RADICALISM OR REFORMISM?
SOCIALIST PARTIES BEFORE WORLD WAR I*
Gary Marks
Burton Craige Distinguished Professor
Department of Political Science
University of North Carolina at Chapel Hill
and
Professor, Chair in Multilevel Governance
VU, Amsterdam
Heather A. D. Mbaye
Assistant Professor of Political Science and Planning
University of West Georgia
Hyung-min Kim
University of North Carolina at Chapel Hill
1
Abstract
This paper seeks to explain ideological variation among socialist, social democratic, and
labor parties across 18 countries in the early twentieth century. On the basis of quantitative and
qualitative analysis, it suggests that the timing of civil liberties (freedom of association and
freedom of the press) and the strength of links with labor unions were decisive for reformism or
radicalism. Both depended on the capacity of the middle class to achieve its legislative
objectives. While it builds on Seymour Martin Lipset’s prior analysis, it qualifies Lipset’s claim
that political citizenship provides a key to socialist orientation. The evidence presented here is
consistent with the idea that socialist radicalism was a response to repression and generalizes
from the American case to theorize the consequences of union-party links.
2
“[C]hanges in class and political relations within industrially developed societies, much like the shifts in
left-wing politics in the United States and Europe, may be analyzed within the framework of an
“apolitical ”Marxism – that is, by accepting the proposition that technological advances and the
distribution of economic classes determine the political and cultural superstructures, but without
assuming that socialism will succeed capitalism” (Lipset 2001: 77).
The ideological development of socialist parties resembles a natural experiment, and as
a result, has relevance for social scientists even beyond its substantive importance. As Seymour
Martin Lipset and Stein Rokkan (1967) recognized in their classic account of party structure, the
sociology of politics in all industrial societies has been framed by conflict between workers and
employers. Unlike the center/periphery and religious cleavages arising from national state
formation and the Reformation, the class cleavage characterizes all industrial societies. The
question we pose in this article is why that conflict varied so decisively prior to World War I.
Why were some socialist parties radical and others reformist?
Socialist, social democratic, or labor parties were formed across Europe, North America,
and Australasia in the decades prior to World War I. All appealed to manual workers as their
core constituency. All began with, or adopted, political programs demanding manhood (later,
universal) suffrage, civil rights, economic equality, and a decisive role for the state in the
economy. Yet the goals that socialist parties pursued varied widely. Some, like the Russian
Bolshevik party, the Finnish socialist party, and the Spanish socialist party had revolutionary
platforms demanding the dissolution of the existing capitalist system, the expropriation of
private property, and an immediate transition to socialism. Other socialist or labor parties, such
as the British Labour party, the Danish socialist party, and the Swiss socialist party, were
1
reformist, campaigning for welfare reforms, the eight-hour day, and more working-class
representation in the legislature.
Since Tilly’s landmark study (1978), social movement theorists have explored how state
repression has shaped social movements and their repertoires (McAdam 1996; Tarrow 1996).
Subsequent research has focused on variation in the process of repression: Is repression brutal or
soft, generalized or selective, legal or illegal, preemptive or reactive, rigid or flexible,
professional or improvised (della Porta and Reiter 1998; McPhail and McCarthy 2005)? This
article disaggregates repression into three substantive categories: denial of civil liberties, denial
of the right to organize in the labor market, and denial of political citizenship—because these
rights are causally, as well as conceptually, distinctive.
The topic of radicalism or reformism has been taken up by political sociologists from the
time of Karl Marx and while we build on their ideas throughout this paper, the explanation we
offer differs from previous ones in three respects. First, we evaluate the validity of what we
regard as the most compelling hypotheses in a quantitative, as well as a qualitative, format. One
reason why there has been negligible systematic testing in this field is that we lack objective
measures, and so our first step is to operationalize and measure socialist ideology and salient
elements of the political context that confronted socialist parties. 1 Second, we reject the notion
that participation in elections dampened the revolutionary ardor of socialist parties and we
qualify Lipset’s claim that political citizenship provides a key to socialist orientation. Basic civil
liberties (freedom of association and freedom of the press) are prior to, and apparently more
decisive than, citizenship. Finally, the explanation we propose goes beyond the widely-held
1
A major exception to the statement in the text is Stefano Bartolini (2000). We draw on Bartolini's data to
measure union-party links.
2
view that socialist ideology can be explained as a response to the structure of political
alternatives. We claim that the structure of the labor movement—in particular, links between the
party and labor unions—constituted a powerful pressure for socialist radicalism or reformism.
I. THEORIZING REFORMISM OR RADICALISM
An astonishing variety of factors have been hypothesized to have an effect on socialist party
orientation. These include economic variables bearing on the structure and class consciousness
of the working class, such as the standard of living, the degree of industrial concentration, the
character of the division of labor and the work process, and the level, pace, and timing of
industrialization; social variables shaping working-class attitudes, mobility, and solidarity,
including the status system, social mobility, the ethnic, religious, and linguistic composition of
the working class, the spatial nexus of work and home, and the development of compulsory
education; and political variables influencing both working-class attitudes and socialist party
ideology, such as the timing of manhood suffrage, state repression, cross-class coalitions, the
degree to which the state was decentralized, the role of intellectuals, and the type, coverage, and
timing of welfare provisions.
Two lines of approach have been taken to come to grips with this diversity of plausible
causes. The first is to make detailed comparisons across a few cases in order to understand the
interaction of multiple factors. By observing change in even a single country, the researcher can
present evidence about how, say, the introduction of manhood suffrage affects the political
orientation of socialist leaders. However, inference is likely to be difficult given the large
number of possible influences and limited variation of relevant conditions across a small
number of cases.
3
A second approach, and the one taken here, is to compare a larger set of cases. While we
sacrifice descriptive specificity, we have the prospect of examining the validity of our
expectations with greater rigor by increasing the breadth of experience that we bring to bear.
However, the number of independent cases at our disposal—37 cases drawn from 18
countries—is still not large enough to control the range of influences listed above. We therefore
focus on two sets of causally proximate variables. The first has to do with the political
opportunities that were available to socialist parties within their respective political systems.
Could workers vote? What channels were open for workers to organize and express their
demands? The second line of argument concerns the structure of the labor movement and the
pressures that labor unions could exert on socialist parties to reform the system of wage labor
rather than campaign for its dissolution.
The political channels available to socialist parties and the structure of the labor
movement resulted from explicable patterns of economic, social, and political development.
What appears to have been crucial was the ability of the bourgeoisie and its middle-class allies
to mount an effective challenge to the hegemony of the monarchy and its agrarian allies.
Political liberalism was designed to promote the interests of the propertied middle classes, but
by denting absolutism it had the unintended consequence of a) extending some minimal,
though contested, domain of freedom of expression and association for workers, and b)
provoking legal ambiguity for workers' combinations in the labor market, thereby providing
them with some breathing space. 2
2
In their explanation of the push for manhood suffrage, Rueschemeyer, Stephens, and Stephens argue
that “[C]apitalist development strengthens civil society and both the middle and working classes, thus
leading to the strengthening of democratic forces,” (1992: 143) but that the middle classes tended to resist
the extension of suffrage to male workers in the decades around the turn of the twentieth century (96ff).
4
The literature on the development of socialist parties has been an extended dialogue
with the ideas of Karl Marx, and while we analyze socialist strategy unburdened by the
assumption that workers would challenge the basic framework of capitalism, the explanation
we propose is Marxian in that it is motivated by class conflict. Classes are groups “endowed
with particular rights” (Marx 1843), the most important of which were property rights over the
means of production and rights underpinning control of the state. Following Lipset, we argue
that prior class conflict between the landed aristocracy and commercial and industrial elites
conditioned the rise of a new class, the proletariat.
Manhood suffrage
The effects of manhood suffrage were a puzzle for socialists in the decades around the
turn of the twentieth century. Some believed that if workers had access to the vote this would
deepen class consciousness. In 1894, Engels noted with satisfaction that British unions were
intent on gaining representation in the House of Commons, ". . . let us have textile workers in
Parliament just as we already have miners there. As soon as a dozen branches of industry are
represented class consciousness will arise of itself” (Lapides 1986: 165).
Against this, many socialists noted that male suffrage might weaken demands for the
wholesale rejection of capitalist society. Lenin argued that German Social Democracy could
benefit from the fact that “the bourgeois-democratic revolution was still incomplete” and that,
conversely, American socialism was weak because it existed in a ". . . firmly established
democratic system… which confronted the proletariat with purely socialist tasks" (Lenin 1907:
362). In an interview with Arthur Ransome, Lenin described England as the “freest (sic) country
in the world," which harnessed the masses to political democracy via a “systematically
managed, well-equipped system of flattery, lies, and fraud . . .” (Cowden 1963: 50).
5
The latter view has predominated among social scientists. Seymour Martin Lipset
summarized this in his presidential address to the American Political Science Association: “The
exclusion of workers from the fundamental political rights of citizenship effectively fused the
struggle for political and economic equality and cast that struggle in a radical mold. Thus, a
large number of European socialist movements grew strong and adopted a radical Marxist
ideology while the working class was still unfranchised or was discriminated against by an
electoral system that was explicitly class or property biased" (Lipset 1983: 7; see also Dahl 1971).
The manhood suffrage hypothesis has been argued on several grounds:
1. The denial of manhood suffrage meant that social democratic parties could not convert
electoral support into political reform. Instead, they were induced to campaign for fundamental
change in the rules of the political game.
2. The denial of manhood suffrage alienated workers by making them feel a class apart. It
legitimated the social democratic claim to represent workers as a class against the ruling class
(Lipset 1983).
3. To the extent that workers were politically disempowered, so laws were biased against them
(Stephens 1979).
Adam Przeworski and his co-authors argue that manhood suffrage undermined the
revolutionism of socialist parties, but for reasons that are distinct from the classic manhood
suffrage hypothesis (Przeworski 1985; Przeworski and Sprague 1986). In their view, the very act
of participating in democratic elections blunted revolutionism. The argument underlying the
electoral participation hypothesis is subtle. Because the unit of electoral appeal is the individual,
socialist parties were induced to frame their appeal in individual, not class, terms. Elections
therefore undermined the ability of socialist parties to extend class solidarity to the political
6
realm. 3 But this is not all. As socialist parties entered the electoral arena, their leaders came to
realize that despite Marxist claims, the working class would never constitute a simple majority
of the voting-age population. If socialist parties limited their appeal to the proletariat they
would be doomed to minority status. Hence, socialists made cross-class appeals. The net result,
according to Przeworski and Sprague, is that elections undermine class cohesiveness by
individualizing representation and by inducing socialist leaders to appeal to non-proletarians. 4
By participating in elections, socialist parties inadvertently weakened the salience of class in
political life. The decision to participate thus marked a fork in the road away from class-based
revolutionism toward integration within capitalist democracy.
Civil Liberties
The civil liberties hypothesis has been argued on the following grounds:
1. When the regime makes it impossible for the party to gain political influence, basic
institutional change becomes a logical precondition for the pursuit of the party's other goals.
3
“People who are capitalists or wage-earners within the system of production all appear in politics
undifferentiated ‘individuals’ or ‘citizens’“(Przeworski 1985: 13).
4
As Przeworski notes, Robert Michels made a similar point, “For motives predominantly electoral, the
party of the workers seeks support from the petty bourgeois elements of society, and this gives rise to
more or less extensive reactions upon the party itself. The Labour Party becomes the party of the ‘people.’
Its appeals are no longer addressed to the manual workers, but to ‘all producers,’ to the ‘entire working
population,’ these phrases being applied to all the classes and all the strata of society except the idlers
who live upon the income from investments” (Michels [1916] 1962: 254, quoted in Przeworski 1985: 26).
7
2. To the extent that the regime denies the party political access, it isolates the party from
diverse influences that could co-opt, moderate, or incorporate the party and its leaders
(Katzenstein 1985).
3. A regime that stonewalls gives the initiative to radicals in the party who can point to past
experience to support their claims that gradual ameliorative reform is impossible.
4. Repression arouses feelings of deprivation and resentment within the party and among
workers. Injustice in a society is likely to be felt all the more acutely when those subject to it are
denied the chance to defend themselves (Tilly 1978; Mikkelsen 2005).
A stream of liberal political theory from John Locke and John Stuart Mill to Robert Dahl
and Seymour Martin Lipset draws on these arguments to make a law-like generalization:
toleration moderates; repression radicalizes. Civil liberties and the rule of law, in this tradition,
create the conditions for pluralistic political competition. The path to socialist reformism begins
not with working-class citizenship and manhood suffrage, but with freedom from state
repression, which is at the core of negative liberty (Berlin 1969; Dahl 1971).
These expectations are echoed in the comparative literature. Lipset believed that “Crossnational variations in working-class political avifctivity were . . . affected by differences in the
extent to which the proletariat was legally free to form class-based organizations and participate
in the economic and political life of their societies” (1983: 6). In his comprehensive analysis of
socialist parties, Bartolini concludes that “Political repression was one of the main determinants
of early socialist movement behaviours and the instrument through which the state shaped the
structure and fundamental forms of labour protest” (Bartolini 2000: 397; Eley 2002: 68).
In the early 1840s, Karl Marx wrote a series of articles on freedom of the press in which
he argued that it was in the interest of the ruling class in the German lands to tolerate freedom
of the press: “Censorship does not abolish the struggle, it makes it one-sided, it converts an
8
open struggle into a hidden one, it converts a struggle over principles into a struggle of
principle without power against power without principle” (Marx 1842: 6).
The effect of civil liberties, or their denial, can be conceived in instrumental terms.
Radicalism, from this perspective, is a response to the absence of legitimate channels, not a
psychological disposition produced by alienation or poverty. Where civil liberties were weak or
absent, social democratic parties were denied legal channels of expression and organization and
were induced to combat the regime to gain the political space necessary to exist, even if they
would otherwise have preferred a reformist course (Bartolini 2000: 379). One might say that
radicalism is reformism under duress.
The implication of this line of argument is that the effect of repression on radicalism is
U-shaped. Extreme repression raises the cost of resistance to such an extent that even the most
angry or alienated person may be beaten into acquiescence. As Charles, Louise, and Richard
Tilly argue in their study of violence in the nineteenth and early twentieth centuries: “Under
heavy repression, collective action subsides, and the effectiveness of those collective actions
which do occur, and do generate violence, declines. . . . Highly tolerant regimes also diminish
the effectiveness of those collective actions which have considerable probabilities of violence.
They do so by multiplying the available paths to any particular objective, thus making the
violence-strewn path less attractive. The maximum relative effectiveness of the high-violence
path probably lies between the extremes of repression and nonrepression, toward the repressive
end of the range” (1975 : 286).
Unions
Labor unions predated socialist parties in every society, and by the turn of the twentieth
century they had far larger memberships and far greater financial resources than socialist
9
parties. Socialist revolutionaries and business union leaders from Lenin to Samuel Gompers
agreed on a possible consequence: Where socialist parties were enmeshed with strong unions,
socialism would be driven in a reformist direction.
One reason for this is that unions are embedded in the capitalist system of wage labor.
The first unions in Britain, France, Germany, and in many other countries, did not begin by
bargaining with employers. They attempted to control the supply of labor at its source,
unilaterally, by setting the rates at which they would offer their labor to employers. Later, craft
unions were joined by groups of less-skilled workers who could never hope to control the
supply of their labor, but who could gain some bargaining power by threatening to withhold
their labor en masse. The first groups to form industrial unions were coalminers and textile
workers (Laslett 2000). Such unions had the numbers to campaign for legislation of working
conditions, but like craft unions, their efforts were directed to improving working conditions
rather than ending the system of wage labor itself.
Unions challenged orthodox Marxism in another, yet more fundamental, way. They
were sectional organizations. Although they were more proletarian than socialist parties, for
they were composed exclusively of wage workers (and led by them!), they did not amalgamate
into class-wide organizations. The first durable unions were composed of workers who
identified their life chances with those of other workers in their occupation. This allowed them
to overcome the free-rider dilemma. Compositors, cigarmakers, shoemakers, handloom
weavers, and coalminers, for example, were located in close-knit occupational communities
which could punish defectors and sustain union membership as a social norm (Marks 1989).
The occupational communities formed by such workers were bounded. While levels of union
membership vary widely across countries and across time, union movements everywhere are
10
composed of multiple, more or less independent, organizations rooted in particular occupations
or industries, often grounded in local communities (Voss 1988; Conell 1990).
In the Communist Manifesto, Marx and Engels predicted that union sectionalism was a
temporary condition. Experience, they believed, would teach unions to consolidate into classwide organizations: "The real fruit of their battles lies not in the immediate result, but in the
ever expanding union of the workers. This union is helped on by the improved means of
communication that are created by Modern Industry, and that place the workers of different
localities in contact with one another. It was just this contact that was needed to centralize the
numerous local struggles, all of the same character, into one national struggle between
classes”(Marx and Engels 1890 [1985]: 90).
Forty years later, Engels described the rise of new unionism as a turning point in British
history because it appeared to be a "general cry for the organisation of all trade unions into one
fraternity and for a direct struggle against capital" (Cowden 1963: 45). When Engels turned
again to the topic, in his preface to the Second Edition of his Condition of the Working Class in
England, he made no mention of his hope for "one fraternity,” but emphasized instead that
general unions provided strong support for a labour party which would be independent of the
Liberal party (1892: 1ff).
Left-wing socialists were convinced that amalgamation was essential if unions were to
represent workers as a class. Eugene Debs, five-time presidential candidate of the American
Socialist party, was incensed by the decentralized craft structure of unions in the American
Federation of Labor: “You have innumerable unions represented there, but no unity. You have
this great body of workers parceled out among scores of petty and purposeless unions, which
are in ceaseless conflict with each other, jealous to preserve their craft identity. As long as this
great army of workers is scattered among so many craft unions, it will be impossible for them to
11
unite and act in harmony together. Craft unionism is the negation of class solidarity. The more
unions you have, the less unity; and here, in fact, you have no unity at all” (Debs 1909: 11). 5 In
1905, Debs set up the Industrial Workers of the World, “a working class organization, so allinclusive, so comprehensive, that it will embrace every man and woman who does useful work
for a livelihood”(1909: 5). The effort failed to attract more than a small proportion of workers or
unions, as did all subsequent efforts by socialists to create class-wide union movements.
The core idea of the union-party links hypothesis is that socialist parties tied to sectional,
economistic unions were impelled towards reformism. Orthodox Marxists believed that
socialists would socialize and educate unionists, but given the preponderant size and financial
strength of unions, it is more plausible to believe that unions would shape the party. The
contrast in resources was striking. Socialist parties collected small annual dues; unions collected
larger dues, on a weekly or monthly basis. Most socialist parties depended on unions to provide
them with members, and where this was not the case, the membership of socialist parties was
but a small fraction of that of unions. Before World War I, socialist parties were premised on the
aspiration to reform or transform the economic, social, and political system. Unions, by contrast,
were concerned with workers’ material rewards, their economic security, their safety, and their
dignity, in short with the conditions of daily life.
However, trade unions, like socialist parties, were sensitive to the legal climate. Union
reformism depended on the right of workers to combine (Marks 1989; Lipset and Marks 2000).
5
John Commons agreed, but saw this as a strength: “Finally, when Gompers and the others built up the
American Federation of Labor, they did not have a centralized big union where Gompers would be a
dictator and have control of the funds and discipline, but a loose federation, with shop autonomy, union
autonomy, craft autonomy—‘autonomy’ everywhere, and only two rules—union card and no dual
unions” (Commons 1926: 286).
12
When unions were free to organize, they led socialist parties to reformism, but when unions
were repressed, socialist parties led unions to revolutionism, or unions took the task upon
themselves by turning to revolutionary syndicalism.
Unions that were not suffocated by the state sprouted up as independent, reformist
organizations. This was regarded as a fact of life by revolutionary socialists and reformist
unionists alike. Lenin argued that unions were capable of nothing more than “economism,” a
derogatory term for wage bargaining. Samuel Gompers, president of the American Federation
of Workers from its inception in 1888 to 1924, came to the same conclusion, though he regarded
it as a virtue. As Dick Geary observes in his survey of working-class political activity in Europe:
"To a large extent labor protest remained purely industrial where it could satisfy its needs
through the application of industrial muscle. The absence of such muscle, however, or its
thwarting by laws and the intransigence of employers, transformed attitudes and the arena of
conflict" (1981: 69).
II. METHOD AND DATA
We evaluate the expectations set out above against a new dataset for eighteen European,
North American, and Australasian countries across two time points, 1900 and 1914. 6 One reason
there has been negligible systematic testing of explanations for socialist radicalism or reformism
is that we lack measures of party positioning. We seek to remedy this by evaluating social
6
Socialist, social democratic, or labor parties existed in all countries at both time points, except in New
Zealand where a labor party was established in 1910; France, where in 1900 there were two socialist
parties, the Parti Socialiste de France and the Parti Socialiste Français; and Russia, where in 1914 there were
also two parties, the Mensheviks and the Bolsheviks.
13
democratic positioning along three indicators: 1) the party’s orientation to the political system;
2) the party’s orientation to the economic system; and 3) the methods that the party advocates
for achieving its goals. We then adjust the scores for dissent. Appendix B details our coding
scheme and the chief historical sources on which it is based. The Cronbach’s alpha statistic for
the three components is 0.93 which is consistent with the claim that they indicate a single
dimension. 7
Our argument is that socialists tended to more reformist when the bourgoisie and its
middle-class allies were able to achieve some of their political objectives. It is not possible to
observe middle-class power directly, and so we estimate it indirectly as a function of the timing
of middle-class suffrage. Our presumption is that the earlier middle-class enfranchisement took
place, the more able was the middle class to achieve its political objectives. One can also make
the argument that the greater the economic and political power of the middle class, the earlier it
was enfranchised. Causal complexity is often problematic for social science, but in this case it is
good because it suggests that the indicator we use is associated with the phenomenon we wish
to measure.
Our measure of union-party links extends Bartolini’s categorization of “cross-linkages”
between socialist parties and trade unions in 13 European countries (2000: 256ff and Table 6.4)
to the 18 countries in our dataset.
7
A principal component factor analysis yields one factor with an eigenvalue > 1 which explains 86
percent of the variance. The component weight of orientation to political system is 0.94, orientation to
economic system is 0.92, and method is 0.92. The correlation of this factor with the summary score in
Appendix A is 1.00. When we use the factor as the dependent variable in the analysis reported in Figure
3, all estimates are unchanged to the second decimal place.
14
We operationalize civil liberties as a function of the elapsed number of years between the
legal recognition of freedom of association and freedom of the press and our survey time point.
Freedom of combination and manhood suffrage are operationalized in the same way. The results we
report measure time elapsed as a logarithmic function of the number of years between a
particular event and our survey time-point on the expectation that the marginal effect of an
additional year’s experience declines as the number of years increases. However, all results
reported in this article hold if we measure time linearly. Table 1 explains how we calculate the
independent variables and table 2 provides the raw data.
[Tables 1 and 2 about here]
By using the date of a legal reform to anchor an independent variable we minimize
coding subjectivity. The potential cost is that this does not fully capture the subtle nature of
class power or of the political space available for working-class organization. However, to the
extent that the error in our operationalizations is random, so our estimates are conservative. It is
more important for us to avoid systematic bias that might have the effect of reducing the
standard errors of our estimates than it is to eliminate the white noise that inflates them (Marks
2007).
The dataset is characterized by cross-sectional dominance: 37 cases in 18 countries across
two time periods. The bulk of variation in socialist party orientation—89.7 percent—is
crossnational, and the same applies to the independent variables hypothesized here. Because
errors in our estimation are likely to be correlated within individual countries, we calculate
cluster-corrected standard errors (Beck and Katz 1995).
15
In the next section we estimate a structural equation model which summarizes the
causal depth of our argument. 8 A structural equation model allows us to encompass both
proximate and distal sources of socialist party orientation. We are therefore able to relax the
assumption that the sources of socialist party orientation are causally equidistant from the
explanandum and to hypothesize, instead, both direct and indirect causal effects. Our purpose
is not only to explain variance among socialist parties, but also to propose a causal argument
connecting a) the political emergence of the bourgeoisie and its middle-class allies to b) the
political space for labor unions and working-class parties which c) provides a setting for
internal pressures and external opportunities to shape socialist party ideology. In the
subsequent section of the article we discuss theoretically interesting and deviant cases.
III. DISCUSSION
Figure 1 confirms the manhood suffrage hypothesis. The univariate association between
manhood suffrage and reformism/radicalism is −0.53 and is significant, correcting for the clustered
character of the data, at the .01 level. Political citizenship opens a channel for working-class
representation, and this moderates socialist demands. However, the timing of manhood
suffrage is blind to the context in which manhood suffrage is introduced. Whereas middle-class
suffrage provided property owners with a channel for legislative influence, manhood suffrage
did the same for individuals who lacked property, and who were therefore dependent on the
mobilization of numbers, not resources. The right to vote is one thing, freedom to articulate and
organize demands in a mass movement is another.
8
Using Amos 7.0.
16
In Austria, Italy, Russia, Spain, and Finland, manhood suffrage went along with the
suppression of civil liberties. In Belgium, Sweden, Norway, the Netherlands, Britain, and
Canada, manhood suffrage was delayed, but in the context of a relatively open society. In the
latter societies, the extension of the suffrage proceeded in steps. Socialist parties were able to
openly campaign for reform in societies that appeared to be moving in the direction of political
citizenship. Many socialists regarded the pace of change as unacceptably slow, but the direction
of reform was clear: with few exceptions, the extension of the franchise was a ratchet-like,
unilinear, process.
[Figure 1 about here]
One can gauge the validity of this line of argument by comparing the markers in Figure
1. Black circles indicate socialist parties in regimes where freedom of the press and freedom of
association were in place before 1880; white circles indicate socialist parties in regimes where
these rights were introduced after 1880. All but one party located below the fit line is
represented by a black circle. All but one party located above the fit line is represented by a
white circle. The pattern displayed in Figure 1 suggests that manhood suffrage is associated with
reformism/radicalism, that civil liberties is powerful in accounting for the residuals, and that three
cases—the Italian socialist party in 1900 and the American socialist party in 1900 and 1914 —
defy this line of explanation.
Before we investigate this more thoroughly, we need to evaluate the hypothesis that
participation in elections blunted the revolutionary impulse of socialist parties. As Przeworski
and others have noted, the decades around the turn of the twentieth century saw a shift
towards reformism in several socialist parties: the mean score on our 12-point scale is 6.1 in 1900
and 5.6 in 1914. The turn to reformism is generally stronger among parties that were more
radical in 1900 than in 1914 (R = 0.39), which is consistent with the thrust of the electoral
17
participation hypothesis, but the timing of electoral participation does not account for this. Nor,
interestingly, does socialist success in elections, measured by the proportion of the vote for
socialist parties. Electorally successful parties in Norway (with 26.3 percent of the vote in 1914),
Finland (43.1 percent), and Italy (17.6 percent) were at least as radical in 1914 as they were in
1900; electorally weak parties in the United States (6.0 percent) and New Zealand (9.6 percent)
became more reformist. 9
The electoral participation hypothesis does not purchase much, if any, cross-sectional
variation in socialist radicalism or reformism. The association is in the expected direction, but it
is weak (R = −0.23). Figure 2 reveals that the timing of electoral participation predicts too little
reformism in societies where civil liberties were in place, and too much reformism in societies
that lacked such liberties.
[Figure 2 about here]
Figure 3 presents a structural equation model that is consistent with the expectations set
out above, including the claim that the middle class set the legal parameters within which
working class organizations could mobilize, and thereby shaped the political orientation of
socialist parties. 10 In the model, the variable middle-class suffrage does not directly affect socialist
9
R = −0.10 for the percentage of socialist vote and change in reformism/radicalism (measured as
reformism/radicalism in 1914 minus reformism/radicalism in 1900). R = −0.17 for electoral participation and
change in reformism/radicalism.
10
Estimates of model fit, as reported in Figure 3, are within acceptable bounds. The X2 or chi-square
statistic is a "badness of fit" measure: a finding of significance indicates that the given model's covariance
structure is significantly different from the observed covariance matrix. To reduce the sensitivity of X2 to
sample size, researchers divide it by the degrees of freedom, where a value < 3 is deemed acceptable
(Kline 1998: 128). The next three indices (CFI, NFI, and IFI) are comparative or incremental fit indices
18
party reformism/radicalism, but exerts an indirect effect via each of the variables that exert
proximate causal influence, the timing of manhood suffrage, a factor summarizing the timing of
civil liberties and the timing of freedom of combination, 11 and union-party links. Together these
variables account for an estimated 68 percent of the variance in socialist party
reformism/radicalism.
[Figure 3 about here]
Table 3 lists the total standardized effects of the independent variables and reveals that
although middle-class suffrage has no direct effect on reformism/radicalism, its standardized
total effect is greater than any other variable in the model. 12 We conceive the legal framework
of civil liberties and freedom of combination to be causally important both for socialist parties
and for labor unions, and so we model this variable as having both an indirect and a direct
effect on socialist reformism/radicalism. Combining these produces a total effect for liberties
that is second only to middle-class suffrage.
[Table 3 about here]
assessing the relative fit improvement of the model implied here compared to the null model. Values for
the CFI, NFI, and IFI range from 0 to 1, where a value > .90 is considered acceptable (Kline 1998; 2005; Hu
and Bentler 1999).
11
The correlation of civil liberties and freedom of combination is 0.90. A principal component factor analysis
yields one factor with an eigenvalue > 1 which explains 95 percent of the variance. The component
weights for civil liberties and for freedom of combination are 0.97.
12
Regressing reformism/radicalism on manhood suffrage, liberties, and union-party links, produces Variance
Inflation Factors (VIFs) of 1.98, 2.92, and 1.78, respectively. These suggest that our estimate for liberties is
the least reliable of those in Figure 3, but that multicollinearity in the model is within acceptable bounds.
19
The implication of our analysis is that Marx was correct in believing that class relations
reflected the productive basis of a society and that the political success of the bourgeoisie and its
middle-class allies in creating a liberal capitalist society shaped the rise of the proletariat. But he
got the sign wrong. The stronger the bourgeoisie, the less, not more, radical were working-class
political parties. No bourgeois; no reformism.
The United States: An exception that suggests a rule
One party, in particular, stands out as being poorly predicted by the timing of manhood
suffrage and the timing civil liberties and freedom of combination: the Socialist party of
America. When we model reformism/radicalism as a function of manhood suffrage and liberties, this
party is considerably more radical than predicted. The residual for the party in 1900 is 2.33
standard deviations above the predicted value and in 1914 the residual is 1.58 standard
deviations.
The republic that was created in 1787 established freedom of the press and of
association, and manhood suffrage for the white majority of the population followed in the late
1820s. Freedom of combination followed in 1842. Yet the socialist party that was formed in 1901
was radical even by continental European standards (Moore 1970: 204-205; Lipset and Marks
2000). Eugene Debs, the party's most prominent spokesman and five-time Presidential
candidate, was a committed Marxist who made no effort to conceal his revolutionary
sentiments in his campaign speeches. Visiting British socialists, from H.G. Wells to Henry
Pelling, were struck by the dogmatic Marxism of the American party. Lenin praised Eugene
Debs on a number of occasions and rarely attacked other American socialist party leaders.
When he condemned opportunists and revisionists, Lenin singled out Arthur Henderson,
leader of the Independent Labour party in Britain, and European social democrats, such as
20
Eduard Bernstein and Alexandre Millerand, but, in America, Lenin's target was Samuel
Gompers, the non-socialist leader of the American Federation of Labor. Before America's entry
into World War I, Lenin believed that the American Socialist party, unlike most European
parties, was capable of becoming a truly revolutionary party.
The American Socialist party was radical in a context which, according to political
opportunity theory, should have induced political moderation. To understand why, one must
engage the structure of the American labor movement and the inability of unions to exert any
moderating influence among socialists. The American Socialist party was not only independent
of the American Federation of Labor (AFL), but there was a well-publicized acrimony between
the leaderships of the two organizations. Eugene Debs, as we have noted, went so far as to set
up a dual labor movement in an effort to outflank the AFL. Samuel Gompers, who led the AFL
for all but one year from its establishment in 1886 until his death in 1924, came to regard
socialism as an ideological vice to be rooted out of unions.
The split in the ranks of labor had several sources. Craft unions, which in the early 1900s
accounted for around two-thirds of AFL members—a large proportion compared to other
industrialized societies—conceived their mission as the defense of skilled workers (Marks 1989;
Conell and Voss 1990). Mass immigration from Eastern and Southern Europe in the decades
around the turn of the century reinforced the distinction between native unionized workers and
non-unionized immigrants. The American working class was religiously, as well as ethnically,
diverse. Several unions were led by Irish Catholics who rejected socialism on religious and
cultural, as well as political, grounds. For their part, socialists failed to reconcile their desire for
an inclusive working-class movement with a penchant for ideological correctness. While some
influential socialists, including Victor Berger, Morris Hillquit, and Frank Hayes, collaborated
with non-socialist unions, they were never able to bring the party with them. They were
21
opposed by those, like Eugene Debs, who rejected the AFL outright and who thought it was
better to have a Marxist party than an inclusive labor party. In any case, AFL leaders preempted
a labor party strategy with their policy of “rewarding friends and punishing enemies” which
signalled an intent to avoid any third-party entanglement (Archer 1998).
Denied union backing, the American Socialist party never became a mass party. Its
membership peaked at 120,000 in 1912, a level that was less than one fifteenth that of the British
Labour party in the same year. The party remained a party of activists and intellectuals,
unalloyed by a large blue-collar base. The ideological effects of this were evident to
contemporary observers in the first decade of the twentieth century, and fatally determined the
party’s policy on intervention in World War I. Socialist leaders in Europe and beyond were
traumatized by the prospect of war in the years before 1914, but once the war began, they were
pressed by unionists to give their support to the war effort. In the United States, however, the
AFL participated in the war effort, but this did not constrain the American socialist party.
Nathan Fine, writing in 1928, observed that the party’s decision to oppose US entrance in the
war was “due primarily to the fact that unlike the parties of Europe the Socialist Party of the
United States . . . was not a mass movement.” According to Fine, the party’s “absence of control
over the trade unions with their bread-and-butter demands, its lack of political strongholds and
a large organization to conserve, all this made the American party primarily a party of
propaganda and education”(1928: 302, 307).
What began in our thinking as an explanation for an exception—the American Socialist
party—appears to be a generalizable phenomenon. To the extent that the United States was
distinctive, our purpose is to explain this from a comparative perspective. The effect of union
links on reformism/radicalism is strong and significant in univariate analysis and under controls.
When we add union-party links to a model with manhood suffrage and liberties, we estimate that
22
the variance explained across the 37 cases in our dataset increases from 54 percent to 68
percent. 13
Middle class weakness and socialist radicalism
Six countries dealt with here did not introduce basic civil liberties until the last decade of
the nineteenth century or later: Russia, Finland, Spain, Italy, Germany, and Austria. In none of
these countries was the middle class able to set the terms of political or economic competition.
Freedom of combination for workers in the labor market was denied for most, if not all, of the
nineteenth century, and where such a right was formally granted, as it was in Germany in 1869
or Austria in 1870, workers had to contend with state repression when they went on strike.
These countries delayed manhood suffrage until the twentieth century, or where they did not—
as in Germany after 1870 and Spain after 1890—elected parliaments were hedged in by an
executive monarchy or by corruption. Beyond these similarities, however, lie contrasting
conditions and divergent trajectories that illustrate both the power and limitations of the theory
we propose.
A reform movement in a society as closed as Russia appeared impractical, if for no other
reason than that it would provoke repression. The paraphernalia of a mass socialist
movement—public meetings, openly elected leaders, membership lists—were inviting targets
for a police state. Lenin’s conception of a professional revolutionary vanguard was premised on
the infeasibility of an open party: “Just try to picture this in the frame of our autocracy!” (Lenin
1902: 171).
13
The variance explained is 52 percent and 64 percent, respectively, when we adjust for chance.
23
The early history of the Russian labor movement exemplifies both the appeal of
reformist unionism for workers and the self-fulfilling consequence of state suppression in
driving unions into the arms of revolutionists. When repression was eased in 1906 and the first
half of 1907, several groups of workers established unions that pressed for better wages and
hours. The printers, in Russia as elsewhere in the vanguard of reformism, managed to set up
formal channels of collective bargaining with their employers. These efforts collapsed when
repression was intensified after 1907. As Victoria Bonnell observes: “Above all, . . . the
Bolsheviks struck a responsive chord among many union activists because the party's approach
coincided with the workers' own experiences after 1912. The autocracy treated economic protest
as political opposition, thereby creating the very conditions that eventually transformed the
unions into revolutionary organizations” (Bonnell 1988: 291).
Until the revolution of 1905, Finnish workers were subject to harsh controls imposed by
Russia, and the response of the Finnish socialist party was to demand revolutionary change. 14
As in Russia, strikes unleashed police repression, and reaffirmed Marx’s dictum that union
activity was merely preparatory to the creation of a class-wide political movement. When
Russian repressive capacity collapsed in 1905, Finnish workers joined with civil servants in a
general strike for civil rights and popular representation. A parliament based on manhood
suffrage was established, but when Russia reasserted its imperial control after 1906, strikes once
again faced police suppression and parliamentary power was curtailed. Participation in
national elections based on manhood suffrage did nothing to tame the revolutionism of the
14
It is worth noting that the causality described here is independent from that in Russia. Finnish socialists
did not follow socialists in Russia, but were influenced by socialists in Germany and Sweden (Knoellinger
1960: 46f).
24
Finnish socialist party. This despite the fact that the Finnish social democratic party received 37
percent of the vote and 40 percent of the seats in the national election of 1907, rising to 43
percent of the vote and 45 percent of the seats in 1913. The party succeeded in becoming a crossclass coalition, attracting a significant proportion of agrarian workers into its ranks, but it
rejected cooperation with other parliamentary groups, and pulled back from its earlier
participation in a national front for Finnish independence.
In Spain and Italy, the aristocracy lost its monopoly of political power in the course of
the nineteenth century, but the middle class was politically and economically weak (Malefakis
1970: ch. 3; Nadal 1973: 556). Systematically rigged elections convinced working-class leaders
that manhood suffrage was fraudulent. In both countries, political corruption led significant
sections of the working class to ideologies rooted in distrust of the state. Anarchism and
anarcho-syndicalism shared a millenarian belief that the masses could be shaken from their
lethargy into revolutionary action by heroic example. Unions were suppressed, organizationally
weak, and politically fragmented, and consequently exercised little moderating influence over
socialist parties.
The response of the Spanish socialist party under the leadership of Pablo Iglesias was to
build its organizational base while espousing a rigid and reductionist Marxism. The party
participated in elections from 1891, but until 1909, it pursued an isolationist policy in the Cortes.
An entrenched agrarian elite was able to sustain its hold on authority behind a façade of liberal
democracy with the support of both the Catholic Church and the bourgeoisie. The result,
government alternation among established landed interests — turno pacifico—produced
immobilism and undermined the case for gradual reform (Heywood 1989).
In Italy, state repression, late manhood suffrage, and weak unionism led to a deeply
divided socialist movement that was initially dominated—contrary to our theoretical
25
expectation—by its reformist wing. The Italian socialist party in 1900 is the second most poorly
predicted of the 37 cases dealt with here (standardized residual = −1.77). The Italian case makes
the point that historical materialism can only take us so far in explaining variation among the
strategies of socialist parties. To go further, one must delve into factors, including culture, that
are independent of the economic basis of a society.
Socialism in Italy was influenced by radical republican followers of Mazzini and
Garibaldi, the founding giants of the Italian state (Steenson 1991: 214ff). Filippo Turati, who led
the Italian socialist party in its early years, was formerly a radical republican who believed that
socialists should ally with liberal sections of the middle class to produce democratic reforms.
Turati wrote to Engels for support, which was forthcoming. Engels shared Turati’s view that
“the Socialist Party of Italy is obviously too young and, considering the whole economic
position, too weak, to be able to hope for an immediate victory of Socialism” (Engels 1894b: 520;
italics in original). And Engels backed Turati’s position that socialists should support a
reforming Republican ministry: “This will give us universal suffrage and greater freedom of
movement (freedom of the press, of organisation, and of assembly)—new weapons not to be
despised”(Engels 1894b: 522).
Gradualism lost its appeal when the historical compromise between reformist socialists
and Giolitti broke down. Reformist socialists were not able to stem a tide of syndicalist strikes
after 1901 and Giolitti’s republican government was unable to maintain a policy of nonintervention in strikes. Maximalist socialists, led by Benito Mussolini, gained support, and the
1912 Socialist congress rejected parliamentarianism in favor of direct revolutionary action. By
this time, the influence of republican reformism among socialists had run its course, and the
Italian socialist party is no longer an outlier.
26
In neither Germany nor France did early manhood suffrage engender reformism. In
Germany, relatively early manhood suffrage provided socialists with representation, but little
more. Men of 25 years or more were given the vote when the German Reich was established in
1871, but the executive was selected by the Kaiser and insulated from the Reichstag.
Constitutional change could be blocked in the upper chamber, the Reichsrat, which was elected
through a curial system that discriminated heavily against workers. Karl Marx described the
regime as “police-guarded military despotism, embellished with parliamentary forms” (Marx
1875: 3). The German Social Democratic party participated in elections, but its leaders claimed
with justice that male suffrage was a tool used by the authorities to bind workers to the regime
while denying them power. There was little political freedom in Germany, even after the most
repressive legislation — the Anti-Socialist Laws — lapsed in 1890. “[T]he persecution not only
failed to destroy the party but radicalized its membership, and led to a process of theoretical
clarification which culminated in the adoption at the Erfurt party congress of 1891 of a
programme written by Karl Kautsky that was avowedly Marxist” (Geary 1989: 119).
Contrasts among German Länder confirm the causal link between civil liberties and
socialist strategy (Bartolini 2000: 321). Socialists in the south did not have to contend with a
reactionary Junker landed aristocracy intent on eliminating socialism as a political force. The
middle class was more diversified and assertive in Bavaria and Baden than in Prussia and,
correspondingly, civil liberties were stronger and more resilient. Socialists in the south of
Germany experimented with reformist policies, and while they were never able to determine
the party as a whole, they were able to break the party ban on coalitions with bourgeois parties
and on voting for state budgets.
Germany was home to the largest Socialist party in the world before 1914 and a beacon
for orthodox Marxism. Yet in the years after 1900, trade unions were a decisive influence in
27
favor of reformism (Mikkelsen 2005). Following the non-renewal of the Anti-Socialist Laws,
increasingly effective and self-confident unions demanded freedom from party interference so
that they could pursue a gradualist strategy in the labor market. After 1905, the unions
thwarted a push to revive the party's revolutionary stance. Most party leaders viewed the
general strike as a potent weapon in their political arsenal, but union leaders insisted that the
strike was an economic weapon that should be reserved for unions. Carl Legien, head of the
socialist Free Union movement, described the general strike as “general nonsense.”
When it comes to France, it is clear that the timing of manhood suffrage and civil
liberties cannot fully capture the ruptures and discontinuities of regime change and its effects
on socialist strategy. Manhood suffrage came early, but it did not lead to reformism. Some nine
million Frenchmen won universal manhood suffrage in March 1848, the overwhelming majority
of whom voted for a government which crushed an incipient socialist state within a state. The
creation of a democratic Third Republic in the early 1870s was preceded by the Paris Commune,
an experiment in populist rule which ended in defeat at the hands of a vengeful government
and the execution of 20,000 sympathizers. This was the bloodiest episode of working-class
repression in any society up to that time. It splintered the socialist movement into several
distinct and competing strands, including reformists, later associated with Jean Jaurès, who
believed that humanitarian socialism could be built on French republican traditions; Marxists of
various tendencies, led by Jules Guesde, united by their rejection of compromise with bourgeois
power brokers; and an insurrectionary stream in the tradition of Auguste Blanqui.
Unions in France did not experience the heavy hand of the state as they did in Italy or
Spain, but they had to contend with legal uncertainty. Strikes were legalized in 1864, two
decades before unions were themselves legalized. Collective bargaining and industrial conflict
outpaced organizational development. Unions, which had to struggle to survive in a twilight
28
legal setting, distanced themselves from socialist parties which from the 1880s had the benefit of
legal tolerance and a relatively broad suffrage. Divisions among socialists reinforced the sense
that links with socialist parties were of limited utility. Many unions turned to revolutionary
syndicalism, an ideology which expressed a deep-seated distrust of the state and electoral
politics, and which made a virtue out of weakness by demanding no more than a loose,
decentralized, and unbureaucratic organizational structure.
In countries where capitalist development went hand in hand with traditional absolutist
controls suppressing worker combination, unions were denied the freedom to bargain
effectively in the labor market and were formed in the wake of socialist parties or rejected the
state altogether. Craft or industrial unions emerged whenever cracks of freedom appeared, but
in Russia, Finland, Spain, and Italy, the preponderant response of ruling elites was to crush
overt resistance. In each of these countries, the ruling class was divided among soft and
hardliners, but the latter had the initiative. In Germany and Austria, unions were established
after socialist parties, and, at least initially, existed in subordination to them (Deutscher 1952:
480-487; La Palombara 1957: 4-7). Later, when unions were given a little breathing room, most
continued to support socialism, but they interpreted socialism as a long-term goal that allowed
them to seek immediate improvements in wages and conditions.
In each of these countries, working-class revolutionism was strengthened by the failure
of the middle class to shape the regime. Socialists were confronted with the task of gaining
liberalism as well as socialism. One logical response was to ally with the middle class, but this
option was pursued only in Italy for a short time. The unwillingness of ruling elites to provide
meaningful channels for working-class demands not only presented a strategic challenge for
socialists, but suppressed the emergence of effective parliamentary representation—and
unions—that could provide the organizational backbone for reformism.
29
Middle class strength and socialist reformism
Seymour Martin Lipset hypothesized that the timing of manhood suffrage provides a
key to the political orientation of a country’s working class. The model in Figure 1 confirms that
the earlier manhood suffrage was granted, the more reformist the socialist party. However, this
general tendency did not hold when unions were cut off from the party (the United States),
when civil liberties were denied (France), or when manhood suffrage was a sham (Germany).
Conversely, as we argue below, late manhood suffrage did not always produce radicalism.
Where the middle class was strong enough to entrench civil liberties and provide unions with
space to be effective in the labor market, reformism could result even if political citizenship was
denied.
In the Low Countries, northern Scandinavia, and Britain, manhood suffrage was not in
place until the late nineteenth or early twentieth centuries. Male workers were excluded for one,
two, or three generations after urban bourgeois and independent farmers were given the vote
(Belgium, 1831; Britain, 1832; Netherlands, 1849; Norway, 1815; Sweden, 1866). The demand for
manhood suffrage pitted the working-class against reactionary forces and made it inevitable
that workers conceive of their political opportunities in broad class terms, but this did not entail
radicalism. In these countries, working-class movements could legitimately voice demands.
Freedom of association and expression were in place before industrialization. Each of these
countries experienced concerted attempts to suppress such freedoms, but these were
unsuccessful in the face of middle-class resistance.
The contrast between early liberties and late manhood suffrage is particularly stark in
Sweden. At the turn of the twentieth century, the Swedish franchise was the most restrictive in
western Europe and a fully parliamentary system had not yet been introduced. The Swedish
30
social democrats modeled their first official program, adopted in 1897, on the Marxist Erfurt
Program of the German social democratic party. But over the next fifteen years the party shifted
in a reformist direction. According to Herbert Tingsten, “The movement’s early adoption of a
moderate and reformist character was due to the strong traditions of freedom and justice which
prevailed in Swedish society. No special legislation against Social Democracy was enacted; the
opportunity to propagandize had few limitations; the state made no systematic attempts to
prevent the growth of the movement through the use of existing legislation” (1973: 712).
Universal male suffrage was not established in Britain until after the First World War.
After prolonged working-class pressure, a Reform Act was finally passed in 1884, but it
excluded domestic servants and recipients of poor relief and was perhaps 35 percent short of
full manhood suffrage. 15 In 1900, the proportion of men over the age of 21 who could vote was
smaller in Britain than in eight of nine European countries for which we have comparable data
(Flora et. al. 1983). However, a strong and independent middle class undergirded a liberal
regime that, apart from a harshly repressive period following the Napoleonic War, generally
tolerated associations and a free press.
A revealing example of the commitment of the English political elite to civil liberties is
its response to Chartism (1838-1848), which was the first mass working-class movement in the
world and which might have engendered large-scale violent conflict. Chartism was a response
to the distress and famine arising from a series of failed harvests and a Tory tax on corn
imports. The palliative, in the eyes of Chartist leaders, was political citizenship. The Charter
consisted of six demands: manhood suffrage, the secret ballot, the end of the property
15
Friedrich Engels observed that non-payment of MPs, and rules for voter registration and candidate
selection, rendered Parliament “a club of the rich ”(Engels 1894a: 536).
31
qualification for MP’s, payment of MPs, equal districts, and annual parliaments. The movement
was uncoordinated, but it was threatening. In the north of England, “physical force” Chartists
(whose slogan was “peacable if we can, forcibly if we must”) were on the verge of violent
revolt.
The government refused to budge on the demand for manhood suffrage or fair elections,
but it avoided provocation. Rather than appoint a reactionary to command the Northern army,
the Home Secretary, Lord John Russell, selected Sir Charles Napier, a man of radical
sympathies. Russell was aware that Napier had actually been invited to be a delegate at the
National Chartist Convention. Not surprisingly, Napier resisted attempts by local magistrates to
mobilize the army against Chartist meetings, such as a massive meeting that was to take place
outside Manchester in 1840. In his diary, Napier recounts that the local justices “were for
stopping the meeting by force, and would have done so without any encouragement; but I
swore if they attempted it not a soldier should quit the barracks till both constables' and
magistrates' heads were broken. This was bravado, for I dare not refuse to obey their orders”
(1857: 53). Napier’s private letters and journal reveal considerable sympathy for the Chartists
and a determination to avoid class war. Napier went, incognito, to the Manchester meeting, and
wrote privately that he saw men “expressing orderly, legal political opinions, pretty much –
don't tell this – very much like my own” (1857: 39). That Lord Melbourne's liberal government
could have chosen such a man to command the army in the potentially explosive north of
England speaks volumes about the self-confident tolerance of the English ruling class at a
critical political juncture.
The failure of Chartism to sustain a durable working-class political movement was
followed from 1851 by the piecemeal establishment of craft unions. The development of
independent union organizations prior to the creation of a working-class party is typical of the
32
English speaking societies. Given freedom to organize and strike, unions tended to develop
segmentally and represent their members in a sectional fashion, wary of both calls for strong
national federation and the leadership of workers' political parties. Where unions were
dominant—as in Australia, Britain, Canada, and New Zealand—the resulting labor or socialist
parties were reformist. In these societies, working-class parties were established by unions,
funded by unions, and continued to be shaped by unions until recent decades. The distinctive
character of the resulting parties is summarized by their self-designation as “labour” parties.
When they eventually took on socialist programs, they did so only after a lengthy period in
which their chief aims were to extend union security and working-class representation within
the existing political system.
IV. CONCLUSION
Seymour Martin Lipset, to whose memory we dedicate this article, described himself as an
apolitical Marxist, and we find this a useful point of departure for explaining variation among
socialist parties. The political struggle of the bourgeoisie and its middle class allies against
landed elites shaped the subsequent struggle of the working class. In order to defend itself from
absolutist repression, the middle class fought for basic civil rights, including freedom of
association and freedom of the press, but once in place, such rights provided working-class
movements with some breathing room. These consequences were largely unintentional. The
propertied middle classes demanded “specific minority rights, as a means of legitimating their
own right to exist” (Lipset, Trow, and Coleman 1956: 15-16). In most cases, they denied the
suffrage to workers and tolerated unions only with reluctance. However, in countries where the
middle class was weak or subordinated to agrarian elites, workers experienced greater
33
repression, had weaker unions, and were denied the right to vote or the right to vote in
meaningful elections.
Charles Tilly noted that “The repressiveness of a government is never a simple matter of
more or less” (1978: 86). This article distinguishes three modes of repression having distinct
causal mechanisms: denial of civil liberties, denial of the suffrage, and denial of combination in
the labor market. In the absence of civil liberties, including the right to associate for political
purposes, socialist parties were driven into illegal opposition. Repression of civil liberties
reinforced those who believed that it was more feasible to overthrow capitalist institutions than
to participate in them. Where civil liberties were completely denied, socialist parties were
forced underground. In Russia, the Bolshevik party became a vanguard party composed of
professional revolutionaries. In Germany, the socialist party weathered repression by dissolving
itself into informal organizations, “micro contexts,” or “cells” that in freer times facilitated
broader mobilization (Aminzade 1995: 42; della Porta 1995).
The denial of the vote to workers because they owned no property confirmed a marxist
class analysis of society and made it difficult or impossible for socialists to affect legislation. The
denial of suffrage had a radicalizing effect, though this was moderated in societies (e.g. in
north-western Europe) where civil rights were entrenched and where reformists could argue
that if workers pressed hard enough, manhood (or universal) suffrage would follow.
Finally, the denial of the right of combination shaped the organizational structure of the
labor movement with decisive consequences for the orientation of socialist parties. Unlike
socialist parties, trade unions defended workers’ interests within the capitalist system of wage
labor. Moreover, union movements were made up of independent organizations representing
distinct groups of workers in particular occupations or industries. Where socialist parties were
34
linked to effective unions, they formed broad class alliances rather than ideologically motivated
sects.
Today, these basic rights are considered components of a single phenomenon, liberal
democracy, but historically they appeared in different sequences with contrasting consequences
for the character of political contestation (Dahl 1971). The socialist experience of the early
twentieth century suggests that basic liberties – the right to organize, and freely communicate,
political opposition – are at the core of reformism for they establish the conditions under which
political citizenship is meaningful. Most contemporary authoritarian regimes with
revolutionary oppositions are politically inclusive, but illiberal – they allow citizens to vote, but
deny them liberty. And like authoritarian regimes a century ago, many authoritarian regimes
today are too weak or illegitimate to stamp out political opposition by brute force.
35
Figure 1. Manhood Suffrage and Socialist Party Reformism / Radicalism
Timing of civil
liberties
12.0
Socialist Party Reformism / Radicalism
Rus1
Rus-B2
Fin1
Fin2
Aut1
Before 1880
After 1880
Fit line for Total
Sp2
10.0
Rus-M2
Sp1
It2
Ger1 Fr-G1
US1
Sv1
Aut2
8.0
Ger2
Neth1 Be1
It1
US2
Fr2
Be2
Fr-J1
Sv2
6.0
Nor2
Neth2
Nor1
Dk1
NZ2
Swt2
Dk2
4.0
GB1
GB2
Swt1
Aus1
Aus2
Can1
Can2
2.0
0.0
0.5
1.0
1.5
2.0
2.5
Manhood Suffrage
Notes:
1=1900; 2=1914
Aus=Australia, Aut=Austria, Be=Belgium, Can=Canada, Dk=Denmark, Fr=France, Fin=Finland, Ger=Germany,
GB=Great Britain, It=Italy, Neth=Netherlands, Nor=Norway, NZ=New Zealand, Rus=Russia, Sp=Spain, Sv=Sweden
Swt=Switzerland, US=United States
Fr-G=Guesdians, and Fr-J=Jaurèsians
Rus-B=Bolsheviks, and Rus-M=Mensheviks
The manhood suffrage axis represents the logged difference between the year of manhood suffrage and the year for
which socialist party reformism/radicalism is measured (1900 or 1914).
36
Figure 2. Electoral Participation and Socialist Party Reformism / Radicalism
Timing of civil
liberties
12.0
Fin2
Socialist Party Reformism / Radicalism
Rus1
Rus-B2
Fin1
Aut1
Before 1880
After 1880
Fit line for Total
Sp2
10.0
US1
Sp1
Fr-G1
Ger1
It2
Sv1
Rus-M2
Aut2
8.0
Ger2
US2
It1
Fr2
Neth1
Be2
Be1
Fr-J1
Sv2
6.0
Nor2
Nor1
Dk1
Neth2
NZ2
Swt1
4.0
GB1
Dk2
GB2
Swt2
Aus2
Can1
Can2
Aus1
2.0
0.0
0.5
1.0
1.5
2.0
Electoral Participation
Notes:
1=1900; 2=1914
Aus=Australia, Aut=Austria, Be=Belgium, Can=Canada, Dk=Denmark, Fr=France, Fin=Finland, Ger=Germany,
GB=Great Britain, It=Italy, Neth=Netherlands, Nor=Norway, NZ=New Zealand, Rus=Russia, Sp=Spain, Sv=Sweden
Swt=Switzerland, US=United States
Fr-G=Guesdians, and Fr-J=Jaurèsians
Rus-B=Bolsheviks, and Rus-M=Mensheviks
The electoral participation axis represents the logged difference between the year in which the socialist party first
participated in national elections and the year for which socialist party reformism/radicalism is measured (1900 or
1914).
37
Figure 3. Sources of Socialist Party Reformism / Radicalism
.45
Manhood suffrage
−.28*
.67**
.83
.98
Civil liberties
Middle-class
suffrage
.92**
.99NT
Liberties
.52**
Freedom of
combination
.91**
−.31*
.84
−.44**
.68
Socialist party
Reformism /
Radicalism
.27
Union-party
links
Notes:
Bold figures are standardized regression weights.
Italicized figures are squared multiple correlations which estimate the total variance explained. NT=not tested, the parameter is constrained to 1 for
the scaling.
All estimates are significant (correcting for clustered errors): ** p < .01, * p < .05.
Model Fit Statistics: X2 = 21.54, df = 7, X2/df = 3.07, CFI = .92, NFI = .90, IFI = .93
X2 = chi-square statistic, df = degrees of freedom, CFI =Bentler’s Comparative Fit Index, NFI = Normed Fit Index , IFI = Incremental Fit Index
38
Table 1. Operationalization of Predictors
Variable
Description
Civil liberties
Time elapsed between survey date and mean of two basic civil liberties:
freedom of the press and freedom of association.
Log10 of {mean of [(survey date minus date of freedom of the press) + (survey
date minus date of freedom of association)] + 1}
Electoral participation
Time elapsed between survey date and date when the party first participated
in national elections.
Log10 of [(survey date minus date of first electoral participation) + 1]
Freedom of combination
Time elapsed between survey date and date of right of combination.
Log10 of [(survey date minus date of right of combination) + 1]
Manhood suffrage
Time elapsed between survey date and date of manhood suffrage.
Log10 of [(survey date minus date of manhood suffrage) + 1]
Middle-class suffrage
Time elapsed between survey date and date when middle class first formed a
majority of the voting population.
Log10 of [(survey date minus date when middle class formed majority of
voting population) + 1]
Union-party links
Institutional cross-linkage between socialist party and trade unions, following
Bartolini’s schema and coding (2000: 256-262). 1 = very weak or nonexistent links; 2 = weak or contingent links (“contingent”); 3 = medium or
strong links where unions are dependent on party
(“dependency/interlocking”); 4 = medium or strong links with equality between
party and unions (“interlocking”); 5=party is dependent on unions
(“dependency/interlocking”). We extend the coding to socialist/labor parties in
Australia, Canada, New Zealand, Russia, Spain, and the United States.
39
Table 2. Predictors of Socialist Party Reformism / Radicalism
Australia
Austria
Belgium
Canada
Denmark
Finland
France
Germany
Great Britain
Italy
Netherlands
New Zealand
Norway
Russia
Spain
Sweden
Switzerland
USA
Freedom of
the press 1
Freedom of
association 2
Electoral
participation 3
Freedom of
combination 4
Middle
class
suffrage 5
Manhood
suffrage
Unionparty
links 6
1830
after 1914
1830
1830
1846
after 1914
1881
after 1914
1830
1900
1848
1830
1814
after 1914
1883
1838
1830
1787
1824
1867
1830
1824
1849
1906
1884
1869
1824
1890
1855
1824
1839
after 1914
after 1914
1864
1848
1787
1901
1897
1894
1900
1884
1907
1893
1871
1900
1895
1888
1890
1894
1918
1898
1896
1896
1900
1824
1870
1866
1824
1849
1906
1884
1869
1824
1890
1855
1824
1839
after 1914
after 1914
1864
1848
1842
18567
18738
1831
18679
1849
1907
181513
1871
1832
186116
1849
1852
1815
190018
189019
1866
1848
1787
1858
1907
1893
188510
184911
190712
184814
187115
1885
1912
1896
1852
189817
after 1914
1890
190920
1848
186021
5
4
4
5
4
3
2
3
5
2
4
5
4
1
2
4
2
1
Notes:
1 Goldstein 1983: 35. 2 Ebbinghaus 1995: 61. 3 Date at which the socialist party first participated in national elections (Mackie and
Rose 1982). 4 Bartolini 2000: 321; Ebbinghaus 1995: 61. We use Ebbinghaus’ date where there is disagreement.
5 Manhood suffrage and middle-class suffrage are from Carstairs 1980; Flora et al. 1983: 89-152; McMinn 1979, quoted in
Molony 2004: 59ff; Rokkan 1970: 80-81; Ward 1950: Ch. 12. 6 See Table 1 for description of coding of union-party links.
7 The timing of manhood suffrage varied across the six Australian colonies. South Australia, Victoria, and New South Wales
legislated manhood suffrage in the 1850s. The smaller colonies were somewhat slower to do so.
8 A curial system restricted the electoral weight of the middle classes. Of the four curia, one was reserved for large landowners
and one for rural inhabitants who paid minimum direct taxes. The urban middle class electorate was confined to the two
remaining curia. 9 Less than half of the enfranchised age group was eligible to vote for the first Canadian election.
10 In 1898, there was a reversion to provincial laws, and although manhood suffrage was adopted in most provinces, it was not
adopted in British Columbia or Quebec. 11 Suffrage for the upper house (Landsting) was narrowed in 1866 to exclude all but the
large landowners and wealthier middle classes. 12 Before 1907, a curial system restricted the electoral weight of the middle class
even more decisively than in Austria or Germany. 13 Suffrage from 1815 to 1848 was effectively restricted to large landowners
and the wealthiest sections of the bourgeoisie. 14 Under the Second Empire, from 1852 to 1869, elections were systematically
manipulated by the government. 15 Universal male suffrage was for the lower chamber, which had little control over government.
16 From 1861 to 1881, suffrage was restricted to less than ten percent of the enfranchised age group. Only those who paid
sufficient taxes and could read and write were eligible to vote. From 1882, more than one quarter of the enfranchised age group
could vote. However, elections were systematically manipulated by the government.
17 Rural areas were under-represented and paupers could not vote. 18 In 1905, the franchise was broadened, but it remained a
curial system that discriminated decisively against the urban middle classes and workers. In 1907, the system was reformed to
further ensure landowner control. 19 Caciquismo transformed a formally democratic system into an oligarchy.
20 Recipients of public poor relief were excluded. 21 In 1830, ten states permitted white manhood suffrage without qualification;
eight states restricted the vote to taxpayers; six imposed a property qualification for suffrage. By 1860, five states limited suffrage
to taxpayers and two imposed property qualifications.
40
Table 3. Standardized Total, Direct, and Indirect Effects on Reformism/Radicalism
TOTAL
EFFECT
DIRECT
EFFECT
Manhood Suffrage
–.28
–.28
Liberties1
–.53
–.31
Union-Party Links
–.44
–.44
Middle-Class Suffrage
–.68
1
INDIRECT
EFFECT
–.22
through Union-Party Links: .52 * (–.44) = –.22
–.68
through Manhood Suffrage: .67 * (–.28) = –.19
through Liberties1: .92 * (–.31) = –.28
through Liberties1 & Union-Party Links: .92 * .52 * (–.44) = –.21
Factor of Civil Liberties and Freedom of Combination
41
Appendix A: Socialist Party Reformism / Radicalism
Year
Orientation
to Political
System
(1-4)
Orientation
to Economic
System
(1-4)
Methods
(1-4)
Dissenting
Factions
Summary
score
(-1 to 1)
(2-13)
Australia: Australian Labour party
1900
1914
1
1
1
1
1
1
0
0
3
3
Austria: Austrian Social Democratic party
1900
1914
3
3
3
3
3
2
1
0
10
8
Belgium: Belgian Workers’ party
1900
1914
2
2
2
2
3
3
0
-0.5
7
6.5
Canada: Canadian Socialist party
1900
1914
1
1
1
1
1
1
0
0
3
3
Denmark: Social Democratic party
1900
1914
2
1
2
2
1
1
0
0
5
4
Finland: Labour Party/ Social Democratic
party
1900
1914
4
4
4
4
3
3
0
0
11
11
France: Parti Socialiste Français (Guesde)
1900
3
3
3
0
9
France: Parti Socialiste de France
(Jaurès)
1900
2
2
2
0
6
France: SFIO
1914
2
2
2
1
7
Germany: Social Democratic party
1900
1914
3
3
3
3
3
2
0
-0.5
9
7.5
Great Britain: Labour party
1900
1914
1
1
1
1
1
1
0.5
0.5
3.5
3.5
Italy: Italian Socialist party
1900
1914
2
3
2
3
2
4
1
-1
7
9
Netherlands: Social Democratic Workers’
party
1900
1914
2
1
2
2
2
1
1
1
7
5
New Zealand: New Zealand Labor party
1914
1
3
1
-0.5
4.5
Norway: Labor party
1900
1914
2
1
2
2
1
2
0
1
5
6
Russia: Russian Communist party
1900
4
4
4
-1
11
Russia: Bolsheviks
1914
4
4
4
-1
11
Russia: Mensheviks
1914
3
3
3
1
9
Spain: Spanish Socialist party
1900
1914
3
3
3
3
2
2
0
1
8
9
Sweden: Social Democratic party
1900
1914
2
1
3
3
3
2
0.5
0.5
8.5
6.5
Switzerland: Social Democratic party
1900
1914
1
1
2
2
1
1
0
0
4
4
United States: Socialist Party of America
1900
1914
2
2
3
3
3
2
1
0.5
9
7.5
42
Bibliography
Archer, Robin. 1998. “Unions, Courts, and Parties: Judicial Repression and Labor Politics in Late
Nineteenth-Century America. “Politics and Society, Vol. 26: 391-422.
Aminzade, Ronald. 1995. “Between Movement and Party: The Transformation of Mid-NineteenthCentury French Republicanism” in J. Craig Jenkins and Bert Klandermans, eds., The Politics
of Social Protest: Comparative Perspectives on States and Social Movements. Minneapolis:
University of Minnesota Press.
Bartolini, Stefano. 2000. The Political Mobilization of the European Left, 1860 – 1980: The Class Cleavage
Cambridge: Cambridge University Press
Berlin, I.. 1969. “Two Concepts of Liberty,” in I. Berlin, Four Essays on Liberty, London: Oxford
University Press. New ed. in Berlin 2002.
Bonnell, Victoria. 1988. “The Representation of Politics and the Politics of Representation,” (Review
Essay) The Russian Review, Vol. 47(4).
Commons, John R. 1926. “Karl Marx and Samuel Gompers.” Political Science Quarterly, Vol. 41(2):
281-286.
Conell, Carol. 1988. “The Local Roots of Solidarity: Organization and Action in Late-NineteenthCentury Massachusetts.” Theory and Society, Vol. 17(3): 365-402.
Conell, Carol and Kim Voss. 1990. “Formal Organization and the Fate of Social Movements: Craft
Association and Class Alliance in the Knights of Labor.” American Sociological Review Vol.
55(2): 255-269.
Cowden, Morton H. 1963. “Early Marxist Views on British Labour.” Political Research Quarterly Vol.
16: 34-52.
Dahl, Robert. 1971. Polyarchy: Participation and Opposition. New Haven: Yale University Press.
Debs, Eugene V. 1909. “Class Unionism.” Speech delivered in South Chicago, Nov. 24. 1905, revised
and reissued 1909. Chicago: Charles H. Kerr and co. cooperative; available at
http://www.marxists.org/archive/debs/works/1905/classunionism.htm and accessed
May 7, 2007. pp 1-20.
della Porta, Donatella. 1995. Social Movements, Political Violence, and the State: A Comparative Analysis
of Italy and Germany. Cambridge: Cambridge University Press.
43
della Porta, Donatella and H. Reiter, eds. 1998. Policing Protest. The Control of Mass Demonstration in
Western Democracies, Minneapolis, The University of Minnesota Press.
Ebbinghaus, Bernard. 1995. “The Siamese Twins: Citizenship Rights, Cleavage Formation, and
Party-Union Relations in Western Europe. International Review of Social History, 40,
Supplement 3: 51-89.
------. 1996. “From Ideology to organization: The transformation of political unionism in Western
Europe.” Pp. 28-59 in The Lost Perspective? Trade Unions Between Ideology and Social Action in
the New Europe, Vol. 2, edited by P. Pasture, J.Verberckmoes, and H. DeWitte. Aldershot:
Avebury.
Eley, Geoff. 2002. Forging Democracy: The History of the Left in Europe, 1850 to 2000. Oxford: Oxford
University Press.
Engels, Friedrich. 1892. Condition of the Working Class in England. 2nd German edition.
-------. 1894a. “Letter to Sorge.” In Marx, Karl and Friedrich Engels. 1953. On Britain. Moscow:
Foreign Languages Publishing House.
-------. 1894b. “Letter to Turati.” Marx, Karl and Friedrich Engels. 1935. Selected Correspondence 18461895; a selection with commentary and notes. New York: International Publishers.
Fine, Nathan. [1928] 1961. Labor and Farmer Parties in the United States, 1828-1928. New York: Russell
and Russell.
Flora, Peter et al. 1983. State, Economy, and Society in Western Europe 1815 – 1975, Vol. I. Chicago: St.
James Press.
------, ed. 1952. Comparative Labor Movements. New York: Prentice-Hall. with John Clarke Adams.
Geary, Dick. 1981. European Labour Protest, 1848-1939. London: Croom Helm.
------. 1989. Labour and Socialist Movements in Europe before 1914. Oxford: Berg.
Goldstein, Robert J. 1983. Political Representation in 19th Century Europe. London: Croom Helm.
Heywood, Paul. 1989. “The Labour Movement in Spain before 1914,” in Labour and Socialist
Movements in Europe before 1914 edited by Dick Geary. Oxford: Berg.
Hu, Li-tze and Peter M. Bentler. 1999. “Cutoff Criteria for Fit Indexes in Covariance Structure
Analysis: Conventional Criteria versus New Alternatives.” Structural Equation Modeling, Vol.
6(1): 1-55.
Katzenstein, Peter J. 1985 Small States in World Markets: Industrial Policy in Europe. Ithaca: Cornell
University Press.
44
Kline, Rex B. 1998. Principles and Practice of Structural Equation Modeling. 1st Edition. New York:
Guilford Press.
-------. 2005. Principles and Practice of Structural Equation Modeling. 2nd Edition. New York: Guilford
Press.
Knapp, Vincent J. 1980. Austrian Social Democracy, 1889-1914. Washington, D.C. : University Press of
America.
Knoellinger, Carl Erik. 1960. Labour in Finland. Cambridge: Harvard University Press.
Lapides, Kenneth. 1986. Marx and Engels on the Trade Unions. Portsmouth : Praeger.
La Palombara, Joseph. 1957. The Italian Labor Movement : Problems and Prospects. Ithaca : Cornell
University Press.
Laslett 2000. John H. M. Colliers across the Sea: A Comparative Study of Class Formation in Scotland and
the American Midwest, 1830–1924. Champaign: University of Illinois Press.
Lenin, V. I. [1902] 1973. What is to be Done? Peking: Foreign Languages Press.
------. [1907] 1962. “Preface to the Russian Translation of Letters by J. Ph. Becker, J. Dietzgen, F.
Engels, K. Marx and Others to F. A. Sorge and Others.” In Lenin, V. I. 1962. Lenin Collected
Works, Vol. 12, pp 359-378.
Les Editions ouvrières. Jaurés et la classe ouvrière. Paris: Les Editions ouvrières.
Lipset, Seymour Martin. 1983. “Radicalism or Reformism: The Sources of Working-Class Politics.”
American Political Science Review, 77: 1-19.
------. 1985. Consensus and Conflict: Essays in Political Sociology. New Brunswick: Transaction printers.
------. 2001. “The Americanization of the European Left.” Journal of Democracy, Vol. 12(3): 74-87.
Lipset, Seymour Martin and Gary Marks. 2000. It Didn't Happen Here: Why Socialism Failed in the
United States. Norton.
Lipset, Seymour Martin and Stein Rokkan. 1967. “Cleavage structures, Party Systems and Voter
Alignments: An Introduction,” in Party Systems and Voter Alignments: Cross National
Perspectives edited by Seymour Martin Lipset and Stein Rokkan. New York: Free Press.
Lipset, Seymour Martin, Martin Trow, and James S. Coleman. 1956. Union Democracy: The Internal
Politics of the International Typographical Union. New York: Free Press.
Lorwin, Val R. 1954. The French Labour Movement. Cambridge, MA: Harvard University Press.
Malefakis, Edward E. 1970. Agrarian Reform and Peasant Revolution in Spain: Origins of the Civil War.
New Haven: Yale University Press.
Marks, Gary. 1989. Unions in Politics: Britain, Germany, and the United States in the Nineteenth and Early
Twentieth Centuries. Princeton: Princeton University Press.
45
-------. 2007. “Triangulation and the Square Root Law." Electoral Studies 25: 1-10.
Marx, Karl. 1842. “On Freedom of the Press.” Rheinische Zeitung No. 132, Supplement, May 12, 1842.
-------. 1877. “Marx to Engels In Ramsgate.“ in Marx, Karl and Friedrich Engels. 1935. Selected
Correspondence 1846-1895; a selection with commentary and notes. New York: International
Publishers.
-------. 1875. “Critique of the Gotha Programme.” Online:
http://www.marxists.org/archive/marx/works/1875/gotha/index.htm In several parts.
Accessed May, 2007.
Marx, Karl and Friedrich Engels. [1872] [1890] 1985. The Communist Manifesto. With an introduction
and notes by AJP Taylor. New York: Penguin.
-------. 1935. Selected Correspondence 1846-1895; a selection with commentary and notes. New York:
International Publishers.
-------. 1953. On Britain. Moscow: Foreign Languages Publishing House.
McAdam, Doug. 1996. “Conceptual Origins, Current Problems, Future Directions,” pp. 23-40 in
Comparative Perspectives on Social Movements, edited by Doug McAdam, John D.
McCarthy and Mayer N. Zald. New York: Cambridge University Press.
McAdam, Doug, Sidney G. Tarrow, and Charles Tilly. 2001. Dynamics of Contention. Cambridge.
Cambridge University Press.
McPhail, Clark and John D. McCarthy. 2005. “Protest Mobilization, Protest Repression and Their
Interaction,” in Christian Davenport, Hank Johnston and Carole Mueller, eds., Repression and
Mobilization. Minneapolis: University of Minnesota Press.
McMinn, W. G. 1979. A Constitutional History of Australia. Melbourne: Oxford University Press.
Michels, Robert. [1916] [1962] 1999. Political Parties : A Sociological Study of the Oligarchical Tendencies
of Modern Democracy. New York: Crowell, a division of MacMillan.
Mikkelsen, Flemming. 2005. “Working-class Formation in Europe and Forms of Integration: History
and Theory.” Labor History, Vol. 46(3): 277-306.
Moore, R. Laurence. 1970. European Socialists in the American promised land. New York: Oxford
University Press.
Nadal, Jordi. 1973. El fracaso de la revolución industrial en España 1814-1913. Revista de Occidente,
Madrid.
Napier, William Francis Patrick. 1857. The Life and Opinions of General Sir Charles James Napier, G.C.B.
Vol. 2, London: John Murray.
46
Osterroth, Franz and Dieter Schuster. 1975. Chronik der deutschen Sozialdemokratie. Bd.1: Bis zum Ende
des ersten Weltkrieges. Bonn: Dietz.
Perlman, Selig and John R. Commons. 1918. A History of Labor in the United States. New York:
Macmillan.
Przeworski, Adam. 1985. Capitalism and Social Democracy. Cambridge: Cambridge University Press.
Przeworski, Adam and John Sprague. 1986. Paper Stones: A History of Electoral Socialism Chicago:
Chicago University Press.
Rokkan, Stein. 1970. Citizens, Elections, Parties. New York: David McKay .
Rueschemeyer, Dietrich, Evelyne Huber Stephens and John D. Stephens. 1992. Capitalist Development
and Democracy. Chicago: University of Chicago Press.
Schakel, Arjan Hille. 2004. The Rise of Social Democracy. MA thesis: Vrije Universiteit Amsterdam.
Shannon, David A. 1955. The Socialist Party of America. New York: Macmillan.
Steenson, Gary P. 1991. After Marx, Before Lenin: Marxism and Socialist Working-Class Parties in Europe,
1884-1914. Pittsburgh: University of Pittsburgh Press.
Tarrow, Sidney. 1996. “States and Opportunities: The Political Structuring of Social Movements,” pp.
41-61 in Comparative Perspectives on Social Movements, edited by Doug McAdam, John D.
McCarthy and Mayer N. Zald. New York: Cambridge University Press.
Tilly, Charles. 1978. From Mobilization to Revolution. Reading, Massachusetts: Addison-Wesley
Publishing Company.
Tilly, Charles, Louise Tilly and Richard Tilly. 1975. The Rebellious Century, 1830-1930. Cambridge:
Harvard University Press.
Tingsten, Herbert. 1973. The Swedish Social Democrats. Totowa, New Jersey: Bedminster Press.
Voss, Kim. 1988. “Labor Organization and Class Alliance: Industries, Communities, and the Knights
of Labor.” Theory and Society, Vol. 17(3): 329-364.
Ward, Norman. 1950. The Canadian House of Commons. Toronto: University of Toronto Press.
Windmuller, John P. 1969. Labor Relations in the Netherlands. Ithaca: Cornell University Press.
47
APPENDIX B: A Measure of Reformism/Radicalism
We conceive reformism versus radicalism as a continuous dimension ranging from the demand for
incremental change within a regime to the demand for its revolutionary transformation. We break
this down into three domains. The first two domains, orientation to the political system and
orientation to the economic system, refer to the substantive character of the demands made by labor,
social democratic and socialist parties concerning the institutions they regard as decisive. The final
domain refers to the means they pursue.
•
Orientation to the political system. To what extent is representative democracy regarded as
legitimate? Is representative democracy worth reforming, or must it be abolished?
•
Orientation to the economic system. To what extent is capitalism regarded as legitimate? Is
it worth reforming, or must it be abolished?
•
Method. Does the party advocate conventional methods of representation or the violent
overthrow of the regime?
Our unit of analysis is the political party at a particular point in time, but we are keenly
aware that political parties harbor dissenting factions, and we adjust for this by evaluating the
strength and ideological orientation of opposition within a party.
Our focus is on the dominant ideological orientation of the party leadership as expressed in
founding documents, party manifestos, leadership speeches, and printed materials. There are
secondary literatures on each of the parties of concern to us, and we rely on these extensively.
Political parties are never monolithic, except by force. Parties are not simply organizations
that compete for authority; they are no less institutional terrains for internal contestation. Their
programs and tactics are often subject to debate and disagreement, and for no party family is this
more true than for socialist parties. Even a party as reformist as the British Labour party contained
marxist voices. In 1906 one of 29 newly elected Labour MPs reported that Das Kapital had the
greatest influence on his thinking. But he was alone: the Bible was most commonly chosen, followed
48
by Pilgrim’s Progress (Stead 1906). Most political parties have a dominant ideological orientation, and
this is the focus of our coding.
We regard the three domains set out above as logically and conceptually distinct. It is
possible, for example, for a socialist party to regard representative democracy as legitimate, but to
advocate the complete abolition of wage labor. Or a party could advocate a revolution in order to set
up representative democracy. Such positions are logically coherent, but they are empirically rare. In
general, a party that takes a radical position in one of these domains takes a radical position on the
others.
This increases the reliability of our coding for we can combine coding across the three
domains to create a composite index of reformism/radicalism. Unless our coding is systematically
biased, the mistakes we make in coding a party in one domain are averaged across three domains.
To the extent that coding errors across the three domains are random, they will tend to cancel.
Our first step is to conceptualize scales for each domain that meaningfully summarize the
variation we have in mind and which also minimize ambiguity (Hooghe, Marks and Schakel 2008).
Below we set these out and examine cases in each category that illustrate the distinctions we make.
Orientation to the political system
1 = representative democracy is legitimate and does not require major reform;
2 = representative democracy is legitimate, though it requires major reform;
3 = representative democracy is illegitimate, requiring major reform;
4 = representative democracy should be abolished and replaced by a dictatorship of the proletariat.
Orientation to liberal democracy provides something of a litmus test for the reformism or
radicalism of socialist, social democratic or labor parties. Did the party demand liberal democracy,
that is, representative democracy in which political parties compete in free elections based on
49
manhood (later, universal) suffrage, or did the party distance itself, or reject, representative
democracy as a bourgeois institution?
The Australian Labor party campaigned for electoral reform, within the context of liberal
democracy. So, for example, it demanded that residential requirements be relaxed to allow transient
workers, such as sheep shearers, to vote. The Labor party campaigned for a variety of incremental
reforms, all of which it believed could be achieved under liberal democracy. Conflict between
workers and employers was deemed inevitable under capitalism, but the state itself was regarded as
a neutral player. 1 [Score = 1]
The British Labour party was established in 1900 as the Labour Representation Committee to
gain greater representation of workers in the House of Commons. As its name suggests, the party
accepted liberal democracy as the frame for its substantive demands. Reform of the political system
was a low priority. The party’s 1900 manifesto called for 16 reforms, the final four of which were
shorter parliaments, adult suffrage, registration reform, and payment of members of parliament.
These demands were almost identical to the unfulfilled demands of the Chartists more than half a
century earlier, but now they were regarded as an extension of principles that were already in place,
and we score the party 1. 2 [Score =1]
1
The notion of the neutrality of the state underpinned the Labor party’s proposal for state arbitration of
labor disputes.
2
The 1900 Labour Party General Election Manifesto contained the following demands: adequate
maintenance from national funds for the aged poor; public provision of better houses for the people;
useful work for the unemployed; adequate maintenance for children; no compulsory vaccination; public
control of the liquor traffic nationalisation of land and railways; relief of local rates by grants from the
national exchequer; legislative independence for all parts of the empire; abolition of the standing army,
and the establishment of a citizen force; the people to decide on peace or war; graduated income-tax;
shorter parliaments; adult suffrage; registration reform; payment of members.
50
The legitimacy of republican government was a major issue for French socialists around the
turn of the twentieth century and the divergent positions taken by Jean Jaurès and Jules Guesde
illustrate well the distinction we make between the intermediate values on our scale. From the late
1890s, Jaurès regarded French Republican institutions not merely as legitimate, but as the
foundation for socialist success. Locked between marxists who regarded republican democracy with
indifference and a revanchist right that wished to destroy it, Jaurès threw himself into the defense of
the Third Republic. Just as the republic grew out of a progressive conception of popular sovereignty,
so socialism, with its concept of economic sovereignty, would grow out of the republic. Jaurès’
notion of republicanism was a radical version of liberal democracy for it contained within it the
notion of inevitable socialist victory and the possibility that the working class would have to engage
in extra-parliamentary tactics, such as a general strike, were the ruling class to revoke workers’
rights. 3 [Score = 2]
Jules Guesde and the Parti Socialiste, by contrast, viewed the Third Republic as illegitimate,
as a capitalist institution that was incapable of ameliorating economic exploitation. Its essential
virtue was that, by obliterating feudalism, it intensified the clash of proletariat and bourgeoisie, and
therefore accelerated revolution. In the words of Guesde: “All French socialists are republicans
because the republic is the only governmental form that places the classes against one another in the
3
“Finally, if the governing class were mad enough to lay hands on the poor liberties that have been won,
the wretchedly insufficient means of action of the proletariat, if they threatened or attacked universal
suffrage, if by the persecution of employers and the police they made the right to unite in trade unions
and the right to strike empty forms, then a violent general strike would be certainly the form that a labour
revolt would take” (Jaurès 1906).
51
brutality of their economic opposition . . .” (Guesde 1891, quoted in Steenson 1991: 157; see also
Howarth 1983: 4; Derfler 1998: 48). 4 [Score = 3]
Only the Bolshevik Finnish and Russian parties, among those we examine, took the extreme
position that representative democracy expressed the dictatorship of the bourgeoisie which would
have to yield to a dictatorship of the proletariat. Writing in 1905, Lenin took a similar position to
Guesde on the issue of republican democracy:
“[Social democrats] have never been afraid of saying, and never will be, that they sanction
the republican-democratic bourgeois order in preference to an autocratic serf-owning
bourgeois order. But they “sanction” the bourgeois republic only because it is the last form
of class rule, because it offers a most convenient arena for the struggle of the proletariat
against the bourgeoisie; they sanction it, not for its prisons and police, its private property
and prostitution, but for the scope and freedom it allows to combat these charming
institutions.” (Lenin 1905)
However, Lenin had a clear conception of the transitional political order that would replace
the bourgeois order: a dictatorship of the proletariat which “imposes a series of restrictions on the
freedom of the oppressors, the exploiters, the capitalists. We must suppress them in order to free
humanity from wage slavery, their resistance must be crushed by force; it is clear that there is no
freedom and no democracy where there is suppression and where there is violence” (Lenin 1917). 5
[Score = 4]
4
The apparent success of Guesde and his allies in setting the terms of unification of the two socialist
parties in 1905 was pyrrhic. The unified party came closer to Jaurès’ position in the years preceding
World War I, as reflected in our coding.
5
We can find no reference to a dictatorship of the proletariat in Guesde’s writing. In a brutal attack on
Karl Kautsky in 1918, Lenin wrote that the fundamental task of the proletariat was ““smashing” the
52
Orientation to the economic system
1 = capitalism is legitimate, and can be improved by piecemeal reform;
2 = capitalism is legitimate, but requires basic reform;
3 = capitalism is illegitimate and reform can be justified only as a step to its abolition;
4 = capitalism is illegitimate and all reform short of its abolition is unproductive.
Except for the first labor party platform in Queensland, the early Australian labor parties
demanded incremental economic reforms—a land tax, maximum hour and minimum wage
legislation, industrial hazard legislation—that accepted the principle of wage labor and private
ownership of the means of production. [Score = 1]
Likewise, the British Labour party campaigned for slightly expanded government
responsibilities within a capitalist economic system that it regarded as legitimate. The first five
demands of the party’s general election manifesto of 1900 were “adequate maintenance from
national funds for the aged poor, public provision of better houses for the people; useful work for
the unemployed; adequate maintenance for children; and no compulsory vaccination.” [Score = 1]
Despite its well-earned reputation for revolutionism following World War I, in its early
years, the Norwegian labor party viewed itself as a “practical, political people’s party” concerned
with “the demands of the day.” “It joined the Second International, and faithfully included the basic
tenets of social democracy in its annual statements of policy. But socialism was primarily a slogan
and a vague hope for the future” (Galenson 1949: 58-59). The party made basic demands for political
regulation of working hours and wages, but these steps were defended in their own terms,
bourgeois state machine” and quotes Marx to the effect that “Dictatorship is rule based directly upon
force and unrestricted by any laws.”
53
irrespective of whether they would lead to the abolition of the system of wage labor. 6 We code the
Norwegian labor party in the first intermediate category on its economic ideology. [Score = 2]
The Swedish party was more radical in that ameliorative economic reforms were viewed
from the standpoint of abolishing capitalism. According to its principal historian, Herbert Tingsten,
“Swedish Social Democracy was certainly not critical or opposed to social reforms of limited scope . .
. But it is characteristic of the early debate that all improvements to this kind were regarded as
comparatively unimportant . . . Social welfare measures were regarded as palliatives whose greatest
value was often said to be that they spurred the workers on to new and more fruitful efforts (1973:
697). [Score = 3]
Once again, the Bolshevik and Finnish socialist parties anchor the radical extreme. Whereas
the Menshevik party conceived of economic reforms as useful steps on the path to socialism, the
Bolshevik party conceived economic reform as a distraction from the revolutionary struggle. Lenin
ridiculed the notion, dear to orthodox German marxists and Mensheviks alike, that trade union
organization bolstered socialist consciousness: “[T]he pompous phrase about ‘lending the economic
struggle itself a political character’, which sounds so ‘terrifically’ profound and revolutionary, serves
as a screen to conceal what is in fact the traditional striving to degrade Social-Democratic politics to
the level of trade union politics” (1902: 5). [Score = 4]
Although the Finnish socialist party was influenced by Swedish and German socialism more
than Bolshevism, it took a similar line, regarding unions and labor market activity not only as
subsidiary to the political tasks of the party, but as futile. The leadership of the Finnish socialist
party procrastinated in forming a union wing, and when the party eventually established a union
federation in 1907, it regarded unions as subordinate to the party (Knoellinger 1960:47). [Score =4]
6
Radical marxists and syndicalists gained support within the party prior to World War I, but were able to
take control of the party only towards the end of the war.
54
Method
1 = the party advocates conventional political activity;
2 = the party advocates unconventional alongside conventional political activity to defend basic
political rights;
3 = the party advocates unconventional alongside conventional political activity as part of a
revolutionary strategy;
4 = the party rejects conventional political activity and advocates the violent overthrow of the
existing regime.
Most labor parties, including the British, Canadian, and New Zealand parties, advocated
conventional liberal democratic methods, including non-violent mass demonstrations, distributing
propaganda, pressuring and petitioning governments, and campaigning in elections. Prior to 1905,
when the party began its turn to the left, this category of reformist parties included the Norwegian
labor party. After suffrage was extended to all Norwegians over 25 in 1898, the Norwegian labor
party advocated a conventional political strategy. While not abandoning its socialist principles, the
party focused on "the demands of the day and the realizable possibilities, the tactics of relationships
with the Liberal Party and the possibility of establishing contact with the masses” (Koht et al, 1937:
170, quoted by Galenson 1949: 58). [Score =1]
Despite its revolutionary rhetoric, the Spanish socialist party adopted cautious political
methods. The tension between formal ideology and actual practice was perhaps more acute in Spain
than anywhere else. In the early 1900s, the party predicted imminent revolution, yet was intent on
protecting its organization from repression by adhering to strictly legal methods. In Evolutionary
Socialism, Eduard Bernstein cited Pablo Iglesias, the leader of the Spanish socialist party, for
abandoning revolutionary violence: “’The bourgeoisie, of whatever shade of opinion it may be,’
declared lately the Spanish socialist, Pablo Iglesias, ‘must be convinced of this, that we do not wish
55
to take possession of the Government by the same means that were once employed, by violence and
bloodshed, but by lawful means which are suited to civilisation’” (Vorwärts 1898, quoted by
Bernstein 1899). [Score = 2]
After 1912, the Italian socialist party moved from a similar reformist position to extreme
revolutionism based in the belief that “it was a matter of certainty that the revolution would
eventually take place and that it would be generated by the fury of proletarian violence" (Sassoon
1996: 75). Benito Mussolini was then a leading maximalist who condemned the aging moderate
leadership in the party and, in his journal, Utopia, tried to summon a wave of revolutionary violence
that would topple the regime. In 1924, Gramsci recollected that as a young socialist he was himself a
maximalist who, “saw a solid wall upon which to turn the full weight of our forces with an energetic
act of will. All or nothing, isn't that what we said?” (Craver 1996). [Score = 4]
Dissenting factions
−1.0 = a large moderate wing or moderate social democratic party confronts the mainstream of the
party
−0.5 = a small moderate wing or moderate social democratic party confronts the mainstream of the
party;
+0.5 = a small radical wing or radical social democratic party confronts the mainstream of the party;
+1.0 = a large radical wing or radical social democratic party confronts the mainstream of the party;
All socialist parties contained factions that opposed the strategic or ideological choices of the
leadership. In some cases, these factions broke off into small competing parties. Most parties had
both radical and reformist factions, but at any given time one wing or the other was usually
ascendant, and we recognize this in our coding by adding 0.5 or 1.0 to a party’s score where the
dominant faction is radical in relation to the party leadership and subtracting 0.5 or 1.0 from a
party’s score where the dominant faction is reformist.
56
We wish to take some account of the relative strength of a faction in relation to the party
leadership, as well as its ideological leaning. A faction scores +1.0 or −1.0 where the secondary
literature reports that it commands around a third or more of delegates to the party congress in the
year of evaluation, or if the faction is perceived to threaten the tenure of the party leadership.
57
Chief sources for coding the dependent variable:
Australia: Burgmann 1985: 196-198; 180-181; McKinlay 1981: 3-8,12-14, 19; Patmore 1991: 80-83.
Austria: Knapp 1980: 41, 42, 60-61, 72, 153, 189-202; Sassoon 1996: 21; Shell 1962: 9-11, 22;
Steenson 1991; Sully 1982: 13-18, 22.
Belgium: Dodge 1979: 51; Fitzmaurice 1996: 27-29, 181-182; Henig and Pinder 1969: 72; Liebman
1979: 111, 128, 130-134, 145, 149, 163; Mabille 1992: 201-202; Strikwerda 1997: 111, 272.
Canada: Heron 1989: 47-49; Palmer 1983: 149-150, 158-161.
Denmark: Christiansen 1978: 304-107; Galenson 1952: 149; Sassoon 1996: 11-13.
Finland: Alapuro 1988: 103-104, 122-127; Nousiainen 1971: 21-26; Puntila 1975: 70-74; Sassoon
1996: 11-13; Soikkanen 1978: 348.
France: Brunet 1989: 21-22; Derfler 1973: 67, 70-73; Derfler 1998; Eley 2002 : 87 -88; Geary 1989:
78-79; 86; Goldberg 1962: 66, 261; Jaurés et la classe ouvriere: 91; Judt 1986: 118-119; Kruezer
2001: 25-30; Moss 1980: 90; Ridley 1970: 54-44, 92-93; Sassoon 1996: 11-13; Steenson 1991;
Williams 1983: 4-7, 14-25.
Germany: Eley 2002: 68; Fletcher 1987: 24; Geary 1989, 129; Groh 1973; Henig and Pinder 1969:
35; Kruezer 2001: 25-27, 30-32; Osterroth and Schuster 1975: 50ff; Sassoon 1996: 11-13;
Schorske 1981; Steenson 1991.
Great Britain: Eley 2002: 64, 89-90; Marks 1989; Rubenstein 2006, 11-50; Sassoon 1996: 15-18;
Thomson 2006: 10-76.
Italy: Davis 1989: 182-194; de Grand 1989: 17-19, 25; 423; Henig and Pinder 1969, 193-196; HiltonYoung 1949: 42-49, 55, 72-75; Neufeld 1961: ch. 8; Steenson 1991.
Netherlands: Eley 2002: 68; Hansen 1977: 204, 200, 221; Henig and Pinder 1969: 263.
New Zealand: Gustafson 1980: 13, 39, 70-76, 82, 84.
Norway: Galenson 1952: 149-151; Sassoon 1996: 11-13.
Russia: Galenson 1952; Gooding 2002: 34- 61; Brovkin 1987: 2; Fineberg 1934: 36-50; Hildermeier
2000: 51-89; Read 1989: 153-154, 180.
58
Spain: Heywood 1989: 231-235; 263; 247-256; Gillespie 1989: 8-9, 17, 25-35; Heywood 1990: 7, 1115.
Sweden: Eley 2002: 67; Galenson 1952: 150-151; Sassoon 1996: 1-13, 21; Tingsten 1973: 149, 150,
168, 192, 378, 396, 422.
Switzerland: Biucchi 1973: 627-653; Henig and Pinder 1969: 365, 376; Kohler 1977: 28, 46; Lang
1977: 140; Masnata 1963: 107-114.
United States: Chester 2004: 39-83; Esposito 1997: 34-36, 45-78, 221-250; Lipset and Marks 2000;
Shannon 1955: 78-79.
59