pathways of influence.

84

 

 

PM
NUM\(H1) O3 IDG (4) ° DUR (5)
-.59 .58
DMW (2)
Figure 4

Alternative Models of Governmental Destabilization
with Path Coefficients

What Stokes proposes is an "inverse" use of path analysis,
in which the researcher begins with the observed correlations
within the system and deduces what the "settings" of the
variables must be in order to produce the observed data. For
each of the possible traverses between X1 and X4, a linear I
combination equation with a weight equal to the product
of the elementary path coefficients is written. While
Stokes’ example is a four-variable model, there are no
inherent difficulties in applying the same technique to a
five-variable model such as ours.

Figure 4 presents the path coefficients for the model.
The question concerning us is whether the use of path
coefficients alters the result obtained by partial correla-
tional analysis. Five alternative paths are posited:

(1) from duration (DUR) to ideological diversity (IDG) to
difference from minimum-winning size (DMW) to number of

parties (NUM); (2) from duration to ideological diversity

85

to post maldistribution (PM) to number of parties; (3) from
duration to ideological diversity to difference from minimum-
winning size to number of parties; (4) from duration to
ideological diversity to difference from minimum-winning
size to post maldistribution to number of parties; and (5) from
duration to number of parties.

Table 3 contradicts the Simon-Blalock findings, but we
must take into account that Path 5 is the NUM—DUR relation
that we have previously explained as a consequence of the

coalition strategies pursued by the major parties.

Table 3

Equations and Path Coefficients for the
Causal Model in Figure 4

 

 

 

Path Equation Combined Path Coefficient
1 P5AP42P21 .16

2 P54P43P3l .03

3 P54P42P21 ~00

4 P54P42P23P31 .00

5 P51 .20

86

 

 

 

Table 4
Regression Analysis of Duration
N217
Variables Multiple R R2 R2 Change Beta
IDG .415 .172 .172 -.455
NUM .528 .279 .106 .475
DMW .576 .331 .052 .304

 

The Amount of Variance in Duration Explained by the Model

 

Finally, we turn to the question of how much of the
variance in the dependent variable is accounted for by the
model of the governmental destabilization process. Table 4
presents the results of a multiple regression analysis of
the national level data. The r squares‘ column shows that
approximately thirty-three per cent of the variance of
governmental duration is explained by the three variables
of our model.

Post maldistribution (PM) was dropped from the equation
due to its low tolerance level. The beta index for PM
is -.016. In a separate equation including radical party
growth/decline (RP), it was found that this variable had a
beta weight of -.140, bearing out the previous pattern of
this variable's insignificance in the destabilization process

model we have tested.

87

Conclusions

 

0f the conclusions that we might draw from the preceding
analysis, the most salient is that governmental instability
is partially the result of a combination of variables, the
most important being ideological diversity. While it is risky
to generalize this pattern beyond the confines of this study,
it would appear, on the basis of the evidence, that ideologi-
cal diversity is the one independent variable in a signifi-
cantly negative relation with duration.

What is remarkable is the seemingly negligible role of
radical party growth/decline in the destabilization process.
Whatever the role of radical party growth in other multiparty
systems, it seems that in the Weimar Republic, from 1919 to
1931, the growth of radical parties did not directly affect
the stability of governments. It is possible, of course,
that radical party growth affected duration through some
intervening variable not included in the initial set of
variables. On the basis of the statistical analysis, however,
Hypothesis 1, that governmental duration declines with
increases in radical party electoral support, must be
rejected.

That governmental duration is negatively related to
ideological diversity in the governmental coalition (Hypo-
thesis 2) can be affirmed from our analysis. None of the
tests for spuriousness affected the strong correlation of
ideological diversity with duration. Ideological diversity

further accounted for half of the total variance explained

88

by all five variables included in the study. Our tests for
spuriousness and path analysis methods revealed IDG to be
the key variable in the multi—path model of the destabili-
zation process, serving as a ”gate" for the other variables.
(excluding number of parties).9

As for the relationship between post maldistribution and
duration (Hypothesis 3), no evidence of any direct relation
was found in the analysis. Partial correlational analysis
failed to support such a hypothesis. Indeed, multiple
controls for IDG, NUM, and DMW reduced the PM-DUR correla-
tion coefficient to -.03. The principal role of PM was
found to be that of a contributing influence on ideological
diversity in the government. Hypothesis 3 can be rejected

with the qualification that it is indirectly related to

 

declining duration.

Increases in the number of parties in the system were
found to have a different effect on the duration of the
national governments from that anticipated. Reductions in
the number of parties indirectly affect the level of ideolo-
gical diversity (through the intervening variable, DMW). Thus,
reductions in the number of parties tended to increase
ideological diversity, while increases resulted in a
lessening of ideological diversity. A logical interpretation
of this finding would be that the political climate of
Weimar Germany encouraged the development of new splinter
parties. These splinter parties, we speculate, served as

replacements for the major parties in periods in which tacit

89
support was necessary for the maintenance of the governmental
coalition. We shall explore this speculation further in
Chapter 4. Hypothesis 4-—that increases in the number of
parties negatively affect governmental duration--is
disconfirmed by the evidence of our analysis.

The difference from minimum-winning size variable was
found to have an indirect relation with duration. The
relation was interpreted by IDG, with the relation between
IDG and DMW being a positive one. We must take into
consideration, of course, the nature of our measure of
ideological diversity, which probably overestimates the
degree of ideological diversity in majority cabinets and
underestimates that of minority cabinets. If our findings
are at all reflective of the actual forces operative within
the governments of Weimar Germany, then we must conclude
that ideological diversity was increased by the greater-
than-winning size characteristics of some of these govern-
ments. However, it is important to note that with a small
sample, like that with which we are working, our statistical
results may reflect the exaggeration of minor variations in
our data. While Hypothesis 5 is confirmed, the indirect
nature of the relation of DMW to governmental duration must
be noted as a qualification of the hypothesis.

Finally, Hypothesis 6 is confirmed by our findings.

The model emerging from the analysis is one in which the
number of parties, difference from minimum-winning size, and

ideological diversity variables proved to be the primary

9O

path to governmental destabilization. 0f the five variables
considered, only radical party growth/decline was found to
have a negligible role in the process. Post maldistribution,
as previously noted, made an independent contribution to
instability through IDG, with that variable acting again in
an intervening fashion.

While all of the variables in our model explain thirty-
three per cent of the variance, it is obvious that a consi-
derable per cent of the variance of governmental duration
(sixty-seven per cent) remains unexplained. Obviously
there remains the problem of intervening variables unaccounted
for in the analysis.10 This is especially evident in the
NUM-DUR relation. Unfortunately, all variables of political
significance have not been identified. Even when identi—
fiable, some of these variables——especially those relating
to the calculation of electoral or coalition strategies—-
are not accessible to the researcher. Moreover, as a number
of environmental variables-~cultural, economic, and foreign
political—-lie outside of the scope of this study, they have
Ilot been treated here. A future task of researchers in the
area of governmental instability in Weimar Germany will be
to integrate the study of foreign and economic variables

with those of a parliamentary-political nature.

91
NOTES
1 . . .
We employ the Simon-Blalack technique in these tests

of spuriousness. See Hubert M. Blalock, Jr., Social
Statistics (New York: McGraw-Hill, 1960), pp. 352-450.

 

2An explanation of this technique is found in Donald
E. Stokes, 'Compound Paths: An Expository Note " American
Journal of Political Science, 18 (February, 1974), 191-214.

 

 

3We will examine the possibility that IDG functions as
a "gate" for the other variables in our model.

LLWe assume recursiveness in the model on the basis of
the sequential order of the events reflected by the variables.

5We originally hypothesized that increases in the party
system result in declining governmental stability (Hypothesis 4).

6This limiting of the number of formal members of a
coalition government should not be viewed as a conscious
strategy of the coalition leaders. Rather, it may well have
been due to the reluctance of major party leaders to enter
the government and risk electoral losses. This trend in turn
necessitated the wooing of small splinter parties in order
to gain their tacit support on parliamentary votes of con-
fidence for the government. This process is discussed more
fully in Chapter 4.

7Blalock, op. cit., p. 453.
8Stokes, op. cit., pp. 200-210.

9Deletion of ideological diversity (IDG) increased the
beta index of NUM to .57. The total variance explained by
all variables (including PM and RP growth/decline) fell to
.227.

10A further consideration is the degree of autocorre-
lation in the data. A test for the presence of autocorre-
lation was performed, using the Durbin-Watson technique.
The test statistic of 1.635 was found to be inconclusive
given sixteen degrees of freedom. A check for homoscedas-
ticity also proved inconclusive.

Chapter 4

THE ROLE OF IDEOLOGICAL DIVERSITY IN
NATIONAL CABINET FORMATION:

1919-1931

Introduction

 

The analysis presented in Chapter III strongly points
to ideological diversity in the governing coalition as the
single most important variable affecting the durability of
a government. Ideological diversity was found to be the
most immediate cause of declining governmental duration,
with two independent paths of causality being traced to this
variable.

In this chapter, we review the political history of the
Weimar Republic by focusing on the groups of coalitions that
controlled the government during different periods. This
historical review, which includes description of the episodic
events which might bring about instability, supports the
finding that ideological diversity was the major political
cause of instability during this period.

Using the typology developed in Chapter II, we examine
each group of coalitions, seeking a pattern of governmental
stability or instability characteristic of each. In studying
each of the four coalition types, we focus on the ideological

tensions present both in the legislature and in the coalition

92

93

itself. Alternative sources of instability, such as foreign
diplomatic or military intervention, are also given considera-

tion.

The Weimar Coalition

 

Five of the seventeen coalition governments formed
during the Weimar Republic were of the Weimar type. This
was a combination of the Social Democrats, Catholic Zentrum,
and the Democrats. These parties were committed to the new
republic and had been instrumental in its foundation in
1919.

The first three cabinets of the republic were Weimar
Coalition cabinets. The issues that confronted these coa-
lition governments were primarily foreign policy issues.
Each cabinet was beset by problems related to the end of the
First World War and the Versailles Peace Treaty. This treaty
had yet to be signed when Scheidemann’s cabinet took office
on February 13, 1919.

In the Reichstag, the Weimar Coalition possessed a
substantial majority of seventy-six per cent of the seats.
However, several of the parties had not approved party lists
for the January, 1919, elections. Many political leaders,
both on the right and left, had adopted a "wait and see"
attitude. In some cases, the November armistice had destroyed
party organizations, especially those on the right. Only
with time would some of these be rebuilt, often under new

labels. Moreover, the Reichstag had for a time ceased to be

 

 

94
the center of German political life. The revolutionary forces
unleashed by mutinies in the German navy and other military
units had yet to be brought under control. Within the ranks
of the Social Democrats (SPD), Rosa Luxemburg and Karl
Liebknecht stood in Opposition to the policies of the majority
SPD leadership and were soon to lead an unsuccessful leftist
revolt, the Spartakist Uprising.

The real support of the Weimar Coalition came not from
the Reichstag, but from the arch-conservative institutions,
the Army and the Prussian Civil Service. With the Kaiser
fled and the Allied armies at the frontiers of the Reich, a
government had to be formed that could negotiate a treaty of
peace with the Allies. But the Army and Civil Service did
not want the onus of having sued for peace to fall on any
party associated with the old Imperial order. Therefore, it
suited their purpose to lend support to the Weimar Coalition
parties, the "outsiders" of the Second Reich. Thus, at its
birth, the Weimar Republic was compromised by collaboration
with the forces of the old order.

The major opposition to the Weimar Coalition came from
the Independent Socialists and the conservative German
Nationalists (DNVP). Of the two, the Independents were the
most influential, for they threatened the majority SPD's image
as a working-class party. The Independents criticized the
SPD’s participation in a coalition of bourgeois parties. The
Nationalists were opposed not only to the Weimar Coalition,

but to the entire concept of republicanism, favoring a return

95

to the monarchy. They were also opposed to any effort to make
peace with the Allied powers that involved concessions by
Germany.

The Scheidemann cabinet (February 13, 1919—June 21, 1919)
was installed for the purpose of dealing with the Allies,
specifically on the question of the acceptance or rejection
of the Versailles Peace Treaty. The major points of
contention were the war guilt clause requiring Germany to
accept the entire responsibility for the First World War,
and the demand that Germany hand over certain individuals
for trial by a special tribunal. Both the SPD and the Zentrum
wanted these two clauses stricken, but they would otherwise
approve the treaty. The DDP was almost equally divided over
the treaty. The vote on acceptance of the treaty led to a
stalemate, with seven for, seven against. Having failed in
its major goal, the cabinet resigned on June 10, 1919.

The Scheidemann cabinet fell not as much from ideological
differences as from inability to come to terms with the foreign
political problems confronting Germany. This indecision
probably cost the Weimar Coalition much of its popular support.

The Bauer cabinet (June 21, 1919-March 27, 1920) that
succeeded Scheidemann’s government carried through the signing
arui subsequent negotiations with the Allies. Some initial
(iiffficulties occurred over the distribution of cabinet posts

in ‘the administration. The Democrats demanded three ministerial

96
positions in the new cabinet. Bauer noted in a letter to
his SPD colleagues:
In an extensive and thorough discussion with the
DDP negotiators, it developed that the Democrats
desire three ministerial posts. It was further-
more assumed that the number of ministers would
not exceed twelve, and that of these six would
be held by our party, and three each would go to
the Zentrum and the Democrats.

Bauer went on to remark that the DDP desired the Justice,
Trade, and Interior ministries, most especially the last.
However, BaUer was unwilling to surrender the Interior
portfolio. He observed that the DDP considered the Justice

and Trade ministries to be essentially bureaucratic ministries

(Fachresorts), not necessarily political posts. The Democrats,

 

Bauer continued, compared their share of the posts unfavorably
with that of the Zentrum, which controlled the Finance and
Postal ministries.

In View of our conclusions in Chapter III, it is interes-
ting to note the role of post maldistribution as an issue in
the negotiations leading to the formation of the Bauer cabinet.
In this instance, the conflict was resolved by the creation
of a new post (Reconstruction) and its offer to the DDP. But
the DDP won control not only of Justice, but of Interior as
well. Perhaps this indicates how quietly desperate the SPD
leaders were to form a cabinet with the semblance of a
consensus.

The Bauer cabinet lasted nine months. This relatively
long duration can be attributed to the lack of ideological

differences within the coalition. Its end came not from

97
internal differences or rival parliamentary opposition, but

from an attempted coup d'etat by para-military forces. The

 

Kapp Putsch was an attempt to seize state power by reactionary
elements. While it succeeded in toppling the Bauer cabinet,

it did not destroy the republic. The coup eventually collapsed,
but the Bauer cabinet was not re—formed. A new cabinet of
Weimar Coalition parties formed under the leadership of

Hermann Mueller (SPD).

Aside from the Kapp Putsch, the other major political
event of 1920 was the national election. The results disappointed
the Weimar Coalition parties, especially the SPD. The Social
Democrats' parliamentary representation dropped from thirty—
seven to twenty-one per cent. Also, the number of parties
increased. Radical parties had the greatest increase: the
Communist Party (KPD) seated a delegation for the first time,
and the German Nationalists expanded their delegation.
Electoral polarization thus seems to have begun as a trend at
the same time as party system expansion. While radical party
growth was not a primary variable in the model, the number of
parties is a variable of importance in the causal model
developed in Chapter III.

The Mueller cabinet (March 29, 1920-June 21, 1920)
resigned after the June elections, having served only three
months. While its resignation had been only a pro forma
gesture, it was more significant than at the time was realized.

The Mueller cabinet was the last SPD-led Weimar Coalition.

98
While two more such coalitions would be formed under Wirth,
they were both led by the Zentrum.

The SPD chose not to enter the government after the 1920
election. Its poor electoral showing convinced the SPD
leaders that their proletarian supporters were deserting
them for the Independent Socialists and the Communists.
Moreover, it was obvious that a new Weimar Coalition would
not possess a parliamentary majority. Only with the entry
of the Independents into the coalition would the SPD
consider entering itself, and this the Independents rejected.

Only after the minority Fehrenbach cabinet (June 25, 1920-
May 4, 1921) resigned did the Weimar Coalition emerge again,
in 1921. This time it was led not by the SPD, but by the
Catholic Zentrum. Wirth, the Chancellor, was a member of the
left wing of the Zentrum and had the confidence of the SPD.

The first Wirth cabinet (May 10, 192l-October 22, 1921)
included a reunified Social Democratic party. A split in the
Independent Socialist ranks had brought a number of Independent
Reichstag delegates into the SPD fold. Thus, a reduction in
ideological divisions in the legislature increased the solida-
rity of one of the coalition parties. It also lessened
tensions within the Weimar Coalition as the SPD was no longer
pressured by its independent splinter. Strengthened by an
increase in its Reichstag delegation, the SPD was prepared to
take again an active part in government.3

Since the SPD had the largest delegation in the Reichstag,

it might appear curious that its share of the cabinet posts in

99

the first Wirth cabinet was only one-quarter. However, closer
examination reveals these to be fairly powerful posts: Vice
Chancellor, Interior, and the Treasury. Control of such
prestigious and powerful ministries blunted any resentment
over the number of ministries held.

The resignation of the first Wirth cabinet was purely
a foreign policy maneuver. The resignation was supposed to
demonstrate Germany’s displeasure at the Allied demand that
Upper Silesia be turned over to Poland. Thus, the short
(five months) duration of the first Wirth cabinet is misleading
if taken as an indication of the stability of that particular
cabinet. The fall of the first Wirth cabinet had no real
origins in internal differences or in parliamentary oppo—
sition. The real forces lay beyond the realm of the Reichstag
and cabinet; they were primarily foreign, as Germany was
still under pressure from the Allied powers to fulfill the
terms of the Versailles Treaty, the surrender of Upper
Silesia being only one condition.

There are, however, grounds for suspecting that more
was at stake in the resignation than a diplomatic gesture.
Elements of the Zentrum may have hoped that the cabinet’s
resignation would be followed by formation of a cabinet
including the DVP."

If such was the case, the tactic misfired. Negotiations
with the DVP revealed that it would not enter a coalition
with the SPD, especially in view of the cabinet’s avowed

intention of carrying through the terms of the Treaty.

100
Membership in such an accommodationist cabinet could have
only an adverse effect on the DVP's electoral chances.

The DVP'S attitude was characteristic of rightist thinking
during the early period of the republic’s history. The DVP
opposition to the Wirth cabinet stemmed from more than its
reluctance to associate with "accommodationist" parties.

The SPD's presence made such a coalition doubly objectionable,
as this party represented organized labor, while the DVP
represented heavy industry and the managerial-entrepreneurial
class.

A grand coalition of all parties from the SPD to the DVP
could come about only in the face of a crisis that unified
all parties with some vested interested in the status quo.
While the French and the monetary crisis would eventually
provide such an occasion, the attitudes of the SPD and DVP
are more characteristically portrayed in their behavior
during non-crisis periods, such as that of the first Wirth
cabinet.

The second and final Wirth cabinet (October 26, 1921-
November 14, 1922) was similar in most respects to the first,
with changes only in personnel. The SPD did enhance its control
of ministries, but only by one new post. The number of
ministries was reduced from twelve to ten. The DDP lost one
ministry, giving up the Justice portfolio to Radbruch of
the SPD. A rough balance was restored, however, when Rathenau
(DDP) assumed the post of Foreign Minister.

The thirteen months of the second Wirth cabinet were

eventful ones. A treaty of friendship was concluded with the

101
Soviet Union, and Foreign Minister Rathenau was assassinated.
Less dramatic but no less significant was the steady infla-
tional spiral that led to the progressive devaluation of the
mark on the international currency exchange. From November,
1921, to N0vember, 1922, the rate of exchange went from 200
marks to the dollar to 3,500.5
In view of such controversial events, it is significant

that the fall of the second Wirth cabinet was a direct result
of none of them, but rather of the attempt to broaden the
coalition to the right. On November 22, 1922, the press
reported:

On the first day of the talks concerning the

reformation of the cabinet, Dr. Wirth has not

succeeded in fulfilling his original plan of

broadening his cabinet to the right to include

some men of the economy; and to bring the present

regime into an unoffigial relationship with the

German Peoples Party.

The negotiations had encountered the opposition of the

SPD to the entry of the DVP into the new government. In the
minutes of the November 14 cabinet meeting, Wirth declared
that all of the bourgeois parties-—including the Bavarian
Peoples’ Party--had declared themselves in favor of the
DVP's entry into the new government. However, the minutes
record a negative response to the query posed to the SPD
representatives as to whether they were prepared to work
with the DVP. Vice Chancellor Bauer confirmed the SPD’s
attitude.7

Having failed in his attempt to broaden the coalition,

Wirth was compelled to carry out his threat of resignation

102

lest the SPD reject the DVP’s entry into the government.
Thus, the fall of the second Wirth cabinet is attributable
to ideological differences of the most basic sort.

The fall of the last Weimar Coalition—based government
illustrates the change of perspective that had occurred within
a year within the DVP. The DVP stood ready to enter the
government (presumably with the SPD), whereas a year previous
it had rejected a similar proposal. The SPD's leadership
had not, however, changed its perceptions of the DVP suffi-
ciently to enter the government with that party.

Underlying the SPD leaders’ attitude was undoubtedly the
distrust of a party representing heavy industry and the
capitalist class. But no less significant were immediate
political considerations, such as the dilution of the SPD’s
power should the DVP enter the cabinet. This would entail
the surrender of ministerial posts and the sharing of policy-
making powers. Also, the sharing of power with the DVP
might have jeopardized the SPD's electoral future, further
reducing its parliamentary delegation and political capital.

In short, the costs of a grand coalition with the DVP
outweighed the benefits, as far as the SPD was concerned.

All of the political considerations reinforced the basic
ideological cleavage between the two parties, making the
formation of a grand coalition an insurmountable task.

These case studies of the five Weimar Coalition cabinets
support the conclusion drawn in Chapter III concerning the

fundamental causes of governmental destabilization in the

103

Weimar Republic. Ideological diversity seems to have played

a role not only in the fall of several of these cabinets,

but also in the extension of existing coalition governments.
Despite the necessity of broadening the Weimar Coalition in
order to provide a parliamentary majority, the parties could
not overcome basic ideological antagonisms. These ideological
differences were exacerbated by more purely political consi-
derations: the distribution of ministerial posts and the
electoral consequences of sharing power with an ideological

opponent.

The Bourgeois Coalitions

 

While the Weimar Coalition probably presents the fewest
examples of ideological diversity operating on governmental
stability, it must be noted that ideological diversity was
low in the weimar Coalition because of the common "outsider"
status shared by the parties. Also, the Weimar Coalition
governments held power during the early period of the republic
(1919-1923). It was during this period that foreign political
considerations weighed heaviest in the calculation of a govern-
ment's chances of endurance.

The Bourgeois Coalition was doubtlessly the most common
of coalition types to arise in the Weimar Republic. It
contained a set of parties having a common electoral base--
the German middle c1aSses--and a common perception of certain
domestic problems, such as inflation.

Aside from such issues, however, bourgeois parties

differed markedly. The DDP, while ostensibly a middle-class

104
party, did not share the ideological perceptions of the
Peoples' Party (DVP) in a number of policy areas: labor, the
role of the government in the economy, and the former
monarchy. Nor did a party such as the Catholic Zentrum
share views on national educational policy with the German
Peoples Party (DVP) and the Nationalists (DNVP).

If these differences suggest nothing else, it is that
the term "bourgeois party" is an amorphous one, lacking
semantic exactness when used in the Weimar context. None-
theless, the term was employed by the\1eft to describe these
parties, and since a better term is lacking, it is used here
to describe the DDP, Zentrum, BVP, DVP, and DNVP.

The first wholly bourgeois coalition government occurred
after the fall of the Mueller cabinet. The Fehrenbach
cabinet (June 25, 1920—May 4, 1921) was a minority government
comprised of the Democrats, Zentrum, and DVP. Too weak to
request a parliamentary vote of confidence, it continued only
at the sufferance of the SPD.

Despite such a handicap, the Fehrenbach government
survived eleven months. Throughout its tenure, the various
member parties attempted to extend the coalition to the
left (SPD) or right (DNVP). Often the barriers to the entry
of a particular party into the coalition were ideological

in nature. The Democrats, for example, refused to serve with

the German Nationalists out of purely ideological reservation.

8

105

The Peoples Party demanded a high price for its member-
ship in the government: the Economic Affairs, Finance, and
Trade ministries. These were viewed as an "economic unity"
by the DVP. But the major problem, insofar as the DVP was
concerned, was that the SPD, exercising a form of remote
control over the actions of the cabinet, vetoed many of the
DVP's nominees for cabinet posts.9 The course of the
Fehrenbach cabinet aptly illustrates the dilemma of the
minority bourgeois coalitions: they were at the mercy of
forces outside the coalition. It was possible for the SPD
to have a hand in the formation of cabinet policy without
formal participation in the cabinet. This informal overseeing
of the cabinet’s work led to tensions between the DVP and
the other parties of the coalition.

The Fehrenbach cabinet ended abruptly with the london
Ultimatum: a demand by the Allies that Germany accept the
terms of the Versailles Treaty. The crisis presented by the
ultimatum demanded a strong majority government, which the
Fehrenbach cabinet decidedly was not. Following its retire-
ment, the first Wirth cabinet was formed.

The Fehrenbach cabinet presents the model of a typical
minority bourgeois coalition government, with its weakness
stemming from both size (its lack of a parliamentary majority)
and the related problems of dependency on a party outside the
Coalition for continued survival. The SPD’s absence did not
diminish its influence (particularly in the matter of ministe-

rial appointments), as the pro-Republican parties (the Zentrum

106
and the Democrats) represented the SPD's interests. This
led to conflict between the other coalition parties and the
DVP. The only factor working for the cohesion and continuance
of the cabinet was the absence of a viable alternative accep-
table to all parties.

One tactic for overcoming the stigma of a bourgeois
coalition was the so-called non-partisan cabinet. This
formula called for the assignment of "experts" to cabinet
posts. However, these appointees were for the most part
members of bourgeois parties, or high civil servants who
sympathized with a particular party and had close ties to it.
This formula, calculated to lessen ideological tensions
inside and outside the cabinet, did not achieve an impressive
record.

The Cuno cabinet (November 22, 1922-August 12, 1923) was
a classic case of the "non-partisan cabinet." What Cuno
constructed was not a non-partisan cabinet, but a cabinet in
which representatives of three bourgeois parties (DDP,
Zentrum, and DVP) participated ”unofficially." The absence
of any appointees from the SPD alienated many Social Demo-
cratic leaders.10

The Cuno cabinet’s downfall came about through a combi-
nation of foreign and domestic developments that demonstrated
the inadequacies of the "cabinet of experts" solution. First
came the diplomatic-military crisis, which resulted from
Germany's declaration of its inability to meet reparations

payments stipulated in the Versailles Treaty. Following this

107
declaration and a conference on the question in Paris,
French and Belgian military units occupied the Ruhr on
January 11, 1923. The Cuno cabinet responded with a proposal
concerning reparations that was insufficient, and a diplomatic

note that the British declared inappropriate.ll

The second crisis was that of inflation. The inflation
of 1923 was one of the worst in Western European economic
history. The Cuno cabinet did nothing to alleviate the
inflation, the major victims of which were the old middle
class and the proletariat. The beneficiaries were debtors,
real estate operators, and industrialists.l2 But all groups,
those who profited and those who suffered, were not equally
represented in the cabinet. In fact, Cuno's appointments
reflected an over-representation of men with industrial
backgrounds.

The background of the Cuno cabinet may or may not have
had an influence on its inability (or unwillingness) to take
action against the growing inflation. The fact remains that
efforts to stabilize the currency came only after the mark
had.become virtually worthless.

The Social Democrats, having tolerated Cuno’s solution
for lack of a viable alternative, were rapidly reaching the
conclusion that any alternative would be better. Trade
Inlion membership, a primary source of SPD recruitment, was
deczllining as workers perceived that no aid was forthcoming
frcnn that quarter. The growth of radicalism as a consequence

Of ‘the dual crises had led to warnings and counter-measures

108
from the Cuno government. These were perceived by the SPD
to be one-sidedly directed at leftists. This only heightened
the proletariat’s infuriation with a government that was
unable or unwilling to solve the nation's economic ills.
Finally, the SPD announced its readiness to participate

in a grand coalition with the DVP. On August 12, 1923, Cuno
reported the situation to his cabinet:

The Chancellor: The Social Democrats reject the

Communist vote of no confidence, but will express

in a special explanation that the cabinet does

not have the confidence of the parties. This

would in itself normally not necessitate a

resignation. But as I yesterday ascertained,

the prospect of a replacement of this cabinet by

an administration comprising a great coalition

has the support of the leading parties. The only

question remaining is whether such a replacement

can succeed. But early today, Herr Hermann Mueller

(SPD) explained to me that the negative side of his

party's resolution was indeed the elimination of

the present cabinet, whereas the positive side was

the willingness to participate in coalition forma-

tion along the lines of a ’grand coalition."l3

The inability of the Cuno cabinet to deal with crises that

threatened Germany’s survival underscored the inadequacies of
the non—partisan solution. Despite its shortcomings, the
Cuno government lasted nine months. This was primarily due
to the differences between the SPD and DVP that prevented
the formation of a grand coalition. The toleration of such
a cabinet in the midst of such a series of national crises
can be viewed as additional evidence of the magnitude of
ideological divisions among the set of parties capable of
forming a stable majority government.

The five remaining bourgeois cabinets of the Weimar

Republic conform for the most part to the Fehrenbach model

109

of a minority government comprised of the Zentrum, DDP,

and DVP. All but one of these followed a majority coalition
that fell over an ideological issue. The pattern of these
coalition governments reaffirms that observed in both the
Fehrenbach and Cuno cases: that bourgeois coalitions were
essentially caretaker regimes serving until such time as a
more broadly based coalition could be forged. Such coali-
tions could endure as little as four months (Luther II) or
as long as eighteen (Bruening I). In any event, negotiations
continued between coalition parties and those non-coalition
parties sought as partners in an expanded government.

Table 5 shows no particular pattern insofar as post
maldistribution and governmental duration are concerned.
Duration does, however, appear to decline in inverse
proportion to the degree of ideological diversity in these
bourgeois coalitions. This suggests that bourgeois
coalitions do not basically differ from the pattern that
emerged in the analysis of the entire population of coali-
tion governments presented in Chapter III. The Marx II
cabinet deviates from the pattern of increases in ideologi-
cal diversity matched with decline in governmental duration.
In this case, however, post maldistribution rose perceptibly.
We could conclude that in this case, post maldistribution
was the instrumental variable. In Bruening I, low IDG and
long duration are the deviant pattern. Here, too, post
maldistribution is relatively high, but its effect on duration

was minimal. These exceptions, however, do not contradict

110

Table 5

Bourgeois Coalitions

 

 

 

 

l . 2 Ideological Duration
Chancellor PM Size Diversity (months)
Fehrenbach -.21 36.0 .50 11
Cuno —.41 56.0 .50 9
Marx I +.24 40.2 .51 6
Marx II -.37 35.0 .49 6
Luther II -.10 33.7 .55 4
Marx III -.01 33.7 .55 6
Bruening I -.18 43.0 .47 18

1

7PM refers to the degree of post maldistribution, a
measure described in Chapters 1 and 111.

2Size is expressed as percentage of parliamentary seats.

the basic pattern of high ideological diversity and low
governmental duration.

While in some instances it was possible to overcome
partisan barriers to coalition formation, some minority
bourgeois cabinets endured for considerable periods without
successfully expanding to a majority coalition. The first
two Marx cabinets, for example, represent a period of
parliamentary deadlock in which neither a grand nor a right
coalition government was feasible because of the attitudes of
the SPD and the Nationalists. Thus, for a twelve-month period,

Germany had an ineffective government.

Ill

The principal reason for the SPD‘S nonparticipation was
its poor showing in the May, 1924, elections, in which it
lost sixty-seven seats in the Reichstag. The pro—Republican
bourgeois parties suffered losses as well, with the DDP
losing nine seats, the Zentrum three, and the Bavarian
Peoples’ Party four. The DVP, however, was the largest
loser, with twenty—one seats lost. Evidently, a considerable
portion of the DVP's electorate rejected the party’s change
of course in entering a grand coalition with the SPD.

The radical parties, on the other hand, made substantial
gains. The Communists (KPD) gained forty-one seats, while
the German Nationalists picked up twenty—eight. The Nazis,
for the first time, made significant electoral gains, with
thirty-two seats. Stresemann (DVP) spoke of the "loss of a

1

solid middle to balance out the extremist parties,’ and a

"severe blow to the political-parliamentary game."l"
Despite its electoral losses, the DVP attempted to seize
the initiative, this time in the formation of a right—majority
coalition. On May 14, 1924, Stresemann issued a call for
all bourgeois parties to close ranks and unite. This signalled
a move toward rapprochement with the German Nationalists.
Stresemann hoped to make his party the nucleus of a majority
right coalition.15
The DNVP, however, was under the influence of its right
wing. Marx, too, had been negotiating with Hergt of the DNVP.

The minimum conditions set by the DNVP for its entry were

unacceptable to both Marx and Stresemann. The DNVP demanded

112
(1) that the Prussian administration resign, and (2) that
it be given control of the Foreign Ministry (then under
DVP control).

An attempt by the DVP to force the issue by demanding
the retirement of the first Marx cabinet misfired. The
issue on which the hopes for a new majority bourgeois coa-
lition foundered was the acceptance of the Dawes Plan. This
plan for the payment of Germany’s reparations was acceptable
to the moderate bourgeois parties, but was rejected by the
DNVP as too accommodationist.

President Ebert commissioned Marx to form a coalition
government of all bourgeois parties, including presumably
the German Nationalists. They, however, continued to insist
on the two conditions mentioned above, as well as on rejec-
tion of the Dawes Plan, acceptance of which was necessary
for any new government. Finally, on June 3, 1924, Ebert
instructed Marx to re-form the old minority bourgeois coali-
tion and reassume the office Chancellor.

Thus, the attempt to form an expanded coalition govern-
ment was blocked by the ideological dogmatism within the
DNVP. But among the moderate bourgeois parties as well,
enthusiasm for an expanded grand coalition with the SPD
was lacking. Marx observed at one point: "A great
coalition is, one must conclude, impossible, as neither the
Peoples' Party nor the Zentrum desire one . ."16

Marx's doubts concerning the feasibility of a grand

coalition reflected his growing realization that neither the

113
SPD nor the DNVP would enter the government under his leader—
ship. Moreover, the DDP opposed any extension of the coali-
tion to the right (34g,, to include the DNVP) out of purely
ideological motives.17

Again, ideological differences had prevented any grand
coalition solution, with the result that a governmental
crisis had lasted a year. Since the distribution of party
strength in the legislature was responsible for the deadlock,
new elections were held in December, 1924. This scheduling
of new elections seemed the only solution to the govern-
mental crisis. While the election did add slightly to the
strength of the republican parties, it did not add noticeably
to the prospects for a grand coalition government. The DNVP
remained the second largest party in the Reichstag, after the
SPD (see Table 6). With such an impressive electoral showing,
it felt ready to enter the government. The result was the
first Luther cabinet (January 9, l925-December 5, 1925), a
right coalition of the DDP, Zentrum, BVP, DVP, and German
Nationalists. While the December, 1924, election broke the
deadlock in the formation of a right coalition government, it
did not resolve the ideological differences that were at the
heart of many of the issues preventing the formation of
durable majority governments.

The next bourgeois minority cabinet occurred after
Luther’s first right coalition was brought down over the
signing of the Locarno Pact, which the DNVP found objectionable.
With the DNVP‘S withdrawal, Luther was left with a rump

coalition of the DDP, Zentrum, BVP, and DVP.

114

Table 6

First, Second, and Third Reichstag Elections

 

 

 

I. Reichstag II. Reichstag III. Reichstag
(May, 1920) (May, 1924) (Dec., 1924)
Parties
Seats Per Cent Seats Per Cent Seats Per Cent
KPD 4 1 62 13 45 10
USPD 84 18 0 0
SPD 102 22 100 21 131 28
DDP 39 8 28 5 32 7
Zentrum 83 18 65 69
81 17 88 19
BVP 16 19
DVP 65 14 45 10 51 11
Rightistsl 6 1 19 4 12 3
DNVP 71 16 95 20 103 22
Nazis 32 7 14 3
Total 459 472 465

 

l"Rightists" is a catchall term covering a number
of splinter groups on the right.

115
The second Luther cabinet (January 1, l926—May 12, 1926)
lasted only four months. It ended on what was termed the
"flag issue." This was initiated by President Hindenburg’s
presidential decree ordering all German embassies to display
the maritime ensign (which consisted of the old Imperial
colors) as well as the Republican flag. This decree offended
all of the pro-republic parties. Luther’s refusal to
rescind the decree increased the SPD‘s distrust of the
Luther government. This led to the introduction of a measure
of no-confidence and the subsequent fall of the second
Iuther cabinet.
The trivial cause for this cabinet’s collapse points

to the continued ideological tensions that persisted into
the middle period of the republic's history, which was
relatively free of foreign intervention and economic
dislocation. The press attributed the fall of the second
Luther cabinet to ideological conflict:

The Bavarian Kurier believes that radical rightists

and socialist tendencies have made an effective

middle-of—the-road policy between the extremes of

the German party politics impossible. The Frankische

Kurier ascribes all of the blame to a parliamentary

system that tolerates no governmental stability

but strives instead for alterations so that again

more of its memberi can hold the Chancellorship and
ministerial posts. 8

 

 

The third Marx cabinet (May 16, l926-December 17, 1926) was
another interim cabinet formed during the third Reichstag period.
It consisted of the major bourgeois parties: the DDP, Zentrum,
BVP, and DVP. The SPD tolerated this coalition government,

assuming that it would eventually be broadened into a grand

116

coalition. The German Nationalists, on the other hand,
hoped to see a coalition of the right emerge.

Marx, realizing the tenuous basis of his coalition, did
nothing that might lead to dissension. He made no new
cabinet appointments, retaining most of the personnel from
the second Luther cabinet.

Despite such cautious behavior, the SPD brought down
the government by forcing a vote of no-confidence that
gained support from the Communists, German Nationalists, and
National Socialists. The SPD hoped that by causing the third
Marx cabinet to fall, it would bring about a grand coalition.
Instead, the government that eventually took power was a
majority right coalition.

The eight—month long third Marx cabinet conforms to the
pattern of the tolerated minority bourgeois cabinet. It
illustrates the uncertainty with which parties had to contend
when calculating strategies. In this case, the SPD leadership
miscalculated when it assumed that by forcing the retirement
of the third Marx cabinet, a grand coalition would result.
Instead, the SPD precipitated the formation of a right majority
government, an outcome much less favorable than that preceding
it.

The last coalition government based on parliamentary
support was the first Bruening cabinet (March 31, l930-October
7, 1931). This bourgeois cabinet was intended as the prelude
to the formation of a government of all bourgeois parties,

including the German Nationalists.

117

The Nationalists, however, made any cooperation impossi-
ble with their adamant opposition to the Young Plan, an
Allied proposal for the liquidation of the remaining
reparations payments and related economic matters. This
opposition was motivated by the Nationalist ideological
rejection of all "accommodationist" policies. By this time,
the DNVP was under the influence of its right—wing leader,
Hugenberg, who had close ties to the rightist para-military

veterans’ organization, the Stahlhelm.

 

Bruening was forced to form a minority government for
lack of a better solution. Unfortunately, he required the
support of President Hindenburg to carry out his social and
economic programs. Without the President’s support, Bruening‘s
cabinet would suffer the same fate as previous minority
bourgeois cabinets. As the programs envisioned by Bruening
were highly criticized within SPD circles, it was likely
that the SPD would initiate a vote of no—confidence to bring
the cabinet down. However, the SPD leaders feared that, if
opposed, Bruenning (or some other Hindenburg appointee) would
resort to dictatorial rule supported by the President under
the emergency powers of Article 48 of the Constitution. The
SPD leadership concluded that a temporary toleration of the
Bruening cabinet was the wisest strategy.

However, on December 2, 1930, President Hindenburg
signed the emergency decree enabling the Bruening cabinet to
rule without a parliamentary vote of confidence. Shortly

afterward, however, the first Bruening cabinet fell as a

118

consequence of forcing policy differences between the DVP
and the Zentrum.

The first Bruenning cabinet contained all of the weak—
nesses noted in the bourgeois coalition type. Its lack of a
parliamentary majority made it vulnerable, while its lack of
ideological breadth allowed it little authority with which
to make policy for broad sectors of the nation.

The pattern emerging from this review of the bourgeois
cabinets of the Weimar Republic is consistent with the
conclusions drawn in Chapter III concerning the importance of
ideological diversity as a major cause of instability. The
major cause of unstable bourgeois minority cabinets was the
reluctance of major parties to risk electoral losses by
accepting governmental responsibilities. The major parties
(especially the SPD) pursued the more prudent strategy of
manipulating the government from outside the coalition.

When a different course of action proved necessary, the
minority government could be brought down by the introduc-
tion of a vote of no—confidence, or merely the threat of one.

While bourgeois coalitions were essentially rump
coalitions that could be extended and enlarged, the ideo-
logical barriers to such extension and enlargement were
often insurmountable. This led to long periods of minority
government, an indication in itself of the magnitude of

ideological conflict within the Weimar party system.

119

The Grand Coalitions

 

The grand coalition occurred rarely in the period
1919-1932. The period of its occurrence was often marked by
a national crisis. In general, it appears to be a coalition
often striven for, but seldom achieved. The causes for its
rarity are fairly obvious, and have already been alluded to
in the preceding discussion of bourgeois coalition govern—
ments. The major cause is ideological diversity among
the parties comprising any such coalition: the Social
Democrats, Zentrum, Democrats, and Peoples’ Party.

Volker Rittberger, in his analysis of the tensions
underlying coalition behavior in the Weimar Republic, notes
three basic reasons why the grand coalition failed to occur
between 1924 and 1928. With little or no modification,
these reasons would apply to any other period of the Weimar
Republic as well. First, notes Rittberger, the SPD was
reluctant to sacrifice its welfare objectives for member-
ship in the government. Second, the supportive interests
behind the DVP and DNVP, as well as the military, strongly
opposed inclusion of the SPD in the governing coalition.
Fear of concessions in the area of social policy (unemploy—
ment insurance, the eight-hour work day) was the principal
inhibitor so far as the business interests were concerned,
while the military feared tighter control by the civil
authorities of recruitment practices. Third, no urgency was
felt by the bourgeois parties insofar as the inclusion of the

Social Democrats was concerned, because the SPD worked informally

l2O

with the coalition parties whenever its policy objectives
were concerned.19

Most of these "reasons" contain some ideological element,
if by "ideological" we consider policy dictated by a set of
values. Only Rittberger’s third reason seems based on
practical political concerns, and even that conceals a
partially ideological issue: the "unnecessary'inclusion
of a party (the SPD) in a coalition when its informal parti-
cipation is assured whenever its interests are at stake.20

Prior to 1924, the grand coalition was formed twice, in
the two Stresemann cabinets of 1923. This grand coalition,
as we have noted, was formed after the Cuno cabinet-—the
nonpartisan coalition experiment--was proved a failure.
Germany's dual crisis made a cabinet of national solidarity
necessary. It is significant, in view of our argument that
ideological diversity was a primary cause of governmental
instability, that the Stresemann cabinet--a grand coalition--
could arise only in the face of such a crisis.

The first Stresemann cabinet (August 13, 1923—October 4,
1923) was the product of Stresemann's continued efforts to
form a grand coalition. While ostensibly supporting the
Cuno cabinet, the DVP delegation had accepted his plan of
forming such a coalition as the fate of the Cuno cabinet
became increasingly uncertain.

The SPD, too, realized the necessity of removing the
incapacitated Cuno cabinet after the British announced, on

August 11, 1923, that they were unable to convince the French

121
and Belgians to consider Germany's reparations proposals.
This meant continued occupation of the Ruhr, with the loss
of the coal and steel that area supplied to German industry.

Seeking to avoid a parliamentary crisis, President
Ebert sought the counsel of the moderate party leaders
concerning a successor to Cuno. The unanimous recommendation
was to empower Stresemann to form a cabinet. Stresemann
accomplished this task within twenty-four hours, having
previously contacted and negotiated with all of the relevant
party leaders.

Within the resulting grand coalition, Stresemann’s
leadership was the crucial factor in assuring the SPD's
continued cooperation with the DVP. He generally avoided
open conflict in cabinet meetings, preferring private
discussions in which a compromise was more likely to result.
He adopted a nonpartisan attitude in chairing cabinet
meetings, frequently siding with the SPD members against
members of his own party.21

A further consideration was the limitation of post
maldistribution in the cabinet. All important interests were
represented in order to make the first Stresemann cabinet
well insulated from attacks by malcontents within the ranks
of the coalition parties. Stresemann himself assumed the
posts of Chancellor and Foreign Minister. In the latter
capacity he could make maximum use of his contacts among the
Berlin diplomatic community. The posts of Vice Chancellor

and Reconstruction were occupied by Schmidt of the SPD.

122

Another SPD man, Sollman, held the Interior ministry. Even
the left wing of the SPD had its representative in the form
of Dr. Hilferding, formerly of the Independent Socialists.
The Democrats controlled the Trade Ministry. The Catholic
Zentrum held the Labor Ministry and the Ministry for the
Occupied Zone, the latter being filled by the former
President of the Rhine Province, Fuchs. Von Raumer of the
DVP took over the Economic Ministry, thus providing repre-
sentation of the right wing of Stresemann’s party.

The Stresemann cabinet, in its short duration, managed
to initiate considerable constructive policy. Stresemann
succeeded in easing the situation in the Ruhr without
surrendering any claims to the French and Belgians, while
at the same time laying the groundwork for constructive
negotiation. On the domestic front, the cabinet moved
toward the acceptance of a plan for the relief of Germany’s
inflation-ridden economy.22

The first Stresemann cabinet fell early in October,
1923, when it became necessary, because of the economic
situation, to request the Reichstag to pass a general
enabling act to obtain the necessary power. The SPD took
exception to the enabling act, for it would give the adminis-
tration authority to suspend the eight-hour work day law,
which the SPD considered one of its most significant achieve-
ments. The DVP, influenced by the employers’ associations,
maintained that, in view of Germany‘s worsened economy, a

longer work day would be necessary.

123

By a slight majority, the SPD delegation voted against
support of the enabling act. This led to the SPD'S with-
drawal from the grand coalition and the fall of the first
Stresemann cabinet. The short duration of the first Stresemann
cabinet points to the degree of ideological difference in
such a broad coalition. The SPD’S attitude toward the DVP
was suspicious and reserved, while the DVP was frequently
insensitive to the needs of the SPD’s labor constituency.

The only cohesive force in the coalition was the common
perception of the seriousness of the foreign and domestic
crises, which obviously required the union of all responsible
political parties.

The second Stresemann cabinet was nothing more than a
continuation of the first, once the enabling act issue was
settled to the satisfaction of all parties. A compromise
was finally reached by permitting extensions of the work day
through shop representation on a shop-by-shop basis. Other-
wise, the cabinet would have full power to deal with the
economic crisis. The second Stresemann cabinet (October 6,
1923—November 23, 1923) took office only two days after the
demise of the first.

The next challenge to the grand coalition came in winning
parliamentary approval for the enabling act. The German
Nationalists made strong efforts to oust the newly installed
administration. They absented themselves--together with the
Communists--from the Reichstag as the vote was taking place.
This effort to deny a quorum failed. On October 13, the

enabling act passed by a vote of 316 to 24.23

124

Following the passage of the enabling act, the cabinet
dealt with the matter of currency reform. Luther, a high
civil servant with bourgeois party connections, had assumed
the post of finance minister after Hilferding (SPD) had
proved too indecisive in choosing a plan to combat inflation.24

The so-called Hilfferich Plan was put into effect. This

established a National Mortgage Bank (Rentenbank) that received

 

an interest-bearing mortgage on all agricultural and industrial
land. The bank then issued interest-bearing National Mortgage

Banknotes (Rentenmarks) as legal tender.25

 

The second Stresemann cabinet lasted only two months
(October 6, 1923-November 23, 1923). If anything can account
for the small difference in duration between the first and
second Stresemann cabinets, it is the degree of post maldis-
tribution in the two cabinets. The second cabinet, as can

be seen in Table 7, had a greater degree of post maldistribution
Table 7

The Stresemann Cabinets

 

 

 

Cab'net Ideological POSt Duration
1 Diversity Maldistribution

Stresemann I .58 + .02 3 months

Stresemann II .58 - .07 2 months

 

than the first. This was due to the alterations in the Finance

ministry (Luther for Hilferding) and the appointment of a

125

non-partisan Minister of Food (Kanitz). These changes
reduced the number of SPD cabinet members from five to three.
This reduction of SPD representation may have proved expe-
dient in the administrative sphere, since Luther was a
better administrator than Hilferding, but the loss of SPD
representation diminished the visibility of the Social
Democrats in the grand coalition, and may have reduced the
party’s commitment to it as well.

In any event, the SPD’s attitude toward the government
became increasingly suspicious. This suspicion was fed by
a situation in the southern German states that remained
unresolved. Thuringia and Saxony were under the rule of
legally installed coalitions of the Communists and the SPD.
In Bavaria, a rightist government openly flaunted its con—
tempt for the national government in Berlin. On November
10-11, 1923, the Bavarian regime had been temporarily deposed
by Hitler‘s National Socialists in the so-called Beer Hall
Putsch. In all these states, radical leftists and rightists
organized armed bands which the state governments refused to
curb. Stresemann concluded that it was necessary to depose
the Saxon government, both as a response to the threat of
civil order and as a warning to the Bavarian administration.26

On October 27, 1923, Stresemann requested the cabinet
to approve the ouster of the Saxon government. The SPD
ministers were persuaded to call for the voluntary resignation
of the Saxon cabinet. In a message to the Saxon Minister-
President Zeigner, the SPD ministers called on him to give his

response to the government’s demand by October 28.

126

When the deadline passed without a response from Zeigner,
Stresemann issued the deposition order on his own initiative
and sent a Reich Commissar to Saxony with military units to
enforce the order. The SPD ministers responded by insisting
that if such force were used in the case of Saxony, it should
be applied as well in the Bavarian situation. Minister of
the Interior Sollman (SPD) presented the view of the SPD in
the cabinet meeting of November 1, 1923:

The present political situation has been aroused

by the precipitous action in Saxony. The entire
disagreement arises out of the military emergency
action and the form that it took. This is the
unanimous view of the Social Democratic delegation.
The Social Democrats should have been warned about
the planned action in Saxony. Had Saxony been given
a time limit by which to comply with which the
Social Democratic ministers agreed . . . there
would have been no questioning of the action by

the cabinet. The Social Democratic delegation
demands again that the martial law (in Saxony)
should be lifted. As for Bavaria, as I have stated
before, the situation is not tolerable. For weeks
now the Reichswehr has removed itself from the
command of the government; breaches of the Consti-
tution are frequent, and mobilization is in progress
along the border. Bavaria and Saxony should be
treated equally. Recruitment to the Reichswehr

has been solely from among radical rightist circles.
This must be changed. In particular, people from
all of the trade union and civil service organiza-
tions should be appointed to the Army.

 

 

It is now a question of whether the bourgeois

parties desire to rule through the winter with

or without the SPD. 1 must add that if the SPD

leaves the cabinet, a pgll-intentioned neutrality

would not be possible.

To this, Stresemann replied that the SPD was over-playing

its interpretation of the events in Saxony and ignoring the
. fact that a newly constituted SPD regime had already been

installed there. He emphasized that his actions had been

127
consistent with the decision reached in the cabinet meeting
prior to the issue of the deposition order. Furthermore,
he criticized the SPD for publication in the party newspaper

(Vorwaerts) of the delegation's demands, noting that this

 

gave the impression of "Marxist influences" on the cabinet.

The attacks on the Army in Vorwaerts, Stresemann opined, could

 

work only to the advantage of those seeking a right-wing
dictatorship. Finally, he refused point—blank to lift the
martial law edict, as the situation in Thuringia was too
sensitive.28

Having failed to achieve any redress of their grievances,
the SPD proceeded to carry through the threat to withdraw
from the grand coalition. Stresemann’s defense of his double
standard in dealing with leftist and rightist-ruled Laender
convinced the SPD leadership that their misgivings concerning
membership in a governing coalition with the DVP had been
well founded. Continuance in such a coalition might well
exacerbate the divisions within the SPD and show bad faith
as far as the SPD state organizations in Saxony and Thuringia
were concerned. Moreover, the consequences of such a policy
on the outcome of the upcoming 1924 election had to be
considered. All of these considerations were reinforced by
the basic ideological antagonisms between the labor-oriented
SPD and the management-oriented DVP.

The fall of the second Stresemann cabinet came with the

culmination of ideological differences in the Saxony crisis.

The two Stresemann cabinets may be viewed as one continuous

128
effort to bridge ideological divisions during a period of
intense crisis. With the easing of the international crisis
and the implementation of a plan to deal with domestic
economic dislocations, the same ideological differences that
had inhibited the grand coalition experiment in the past
emerged again to disrupt the Stresemann cabinets.

With the easing of the international and domestic crises,
Germany entered a period of relative stability in the years
from 1924 through 1928. The nature of coalition governments
during this period supports the contention that grand coali-
tions were inhibited by ideological cleavages. The composi-
tion of cabinets during the 1924—1928 period was primarily
that of bourgeois and occasionally rightist parties, whenever
moderate interests within the DNVP prevailed.

The next attempt at a grand coalition government came
in June,l928, this time under SPD auspices. The election of
May, 1928, gave the Social Democrats twenty-eight per cent
of the Reichstag seats, making it the largest delegation.

But the increase in SPD parliamentary strength was only

part of a surge in leftist voting: the KPD increased its
strength to ten per cent. Among the bourgeois parties,

this leftward tendency took its toll. Both the DDP and the
DVP lost two per cent of their seats, while the German
Nationalists lost six per cent of their delegation (from
twenty to fourteen per cent of the Reichstag seats). With
the largest delegation, the SPD received presidential support

for the formation of a new cabinet, according to the accepted

129

formula of empowering the largest party in the legislature
to initiate negotiations toward the formation of a government.

Despite the SPD‘s gains, the old Weimar Coalition (SPD,
DDP, and Zentrum) lacked a parliamentary majority, with a
combined strength of forty-eight per cent. A left coalition
was equally impossible, as together the Communists and SPD
controlled only thirty-eight per cent of the parliamentary
seats. There remained only the alternative of a grand coali-
tion.

Hermann Mueller (SPD) undertook negotiations with the
DVP. The Peoples‘ Party required a change in the Prussian
Lppd cabinet prior to any participation in a grand coalition
at the national level. Curtius (DVP) agreed to serve as
Economics Minister in any such grand coalition government,
but this did not constitute a DVP commitment to such a coali-
tion, as Stresemann was still the key party leader.

The DVP‘s demand for reorganization of the Prussian
cabinet, with increased DVP representation, was a logical,
if not an acceptable, development. Prussia, in terms of
population, land area, and administrative apparatus, was the
most powerful of all German states. Control of Prussia was
tantamount to control of Germany, as it constituted the
essential core of the German state. As Prussia was then
‘under SPD administration, the DVP‘s request led to a conflict
between the SPD Prussian party organization and the national
Reichstag delegation. Dr. Braun, Social Democratic Minister—
President of Prussia, announced that Prussia would not allow

the exigencies of national politics to dictate its form of

130
government.

A further complication to the negotiations was added
by the conflict in the Reichstag over the proposed construc-
tion of a new battleship. While the DVP supported this
project, the SPD saw it as a violation of the Versailles
Treaty. This dispute added yet another ideological cleavage
to those already existing between the two parties.

In the face of such obstacles, the best solution that
Mueller could achieve was an "unofficial grand coalition"
cabinet comprised of representatives of the SPD, DDP,

Zentrum, and DVP, but lacking any official commitment to the
government by the respective Reichstag delegations. The
effort to form a parallel grand coalition in Prussia foundered
on the opposition not only of Dr. Braun, but of the Prussian
DVP delegation as well.

The distribution of ministerial posts within the cabinet
reflected the strong commitment of the SPD, contrasting sharply
with the lukewarm commitment of the Zentrum and DVP. The
Social Democrats controlled the ministries of Finance,
Interior, and Labor, in addition to the Chancellorship,
while the Zentrum held only the Trade Ministry. Stresemann
retained the Foreign Ministry for the DVP, while the
Democrats accepted the Justice and Treasury portfolios.29

Despite its inauspicious beginnings, the Grand Coalition
of 1928 lasted two years. Several explanations could be
offered as to why this particular attempt at a grand coali-

tion fared better than the previous two attempts. First,

.I. I...-

131

the unofficial nature of the parties' commitment to the
cabinet removed a source of tension between the party dele-
gations and the cabinet ministers. Second, the 1928
coalition was led by the largest parliamentary party, which
had not been the case in the Stresemann cabinets. Third,
the nation had grown tired of minority bourgeois governments
(which had been the norm in the 1924-1928 period), and
despite the public‘s distrust of party politics, it desired
a stable government. Fourth, the right and left opposition
pursued the same strategies, frequently voting together
against the government-supported legislation. This acted
as a cohesive influence on the grand coalition, both within
the Reichstag and within the cabinet.

Aside from these influences working for cohesion, the
same divisive forces present in the 1923 grand coalition
were at work in the Mueller cabinet. On the left, the trade
unionists complained of the above—average unemployment in the
face of Germany’s economic recovery. The SPD persisted in
its demands to have the employers' contributions to the
social insurance increased, while the employers’ groups within
the DVP insisted that the compensation benefits be reduced.
A compromise plan for such increases in the employers'
contributions still did not satisfy the DVP.

Such social policy differences, with strong ideological
overtones, caused stress within the cabinet. The diplomatic

talents of Mueller and Stresemann were constantly called upon

132
to settle differences. Stresemann’s failing health impaired
his effectiveness in dealing with such internal tensions.
Moreover, the right wing of the DVP was increasingly
attracted to the German Nationalists.

While Mueller constantly sought to place the cabinet’s
rule on a firm parliamentary basis, the efforts did not
succeed. In fact, the coalition regressed. In January, 1928,
an effort to obtain a parliamentary vote of confidence en-
countered the DVP’s renewed demand for a reformation of the
Prussian cabinet. Simultaneously, the Zentrum demanded
additional representation in the cabinet. The DVP objected
to the Zentrum's demand for three additional posts, with the
result that the Zentrum, never enthusiastic about the grand
coalition, withdrew its members.

By determined and continued effort, Mueller succeeded
in bringing the Zentrum back into the government. The DVP
moderated its demands, settling for concessions in the areas
of taxes and financial policy.

Stresemann’s illness worsened, and the rightist
tendencies within his party became ominously more blatant.
On October I, 1929, Stresemann died, leaving Mueller as the
only moderating influence within the coalition government.
With Stresemann gone, the rightists within the DVP dominated
the party. The DVP remained within the government until the
ratification of the Young Plan, but then provoked a parlia-
mentary crisis by demanding a change in the social insurance

law in March, 1930. This pretext for a cabinet crisis led to

133

the fall of the second Mueller cabinet. In fact, the rightist
parties had planned the replacement of the grand coalition
with a rightist government for some time, and the Bruening
cabinet was the product of their efforts.30

The Grand Coalition of 1928 was a durable, if not
stable, coalition government. The various crises through
which it passed often threatened to destroy its cohesion.
The dissatisfaction with the distribution of posts within the
Zentrum led to that party's seven—month withdrawal. The
talents of Mueller and Stresemann proved decisive in
settling the dispute. Stresemann's demise deprived the
grand coalition of one of its major supporters and defenders,
with the consequent fall of the cabinet six months later.

The three examples of grand coalitions we have examined
have certain features in common. First, all were composed
of the same parties: SPD, DDP, Zentrum, BVP, and DVP.
This array of parties contained serious ideological conflicts,
especially in the case of the SPD-DVP relationship. Second,
all were prone to dissension on policy issues with high
ideological content: the eight-hour work law and the
the treatment of leftist and rightist radicals in the case
of the Stresemann cabinets; the social insurance, defense spending
(battleship construction) and the Young Plan in the second
Mueller cabinet. Third, in the case of the second Stresemann
and second Mueller cabinets, post maldistribution was a
disruptive factor in the coalition relationship. In the

latter cabinet, the disproportionate representation of the

134
Zentrum led to that party‘s temporary retirement from the
cabinet. Fourth and finally, all of the grand coalitions
had skilled statesmen in positions of responsibility who
were capable of settling intra-cabinet disputes with
objectivity and a minimum of partisanship. The loss of
such a political leader (Stresemann) was a contributing
cause in the demise of the second Mueller cabinet.

The similarities found in grand coalition cabinets
again point out the primacy of ideological differences among
the parties as the most salient of destabilizing influences
in these governments. Post maldistribution was a contribu-
ting factor in some but not all of the grand coalitions.
While all of these coalitions were larger than minimum-
winning size, this variable did not appear to contribute to

instability.

The Right Coalitions

 

The right coalition was the rarest of coalition types
to arise at the national level of Weimar politics. Unlike
the grand coalition, it occurred in periods of normalcy, in
which the economy was relatively healthy and no serious
political issues were before the Reichstag. Thus, we find
the two right coalitions falling in the period 1924-1928, the
stable period of the Weimar Republic.

The characteristic feature of the right coalition was
that it was a coalition of bourgeois parties, including the

German Nationalists, In both cases it was a majority coalition,

135

falling within the range of fifty-one to fifty—five per cent
set by Riker for a minimum-winning coalition (allowing for
imperfect information). Considering the similarity of
ideologies represented in such a coalition, together with
the size factor (minimum winningness), the right coalition,
by all theoretical expectations, should have been durable.

Table 8 presents the data for the two right coalition
cabinets. For the first Iuther cabinet, post maldistribution
was relatively high, ideological diversity somewhat strong,
difference from minimum-winning size negligible, and duration
relatively long. In the case of the fourth Marx cabinet,
all of the independent variables have lower values than in
the first Luther cabinet, and duration is longer by six
months. This suggests that lower ideological conflict, as
shown in the findings presented in Table 8, favorably affected
the stability of the Marx cabinet.

First, however, we shall examine the first Luther cabinet
to determine the reasons for its longevity. This cabinet
was the first right coalition to be formed in the Weimar
Republic. Its formation and history can clarify the origins

of the right coalition as an alternative to other coalition

 

 

 

forms.
Table 8
The Right Coalitions
Post . .
Cabinet Maldistri- Igigiggigal Mgifffifengfi Duration
bution y ' in“ g
Luther I -.16 .61 +.o3 11 months

Marx IV -.03 .59 .OO 18 months

 

136

The first Luther cabinet (January 9, l925-December 5,
1925) arose after the fall of the second Marx cabinet. It
was largely at the insistence of the Peoples’ Party that the
right coalition was formed. The DDP opposed the DNVP's
entry, while the Zentrum stated that the DNVP's membership
in the cabinet must be conditioned on the SPD's promise of
cooperation.

Koch-Weser, a leading member of the DDP, noted in his
diary the impasse that the negotiations had reached, pointing
to the ideological barriers inhibiting agreement:

The same fundamental differences of opinion
exist now as before. The Democrats will not
make a match with the German Nationalists.
The German Peoples Party will not join the
Social Democrats and also will not accept a
minority cabinet. The Zentrum apparently
wants nothing. It will not join with the
German Nationalists; it also does not want

a minority cabinet; it even will not-—as
Spahn and Becker (Zentrum) today declared—-
join with the Social Democrats. In the
discussions today I suggested that one must
either set up the Weimar Coalition or else form
a cabinet out of the Zentrum and Democrats.
But the old, hazy notion of a "nonpartisan
cabinet" remains . . . . In the case in
question, the danger is only that, in the
name of a "nonpartisan cabinet" one creates
a covert cabineg of the bourgeois bloc
(Buergerblock). 1

 

Koch-Weser obviously feared a repetition of the Cuno
cabinet under different auspices. The multiple crises of
1923 had shown this to be a poor alternative. But with the
economic recovery well under way and with better relations
with the Allied nations established, the nonpartisan

solution again arose as a possible form of government.

137
By January 9, 1925, all efforts to form a coalition

government had failed. Marx, who had been attempting to
expand his coalition government first to the left, then to
the right, admitted that his efforts had failed. President
Ebert then turned to Dr. Iuther, a nonpartisan civil servant
who had served as Finance Minister under the second Stresemann
cabinet. Luther had been responsible for the implementation
of the Hilfferich Plan, the plan for the stabilization of
the currency and the ending of inflation. Luther had the
confidence of the bourgeois bloc parties and at least the
toleration of the DNVP.

With Ebert’s commission, Luther set about negotiating
toward a bourgeois coalition with inclusion of the DNVP.
In order to lesson ideological antagonisms, the coalition
would be officially nonpartisan, with representatives of
each of the parties assigned to the cabinet. Thus, Luther’s
cabinet was not backed by any official commitment of the
coalition parties, and was technically only a cabinet of party
spokesmen acting as liaisons between the government and the
party Reichstag delegations. On January 19, however, the
cabinet received a vote of confidence in the Reichstag.

Stresemann’s retention of the Foreign Minister’s
portfolio lent not only prestige to the cabinet, but conti-
nuity to the foreign policy of the government. Also, it
secured the cooperation of the DVP, without which no viable
coalition of the right was possible. Moreover, Stresemann

added his considerable talent in achieving compromise.

138

One source of support for the Luther cabinet was removed
on February 28, 1925, with the death of President Ebert.

He had been instrumental in bringing about the right
coalition, and his death meant that a moderating influence
on the SPD was removed.

The SPD moved to the attack with the issue of an
emergency tax decree. Luther favored such a decree as a
means of dealing with continued inflation. The SPD's
proposal for a lifting of the decree was viewed by the
rightists within the cabinet as a tactic for dividing the
coalition along ideological lines. Luther rashly threa—
tened to resign should the coalition parties fail to back
his decree. While the cabinet survived this particular
crisis, Luther's action had set a precedent that would later
be followed by Bruening.

The relations between the DNVP and the other coalition
parties were never amiable, as the German Nationalists were
viewed as reactionary by the Zentrum and various factions of
the other bourgeois parties. Stresemann had pursued his
foreign policy objective of obtaining the end of Allied
occupation. The Locarno Conference of 1925 brought the
fruition of his efforts. The Allies agreed to the end of
occupation conditional upon Germany’s assurances that there
would be no forceful revision of the Reich's boundaries.
Prior to the Foreign Minister's departure, however, the DNVP
ministers had insisted that the negotiations be conducted

on a non-binding basis. Nevertheless, so attractive were

139
The Allied proposals that the German delegation agreed to
initial the pact. Elements of the DNVP, fearing that parti-
cular clauses of the pact could be construed as unconditional
acceptance of Germany's existing boundaries, opposed the
signing. The pro-Lucarno faction within the DNVP lost
influence, and as a result, the German Nationalist ministers
submitted their resignations.

The departure of the DNVP marked the end of the first
Luther cabinet. The DNVP, never fully committed to the cabinet
membership, found the Lucarno Pact unacceptable for the same
reasons it had opposed previous treaties with the Allied
powers: the DNVP was suspicious of any agreement involving
concessions by Germany, especially territorial concessions.
Their nonaccommodationist stance made it extremely difficult
for any government including the DNVP to carry out meaningful
and fruitful negotiations on such questions. Thus, the
presence of the DNVP in any cabinet set limits on foreign
policy options, as well as in domestic policy areas where
the DNVP‘s interests were concerned.

In contrast to the causes of the first Luther cabinet's
fall, the events leading to the end of the fourth Marx cabinet
are not complex. The fourth Marx cabinet (January 31, 1927-
June 12, 1928) began as the result of Marx’s efforts to
broaden the bourgeois coalition of his third cabinet.

This was one of the longest series of cabinet formation
negotiations in the history of the Republic, lasting from

November, 1926, to January, 1927. Marx succeeded only with

140

the help of President Hindenburg, who desired the right
coalition over other alternatives. Eighteen months later,
the cabinet resigned, after the May 4, election.

Hindenburg and his advisors saw the governmental crisis
as an opportunity to bring the Nationalists back into the
government. But the crucial party to such a coalition, the
Zentrum, was not unanimous in its desire for a right coali-

ll

tion. Marx informed Hindenburg that his party needed a

pause in which to orient itself to the new conditions."32
The liberal and conservative wings of the Zentrum were
divided over the question of whether or not to join the
DNVP in a cabinet. Wirth, the leader of the liberal faction,
voiced his reservation about a right coalition and the
possibility of combined attack by leftist parties against
such a coalition. Stegerwald, on the other hand, saw a
right coalition as the only alternative to new elections,
which might result in further losses for the Zentrum.33
Finally, the executive committee of the Zentrum formu-
lated the conditions under which they would participate in a
government with the DNVP. These were the following:

1. Recognition and consistent extension of
existing international agreements,
especially the Locarno Treaty.

2. Continuance of the previous foreign policy.
Loyal cooperation in the league of Nations.

3. Unconditional defense of the existing
republican constitution and the national
colors.

4. Action against all alliances and efforts
to overthrow the existing state. A ban on
all civil servants who participate in such
alliances or efforts.

5. The forbidding of all members of the Army to
join or participate in political organizations
of any persuasion.

 

1 .NH‘? .....‘N. u

f... ,.

 

141
6. The swift fulfillment of legislative
action in the areas of cultural and
educational policy, under the protec-
tion of the constitutional equality
of parochial schools and the assurance
of parental rights.34

The above conditions reflect the cautious approach the
Zentrum took toward an alliance with the German Nationalists
in view of the latter‘s performance under the first Luther
cabinet. The Zentrum demanded assurances that the DNVP
would not again disrupt the continuity of German foreign
policy with its reactionary opposition to the Locarno Treaty
and similar agreements. Moreover, the strong Nationalist
sympathies of elements of the Army and Civil Service put the
Zentrum on guard against a possible repetition of the Kapp
Putsch. Finally, the Zentrum’s firm commitment to parochial
education led its party leaders to require assurances that
Protestant interests within the DNVP would not disrupt
legislative efforts in that area.

After an eighteen-month duration, the fourth Marx
cabinet ended abruptly with the elections of May, 1928, in
which the SPD scored a decisive increase in its legislative
representation and the bourgeois and rightist parties
suffered a decrease in their delegations. It appeared that
the electorate had rejected the right coalition formula, and
desired a return to more moderate governmental alternatives
with a definite role for the SPD. This eventually led to the
formation of the 1928 Grand Coalition under Mueller.

One explanation for the longer duration of the Marx

cabinet is the fact that the Zentrum entered the coalition

with firm assurances on basic ideological issues from the DNVP.

142

Such an understanding was lacking in the case of the first
Luther cabinet. Another contributing factor was the official
status of the Marx government, as contrasted with the non-
partisan Iuther cabinet. Nonpartisan cabinets, whether of a
bourgeois or rightist coalition type, fared worse than those
firmly and officially backed by the Reichstag delegations.
While the nonpartisan status of a cabinet temporarily
eased ideological tensions, its long-term effect was to
diminish any feeling of responsibility for the fate of the
cabinet within the supportive parties. In addition to the
above considerations, there remain the overall lower ratings
of ideological diversity, post maldistribution and difference
from minimum-winning size within the fourth Marx cabinet.

In summary, the Marx cabinet possessed decided advantages
over the first luther cabinet, not the least of which was

the advantage of profiting from the latter‘s experience.

Summary
By far the bulk of the evidence from this survey of the

case histories of Weimar cabinets supports the contention
that ideological diversity was the primary cause of instability.
The disproportionate distribution of cabinet posts among the
parties frequently contributed a source of friction in the
cabinet that, when combined with existing ideological
antagonisms, exacerbated the coalition relations.

No type of coalition examined here was immune to the
effects of ideological diversity. Grand coalitions, being

the largest and most diverse, seemed to suffer the most from it.

143

Nor did the use of a non-partisan label on what was for all
practical purposes a bourgeois coalition government spare such
a government from the effect of ideological diversity,
although in this case the pressures originated from outside
the cabinet.

Minimum-winning size was no guarantee in itself of a
stable government. Indeed, minority coalition governments
frequently lasted long periods, due to the indecision and
timidity of certain of the major parties. But such timidity
and indecision can again be traced to pressures of an
ideological nature stemming from the factions within the
various parties and from the parties' constituencies.

In summary, no evidence disconfirms the results obtained
in the data analysis of Chapter III. Regardless of the
social and economic conditions in the German state between
1919 and 1931, coalition governments fell most frequently due
to ideological conflicts within the party system rather than
to any extraneous social, economic, or foreign political

conditions.

144
NOTES

1The term "outsider" is used by Peter Gay to describe
the status, as well as the mentality, of the Social Democratic
and Zentrum politicians throughout the weimar period. Gay
notes that under the Second Reich, Catholics and Socialists
were systematically eliminated from power, while the Reichstag
and other institutions created by Bismarck functioned, to
quote the Social Democrat Wilhelm Liebknecht, as "fig leaves
for absolutism." Gay observes: "Surely, the political
mentality cannot train itself in an atmosphere of persistent
frustration, or with the sense that it is all a sham. When
the democratic Weimar Constitution opened the door to real
politics, the Germans stood at the door, gaping, like
peasants bidden to the palace, hardly knowing how to conduct
themselves." See Peter Gay, Weimar Culture: The Outsider
as Insider (New York: Harper and Row,I968), pp. 71-72.

2Federal Archive, Koblenz, Germany (henceforth abbre-
viated to FA, Koblenz), Vol. R 43 1/1304, pp. 29-31.

 

 

3Alfred Kastning, Die Deutsche Sozialdemokratie zwischen
Koalition und Opposition (Paderborn: Ferdinand Schoeningh,
1970), pp. 102¥104.

4E. Laubach, Die Politik der Kabinette Wirth 1921/22
(Luebeck and Hamburg: Matthiesen Verlag, 1988):

 

 

 

5Erich Eyck, A History of the Weimar Republic, Vol. I
(New York: Athenseum, 1970), pp. 19o-225.

 

6Deutsche Allgemeine Zeitung, No. 497, November 14,
1922.

 

7FA Koblenz, R 43/1305, 15. Cabinet meeting minutes of
November 14, 1922.

8The DDP leaders considered the German Nationalists
unreliable as a result of the Kapp Putsch, which saw the
participation of rightist elements with close ties to the
DNVP. This attitude is apparent in the diary entries of
Koch—Weser in Nachlass Koch-Weser, Federal Archive.

 

9The discussion between Zentrum and DVP negotiators on
June 23, 1920, points out the antagonisms arising out of the
distribution of posts: "The representatives of the German
Peoples' Party again stipulated that their party especially
valued the three ministries--Economic Affairs, Finance, and
Trade--together with the Postal Ministry, and regarded these
ministries as a single economic unity and would fill them
with its own candidates . . . . It was further intimated

145

that the DVP's demand to fill technical posts with party
members was contrary to the DVP's own criticism of the old
coalition, and now the DVP was creating the same difficulties
in the cabinet formation process by recommending candidates

of the t e to which it had previously objected. FA Koblenz,
R 43/1304? 84-85.

10In an interview with a French journalist, the SPD
leader Breitscheid emphasized that within SPD circles, Cuno
was considered . . . presumptuous . . . without the slightest
qualifications for the Chancellorship. Furthermore, he
stated, the party viewed the cabinet with suspicion, as it
included only ministers with contacts among the bourgeois
parties. FA Koblenz, R 43 I/l305, 55-59. This report was
filed through the German Embassy in Paris.

ll
Eyck, op. cit., p. 245.

128. William Halperin, Germany Tried Democracy: A
Political History of the Reich from 1918 to 1933 (New York:
W. W. Norton Company, 1946), pp. 252-254. A slightly
different interpretation of the consequences of the infla-
tion is found in A. J. Nichols, Weimar and the Rise of Hitler
(London: MacMillan, 1968), pp. 102-108. Nichols states
that because of the liquidation of bank reserves and the
drying up of credit, industrialists could not expand produc—
tion.

13 . .

FA Koblenz, R 43 I/l305, 63. Halperin, op. Cit.,
p. 258, states: 'On August 11, 1923, the Social Democrats
sealed Cuno's doom by declaring in the most unequivocal
manner that their patience was at an end. The country, they
proclaimed, needed a stronger government, one that could
command the confidence and support of the masses ofthe nation.

 

 

 

 

Roland Thimme, Stresemann und die Deutsche Volkspartei
1923/25, Pp- 67ff.

15Michael Stuermer, Koalition und Opposition in der
Weimarer Republik 1924-1928—(Duesseldorf: Droste Verlag,
'1967), p. 42. Since the breakup of the grand coalition,
Stuermer notes, the DVP had been drawing steadily closer to
the Nationalists. Stresemann opposed this tendency, although
mostly due to foreign policy differences with the DNVP, not
because of domestic social policy differences.

l6Berliner Boersenzeitung, No. 229, May 16, 1924.
Except from the ministersr counsel meeting of October 1, 1924.

1

7Eyck, op. cit., p. 321. Koch-Weser was "convinced
that the Democra ic Party would deny its very essence if it
were to form a coalition with the Nationalists."

 

 

 

 

146

18FA Koblenz, R 43 1/1307, 176-177. This press report
was filed by Haniel, the Federal Representative in Munich,
on May 15, 1926.

19Volker Rittberger, "Revolution and Pseudo-Democratiza-
tion: The Weimar Republic," in Gabriel A. Almond et al.,
editors, Crisis, Choice, and Change: Historical Studies of
Political Development (Boston: Little, Brown, and Company,

1973) , pp . 380-381.

20The inclusion of the SPD in a bourgeois coalition
would add an ideological dimension to the coalition in that
the SPD would, by virtue of its formal membership, be required
to take a position on issues that might unnecessarily lead
to conflict. By remaining outside the coalition, the SPD
was able to oversee the business of the cabinet without
commiting itself on all policy issues.

21H. A. Turner, Stresemann and the Politics of the Weimar
Republic (Princeton: Princeton University Press, 1963),
p. 116. Also, FA Koblenz, R 43 1/1305, 152.

22The Hilfferich Plan entailed the establishment of a
National Mortgage Bank (Rentenbank) which would receive an
interest-bearing mortage on all industrial and agricultural
land in the Reich. The bank would then issue notes backed
by this mortgage. This plan succeeded, but more for its
psychological impact than for the actual economic thinking
underlying it.

 

 

 

 

 

23Turner, op. cit., p. 120. Among the seven abstainers
were members of the right wing of the DVP.

24Eyck, op. cit, p. 258: "The decision was up to Hilferding,
and now, unfor unately, it became apparent that he was far
stronger as a critic than as a constructive thinker. He was
sufficiently clever and well versed in the theories of
finance to find the weaknesses in every proposition and to
examine them with patient care. But time was running out,
and a decision had to be made immediately. To the despair
of his colleagues, Hilferding simply could not make up his
mind.

25The first version of the Hilfferich Plan called for
linking the new currency to the price of one pound of rye.
Since rye was the chief product of German agriculture,
Hilfferich believed that such a plan would find immediate
support from the farming interests, thus insuring the
nation’s food supply. Eyck, op. cit.

 

 

147

26Turner, op. cit., p. 126: "In view of this turn of
events, Stresemann ecided it would be necessary to depose
the new Saxon government, which had stubbornly refused to
take measures against armed bands that had clashed with Army
units . . . . He regarded such a move as the potential key
to the national government's predicament and hoped it would
not only snuff out the threat of a Communist revolution but
also facilitate a settlement with Munich. An important
factor in the unyielding attitude of the Bavarian authorities
was, in his opinion, their apparently genuine belief that
his cabinet was operating under Marxist preasure." Turner's
view of Stresemann's motives is substantiated by Stresemann's
remarks in the cabinet meeting of November 1, 1923.

27FA Koblenz, R 43 I/l305, 199-201. The minutes of the
cabinet meeting of November 1, 1923, reveal that Stresemann
was worried that certain interests, presumably the Bavarian
government, would receive the impression that his government
was under Marxist pressure: "It is unfortunate that the SPD
demands have been published . . . . The impression that the
cabinet is under Marxist pressure must be avoided under any
circumstances. This appearance would be given if the cabinet
were to accept the SPD demands (in the form of an ultimatum).
It is altogether impossible to accept ultimatums from any
delegation."

28lbid.

 

29The Zentrum was unenthusiastic about the grand coali-
tion, for they had lost the initiative to the SPD with the
failure of Marx's effort to expand his fourth cabinet.
Moreover, President Hindenburg had rejected Wirth as the
nominee for Vice Chancellor, thus insulting the left wing of
the party. Rudolf Morsey, Protokolle der Reichstags Fraktion
und des Fraktions Vorstands der Deutschen Zentrumspartéi:
1926-1933 (Mainz: Matthias Gruenewald verlag, 1969), pp. 217-
226.

 

 

 

3OHermann Bruening, Memoiren: 1918-1934 (Stuttgart:
Deutsche Verlags Anstalt, 1970), pp. 148-156. Various men
close to Hindenburg--Schleicher, Meissner, and Groener--had
approached Bruening with a proposal for a rightist cabinet
to replace the Mueller grand coalition long before its fall.
As leader of the Zentrum's Reichstag delegation, Bruening
was the logical choice as the leader of such a coalition.

 

31FA Koblenz, Nachlass Koch-Weser 32, diary entry of
January 3, 1925, pp. 9-13.

32Morsey, op. cit., p. 90. Minutes of the meeting of
the executive committee on January 20, 1927, 9:30 p.m.

 

33Ibid., p. 91

341bid.

 

Chapter 5

SUMMARY AND CONCLUSIONS

Summary

Our analysis of governmental instability in Weimar
Germany has pointed to the role of ideological diversity in
the governing coalition as the most powerful predictor of
instability. Supportive of this finding from our statis-
tical analysis is the documentary-historical analysis of
Chapter IV, in which we noted that the repeated pattern of
efforts to extend the coalition "spread" across the spectrum
of "acceptable" parties only resulted in short-lived govern-
ments where these efforts succeeded. The rarity of majority
cabinets is itself an indication of the power of ideological
restraints operating to inhibit stability.

A further finding was the remarkably minimal role of
radical party growth in the governmental destabilization
process. We should note, however, that our study considers
only those coalition governments which were formed while
normal parliamentary government existed (1919-1931). After
1931, parliamentary rule was suspended and cabinets were
formed with only the emergency powers of the President (under
Article 48 of the Constitution) to support them. The post-

1931 period also saw a dramatic increase in the parliamentary

148 L [14W w 1171’

 

149

representation of radical parties, namely the Communists and
Nazis. Following the Nazis' electoral victories of 1932,
President von Hindenburg, in what can only be described as
a half-hearted and ill-considered return to constitutional
procedures, empowered Adolf Hitler to form a new government.
This move ineluctably led to the end of the Weimar Republic.
Still, for the 1919-1931 period, the evidence suggests that
radical party growth did not affect governmental duration.

Post maldistribution, while not directly affecting
governmental duration, did correlate significantly with our
major explanatory variable, ideological diversity. From our
analysis, it would appear that post maldistribution’s primary
role was that of an exacerbator of already existing ideological
tensions in the governmental coalitions. The documentary
evidence suggests, however, that as the Republic entered the
middle and late periods, post maldistribution ceased to be
the source of tensions that it had been in the early phase
of the Republic. This was due to the parties‘ perceptions
of the diminishing value of ministerial posts, coupled with
the realization that extra-governmental interests--the army,
civil service, trade unions, agricultural interests and
industrial associations--he1d control of the ministerial
departments, making their ostensible control by the parties
either unnecessary or meaningless. By late 1931, and perhaps
even earlier, the parties had ceased to perform the political
brokerage function. Cabinet formation and parliamentary
coalition maintenance had become largely meaningless by 1932,

with the interests largely by-passing the parties.

150

The size of the coalition governments formed is also
partially accounted for by the diminishing value of inclusion
in the government. We noted in Chapter IV the sizable number
of bourgeois minority coalitions compared to majority coali-
tions of the weimar, Grand, and Right types. In view of
Riker's proposition that larger-than-winning coalitions
invariably "pare down" to minimum-winning size, it is signi-
ficant to note that the minutes of cabinet meetings, party
conferences and other documentary evidence suggests that the
predominant desire on the part of the coalition leaders was
to widen the governmental coalition so as to provide the
greatest degree of legitimacy to its rule. Reluctance,
however, best characterizes the attitude toward cabinet
membership manifested by the major parties, who perceived
the eventual consequences of membership in the government to
be the loss of electoral support.

The relation between the difference from minimum-winning
size (DMW) and ideological diversity in the government (IDG)
variables is partially accounted for by the nature of the
measure of ideological diversity, which tends to overestimate
the ideological diversity of larger-than-winning coalitions
while underestimating that of minority coalitions. Still,
the fact that minority coalitions tended to last longer than
majority coalitions is certainly evidence of the greater
instability of larger-than-winning coalitions as we have

defined stability. Whether, in fact, the minority coalitions

 

were as unstable as majority coalitions on other dimensions

III. #1 ,i ;

 

lifil I...r>1(l—i.
4

. av‘

 

151
than duration is a question that only further research into
the governments of the Weimar period can answer.

The number of parties (NUM) variable is of key impor-
tance in our study, for it is definitely related to govern-
mental duration and to ideological diversity through the
process of destabilization presented in our model. We have
deemphasized the direct NUM-DUR relation because it is a
difficult one to interpret. In Chapter III, we concluded
that this relation may point to the role of splinter parties
as replacements for the major parties (especially the SPD
and the DNVP) whenever these chose to stay outside the
governing coalition. This decision by the major parties did
not preclude their tacit support of weak minority coalition
governments as the only viable alternative to their own

participation.

Hypotheses

 

The six hypotheses of our study were tested in Chapter
III. We shall briefly review these hypotheses and our research
findings.
H. 1: That governmental duration declines with
increases in radical party electoral
support.
No support was found for this hypothesis. Partial correla-
tion analysis of the radical party (RP) and duration (DUR)
relation produced no coefficient higher than .04 for the

relation. Multiple regression analysis produced a partial

slope (beta) of -.18 (with IDG excluded from the equation),

 

152
and an r-square of .01. This further supports our conclu-
sion that radical party growth/decline was a negligible
influence on governmental duration at the national level
of German politics from 1919 to 1931.
H. 2: Governmental duration is negatively

related to ideological diversity in

the governmental coalition.
Strong support for this hypothesis was found in the statis-

tical analysis. Partial correlation analysis and multiple

regression analysis both point to the importance of the

 

ideological diversity (IDG) variable in the decline of govern-
mental stability. Aside from the number of parties (NUM)
variable, no other variable in the study exhibits such a
strong relationship with governmental duration as ideologi-
cal diversity in the government. Unlike the number of

parties variable, ideological diversity is negatively related
to duration, which is in the direction predicted by H. 2.

We must conclude that ideological diversity is the best
predictor of declining duration in this study.

H. 3: Governmental duration is inversely
related to post maldistribution.

Post maldistribution (PM) was found to be not significantly
related to governmental duration, although in the tests for
spuriousness, controls on three variables (NUM, DMW, and RP)
produced partial correlation coefficients approaching, but
not attaining, a level of significance for sixteen degrees
of freedom. H. 3 must be rejected on the basis of the

statistical tests. It should be noted, however, that post

maldistribution is significantly correlated with ideological

 

I‘ll!!! Ill! IILIII!

 

 

153
diversity, the major explanatory variable in this study.
H. 4: Increases in the number of parties

in the national legislature contri-

bute to governmental instability.
The nature of the relationship between the increases in party
system size and instability has been previously noted as a
mysterious one. The problem in interpreting the fairly
strong statistical relationship between the number of
parties (NUM) variable and duration lies in the expansion
of the parliamentary party system and increasing duration.
Our tentative conclusion is that the new parties were
primarily splinter groups which provided reinforcements
for the bourgeois minority coalitions when these suffered
crises due to the reluctance of other moderate parties to
risk loss of electoral support in joining the government.
This conclusion is open to question, and the final answer to
the nature of the relationship requires further research.
However, the grounds for the rejection of H. 4 are very
clear, for the hypothesis posits a negative, not a positive,
relationship between the increase in parliamentary parties
and governmental duration.

H. 5: As a coalition government‘s parlia-

mentary support deviates from minimum-

winning size, its duration declines.
The findings of Chapter III show no significant relation
between deviations from minimum-winning size (DMW) and govern-
mental duration. On the basis of our analysis, H. 5 must

be rejected. In the model of the destabilization process,

the difference from minimum-winning size variable plays the

 

ll b.1- I.

154
role of intermediary between the number of parties (NUM)
variable and ideological diversity (IDG). The DMW variable
is thus indirectly linked to declining duration through
ideological diversity. The explanation for increasing
differences from minimum-winning size contributing to
increasing ideological diversity, which, in turn, negatively
affects duration, is found in the mathematical nature of the
IDG measure, which is sensitive to increases in the number
of governmental parties. The ideological diversity of
minority cabinets is thus underestimated, which accounts for
DMW‘s indirect linkage with relatively long duration.
H. 6: Declining governmental duration is

produced by a combination of some or all

of these five variables.
This hypothesis finds confirmation from the findings of this
study. The pattern of the governmental destabilization
process in Weimar Germany was one in which the number of
parties (NUM) was the initiator of a chain of circumstances
that led to declining duration of governments. The most
important of the variables in this chain is ideological
diversity in the government, which acted as a primary
"gate" for the other variables: difference from minimum-

winning size (DMW), post maldistribution (PM), and the number

of parliamentary parties (NUM).

Conclusions

 

Our major finding regarding the role of ideological

diversity in the governmental destabilization process provides

 

fit-.. .
....

hill in“. all,“

.I‘

155
additional evidence of the importance of this variable in the
Weimar political system. Weimar Germany was a system rife
with political conflict. That such conflict should affect
the stability of governments formed between 1919 and 1931
points to the pervasive nature of ideological conflict in
this period. The minimal roles of the other variables--post
maldistribution, difference from minimum—winning size, and
radical party growth/decline--can be partially accounted for
by the strategies pursued by the parties and their priorities.

In Chapter 1 we examined the Groennings model of coalition
formation. Groennings makes no prediction concerning the
importance of the four categories of variables (situation
variables; compatibility variables; motivational variables;
and interaction variables).l While not all of these cate—
gories are represented in our study, the fact that ideological
diversity in the government (IDG)--as a representative of the
compatibility variable group--was clearly the most significant
predictor of instability may indicate something about the
importance of compatibility in the Weimar political system.
.Additional research is required in order to clarify the
importance of compatibility as compared to the other variable
categories.

Groennings' motivational variables were the least
important of those included in our study: the post maldis—
‘tribution (PM) variable’s relative lack of relationship with
Ciuration is indicative of this. This we explain as due to
tihe low motivation of parties to enter the government when

tShe electoral costs were prohibitive.

156

We should note that the importance of situational and
interaction variables is difficult to assess from our data
analysis. Situational variables like those described by
Groenning (stability of the situation, numerical strength of
the parties, position of parties, constitutional variables,
conventions or informal rules, external pressures and values
and norms of the political culture) are so numerous and
diverse as to constitute several dimensions within themselves,
not to mention that in our study several of these variables
were incorporated into several of our measures of other
variables. Interaction variables (search strategies, influence
strategy and propaganda) were not included among our set of
variables, making any statements concerning their importance
purely conjectural.

Our study does illustrate an important point concerning
the problem of assuming as rational the quest of the politi-
cal parties for the offices and ministerial posts within the
cabinet. In our study, such an assumption would clearly be
short-sighted, for the behavior of the parties points to
their apprehensions concerning the loss of electoral support
should the voters hold them responsible as governmental
parties for the problems of the state and the economy. Parties
such as the Social Democrats (SPD) sought to minimize their
electoral losses by remaining outside the government, trading
their parliamentary support to such weak governments in
return for policy pay-offs. The Nationalists (DNVP) followed
‘this same strategy vis-a-vis weak bourgeois coalition govern-

.Inents with somewhat less success.

 

 

 

 

III». 1, nu I
flu
‘l I

o:

1‘.

Il|.|I|I.

157
DeSwaan observes the following concerning the pattern of
support from nongovernmental parties:
Investigation of the actual parliamentary
coalitions that supported the eight minority
cabinets formed before Bruening I (3/30)
shows that the Social Democrats did, in fact,
through support or abstention uphold the
cabinets of Fehrenbach, Cuno (?), Marx I,
Marx II, Luther II, and Marx III; the Indepgn-
dent Socialists saved Wirth I and Wirth II.
This observation tends to support our contention that the
"tacit support" of the nongovernmental parties was a major
factor in the survival of minority coalition governments.
Such behavior by the major parties may also account for the
low predictive powers of game theoretic models when applied
to the Weimar coalitions.3
We have seen that the coalition alternatives under the
Weimar system were essentially four: (1) the Weimar Coalition;
(2) the Grand Coalition; (3) the Right Coalition; and (4) the
Bourgeois Coalition. The latter was the most common and
in all cases was a minority governmental coalition. It was
the Bourgeois Coalition, however, that endured longer on the
average than any of the other three types. Only the tacit
support of the major non-radical parties in the Reichstag
could account for this phenomenon.
Each of the four coalition types was minimal in ideolo-
ggical range, i.e., each was composed of parties proximate to
one another on some ideological continuum. We note, however,

‘that few of these coalitions were of minimum-winning size.

ZIn.the fact of an economic and foreign-military crisis like

 

 

 

1!.- u
‘l...

158
that of 1923, the objective of the moderate party leaders
was to build the largest consensus coalition possible. The
three Grand Coalitions were constructed in order to distri-
bute the responsibility of government as widely as possible
among the pro-Republican parties. Whether this behavior is
rational in other political cultures is, of course, something
that only further research can determine.

Finally, we note the curious relation between the number
of parties (NUM) and governmental duration. The positive
nature of this relation is frankly puzzling, for it contra-
dicts the notion that increases in the size of the party
system have a detrimental effect on governmental stability.
We suggest that this relationship can be explained by the
role of the newly emergent splinter parties (such as the
Economic Party—~Wirtschaftspartei) as "recruits" for weak
but long—lived minority coalitions. The Bavarian Peoples
Party (BVP) played this role early in the Weimar period, and
was a frequent member of Bourgeois and Right coalitions.

As this study is restricted to a single case of govern-
mental instability, it is difficult to draw generalizable
conclusions from our results. Further research should
reveal which of our conclusions are ethnosyncratic with
:respect to Weimar Germany and which possess a wider appli-

cation to parliamentary systems.

159
NOTES

lSven Groennings, "Notes toward Theories of Coalition
Behavior in Multiparty Systems: Formation and Maintenance,"
in Sven Groennings et al., The Study;of Coalition Behavior
(New York: Holt, Rinehart, and Winston, Inc., 1970),
pp. 445-465.

2Abram DeSwaan, Coalition Theories and Cabinet Formations
(New York: American Elsevier Company, 1973), p. 173.

3Ibid., Chapter 9. Axelrod finds that of the ten
theories operationalized and tested using Weimar data, the
ordinal minimal range theory--which excludes coalitions
with actors that are unnecessary while adding to the
coalitions‘ minimal range along the policy scale--best
predicted the coalitions formed, and no theory approaches
statistical significance. Axelrod states that "Weimar is
a difficult country for the theories."

 

19““

, , .‘IHH 1. .

. id—

. 4.1V.

.l‘.

BIBLIOGRAPHY

BIBLIOGRAPHY

A. SINGLE-VOLUME WORKS

Almond, Gabriel and Sidney Verba. The Civic Culture. Boston:
Little, Brown and Company, 1963.

 

, Scott C. Flanagan and Robert J. Mundt, eds. Crisis,
hoice and Change: Historical Studies of Political
Development. Boston: Little, Brown and Company, 1973.

 

 

Axelrod, Robert. Conflict of Interest. Chicago: Markham
Publishing Company, 1970.

 

Bendix, Reinhard. "Social Stratification and Political Power,"
in R. Bendix and S. M. Lipset, eds., Class, Status and
Power. Glencoe: The Free Press, 1956.

Blalocké Hubert. Social Statistics. New York: McGraw-Hill,
19 O.

 

Bracher, Karl Dietrich. Die Aufloesung der Weimarer Republik:
Eine Studie zum Problam des MaChtverfalls in der Demokratie.
Rev. ed. Stuttgart: Ring-Verlag, 1957.

Bruck, W. F. Social and Economic History of Germany from
William II to Hitler: 91888-1938. Iondon: Oxford
Uhiversity Press, 1938.

 

Bruening, Hermann. Memoiren: 1918-1934. Stuttgart:
Deutsche Verlags Anstalt, 1970.

 

Converse, Philip E. "The Nature of Belief Systems in Mass
Publics," in David Apter, ed., Ideology and Discontent.
New York: 1964.

Cutright, P. "National Political Development: Its Measure-
ment and Social Correlates," in Nelson Polsby et al.,
Politics and Social Life. Boston: Houghton MiffIin
Co., 1963.

 

Dahl, Robert A. "Patterns of Opposition," in Robert A. Dahl,
ed., Political Oppositions in western Democracies. New
Haven: Yale University Press, 1966.

160

 

 

“I... ,r..
.4

p I’v

- .. u was

 

 

161

DeSwaan, Abram. Coalition Theories and Cabinet Formation.
New York: American Elsevier COmpany, 1973.

Eckstein, Harry. Division and Cohesion in Democracy.
Princeton: Princeton University Press, 1966.

Flechtheim, Ossip K. "The Role of the Communist Party,"
in John C. Conway, ed., The Path to Dictatorship,
1918-1933. New York: Anchor BoOks, 1966.

 

 

Gay, Peter. Weimar Culture. New York: Harper and Row,

1968.

 

Groennings, Sven. "Notes Toward Theories of Coalition
Behavior in Multiparty Systems: Formation and
Maintenance," in Sven Groennings et al., The Study of
Coalition Behavior. New York: HoIt, Rinehart and_’
Winston, 1970, pp. 445-465.

 

 

Halperin, S. William. Germany Tried Democracy. New York:
Norton, 1965.

 

Haungs, Peter. Reichspraesident and Parlamentarische
Kabinetteregierung: Eine Studie zum Regierungssystem der
Weimarer Republik in den Jahren 1924-1929. Cologne-Upladen:
West’Deutscher Verlag, 1968}

Herberle, Rudolf. From Democracy to Nazism. Baton Rouge:
Iouisiana State UniverSity Press, 1945.

 

Hermans, Friedrich A. Europe under Democracy or Anarchy?
South Bend: Notre Dame UniverSity Press, 1940.

Hunt, Richard N. German Social Democracy 1918-1933. Chicago:
Quadrangle Books, 1964.

Huntington, Samuel P. Political Order in Changing Societies.
New Haven: Yale University Press, 1968.

Jackson, Robert J9 and Michael B. Stein. Issues in
Comparative Politics. New York: St. Martin’s Press,

1971.

Kastning, Alfred. Die Deutsche Sozialdemokratie zwischen
Koalition und Opposition. ‘Paderborn: Ferdinand
Schoeningh, 1970.

 

 

:Laubach, Ernst. Die Politik der Kabinette Wirth 1921/22.
Luebeck and Hamburg: Matthiesen Verlag, 1968.

-Idpset, S. M. Political Man. New York: Doubleday, 1959.

 

AflacRae, Duncan. Parliamenp, Parties and Society in France:
1946-1958. New York: St. Martin‘s Press, 1967.

 

 

 

1

\

"-‘1

162

Michels, Robert. Political Parties. New York: Dover, 1959.

 

Milatz, Alfred. Waehler und Wahlen in der Weimarer Republik.
Neuwied: Raiffeissendrueckerei, 1968.

 

Morsey, Rudolf, ed. Protokolle der Reichstags Fraktion und
des‘Fraktions Vorstands derIDeutschen Zentrumspartei:
l926-l933. *Mainz: Matthias Grunewald Verlag, 19697

 

 

 

Olson, Mancur, Jr. The logic of Collective Action.
Cambridge, Mass.: *Harvard University'Press,*I965.

 

Puender, Hermann. Politik in der Reichskanzlei 1929/32.
Stuttgart: DeutSche Verlags Anstalt, 1961.

 

Rae, Douglas. The Political Consequences of Electoral Laws.
New Haven: Yale UniverSIty Press, 1967.

 

and Michael Taylor. The Analysis of Political
Cleavages. New Haven: Yale UniversityIPress, 1970.

 

 

Riker, William H. The Theory of Political Coalitions. New
Haven: Yale UniVersity Press, 1962.

 

Rittberger, Volker. "Revolution and Pseudo-Democratization:
The Weimar Republic," in Gabriel A. Almond et al., eds.,
Crisis, Choice, and Change: Historical Studies of Politi-

calDevelopment. IBOston: Little, Brown, and Company,
9 pp- 2 ‘ 91-

Sartori, Giovanni. "European Political Parties: The Case of
Polarized Pluralism," in Joseph LaPalombara and Myron
Wiener, eds., Political Parties and Political Developmepp.
Princeton: Princeton University Press, 1966, pp. 137-176-

Schorske, Carl. German Social Democracy 1905-1917. New York:
Harper, 1955.

Stjernquist, Nils. "Sweden: Stability or Deadlock?” in
Robert A. Dahl, ed., Political Oppositions in Western
Democracies. New Haven: Yale UniVerSity Press, 1966.

 

Stuermer, Michael. Koalition und Opposition in der Weimarer
Republik 1924-1928. Duesseldorf: Droste, 1967.

 

'Ihimme, Roland. Stresemann und die Deutsche Volkspartie
1923/25. No publication information available.

Trimm, Helga. Die Deutsche Sozialpolitik und der Bruch der
Grossen Koalition im Maerz 1930. Duesseldorf: *Droste
Verlag,71952.

 

 

 

 

 

163

Turner, Henry Ashby. Stresemann and the Politics of the Weimar
Republic. Princeton: Princeton University Press, 1963.

weber, Hermann. Die Wandlung des Deutschen Kommunismus.
Frankfurt am Main: EuropaisChe Verlagsanstalt, 1969.

VEheeler-Bennett, J. W. The Nemesis of Power: The German
Army in Politics 1918-1945. New York: Viking Press, 1964.

B. MULTI-VOLUME WORKS

IL'e Bueggerlichen Partien in Deutschland. Vol. I. Leipzig:
VEB Bibliographisches Institut, 1968.

.Esrck, Erich. A History of the Weimar Republic. Vol. I. New
York: Atheneum, 1970.

llcowell, A. Lawrence. Governments and Parties in Continental

Europe. Vol. 1. Cambridge: Harvard University Press,
9 .

£313resemann, Gustav. Vermaechtnis: Der Nachlass in Drei
Baenden. Vols. I, II, and III. Berlin: 1932-33.

C. GOVERNMENT PUBLICATIONS

IPeederal Archive, Koblenz (FA Koblenz). R43 I/Vols. 1020 and
1021. Discussions of the Parliamentary and Party Leader-
ship in the Reichstag.‘

_;, . R43 I/Vols. 1304-1309. Cabinet meeting minutes,
documents, and newspaper clippings relevant to cabinet
formation from 1920 to 1932.

__ . R43 I/Vols. 2654-2656. Papers pertaining to the
German National Peoples Party (DNVP) and the German
Peoples Party (DVP).

. R43 I/Vols. 2657-2659. Zentrum Party documents,
1919-1933-

. R43 I/Vols. 2661-2663. Democratic Party (DDP) and
Social Democratic Party (SPD) papers, 1919-1933.

_ . R45 II/Vol. 32. Minutes of the meetings of the
German Peoples Party (DVP) delegation to the Reichstag.

Ifeeierabend, Ivo K. et al.

164

. R45 II/Vol. 44. Minutes of the Central Committee of
e DVP, December 14, 1929.

__7 . Nachlass Koch-Weser, VOls. 22-39. Diary entries from
November 11, 1919, through October 21, 1930.

C}ermany, Buero des Reichstags. Reichstags Handbuch, I.
Wahlperiode, 1920. Berlin: 1920.

. Buero deS Reichstags. Reichstags Handbuch, II.

Wahlperiode, 1924. Berlin: 1924. p-

D. PERIODICALS *

Etrowne, Eric C. and Mark N. Franklin. "Aspects of Coalition 7
Payoffs in European Parliamentary Democracies," American .{i
Political Science Review, 67 (June, 1973), 453-469.

 

 

Ekzkstein, Harry. "Authority Relations and Governmental Per-

formance: A Theoretical Framework," Comparative Political
Studies, 2 (October, 1969), 269-325.

______ "Aggressive Behavior within Politics,
1948—1962: A Cross-National Study," Journal of Conflict

Resolution, 10 (September, 1966), 249-271.

(Ecrumm, John G. "Theories of Electoral Systems," Midwest
Journal of Political Science, 2 (1958), 357-376. '

.Igeik, Robert M. "A Measure of Ordinal Consensus," Pacific
Journal of Sociology, 9 (1968), 85-90.

Iluoomis, Charles P. and J. Allen Beagle. "The Spread of
Nazism in Rural Areas," American Sociological Review 4 (1946),

729-730-

Dieubauer, Deane E. "Some Conditions of Democracy," American
Political Science Review, 61 (December, 1971), 607-6I9.

CD’Lessker, Karl. "Who Voted for Hitler? A New Look at the
Class Basis of Nazism," American Journal of Socilogy, 74

(JUlY: 1968): 63'69-

Eihively, w. P. "Party Identification, Party Choice and Voting
Stability: The Weimar Case," American Political Science
Review, 65 (December, 1972), 1203-1225.

Ehzhnaiberg, Allan. "Commentary and Debate: A Critique of
Karl O’Lessker’s 'Who Voted for Hitler?‘" American Journal
of Sociology, 74 (July, 1968), 74-89.

 

165

Stokes, Donald E. "Compound Paths: An Expository Note,"
American Journal of Political Science. 18 (February,
1974), I91-214.

 

E. UNPUBLISHED WORKS

13urnham, Walter Dean. "Political Immunization and Political
Confessionalism: Some Comparative Inquiries," a paper
presented at the 1970 International Political Science
Association Convention.

Ckameron, Stefanie. "Nazi Ascendancy and the Weimar Party
System," a paper presented at the Midwest Regional Meeting
of the International Studies Association, Toronto, 1972.

Imeiserson, Michael. Coalitions in Politics. Ph.D. disserta-
tion, Department of Political Science, Yale University,

1966.

 

 

IRIett, Samuel. The Social Basis of Nazism and Communism in
Urban Germany. M. A. Thesis, Department of Sociology,
Michigan State University, 1948.

 

APPEND IX

 

COALITION GOVERNMENTS IN WEIMAR

166
APPENDIX

GERMANY, 1919-1931

 

 

 

 

Ideolo-
Cabinet Coalition Type Size DItggE figit Duration
sity
1. Scheidemann SPD-DDP Weimar .77 .56 -.08 4 months ‘
Zentrum
2. Bauer SPD-DDP Weimar .77 .56 -.O6 9 months _
Zentrum ;
firs
3. Mueller I SPD-DDP Weimar .77 .56 -.01 3 months v1
Zentrum ,
4. Fehrenbach DDP—DVP Bourgeois 36 .50 -.21 11 months
Zentrum
5. Wirth I SPD-DDP Weimar .44 .56 -.07 5 months
Zentrum
6. Wirth II SPD-DDP Weimar .44 .50 -.10 13 months
Zentrum
7. Cuno DDP Bourgeois .40 .50 -.41 9 months
Zentruml
BVP-DVP
8. Stresemann SPD-DDP Grand .58 .58 +.02 3 months
I Zentrum
DVP
9. Stresemann SPD-DDP Grand .58 .58 +.02 1 month
II Zentrum
DVP
10. Marx I DDP Bourgeois .40 .51 +.24 6 months
Zentrum
BVP-DVP
ll. Marx II DDP Bourgeois .30 .49 -.37 6 months
Zentrum

DVP

167

 

 

Ideolo-
gical Post
Cabinet Coalition Type Size Diver- Mal. Duration
sity

 

12. Luther I DDP Right .55 .61 -.16 11 months
Zentrum Hr -
BVP-DVP ‘
DNVP

l3. Luther II DDP Bourgeois .34 .55 -.10 4 months
Zentrum
BVP-DVP t

 

l4. Marx III DDP Bourgeois .34 .55 -.01 6 months Pf
Zentrum
BVP-DVP

l5. Marx IV DDP Right .54 .59 -.03 5 months
Zentrum
BVP-DVP
DNVP

16. Mueller II SPD Grand .59 .53 -.07 21 months
Zentrum
BVP-DVP
DNVP

l7. Bruening DDP Bourgeois .34 .47 -.18 20 months
Zentrum
BVP-D
Wirt.

 

lThe Bavarian Peoples Party (BVP) split off from the Zentrum
in 1920. It was farther to the right than the Zentrum on many
issues.

2The Economic Party (Wirtschaftspartei) was a splinter of the
right, acting as a representative of small business. Only in the
Bruening Cabinet (1931) was it included in a national government.

 

 

   

M'TITI'IILIWJLIJMIfllfiujffl'ffll'l’tjlififlfijflm”