(New Studies in Sociology) Nicos P. Mouzelis (Auth.) - Politics in The Semi-Periphery - Early Parliamentarism and Late Industrialization in The Balkans and Latin America-Macmillan Education UK (1986)

Politics in the Semi-Periphery
Politics in the
Semi-Periphery
Early Parliamentarism and Late
Industrialization in the Balkans
and Latin America
Nicos P. Mouzelis
Macmillan Education
ISBN 978-0-333-34934-2
ISBN 978-1-349-18019-6 (eBook)
DOI 10.1007/978-1-349-18019-6
Nicos P. Mouzelis 1986
Softcover reprint of the hardcover 1st edition 1986 978-0-333-34933-5
All rights reserved. For information, write:

St. Martin's Press, Inc., 175 Fifth Avenue, New York, NY 10010
Published in the United Kingdom by Macmillan Education Ltd.
First published in the United States of America in 1986
ISBN 978-0-312-62886-4
Library of Congress Cataloging in Publication Data
Mouzelis, Nicos P.
Politics in the semi-periphery.
Bibliography: p.
Includes index.
1. Political participation-Balkan Peninsula-History.
2. Political participation-South America-History.
3. Civil-military relations-Balkan Peninsula-History.
4. Civil-military relations-South America-History.
5. Balkan Peninsula-Industries-History. 6. South
America-Industries-History. I. Title.
JF2011.M68 1985
323' .042
84-27738
ISBN 978-0-312-62886-4
To the memory of my father
Contents
Acknowledgements
xi
General Introduction
xiii
PART I
1
Modes of Transition to Post-oligarchic Politics:

The Demise of Oligarchic Parliamentarism in the
Balkans and Southern Latin America
Introduction
1.1 Socio-economic processes leading to the early
demise of oligarchic parliamentarism
1.2 Routes to post-oligarchic politics
1.2.1 Urban populism
1.2.2 Peasant populism
1.2.3 The transformation of clientelistic networks
1.3 Initial stages of industrialisation and the incorporation of the industrial working classes
1.3.1 Industrialisation
1.3.2 The working-class movement in the Balkans
1.3.3 The working-class movement in southern
Latin America
62
Conclusion
72
Theoretical Implications (1):

Clientelism and Populism as Modes of Political
Incorporation
73
2.1 On the concept of incorporation

2.2 Populism and clientalism as modes of political
inclusion
2.3 Some basic differences between clientelistic and
populistic incorporation
2.4 From clientelistic to populistic in politics
2.5 The debate on the nature of populism
vii
3
7
15
15
29
39
50
51
55
73
76
78
83
88
vm
Contents
PART II
3
Routes to Military Dictatorship:

A Comparative Essay on Argentina, Chile and Greece
3.1 The military in the early post-oligarchic era
3.1.1 The military's role during the transition
3.1.2 Variations in civil-military relations: the
conservatives' post-oligarchic electoral
strength and army dominance
3.2 The post-war expansion of industrial capitalism:
restricted and uneven development
3.3 On the basic political structures of Argentina,
Chile and Greece
3.3.1 The limiting framework
3.3.2 Vertical incorporative modes of inclusion
3.3.3 The major political contradiction
3.4 On the post-war rise of military dictatorial regimes
3.4.1 Greece
3.4.2 Argentina
3.4.3 Chile
3.4.4 Conclusion
3.5 Postscript: a short note on the structure and
dynamics of the post-war military dictatorships in
Argentina, Chile and Greece
97
97
99
105
113
122
122
126
129
134
134
145
158
170
177

Praetorianism, Bureaucratic Authoritarianism, and
Problems of Reductionism in Marxist Political Theory 184
4.1 Military interventions and theories of political
modernisation
4.2 The rise of bureaucratic-authoritarian regimes:
notes on the relevant literature
4.3 On the relationship between the economic and the
political: problems of reductionism in Marxist
political theory
4.4 Relations of production and relations of domination:
the relevance of the distinction for the study of
political development in the semi-periphery
184
189
199
206
Contents
General Conclusion
Notes and References
Index
ix
219
224
270
Acknowledgements
I would like to thank the foilowing persons who, at one time or

another, have read parts of this work and offered me comments
that proved very useful: Omar Bayoumi, Rodi Calderon, George
Catiphores, Percy Cohen, Nikiphoros Diamandouros, Nicos Garganas, Mike Gowan, John Hall, Vasilis Kapetanyiannis, Cristobal
Kay, Christos Lyrintzis, Shelia Massoliver, David Rock, David
Rosenberg, Ian Roxborough, Philip Schmitter, Leslie Sklair,
Anthony Smith, Constantine Vaitsos, Thanos Veremis, and Costas Vergopoulos.
My particular thanks go to Gustavo d' Angelo and Luis Ortega
as well as to Jose Munoz, Jose-Miguel Albala and Ximena
Nascimento for their constructive advice and help with Spanish
texts.
Very much gratitude is due also to George Dertilis and Mick
Mann for very carefully going over the entire manuscript and
giving me both useful advice and moral support.
My warm thanks go to Ellen Sutton for her excellent language
editing and for preparing a bibliography which, for reasons of
space, had to be dropped.
I would finally like to express my gratitude and love to Claire
Baron for her constant moral and practical support.
N.P.M.
xi
While it may be conceded that there are systematic differences

between western European capitalist development on the one
hand, and on the other the development or underdevelopment of
all those countries that have tried to 'catch up' with the West a
century or more later, there is nevertheless such a complexity and
enormous variety of political and economic systems in the socalled third world that any attempt to discuss the nature and
structure of the state and politics in these societies without
differentiating between them is bound to fail. It can only result in
generalisations which are either platitudinous or inconclusively
vague.
In view of these difficulties the present work attempts to go
beyond the third-world blanket approach, by focussing on a
limited number of societies that have certain crucial features in
common with respect to the timing of their political independence,
their adoption of parliamentary institutions, and their development of industrial capitalism.
Despite the geographical distance and the obvious differences in
cultural and historical backgrounds, Greece and (to a lesser
extent) the major northern Balkan societies before their post-war
collectivisation 1 show marked and significant similarities with the
'advanced' countries of Latin America's southern cone. 2 Until the
beginning of the nineteenth century, Balkan as well as Latin
American societies were subjugated parts of huge patrimonial
empires (the Ottoman and the Iberian respectively), and as such
never experienced the absolutist past of western and southern
European societies. 3 Moreover, both the Balkan and southern
Latin American countries acquired their political independence in
the nineteenth century and immediately set about implementing
western parliamentary forms of political rule. Notwithstanding the
xiii
xiv
constant malfunctioning and fragility of such forms, their parliamentary institutions evinced a surprising degree of resilience,
surviving and functioning more or less intermittently from the
second half of the nineteenth century until the 1930s in the case of
the northern Balkan societies, and until the rise of military
authoritarian regimes in the 1960s and 1970s for Greece and the
southern cone countries - the latter's dictatorial regimes, as the
Greek and Argentinian cases suggest, not necessarily entailing the
irreversible demise of parliamentary democracy.
On the economic level, despite their relatively late start and
their failure to industrialise in the last century, Greece, the
northern Balkan and the southern Latin American countries all
managed, through the development of their export sectors, first to
build up a significant economic infrastructure, and then to achieve
a rather impressive degree of industrialisation during the inter-war
and post-war years.
This group of societies might be accurately characterised as
'late-late' industrialising capitalist societies4 with early and persistent
quasi-parliamentary politics. Since this is a rather cumbersome
label, the term parliamentary semi-periphery, or simply semiperiphery, will be used here. I am fully aware of the difficulties
inherent in the centre-periphery terminology, especially as these
concepts have been applied in the tradition of writings following
Gunder Frank's and Emmanuel Wallerstein's theories. 5 Given
however that none of the concepts pertaining to the so-called
third-world capitalist societies (that is, developing, underdeveloped, dependent, backward, and so on) are free of such
difficulties, I have opted for the centre-periphery terminology with the proviso that my use of it is not taken to imply acceptance
of Frank's or Wallerstein's views on capitalist development and on
the specific mechanisms which, on the world-economy level,
create centres, peripheries, and semi-peripheries. 6 Neither is the
label of parliamentary semi-periphery meant to imply an alternative theory of development/underdevelopment, or to operate as an
exclusive category referring to a specific set of societies strictly
different from the rest of the third world. (In fact, some of the
generalisations formulated here are not without relevance to other
societies which are either less industrialised or which have adopted
parliamentary-democratic forms of rule somewhat later.)
In brief, the term parliamentary semi-periphery carries neither
xv
any great theoretical pretensions, nor is it the product of a

complex and rigorous typology; it is simply being used here as a
kind of shorthand, as a convenient term for referring to a number
of societies all of which, unlike most other third-world countries,
have experienced both advanced industrialisation and a long
history of parliamentary-democratic rule. This combination of
early and persistent parliamentarism and late industrialisation
makes these countries quite comparable to western European
societies. In view of this, certain economic and political patterns in
the parliamentary semi-periphery are more or less explicitly
compared in several parts of the book with those of the centre at
similar stages of development. I realise of course the dangers of
oversimplification and overgeneralisation entailed in such a comparative perspective of centre v. semi-periphery. Nevertheless, I
hope to show that, despite these difficulties, adopting a broad,
historically-oriented perspective does help towards a better understanding of the actual socio-political developments experienced by
these countries. A similar though less systematic effort has been
made to show some of the differences between the semi-periphery
and the more peripheral capitalist societies.
Having clarified my use of the term parliamentary semiperiphery, it remains to outline the chief concerns of this book. Its
major purpose is to show the fruitfulness of a Balkan-Latin
American comparison by exploring the complex ways in which
long-term, broad changes in the economy and society can help us
explain certain crucial political transitions which occurred at
approximately the same time in the countries under consideration.
The patrimonial type of rule experienced during their preindependence period by both Balkan and southern Latin American societies was hostile to the autonomous consolidation of corps
intermediaires between the ruler and the ruled. Despite significant
differences in patterns of landownership for instance, in terms of
state-civil society relationships neither Ottoman nor Iberian rule
was prepared to tolerate any political autonomy of merchants or
aristocrats vis-a-vis the central authority of the ruler. Whatever
associations or collective bodies did represent, non-state interests
were despotically controlled from above - at least when the ruler
was strong (and so were the various officials manning the highly
particularistic state apparatuses 7). This situation is of course in
sharp contrast to the subtle and, for some historians, unique
xvi
balance of power between the ruler and the aristocracy (and later
the bourgeoisie) in western European absolutism. 8
The authoritarian/despotic features of the patrimonial state in
the Balkans and southern Latin America did not disappear once
these countries had acquired their political independence - not
even when they adopted western parliamentary institutions during
the second half of the nineteenth century. Indeed, the nineteenthcentury parliamentary regimes in the semi-periphery were based
on very restricted popular participation, and on an authoritarian
particularistic state controlled by a handful of notable families such families being able to keep the parliamentary system functioning stably by manipulating the electorate through a variety of
legal and illegal means. However, with the fuller integration of
these semi-peripheral societies into the world economy, their
restrictive parliamentary regimes began to weaken as processes of
market, state and city expansion undermined traditional mechanisms of political control and generated new political forces which
eventually challenged the oligarchic monopoly of state power.
Part I of the book explores the demise of oligarchic parliamentarism and the mode of transition to broader forms of political
participation - a transition which in all the countries under
consideration took place during the first three decades of the
twentieth century. Three distinct forms of transition are identified:
(i) urban populism (the southern Latin American pattern), (ii)
peasant populism (the pattern in the northern Balkan societies),
and (iii) the broadening of political participation through the
extension/transformation of already existing political patronage
networks (the Greek pattern).
One of the major arguments to be developed extensively in Part
I is that, despite the wide variations in the mode of transition, in all
the semi-peripheral societies under study the demise of oligarchic
parliamentarism occurred in a predominantly pre-industrial context, and consequently the opening-up of the political system was
not marked, as in several western European countries, by the
active participation of the industrial classes (particularly by massive and relatively autonomous working-class organisations). Instead, the new middle- and lower-class participants and their
organisations were brought into the political arena in a more
dependent/vertical manner, and this type of vertical political
inclusion, which I call incorporative, did not eliminate the particu-
xvii
laristic/personalistic features of parliamentary politics; neither did

it manage to check the authoritarian/despotic tendencies of the
state. Moreover, when in the period after the 1929 Depression
industrial capitalism gained momentum in these countries, its
timing and structure meant that it did not weaken but rather
transformed and reinforced these authoritarian structures inherited from the pre-independence era.
Therefore, both the post-oligarchic broadening of political
participation and the subsequent acceleration of the industrialisation process contributed, in different ways, to a polity with a highly
unequal distribution of political power between rulers and ruled.
This unequal distribution of power and the institutional structures
associated with it constituted a negative legacy; it provided a very
inadequate basis for dealing democratically with the staggering
problems that post-war semi-peripheral societies had to face once
their economies were becoming more industrialised and their
politics more fully marked by high levels of mass participation.
In Part II of the book the focus is restricted to only three
societies of the parliamentary semi-periphery (Greece, Argentina
and Chile) in order to show in greater detail the precise relevance
of the political arrangements set up in the early post-oligarchic
period for understanding the advent of highly repressive military
dictatorial regimes in the post-war years. In identifying the basic
structures and long-term processes which can help to explain more
adequately the post-war rise of military dictatorships in Argentina,
Chile and Greece, I have attempted to show both the elements
held in common by these three countries as well as the systematic
variations in their socio-political trajectories.
As I hope will become clear, the aim of this work is not to
neglect the fundamental and fairly obvious differences that exist
between the societies under consideration by fitting them into a
procrustean bed of structural similarities. The aim is rather to
show that those features they have in common provide a basis for a
systematic comparison and explanation of their differences. Given
the book's broad scope and the number of cases involved (especially in Part I) there is no attempt to deal in strictly chronological
fashion with the various phases of economic and political development from the nineteenth century onwards; neither are events in
each of these countries analysed exhaustively and methodically.
Instead, examples from individual countries are used selectively to
xviii
illustrate some more general trends. For instance, in discussing (in

Chapter 1) peasant populism as a form of transition to postoligarchic politics, Bulgaria's peasant movement (as the most
powerful and representative in the Balkan peninsula) is the only
one discussed in ~orne detail. The same is true about the analysis of
urban populism, where I consider in detail only the Argentinian
case. It is in this way that I have tried to resolve the methodological dilemma of extensive but superficial coverage versus a restrictive analysis in depth.
In other words, the present work is not a detailed history of
economic and political developments in the parliamentary semiperiphery. It simply tries, by means of a broad and historicallyoriented comparative analysis to identify certain common patterns
of development, as well as certain significant variations in the
social structures of the societies involved. A greater awareness of
these common patterns and variations is, I think, helpful for a
better understanding of both the overall highly chequered trajectory of these countries' parliamentary institutions, and of the two
critical turning points in this trajectory: the early-twentiethcentury demise of oligarchic parliamentarism that led to a significant broadening of political participation and to the establishment of different types of inc/usionary political regimes; and the
post-war rise of exclusionary military regimes that tried through
systematic repression to keep the masses out of active politics.
As the List of Contents shows, both Part I and Part II have a
theoretical chapter which tries to draw out some of the implications of the preceding empirical analysis for the relevant literature.
So Chapter 2 (Part I) clarifies and develops theoretically the
concepts of populism, clientelism and political incorporation in the
light of the concrete analysis of modes of transition to postoligarchic politics discussed in Chapter 1. This theoretical elaboration is intended both as a contribution to the general debate on the
nature of populism and clientelism, as well as to prepare the
ground for the analysis of post-war political developments that
follows. Chapter 4 (Part II) deals with certain crucial debates in
the literature on military interventions generally, and on the
post-war rise of military dictatorial regimes particularly. Given the
book's constant endeavour to relate developments in the economy
to those taking place in the polity, the same chapter also discusses
the more general and abstract issues of economic reductionism in
xix
Marxist theory and their relevance for the study of the two types of
political transition under consideration.
In sum, the book is intended as a modest contribution to two
interrelated areas of study. These are:
1. The comparative, cross-regional analysis of political development in a number of countries which, in terms of both their
capitalist industrialisation and their early adoption of parliamentary institutions, come quite close to the western
bourgeois parliamentary democracies.
2. The more theoretical writings in political sociology that deal
with such issues as the fragility of democratic parliamentary
regimes in third-world countries, the post-war rise of
bureaucratic-authoritarian military rule in several late industrialisers, as well as (on a more abstract level) the nature of
populism and political clientelism, and the adequacy of certain
paradigms (particularly the Marxist one) for the study of
politics in peripheral and semi-peripheral capitalist societies.
This multiplicity of aims entails the risk, of course, of falling
between two or more stools. I have tried to minimise this risk both
by giving more weight to the empirically-oriented comparative
chapters, and by linking these chapters as closely as possible to the
theoretical discussions that follow them. In any case, the present
work does not claim to offer any definite or final solutions to the
empirical and theoretical issues it raises; it merely presents a
number of tentative ideas which, I hope, may stimulate empirical
research and advance the more abstract debates on some basic
themes and concepts of political theory.
A final word concerning the format and general organisation of
the book. As I have tried to make the more empirical essays
(Chapters 1 and 3) fairly self-contained, a certain amount of
repetition has been unavoidable. At the same time, so as to render
the major themes and arguments as concise as possible, I have
used the Notes and References extensively to develop certain
secondary arguments and, more generally, to provide additional
information both theoretical and empirical. Although a long Notes
and References section is not popular with publishers nowadays, I
strongly believe that the reader should be given the choice of
different readings according to his or her interests and time
available.
PART I
Chapter 1
Modes of Transition to
Post-oligarchic Politics:
the Demise of Oligarchic
Parliamentarism in the Balkans
and Southern Latin America
Introduction
As already mentioned in the General Introduction, a distinctive

characteristic of the societies under study is that not only did they
obtain their independence in the nineteenth century, but also that
during the second half of that century they all had long periods of
parliamentary rule, albeit of an oligarchic, restrictive type. By
oligarchic parliamentary rule I mean a system of government
where active politics was the concern of a handful of notable
families, these families managing to maintain a liberal, pluralistic
system of representation (with the usual freedoms of speech,
association, and so on), while at the same time keeping the bulk of
the lower classes excluded from the political arena. The means for
keeping the people outside politics were either legal (for example,
voting rights being conditional on property and/or educational
qualifications) or, as was the case in Greece where male suffrage
became universal as early as 1864, they resided in the capacity of
the local potentates to control lower-class votes more or less
automatically through fraud, coercion, or other forms of political
manipulation.
It follows from the above, that in oligarchic parliamentary
regimes the typical political forces consist of loose associations of
notables, of political clubs rather than well-organised parties with
a mass following. 1 In all the countries under consideration this
3
restrictive form of parliamentary government began to weaken

towards the end of the last century, followed by the actual
breakdown of the political oligarchy's monopoly and the opening
up of the representative system to new political elites during the
first three decades of the twentieth century. In Greece, for
instance, the turning point was the military coup of 1909 and the
subsequent creation and rapid rise to power of Venizelos's Liberal
Party; in Bulgaria it was the ascent of Stamboliiski's Agrarian
Union to power in 1919; in Argentina, the 1912 adoption of Saenz
Peiia's electoral reform and the subsequent rise to power of
Yrigoyen's Radical party in 1916; in Chile it was the army's
intervention of 1924 that broke up the oligarchic system, and so
on.
A major point I would like to develop in this chapter is that,
given the 'late-late' industrialisation in both Latin America and the
Balkan peninsula, the demise of oligarchic parliamentarism and
the transition to broader forms of political representation occurred
at a time when industrial capitalism2 was both weak, and its
establishment such that its effect on the transition is in no way
comparable with the impact the western European industrialisation had on the process of political development.
In western Europe, capitalist industrialisation was one of the
main processes leading to the transition from a restrictive/oligarchic system of government based on clubs of notables to one based
on broadly organised political parties. In fact, alongside the
development of industry, of national networks of communication
and markets, a process of large-scale political mobilisation took
place which drew more and more people into the national political
arena, transforming them from 'subjects' to 'citizens' ,3 and this
drawing-in process resulted in an expansion of the bureaucratic
state as well as the broadening of political participation. 4 Whether
one looks at societies (for example, England, Switzerland, Belgium, the Scandinavian countries) which had a relatively smooth
transition to broader forms of political participation, or at those
with a more chequered path (France and Germany), the point is
that in all these societies broadly organised political parties
emerged several decades after industrialisation had advanced
considerably. 5
Exactly the reverse is true for the countries of the parliamentary
semi-periphery. There the demise of oligarchic politics and the
Modes of Transition to Post-oligarchic Politics
transition from political 'clubs' to parties occurred before these

countries experienced large-scale industrialisation. In fact, not
only in the nineteenth century, but even in the first three decades
of the twentieth (the period of the demise of oligarchic politics),
neither the Balkan nor the Latin American societies achieved
industrialisation levels comparable to those of western Europe
during her transition (in the second half of the nineteenth century)
from oligarchic to non-oligarchic politics. Even in Argentina, by
far the most industrialised of the semi-peripheral countries where
the economy underwent a major process of import-substitution
industrialisation before 1929, the 1930 share of industrial production in the Gross National Product was only 22.8 per cent. What is
more important, before 1929 most of Argentina's industrial activities were closely linked to the export sector, consisting of processing local raw materials and semi-manufactured imported consumer goods. Moreover, the vast majority of industrial establishments
was of artisanal!family units employing no or very few wage
labourers. 6 The same is true for Greece, which was the most
industrialised of the Balkan societies during the inter-war period. 7
This difference in timing, this disparity between the decline of
traditional/restrictive political structures and the development of
capitalist industry, is a key consideration for understanding the
shape that political institutions subsequently took in the semiperiphery. If this point has not so far been stressed and discussed
extensively in the relevant literature, it is because neither the
neo-evolutionist nor the neo-Marxist approach to development
pays much attention to differences of this kind. In fact, as has
often been pointed out, the neo-evolutionist tradition has a
tendency to see 'modernisation' as a unilinear process which
affects in uniform fashion the economic, political, and cultural
spheres. It assumes that, as western technology and culture are
being diffused to the rest of the world, 'developing' or 'modernising' societies are gradually moving up the evolutionary ladder to
resemble more and more the western societies in their economies,
polities and cultures. From that viewpoint industrialisation and
urbanisation, as well as political and cultural modernisation, all go
hand in hand. 8
Marxist-oriented theorists have, of course, rightly rejected this
simplistic unilinear concept of development. They have stressed
that, beyond the superficial veneer of westernisation, the actual
structures of the economy and the way in which they have grown
historically within the world-capitalist system show quite clearly
that peripheral countries follow a trajectory which is qualitatively
different from that of the West. Due to the persistence of highly
exploitative relationships between the capitalist centre and the
periphery, peripheral countries have either failed to industrialise
at all, or their late and dependent industrialisation has shaped
their economies in such a way that they seem geared more to the
developmental requirements of the centre than to the needs of
their indigenous populations. 9 In the neo-Marxist view, therefore,
peripheral countries are not developing but underdeveloping:
their growing 'modernisation'- consisting of the servile adoption
of western institutions and culture - is seen as a smoke-screen for
the structures of underdevelopment which prevent the satisfactory
solution of such fundamental problems as mass poverty and
unemployment. Hence the Marxist distinction between modernisation and development - where modernisation means superficial westernisation, and development a relatively autonomous
industrial growth of the western type. 10
Although the neo-Marxist critique of modernisation theories
and its uncompromisingly anti-evolutionist stand are indeed a step
in the right direction, this school of thought tends to underemphasise the importance of changes in the political and cultural spheres
in so far as these changes are not related to the industrialisation
process. For instance, the early importation and adoption of
western liberal ideas and parliamentary institutions by nineteenthcentury pre-industrial Balkan and Latin American societies is
often seen as a mere facade, as an epiphenomenon which had very
little impact on the profoundly undemocratic, authoritarian structures of domination prevailing in these societies. 11
Now there is no doubt that parliamentary institutions in the
semi-periphery never functioned in the way they did in western
Europe. But this does not mean that their role in these political
systems was or is merely decorative. Party competition and the
early introduction of universal suffrage or of a relatively broad
franchise in the semi-periphery may not have brought about the
institutionalisation of civil liberties that characterise western parliamentarism; it did, however, create a base for the early organisation of party-political structures and the gradual imposition in
these countries of the rule of law. 12 For instance in countries like
Argentina, Chile, and to a lesser extent Brazil, the relatively stable

period of oligarchic parliamentarism together with other processes
weakened the caudillismolcoronelismo (a mode of arbitrary rule
by local chieftains) which was quite prevalent in nineteenthcentury Peru, Bolivia, and Venezuela. Moreover, in all the
countries of the parliamentary semi-periphery, the early adoption
of a multi-party system of rule led, in conjunction with other
factors, to the relatively early mobilisation of non-oligarchic
political groups and the broadening of political participation
before the advent of industrialisation.
In other words, the adoption of western parliamentary institutions in nineteenth-century semi-peripheral formations was not a
mere fa<;ade. Electoral competition, however intermittent or
distorted by large-scale fraud and coercion, was a strong force
contributing to the early weakening of restrictive/oligarchic structures of domination. Moreover, on the ideological level, democratic/liberal ideals and themes, however formalistic and however
much they clashed with the underlying authoritarian structures of
domination, did strike roots in these polities; they did constitute a
serious obstacle to the institutionalisation and legitimacy of longterm dictatorial rule.
1.1 Socio-economic processes leading to the early demise of
oligarchic parliamentarism
A. It was not, of course, merely the adoption of parliamentary

politics that brought about the early demise of oligarchic politics in
the parliamentary semi-periphery. Various other processes were
having similar effects.
Although western industrial capitalism - contrary to Marx's
predictions 13 - and to the analyses of neo-evolutionist modernisation theorists - did not automatically spread to the rest of the
world, its dominance in the West had profound repercussions on
the social structures of peripheral formations. The intensification
of western Europe's search for raw materials and new markets, at
a time when it had a dominant position in the international
division of labour, meant the restructuring of peripheral economies to suit the developmental needs of the capitalist centre. It
meant the commercial penetration of the semi-periphery and the
monetisation of at least certain sectors of the economy, the

development of national markets, the growth of social overhead
capital (roads, railways) in the agricultural and mineral export
sectors, and consequently the accentuation of the processes of
urbanisation, state expansion, and so on.
For instance, although by the beginning of the nineteenth
century western industrial capital had partially eliminated the
textile and shipbuilding industries that had previously flourished
on Greek territory, 14 the last quarter of the century saw an influx
of foreign and diaspora capital which, despite its extremely
exploitative character, contributed significantly to the development of Greece's social overhead capital. Combined with territorial expansion and demographic growth, this inflow of capital
accelerated agricultural commercialisation and urbanisation. 15 An
analogous point can be made about the rest of the Balkans 16 as
well as several Latin American countries which, towards the
second half of the nineteenth century, experienced a rapid growth
of their export sectors, an intensification of foreign investments,
and a fuller integration of their economies into the world market.
Such developments were particularly important in the southern
cone countries of Latin America, where the strong western
demand for their agricultural and mineral exports and the largescale settlement of south-European immigrants had contributed to
the rapid commercialisation of agriculture, the growth of social
overhead capital, and so on. 17
Moreover these countries, unlike economically more backward
Latin American societies (for example, Peru, Bolivia), did not
have large indigenous populations practicing subsistence agriculture. In societies of the latter type the large-scale persistence of
non-commercial forms of cultivation considerably retarded the
development of a home market and of national networks of
communication. By contrast, the absence of large indigenous
populations in Argentina, Chile, and southern Brazil actively
facilitated the integration of national territories via the rapid
formation of national markets.
While the late nineteenth-century semi-peripheral societies
therefore had, very obviously, pre-industrial economies, their
fuller integration into the world market resulted in their acquiring
social structures which, in a variety of ways, were more differentiated and 'modernised' than those of pre-industrial western Euro-
pean societies. I shall try to illustrate this fundamental point by

assessing the levels of urbanisation and state expansion in the
centre and the semi-periphery at comparable stages of industrialisation.
B. Let us take processes of urbanisation first. In the United
Kingdom in 1815, that is, at a time when industrialisation was
quite advanced, only 10.2 per cent of the population lived in cities
of 50 000 or more inhabitants. In Germany in the 1860s, the figure
for cities with populations of 20000 and over was 10.1 per cent,
and goes down to 3.4 per cent for cities of 100 000 and more. 18 In
Chile on the other hand, around 1920, that is, at a time when the
country's industrialisation was relatively weak, the proportion of
the population in all cities of minimum 20 000 inhabitants was 21.2
per cent and in cities of 100 000 it was 18.2 per cent. In Argentina
for the same period the figures are 32.7 per cent for all cities of
50 000 and over, and 27.1 per cent for cities of 100 000. 19 In Brazil,
urbanisation levels were lower: 11.2 per cent (cities of 50 000 plus)
and 9.5 per cent (cities of 100000 plus). 20 However, given that
Brazil had (and still has) a much more heterogeneous social
structure than either Chile or Argentina, comparisons with data on
the national/aggregate level are misleading. Southern Brazil, a
politically and economically very dominant region during the
inter-war years, was in terms of urbanisation levels, socio-political
mobilisation, and the demise of oligarchic politics quite similar to
her southern neighbours.2 1
The Greek urbanisation levels are comparable with the Latin
American ones. In 1920 the proportion of the population living in
cities of 20 000 or more inhabitants was 17.6 per cent, and in cities
of 100000 and more 12.6 per cent. 22 The other Balkan societies
show lower levels. Bulgaria for instance, the country whose
peasantry resisted the 'modernising' effects of western penetration
most strongly, had in 1920 only 9 per cent of its population living in
cities of 20 000 or more, and a mere 3.1 per cent in cities of
100000. 23 However, even these levels are quite impressive if it is
remembered that in 1920 Bulgaria's industrial base was proportionally much smaller than that of Britain in 1815 and Germany
in 1860. If reliable urban population statistics existed for the
pre-industrial periods in Britain and Germany, then even Bulgaria
would be relatively advanced as far as urbanisation is concerned.
10
This becomes obvious if Bulgaria is compared with a European

late industrialiser such as Sweden, where in 1862 the population
living in cities of 25 000 and more was only 3. 7 per cent, and in
cities of 50 000 plus it was 3.1 per cent. 24
C. If urbanisation, linked as it was with the commercialisation of
agriculture, 25 became a strong process undermining the foundations of oligarchic politics, on the political level the relatively early
expansion of the state was another powerful mechanism working
in the same direction. In fact, commercialisation and urbanisation
are in many ways linked with processes of state expansion and
penetration. 26 Although it is true that state intervention in the
parliamentary semi-periphery increased dramatically after the
1929 Depression (with the acceleration of import-substitution
industrialisation), even before this crucial turning point the role of
the state in the economy was quite considerable - and in any case
much more important than in pre-industrial Europe.
Brazil, for instance, experienced a rapid growth of state intervention during the entire period of the First Republic (18891930). Given the absence of those self-regulating mechanisms
found mostly in the developed capitalist economies, 27 the need to
build an infrastructure for the expanding agricultural/export economy, to protect agricultural prices, to regulate currencyexchange rates, and so on, pushed the government towards
increased intervention - despite the commitment of major politicians to the principle of laissez-faire and federal decentralisation.Z 8
According to an estimate based on Brazil's GNP for the years
1907, 1919, and 1926,
in terms of government revenue as a percentage of the GNP,
the Brazilian state (federal, state and municipal government)
utilized a greater share of the nation's resources than the
governments of the United States and Great Britain. The
revenue went increasingly into fixed capital rather than consumption or transfer payments. 29
Just as rapid commercialisation, even in the absence of industrialisation, is related to state expansion, so is rapid urbanisation.
It is not simply that in the large urban concentrations the state is
obliged to make massive expenditures to ensure at least the
11
minimum of municipal facilities (transport, sanitation, and so on),

but also because expanding state employment is an easy way of
reducing the massive urban unemployment created by the influx
from the countryside, above all in conditions where industry is
weak and/or capital-intensive. In other words, a small industrial
sector with low labour absorption capacity, together with a
massive population influx into the cities, make the over-inflation of
the state apparatus and the service sectors a generally foregone
conclusion. So in Greece during the 1870s the number of civil
servants per 10 000 of the population was approximately seven
times higher than in the United Kingdom; 30 another calculation
for roughly the same period gives a quarter of the non-agricultural
labour force as Greek state employees. 31 In Chile between 1930
and 1949 public employment more than doubled - a rate of
increase which surpassed by far the increase in employment in
industry, mining and agriculture. 32
Apart from the pressures from the commercialised economy and
the urban centres, another reason for the rapid state growth in
both Latin America and the Balkans was the attempt by the
respective governments to imitate the institutional structures of
the (already industrialised) western European parliamentary
democracies. As Stavrianos puts it in his historical analysis of the
Balkans:
State structures developed in the West naturally and harmoniously with the growth of economic life. But in the Balkans the
machinery of government was copied from the West and was
superimposed with all its elaborateness and costliness upon an
underdeveloped agrarian economy. At the same time the
pressure of the unemployed university students for government
positions also tended to swell the ranks of civil servants. The
result was that all the Balkan bureaucracies became grotesquely
overstaffed. Many more officials per capita were to be found in
the Balkan states than in the western ones, despite the weakness of the Balkan economies. 33
A more systematic, although methodologically problematic34
way of comparing state employment in the centre and the parliamentary semi-periphery is to look at the size of their respective
public administration at a given time. For instance, the proportion
12
of the labour force employed in the public administrations of the

centre and the semi-periphery just before or around the First
World War demonstrates that in the latter, despite extremely low
levels of industrialisation, the state apparatuses had reached or
even surpassed in size the public bureaucracies of the already
industrialised western societies. In the United Kingdom (1911) the
proportion of civil servants in the total labour force was 1.1 per
cent, in Germany (1907) and in France (1911) 1.4 per cent. By
contrast, in Argentina (1914) it was 3.3 per cent, and in Chile
(1920) 1.8 per cent. 35
A comparison of state tax-raising capacities- that is, the state's
ability to extract resources for self-maintenance, expansion, and
goal achievement - leads to similar conclusions. Although of
course there is a wide margin of error in calculations of this kind,
the fundamental point emerges quite clearly: at a time when the
societies of the parliamentary semi-periphery had barely begun on
their industrialisation, their states' tax-raising capacities were
already comparable to those of the industrial centre. So in 1913
United Kingdom government revenue per capita of the population
was equivalent to $20.14; in France at roughly the same time it was
$24.23, in Sweden $12.73, and in Germany $13.34. Argentina's
government revenue in 1913 per capita was $24.42, Chile's $16.39,
and Greece's $19.49. 36
Although it is difficult, if not impossible, to find state expenditure figures for pre-industrial Europe that are both reliable and
comparable with the figures given above, it is not hard to see that
the differences between the eighteenth/early nineteenth-century
western European states and the early twentieth-century Balkan
and Latin American states were considerable - both in terms of
revenue extracted and in terms of personnel employed. 37 The
difference is also reflected by, as well as related to, differences in
the educational field. For instance, most of the societies under
study here were more advanced in terms of school registration
around 1920 than European societies had been in the 1850s and
1860s. In France (1850), primary-school registration came to
3 322 000 or 8.23 per cent of the population. In Germany (1867)
the figure was 12.49 per cent, and in the United Kingdom (1860)
only 3.5 per cent. 38 In Argentina on the other hand, in 1920 those
registered in primary schools constituted 13.56 per cent of the
population, in Chile for the same period 10.59 per cent, in Greece
13
8.88 per cent, and in Bulgaria 13.47 per cent. 39

The differences are even more striking in the figures for
secondary education. In France (1863), secondary-education registration in relation to the total population was 0.68 per cent and in
Germany (1867) 0.12 per cent. 40 This compares with a 1920 figure
for Argentina of 0.45 per cent, for Chile of 1.4 per cent,
Yugoslavia 0.56 per cent, Bulgaria 1.01 per cent, and Greece 1.67
per cent. 41
D. One could go on establishing systematic differences between
centre and parliamentary semi-periphery by choosing a variety of
other indexes, but I think that even the limited data given above
show quite clearly that the early twentieth-century pre-industrial
semi-peripheral societies were more differentiated or 'modernised'
in terms of market, city and state expansion than pre-industrial, or
even early industrial Europe.
The connection with the early demise of oligarchic politics is
obvious. Although processes such as urbanisation and state expansion do not automatically bring about the demise of oligarchic
politics, in as far as oligarchic parliamentarism is predominantly
based on the tight control exercised on the lower rural classes by
local potentates who are relatively free from centralised state
interference, urban and state expansion does weaken their local
power and provides a favourable framework for the broadening of
political participation.
A final factor should be mentioned in this context which in its
turn contributed to the weakening of localistic controls and hence
to the creation of national arenas in the parliamentary semiperiphery. This is the relatively early adoption (around the turn of
the century) of the principle of universal military conscription.
This happened not only in the Balkans which, due to the Balkan
Wars and the First World War, were on a constant war footing,
but also in the southern cone countries where the need for so
drastic a measure was less obvious. Chile, for instance, initiated
general conscription in 1900 and Argentina in 1901. Although,
particularly in these Latin American countries, general conscription was not rigorously applied (the well-off could easily avoid
it), 42 it was a type of mobilisation that operated as something of a
school for the lower classes, contributing to the weakening of
14
traditional ties and gradually shifting loyalties from the local to the
national centre.
To stress the point once more, localism being slowly superseded
by the creation (through army, public administration, city and
market expansion) of national arenas was not directly responsible
for the break-up of the oligarchic monopoly of power; it simply
provided a framework of objective conditions favourable to the
oligarchy's overthrow. Within such a framework, and given the
competitive parliamentary system, it becomes increasingly tempting and feasible for leaders and groups outside or on the
periphery of the political establishment to assault the oligarchy's
monopoly of power by appealing to broader social strata and by
bringing them to play a more active role in politics. This task is, of
course, made easier when specific conjunctural developments deal
a sudden blow to the declining structures of oligarchic control. For
instance in Chile, the catastrophic decline of nitrate exports after
the First World War discovery of synthetic substitutes markedly
enfeebled the oligarchic political groups by dramatically cutting
down their major source of state funds and patronage. Similarly,
Greece's humiliating defeat in the 1897 Greco-Turkish war dealt a
severe blow to the ancien regime political world and paved the way
for the 1909 military coup which put an end to oligarchic parliamentarism.
Finally, as will be argued further below, it is not only that
conjunctural developments related to a country's specific history
are relevant for understanding the timing of an oligarchy's decline,
but - more importantly - the actual demise of oligarchic rule can
take varied forms, these variations being directly or indirectly
linked with differences in the countries' social structures.
To sum up: the argument in this section has been that in the
Balkans and in the southern cone of Latin America, despite low
levels of industrialisation during the nineteenth and early twentieth centuries, the increasingly intensive integration of these
economies into the world capitalist market generated very powerful processes of commercialisation, urbanisation, and state expansion. These processes, occurring within the political context of a
restrictive, albeit competitive, parliamentary system, undermined
the narrow basis of oligarchic rule even before the emergence of a
numerically strong industrial proletariat and an industrial
15
bourgeoisie. Whereas in western Europe the demise of oligarchic

politics came after the development of industrial capitalism, in the
semi-periphery it preceded it.
In such a situation it is not surprising that the transition from
oligarchic to mass politics was not characterised by the relative
dominance of autonomous trade unions and massive working-class
parties (as it had been in the West), as by the persistence, in
modified fashion, of the particularistic/personalistic organisational
forms of the oligarchic period and/or by the emergence of strong
populist movements in towns and countryside. To be more precise, that transition was brought about above all by the development of urban populism in southern Latin America, by the
emergence of strong peasant movements in the northern Balkans,
and by simply extending/centralising the traditional forms of
clientelistic politics in Greece. It is worth analysing these three
different modes of transition in some detail.
1.2 Routes to post-oligarchic politics
1.2.1 Urban populism
In Latin America's parliamentary semi-periphery, where urbanisation was more highly developed than in the Balkans and where the
landowning classes were politically and economically more powerful, it was predominantly in the towns that intense agitation
against oligarchic political rule developed and where populism
flourished during the first half of this century (the rural populations being firmly controlled by clientelistic or more coercive
means).
Although populism as a style of political mobilisation and
organisation appeared in the Latin American semi-periphery
before the demise of oligarchic parliamentarism, its full development came after the actual break-up of the oligarchy's political
monopoly, that is, after the civilian opponents of oligarchic rule
had, often with the army's decisive support, achieved an initial
broadening of political participation. From this point of view
populism can be seen as a means of consolidating such broadening
as had already occurred. In the face of the post-oligarchic persist-
16
ence of the economic and (to a lesser extent) political power of the
landowning classes, populism (in the form of political mobilisation
of the urban middle and lower strata from above) was a mechanism whereby the 'new men' strove to protect their gains against
attempts at oligarchic restoration.
In discussing the mode of transition to post-oligarchic politics in
southern Latin America I shall primarily focus on Argentina both because among the three countries under review it was the
first which saw the demise of oligarchic parliamentarism (in 1916),
and because it was precisely in Argentina that urban populism
achieved its most developed form.
A. Oligarchic parliamentarism began to function stably in
Argentina after the long period of national reorganisation which
lasted from the overthrow of Juan Manuel de Rosas (1852) to Julio
Angentino Roca assuming the presidency in 1880. In that year
Buenos Aires became the federal capital of the republic, this
federalisation meaning the final victory of the export-oriented
landowning interests of the pampean regions over the landlords of
the interior (who were less integrated into the export trade and the
world market). The year 1880 therefore marks the end of regional
conflicts and the consolidation of a national bureaucracy and
army. The Indians were vanquished, regional caudillo armies
eradicated, and the rule of law more firmly established. These
conditions, in combination with a very favourable economic
conjuncture (including the spectacular increase in the western
demand for primary products) greatly facilitated the development
of an export-oriented economy. National markets and national
communication networks were established, and foreign and indigenous capitalists entered an era of high prosperity. 43
Since the big landowners of the littoral constituted the backbone
of oligarchic Argentina, I shall begin by saying a few words about
the development of big landed property in that country, a development which is in striking contrast to the trajectory of the agrarian
question in the Balkans. Whereas in the latter, as we shall see
below, the tendency from the nineteenth century onwards was
towards small landholdings as the predominant form of landownership, the southern cone countries of Latin America exemplified
the opposite: a very strong trend towards extreme land concentration.
17
As early as 1826 the Bernardino Rivadavia government in

Argentina passed a law which allowed the state to lease out vast
tracts of land which, eventually, were bought by the lessees at
extremely low prices. It was particularly during Rosas's rule that
massive land sales took place, and the practice was continued by
the governments that succeeded him. Given the prevailing type of
agricultural activity (mainly cattle raising), the thinly-spread rural
population, and the favourable attitude of the state, a handful of
creole families very rapidly managed to amass vast landed estates.
These lands, which initially had a very low market price, acquired
great value with the development of export trade during the
second half of the nineteenth century. 44 This was particularly so in
the pampean regions as the most closely integrated into the world
market. In consequence, land accumulation and the spectacular
rise of land prices transformed the big landowners into a landowning bourgeoisie whose economic power was immense, even by
Latin American standards.
It has been estimated that as late as 1928, a group of 50 families
in the Buenos Aires area between them owned approximately 11
million acres, that is, 13 per cent of the land of Argentina's richest
province. 45 The astronomical value of these lands aside, the
annual income derived from them amounted to a phenomenal
sum. According to another calculation, the combined annual
incomes of the six largest landowners of the Buenos Aires
province in the year 1916 were much higher than the combined
annual budgets of the major Argentinian ministries (Foreign
Affairs, War, Agriculture, Treasury, and Public Works). 46
Moreover, most of the big owners were absentee landlords who
lived abroad for most of the year, leaving the management of their
huge estates to local managers or renting parts of their land to
tenants. In view of the high land prices and the lack of cheap credit
or more general government assistance, such tenants could not
easily become landowners themselves. In contrast to the situation
in the Balkans, therefore, small peasant-owners were quite
peripheral in the Argentinian countryside in terms of both numbers and socio-economic organisation. Also, since land hunger is a
characteristic typical of the small peasantry - not of tenants who
are more concerned with low rents, nor of hired labour which
aspires to high wages - there was no real pressure from within
rural society for land distribution. Neither, of course, was there
18
state pressure for such reforms, for the links between Argentina's
agrarian bourgeoisie and the oligarchic state were very close.
In fact the core landowning families, although not always
directly involved in politics, wielded enormous political power
through a variety of channels. For instance, they were members of
the prestigious Argentine Rural Society which was extremely
powerful, especially during the oligarchic period. The landowners
of the interior, although economically less influential, also enjoyed
considerable political kudos, as did the owners of merchant or
finance capital. To use a somewhat different terminology (see
Chapter 4), the links between the holders of the means of political
domination and those of the means of production/distribution
were very strong; the latter, even when not directly exercising
power, strictly limited the formers' room for manoeuvre. This
brings us to a brief examination of the system of oligarchic politics
in Argentina.
B. Unlike Chile's 'parliamentary republic' and Brazil's Old Republic, Argentina's oligarchic parliamentarism took a very centralised form. After the post-1880 effective centralisation of the state,
the president's powers expanded at the expense of the power of
congress and of local legislatures. Even so, the political system was
to a large extent based on local oligarchies centred around the
provincial governors. It was the governors who, through the
establishment of national coalitions, were the main factor in the
election of the president. In fact, the presidential candidates were
less concerned with capturing the popular vote than with securing
the support of these governors. Although the 1853 constitution
had institutionalised the principle of universal male suffrage, vote
participation was extremely low47 as the oligarchy controlled the
electoral process through clientelistic or more fraudulent means.
In this process the governors played a crucial part:
The governors had a network of caciques in the different
electoral districts of their province. The cacique, or a man of his
entourage, secured the position of mayor, supposedly through
direct popular election. However, the governor of the province
had the right to nominate the chief of the local police, the tax
collector and, with the consent of the city council, the justice of
the peace. These officials plus the mayor were then in charge of
the local 'situation' and without their support and explicit
19
approval it was impossible to win any elections. The system

worked so nicely that these four local officials did not need
special incentives or bribes to jail opposition leaders, make all
kinds of electoral frauds, and replace the electoral act by a
tragicomic anecdote of a small provincial town. 48
As in most other countries of the parliamentary semi-periphery,
this system of restrictive representation was gradually undermined
by the fast development of the economy and the concomitant
spectacular growth of the urban population towards the end of the
last century. In this respect the massive influx of immigrants from
Europe was of particular importance. Especially prior to the First
World War, the migratory flow was of extraordinary dimensions:
between 1881 and 1915 five million immigrants entered
Argentina. 49 Given the highly unequal and rigid agrarian structure, the majority of these migrants went to the towns of the
littoral, taking over most of the non-agricultural occupations and
contributing to the expansion of both the urban working and
middle classes.
These developments rendered increasingly incongruous a system of rule which excluded from political participation not only
the foreign migrants but also the majority of the native-born urban
middle and lower classes. The first cracks in the oligarchic system
appeared as early as 1890. The severe economic crisis of that year,
which led to the devaluation of the peso and a drastic deterioration
of middle- and working-class incomes, generated widespread
dissatisfaction and resulted in the creation of the Union Civica, a
party formed by a variety of disaffected elements of both upperand middle-class origin. The crisis also brought the uprising of July
1890 which precipitated the resignation of president Juarez
Celman.
In 1892 the fraction of the Union Civica that was led by Leandro
Alem, and represented the less conservative elements, created the
Union Civica Radical which, under the subsequent leadership of
Yrigoyen, was to play the chief role in opening up the political
system. The Radical Party, although it included landowners
peripheral to the politically dominant groups, increasingly
oriented itself towards the new social forces generated by the rapid
growth of the urban centres. The 1893 and 1905 Radical uprisings
(in which army officers took part) may be regarded as a final
20
flickering of the pre-1880 civil-war tradition through which interelite disputes used to be solved;50 but they can equally be seen as
signs of the growing crisis of a restrictive system of rule which, by
denying political expression to the rapidly expanding urban middle
sectors, was becoming more and more incompatible with the
highly differentiated and dynamic economic and social system.
After 1905 the Radical Party stopped being merely an arena for
intra-elite squabbles and began much more systematically to
mobilise popular support and to create the first nation-wide
organisational party structure. By its tactics of 'revolutionary
absenteeism' from the electoral process, it put the legitimacy of
elections into question and made the corrupt and fraudulent
means through which the oligarchy was maintaining its control
both more salient and less acceptable. At the same time the
radicals began to attract numerous supporters from the ranks of
the public administration, the police, and the armed forces. In
addition, growing agitation on the political level was coupled with
growing trade-union militancy which, especially during the first
decade of the twentieth century, reached a very strong and violent
level to which the oligarchic government reacted in an extremely
brutal and repressive manner.
All these developments, and a certain unrest within the army, 51
kept aggravating the divisions inside the ruling conservative party
(the Partido Autonomista Nacional) and gave increasing political
weight to a number of 'enlightened' conservatives. These, the
so-called modernists, were arguing that the only way of coping
with the 'social question' and with the rising tide of radicalism was
to allow a limited opening-up of the political system that would
bring in the middle classes electorally. Such an inclusion, they
suggested, would take the steam out of the radical movement, and
yet would by no means endanger the traditional political hegemony of the conservative forces. After Roca's failure in 1903 to get
his chosen candidate appointed as president, the modernists were
in the ascendancy, and in 1910 they succeeded in electing Saenz
Pefia as president. In 1912 Saenz Pefia introduced a new electoral
law which became a crucial landmark in the country's political
history.
Although Saenz Pefia's law was not the first attempt at electoral
reform (there had been a more timid and less successful one in
1902), and although it did not give the vote to foreign-born
21
immigrants, it did have a dramatic impact on Argentina's political

system. The introduction of the secret ballot (with voting now
compulsory for all men of eighteen years and above who were fit
for military service), the establishment of more accurate electoral
rolls, and a variety of other measures did break the monopoly of
the traditional political oligarchy and made possible Hipolito
Yrigoyen's election to the presidency in 1916 and the entrance of
new men into power. 52 The rise of the Radical Party did not, of
course, completely displace the older, conservative parties. Even
if the old political guard had lost its previous hold on government
posts and patronage, it still exercised considerable influence,
particularly in the senate. Moreover, one must not forget the
influence of landowning groups within the Radical Party: it is not
accidental that five of the eight members of Yrigoyen's first cabinet
were members of the Sociedad Rural. 53 As a matter of fact
Yrigoyen's policies accurately reflected the heterogeneous character of his party. Once in power, he tried to satisfy both the
economically dominant landed/export interests and the newly
enfranchised middle classes.
To please export-oriented estate owners, Yrigoyen refrained
from any serious effort to redress the huge inequalities prevailing
in landownership. 54 He also accepted as a matter of course the
landowners' concept of Argentina's economy as first and foremost
based on primary exports, with industry playing a secondary,
ancillary role. In fact, except for a short period towards the end of
the Radicals' long rule, their policies showed none of the type of
economic nationalism that flourished during the early Per6nist era;
neither was there any serious attempt to introduce effective
protectionist measures for the development of indigenous industry. Even the interruption of commodity imports during the First
World War resulted in no systematic effort to change the tariff
system and to promote import-substitution industrialisation.
With respect to the middle classes, the relatively weak industrial
development and the dependence of industrialists on the growth of
the export economy meant that their demands could, to a certain
extent, be reconciled with the landowners' economic interests. In
any case, middle-class demands did not focus on industrial but on
state expansion, on the creation of professional and bureaucratic
jobs for the growing urban populations. So Yrigoyen tried to keep
the middle classes happy by the large-scale creation of government
22
posts which, together with other favours, were massively distributed through the Radical party's nation-wide patronage network.
As long as economic growth could be sustained, this type of public
largesse towards the middle classes did not seriously clash with
landed interests. In times of economic crisis, however, it was much
more difficult to please both the upper and the middle classes, and
so it is not surprising that Yrigoyen fell soon after the 1929 world
economic crisis. 55
Looking at the over-all picture, one could well say that the
Radical policies between 1916 and 1930 did not seriously challenge
the prevailing relations of production. This was not at all surprising, given the strength of the landed/export interests, the lack of
popular pressures in the countryside for agrarian reforms, and the
relative complementarity in economic orientation of the big
landowners and the weak industrial bourgeoisie whose enterprises
were closely linked to primary exports.
Nevertheless, Yrigoyen's would-be liberal radicalism did bring
about a serious restructuring of the relations of domination in
Argentina's polity. Not only did it break the restrictive political
monopoly of a handful of patrician families, it also profoundly
changed the country's patterns of political organisation and mobilisation. From the point of view of this essay this aspect of
Yrigoyenism requires further discussion.
C. Even before 1912, but more dramatically so after the introduction of the law on electoral reform, the Radical Party developed into a national party - becoming the first Argentine party
with a nation-wide organisation. (The Socialist Party, which was
founded as early as 1896, never really managed to extend its
influence beyond the Buenos Aires area and, unlike its Chilean
counterpart, remained a regional party.) 56 This national organisation consisted of committees which, especially during elections,
were highly active all over the country. The Radical committees,
and the ward bosses who usually controlled them, played a crucial
role in gaining and maintaining popular support for the Radical
Party by distributing a variety of favours and by introducing a new
style of politics. The committees were organised
geographically and hierarchically in different parts of the country. Thus there was a national committee, provincial commit-
23
tees, or in the case of the city of Buenos Aires the committee of

the federal capital, precinct and ward committees, and during
election periods a chain of sub-committees, serving smaller
areas of each precinct ...
The committees' activities reached a peak during election
periods. In addition to the conventional street-corner meeting,
the posting of manifestos and the distribution of party pamphlets, they were used as centres for the distribution of charity
handouts to the voters. In Buenos Aires, in 1915 and 1916, the
committees founded children's cinemas, held music-hall concerts, distributed Christmas gifts and supported the annual
carnival celebrations. Many of the committees also founded
medical clinics, legal advice centres and libraries the cost of
which was borne by active members. They also provided cheap
foodstuffs, the pan radical and the carne radical as they became
known. 57
It might be objected that these committees presented no great
organisational break with the past since they more or less repeated, if on a larger scale, what the more traditional caciques had
been doing during the oligarchic period. However, although the
particularistic exchange of favours for votes was similar in both
cases, there were nevertheless fundamental differences between
the face-to-face patron-client relationships of the traditional type,
and the party-based and more collectively oriented patronage of
the Radical Party.
True enough: in many cases the provincial committees were
controlled by local landowning groups which would disregard
formal election procedures, could co-opt officials, and operate the
committees like traditional clientelistic networks. But this type of
local sabotage of the new organisational patterns was attenuated
to some extent by the sharp increase in federal interventions (a
constitutional provision which allowed the president to replace
provincial authorities by a centrally appointed interventor). 58
Moreover, in the urban areas, and especially during Yrigoyen's
second presidency (when the middle-class elements in his party
began to dominate the landowning ones), the committees adopted
a political style which was quite different from that of the
oligarchic parties. Their officials were often elected by the party
members, and these party officials were frequently university-
24
educated, professional people possessing very different skills from

those of the traditional cacique; they were geared less to the
requirements for face-to-face patron-client relationships and more
to the needs of mass politics. This important organisational
difference was well reflected in the social composition of the
Radical political elite during Yrigoyen's second presidency (192830).
Among the Yrigoyenistas in Congress the sons-of-immigrants
group was now represented in considerable numbers. Many of
them had made their way up from precinct committees, especially
those representing the federal capital. They were mainly professional men with university educations, though most owed their
positions simply to their control over the local party machines.
This was in marked contrast with the position in 1916, when the
overwhelming majority of Radical legislators were landowners.
(my italics) 59
Apart from the committees and the nation-wide coverage of the
party organisation, the other fundamentally new dimension of
Yrigoyen's political style was, of course, the intensely personalistic
and plebiscetarian style of leadership which he developed much
more than any previous president:
Although with Roca, Juarez Celman, Figueroa Alcorta, and at
an earlier date with Rosas, there had been a tendency to
personalistic policies and issues, with Yrigoyen this became one
of the central stylistic elements of Argentine politics. It became
the accepted convention for the Radicals to prefix their actions
and statements with lengthy panegyrics to their leader. Equally
the opposition reserved its most biting attacks for the president
himself. 60
This new style and the highly paternalistic policies associated with it were largely related to the extensive appeal the
Radicals had among the urban masses. In fact, the extraordinary
mystique which surrounded the person of the leader and its
nation-wide projection through the party organisation meant that
the new men in charge of local branches - unlike traditional
clientelistic bosses who had a relatively autonomous base - derived
their authority from the party machine, and ultimately from the
person of the leader himself. Having no roots outside the party
25
organisation, and operating in a situation where the party structure was to some extent dependent on the leader's charismatic
appeal, they were much more at the latter's mercy. 61
This became very clear after the split of the Radical party in
1924, when Marcelo Alvear's conservative followers founded the
Union Civica Anti-personalista and challenged Yrigoyen's leadership. This challenge failed. In the presidential elections of 1928
Yrigoyen more than kept his popular base, and in a landslide
victory received 57.4 per cent of the vote. As far as vote-catching
was concerned, Yrigoyen no longer needed the more traditional
party elements. His strength did not lie in the support of traditional notables, but in his direct popular appeal to the urban populations and in a national organisation largely based on it.
The above populistic/plebiscitarian organisational elements,
although quite marked in the 1920s, were of course to reach full
development with Peron's ascent to power fifteen years later.
D. This last point suggests a brief account of developments
during and after Yrigoyen's fall. As already mentioned, Yrigoyen's second presidency encountered severe difficulties when
the economic prosperity of the 1920s came to an abrupt end with
the Depression. This crisis, much more than the preceding one
(from 1913 to 1917), meant that the president could no longer
sustain his juggling act between the landowners and the urban
middle classes. Besides, the difficult economic situation created a
climate within which other discontents (objections to the president's aloofness and senility, to his partisan interference in the
army's promotion procedures, to his tendency to use his presidential powers to eliminate opposition) 62 acquired particular saliency.
All these developments drove the conservatives and the antipersona/isla radicals to flirt with the idea of military intervention
and to encourage disgruntled or extreme right-wing officers in this
direction.
The military coup of September 1930 put an end to fourteen
years of uninterrupted Radical government. It can be seen- if not
at the level of intentions, at any rate at the level of unintentional
consequences - as a restoration of restrictive quasi-oligarchic
rule. 63 It was certainly an attempt by conservative groups (both
within the military and within the economically dominant classes)
to recapture political control of the state by force, that control
26
which they had lost with the introduction of Saenz Pefia's electoral
reform bill. Saenz Pefia had hoped that the cautious opening-up
of the political system would integrate the urban sectors without
the landowners losing their political hegemony. But, given the
organisational and populistic dimensions of Yrigoyenism, and the
conservatives' failure to match the Radicals in organisation and/or
populistic appeal (see Chapter 3, section 3.1 on this point), Saenz
Pefia's move turned out to have been a serious miscalculation. 64
From 1930 to 1943 the conservatives made a desperate effort to
correct this mistake by establishing - after a short period of
military rule (1930-32) - a 'neo-oligarchic' system of restricted
representation based on the time-honoured techniques of political
corruption, fraud and coercion. Of course the new restrictive
system of representation could not be and was not a straightforward return to the pre-1912 order. In view of the greater politicisation of the urban population meanwhile, the conservatives could
only maintain political control by using increased doses of coercion
- and this lack of political legitimacy made it quite easy, after the
1943 military coup, for Peron to rise to the leadership and for
oligarchic parliamentarism to decline irreversibly.
It goes beyond the scope of this essay to give a detailed account
of the Peronist movement and the profound transformations it
brought about in Argentina. From the point of view of our
problematic all that needs to be stressed briefly are certain
fundamental continuities and discontinuities between Yrigoyen's
and Peron's populism. Given the continuing economic strength of
the landed/export interests, Peron adopted a dual policy vis-a-vis
the dominant classes:
(a) despite his initial economic nationalism and his channelling of
resources from agriculture to industry, Peron (like Yrigoyen)
refrained from interfering with the prevailing relations of
production in the countryside;
(b) much more intensively than Yrigoyen, Peron created a very
strong populist base, focussing his energies and his charismatic
appeal not so much on the middle classes as on the growing
working class, especially (but not exclusively) on the indigenous elements moving from the countryside to the evergrowing towns. 65 By unionising and gaining the almost unquestioned political support of these people, and by developing populistic/plebiscitarian techniques of organisation and
27
leadership to an unprecedented level, he gave the coup de

grace to the political hegemony of the old creole patrician
families.
E. The transition to the post-oligarchic period of the other two
Latin American countries in our sample, although in many respects very different, does have some elements in common with
the Argentinian case - elements which, in terms of the mode of
this transition, clearly differentiate the Latin American from the
Balkan semi-periphery. I shall merely sketch these elements
below.
(i) Brazil. In Brazil the processes undermining oligarchic parliamentarism were very long drawn out. On the political level,
however, it was the tenentes movement of the 1920s, culminating
in the 'revolution' of 1930, which marks the beginning of the end
of oligarchic politics. In October 1930, when the Brazilian army
brought into power Getulio Vargas, the defeated leader of the
Liberal Alliance, it initiated a process which transformed an
essentially regional revolt (that is, Rio Grande and Minas against
Sao Paulo) into a political upheaval that was to undermine the
whole system of oligarchic politics. This new political situation,
which clarified itself after 1932 and was further accentuated during
the Estado Novo period (1937-45), was characterised by the
ending of the old republic's extreme regionalism, the transformation of the army into a national institution and, above all, the
spectacular centralisation of the state and its increased interventionism in the economy and polity. 66 All of these developments led
to the break-up of the political monopoly held by the agrarian
oligarchy, and the entrance of new men into the corridors of
power.
Although Vargas's movement and his forced installation as
president in 1930 enjoyed much popularity, 67 he was not brought
to power by a populist movement. The breakthrough or turning
point came clearly with the intervention of the army, not as the
result of populist mobilisation. Once Vargas was installed in office
however, populism as a distinct style of political mobilisation and
organisation, did begin to develop. It first emerged in the rapidly
growing urban centres of southern Brazil; 68 after the end of the
Estado Novo it developed on a much larger scale when Vargas
28
attempted to continue to rule in a parliamentary context by

making a populistic appeal to the urban middle and workin~
classes.
As in Argentina, the Brazilian leader too was trying, both
during and after the Estado Novo, to consolidate his position and
more generally the post-oligarchic broadened system of political
participation (a) by compromising with the landowning classes and
refraining from questioning the prevailing relations of production
in the countryside; and (b) by bringing new urban strata into the
political arena as a counterweight to the oligarchy's persisting
economic power. It was this dual policy which informed not only
his attempts at controlling and obtaining the support of existing
unions but, more importantly, his systematic endeavours to unionise and bring into the political arena the thousands of rural
immigrants pouring into the cities.
(ii) Chile. Chile again showed similar patterns of transition to
post-oligarchic politics, although populist mobilisation was not as
intense as in Argentina. Here also, although Arturo Alessandri
had for several years been challenging Chile's political oligarchy
through a charismatic/populist appeal to the politically excluded
classes, 69 the real breakthrough came with the army's reformist
intervention in 1924; in this case it was directed against an
oligarchically controlled congress dragging its feet over legislation
on the 'social question'. 70 The intervention paved the way for
Ibanez's rise to power (1924-1931) and the entrance of the novi
homines into politics. 71 In more attenuated form Ibanez was to
initiate the pattern of consolidating post-oligarchic rule which was
followed later by Vargas and Per6n: a compromise with the
dominant agrarian interests by a policy of non-intervention in
respect to agrarian structures, as well as an attempt at securing an
urban mass base by co-optive/repressive social legislation, the
creation of new unions, and so on.
The populistic style of politics initiated by Alessandri and
continued by Ibanez in the 1920s was finally taken up by Marmaduke Grove in the 1930s. Grove, despite his leadership of the
Socialist party, developed a charismatic style of politics not unlike
that of the late Vargas and Per6n. AsP. Drake points out:
As an anti-status quo party, the Chilean Socialists broke into

the national arena in the 1930s with intensely populist appeals
29
behind a charismatic caudillo from the armed forces ... While

emphasising the workers and revolutionary socialist objectives,
the party also developed a mixed constituency encompassing
the middle classes and programs seeking to reconcile industrial
growth and welfare reforms. As a mixed breed of socialism and
populism, the PS rapidly became one of the most dynamic mass
movements yet to appear in the hemisphere. 72
The Socialist party's strong populist mobilisational features were
to become more tenuous after the socialists' participation in the
Popular Front of 1938. 73
In more general terms, and notwithstanding the abovementioned strong populist tendencies in Chile's post-oligarchic
political system, the (by semi-peripheral standards) exceptional
autonomy of the Chilean working-class movement meant that
there was less political margin for the development of populism
from above than in Argentina and Brazil. From this point of view
the Chilean left-wing forces acted as the counterweight to the
oligarchy's power which in Brazil and Argentina was provided by
populism. It is precisely because the consolidation of the postoligarchic system in Chile was not exclusively achieved through
populism but also through a working-class movement relatively
autonomous from state tutelage that the pre-dictatorial political
structures of Chile came closer to those of Western Europe. 74
1.2.2 Peasant populism
A. Whereas in the Latin American parliamentary semi-periphery
it was urban populism that was a major feature in the immediate
post-oligarchic period, in the northern part of the Balkan peninsula it was above all the peasantry that, mobilised in a populist
manner, strongly affected post-oligarchic events. It is not only the
actors mobilised who were dissimilar, but also the manner of
mobilisation (from below, rather than from above by the state),
and the role that populism played in the subsequent political
developments.
A major difference in the respective social structures of
nineteenth-century Latin America and Balkan societies lies in the
power position of the landlord classes. In contrast to Latin
America, in the Balkans (where the best lands were occupied by
30
the Turks) the autochthonous landowning classes were relatively

weak. The gradual decline of the Ottoman empire and the rise of
Balkan nationalist movements brought about the more or less
sudden withdrawal of the Turkish landlords. This enabled the
peasants to take possession of the land they had been cultivating
for their foreign overlords before the relatively weak autochthonous Balkan landlords could get hold of it. 75 The more precipitous
the Turkish withdrawal, the easier the take-over for the peasants.
This explains, for instance, why from the very start Bulgaria which gained independence as a result of Russia's defeat of the
Turks in 1874- had a much more egalitarian agrarian structure
than either Greece or Serbia. 76
Notwithstanding such variations, in general it is quite clear that
in the Balkans the position of the economically dominant classes in
the countryside and their hold over the rural populations was
much shakier than it was in Latin America. While this was true in
the nineteenth century, it became even more so after the important agrarian reforms which took place in all Balkan countries in
the inter-war period. 77
B. Another major feature of the nineteenth and early twentiethcentury Balkan societies was that, with the exception of Greece,
they were all much less urbanised and industrialised than those of
the Latin American semi-periphery. Since some comparative data
on urbanisation have already been given, I shall merely add a brief
reference to differences in terms of industrialisation.
Despite the fact that neither Greece nor the Latin American
semi-periphery were effectively industrialised around the turn of
the century, their levels of industrialisation were higher than those
of the northern Balkans in the early twentieth/8 at any rate up to
1930. True enough, after the 1929 crisis, as the process of
import-substitution industrialisation was accelerated all over the
Balkan peninsula, the difference in terms of the industrial sector's
contribution to the GNP between Greece and her Balkan neighbours became less marked. 79 It must be stressed, however, that
even in the 1930s and 1940s the persistently low levels of urbanisation in the northern Balkans, as well as the low levels of migration,
were keeping the vast majority of the active populations in the
primary sector. 80
This fundamental difference with regard to the structure of the
31
labour force between Greece and the southern Latin American

societies on the one hand, and the northern Balkan countries on
the other, a difference which persisted despite comparative levels of
import-substitution industrialisation, is extremely important for
understanding the numerical and political weakness of the middle
classes in the northern part of the Balkan peninsula.
To be more specific, the low levels of urbanistion, commercialisation, and industrialisation in the northern part of the Balkans meant not only a weak industrial bourgeoisie but a very small
middle class in general (professionals, merchants, administrators,
and so on). In the early twentieth century it was not only the
landlord classes in the northern Balkans which were weak but - in
comparison with Greece and the Latin American semi-peripheral
societies - the urban middle classes too were both numerically
small and politically feeble. This meant of course that there were
no strong integrating links between villages and towns, and that
the over-all control the political oligarchy was able to exercise over
the peasants was quite fragile and precarious. In order to make
this important point clearer, I shall focus on Bulgaria, the country
which developed the strongest peasant movement during the
inter-war years in the Balkan peninsula.
In the early part of the century the Bulgarian peasants still
maintained the zadruga family system 81 and hence their integration into the market economy was less than that of, for instance,
the Greek peasants. The contrast emerges quite plainly from the
different life styles of the Greek peasants and those of Slav origin
living in northern Greece during the inter-war period. The Greek
peasants of the Peloponnese, involved in the market economy
from very early on, had developed much more individualistic,
'modem' attitudes. Their attachment to the soil was looser, their
orientation more urban, and they were quite ready to develop a
consumer mentality. Their Slav counterparts on the other hand
who still tilled the land under the zadruga kinship system looked
less to the market for their food and clothing, remained fiercely
attached to the land, and were openly hostile to urban values. 82
The two countries' inter-war agrarian reforms only accentuated
this difference in degree of commercialisation. Mitrany notes that
after the reforms many of the Slav smallholders actually reverted to
subsistence farming, so deliberately trying to avoid market hazards.
In Greece the commercialisation of agriculture had meanwhile
32
advanced to such a degree that it pre-empted a similar return to

subsistence farming. 83
In the Bulgarian villages the social structure, due to the absence
of marked inequalities in landownership and the low level of
commercialisation, was both more egalitarian than in Greece 84
and less differentiated in terms of occupations. It was typical
around the turn of the century that those Bulgarian villagers who
actually did engage in non-agricultural/pastoral activities were not
only few in number but also poorly integrated into the rural
community. This becomes more readily understandable if one
remembers that prior to independence the few towns that existed
in the northern Balkans were populated mainly by non-Bulgarians
(Jews, Greeks, and so on.) 85
Even in the first quarter of this century the Bulgarians had
remained agriculturalists to such an extent that in the large
villages too, which did have a little occupational differentiation,
the local merchant, the usurer, the baker, shoe-maker, blacksmith were of foreign origin (Greeks, Vlachs, Epirotes). 86 The
division into peasants and non-peasants was so trenchant that even
the few Slavs in non-agricultural activities were regarded as
foreigners. 87 Therefore, the marked hostility between the peasant
community and urban merchants aside, the village was sharply
split into people who were cultivating the land and people who
were not. Such animosity of course contributed to making it more
difficult for the latter to exercise any real leadership and so
becoming efficient intermediaries between the village and the
town.
This enormous gap between town and village was reflected on
the level of political organisation. Bulgaria having attained independence relatively late- in 1874- her state institutions were less
developed and did not penetrate the countryside to the same
degree. One way of measuring this is to assess the state's taxraising capacity, and in this respect Bulgaria was far behind
Greece. In the 1930s the Greek state managed to extract 83.40
French francs per head (calculated on the basis of purchasing
power prior to the First World War), whereas the Bulgarian state
raised only 28.08 francs. 88
A comparison between Greece and Bulgaria in terms of political
parties shows similar differences. In Greece, political organisations based on powerful local oligarchies89 had been set up from
33
the moment the country first attained independence. The stronger

and more centralised the state became, the more these oligarchies
endeavoured to offset their loss of local autonomy by acquiring
control of the state from within, and this gave rise from very early
on to political factions with strong and extensive links in the
countryside. The introduction of parliamentary politics in Greece
in no way impeded these political factions: they continued to
thrive as pressure groups and patronage agencies, successfully
controlling and incorporating the peasantry into the central state
institutions. In Bulgaria on the other hand, where the ruling
classes had neither accumulated the wealth nor achieved the
power of their Greek counterparts, such roots as the nineteenthcentury oligarchic political parties had in the countryside remained
relatively weak. When E. Dicey visited Bulgaria a decade after her
independence he was struck by the lack of linkages and the
political vacuum between the state and the peasantry:
Except in the large towns, very little interest is taken in politics.
To the great mass of the electorate it is a matter of utter
indifference who their representatives might be. The difficulty
is to get the electors to vote at all; and in the majority of
instances the representatives (in parliament) are virtually
nominated by the government of the day. 90
In conclusion, the fact that neither the rural nor the urban
economically dominant classes in the northern Balkans could
exercise any strong control over the peasantry - at a time when the
latter started being exposed to the vagaries of national and
international market forces - explains why agrarian populism
could spread so widely and quickly in the northern part of the
peninsula.
C. This brings us to a more detailed analysis of peasant conditions during the transition period from oligarchic to post-oligarchic
politics. Any reference to Balkan peasant conditions towards the
end of the nineteenth century and the early twentieth must
necessarily start with the changing relationship between the southeastern and western European countries before the turn of the
century.
Whereas, to begin with, the western powers had found it
34
expedient to use diplomatic channels and military coercion to

further their interests in the Balkans, by the end of the century
their tactics had become those of economic imperialism. This was
the time of intensive penetration of the Balkan peninsula by
western capitalism, initially through loans to governments, as well
as investments in transportation networks geared to the strategic,
political and economic requirements of the western powers. 91 The
Balkan states were chronically short of not only social overhead
capital but also of funds for their rather large armies and their
much inflated and corrupt civil services. There is little profit in
such investments, and since even infrastructural projects are slow
to yield an appreciable return, Balkan countries became more and
more deeply indebted to their western creditors. 92
The section of the population that suffered the brunt of this
disruptive impact of imperialist penetration was the peasantry.
The cheap manufactured goods from the West arriving in the rural
areas via the transport routes built with western capital, as well as
the lack of cash with which to meet their governments' ever more
exorbitant tax demands, increasingly forced the peasants to turn
from subsistence farming to market production. Moreover the
inter-war agrarian reforms in the Balkans did little to improve the
peasants' lot. Although it did give the moribund landlord classes
the coup de grace, it provided none of that state help for the
peasantry (loans, technical assistance, educational facilities) which
is a fundamental precondition for successful land reform.
A further aggravation of the already precarious economic
condition of the Balkan peasants was a spectacular population
growth, which began in the early nineteenth century and accelerated in the inter-war years. 93 In consequence, the anyhow very
small landholdings were splintered into ever smaller fragments,
and agricultural productivity dropped to still lower levels. The
peasants' already low standard of living kept declining further, and
approximately half the rural population became redundant. 94
Given that the above processes occurred in a context where, as
mentioned earlier, both the remaining landlords and the urban
middle classes were weak and unable to prevent the large-scale
mobilisation of the peasantry by intellectuals, the emergence of
agrarian movements and parties with both anti-landlord and
anti-urban, anti-industrial platforms is hardly surprising. In fact,
some of these movements aimed simultaneously at further drastic
35
agrarian reforms as well as, rather unrealistically, maintaining

intact the peasant economy and society by keeping industrialisation and the growth of the state and the cities to a minimum. 95
In addition to the favourable conditions outlined above for the
development of agrarian populism, important conjunctural developments must also be taken into account of course. Although
some of the agrarian movements had preceded the Balkan Wars
and the First World War, it was the six-year period (1912-18) of
almost continuous war-footing that was the catalyst for large-scale
mobilisation of the peasantry and the emergence of agrarian
parties as major political forces.
D. Let us look more closely at the development of Bulgaria's
Agrarian Union, the strongest peasant party in the inter-war
Balkans. During the turn of the century, in addition to other
difficulties (such as the decline in world grain prices, the spectacular growth of usury in the countryside, and so on) the peasants had
to contend with a series of bad harvests that drove them almost to
starvation. It was at this point (in 1899) that the IvanchovRadoslavov government chose to reintroduce a tithe on agricultural products, a measure which aggravated peasant conditions
further. The founding of the Bulgarian Agrarian Union (BANU)
in 1900 was more or less a direct reaction to this tax in kind. Unlike
Argentina's Union Civica, the Agrarian Union was not founded by
disaffected landowners but by non-elite intellectuals - mainly
school teachers struggling along on meagre incomes and in difficult
work conditions. 96 It was these relatively educated people who
were to provide most of the Agrarian Union's cadres, greatly
influenced by populist and socialist ideas imported from Russia
(particularly the populist ideas of Herzen, Michailovsky and
Lavrov). 97 When BANU changed from a merely educationaleconomic interest group into a political organisation, some of the
populists objected to this change and left the organisation. Nevertheless, by having drawn attention to the plight of the peasantry,
populist ideas contributed significantly to Bulgaria's peasant
movement.
Once BANU was established and following some initial successes, the Union entered on a brief period of decline. It was revived
under Alexander Stamboliiski's energetic leadership from 1906
onwards, thereafter developing into a serious political force. In the
36
1908 elections BANU won enough votes and seats in the national
assembly to become the main opposition party. There can be no
doubt that this success was due partly to a vigorous attempt to
build up a national party organisation based on local branches, the
druzhbi. Whereas at the beginning a druzhba might include
members who belonged to other parties, after 1905 a formal
pledge was required that ruled out party membership elsewhere.
In this way the Agrarian Union not only acquired a more distinct
political profile, it also began to weaken the clientelistic networks
of the traditional parties. New resolutions further formalised and
strengthened the party organisation. So
a minimum of ten members were required to elect officers on
November 21 each year; and to hold monthly meetings during
the rest of the year to discuss union affairs and questions of
general interest to the peasantry. They were urged to form
cooperative societies, to establish reading rooms where Agrarian literature would be available and to form new druzhbi in
neighbouring villages. 98
The druzhbi spread very rapidly in the Bulgarian countryside and
became important centres of village life.
The heart and lifeblood of this organisational structure was
Stamboliiski's charismatic personality which dominated party
affairs and made Union cadres dependent on his support. An
indication of this control over the movement was his recommendation to the Supreme Union Council, prior to the 1923 elections,
not to re-nominate more than half of the Union's parliamentary
group and to appoint in their place younger men from the local
branches. Although this amounted to a drastic purge, the departure of those who had been sacked did not - since they had no
independent power base outside the organisation - negatively
affect the election results. On the contrary, Stamboliiski in this last
election before his assassination actually won a landslide victory.
During the war years Stamboliiski adopted an anti-war pacifist
attitude in opposition to the throne and the political and military
establishment. The First World War ended with Bulgaria on the
losing side. The popular discontent created by the great devastation of the countryside and the severe restriction of the country's
frontiers, discredited the old political oligarchy and left the
37
road wide open for the Agrarian Union's rise to power. During
the elections of August 1919 the old parties were routed. BANU
received the highest percentage of votes, and together with two
other anti-oligarchic parties (the communists and broad socialists)
commanded an absolute majority of votes and 71 per cent of the
seats in the national assembly. (In the two subsequent national
elections of 1920 and 1923 the Agrarians further improved their
electoral performance.)
During its stay in power (1919-23), BANU initiated a series of
radical reforms a~med at the effective improvement of peasant
conditions. To take the agrarian reforms first: although it was
small landholdings that had prevailed before the wars, there was
some tendency towards land concentration as money lenders,
traders and politicians were trying to create large estates. 99 Since
the number of landless peasants was further swollen by the influx
of roughly half a million refugees from territories lost by the war,
Stamboliiski initiated highly drastic measures of land distribution,
as a result of which more than 80 per cent of the peasants came to
own the land they tilled, and only 2.6 per cent were renting all the
land under their cultivation. 100 There were also important reforms
in education, in the field of co-operatives, and in taxation (to shift
the tax burden from the rural to the urban population). Another
striking measure was the introduction of a compulsory labour
service as a substitute for the conventional military service forbidden by the Allies (in the Treaty of Neuilly) after Bulgaria's First
World War defeat.
It is interesting to note that in implementing the above measures
the Agrarians tried to by-pass the pre-existing state bureaucracy
and the established authorities generally. For instance, they
appointed organs independent of the local government administration to carry out the agrarian reform programme. They also
occasionally used the famous orange guards, paramilitary proAgrarian Union troops, instead of the mistrusted police and army.
Stamboliiski also tried to reduce the entrenched authority and
powers of the judiciary and to make it more responsive to
peasants' needs:
Most ministries and their major subdivisions, such as Compulsory Labour Service, established internal machinery to arbitrate
disputes in order to avoid involvement in the courts. Seeking to
38

make justice cheaper, more accessible, and less mysterious to
the peasant, the Agrarians also undertook a major reform of
the lower court system. Since they were convinced that most
professional lawyers were parasites at best and dangerous
enemies at worst, a major thrust of their reform was to limit the
lawyer's right to appear before certain courts. 101
Finally, Stamboliiski not only made draconian cuts in military

expenditures, he also reduced the army's size below the already
low level stipulated by the Treaty of Neuilly. He chose a civilian
for the portfolio of the Ministry of War, and rigorously pursued a
pacifist policy which, naturally, was anathema to the military
establishment.
Needless to say, all these measures greatly enraged the throne 102
and the army. The latter's secret organisation, the Military League
created as early as 1919, soon managed to acquire control of the
Union of Reserve Officers (representing thousands of officers who
had to leave the army soon after the Neuilly Treaty had severely
cut down its size), and in September 1923 it put an abrupt end to
Stamboliiski's government. This coup and the savage assassination
of Stamboliiski demonstrates the extent to which the weak Bulgarian upper classes had felt themselves threatened by the political
rise of the peasantry. Despite that setback, and despite the
subsequent harassment and persecution of Agrarian Union members, the movement made a remarkably rapid recovery. In the
elections of 1931, the moderate wing of the Agrarian party, in
coalition with a group of liberal bourgeois politicians, succeeded
once more in forming a government which lasted until the military
coup of 1934 which, less than a year later, led to the establishment
of a royal dictatorship that continued until 1943.
A similar development pattern of agrarian populism obtained in
Yugoslavia where the Peasant Party, organised by the famous
Radich brothers, constituted the major political force in Croatia
and played a crucial role in inter-war Yugoslav politics. Moreover
in Yugoslavia, as in Bulgaria, the immediate post-war years of
reforms and the rise of agrarian parties was followed (in the
post-Depression period) by the imposition of a monarchical dictatorial regime which lasted until the Second World War. 103
39
1.2.3 The transformation of clientelistic networks

A. In Greece, contrary to both the northern Balkans and the
southern cone countries of Latin America, the transition from
oligarchic to broader forms of political representation occurred
without the development of any strong populist movement in
either the countryside or the cities. Although conjunctural events
may have had something to do with this peculiarity of Greek
politics, I think that it is, above all, certain features of Greece's
social structure which can throw some light on this interesting
question.
As mentioned already, during the nineteenth century and up to
the Second World War, Greece's agrarian structures were quite
similar to those in the rest of the Balkans. As the Turkish
landlords fled the Peloponnese, most of their chijiik estates despite extensive land usurpations by both Greek landlords and
peasants - became 'national lands' under the direct ownership of
the Greek state. 104 The latter reneged on its promise to distribute
these lands gratis to the landless peasants who had fought in the
war for Greek independence; instead, the Dotation Law of 1835
gave them the right to purchase small plots of national land against
low annual payments over a 36-year period. This law was not very
popular, and its implementation proceeded at a snail's pace; 105 but
the Greek government's refusal to sell lands by auction (a Ia Rosas
in Argentina) was a decisive step hindering the emergence of big
landed property in the newly-born Greek state. 106 Moreover,
prime minister Koumoundouros further reinforced the pattern of
small peasant landholdings by a massive distribution of national
lands in 1871. 107 The Peloponnese, therefore, was dominated by
small landholdings in the nineteenth and twentieth centuries.
The situation was different in Attica, Euboea and the central
and northern provinces of Greece. Due to the successive later
incorporations of these territories into the Greek state, 108 special
treaties with Turkey safeguarded the rights of Turkish chiflik
owners. Despite this protection the Turks sold their lands cheaply
to rich Greeks - mainly diaspora Greeks from Constantinople,
Smyrna and Alexandria. Given the late date and the mode of
acquisition of these estates, their Greek owners never achieved the
type of political control over the state that their Argentinian or
Chilean counterparts enjoyed during the oligarchic period. By the
40
time they became chiflik owners the state apparatus had already
been appropriated by the mainly Peloponnesian notable families,
whose political hegemony was based not so much on landownership but on the role they had played during the War of
Independence and, later, on state patronage. 109
As a result of this relative political weakness of big landed
interests in early twentieth-century Greece, premier Venizelos was
able to consolidate the broadened, post-oligarchic political system
without large-scale populist mobilisation, that is, without any
sustained appeal to the masses over the heads of established
authorities and clientelistic networks. Accordingly it was relatively
easy for him to introduce a land reform law which became
particularly effective after the massive influx of Greek refugees
from Asia Minor in 1922. 110 The desperate need to accommodate
this huge mass of uprooted people had as one of its consequences
the acceleration of the land reform programme, to the extent that
by 1936 a total of 425 000 acres had been distributed to 305 000
families. 111 These developments dealt the final blow to big landed
property in Greece. From then on, and quite irreversibly, the
small private landholding was to be the dominant form of cultivation in the Greek countryside. This type of drastic land distribution, effectuated without any major peasant mobilisation and
against only lukewarm landlord resistance, would have been
impossible in countries (like those of the southern cone of Latin
America) where the landowning classes were both economically
and politically more powerful during the oligarchic period.
B. If Greece had a pattern of land reform which, in the end
result, was quite similar to that of the rest of the Balkans, she
much more closely resembled countries like Chile and Argentina
in terms of high rates of urbanisation, market and state
expansion. 112 The significant difference between Greece and her
northern neighbours in these respects has its historical roots in the
extraordinary development of the Greek merchant class during the
period of Ottoman rule. An interesting phenomenon in the
economic history of the Ottoman empire was the early ascendancy
(in the fifteenth and sixteenth centuries) of the Greek-Orthodox
merchant, not only in the Balkans but in most of the empire's
territories. In the eighteenth century Greeks (and to a lesser
extent Jews and Armenians) managed to control a considerable
41
portion of the empire's internal and external trade. 113 Other

Balkan merchants appeared on the scene much later, and never
managed to rival the Greeks in terms of power and wealth. The
Bulgarian merchant class, for instance, only just began to develop
in the eighteenth century, by which time commerce between
Constantinople and Bulgaria was already firmly in the hands of
Greeks and Armenians who had a virtual monopoly of trade
between the Ottoman capital and the eastern Balkans. 114
One clear indication of the Greek merchants' supremacy in the
Balkans during the eighteenth/early nineteenth centuries is that at
that time Greek was the language of commerce and culture all
over the peninsula. Non-Greek merchants considered it prestigious to be able to speak Greek and were proud if they were
mistaken for Greeks. 115 This situation came to an abrupt halt with
the rise of nationalism in the various societies of the northern
Balkans. 116
Moreover, the Greek merchant class played a very important
role in the early awakening of Greek nationalism. It was Greek
traders, especially those settled in the Danubian principalities and
in the European capitals, who were responsible for importing
revolutionary ideas from France, and who provided both leadership and considerable material help to the forces fighting for
independence from the Turks. 117 Their extensive contacts with the
West, and the schools they founded at home to spread rationalist
and anti-clerical teachings, greatly contributed to the early westernisation - 'modernisation' - of Greek society and culture. How
much this development owed to the Greek merchant bourgeoisie
can be understood better if it is remembered that high churchmen
and big landowners - the more traditional Greek social elite were rather lukewarm in their attitude to independence. Having
much more to lose, they joined the struggle only when it had
become clear that the status quo could not be upheld. Their
ambivalence left the field open for the mercantile bourgeoisie to
set the new values and standards, and so almost from the
beginning to shape the major institutions of Greek society along
liberal-bourgeois lines. Finally, it should also be stressed that after
the establishment and consolidation of the modern Greek state the
Greek diaspora bourgeoisie (which in terms of financial power was
much more important than the indigenous one) maintained its
very strong links with the mother country and continued in many
42
ways to influence developments in mainland Greece. 118

I think that the above explains to a great extent why in the late
nineteenth/early twentieth century Greece was a more urbanised,
occupationally more differentiated society, with a more developed
middle class than Bulgaria and the other Balkan societies.
There is another peculiarity of Greece's social structure that is
highly relevant to our argument: in spite of rapid urbanisation and
the growth of a very sizeable middle class partly linked to an
overinflated service sector, Greece's 'marginalised' urban population never attained Latin American proportions. Athens, for
instance, never experienced the growth of slum areas on the scale
of Santiago or Sao Paulo. From this point of view a diagram of
Greece's urban structure looked then and looks still like an
inverted pear: very broad in the middle and quite narrow at the
bottom.
This is explained in part by the massive labour migration from
Greece which began as early as the nineteenth century and was
related to and facilitated by the vast Greek diaspora which, as
noted above, always kept up strong ties with the homeland. For
instance, the period we are concerned with here saw a very strong
migratory flow from Greece to the United States. It began around
the turn of the century and, for various economic and social
reasons, acquired considerable momentum during the first two
decades of the twentieth, slowing down only with the restrictive
immigration policies of the American government in the early
1920s. So in 1907 alone, 36 580 Greek citizens emigrated to the
US, and by the end of 1932 there were almost half a million Greek
immigrants in the States. In no other Balkan country did external
migration take on such proportions. 119
C. Having noted some very general ways in which the social
structure of Greece differed from both the Latin American
semi-periphery and the rest of the Balkans, let us now look at the
demise of oligarchic politics in Greece. Processes of urbanisation,
market and state expansion, in combination with the early adoption of parliamentary institutions and universal suffrage had, as
already mentioned, begun to undermine the basis of restrictive
oligarchic politics long before the country's effective industrialisation.
During the second half of the nineteenth century, the closer
43
integration of Greece's economy into the world market, the

country's territorial growth, and the Trikoupian reforms of the
1880s (concerning the attraction of foreign capital, social overhead
investments, administrative rationalisation, and so on) all led to
the formation of a more complex and differentiated economic
system generating new social forces of a predominantly middleclass' character:
The last forty years of the nineteenth century saw a very
considerable expansion of the world of professional men and
salaried clerks working in financial, commercial and government institutions. These, with the great number of small
wholesale and retail traders, formed the nucleus of a growing
urban middle class wearing European clothes, priding themselves on having an education and attitudes similar to their
western counterparts. 120
As in the Latin American southern cone countries, these new
social forces were eventually to challenge the monopoly of power
being held by a handful of oligarchic families. The gradual
processes of erosion of oligarchic parliamentarism were accelerated by conjunctural events, such as Greece's defeat in the
Greco-Turkish war of 1897, the severe economic and financial
crisis that followed the war, 121 the subsequent failure of successive
governments to strengthen and modernise the army, the unsuccessful management of army affairs by the royal princes, and so
on. 122 It is these circumstances which resulted in the 1909 military
intervention (or 'revolution' as it is often referred to in Greek
historiography), which marks the end of oligarchic parliamentarism (the paleokommatismos).
Although the instigators of the 1909 coup were motivated
primarily by grievances related to their careers and to the generally unsatisfactory state of the armed forces, 123 in view of the
gradual erosion of oligarchic parliamentarism and the people's
disillusionment with the political oligarchy after the 1897 defeat, 124
the army's intervention led to broader, mostly unanticipated
consequences. Like similar reformist army interventions in interwar Latin America, the Greek coup of 1909 opened up the road
for the entrance of new strata into the political arena. The army,
by summoning Venizelos (the charismatic Cretan statesman) to act
44
as its political advisor, initiated a series of moves which brought

about Venizelos's spectacular entry into Greek politics and his
newly founded Liberal Party's unprecedentedly large election
victory in December 1910. The paleokommatiko?, the old political
oligarchy, were routed and the door to active politics was thrown
wide open to the rising middle classes. Even if the old notable
families (the so-called tzakia) which had monopolised nineteenthcentury political life were not swept away altogether, they were
certainly forced to share power with the new men - lawyers,
doctors, businessmen and so on. 125
In contrast to developments in the rest of the Balkans, the
broadening of the political system did not lead to a strong peasant
movement in post-1909 Greece. As noted above, the land-reform
programme, initiated in 1917 and seriously implemented after the
massive influx of Asia Minor refugees in 1922, was effected
without large-scale peasant mobilisation. Agrarianism was not, of
course, entirely absent in inter-war Greece. Especially in the
newly acquired northern provinces where, after the land reforms,
conditions in the countryside were more egalitarian, the Greek
Agrarian Party did manage to play some role in politics. However,
even in the New Territories the agrarian movement was not only
insignificant if compared with the Bulgarian one, but its influence
was drastically curtailed by the accentuation of the political
polarisation between Venizelists and royalists in the early 1930s.
Looking at the over-all picture, it can confidently be argued,
therefore, that, for the whole of the inter-war period, the peasants
were firmly kept within the clientelistic networks of the two major
bourgeois parties (the Venizelist Liberals and the Populist Party),
which were profoundly divided over the issue of the throne. Given
the numerical strength of the urban middle classes (relative to the
northern Balkan societies) and the depth of the dichasmbs (the
split over the monarchy), the peasants were drawn into an
intra-bourgeois conflict which had very little relevance to their
own class interests. 126 This was also the case with regard to the
political inclusion of the 1.5 million refugees from Asia Minor who
settled on the mainland after Greece's defeat by Turkey in 1922.
The majority of them considered the royalists as wholly responsible for the 1922 debacle and gave their massive vote to Venizelos
who, although not directly responsible, had actually been the
principal architect of Greece's Asia Minor adventure. 127
45
Venizelos therefore, the liberal politician whose personality

dominated inter-war Greek politics, managed to steer Greece
through its immediate post-oligarchic phase without resorting to
either peasant mobilisation a Ia Stamboliiski, or urban populist
solutions aIa Peron or Vargas. The rising middle classes which his
Liberal party mostly represented managed to consolidate their
position vis-a-vis the ancien regime oligarchy and to introduce the
usual reforms of the immediate post-oligarchic phase (social
legislation, import-substitution industrialisation, educational reforms, and so on) without using the state apparatus for creating a
supportive urban populist base. Such a populist base was both
unnecessary (given the low possibility of an oligarchic restoration
on the model of Argentina) and rather difficult to set up (in the
absence of an extensively marginalised urban population).
D. Finally it should be stressed that the Greek type of transition
to post-oligarchic politics differed not only from the patterns in the
northern Balkans and southern Latin America, it also diverged
from the western European pattern: the broadening of the political system and the inclusion of new strata in the political game was
not effectuated through autonomous and massive trade unions and
working-class parties, but by clientelistic means. Of course, the
fact that clientelism persisted during the whole of the inter-war
period does not mean that there were no differences between
nineteenth-century oligarchic clientelism and the political patronage system of the post-oligarchic period. The huge expansion of
the state bureaucracy and the centralisation of the major parties
altered the more traditional type of clientelism under which local
notables had had such tight control over their voters and enjoyed
such high autonomy vis-a-vis the national leaders that parties on
the national level were simply loose coalitions of political barons.
The entrance into politics after 1909 of new social elements and
the decline of the tzakia families meant the strengthening of the
parties' central organisation as orientations, allegiances, and resources shifted from the local to the national level. These developments in turn meant that traditional clientelism had to give way to
more centralised forms of party and state-oriented patronage. The
autonomy of the local notables having been eroded, they had
increasingly to share brokerage functions at the local level with
new patrons, lawyers, doctors, state employees, and so on. 128
46
The peasants meanwhile, having become increasingly dependent on the state for a variety of services, had to resort to a
number of different patrons for their different needs and requirements. This did not mean that the clients necessarily acquired
greater autonomy, or that their power relationship with the
patrons became less asymmetrical. It rather meant that, as the
village community was losing its self-containment, peasants as well
as traditional political patrons became more dependent on the
state and on the national party organisations. In other words, the
loss of autonomy of the traditional 'monopolistic' patron was not a
gain for the peasant so much as for the state and for those who
controlled its expanding resources.
On the national level, the passage from traditional to more
centralised forms of clientelistic politics meant also that the party
leader, much more so than during the oligarchic period, acquired a
mass following which, especially in the urban centres, was no
longer directly related to clientelistic networks. Given this new
context, the political debate acquired a more distinctive flavour of
class and national issues. For despite the absence of strong
non-personalistic party structures within the bourgeois parties,
state expansion, the development of communications and national
markets brought, among other things, the emergence of 'national
constituencies', of a nationwide 'public opinion' which, over and
above clientelistic considerations, began to have an important
impact on the shaping of political issues. 129
So, for instance, the dichasmos, the major inter-war political
split between Venizelists and anti-Venizelists, was obviously not
the result of mere party squabbles over the distribution of spoils.
After the fundamental disagreement between Venizelos and King
Constantine over the foreign policy to be adopted during the First
World War, the issue of the monarchy became crucial; and given
the violence and fanaticism with which it was being fought, moving
from one hostile camp to the other was not as easy as clientelistic
theory might suggest. 13 Clientelistic 'instrumentalism', in other
words, was drastically attenuated by ideological stance, for or
against the monarchy. 131
Moreover, it was not only ideological differences that restricted
purely clientelistic/instrumentalist orientations; class locations and
cleavages in the post-oligarchic period were more directly related
to political conflict. For instance the split between Venizelists and
47
anti-Venizelists, despite its strongly personalistic flavour, can

partly be explained in terms of class interest. Venizelos's Liberal
Party had its social base in the entrepreneurial bourgeoisie, those
fractions of the petty bourgeoisie (for example, shopkeepers) who
were able to profit from the expansion of the market and the
general modernisation of the economy, and (initially at least) in
the urban working class. Venizelos also had the support of rural
smallholders in the more recently acquired northern provinces
who were the beneficiaries of the Liberals' radical agrarian reforms and of their policy of protectionist measures favouring the
cultivation of, for instance, wheat for the domestic market.
The anti-Venizelists, on the other hand, were supported by the
tzakia families, the declining landowning classes, the investors and
financiers associated with the National Bank of Greece, and by the
majority of peasants in the provinces of the Old Kingdom. (These
peasants had profited from the 1871 distribution of national lands
and were integrated into the market and into the clientelistic
networks of the tzakia families much earlier. They specialised in
'luxury' produce for export, and were much more differentiated
and hierarchically organised than their counterparts in central and
northern Greece.) 132
Both the increasing importance of ideology and the more direct
linkages between class positions and political practices meant that
national and/or class issues and broad economic developments
could more easily cut through clientelistic networks and traditional, particularistic ties. This should not, however, lead one to the
conclusion that clientelism was on the wane in post-oligarchic,
inter-war Greece. A look at the structures of the two major
parties, both on the local and national level, shows that they were
clearly clientelistic - in the sense that the core party-component
consisted of powerful clientelistic factions which systematically
and successfully boycotted any attempts at introducing more
formal/bureaucratic procedures of party organisation.
For instance Venizelos's Liberals, the least clientelistic of the
two major parties, had developed local associations both in the
provinces and in Athens which could have become the nuclei of a
more modem/bureaucratic party structure. But the political bosses
saw such associations as a threat to their own power position and
quite effectively underm9Ied their operations. In contrast to
Yrigoyen's radical committees and to Stamboliiski's druzhbi,
48
Venizelos's liberal associations did not succeed in gradually displacing the old clientelistic bosses - the opposite happened: the
bosses sabotaged the associations. Similarly on the national level,
Venizelos's repeated attempt to reform the Liberal Party organisation and reduce its clientelistic aspects (e.g. by introducing a
written constitution, formal procedures of recruitment, etc.) were
doomed to failure by the sustained opposition of the party's
dominant clientelistic chiefs, especially those from the Old
Kingdom. 133
In summary: politics generally in inter-war Greece, and political
parties in particular, retained their strongly clientelistic/particularistic character - notwithstanding significant variations in the
strength of clientelism (which was greater in the Old Kingdom, in
rural areas, and in the more conservative parties), and despite the
growing importance of ideology and the closer links between class
cleavages and political conflicts. The significant changes that did
take place in political organisation between the oligarchic and the
post-oligarchic period indicate not so much an irreversible decline
of the patronage system, as rather its transformation: they were
concomitants of the shift from restrictive/oligarchic to a more
centralised, party-oriented clientelism.
In explaining the various forms of transition from oligarchic to

post-oligarchic politics in the countries under consideration I have
emphasised some basic differences in their social structures (for
example, the degree of urbanisation) and how these differences
were related to the nature of their dominant classes and the types
of control they exercised over the urban and particularly the rural
populations. In doing so I did not intend to give a full and detailed
explanation of the transition, but simply wished to point out
certain broad structural features which, while not themselves the
cause, did facilitate certain types of political solutions. My major
argument in this respect can be schematically summarised in the
following diagram (which has no claims to be applicable to
countries other than those considered in this chapter).
As indicated by this diagram, I hold that in Latin America the
exceptional economic strength of the landed/exports interests and
their relatively strong political sway over the state necessitated 134
49
Modes of transition to post-oligarchic politics

Urban middle and upper-middle classes
STRONG
WEAK
Peasant populism
(northern Balkan
pattern)
E-
C/l
Urban populism
(southern cone
Latin American
pattern)
Non-populistic
transformation
of clientelism
(Greek pattern)
the populist mobilisation of the urban middle and lower classes for
the broadening and/or consolidation of the post-oligarchic political
system. Given the highly advanced stage of urbanisation in the
southern cone, the objective conditions for such a mobilisation
were already present, and in time charismatic leaders appeared
who took advantage of these conditions to build themselves an
urban populist base as a counterweight to the persisting power of
the landowners - that is, as a means of consolidating the postoligarchic broadening of political participation.
In the northern Balkans, on the other hand, given the landowners' economic and political weakness and the low levels
of urbanisation, neither themselves nor the numerically weak
urban middle and upper classes were able to control the small
peasant owners who, at the beginning of the present century,
constituted the vast majority of the total population. As
oligarchic parliamentarism was undermined, partly through the
long-term structural processes common to all countries of the
parliamentary semi-periphery (closer integration into the world
market, commercialisation of agriculture, development of railroads, and so on), and partly through conjunctural developments,
progressively-minded intellectuals managed to break the oligarchic system by mobilising the peasantry against both the weak
landlords and the equally weak urban upper and middle classes.
Unlike Latin American urban populism, Balkan peasant populism
was not, on the whole, based on a multi-class alliance; and unlike
the post-liberal, 'classical' populism of Peron and Vargas, agrarian
populism was less interested in rapid industrialisation and more in
the maintenance of the peasant community or, more realistically,
in a very slow and smooth transition to industrialism.
50
To explain the northern Balkan peasantry's radical rejection of

the commercialisation of agriculture and of other processes which
disrupted their traditional style of life, I have drawn attention to
both conjunctural factors and particularly to certain structural
features that are also found in other studies of peasant radicalism:
the egalitarian, non-differentiated character of the village which
leads to what Barrington Moore has called 'radical village
solidarity'; 135 the weakness of upper-class controls; 136 the existence of a disaffected intelligentsia with an effective mobilising
ideology, and so on. 137
Finally, in Greece the weak political control over the state by
the owners of the late-formed chifiik estates meant that the
oligarchic political monopoly of the tzakia families (a monopoly
not predominantly based on landownership) was broken and the
post-oligarchic, more open system consolidated without any
marked populist mobilisation either among the peasants or among
the rising urban classes. In other words, the tzakia families, due to
their weak economic base, and the relatively new chiflik owners,
due to their weak political control, could not and did not offer the
resistance of their counterparts in the Latin American southern
cone countries. The relatively large urban middle classes, therefore, managed to broaden the political system and to introduce
social reforms as well as land redistribution within a political party
system which, although more centralised, retained most of its
ancien regime clientelistic features. In that sense the Greek
transition to post-oligarchic politics was the 'smoothest'- that is to
say, its continuities (both party-organisational and ideological)
with the oligarchic system were stronger than in the other countries under consideration. 138
1.3 Initial stages of industrialisation and the incorporation of the

industrial working classes
In the previous sections of this chapter the focus has been on the
different types of political transition from oligarchic to broader
forms of participation - a transition which, as noted, occurred in
the parliamentary semi-periphery before the large-scale development of capitalist industry. The point I would like to develop in
this section is that even when industrial capitalism was gaining
51
momentum, especially in the post-1929 period, the state's well

entrenched incorporative tendencies and its leading role in the
industrialisation process meant that it could easily undermine the
autonomy of working-class organisations and bring the growing
number of industrial workers into the post-oligarchical political
arena in a vertical, dependent manner. Before examining this
central problem, however, let us have a brief look at the initial
stages of industrialisation in the semi-periphery.
1.3.1 industrialisation
A. Despite its secondary role, capitalist industry did make some
hesitant advances in several of the semi-peripheral economies
before the 1929 crisis. Already in the nineteenth century the
development of import/export trade had created opportunities for
the establishment of industrial nuclei linked to export activities,
and to the growing demand in the rapidly expanding urban centres
for light industrial consumer goods. 139 As is well known, with
some exceptions 140 this type of industrialisation was given an
important boost by the First World War and its economic repercussions.
In the Latin American southern cone especially, the period
between the First World War and the 1929 Depression saw a rapid
growth of the so-called traditional industries (food, beverages,
textiles, leather, wood, and so on) which require relatively simple
technologies and know-how. In that sense, import-substitution
industrialisation in several Latin American economies - but especially in Argentina and Chile- had started much before the 1929
crisis. 141 In the northern Balkan countries, war conditions created
some limited opportunities for industrialisation, but the direct
involvement of these societies in the hostilities and the ensuing
destruction meant that it was only in 1926/27 that, profiting from
the pre-Depression economic boom, their economies could be
reconstructed and achieve pre-war levels of production. 142 In
Greece, finally, industry received a strong boost after the country's
defeat by Turkey in 1922. As a result of this defeat, more than one
million refugees from the previously flourishing Anatolian Greek
communities poured into Greek territory. The settlement of large
sections of these uprooted people near the main urban centres
meant the availability of abundant and, at least initially, cheap
52
labour as well as of entrepreneurial skills, at a time when Greece

was experiencing a massive injection of foreign funds in the form
of international aid for the refugees, private investments in public
works, and so on. This combination of entrepreneurial skills,
cheap labour and abundant capital gave a strong impetus to Greek
industry. 143
However, a factor that was even more important for the
pre-1929 semi-peripheral economies than the effect of war conditions was that by the end of the nineteenth century all of them, as
already mentioned, had in different degrees developed quite
extensive infrastructures (roads, ports, railways) which, even
if geared to the promotion of exports (particularly so in Latin
America) and/or the strategic designs of the Great Powers (in the
Balkan case), provided a favourable ground for the rapid development of industrial capitalism in the post-Depression period.
So in the Latin American semi-periphery it was not only the
temperate-zone countries (Argentina, Chile) which had developed
considerable social overhead capital towards the end of the
nineteenth century, but even Brazil, despite the fact that her
exports of tropical commodities require as a rule less complex
infrastructures. The Balkan economies too, despite their lower
volume of exports 144 and lower urbanisation rates, experienced
significant economic growth and infrastructural development towards the end of the nineteenth and at the beginning of the
twentieth century, to such an extent that, in terms of communication networks for example, they ranked higher than the semiperipheral economies of Latin America. 145
B. Notwithstanding the fact, therefore, that in the late
nineteenth/early twentieth century all Latin American and Balkan
semi-peripheral societies had a clearly pre-industrial character,
their economies were advanced enough to be able to respond
positively to the new and difficult conditions created for their
export-oriented economies by the world economic crisis. The
dramatic decline of their exports, as well as the widening gap
between agricultural and industrial products, 146 caused semiperipheral societies to experience a drastic reduction of their
capacity to import industrial goods from the West, and so more or
less forced them to develop their own industries on a much larger
scale.
53
Although the decline of exports as well as the process of import

substitution had started before the world Depression, 147 by dramatically accelerating both processes the international economic
crisis became a decisive turning point and, in a way, marked the
end of the outward-looking character of the semi-peripheral
export economies. From that time on, the hitherto feeble process
of industrialisation lost its more or less haphazard character as the
state, in much more vigorous and interventionist manner, promoted the development of industry by adopting a series of measures
to assist industrial capital: a higher and more coherent protectionist policy, 148 low tariffs for imports of capital goods, grants and
credit facilities to major industrialists, and so on. In these favourable circumstances, and given that the industrialising effort was
limited to the production of consumer rather than intermediary
and capital goods, industry was able to advance at a very rapid
pace during the early pre-Second World War phase of import
substitution.
The pace and pattern of capitalist industrialisation varied considerably from one semi-peripheral country to another, of course.
Argentina for instance, which was the most industrialised of the
Latin American societies with very fast growth rates between the
First World War and the Depression, adapted well to the postcrisis conditions and her industry grew considerably, despite the
fact that, as already mentioned, landed interests regained appreciable political power after the 1930 'oligarchic restoration'. The
export difficulties and the pragmatic outlook of those in charge of
economic policy 149 led to a successful reorientation of the
economy. 150
In Chile, on the other hand, the export economy suffered much
more severely from the Depression, to such an extent as to
preclude the vital imports of machinery necessary for the country's
industrial development. As a matter of fact it was only after 1945
that Chile's industry could begin to grow at reasonably fast
rates. 151 In Brazil, the rapid recovery of the export sector in the
post-Depression period allowed the build-up of a considerable
industrial base which began to expand, especially after 1945, at
such a rapid rate that it outpaced both Argentina and Chile. 152
In Greece a series of pro-industry governmental measures (such
as the abolition of free drachma convertibility into gold, quantitative restrictions on imports, and so on) provided a framework
54
conducive to an effective breakthrough into the industrial sector.

According to League of Nations statistics for the first postDepression decade, Greece's increase in industrial production was
one of the highest in the world. 153
In the northern Balkans, where the economic depression generated conditions which led to a considerable development of
industrial capitalism, the pattern of industrialisation was nevertheless different. Here, as we have seen, the overwhelming majority
of the labour force remained in agriculture, the significant advances in industry notwithstanding. So, for instance, in Bulgaria as
late as 1946 the industrial sector only employed 9.8 per cent of the
labour force, the tertiary sector 11.6 per cent and the remaining
75.5 per cent of the work force was still employed in agriculture. 154
The pattern was much the same in Yugoslavia. 155 In contrast to
this, Argentina in 1947 had only 26.4 per cent of her work force in
the primary sector, industry was absorbing 29.4 per cent and the
service sector 38.2 per cent. 156 In Chile in 1940, the distribution
was 41.2 per cent in agriculture and mining, 20.4 per cent in
industry, and 38.2 per cent in services. 157 The southern regions of
Brazil had a pattern of industrial development not dissimilar to
that of Argentina and Chile. Finally, as far as Greece is concerned,
the labour-force distribution was between the northern Balkan
and the Latin American pattern: in the late 1940s half of the work
force was in the primary, 18.9 per cent in the secondary, and 25.3
per cent in the tertiary sector. 158
In terms of industry's contribution to the gross domestic product, however, the lead of Latin American industrialisation
appears less impressive. Towards the end of the 1930s, the
industrial sector's contribution to the GDP was approximately 25
per cent in Argentina (1937), 11.3 per cent in Chile (1937), and
13.1 per cent in Brazil (1937) 159 - whereas it was 22 per cent in
Yugoslavia (1937), 21 per cent in Greece (1938), and 17 per cent in
Bulgaria (1939). 160
The difference in the northern Balkans between the distribution
of the labour force and of national income among economic
sectors indicates that these countries, even if they did not lose their
marked agrarian character, did undergo an important process of
capitalist industrialisation - a circumstance which clearly sets them
apart from countries like Bolivia and Peru where post-Depression
industrial performance was much lower. 161 As two influential
economic historians have put it:
55
It becomes obvious that population distribution by occupation

does not afford an accurate picture of the structural changes in
the economy of the countries in question (northern Balkans),
for it shows less change than actually took place. 162
In other words, northern Balkan societies did undergo a considerable process of import-substitution industrialisation, albeit under
conditions radically different from those prevailing in the more
highly urbanised Latin American semi-periphery. In the northern
Balkans, different relations of production in the countryside, and
their geographical distance from the major Mediterranean sea
routes kept rural-urban migration low, both before and after 1929;
this resulted in a type of industrialisation that was marked by the
absence of very large urban industrial concentrations, by a less
inflated tertiary sector, and consequently by levels of underemployment which were much higher in agriculture than in the
cities. 163
After these few remarks on industrialisation, let us now see how
the industrial working classes were brought into the early postoligarchical political arena. The general point here is that in the
parliamentary semi-periphery the industrial working classes made
their appearance at a time when the state apparatus had already
acquired impressive dimensions and, given technological developments, an enhanced capacity for intervention and repression.
When in these circumstances the working-class movement began
to shift from artisanal to modern trade union and political party
organisations, powerful forces were already operating and attempting either to suppress such organisations or to undermine their
autonomy and bring them under state tutelage. The techniques
and the intensity of these processes of suppression and/or incorporation varied quite considerably from one period and one area
to the next, of course.
1.3.2 The working-class movement in the Balkans
A. As already discussed above, even after the intensification of

import-substitution industrialisation during the 1930s the industrial labour force in the northern part of the Balkans was relatively
small, as the bulk of the working population remained in the
agricultural sector. It is not therefore surprising that the tradeunion movement did not attract a mass following in the inter-war
56
years. In the northern Balkans, as in all the societies in our

sample, the artisan-based mutual-aid associations which had been
a marked feature of the late nineteenth century began to give way
to modern trade unions during the first two decades of the
twentieth century, and increasingly so after the emergence of a
small industrial proletariat in the period directly after the First
World War. Trade-union membership was very low however- not
only at the beginning of the century but throughout the inter-war
period.
For instance, in 1903 the whole of Bulgaria's trade-union
movement could only count 1300 members, many of them from
semi-artisanal trades such as tailoring or shoe-making. Just before
the First World War, the two major trade-union federations (the
Free Federation controlled by reformist socialists), and the General Federation (controlled by revolutionary Marxists) had a total
membership of less than 10000; 164 before the abolition of tradeunion autonomy by the 1934 dictatorial regime the two major
federations between them had fewer than 50 000 members. 165
If the inter-war trade-union movement in the northern Balkans
was numerically weak, the development of communist parties was
quite different. Socialist ideas reached the northern part of the
Balkan peninsula via Russia rather than France or Germany. For
instance, most Bulgarian intellectuals who played leading roles in
their country's socialist and communist movements had studied in
Tsarist Russia or were influenced by Russian revolutionaries.
Also, as in pre-revolutionary Russia, Bulgaria's social democrats
belonging to the Second International were divided into a revolutionary and a reformist wing (the so-called 'narrow' and 'broad'
socialists respectively).
After the Russian Revolution, the communist movements, in
both Bulgaria and Yugoslavia, that emerged out of the pre-war
socialist/social-democratic parties developed into important political forces. In Bulgaria the 'narrow' socialists (who broke with the
Second International and joined the Comintern in 1919) managed
during the national elections of the same year to elect 47 deputies
(out of a total of 233) to parliament (the Sobranie). Similarly in
Yugoslavia (or the Kingdom of Serbs and Croats, as the country
was known before 1929) 166 the Socialist Workers' Party of Yugoslavia was established by merging the pre-war social-democratic
parties of the various ethnic groups (Serbs, Croats, Slovenes, and
57
so on). The resultant party joined the Comintern, and in the 1920
national elections obtained 200 000 votes and 58 deputies (out of a
total of 419) in parliament.
The spread of communism in the northern Balkans was not,
however, related to either rapid industrialisation or to processes of
proletarianisation in the countryside. (In fact most of the communist parties' rank and file were not proletarians but peasant
smallholders.) 167 It had more to do with geographical proximity,
ethnic/cultural affinities and, above all, with the traditional Russophilism of the Slav-speaking populations of the Balkan peninsula. That the attitudes of the people of Bulgaria remained positive
towards Russia, despite the latter's blatant and continual interference in the country's internal affairs, is explained by the fact that
Bulgaria's independence came as a direct consequence of Russia's
victory over Turkey in 1878.
For all that, the success of the communist ideology in the
predominantly agrarian northern Balkan societies should not be
exaggerated. Its appeal, while important, was not predominant in
the inter-war years. As has been pointed out already, the major
post-oligarchic force that emerged in the northern Balkans after
the First World War was not communism but 'peasantism': that is,
massive populist peasant movements that were hostile to both
communist and bourgeois/urban parties. Especially in Bulgaria,
neither the small urban and rural bourgeoisie nor the Moscowcontrolled Communist Party were able to win over the majority of
the peasants; and on the international level, the Comintern's
attempt to create a Red Peasant International in the 1920s was
much less successful than Stamboliiski's 'Green International'. 168
A look at the relationship between the communists and the
trade-union movement shows that both in Bulgaria and Yugoslavia communist control over the tiny movement was considerable
(though more so in unions of wage earners and less in those of
artisans or civil servants), until these communist parties were
outlawed (in 1921 in Yugoslavia, in 1925 in Bulgaria). While
operating underground they, of course, set up front organisations
which were active in the trade-union movement as in parliament.
But given the hostile environment with which they had to contend,
these front organisations did not manage to exercise the same
influence as before. For instance, in Yugoslavia the Independent
Workers of Yugoslavia Party (a front for the banned communists)
58
managed in the 1923 elections to get only a small fraction of the

votes the Communist Party had obtained three years earlier.
The Labour Party of Bulgaria, set up in 1926 as a front after the
communists had been outlawed, enjoyed a relatively more successful career. Despite constant harassment it managed to gain
considerable electoral support in both villages and towns, especially after the 1929 economic crisis and the concomitant rapid
deterioration in the peasants' living conditions. As the moderate
agrarians were at that time part of the governing coalition, the
Labour party managed to gain some of the peasants' protest votes.
(In the June 1931 national elections, the Labour party, under the
new label of Toilers' Bloc, received 160000 votes and 31 seats in
the Sobranie. 169 ) In the trade-union movement the frontorganisation tactics of the Bulgarian Communist Party were less
successful however. At the time of its dissolution in 1924 the
communist-controlled General Trades Union Federation had had
35 000 members, but the Independent Trades Union Federation
set up by the communists in 1925 as a front organisation never
managed to recruit such high numbers. At its first Congress in
November 1927 it had approximately 7000 members and by the
time of the establishment of the 1934 dictatorship its membership
was 11 OOOY0 Needless to say, with the over-all suppression of
parliamentary politics in Yugoslavia in 1929 and in Bulgaria in
1934 the independent trade unions ceased to exist, and both
working-class and peasant parties had to operate clandestinely.
B. The major difference underlying the post-oligarchic development of the working-class movements in Greece and the northern
Balkans was that the Communist Party was less strong in Greece in
the inter-war years. So in the early 1920s, at a time when the
Bulgarian Communist Party could obtain 20 per cent of the
popular vote and become the main opposition party to Stamboliiski's agrarian government, the Greek communists failed to put a
single member into parliament. (This was true in both the November 1920 and the 1923 elections.) It was only in 1926 that the
communists had their first election success and obtained 10 seats
(out of a total of 286) in the Greek parliament. In 1936, with 5.7
per cent of the total vote and 15 parliamentary seats, the Greek
Communist Party's parliamentary strength reached its zenith; and
in spite of the relatively small number of deputies, the communists
59
managed to play a crucial balancing role in parliament where

neither of the two major bourgeois parties (Liberals and Populists)
had won a working majority. 171
What was the relationship between the communists, their
socialist predecessors, and the trade-union movement? Some
strike activity notwithstanding, prior to 1909 most of the Greek
workers were organised in mutual-aid societies. It was only after
1909 that trade-union organisations and labour centres made their
appearance in certain major Greek towns, with a number of
individuals who, inspired mainly by French and German socialist
ideas, were trying to influence the nascent trade-union movement.
These socialist intellectuals, unlike their Bulgarian counterparts
did not, however, manage to constitute themselves into ideologically and organisationally cohesive political groups. For instance,
although Greece already had both reformist and radical or revolutionary socialist intellectuals, 172 there was none of that highly
institutionalised split which differentiated Bulgaria's 'narrow' and
'broad' socialists - a split that already at the beginning of the
century was reflected in two competing federations within the
trade-union movement. This means that although in Greece the
inter-war trade-union movement was numerically larger than in
Bulgaria, 173 the latter's proximity to Tsarist Russia and the
consequent strong links of her intelligentsia with revolutionary
circles made her socialist groups more coherent, more determined,
and from the very start more capable of taking partial control of
the trade-union movement.
In Greece the early socialist groups were not only less cohesive,
but the break-up of oligarchic parliamentarism had occurred ten
years earlier than in Bulgaria. In the 1920s therefore, Greece
already had a reformist government with the political will and
capacity to introduce significant social legislation for the improvement of working-class conditions. This combination of a relatively
weak socialist tradition and a relatively early post-oligarchicreformist government meant that the latter could exercise greater
control over the infant trade-union movement than was the case in
Bulgaria (where at that time a more active and better organised
socialist intelligentsia existed, alongside a government that was
indifferent or hostile to working-class demands).
As soon as Venizelos came to power in Greece he tried to cope
with the 'social question' by applying the paternalistic carrot-and-
60
stick techniques that are typical of most semi-peripheral societies

in their post-oligarchic phase, that is, granting certain rights and
benefits to the working class, while at the same time instituting
restrictive rules to curtail trade-union autonomy. So the decade
after 1909 saw the implementation of a series of laws covering
working hours, woman and child labour, industrial accidents, and
so on, coupled with provisions for putting the trade-union movement under state tutelage. 174
This co-optation/incorporation approach worked pretty well up
to the end of the First World War (during the war years some
socialist intellectuals were persecuted because of their anti-war
attitudes). In fact, the way in which the General Confederation of
Greek Labour (GSEE) was instituted in 1918 is a very clear
illustration of the degree of control the Venizelos government
exercised over the working-class movement. The initial and decisive push for the creation of the GSEE did not come from the
trade unions themselves so much, or from the socialist intelligentsia, but from the Liberal government which wished for
reasons of both internal and foreign policy to set up a unified
trade-union movement. In addition to considerations of tighter
state control, Venizelos believed that western Europe's labour and
socialist movements would influence the outcome of the future
European peace talks, 'in which case the support of Greece's
labour and socialist movements for the war aims would be
indispensable. ' 175 It was for the same reason that the various
socialist groups were allowed to unite (forming SEKE, the Socialist Labour Party of Greece) and to establish links with the GSEE.
However, once the support of the trade-union and socialist
movements in respect of the war aims was deemed no longer
necessary, the socialists were persecuted and the government
systematically eliminated any feeble control they had managed to
obtain over the unions. 176
With the creation of the Greek Communist Party, however,
governmental control of the trade-union movement became more
problematic. In 1920 the Socialist Labour Party of Greece decided
to withdraw from the Second International and to join the
Comintern (the official title of Communist Party of Greece was
adopted in 1924). In the same year Venizelos was defeated in the
national elections and his Liberal Party lost control of the GSEE.
From 1920 until the establishment of the short-lived Pangalos
61
dictatorship (January to August 1926) it was the communists who

controlled the GSEE. The dictatorship then outlawed the Communist Party and imprisoned or exiled its leaders. Although a
general amnesty was declared as soon as the dictatorship fell, the
communists never managed to regain control of the General
Confederation and, in view of that failure, they decided to form a
rival federation (the Unitary Confederation of Labour) in 1929.
During the Depression years with the rise in the cost of living and
increasing unemployment, the Communist Party inaugurated a
successful drive for a 'united front from below', as a result of which
the two confederations began to co-operate in the co-ordination of
strike activities. The radicalisation of the working-class movement
reached its peak in 1936 with a massive strike of tobacco workers
in Macedonia which led to the death of 22 strikers, and a
subsequent general strike initiated by both federations. One day
before the launching of a second general strike, John Metaxas
stepped in and established a dictatorial regime which put an abrupt
end to parliamentary politics and trade-union autonomy in interwar Greece. 177
What stands out most clearly in a comparison of the inter-war
working-class movement in Greece and the northern Balkans is
the exceptional electoral strength of communism in the latter, a
strength which was only partially based on the industrial working
class. It was partly as a result of that strength that the communist
parties of Bulgaria and Yugoslavia had already been outlawed by
the 1920s, whereas in Greece (with the exception of the brief
dictatorial interlude under Pangalos) the communists were
allowed to operate legally until the end of the inter-war parliamentary period in 1936. This difference notwithstanding, it must be
stressed that in all three countries, but particularly in the northern
Balkans, the industrial proletariat remained numerically weak (in
comparison with both western and central Europe) - even after
the post-1929 intensification of the import-substitution industrialisation process. The highly politicised trade-union movement,
whether controlled by communists, socialists or liberal reformists,
did not achieve the type of massive strength and autonomy
vis-a-vis the state that western European trade unions had
achieved towards the end of the nineteenth and the beginning of
the twentieth century. This was true in the 1920s and even more so
in the 1930s when, in all three countries, the rise of right-wing
62
dictatorships completely destroyed the meagre and precarious

autonomy which part of the trade union movement had acquired.
1.3.3 The working-class movement in southern Latin America
In the Latin American southern cone too the inter/war workingclass movement was unable to achieve a high degree of autonomy,
although the way in which it was controlled by the state took a
somewhat different form. Given the higher rate of urbanisation
and the greater strength of the landowners, the decline of oligarchic parliamentarism was followed by fairly systematic attempts to
not simply control the existing trade unions, but also to create new
unions from above, by organising, and granting, considerable
social benefits to the mass of workers leaving the countryside to
seek work in the big cities. The new, state-created unionism was
used by the post-oligarchic political elites both as a means for
undermining the strength and autonomy of the old, socialist or
communist controlled unions, and as a weapon against the still
powerful landowning oligarchy. This pattern of co-optation/incorporation first emerged clearly in Chile with Ibanez in the 1920s,
but it was only later, with Vargas in Brazil and Peron in Argentina,
that it took its most developed form.
A. Chile had, by Latin American standards, one of the largest
and best-established labour movements in South America. Not
only did the relatively early urbanisation create a sizeable working
class in the big cities (particularly in Santiago and Valparaiso), but
large concentrations of wage earners emerged very early on with
strong, militant organisations around the important nitrate fields
in the north and the copper mines in the south. The first central
labour organisation, the Gran Federacion Obrera de Chile
(FOCH), was set up as early as 1909 and in 1916 came under the
control of Luis Emilio Recabarren's Partido Socialista Obrero
(PSO). In 1922, after the PSO had joined the Comintern, FOCH
joined the newly-organised Red International of Labour Unions.
However, as in most countries of the parliamentary semiperiphery, the major initiative for the break-up of the oligarchic
system and the subsequent enactment of social legislation did not
come from the trade-union movement:
63
organised labour was a spectator rather than active participant

in these events. A new constitution, a labour code, social
legislation - all were passed without consultation with the
unions. 178
In fact it was Alessandri's government in the early 1920s that first
tried to introduce extensive labour legislation. But, as mentioned
earlier, in view of strong congressional opposition it was only
under military pressure that the Labour Code was finally enacted
in September 1924. In 1925, when Alessandri returned from exile,
he tried to implement some of the clauses of that Code, but it was
only under Ibanez's authoritarian rule (1927-31) that comprehensive labour measures were significantly extended and consolidated. These measures combined giving considerable social benefits to the workers with the persecution of communists, the
suppression of the communist-controlled FOCH, and the creation
in its place of hundreds of 'legal unions'. These legal unions were
co-ordinated from the top by the creation of two state-controlled
central labour organisations which in 1931 merged into the Confederaci6n Nacional de Sindicatos Legales (CNSL).
On the whole the unions were ambivalent about the Labour
Code. Some were hoping that by becoming 'legal' they could use
its provisions to bring more workers into the trade-union movement and subvert state controls from within; others, after Ibanez's
manipulative use of the Code, were afraid of its incorporative
effects and decided to continue operating outside the legal
framework. 179
After Ibanez's fall the communists resurrected FOCH and kept
it under their control, whereas the newly-founded Partido Socialista (set up in 1933 by backers of Marmaduke Grove in the 1932
presidential elections) managed very quickly to dominate the
'legal' Confederaci6n which was the largest central labour
organisation. 180
In 1935, following the Comintern's new 'popular front' line, the
communists launched a drive for the unification of the trade-union
movement which, a year later, resulted in the Confederaci6n de
Trabajadores de Chile (CTCH) within which socialists and communists fought incessantly for supremacy until in 1946 it split into
separate socialist and communist-controlled confederations. 181
64
There are two major points worth emphasising about Chile's

inter-war working-class movement. The first is that, given the
country's exceptionally strong/autonomous working-class movement, urban populism from above (as a mode of transition to
post-oligarchic politics) was not as accentuated as in Brazil and
Argentina. To be more specific, Ibanez's attempt in the 1920s to
undermine the old, communist-controlled trade unions by the
creation of new, state-controlled unionism did not succeed. The
communists survived Ibanez's persecution and after his downfall
emerged as a major political force controlling a large section of the
working-class movement. Moreover, the new Socialist Party, by
almost immediately after its creation assuming control of the
'legal' confederation (CNSL), made sure that the state could not
use the legal unions as a means for subverting the autonomy of
those communist and syndicalist-controlled unions which refused
to become legal. As Alan Angell puts it:
There was not in Chile, as there was in Argentina, a sharp
break between an 'old class' of labour leaders and a 'new class'
formed by different circumstances in response to different
pressures (such as massive industrialisation or the emergence of
leaders like Peron or Vargas). The continuity of ideas, attitudes, structures and leaders make it important to understand
the tradition of the Chilean labour movement. 182
In consequence, when Ibanez returned to power he could not given the exceptional strength of the working-class movement and
despite his populist/authoritarian tendencies - do in the early
1950s what Vargas had done in the 1930s and Peron in the 1940s:
he could not by means of unionising internal migrants transform a
large section of the labour movement into an administrative
extension of the Ministry of Labour.
Even so, the importance of a second major feature of Chile's
trade-union movement must not be underestimated. This was the
highly restrictive impact of the 1924 Labour Code on its development. It had in fact been one of the chief aims of the Code to
curtail the financial strength and independence of the unions. So,
for instance, it stipulated that in principle the unions were not
allowed to set aside funds for strike activities; neither were they
permitted to pay dues to the federation to which they belonged.
65
Moreover, their finances became subject to control by labour

inspectors, and most of their funds had to be placed in government-supervised bank accounts. The ensuing financial weakness of
the unions made them come to depend for their funding on the
political parties, and in this way one of the unintentional consequences of the 1924 Labour Code was to accentuate the politicisation
of the trade-union movement. 183
Another major objective of the Code was to severely restrict the
process of collective bargaining. Under its provisions the local
plant union was considered as the main negotiating unit. Agreements had to be negotiated not on an industry-wide, national level
but locally between the employer and the workers of a particular
plant. This system of bargaining tended to work against the
creation of larger units, for if agreements have to be negotiated
locally, then the basic unit will be the local one. 184 This tendency
towards fragmentation was further reinforced by the Labour
Code's rigid distinction between white-collar and blue-collar unions, with the former receiving preferential treatment in terms of
social benefits, minimum salary levels and so on. In conclusion, if
one takes into account the usual structural obstacles to the
development of strong trade unions that obtained in Chile just as
in all other semi-peripheral economies (for example, the existence
within the industrial sector of a plethora of small artisanal familyrun units), the 1924 Code emerges as a factor further reinforcing
such structural obstacles and so undermining trade-union strength
and autonomy.
B. Chile's 'pioneering' pattern of co-optation/incorporation as it
developed in the immediate post-oligarchic period was adopted
and further refined in the 1930s and 1940s by several other Latin
American countries. In Brazil for instance, the dictatorship imposed by the civilian-military alliance in 1930 allowed Vargas to
rapidly bring in a series of measures which granted considerable
social rights to the workers, while simultaneously establishing
strict state controls over the labour movement. In fact, Vargas
created the first Ministry of Labour, as well as retirement and
pension funds that covered a large number of workers. Other
measures were aimed at improving work conditions, providing
cheap housing loans to workers, fixing minimum wages and so on.
On the other hand, as early as 1931 the government withdrew its
66
official recognition from all unions and thereafter granted legal

status only to unions of its liking. The 1934 constitution systematised all labour legislation proposed since 1930, and set up an
incorporative system of controls which enabled the government to
intervene in the affairs of the unions by checking their records,
attending their meetings, and even approving/disapproving their
elected leadership.
These developments transformed the government from 'a declared enemy of organised labour into its patron, protector and
domesticator' .185 The anarcho-syndicalists and communists who,
since 1929, had controlled the Confederacao General dos Trabalhadores do Brasil of course reacted to Vargas's attempts at 'domestication'. In 1935 they founded the Allan~a de Liberacoa Nacional
which, under the leadership of Luiz Carlos Prestes, attempted an
unsuccessful military insurrection in Rio de Janeiro, Rio Grande
do Norte, and Pernambuco. After this failure the Alliance was
liquidated, Prestes went to jail and the communists were severely
persecuted.
With the establishment of the Estado Novo in 1937 both the
carrot and the stick aspects of Vargas's previous labour policies
were further amplified. The 1937 constitution, the labour clauses
of which had been inspired by Mussolini's labour code, made
strikes and lock-outs illegal. It imposed compulsory arbitration by
special labour courts headed by state-appointed judges, and it
instituted a state-monitored system of union finance. On the other
hand, a series of laws were passed to enhance workers' welfare (an
eight-hour day, paid holidays, medical services, and so on).
Although towards the end of the Estado Novo period growing
inflation and declining wage levels created signs of unrest within
the trade-union movement, on the whole Vargas managed during
his first period of rule (1930--45) to put the relatively weak
working-class movement firmly under his thumb. What helped his
incorporative efforts considerably were not only his social-welfare
measures but also his active encouragement of the creation of new
unions. This brought thousands of rural immigrants into the
movement, whose paternalistic attitudes and newly-acquired
urban status and privileges made them ardent Vargas
supporters. 186
Towards the end of the Estado Novo in 1945 there was an initial
attempt at liberalisation. A political amnesty was declared and the
67
communists, in collaboration with Vargas's supporters, in 1946

created the Confederacao dos Trabalhadores do Brasil (CfB). For
a while the communists gained considerable influence under
Prestes's charismatic leadership, both in the sphere of electoral
politics and in the trade-union movement. However, right from
the beginning of the Euricio Dutra regime the communists decided
to operate within the Estado Novo framework of legallyrecognised unionism rather than create parallel, independent
unions. 187 Moreover, since the basic anti-communist provisions of
the Estado Novo legislation were never abolished, the Supreme
Court could declare both the Communist Party and the CfB
illegal in 1947. Although the communists continued to operate
through a variety of front organisations, they lost most of the
influence they had gained between 1945 and 1947.
When Vargas came back to power in 1950 he resumed, this time
in a more liberal framework, his basic strategy of co-optation/
incorporation in relation to the trade unions. The communists who
had already supported him towards the end of his Estado Novo
rule and who closely co-operated with him during the Dutra
presidency continued, despite occasional criticism, to work with
the most demagogic fringe of Vargas's followers right down to the
old dictator's death, and had sought to be more Vargasista than
Vargas after his suicide. 188
Given all of the above, it is not surprising that up to the 1964
military dictatorship several provisions of the Estado Novo legislation were still in force in Brazil; and that, despite the creation of
'parallel' unions, many unions still resorted to the legally prescribed labour courts rather than to collective bargaining as a
means of improving wages and work conditions. 189
C. In Argentina, finally, given the massive immigration of workers from Europe with anarcho-syndicalist and socialist ideas, the
trade-union movement developed very early. The first federation,
the Federacion Obrera de Ia Republica Argentina (FORA) was
founded as far back as 1890. During the first two decades of the
twentieth century the syndicalists were more influential in the
trade unions than the socialists; in the 1920s and 1930s however
the latter managed to prevail over both syndicalists and
communists. 190 They were also able during Yrigoyen's rule to
achieve some representation in congress: in the 1924 elections they
68
had 13 deputies and two senators; and in 1932, given the radicals'
abstention from the elections, the number of socialist deputies
more than trebled. All the same, with few exceptions the Socialist
Party did not manage to spread its influence beyond the Buenos
Aires area and, with the drastic curb on external migration from
1930 onwards, it entered into a phase of long-term decline. What
contributed to this decline were its poor organisation, its internal
divisions, and its persistent internationalist orientation at a time
when foreign/European ideas had little appeal to first-generation
Argentinians born to immigrant parents, or to the growing mass of
internal migrants pouring into the cities. 191
With respect to the state's post-oligarchic policies vis-a-vis
labour, Yrigoyen's rise to power meant that the radicals adopted
an attitude towards the unions that was rather less hostile than that
of the former oligarchy. Yrigoyen introduced a rudimentary
programme of social legislation and, at least initially, looked
favourably on strikes aimed at improving working-class conditions. At other times, however, he did not hesitate to suppress
workers' agitation by the oligarchic methods of brute force. (This
reversal became particularly evident during the oligarchic restoration in the 1930s and early 1940s.) As already mentioned, his
populist successor, having a less ambivalent attitude towards the
unions, and seeking working-class rather than middle-class support, managed to use the 'carrot and stick' method much more
systematically.
After the military coup of 1943 Peron gradually emerged as the
leading force within the government. He quickly drew up a series
of legislative measures which the trade unions had unsuccessfully
tried to introduce for a decade or more. These measures improved
and extended the social-security system which by 1946 covered 1.5
million persons; it protected urban workers against accident and
unfair dismissal, raised retirement benefits, and so on. After the
victory of the Peronist coalition in the 1946 elections Peron not
only extended his progressive labour legislation, but also initiated
a series of economic policies which led to a number of pay
increases for both skilled and unskilled workers. From 45.2 per
cent in 1946, industrial wages and salaries rose to 50.2 per cent of
the over-all national income in 1948, and to 56.7 per cent in 1950.
What about the 'stick' side of the equation? Although in 1943
Peron had suspended the unpopular Law on Professional Associa-
69
tions which had allowed the state an extensive say in union affairs,
in 1945 he reintroduced it in a new version which increased state
control over the unions. The new law empowered the government
to grant or withdraw the unions' legal.status, and stipulated that a
single union must represent all workers in any one industry. By
means of this incorporative framework, Peron managed to draw
into the trade-union movement the growing mass of labour moving
from the countryside into the cities. Between 1946 and 1951 the
number of unionised workers quintupled: from 500 000 to 2.5
million. With the overwhelming support of the newly-unionised
workers, and by the skilful manipulation of both licit and illicit
weapons, Peron succeeded in neutralising all those forces within
the trade union movement (mainly socialists and communistcontrolled unions) which were opposed to his populist co-optation/
incorporation tactics. For instance, 'recalcitrant' unions were
systematically replaced by rival, pro-Peron unions, the latter
receiving legal recognition and being granted the exclusive right to
negotiate with employers. 'Recalcitrant' trade-union leaders,
especially communist ones, were often imprisoned, exiled, or
liquidated. Given these tactics and in view of Peron's enormous
popular appeal, neither the communists nor the socialists were
able to put up an effective resistance to his brand of populism.
Opposition from within the trade unions only started to grow again
after 1950 when, due to the deteriorating economic situation,
Peron was forced to reverse his pro-industry, pro-worker
policies. 192
Finally, what needs stressing in the case of Argentina is that,
although the pre-Peron unions and left-wing parties were stronger
than in Brazil, Peron succeeded even more trenchantly than
Vargas in undermining the autonomy and dynamism of the old
unions. This was due not only to favourable over-all conditions
(the massive influx of rural migrants, the declining attraction of
internationalist/socialist ideas, and so on), but also to Peron's
highly charismatic appeal to the masses and to his extreme
ruthlessness against his left-wing opponents.
This brief survey of the working-class movements in the parliamentary semi-periphery makes it quite clear that neither in the
70
Balkan peninsula nor in southern Latin America were the political

and trade-union organisations of the working classes able during
the early formative years of the post-oligarchic period to develop
an autonomous power base and to put up effective checks to the
state's incorporative tendencies. 193 In the northern Balkans where
communist parties were relatively strong, their outlawing in the
1920s diminished both their over-all political appeal and the
influence within the trade-union movement. In Greece where both
socialists and communists were weak, the state (particularly during
the early post-oligarchic period) could subjugate a sizeable part of
the trade-union movement through the usual co-optation/incorporation techniques. Moreover, in all three Balkan countries the
post-1929 intensification of import-substitution industrialisation
increased the ranks of the proletariat and the militancy of unions but at the same time the royal dictatorships that emerged put an
end to such militancy and eliminated the little autonomy that
Balkan trade-union movements had enjoyed in the previous
decade.
In the southern cone countries of Latin America, where urbanisation was more intense, post-oligarchic populist leaders managed (with the partial exception of Chile) to undermine the
autonomy of the trade-union movement by the mobilisation and
unionisation from above of all those who were abandoning the
landlord-dominated countryside in the hope of finding work and a
better life in the cities.
Before concluding this section it should perhaps be noted that it
was not only the industrial working classes that had a verticaV
dependent relationship to the post-oligarchic state; the industrial
bourgeoisies, though in a different way, of course, also stood in a
much more dependent relation to state economic policies than
their western counterparts. 194 In view of the difficulties of competing with the industrialised West it was not only the establishment
of these industrial bourgeoisies195 but also their continuation
which had to rely on state protectionism and heavy subsidisation.
Their enormous privileges and profits were due much less to their
competitiveness than to the hot-house conditions in which they
existed and functioned. It is not therefore surprising that the more
liberal and reformist ideas that began to spread in the parliamentary semi-periphery after the Napoleonic Wars were not championed by industrialists but by intellectuals, bureaucrats, and the
71
military . 196 Neither did industrialists portray the technological

inventiveness of their European counterparts. As Claudio Veliz
very aptly puts it in a reference to Latin American industrial
capital:
The flow of state capital into industry was guided by people who
belonged largely to the same social groupings of those who
occupied key posts in the administrative and political structure.
The wealth they acquired resulted almost inevitably from their
holding strategic positions in the critical sectors of economic
and political decisions at a time that, from their personal point
of view, must have been both fortunate and fortuitous.
Together with their friends, relatives and associates, they soon
constituted a kind of industrial clientele deeply conscious of the
fact that their well-being depended in large measure on their
remaining a loyal and co-operative extension of the central state
apparatus. Their newly acquired wealth was not industrial
capital resulting marginally from a style of life; there was no
cultural or religious infrastructure behind their association with
industrial ventures, there had been no commitment to austerity
or general reforms and improvement in the educational system
leading to substantial technological innovation; but most important of all, there was no cultural ethos associated with the task
of industrialisation. 197
What to me seems even more important than the lack of a
'cultural ethos associated with the task of industrialisation' are the
severe structural limitations within which Latin American and
Balkan industrial capital had to operate, that is, competing
internationally with the highly industrialised western European
economies from a subordinate position in the international division of labour, the impossibility of building up an autonomous
technological base at a time when industrial technology was
already very complex and therefore beyond the know-how of
indigenous craftsmen, and so on. 198 But whether the stress is on
the cultural orientations or on the structural international and
national position of industrial capital in the semi-periphery, the
fundamental point remains that, being weaker and more
dependent 199 than its western counterpart, it played a lesser role in
reforming or 'rationalising' the post-oligarchic state administration
72
and political system in general - a system which retained most of

its particularistic, personalistic features from the oligarchic
period. 200
Conclusion
In the General Introduction it was noted that the Balkan and Latin
American southern cone societies before their independence were
parts of two large patrimonial empires within which those who
controlled the means of domination were consistently hostile to
the formation of autonomous power groups mediating between
the state and the people. These marked incorporative features of
the patrimonial state persisted, albeit in different form, in the
post-independence period, and were consolidated, and in some
respects reinforced, by subsequent political and economic developments.
More specifically, given that the demise of oligarchic parliamentarism occurred in a predominantly pre-industrial context, the
opening up of the political system was not, as in western Europe,
characterised by the active participation in the post-oligarchic
political arena of the industrial classes (particularly massive and
autonomous working-class organisations). Instead the new participants were brought into the political game in a more dependent/
vertical manner, through populistic and clientelistic means.
Moreover, the rapid post -1929 industrialisation which, particularly
in Greece and the southern cone countries, considerably increased
the ranks of the industrial proletariat, did not lead to the formation of autonomous trade-union organisations able to put an
effective check on the state's incorporative tendencies. With few
exceptions, the post-1929 working-class movements in the semiperiphery were suppressed or manipulated/controlled from above
through a variety of means. In other words, both the postoligarchic broadening of political participation and the post-1929
rapid industrialisation did not weaken but rather reinforced/consolidated the authoritarian, incorporate state features that the semiperipheral countries had inherited from their pre-independence
days.
The relevance of the above for the post-war rise of highly
repressive military dictatorships in several semi-peripheral
societies will be discussed in Part II.
Chapter2
Theoretical Implications ( 1):
Clientelism and Populism as
Modes of Political Incorporation
This short chapter will further develop some of the key concepts
used in Chapter 1, as well as relate them briefly to the more
general theoretical debates in the relevant literature. With this aim
in mind I shall begin with some remarks on the concept of
incorporation - a concept used extensively in the preceding pages.
2.1 On the concept of incorporation
At the risk of overgeneralisation it can be argued that in several

societies of the capitalist centre the development of strong trade
unions and working-class parties, at a time of restricted levels of
state expansion and against a historical background encouraging
the formation of a corps intermediaires between state and people,
resulted in the establishment of strong and autonomous civil
society. (The emphasis on both strength and autonomy is important here, since these two variables are not always consonant. So,
for instance, there may be trade unions not vertically controlled by
the state which are weak in terms of numbers and/or resources;
whereas at a later stage, increases-in membership/resources can, as
in Per6nist Argentina, coincide with a strengthening of state
tutelage). Within a parliamentary context a strong and autonomous civil society consists of a variety of interest groups representing both upper- and lower-class, as well as non-class interests.
Such groups are capable of setting serious limits to state encroachment and manipulation, and so contribute to the establishment of
73
74
the principle of civilian supremacy over those holding the means of

coercion and administration. As far as working-class interest
groups are concerned, such a situation implies that they are
brought into the expanding political arena in a less vertical, more
autonomous fashion. By autonomy here I mean both the relative
independence of lower-class organisation vis-a-vis the state, as
well as their independence vis-a-vis leaders attempting to undermine or bypass organisational structures in their efforts to mobilise
and lead the masses in plebiscitarian, paternalistic fashion. For
lack of a better term one can call this more horizontal mode of
lower-class inclusion into politics integrative, in order to distinguish it as clearly as possible from more vertical forms of inclusion
that are more typical of the capitalist periphery and semiperiphery. In the latter case, not only did the pre-independence
state inhibit the development of autonomous corps intermediaires,
but the timing and structure of industrialisation 1 after independence strongly accentuated the paternalistic/authoritarian tendencies of the state (as well as the tendencies of leaders of the lower
class to relate to their followers in similar fashion).
In fact, the state's tendency to inhibit the formation of autonomous interest groups appears to be a constitutive feature of the
semi-peripheral polity. The precise form taken by state authoritarianism changes of course from pre-independence to postindependence, and from oligarchic to post-oligarchic parliamentarism; but throughout the different forms of political representation runs the state's persistent hostility towards social forces
attempting to relate to it from a relatively autonomous position.
This being so, lower-class organisations in the semi-periphery are,
particularly as they grow in numbers and organisational resources,
beset by an 'overdeveloped' 2 state constantly attempting to impose
its tutelage over them.
In this situation, and with regard to the over-all process of
broadening political participation in the parliamentary semiperiphery, one can characterise as incorporative the way in which the
lower classes were brought into politics and the way in which their
representative organisations were and still are related to the state.
The concept of incorporation is worth developing a little further. As used in this study, incorporation is quite distinct from both
the notion of corporatism as a regime form, and from what Philip
Schmitter has labelled societal corporatism. 3 Contrary to regime
Theoretical Implications (I)
75
corporatism (or state corporatism, to use Schmitter's term),

incorporation does not imply the total abolition of parliamentary
democracy. In state corporatist regimes (for example, Mussolini's
Italy) the incorporation of the lower classes is not only more
pronounced, it also takes a de jure form: spontaneously formed
interest groups and associations are not merely controlled from
above, they are outlawed and replaced by a limited number of
authoritatively recognised groups that interact with government
apparatuses in defined and regulated ways. 4 In consequence,
despite the persistence of limited social pluralism, there is no
genuine political pluralism - a situation which sharply differentiates state corporatist regimes from semi-peripheral parliamentary
democracies, with the latter enjoying a higher degree of both
social and political pluralism.
Incorporation must also be distinguished from societal corporatism. This refers to the type of negotiated agreements which the
state in the western parliamentary democracies is increasingly
trying to establish with trade unions and employers' associations.
Even in cases where such arrangements have led to the quasielimination of strikes (for example, Switzerland), such situations
are very different from the incorporative type of control over trade
unions found in the periphery and semi-periphery. In the one case,
the corporatist regulation of the working classes is based on
collective agreements freely negotiated in a situation where trade
unions are relatively strong and autonomous; in the other, the
trade unions have not reached a comparable level of autonomy
and cannot, therefore, negotiate from a position of strength.
In other words, corporatism as generally used always implies de
jure arrangements between the state and various civil-society
associations for the purpose of avoiding 'social strife': in the case
of state corporatism the legal arrangements are imposed by force
from above; in the case of societal corporatism they emerge
through negotiations and eventual agreement, with the parties
concerned maintaining their relative autonomy vis-a-vis the state.
On the other hand, incorporation as used here refers to the de
facto control exercised by the state over associations which, while
on paper free from legal commitments to keep the 'social peace',
are weak and therefore easily subjected to state manipulation and
control.
A final point concerning the concept of incorporation: incor-
76
porative relations of domination should not be seen as necessarily

leading to state corporatism. A look at the long-term trajectories
of Argentina, Chile and Greece, for instance, shows that although
there were attempts in all three countries to establish state
corporatism, these attempts invariably failed. So when in the
post-war period the typical incorporative control mechanisms of
the parliamentary system were shattered, they were replaced, not.
by a state corporatist regime, but by the imposition of a military
dictatorship (see Chapter 3).
2.2 Populism and clientelism as rnodes of political inclusion
A. Having clarified the concept of incorporation, let us now turn

to the two modes of incorporation which are typical in the polities
of the parliamentary semi-periphery: clientelism and populism.
Starting with clientelism, a distinct and, from the point of view
of the maintenance of the status quo, rather safe solution to the
problems created by political mobilisation and enlarged political
participation in peripheral and semi-peripheral formations is the
use of vertical networks of patron-client relationships for bringing
lower-class strata into national politics. Indeed, extensive clientelist networks, cutting across and hindering the direct producers'
horizontal organisation along class lines, were a typical form of
organisation in oligarchic parliamentary politics.
Contrary to the belief of neo-evolutionist political scientists,
'modernisation' does not eliminate clientelism; patronage networks tend to persist in a modified, less traditional form even after
the decline of oligarchic politics and the development of industrial
capitalism. As many empirical studies have shown, the entrance of
the urban middle and lower classes into politics is perfectly
compatible with the continuation of vertical clientelistic forms of
organisation in many peripheral and semi-peripheral capitalist
formations. 5
Of course, city and market expansion and the advent of mass
politics have considerably changed the traditional forms of patronage that characterised oligarchic politics. The rising dominance of
the capitalist mode of production, state expansion, the emergence
and political activation of new social strata eager to break the
oligarchy's political control - all these have weakened the
77
exclusive patronage monopoly that local potentates had exercised

over the rural populations. As I have argued for the case of
Greece, these changes contribute to the multiplication and diversification of patrons at the local level, the centralisation of
clientelistic parties, and to the more direct involvement of clients
with party and state bureaucrats. In other words, such processes
bring about what A. Weingrod has called a shift from oligarchic/
traditional to state/bureaucratic forms of patronage. 6
Still, significant as these changes may be they do not alter the
basic effect of clientelistic politics on the maintenance of the status
quo: even in its modern, centralised and diversified forms clientelism continues to operate as a means for the vertical inclusion of the
working population into active politics. If and when it successfully
cuts across and weakens horizontal modes of political integration,
it safeguards the status quo against any serious threat from below,
and it draws the new political entrants into types of conflict where
fundamental class issues are systematically displaced by personalistic politics and particularistic squabbles over the distribution of
spoils. 7
B. Another mode of political inclusion prevalent in peripheral
and semi-peripheral capitalist formations is populism. Given the
huge disruption in the periphery created by the process of capital
accumulation and the relatively abrupt 8 entrance of the masses
into politics, clientelist networks often fail to accommodate the
new political participants in their vertical structures. Since the
western-type dominance of strongly institutionalised horizontal,
integrative organisations is lacking, populism can provide another
framework for bringing in the lower classes. Populism- whether in
the form of distinct organisations like parties, or in the more
diffuse form of movements occurring either as part of or separate
from non-populist organisations- always involves a specific type of
political mobilisation of the masses and their involvement in
politics: a type of political mobilisation/inclusion which, although
vertical, is quite distinct from the clientelistic one.
Certainly (as shown in Chapter 1) as far as the transition from
oligarchic to mass politics in peripheral quasi-parliamentary regimes is concerned, the passage from 'traditional' to 'bureaucratic'
clientelism and the rise of populist movements can be seen as two
different ways of undermining oligarchic politics and bringing
78
about or consolidating less restrictive forms of political

participation. 9 In both cases the end result is the break-up of the
political monopoly of a small number of powerful families, the
entrance of new men into the power game, the transition from
political clubs or coteries of notables to more extensive forms of
political participation, and the introduction of a series of reforms
in the fields of labour legislation, education and so on.
But despite such similarities there are fundamental differences
between clientelistic transformation and populist mobilisation as
mechanisms of political inclusion. In the clientelistic case, the new
men of power manage to break the oligarchic stranglehold by
activating, extending and reorganising to their own benefit already
existing patronage networks - and in that sense there is no radical
departure on the organisational level from the oligarchic type of
authority relationship. In the populist case on the other hand, the
new men achieve their entrance into the corridors of power,
and/or the consolidation of their power position within the nonoligarchic political arena, through a type of mobilisation which,
both ideologically and structurally/organisationally, makes a more
radical break with ancien regime politics.
2.3 Some basic differences between clientelistic and populist
incorporation
An attempt to differentiate in ideal typical terms 10 between

the populist and the clientelist modes of inclusion brings out the
obvious fact that the ideological themes of the populist discourse
focus predominantly on the antagonism between the 'people' and
the 'establishment', the poor versus the rich, and so on, 11 themes
which, as a rule, play a lesser role in non-populist ideologies
(either because they are not emphasised, or because they are not
fully systematised and incorporated into an over-all Weltanschauung). In Greece, for example, Venizelos's 'modernising'
programme and ideology were not so very different from the
dominant ideologies of the oligarchic period. In fact, many Greek
historians quite rightly see Venizelos's reforms as the continuation
or culmination of the ideas and reforms initiated by the great
bourgeois statesman Harilaos Trikoupis in the nineteenth century.
It is perfectly true of course that Venizelism was intensely critical
A.
Theoretical Implications (1)
79
of the corruption and moral degeneracy of the old political

establishment (the paleokommatikoi), that it exhorted the people
to cast off the yoke of the tzakia families, and advocated greater
social justice. But it put much more emphasis on modernisation
and on the heroic panhellenic irredentism of the Megali Idea than
on popular reforms. 12
Stamboliiski's peasant populism in Bulgaria on the other hand,
and his vehemently anti-establishment, anti-western, anti-town
ideology did constitute a sharp break with the prevailing ideologies
of the oligarchic period. This is borne out particularly by his
consistent hostility for and opposition to the militarism and
chauvinistic nationalism projected by King Ferdinand and the
more traditional Bulgarian bourgeois parties. Stamboliiski's
emphasis that the welfare of the people should take precedence
over schemes of national aggrandisement, his determination to
maintain the autonomy and vitality of the community of small
rural proprietors even to the detriment of rapid industrialisation,
his attempts to forge a 'third road' between international capitalism and communism by building a. 'green international', all this
brings the themes of 'the people' and of 'anti-elitism' into the
centre of his ideological discourse. Moreover, these themes were
integrated into an over-all theory of society based on 'estate'
principles of social organisation. 13
With the ideological differences between populist and nonpopulist reform parties quite obvious, it is equally possible to find
differences on the level of authority relationships and the concomitant organisational structures. In fact on the organisational level,
given the intensive mobilisation element and the relatively abrupt
entrance of the masses into politics, it is plebiscitarian leadership
rather than intricate patronage networks which provides the basic
framework for populist political incorporation. As a rule populist
leaders are hostile to strongly institutionalised intermediary levels,
whether of clientelistic or the bureaucratic type found in western
European political parties or trade unions. The emphasis on the
leader's charisma, on the necessity for direct, non-mediated
rapport between the leader and 'his people', as well as the
relatively sudden process of political incorporation, all lead to
organisational forms of a fluid, gelatinous character. Even in cases
of populist movements with strong grass-roots organisations, in so
far as the rank and file's allegiance is centred on the person of the
80
leader, local or intermediary cadres are left without a structural

basis for establishing some degree of political autonomy vis-a-vis
that leader: most of their power and legitimation is derived more
or less directly from his personal charisma.
B. Peron provides the clearest and most striking illustration of
the downgrading of organisational intermediaries and of direct
rapport between a populist leader and his followers. As already
mentioned in Chapter 1, his control of the Argentinian workers
was to a large extent achieved over the heads of the already
established communist and socialist-controlled trade-union organisations. The latter were understandably apprehensive of the
overwhelming power Peron was exercising over the workers,
especially over those newly arrived from the country's interior.
Those trade-union leaders who at the beginning had tried to resist
Peron openly were eliminated or marginalised; and those who,
realising the strength and irreversibility of the Peronist movement,
tried to exercise their influence covertly, very soon had to admit
that their attempts at establishing relatively autonomous workingclass organisations within Peronism were doomed to fail.
To give a specific example: after the events of October 1945
which forced the military junta to free Peron, many hesitant
trade-union leaders, impressed by Peron's working-class support,
decided to help in the creation of the Partido Laborista, in this way
hoping to set up a genuinely independent political organisation
based on the mobilised masses. Once the party was established, it
played a crucial role in bringing about Peron's victory in the
February 1946 elections. In fact the Laboristas, as the main
political force in the Per6nist electoral alliance, obtained 65 per
cent of the Per6nist vote and so managed to control the majority of
seats in congress. Yet when subsequently those close to Peron
created the Partido Unico in a determined bid to assimilate/
subjugate the Labour Party, few of the Laboristas managed to
muster any effective opposition. Although the cadres of the
Labour Party resisted the decision to merge with the Partido
Unico, their defiance was not very successful, not because they
were coerced into complying (at the early stages of the Peronist
regime repressive mechanisms of control were not yet very prominent), but because they realised that they did not have the support
of their rank and file. Even the most popular of the labour leaders
81
failed to mobilise workers' support. The working classes, and

especially the 'new' working classes recently arrived from the
countryside, were in direct rapport with their charismatic leader,
and this direct link emasculated the strength and autonomy of
pre-existing working-class organisations. 14
If this almost 'pure' type of populist relations of domination is
contrasted with an equally extreme clientelist case, the organisational differences between the two become perfectly clear. One
need only think of the well-documented cases of 'captive voters',
where local clients owe personal allegiance to the local patron
rather than to the national party leadership, and to such an extent
that the patron can even change parties without losing his political
clientele. Although this type of extreme political localism is
generally more prevalent at the stage of oligarchic politics, the
local patrons retain a great deal of their autonomy and room for
manoeuvre even in post-oligarchic, more centralised clientelistic
parties. 15
C. The persistence and dominance of clientelistic elements in the
post-oligarchic bourgeois parties of inter-war Greece has already
been noted in Chapter 1. With reference specifically to Venizelos's
Liberal Party, attempts at building up a more formal/bureaucratic
organisational structure were systematically boycotted at both
local and national level. I shall now further elaborate this point
which indicates clearly the different type of authority relationship
between leaders and cadres in the clientelistic and populist cases
respectively.
Venizelos, from the start of his political career in mainland
Greece, had the ambition to go beyond the existing parties of
notables and to create his country's first modern bourgeois mass
party. This intention emerged quite clearly in the elections of
November 1910 when many Liberal candidates were nominated in
non-clientelistic fashion by various interest groups. 16 This practice
- which could eventually have led to a party organisation closely
linked with, or even based on, organised collective interests- was
soon abandoned, and the nomination of candidates reverted to
informal clientelistic negotiations between the leader and powerful
clientelistic factions. After this first abortive attempt Venizelos
tried several more times to create modern party structures but,
faced with the adamant opposition of strong clientelistic elements,
82
had to give up his attempts at party reforms. As George Mavrogordatos puts it:
Structural reform could, therefore, only be imposed on the
clientelistic component of the party at the risk of exacerbating
existing cleavages and upsetting the fairly close balance of
forces with Antivenizelism, if dissatisfied local factions were not
to walk out en masse. Such a risk was evidently unacceptable to
Venizelos, whose fundamental pragmatism had led him to
welcome several supposedly reform-minded local bosses in 1910
and later. 17
Now the point is that the threat of dissatisfied local factions
'walking out en masse' was a very effective one in the case of
Venizelos because, unlike Peron, he could not so easily bypass
their authority and appeal directly to the people. In other words,
local clientelistic factions occupied a much more powerful and
autonomous position within Venizelos's party than trade-union
leaders and other key cadres occupied in Peron's Partido Unico.
This difference was not, I think, due to the fact that Peron
possessed greater charisma than Venizelos. It was due rather to
the different objective, structural conditions pertaining in the two
countries - structural conditions which in the case of Greece
allowed charismatic leaders less room for manoeuvre.
Finally, it should be mentioned that the dominant power of
clientelistic factions within Greece's major bourgeois parties was
not limited to the inter-war period. So in post-war Greece and until
the establishment of the military regime in 1967 the two big
bourgeois parties (the Centre Union as heir to the inter-war
Venizelist party, and the conservative Radical Union party)
retained their marked clientelistic features. The core of their
organisational structures consisted of extensive clientelistic networks controlled by influential political figures who were neither
totally dependent on the party leader's goodwill, nor constrained
by formal party structures and regulations. As K. Legg wrote in
the 1960s:
For the most part, Greek party structures still do not reach the
local level. If political organisations are found here, they are
usually the clientage groups of local notables or individual
83
deputies. A voter's electoral decision is largely based on

personal ties to a local leader, and he may not even be aware of
the parliamentary faction that his vote will ultimately support.
The membership of all major party groupings is calculated by
the number of voters who supported each at the last election.
There are very few actual members of the national political
parties. For example, in the spring of 1965 the total printing of
the sporadic ERE party bulletin was 800 copies, including those
that went to the 100 ERE deputies in parliament. 18
Of course, neither the right-wing ERE (Radical Union) nor the
Centre Union parties were entirely based on clientelistic networks.
Another important component of their structures was provided by
what Legg calls the 'personal-fragile' aspects of party organisation.
By this he means deputies who, for one reason or another, 19 had
no personal clienteles and owed their nomination as candidates
and their subsequent election to the party leader. Contrary to
strictly populist organisations, in Greece this 'personal-fragile'
aspect of party organisation was not very strong in either the
inter-war or the post-war (pre-junta) period.
2.4 From clientelistic to populistic politics
As the foregoing clearly indicates, political organisations in the

post-oligarchic polity of the parliamentary semi-periphery portray
a mixture of clientelistic, populistic, as well as bureaucratic/
non-personalistic elements; and very often the articulation of such
elements is, in concrete cases, extremely complex, especially so
during periods of transition when new organisational patterns
emerge and coexist uneasily with older forms. M.L. Connif's study
of urban politics in Brazil20 provides an excellent detailed example
of the transition from traditionaVclientelistic to populist politics in
Rio de Janeiro, and of the intricate mixture of organisational
forms that characterised it.
A. The traditional, oligarchic system of politics in Rio was
primarily based on a complex hierarchy of patrons who dispensed
a variety of favours to their clients in exchange for these clients'
votes. The key figure in the clientelistic hierarchy was the chefe
84
politico, the urban equivalent of the rural coroneis. Chefes were

usually professional people (doctors, lawyers) whose occupation
brought them into daily contact with a large number of people and
who were, therefore, able to dispense free or cheap services to the
needy and so gained popular support. Above the chefes were the
figuriios, the grand personages of patronage politics who, with a
considerable number of chefes under their control, were part of
the top political elite with access to the presidential circle. Below the
chefe was the cabo eleitoral, or precinct captain, whose relation to
the chefe was similar to the chefe's vis-a-vis the figuriio. The cabo
was in charge of the day-to-day management, the nitty-gritty of
grass-roots politics.
The cabo dealt in small favours - a paved street, or protection

against vandalism - but his importance was enhanced by his
constant availability to the local populace. He was usually a
long-time resident and a community leader, acquainted with all
on a first-name basis. As elections approached, he carefully
reviewed his constituents' registrations and provided new ones
when necessary. Voter registration was complicated, and if the
cabo himself was not a public notary, he was the close friend of
one. Occasionally a prospective voter would be illiterate and
would have to be brought to sign his name to qualify. When the
campaign heated up, the cabo established a political centre and
distributed printed material, including his chefe's ballots ...
Finally, on election day, the cabo was responsible for getting
out the voters and verifying that their ballots were cast
properly.Z 1
Lastly, separate from the figurao-chefe-cabo triad there were a few
'outsider' politicians who consistently opposed the 'corrupt' clientelistic system and proposed a variety of reforms, not only of a
political but also of a socio-economic nature (educational reforms,
advanced social legislation, and so on). In a way these politicians
were the harbingers of the post-1930 reform movements. 22
This traditional system of clientelistic politics began to weaken
in the 1920s and early 1930s as a variety of long-term processes (for
example, rapid urbanisation) and more specific political developments (for example, the tenente movement) spectacularly in-
85
creased the number of active voters - a broadening of political

participation that was making the continued operation of the
traditional clientelistic system of vote management particularly
difficult. As the number of voters increased, and as the post-1930
electoral reforms (such as the introduction of the secret ballot)
were impeding the face-to-face operations of the chefes and cabos,
other forms of political organisation emerged and became quite
prominent, such as interest-group voting. So trade unions, large
business organisations and other associations would try to exert
direct influence on the way their members voted, thus bypassing
the traditional clientelistic networks. 23
It was in this new context that Dr Pedro Ernesto, Mayor of Rio
from 1932 to 1936 and a tenente supporter, managed to build up
the first large-scale populist party (the PADF) in Brazilian politics,
doing on the level of urban politics what Vargas was to do a decade
later on the national level. In fact Ernesto managed to attract
votes from the middle classes, from the newly-formed working
class, and from the marginal populace of the favelas by initiating a
new style of politics based on charismatic leadership and the
extensive use of the media, appealing directly to the electorate as
the medico bondoso, the humanitarian father/doctor exclusively
concerned with his people's needs.
In creating his party organisation, Pedro Ernesto had of course
to make use of the traditional chefes/cabos as well as of the various
emerging interest-group associations. However, what linked all
these disparate elements together, and what was responsible for
the mass appeal of the new party, was the leader's new style and
the skilful presentation of his charisma by the party managers, a
new breed of political operators quite distinct from the traditional
chefes. As Connif points out, within the new organisation it was
increasingly the party managers (with their skills of speech-writing
and media management) rather than the traditional chefes who
played the crucial co-ordination roles and provided the links
between the leader and the public. In contrast to the chefes, these
party managers did not have electoral followers of their own.
Their power and authority derived from the party and, ultimately,
from the leader. Connif writes:
The critical difference was that whereas chefes could always
return to their constituencies for renewed support and were
86

hence somewhat independent of higher authority, the managers
owed their political existence to the party. Their only outside
appeal was to national leaders. 24
This again illustrates the fundamental difference between clientelistic and populistic organisations. The cadres within the clientelistic structure of authority have an autonomy that does not exist in
populism. The difference between the party manager and the
chefe is like the difference between the foreman and subcontractor of the early factory system. The first is an employee
who derives his authority from his boss, whereas the second is a
more or less independent operator who strikes a deal with a larger
entrepreneur from a position of relative autonomy.
A possible objection to attempts at clearly differentiating between populistic and clientelistic types of organisation is that the
former can be as particularistic in their approach to voters as the
latter: in the parliamentary context both of them are in the
business of exchanging favours for votes. Nevertheless, the mode
of exchange is different. In the purely clientelistic case, at grassroots level, the exchange takes place between a 'political subcontractor' and the voter. In the typical populist case, it is not only
that the sub-contractor has been replaced by a party employee, but
also that this employee (unlike cadres in relatively autonomous
party organisations) depends for his position on the leader's
goodwill.
B. The distinction I am trying to draw bears some resemblance to
Weber's between feudalism and patrimonialism as two ideal
typical sub-categories of traditional domination. According to
Weber, a central difference between feudal and patrimonial
administrative structures is this: within the latter, office holders
not owning or effectively controlling the means of administration
are much more dependent on the ruler's arbitrary will than within
feudalism. 25 Now if clientelism and populism are viewed as
relations of domination, typical not of traditional, pre-industrial
societies but of parliamentary societies with late capitalist development, then interesting analogies can be established between feudal
lords and, say, provincial notables with strong electoral fiefs.
Furthermore, the relationship between a powerful populist leader
like Peron and his cadres can also be seen as an essentially
87
patrimonial relationship between a ruler (unrestricted by either

bureaucratic rules or feudal autonomies) and his pliable officials.
In the same way that 'feudalising' tendencies emerge in declining
patrimonial systems of domination, so in populistic organisations
or politics 'clientelisation' processes occur, especially when the
populist leader and his central organisation fail to subjugate or
permanently replace local bosses with their own political cadres.
The difference between traditional and modern patrimonialism is,
of course, that whereas the traditional ruler derived his legitimacy
from ascriptive/traditional sources, the authority of the modern
patrimonial/populist leader is based on his charismatic ability to
appeal to and mobilise the masses against a political, economic or
cultural establishment.
This line of thought brings into focus interesting parallels
between the ideological and organisational aspects of populism.
For instance, it has often been pointed out in the relevant
literature that populist ideologies, emerging in a situation of
backwardness and of attempts to catch up with the already
modernised West, are Janus-faced: they look towards both the
past and the future, they contain 'traditional' as well as 'modern'
themes. 26 So populist ideologies are always concerned with finding
a course of modernisation that is not a blind imitation of the
western capitalist trajectory, a course that would avoid the disruptions and inhumanities of the Industrial Revolution and pay more
attention to indigenous/traditional values and forms of social
organisation. (For instance there is a strong 'indigenist' theme in
both the East European peasant populist parties and in the Latin
American urban populist movements.)
If, as I am arguing, populism involves a specific type of
authority relationship between leaders and led, such a mixture of
traditional and modern elements exists also on the organisational
level. For instance, it has frequently been emphasised in the
literature on Latin American populism that the type of relationship between populist leaders and their urban followers is a
fairly straightforward transposition of traditional, paternalistic
authority patterns from the countryside to the city. The father
image on which the leadership style of Peron or Vargas was based,
for example, had particular appeal to all those rural emigrants who
had recently come to the urban centres. It came quite naturally to
these people, used to the appeal of the traditional patron-
88
protector, to vest their trust in and support a populist leader. 27

In the urban context, these traditional elements were of course
articulated with more modern ones. It was not only that efforts
were made to bring people into the political arena rather than to
keep them out (a clear indication of 'modernity'), but also that the
leader's legitimation was no longer based on ascriptive criteria but
on performance. It can therefore be argued that populism is
Janus-faced not only in terms of ideological content, but also in
terms of organisational form/structure.
2.5 The debate on the nature of populism
A. I believe that the above considerations are applicable in the
context of the more theoretical debates on the general nature of
populism. The relevant literature presents three main positions on
the issue of the general definition or/and theory of populism.
There is, first, the totally negative position which argues that
populism refers to such a wide variety of phenomena (for example,
the Russian narodnik movement, the late nineteenth-century
North American agrarian movements, the East European peasant
parties, third-world populist regimes, Bonapartist types of dictatorships, and so on) that it is quite impossible to find any features
that they all hold in common which could be called populist, and
which would justify the scientific usage of the term. In consequence, as argued by this school of thought, the word populism
should be dropped from the social-sciences vocabulary.
The second position opts for a very narrow definition of
populism, restricting its scope to one of the major genera of
populist phenomena. For instance, there have been attempts to
identify populism exclusively with movements and ideologies
focussed on the peasantry or, more generally, on rural cultivators;
and conversely, there have been attempts to see populism as an
urban-based, multi-class movement of the type encountered in
Latin America. 28
Finally, there have been efforts to construct more inclusive
definitions/theories to account for most of the movements conventionally labelled as populist. So Ernesto Laclau argues that in the
same way that class cleavages and contradictions give rise to class
'interpellations' (that is, ideologies which call on, or interpellate,
89
people as class subjects), on the political level, the objective

cleavage between people and the power bloc gives rise to popular
interpellations. For Laclau, populism is the articulation of popular
interpellations/themes in opposition to the power bloc. This
antagonistic articulation of popular themes may be realised by the
exploited classes, in which case the populism is of a left-wing,
progressive type; or it may be realised by peripheral factions of the
dominant classes in their attempts to restructure the power bloc to
their advantage, in which case the populism is reactionary (for
example, fascist). 29
Margaret Canovan's more recent and less theoretical formulation identifies as many as seven different types of populism:
revolutionary intellectual populism, peasants' populism, farmers'
radicalism, populist dictatorship, populist democracy, reactionary
populism, and politicians' populism. After examining each separate type she concludes that there are two elements that are shared
by all types of populism: 'All forms of populism without exception
involve some kind of exaltation of and appeal to the 'people' and
all are in one sense or another anti-elitist'. 30 Despite their very
different theoretical approaches, both Laclau and Canovan seem
to agree on the core elements of populism: an appeal to the people
(Laclau's popular interpellations), and anti-elitism (Laclau's articulation of popular interpellation against the power bloc).
B. The difficulty with both of these definitions of populism is that
they are so general that they apply to almost any modern, radical
political movement. For instance, the appeal-to-the-people and
the anti-elitist orientation are major elements in the ideologies of
the non-governing communist parties all over the world. They
were equally found in the rising late nineteenth-century western
European socialist and social-democratic parties and, in more
attenuated form, in a variety of clientelistic parties in their
anti-oligarchic, reformist period. All of these cases have involved,
to use Laclau's terminology, a clear attempt at articulating popular
interpellations/themes in a way that is antagonistic to the power
bloc; or, in Canovan's terms, the attempt has been to appeal to
and mobilise the people against a political and/or economic
establishment.
I would suggest that the endeavour to pinpoint the decisive
difference between populism and the above-mentioned parties/
90
movements cannot be restricted to the level of ideology; one must

also take into account the organisational structures of such movements, and particularly the type of authority relationship existing
between leaders, cadres and followers. As I have pointed out
already, populist movements, in their logically pure form, portray
a direct and unmediated rapport between the leadership and the
rank and file, and it is this emphasis which, on the organisational
level, weakens intermediary authority levels. It is not that, in a
purely populist situation, organisational structures or administrative intermediaries are non-existent, but rather that the legitimation and power of the administrative intermediaries is rooted
neither in an autonomous local base (as in the typically clientelistic
party), nor in relatively autonomous bureaucratic structures that
cannot be easily brushed aside by a leader's direct appeal to the
masses. In other words, organisational structures do exist in
populist movements, just as there are charismatic personalities in
non-populist parties. The chief difference lies in the fact that the
populist leader, in the case of a clash with his organisational
cadres, can easily push them aside; whereas the non-populist
leader, whether charismatic or otherwise, cannot.
In conclusion, it seems to me that for an effective definition of
populism as a type of social movement one must incorporate, in
addition to the ideological elements mentioned by Laclau and
Canovan, the type of plebiscitarian relationship between leader
and led - a relationship which has important organisational
consequences. In other words, what needs to be realised is that the
'people' and 'anti-elitist' themes of Laclau's and Canavan's formulations are of a specific kind: that on the organisational level
they conduce to a type of authority structure that is quite distinct
from that of other radically-oriented popular movements and
parties. Any failure to take into serious account the organisational/
authority aspects of populism not only results in the populist
phenomenon appearing as a set of disembodied ideological
themes, 31 it also tends to dilute the specificity of the concept.
The solution I suggest in these pages avoids the shortcomings of
both the very restrictive and the very general definitions of
populism. It is wide enough to encompass, for instance, both rural
and urban populism; and on the other hand it is specific enough to
make a clear distinction between populist and non-populist radical
movements. It is true of course that, although there is an obvious
91
logical link between the ideological and organisational aspects of

populism, these two dimensions may not be present to the same
degree. There are movements with populist ideological themes,
the authority structures of which do not reflect strong plebiscitarian elements, and there are also plebiscitarian movements with
non-populist ideologies. Moreover, a movement which in its early
stages portrays all the relevant populist characteristics may be
transformed during later stages and lose one or both of its populist
dimensions. So, for instance, (as mentioned in Chapter 1) Chile's
Socialist Party, its denominational label notwithstanding, had very
strong populist features under Grove's leadership in the 1930s (on
both the ideological and organisational levels); but once it had
joined the popular-front coalitions in the 1940s these features were
greatly attenuated. 32
None of the above difficulties is specific to the concept of
populism, of course. All general concepts in the non-positivistic
social sciences (for example, bureaucracy, nationalism, communism, and so on) encounter similar complications when it comes to
applying them to concrete fields of study. Such complications
become manageable, however, once the ideal-typical character of
these concepts is understood, and once it is accepted that there
need not be total coincidence between lay/conventional and
scientific definitions of a phenomenon. The Co-operative Commonwealth Federation in Saskatchewan for example, the strong
internal democratic procedures of which were analysed by S.M.
Lipset, 33 had a populist ideology but a non-populist organisational
structure; on the other hand, Stamboliiski's Agrarian Peasant
Union or Peron's Partido Unico had a populist ideology and a
clearly plebiscitarian organisational structure. Following the definition proposed here, the Bulgarian and Argentinian examples
are clearly populist movements, whereas I would call the Canadian
one a farmers' radical movement with populist ideological elements.
In terminating this theoretical discussion on the general concept
of populism I would like to emphasise once more that in this book
I view populism in relation to processes leading to the broadening
of political participation in a number of countries characterised by
early parliamentarism and late industrialisation. More precisely I
view populism as a mode of vertical inclusion/incorporation of the
lower classes into the political arena during the transition from
92
oligarchic to post-oligarchic parliamentary politics. This does not

mean, of course, that this is the only way of studying populist
movements. There are, for instance, forms of populism which are
not related to the oligarchic/post-oligarchic transition, either because they occur at much earlier or later periods (for example,
Rosa's populism in nineteenth century Argentina or Andreas
Papandreou's socialist populism in present day Greece), or because they take place outside parliamentary contexts altogether.
I would argue, however, that what I have said here about the
organisational dimensions of populism applies to all types of
populist phenomena irrespective of whether or not they are linked
with the oligarchic/post-oligarchic transition. For instance, even
when one looks at what Canavan calls populist dictatorships, one
can identify, in addition to popular ideological themes, a leadership style leading to systematic attempts at by-passing intermediary groups and at undermining autonomously constituted administrative/organisational structures between rulers and ruled- groups
and structures which may enjoy considerable autonomy in nonpopulist dictatorships. Needless to say, a major task for the further
theorisation of populism is to go beyond general formulations and
to construct a logically congruent and empirically adequate typology, as well as sub-theories focussing on particular types of populist
phenomena. Although this paper has not addressed this problem,
it suggests that a useful way of going about building such subtheories is by paying attention
(a) to forms of organisation which, although not populist, are
closely linked with populist ones;
(b) to long-term processes of socio-economic change and the
way in which they are interrelated with the rise and fall of
populist movements in a limited number of comparable
cases.
C. Finally, I would like to show briefly that the distinction
between clientelistic and populistic modes of political inclusion
developed in this section is relevant not only for the general theory
of populism but also for that of clientelism. Given the scope of this
chapter I shall neither review nor systematically criticise the
complex debates on the concept of clientelism. I would simply like
to draw attention to the fact that clientelism, just like populism, is
most often defined in such a context-less and universal manner
93
that, as a concept, it loses much of its heuristic value and utility.

The definition of clientelism usually focusses on the dyadic relationship between a patron and a client - a relationship which is
typically particularistic/personalistic and which is based on the
exchange of goods and services between two parties of unequal
status, power, and wealth. In the specific case of political clientelism in a parliamentary context, the client offers his political
support/vote in exchange for various goods or services. 34
Now in the same way that the theory of populism, when it
emphasises exclusively ideological elements, is incapable of distinguishing populistic from non-populistic radically oriented movements or parties, so the typical definition of political clientelism in
its narrow focus on the patron-client relationship is not capable of
drawing a clear analytical line between clientelistic and nonclientelistic parties involved in particularistic exchanges with their
supporters.
If, however, the focus of analysis is shifted to the macro-level,
and the patron-client relationship viewed as part of the broader
system of domination in which it is embedded, then it becomes
quite clear that there is another crucial dimension that must be
taken into account when defining a political party as clientelistic:
this is the relation of local patrons to the over-all hierarchy linking
rural and urban voters to the national power centres. From such a
perspective clientelism, as mentioned already, implies a certain
autonomy of the local patron vis-a-vis the national party organisation and leadership - an autonomy based on his capacity to act as a
relatively independent political entrepreneur or sub-contractor,
rather than as an interchangeable cog in the party organisation. 35
In the parliamentary semi-periphery this organisational autonomy of the local patron was very high during the oligarchic period.
In the post-oligarchic era local potentates lost part of their
autonomous base as party structures became more centralised and
as the state penetrated deeper into the political periphery. But
even in the post-oligarchic period, as I have argued for the Greek
case, the national leadership of the political parties that retained
marked clientelistic features did not succeed in subjugating local
patrons totally and in replacing them or transforming them into
dependent party cadres. When national party leaders do manage
to do so, that is, when party cadres entirely derive their power
from the party organisation or the person of the leader, then
94
clientelism is giving way to other forms of political organisation.

From the above point of view post-oligarchic clientelistic parties
can be seen as representing a form of political centralisation/
'modernisation' which, in comparison with populistic or integrative modes of inclusion, shows greater continuity with ancien
regime parliamentary politics. It is a form of political development
that brings the middle and lower classes into the national political
arena relatively smoothly, that is, in a way which least disrupts the
ideological discourses and the organisational arrangements of the
oligarchic period.
The theoretical clarification and elaboration of the concepts of
incorporation, clientelism and populism given here will help in the
examination in the following chapter of some basic features of the
structures of domination in the semi-periphery, and of how these
features relate to the rise of military dictatorial regimes in Argentina, Greece and Chile.
PART II
Chapter 3
Routes to Military Dictatorship:
A Comparative Essay on Argentina,
Chile and Greece
In chapter 1 the focus was on the fact that the state, city and
market in the parliamentary semi-periphery had expanded before
industrial capitalism could experience any large-scale development, and it was argued that thi~ sequence had a profound impact
on the formation of post-oligarchic political institutions. In the
present chapter I shall try to show systematically and in some
detail how the above is relevant to the emergence of military
dictatorial regimes in three specific countries of the parliamentary
semi-periphery: Argentina (1966), Greece (1967), and Chile
(1973). For the sake of convenience I shall, when referring to all
three of these societies, use the abbreviation ACG - Argentina,
Chile, Greece.
Since my main concern is with the rise of military regimes, and
believing that the roots of such regimes can be traced back to
pre-war developments, I shall begin by discussing the role of the
military during the transition from oligarchic to broader forms of
political participation.
3.1 The military in the early post-oligarchic era
As discussed earlier, the demise of oligarchic rule in the parliamentary semi-periphery occurred in a predominantly preindustrial context, when trade unions and working-class parties did
not play a central role in shaping post-oligarchic political structures. In the absence of autonomously constituted working-class
organisations, and in the more general context of a dependent civil
97
98
society, it is not surprising that the army, already highly conscious

of its corporative interests, was drawn into the centre of the
political stage. In fact, the process of army professionalisation,
which both in the Balkans and in southern Latin America began
towards the end of the nineteenth century and in the early
twentieth under the guidance of western European military missions, meant the creation of military academies and the adoption
of more bureaucratic, universalistic criteria for recruitment, training, promotion, and so on. 1
Given the over-all social context at the time, this professionalisation did not bring about the insulation of the military from
active politics. One of the results of the army's modernisation in
the parliamentary semi-periphery, combined as it was with quantitative growth and the officers being recruited increasingly from
non-oligarchic strata, was that it weakened the nineteenth-century
fusion of civilian and military elites that had been based on a
shared aristocratic background and/or orientation. This in turn
meant that while officers in the nineteenth century had tended to
intervene in politics as individuals (that is, without strong feelings
of corporate identity), during the first few decades of the twentieth
century they did so as a relatively cohesive interest group with
specific professional demands and with a predominantly middleclass, anti-oligarchic outlook. 2 In consequence, (as already mentioned in chapter 1),3 the military played a crucial role, in Greece
and Chile directly, in Argentina indirectly, in breaking up oligarchic parliamentarism.
It must be stressed, however, that the army's central position
within the post-oligarchic polity in the ACG countries should not
be seen as the result of a one-way military penetration, that is, not
as the result of a unified military force pushing aside passive
civilian groups who may have resented, but were unable to
prevent, military involvement. It was not simply that the army
intervened in civilian affairs; civilians were constantly interfering
with and undermining military non-involvement, both by striving
to get officers loyal to them appointed to top military posts, as well
as by using their military clients or patrons as a weapon against
their civilian adversaries. 4 In such a situation it is not to be
wondered at that army coups were often based on military-civilian
coalitions. 5
Routes to Military Dictatorship
99
Such a marked politicisation of the ACG military meant, on the

one hand, that these armies acquired powers and even rights of
intervention in, and manipulation of, the political process not
possessed by modern western European armies; and on the other
hand, the price they paid for this privileged monitoring position is
that military institutions and organisations became more open to
an permeable by civilian political forces and cleavages. However,
military involvement in civilian affairs being a two-way process, it
is much easier to overcome fragmentation and to establish at least
a modicum of institutional unity within the armed forces (as
unitary, hierarchically ordered formal organisations) than within a
highly fragmented civil society. If furthermore it is remembered
that it is the military who control the major means of coercion,
their considerable weight within the polity becomes perfectly
understandable.
3.3.1 The military's role during the transition
Since the army's post-oligarchic power position is crucial for

understanding the imposition of post-war dictatorships, I would
like to take a closer look at military developments in ACG during
the transition to post-oligarchic politics.
The example of Argentina demonstrates the basic pattern of
army dominance in the transition from oligarchic to mass politics
most clearly. Given the massive immigration from Europe and the
precocious 'modernisation' of Argentinian society, radicalism became a growing political force and anti-oligarchic ideas started
permeating the armed forces as early as the end of the nineteenth
century. In consequence a considerable number of officers of all
grades participated in the radically-inspired abortive coups of 1893
and 1905. Moreover, the fear of a new military intervention in
favour of Yrigoyen's radical demands was one of the reasons
which pushed Saenz Pefia into introducing his famous 1912
electoral reform law which, at least temporarily, contributed to the
broadening of political participation and the rise of the Radical
Party to power in 1916. 6
While it is true that 14 years later the military intervened more
overtly and directly in order to oust Yrigoyen and reverse the
A.
100
process of opening up the political system, it must be emphasised

that the 1930 restoration of oligarchic rule was brought about by
the clever manoeuvering of small army cliques which did not enjoy
wide support within the armed forces; the majority of officers were
either indifferent or pro-Radical. 7 However, once the 1930 'Liberating Revolution' had removed Y rigoyen from power, the officers
as a whole had to choose between General Uriburu's and Justo's
projects, the former favouring the establishment of a corporatist
regime, and the latter a return to restrictive parliamentarism.
General Justo eventually prevailed, and the army became the
main guardian of a highly fraudulent parliamentary system. 8
Although this system did not bring back Argentina's ancien regime
structures of domination, it effectively restricted broad political
participation by the large-scale and systematic use of illegal and
quasi-legal means of political exclusion (for example, electoral
manipulation, and so on).
The crucial point here is that ever since then the armed forces
have been the decisive force and main regulator in Argentina's
political life. Before 1930 it was the civilian politicians who called
the tune, even when they were seeking the support of the military;
since 1930 the situation has been reversed, with the civilians
having become the junior partners in civilian-military ventures. 9
As will be argued more extensively below, this is quite understandable in view of the failure of the conservative civilian forces, both
before and after 1930, to acquire a broad popular base. Given this
failure, their only chance of maintaining political power was to rely
on the army's role as the ultimate guarantor of a restrictive system
of rule based on institutionalised fraud and other incorporative
mechanisms of domination. 10 The weaker the oligarchic governments, the more generous they were to the army and the more
willing to give in to its demands. 11
With the advent of Per6nism a serious attempt was made to curb
the armed forces' dominant position within the polity. During the
upward phase of Peron's rule (1946-51), the populist leader tried
to apply the same carrot-and-stick strategy to the army that was
working so well with the trade-union movement: on the one hand
he granted the military all sorts of favours (huge pay raises,
appointments of officers to top civilian posts, and so on), and on
the other he tried to undermine its power by encouraging the
rivalries between various services, by supporting the demands of
101
non-commissioned officers against strong opposition from their

superiors, and so on. 12 When after 1951 the regime began to be
confronted by growing economic difficulties, the army started to
fret under Peron's tutelage while Peron at the same time was
intensifying his attempts at incorporation. For instance, after
General Menendez's abortive revolt in 1951 Peron not only purged
a very large number of officers, he also tried to make some crucial
institutional changes: he removed the coast guard from army
supervision and placed it under the control of the Ministry of the
Interior; he also increased the active members of the police force
so that it ended up having a larger number of men than the army.
After a second unsuccessful military revolt in 1952 Eva Peron
armed the pro-Peron unionised workers, and a year later Peron
initiated a sweeping campaign for thoroughly indoctrinating all
officers and conscripts with 'justicialism'. However, Peron's desperate attempts at the Peronisation of the armed forces failed.
The military never did become a passive tool of justicialist rule,
and in 1955 it reaffirmed its dominant position within the state by
ousting Peron and making another attempt at a closely monitored
experiment in 'guided democracy' (see below).
B. In Chile, army dominance was not as acute as in Argentina,
although the country never had the type of civil-military relationship that characterises western European polities. The first
important point about military development in Chile is that, after
the immediate post-independence struggles and during most of the
nineteenth century, the Chilean military were under effective
civilian control. 13 This nineteenth-century civilian supremacy over
the military began to weaken with the professionalisation of the
army and the various other processes which, during the turn of the
century, eroded the rule of Chile's oligarchy.
As early as 1907, army officers in the Santiago garrison created
the Liga Militar as a pressure group for the promotion of the
army's and the officers' collective interests. 14 Following the familiar pattern common to all three ACG countries at the end of their
oligarchic phase, junior officers started having strong misgivings
about not only the army's condition but also about the over-all
social and political system - denouncing the corruption of politicians, the malfunctioning of parliamentary institutions, growing
social injustices, and so on. As these complaints grew stronger and
102
found public expression, several conspiracies were hatched (as in

1912 and 1919) and secret officers' organisations sprang up (the
Army Society for Regeneration, and the Junta Militar). 15 Nevertheless, it was only after the anti-oligarchic and populist Alessandri had become president (1920), and after his reformist projects
had been thwarted by congress, that the army intervened directly
to break parliamentary obstructionism.
After numerous changes in the top political leadership monitored by a military junta, Colonel Ibanez was elected president in
July 1927, 'inaugurating the first authoritarian problem-oriented
military regime in modern Latin American history'. 16 Subsequently, the fall of Ibanez in July 1931 plunged the country into an
eighteen-month turbulent period which ended with Colonel
Grove's short-lived Socialist Republic (established by military
coup in 1932 and overthrown by another coup in October of the
same year). With the return of Alessandri to the presidency in
December 1932 the traumatic years of military coups and countercoups were over, 17 and Chile entered on four decades of postoligarchic parliamentary stability (1932-73).
What was even more decisive than the civil-war spectre for
re-establishing civilian control over the military after 1932 was the
existence of a (by Latin American and Balkan standards) strong
party system, capable of assimilating new groups into the political
arena without disrupting the multi-party parliamentary system,
that is, without creating conditions inducive to army
intervention. 18 Contrary to the situation in Argentina, for instance
- where the conservative classes, unable or unwilling to build a
popular base, had no other way of coping with broadening political
participation than to call in the military for help - in Chile their
counterparts opted for a strategy of electoral competition and of
drawing part of the newly activated masses into their camp
(see 3.1.2 below). 19
To put the above into a more balanced perspective, it should be
stressed, however, that if the Chilean military's position within the
state was not, in the post-1932 period, as dominant as in Argentina, neither was civilian supremacy as strong as in Anglo-Saxon
parliamentary polities. The widely accepted view of the post-1932
Chilean army as a model of democratic ethos and of professional
neutrality vis-a-vis politics is largely a myth. 20 As Jorge Nef points
out:
103
The first fact which puzzles students of the Chilean military,

particularly historians, is that its attempts to intervene in
politics have been more frequent than many are willing to
admit. Leaving aside the period of overt participation between
1924 and 1932, almost all the governments between 1932 and
1970 faced abortive military uprisings. 21
During the forty-year period of relatively stable parliamentary
rule, the military frequently attempted to influence and shape
political events behind the scenes. For example, when in the 1938
presidential elections the Popular Front candidate Pedro Aguirre
Cerda won by a very small margin, his opponent Gustavo Ross
Santa Maria, as well as Alessandri as the outgoing president,
appealed to the military to cancel the election results; Aguirre
Cerda himself also went to the military in the hopes of dissuading
them from doing so. After the Popular Front had called on its
supporters to mobilise themselves so as to stop a military intervention, General Arriagada finally opted for the constitutional solution, Aguirre Cerda became president, and legal continuity was
assured. What is important from our point of view is, of course,
that six years after Alessandri had tried to curb the power of the
military and to establish civilian control, he himself appealed to the
armed forces high command in an endeavour to prevent Aguirre
Cerda from becoming president. 22
Moreover, the army's attempts at intervening in politics were
not limited to this form of action. Before as well as after 1938 there
were several conspiracies to overthrow the parliamentary order. 23
Despite the fact that all of them were abortive, their frequency
(for instance, until the start of Ibanez's second presidency in 1952
there were conspiracies in 1938, 1939, 1946, 1948 and 1951), and
the usual leniency with which most of the conspirators were
treated, indicate the continuing strong appeal that authoritarian
solutions had in certain military quarters, and the unwillingness or
inability of civilian politicians to effectively curb such authoritarian
tendencies.
In fact, with Ibanez's return to the presidency the military's
authoritarian tendencies were significantly reinforced. As soon as
Ibanez had won the 1952 presidential election a military lodge was
created, (the PUMA), for the purpose of assuring that their hero's
electoral victory would not be thwarted by congressional opposi-
104
tion. Three years later a similar group of officers (La Linea Recta)
urged the president to purge the army of all anti-lbaiiista officers.
La Linea Recta pledged to support any anti-constitutional measures the president might wish to introduce in order to establish
order, and worked out a detailed blueprint for a new corporatist
system of rule24
All of the above indicates quite clearly that, beneath the surface
of constitutional continuity and civilian hegemony of the post-1932
era, the Chilean military continued to exercise considerable political influence. Although their political role was not as accentuated
as that of their Argentinian and Greek counterparts, they were
during those forty years of parliamentary stability far from establishing the type of power relationshp vis-a-vis the civilian authorities that is found in post-war western parliamentary democracies.
C. In Greece after the 1909 military intervention and the rise of
Venizelos to power, the armed forces, reorganised and imbued
with new morale, fought victoriously in the Balkan Wars (191214) and in the First World War. Due to these wars the army's size
increased spectacularly?5 At the same time, as more candidates
from the middle classes were accepted into the Military Academy
and as, due to the long war years, promotion became easier, the
officer corps lost its upper-class complexion and acquired a more
middle-class orientation, 26 to emerge as a pressure group anxious
to promote the professional interests of its members.
This emergence of the military as a distinct interest group - in a
context of a weak and heteronomous civil society during a period
of political transition - made army intervention in politics a
quasi-certainty. Of particular importance in this respect was the
military coup after Greece's defeat by Turkey in 1922. This crucial
intervention led to the execution of members of the pro-royal
civilian leadership who were deemed responsible for the military
fiasco. It also led to the abolition of the monarchy and the
establishment of an eleven-year republic. From 1922 onwards,
civilian control over the military was greatly weakened as army
officers kept interfering in politics directly by replacing one set of
civilian leaders with another, or indirectly through threats and
pressures. 27
The precise extent of the military's autonomy or dominance
varied with circumstances, of course. For example, Venizelos's
105
return to active politics in 1927 temporarily re-established some

degree of civilian control. But even allowing for these fluctuations
in army dominance, there can be no question that the Greek
military, especially after 1922, were not only fully involved in
politics, but also constituted a major force in the political arena.
This was also true in the wake of the Metaxas coup of 1937 and the
establishment of a monarchical dictatorship (1937-41).
From 1937 onwards the role of the army began to change in a
way that presaged the post-civil war developments. This change
had a structural basis. Although the issue of the monarchy was still
all-important and directly related to the Metaxas coup, the influx
of Asia Minor refugees, as well as the rapid industrialisation and
urbanisation, seemed for the first time to pose a real threat to the
establishment. 28 The Left managed to recruit a small part of the
refugees and the urban working class and, although the Communist Party received a mere 5.7 per cent of the vote in the 1936
elections, it played a crucial role in parliament as a balancing force
between the two big bourgeois parties. While the chief aim of
Metaxas (and his royal patron) was to boost the monarchy, 29
dealing with the growing social unrest from below came a close
second on the dictatorship's list of priorities. In consequence
Metaxas's purge of all republican officers from the army was
followed by the institution of widespread repression and all the
related techniques - techniques which were to be fully developed
by the 1967 regime.
3.1.2 Variations in civil-military relations: the conservatives'
post-oligarchic electoral strength and army dominance
Having briefly reviewed the army's post-oligarchic role in ACG

politics, I would like to focus here on a dimension of civil-military
relations which I believe to shed some light on the significant
variations in army dominance in the three cases under review.
The degree of military dominance in ACG's politics is inversely
related to the capacity of the oligarchic forces to organise themselves politically and to compete in the post-oligarchic political
arena. As shown above, after the demise of oligarchic parliamentarism the military became most powerful in Argentina, a country
whose conservative political forces experienced a very abrupt and
dramatic collapse of their electoral base after the introduction of
106
the Saenz Peiia electoral reforms. In Chile on the other hand,

where not only was the working-class movement more autonomous but the post-oligarchic conservatives managed to constitute
themselves into a strongly organised political force, the army's
political power and involvement was much more limited. The case
of Greece fits neatly between the two extremes: the decline of her
ancien regime politicians was not as accentuated as in Argentina,
and in consequence the Greek military, at least until the civil war,
did not achieve the kind of overwhelming dominance the Argentinian army enjoyed from 1930 onwards - but neither were the
Greek military's powers of intervention in politics as restricted as
in Chile. As the connection between army dominance and the
vote-winning capacity of the post-oligarchic conservative parties is
crucial for understanding civil-military relations in ACG, it is
worth discussing this subject a little more.
A. In Argentina, before the Saenz Peiia electoral reforms, it was
fairly easy for the landowning/export interests to translate their
enormous economic power into parliamentary political control.
Although the 1853 constitution had introduced universal male
suffrage, it said very little about electoral procedures; these were
left to be specified by congress which, in a series of legislative
measures, fashioned an electoral system that put a damper on the
effective political participation of the lower classes. So for instance
the management of elections was the responsibility of area officials
easily able to manipulate the local electorate; 30 and since voting
was not compulsory and public, it could obviously be limited to a
very small number of people. 31
The nature of the major economic rural activities (cattle-raising
and, later, extensive wheat cultivation), the mode of colonisation
and, above all, the type of land tenure discouraged the massive
influx of foreign immigrants and the creation of highly populated
settlements in the countryside. Given the great difficulties poor
immigrants had in acquiring a piece of land of their own, the
exceptionally harsh life in the pampas, and the notorious lack of
such amenities as schools and hospitals, 32 most immigrants
avoided the countryside or were prepared to work there only
temporarily. 33 It is not surprising therefore that from very early
on, much earlier than in Chile or Greece, the majority of the
population concentrated in a few urban centres.
107
This feature of the Argentinian social structure is highly relevant

for understanding the electoral decline of the country's conservatives in the immediate post-oligarchic period. The countryside
being relatively unpopulated, they did not have the large rural
constituencies at their disposal that, as demonstrated by postoligarchic Chile, are so easily amenable to landowners' clientelistic
controls. This fundamental debility of the Argentinian conservatives was further accentuated by their marked fragmentation. This
was due to the way in which the all-powerful landed interests of
the Buenos Aires area had imposed national unity in 1880 by
creating a centralised federal structure that discriminated against
the economic interests of the landowners of the interior. In view of
this it would have been very remarkable if the landowning classes
had managed to build up an effective national party organisation.
Conservatism has always been an alliance of individual provincial parties. This was true of the Federal, Unitary, National,
Autonomist, National Autonomist and National Democratic
Parties which existed prior to 1955. After that date the conservatives gave up all pretence of being a unified national party
and formed a National Federation of Parties of the Centre
(FNPC); however, not even this organisation was able to
affiliate all the local conservative parties ... At the time of the
1966 coup there were perhaps 20 conservative parties. 34
It was because of this type of political fragmentation and the
conservatives' lack of a populous constituency in the countryside
that they could not follow Disraeli's example of building a massive
popular base in the immediate post-oligarchic period. Their hostility to open electoral competition therefore had to do less with their
alleged authoritarian orientation than with the existing objective
structural conditions which did not favour the large-scale development of a conservative electorate.
This fundamental electoral weakness of the upper classes did
not, of course, mean that they were without political influence.
Their immense economic power and their unchallenged cultural
hegemony enabled them to affect political developments strongly,
if indirectly. It is precisely this combination of electoral weakness
with economic and ideological dominance that goes a long way to
explain why the landowning interests never really accepted the
108
rules of the parliamentary game - or rather why they accepted

them with the important proviso that the leadership in charge of
the game had to be to their liking; whenever the 'wrong' type of
leaders came to the fore, they resorted to their friends in the
military. The latter, given the general tendencies of the army in
the post-oligarchic semi-periphery, quite naturally fell into place
to close the political vacuum at the right of the political spectrum
and, as already mentioned, came to occupy the dominant power
position in the Argentinian political system from 1930 onwards.
B. A brief look at the Chilean case shows that the opening-up of
the political system was not only less abrupt, but also that the more
favourable structural conditions enabled the ancien regime political forces to build up a sound electoral base in the countryside and
to organise themselves into a credible political power in the
post-oligarchic parliamentary arena.
As far as the gradual nature of the opening-up process is
concerned, whereas in Argentina in 1916 (after the introduction of
the Saenz Pena law) the number of voters rose from 9 per cent to
30 per cent of the adult population, in Chile as late as 1949 only 20
per cent had the right to vote, and out of that number only 10.2 per
cent were actually registered in the electoral lists. 35 Moreover, the
shift of the population from the rural to the urban areas was more
gradual in Chile. As early as 1914, Argentinians living in cities of
20 000 inhabitants accounted for 38 per cent of the population, but
only 28 per cent in Chile at about the same time. 36
In more general terms, the Chilean countryside from the
nineteenth century onwards was much more populated than the
Argentinian one. This difference is understandable if one considers how Chile's landowners reacted to the world demand for their
country's agricultural produce during the second half of the
nineteenth century. Pressures for increased output were met by
enlarging the areas under cultivation and by extending and reinforcing the traditional service-tenancy system of labour exploitation (the inquilinaje). In fact, the Chilean hacendado tried to cope
with the growth of external demand both by squeezing more
labour services out of the inquilinos 37 and by increasing their
numbers. He managed the latter by attracting the considerable
floating population of rural day workers with more permanent
service contracts. For instance, although from 1865 to 1930 the
over-all rural population increased only slightly, the number of
109
inquilinos doubled. 38 At the same time the working conditions and

income of the service tenants deteriorated considerably.
(This backward-looking solution to the pressures of increasing
international demand was less successful in Argentina where the
indigenous pool of cheap labour was much smaller. Argentinian
landowners did try to attract labourers from Italy and Spain, but
not only were these immigrants less willing to settle in the
countryside permanently, they also commanded much higher
wages than in Chile). 39
The extension and strengthening of the traditional forms of
labour exploitation in Chile explains to some extent why, in
contrast to their Argentinian counterparts, the Chilean landowning classes were more willing to accept the post-oligarchic openingup of the parliamentary system and to exercise political control
through the electoral mechanisms. In fact, within the highly
self-contained, traditionally oriented hacienda, the economic,
ideological and political control of the hacendado over his inquilinos, peons and empleados was a foregone conclusion. As Bauer
points out:
Land ownership was not just a source of wealth or prestige but

also a political base ... Given their free choice, most rural
workers, or at least the service tenancy, would probably have
supported their patron against an outside candidate. But even if
they did not, the landowner had formidable powers of suggestion. The threat of expulsion or fines was undoubtedly a
sufficient incentive for workers who, after all, saw little point in
making a choice. Although the way rural politics actually
worked has not been fully described, there apparently were
cases in which the rental contracts of rural estates stipulated
that the owners reserved the right to use (disponer) the votes of
inquilinos . ... And if customary authority or coercion were not
adequate to gain the support of rural voters, the landowners,
who especially after the 'autonomous municipality' was established in 1891 dominated local government and the police force,
could simply call upon the local authorities to expel rival
politicians or trouble-makers. 40
This exceptional political strength of the landowner can be
explained not only in terms of the internal social structure of the
hacienda but also in terms of its weak links with the outside world.
110
Unlike other peasant societies, rural Chile did not experience the
large-scale development of independent village smallholders' communities that, in certain conditions, can provide a base for rural
radicalism or even peasant revolution (as in Mexico for example).
Chilean minifundists, often dependent on the hacienda for supplementing their meagre incomes, did not form themselves into a
well-organised political force, and in the traditional minifundia
areas the pattern of electoral behaviour was actually quite similar
to that of areas where large estates predominated. 41
Finally, it must be realised that this tight political control over
rural labour by the big landowners persisted long after the demise
of oligarchic parliamentarism. Although the demise of oligarchic
politics and the subsequent world economic crisis did, of course, to
some extent weaken the hacendado's hold over the peasantry, this
slackening of traditional control mechanisms was a very slow and
long-drawn-out process. 42 Chilean landlords, in fact, managed to
retain the political support of their rural subordinates well into the
1950s and early 1960s. A very important reason for this is that the
constitution of 1925 (and especially the way certain of its clauses
were implemented) overwhelmingly favoured the rural areas.
Rural overrepresentation in congress surviving right to the military
coup of 1973 is shown, for example, by the fact that whereas in the
congressional elections of 1973 the minimum number of votes for
electing a deputy in the Third District of Santiago was 73143, in
Aysen (southern Chile) it was only 3918 votes. 43
An even more important reason for the very gradual electoral
decline of the rural upper classes in Chile was the establishment of
a 'social pact' between landowners and urban-based political
groups, on the basis of which unionisation of the rural labour force
was effectively prevented until the 1960s. 44
All the above circumstances explain why, unlike their Argentinian counterparts, the Chilean Right was not subject to an abrupt
electoral decline in the early post-oligarchic period. Instead, after
the turbulent years of dictatorship and frequent military interventions (1924-32), the conservatives and the liberals, the two major
parties of the oligarchic period, in alliance with each other,
worked out an ideological programme better attuned to the
post-Depression economic realities, and showed considerable flexibility in their dealings with the political forces of the Centre and
the Left. 45 Having in this way secured an electoral base and being
111
highly suspicious of the military's predominantly anti-oligarchic

orientation and of their abortive attempts to build a socialist
republic, the Chilean conservatives opted for electoral competition and for a drastic restriction of the army's involvement in
politics.
C. In Greece the post-oligarchic civil-military relationships occupied a place half-way between those of Argentina and Chile.
Although the rise of Venizelos's Liberal Party after the 1909
military coup significantly reduced the power of the oligarchic
political forces (the paleokommatikoz), the latter's electoral decline was not as abrupt as in Argentina. As I have pointed out in
chapter 1, in Greece the transition to post-oligarchic politics was
very smooth, in the sense that the new entrants to power managed
to consolidate their post-oligarchic power position without major
populist mobilisation in the cities or the countryside. Given the
large-scale persistence of clientelistic politics in the post-oligarchic
period, as well as a relatively slow shift of the labour force from
the rural to the urban areas, 46 the conservatives had ample
opportunities to maintain/transform their patronage networks and
to compete successfully in the post-1929 electoral arena. Of
course, given the agrarian reforms and the irreversible decline of
big landed property, their control in the countryside was not as
straightforward as in inter-war Chile. It was rather that in those
years between the wars liberals and conservatives each had their
own fairly stable electoral constituencies in different regions of
Greece. (As outlined in Chapter 1, the smallholders of the
northern provinces, having benefitted from Venizelos's agrarian
reforms, tended to vote for the Liberal Party, while those in the
Peloponnese were more conservative in outlook.) On the other
hand the Greek conservatives did not have to face the rapid
growth of a rural proletariat working on big estates. Besides, in
both the villages and cities of inter-war Greece the electoral appeal
of the socialists was minor, and that of the communists insignificant. In these circumstances the conservatives, under the banner
of the Popular Party, were one of the two major parliamentary
forces (the Liberals being the other), and although both major
bourgeois parties frequently appealed to one army faction or
another for support, in contexts of relatively free competition for
votes the conservatives were not as hopelessly weak as their
112
Argentinian counterparts. In the 1920 elections for example, the

Greek pro-royalist conservative forces managed to defeat the
Venizelists in a (by Balkan and Latin American standards) relatively free electoral contest. 47
Finally, during the inter-war period the Greek conservatives,
operating in a context where more than half of the labour force
was still in agriculture and where clientelism was prevalent as a
mode of political inclusion, managed to retain widespread electoral support- particularly in the countryside. 48 As will be seen in a
subsequent section, it was not until much later (after the 1960s)
that the political Right in both Greece and Chile rapidly lost its
rural electoral base.
Having discussed some basic reasons for the differences in the

conservative reactions to the post-oligarchic electoral challenge in
ACG, and the relevance of these reactions to the military's
involvement in politics, I would like to close this section by
emphasising once more what these three countries had in common
in the area of civil-military relations.
Despite the fact that during the immediate post-oligarchic
period the three societies under examination varied considerably
in respect of their armies' dominance in politics, none of the three
approximated to the type of institutionalised subordination of the
military to civilian authority that prevailed in western parliamentary democracies. Not only did the military frequently intervene in
the political process but, especially in Argentina and Greece, they
became a major arbiter/regulator of that process. However, as will
be argued later, despite the army's central role in post-oligarchic
politics its dominance was tempered by the importance of the
relatively large middle classes in these societies. In other words, if
the political representatives of the civilian forces in post-oligarchic
ACG failed to establish their hegemony over the military, neither
did the military manage to subjugate these forces sufficiently to
establish its own clear and irreversible dominance within the state.
113
3.2 The post-war expansion of industrial capitalism: restricted

and uneven development
After having examined the military's situation in the early postoligarchic period, and before analysing their role in post-war
political developments, we shall first take a look at the post-war
trajectory of ACG's economic development. Since Chapter 1
outlined the import-substitution process of the inter-war years in
the Balkans and Latin America's southern cone, the focus here
will be on post-war industrialisation in ACG and its impact on
basic political structures. Particular emphasis will be laid on those
dimensions that are directly relevant to the main problem under
consideration: the rise of military dictatorships.
A. The Second World War, by severely hampering Chile's and
Argentina's ability to import western industrial goods, further
intensified these countries' import-substitution processes. Moreover, given the general economic boom of the immediate post-war
years, both these countries experienced a considerable growth of
their industrial sectors and of their economies in general, even if
this was not as fast as their Brazilian neighbour's. 49
This last point is particularly significant for Argentina as the
leading economic power in pre-war Latin America. According to
some writers, Peron's nationalist/populist policies were largely
responsible for the country's post-war loss of economic dynamism.
It was unfortunate that, at a time when the world market was again
relatively receptive to the export of agricultural produce, Argentina (contrary to the policies adopted by Australia, for example)
should have turned her back on the production of exportable
goods (both agricultural and manufactured). Peron during the
1940s actually discouraged agricultural production, without at the
same time giving any help to the relatively difficult type of
manufacturing that could eventually have led to the production of
exportable industrial goods. Instead, his early economic policies
provided excessive economic protection for the technologically
backward industrial branches that, characteristically, had predominated in the early phases of import substitution but which in
the 1940s needed rather less protection and more modernisation.
In over-all terms, Peron's attempts to shift resources away from
114
agriculture resulted in the growth not so much of the manufacturing sector as of services (particularly the state bureaucracy) and of
the construction industry. 50
In Greece, the German occupation and the subsequent prolonged civil war had brought unprecedented destruction of the
economy. Despite this set-back, however, in the context of the
general post-war European expansion and with the help of massive
American aid, economic recovery was very rapid. With an average
growth rate of approximately six per cent, pre-war levels of
production were reached again in the mid-1950s. By the end of
that decade, industrial production was double that of 1938, and
industry's contribution to the GDP had reached the 25 per cent
level. 51
Another significant structural feature of post-war importsubstitution industrialisation, both in Greece and in the southern
cone countries of Latin America, was the ever growing role of the
state in the economy. State economic interventionism was not, of
course, a new phenomenon. As already mentioned, it was not only
in the post-Depression period that the state had played a major
part in the developing capitalist industry, but even in the
nineteenth century it had had a decisive hand in creating the
infrastructure which was the fundamental prerequisite for that
industry. State involvement in the post-war era, however, attained
hitherto unknown dimensions. It was not only that state control of
the over-all economy was greatly tightened up, 52 but also that
through the creation of specialised agencies under its control53 the
state began to effect direct investments in a variety of economic
projects. 54
B. Both in Greece and in the southern American cone in the late
1950s and early 1960s the type of economic development which
had characterised the post-1929 era and the immediate post-war
years seemed to have reached its limits. The process of industrialisation based predominantly on the production/substitution of
goods by means of simple technologies and geared to the satisfaction of internal demand now encountered serious difficulties. The
hothouse conditions in which the early import-substitution projects had been nurtured prevented the indigenous high-cost industries from being able to compete internationally. So when there
were fewer possibilities for continuing this easy, as it were
115
horizontal type of industrial expansion (that is, when the possibilities of predominantly servicing the internal market began to
diminish), this increasingly resulted in a bottleneck which, in order
to be overcome, required a deepening of the industrialisation
process. This meant a shift of emphasis to more complex and
difficult forms of industrial investment capable of leading to a
more vertical expansion of industry (to the production of, for
instance, intermediate and capital goods), and/or to the manufacture of exportable industrial goods.
The difficulties in industry were aggravated in all three ACG
countries by an overinflated service sector, with low-productivity
jobs in the state bureaucracy and elsewhere proliferating at great
speed, and by a highly inefficient agricultural sector. The relatively
low productivity of the latter was due not only to generally adverse
world-market conditions and to often inept state policies, but also
to the type of relations of production prevailing in the countryside.
Although these were different in each specific case, they all
prevented the optimum use of existing human and non-human
resources.
In Greece the pattern of small ownership and extreme land
fragmentation prevented the creation or consolidation of larger
units of production which could reduce costs and lead to a more
efficient form of mechanisation. 55 In Argentina on the other hand,
the short-term tenancy contract which the big landowners kept
imposing on their tenant farmers prevented the latter from making
necessary long-term investments in the plots they were
cultivating56 Finally in Chile, although the predominantly foreigncontrolled mining sector was relatively dynamic, agriculture in the
1950s (in the absence of any serious land reform) was still marked
by the dominance of the big latifundists and by the typical
latifundian/minifundian combination which, as in other parts of
Latin America, was putting serious obstacles in the way of
agricultural modernisation. 57
It was this combination of a costly, non-competitive industry,
inefficient agriculture, and a huge low-productivity tertiary sector
which created severe economic problems, especially in the southern Latin American countries. In both Argentina and Chile the
crisis of the import-substitution process manifested itself in rapidly
rising inflation and growing balance-of-payment deficits. Inflationary tendencies were by no means a novelty in these countries; but
116
whereas during the inter-war years inflation had not seemed to

slow down industrial growth, in the early 1960s, especially in
Chile, high inflation rates sharply aggravated economic
stagnation58 and, despite attempts to encourage private investments, capitalists did not respond to the generous incentives
offered by the Jorge Alessandri administration (1958-62). 59
In Greece, where industrialisation was less advanced, the import-substitution difficulties did not appear in acute form,
although here too there were serious balance-of-payment difficulties and a drop in manufacturing investments at the end of the
1950s. As in Chile, neither the incentives offered by the conservative government (under Karamanlis) nor the strict control exercised by the state over the economy could succeed in directing
Greek capital into technologically more complex manufacturing
sectors. 60
The difficulties and bottlenecks mentioned above are often
interpreted in the relevant literature as indicating the 'exhaustion'
of import-substitution industrialisation. Now one might well ask to
what extent the economic adversities of the late 1950s and early
1960s genuinely reflected such an exhaustion, that is, to what
extent it was severe structural obstacles that were blocking further
industrialisation along the same lines rather than transitory difficulties which could have been overcome by appropriate state
policies and without resorting to the multinational solution. 61
Whatever the conclusion of such an enquiry, the fact remains that
during this crucial period there were significant developments
within the world economy (spectacular growth of multinational
corporations, geographical and functional diversification of their
operations, their increasing willingness to make direct industrial
investments in third-world countries, and so on) which, in conjunction with the import-substitution difficulties in several of the
semi-peripheral economies, led to a new phase of industrial
capitalism in the latter. This new phase was characterised by an
intensification of foreign-capital investments (especially of US
origin) that were not portfolio investments but directed into the
key manufacturing industries where indigenous capital was unable
and/or unwilling to go without multinational support. So with
respect to the three ACG cases, Frondizi in Argentina, Frei in
Chile and Karamanlis in Greece, all managed to attract considerable amounts of foreign capital in the early and middle 1960s. 62
117
To be more specific, in Argentina the modest inflow of direct

foreign investments in industry during the 1930s and early 1940s
had come to an abrupt halt with the rise of Peron and his initiation
of ultra-nationalistic, anti-foreign capital policies. Although in the
early 1950s Peron was forced to change his hostile attitude vis-a-vis
foreign capital, it was not until the early 1960s that there was a
substantial resumption of direct foreign investments in technologically complex manufacturing branches (chemicals, petroleum refining, automobiles, non-electrical machinery, and so on). 63 In
Chile as well, although the Alessandri administration took several
measures in the 1950s to attract foreign capital, it was only during
the Frei administration (1964-70) that foreign-capital investment
in industry increased substantially. In fact, while between 1961 and
1964 the percentage of foreign to over-all investment in fixed
industrial capital was only 6 per cent, it rose to 20 per cent in the
1965-70 period. 64
In Greece, legislation for attracting foreign capital had been
initiated in the 1950s, but it was in the early 1960s that foreign
capital, taking advantage of the considerable privileges granted it
by the Greek state, increased its investments. Whereas in 1960 the
yearly influx of foreign capital amounted to approximately $11.5
million, in 1963 it went up to $50 million, and in 1966 to $157.5
million. 65
These increases in foreign-capital investments are not so impresssive when seen in relation to the over-all investment level of
the economies concerned. 66 However, the fact that they were
often oriented towards fields requiring complex technologies and
advanced know-how meant that their impact, at least up to the
1974 world recession, was much greater than their relatively small
size would suggest. So in the 1960s in Argentina there was a major
acceleration in consumer-durables and heavy-industry investments, whereas Chile's economic growth received a boost when
new capital oriented itself towards car assembly and branches of
the chemical industry. 67 In Greece during the same period there
was a definite shift in emphasis from the traditional industries to
more dynamic and complex forms of manufacturing (chemicals,
the metal construction industry, and so on), resulting in a concomitant change in the composition of exports: traditionally an exporter of agricultural produce and unprocessed minerals, Greece now
began to export considerable quantities of industrial goods. 68
118
These developments in ACG not only gave a much-needed push

to the industrialisation process, they also contributed to its relative
deepening, bringing these economies, superficially at least, nearer to
those of the capitalist centre. This is true not only of Argentina,
where the economy in some respects resembled the model of the
capitalist centre even before the 1960s, but also with respect to
Chile and Greece. In terms of the over-all economic structure, in
the early 1970s industry's contribution to the GNP was 29 per cent
in Chile, 31 per cent in Greece, and 45 per cent in Argentina;
whereas agriculture's share was 10 per cent in Chile, 17 per cent in
Greece, and 13 per cent in Argentina. 69 Moreover, dramatic
changes took place in Greece's labour-force distribution as rapid
industrial growth, migration and urbanisation led this country to
experience a drastic decline of her agricultural labour force, thus
making her occupational structure more similar to that of Chile
and Argentina. 70 As a result of these developments, the agriculturallabour force employed today in all three countries is less than a
quarter of the total work force, industry contributes more to the
GNP than the primary sectors, and the majority of the population
is not only urban but primarily concentrated in a few overpopulated large cities. 71
C. A point that requires special emphasis, however, is that the
type of industrialisation experienced in ACG, even in the 1960s,
was different from that of the capitalist centre, and such differences are crucial for understanding the structure of ACG politics.
Since capital accumulation in these countries was not as indigenous a process as in the West, and given the timing of their
industrialisation, it is not surprising, first, that industry should
have developed in breadth rather than depth; 72 and secondly, that
the over-all economy had a comparatively more disjointed, disarticulated character73 - in the sense that the 'forward' and 'backward' linkages between the economic sectors were not as strong
and/or as self-regulating as in the western capitalist economies. 74
An obvious indication of this is the lack of a relatively independent technological base and the very limited development of
capital-goods industries, even in the larger and more developed
Argentine economy. 75 Such a state of affairs tends to lead to
serious balance-of-payment difficulties, particularly at periods of
119
rapid industrial growth which necessitate the large-scale importation of machinery. 76

Another indication of relative disarticulation in the ACG economies is the fact that capitalist industrialisation proper took on a
more restricted and uneven character: restricted in the sense that,
despite the dominance of the capitalist mode of production in
industry, wage labourers working in capitalist enterprises77 constitute only a small fraction of the industrial labour force; and uneven
in the sense that there is marked organisational heterogeneity both
within the industrial sector as a whole and between sectors, a
heterogeneity resulting in huge productivity differentials and
imbalances/disruptions.
This becomes quite obvious when, again adopting a historicalcomparative perspective, the strikingly different ways are examined in which, for instance, labour, becoming redundant in
agriculture, has been employed in the capitalist centre and in ACG
respectively. Around the turn of the century, industrial employment in the UK constituted slightly more than half of all urban
employment. Despite various fluctuations this was still practically
the same 50 years later (51.9 per cent in 1951). In France the
pattern was similar: industrial employment was 51.4 per cent of all
urban employment in 1881, and 51.3 per cent in 1954. In Latin
America's southern cone countries on the other hand, the highest
ratio of industrial to over-all urban employment was not only
much lower than it had been in the West, but in the long term it
appears to be shrinking. So in Chile the share of industrial
employment in all non-agricultural employment was approximately 35 per cent in 1925, and by 1968 it was down to 22 per cent. In
Argentina during the same time span it dropped from 30 to 26 per
cent. 78 There appear to be no comparable data for Greece, but in
view of the similarly capital-intensive pattern of her industrialisation, the labour-absorption capacity of the industrial sector is
equally low. (Unemployment in the post-war years was less severe
than in the southern cone countries, however, because of the
massive labour migration to western Europe). 79
Moreover, not only is ACG industry as a whole fairly restricted
as far as labour absorption is concerned but, within industry,
capitalist-factory employment is not as important as it is in the
western economies. This means that industrial wage earners
120
constitute an even smaller fraction of the total labour force than is

inherent in the low labour-absorption of industry. Or, to put it in
mode-of-production terms, the capitalist mode of production in
industry, despite its dynamism, coexists with a very large simplecommodity sector which continues to engage a very high proportion of the industrial labour force in primarily artisanal activities.
This leads of course to a marked organisational and technological
heterogeneity of the productive structures. On the one hand there
are a few large capitalist enterprises which employ a considerable
section of the wage earners and contribute by far the largest share
of total industrial output; and on the other there is a plethora of
small, family-oriented, artisanal units. In fact, one of the most
striking characteristics of peripheral and semi-peripheral industry
is the persistence and even proliferation (especially in the more
traditional industrial branches) of small inefficient units existing
side by side with the huge firms that are usually state or foreigncontrolled. 80 (The term dualism quite adequately reflects this
hiatus within industry but, because it is often associated with
neo-evolutionist views of development81 I shall use the more
awkward term of 'heterogeneity of productive structures'.) And
although this type of industrial structure is not as accentuated in
Argentina, Chile and Greece as in typically peripheral economies,
it does constitute an important feature of these countries' economic organisation.
For instance, in the middle of the 1960s, out of all industrial
establishments with more than five workers, those employing
more than 100 accounted for 6 per cent in Chile and 10 per cent in
Argentina. In both countries these relatively large units gave jobs
to more than half of the industrial labour force and they contributed more than 65 per cent of the total industrial value-added. 82
The situation is similar in Greece. In both the inter-war and
post-war periods, a few big firms have coexisted with a plethora of
small, family-based artisanal units. For a variety of reasons, any
increase in the demand for indigenous industrial products generated a proliferation of additional tiny units instead of a consolidation/enlargement of existing ones. 83
Needless to say, the above imbalances, and especially so the
marked organisational and technological heterogeneity of productive structures (which often implies high productivity differentials
among and within sectors), are important factors leading to big
121
and growing inequalities, particularly during periods of rapid

economic growth. 84
A quick look at the post-war income inequalities in ACG shows
that, despite progress in terms of reducing absolute levels of
poverty, there is a huge difference between the bottom and the top
of the income scale. Comparing the centre with the semiperiphery, inequalities in ACG seem to be much sharper at the top
of the social pyramid. So in the 1960s and early 1970s, the top 20
per cent of the ACG population received 50 per cent or more of
the national income, whereas in the capitalist centre their average
was well below the half-way mark. 85 The percentage of the
national income received by the top 5 per cent of the social
pyramid makes for an even more lop-sided income-distribution
picture: this highly privileged section of the population received
(before tax) 22.6 per cent in Chile (1968), 29.3 per cent in
Argentina (1961), and 23 per cent in Greece (1957). 86
On the other hand at the bottom of the income pyramid the
differences between centre and semi-periphery are relatively less
accentuated. In Chile for instance (1968) the bottom 20 per cent
received 4.5 per cent of the national income, and in Argentina
(1961) 7 per cent. With the equivalent figures being 6 per cent for
the UK (1968), 5.9 per cent for Germany (1970), and 6.7 per cent
for the United States (1970), the contrast here is less sharp. 87
It must be remembered, however, that the social composition of
the bottom income groups in the centre and semi-periphery
respectively is dissimilar: in the centre the bottom quintile includes
a great number of persons who are either retired or whose low
remuneration is temporary (for example, young workers); in the
semi-periphery the majority in the bottom 20 per cent income
bracket consists of employed workers with permanently low
incomes. 88 Despite this qualification there is no doubt that in the
ACG countries the post-war, foreign-led industrialisation resulted
in a situation where some crumbs of the considerable wealth
generated by relatively high growth rates have trickled down to the
social bottom of the pyramid.
This type of unequal distribution, characterised by the elimination of starvation levels of poverty and, at the same time, an
excessively high concentration of wealth at the top, naturally
creates very favourable conditions for the rapid radicalisation of
the underprivileged, especially when it occurs in a context of mass
122
parliamentary politics. In the West, the 'hump' period of industrialisation (typified by huge disruptions and rapidly widening
inequalities) occurred at a time when the majority of the population remained outside the political arena (late eighteenth/early
nineteenth century). By the time popular demands for broader
political participation became imperative, economic inequalities
had reached a plateau, thus greatly facilitating the problem of
distributing political rights. 89 In ACG, by contrast, inequalities
have been growing rapidly during the post-war period, particularly
so during the 1960s' intensification of the process of foreign-led
industrialisation. 90 By that time, given the relatively early demise
of oligarchic parliamentarism, large sections of the population had
already entered the political arena. In such a situation income
concentration at the top, being both more visible and less acceptable, accentuates the economic contradictions of the system and
creates conditions facilitating 91 the political radicalisation of the
masses. In other words, if the inter-war broadening of political
participation in ACG has led to an accentuation of the incorporative features of the state, and so to a greatly unequal distribution
of political power between rulers and ruled, the post-war highly
uneven capitalist industrialisation has led to an equally uneven
distribution of economic rewards between economically privileged
and underprivileged groups. The failure of the ACG countries to
distribute more equally and broadly both political/civil and social/
economic rights92 must be seriously taken into account in any
attempt to explain the chronic malfunctioning of these countries'
parliamentary institutions. This point brings us to a more systematic consideration of the basic features of the ACG polities.
3.3 On the basic political structures of Argentina, Chile and

Greece
Having briefly discussed the timing of capitalist industrialisation
and its relatively restricted and unequal character, let us now see
what structural limits this type of development sets on the political
level.
3.3.1 The limiting framework
To begin with, the restricted character of capitalist development,

as well as the marked organisational and technological heter-
123
ogeneity between sectors, present obstacles to the autonomous

collective organisation of the working classes. Late capitalist
expansion being so restricted means that wage earners working in
relatively large economic units constitute a small fraction of the
working population; the majority of the labour force operates in
work contexts where relatively autonomous trade-union structures
can emerge only with difficulty. 93 Moreover, the greater heterogeneity of productive structures, the large productivity differentials, and the ensuing income inequalities as well as the marked
differences in work conditions between technologically advanced
and backward sectors, all tend to create severe cleavages within
the working class. Such structural cleavages exist in the capitalist
centre too, of course, but they are more acute in the semiperiphery and to that extent it becomes much more difficult to
mobilise and organise the workers in such a way as to put an
effective check to the state's incorporative tendencies, and/or to
the manipulation of the rank and file by populist/paternalistic
leadership.
If additionally one takes into account what has already been said
about the pre-independence patrimonial legacy, the expansion of
the state before industrialisation and its authoritarian orientation
vis-a-vis working-class organisations, it becomes obvious why a
less dependent, western-style inclusion of the lower classes in
politics, which I have called integrative, is less likely to occur in
ACG. In fact, to the extent that the state from the outset played
the central role in the development of national industry, it was
only natural that it should attempt to control all its different
aspects, including of course the newly-created proletariat.
However if, among other factors, the late and restricted character of capitalist industrialisation provides a less favourable ground
for the development of integrative modes of working-class inclusion, it does provide the conditions for the rapid political mobilisation and radicalisation of the lower classes. The early inclusion of
the ACG countries into the world market and the resulting rapid
urbanisation, their early adoption of parliamentary institutions, as
well as their recent industrialisation have inevitably generated
processes of large-scale political mobilisation as the majority of the
population was drawn into national politics. Since this drawing-in
process occurred in circumstances of growing economic and social/
political inequalities, it has been highly conducive to the fast
radicalisation of the masses. Nevertheless, such radicalisation does
124
not lead to a revolutionary type of mobilisation threatening the

bourgeois order. This is a point which deserves some discussion.
Given the high concentration of the population in a few urban
centres, and the enormous capacities of the modern state for
repression, the present chances of a revolutionary overthrow of
the capitalist state in ACG are rather low. Not only is the type of
state collapse associated with classical revolutionary situations (for
example, those of Bourbon France or Tsarist Russia) 94 not very
likely, but ACG also lack those isolated but at the same time
highly populated regions which, lying beyond the easy reach of
centralised state control, can become the breeding ground for
effective guerrilla/revolutionary activities against the political
centre. 95
Moreover, the exceptionally high urbanisation in these countries reduces their revolutionary potential in several other ways.
First of all, a large-scale exodus to the cities often operates as a
strong safety valve for revolutionary propensities in the countryside. Despite the great dislocation created by rapid urbanisation
(especially if it is accompanied, as in ACG, by capital-intensive
industrialisation), massive peasant migration can in certain circumstances be a substitute for rural revolt or revolution. 96
The chances of revolutionary upheavals are considerably higher,
of course, when the radical, anti-oligarchic phase of the rising
urban middle classes coincides with the radical mobilisation of the
peasantry. However, as pointed out earlier, neither Greece nor
the Latin-American cone countries experienced such an antioligarchic rural-urban alliance. In Greece the relative weakness of
the landowning classes meant that the break-up of the restrictive
oligarchic political system was achieved without any populist
mobilisation in either the towns or the countryside. In southern
Latin America the urban middle classes managed to accommodate
themselves and to consolidate their power position in the postoligarchic political framework by mobilising only the urban masses
- leaving the rural populations under the control of their traditional masters (see Chapter 1.2). When later the lower classes in
the countryside (particularly in Chile) began to mobilise radically,
the middle classes had long ceased to be political outsiders and had
become a main bulwark against any popular threat to the relations
of production or domination.
125
This post-oligarchic conservative role of the middle classes in

ACG97 has been, and is being, performed quite effectively due to
the relative strength of their political organisations. Even if the
urban middle classes in the three countries cannot, in terms of
political organisation, be compared with their western European
counterparts, they are much stronger than those in less urbanised
and industrialised peripheral formations. It is not only the high
level of urbanisation which explains the strength of urban political structures and organisations in ACG, but also the timing
of urban growth. Countries where the levels of urbanisation went
up only very recently (for example, Venezuela, where the urban
population doubled between 1950 and 1970) have less established
party structures than older urban societies (for example, Chile and
Argentina). 98 So the ACG countries' relatively early urbanisation
and their equally early establishment of a restrictive parliamentary
system resulted in a situation where political parties, despite their
particularistic character, could and did take strong roots. 99
Another, related factor which diminishes considerably the
chances of an anti-capitalist revolution in ACG is the fact that
the indigenous dominant classes were not controlled by foreign
interests to the same degree as, for instance, in pre-revolutionary
Cuba and several Central American 'banana republics'. In the
latter, the very high degree of foreign penetration and control of
both the polity and the economy made it very much easier for the
indigenous ruling classes to be identified with foreign domination,
and therefore to mobilise the rural and urban masses for nationalistic/revolutionary ends. The dominant classes in ACG on the
other hand were not simply puppets manipulated by foreign
interests, despite the importance of foreign interests in their
export sectors, and despite the foreign-led character of their
post-war industrialisation. 100
The greater strength of the indigenous dominant classes in ACG
therefore, the early and rapid development of urbanisation and
later of industrialisation, and the establishment of parliamentary
party systems as early as the nineteenth century, have resulted in
the creation of a numerically large and (by third-world standards)
relatively well-organised middle class with political formations that
have operated as effective buffers against the revolutionary potential of the lower classes.
126
The same set of circumstances explains why not only a revolution from below, but also a revolution from above is highly
unlikely in the ACG countries. While the dominance of the
military within the post-oligarchic power bloc is real enough, it is
not an overwhelming dominance. Relatively strong civilian forces
(compared to more peripheral formations) curb the army's proclivity for institutionalising military rule irreversibly, or for acquiring the type of autonomy vis-a-vis civil society seen, for instance,
in Peru (1968-75) or in Kemalist Turkey. 101
In other words, even if the political representatives of the
civilian forces in post-oligarchic ACG did not manage to keep the
armed forces and the state bureaucrats under the kind of hegemony that has prevailed in Anglo-Saxon parliamentary democracies, they have been strong enough to prevent a situation in which
the military or state bureaucrats can establish an overwhelming
dominance within the state. In fact, the lack of civilian hegemony,
in a context of civilian forces strong enough to prevent the holders
of both the means of coercion and of administration from establishing their own hegemony, constitutes a characteristic which
differentiates ACG from the western parliamentary democracies
and from the type of peripheral polity characterised by permanent
military or one-party rule.
In conclusion, an analysis of the timing and structure of capitalist industrialisation in relation to political developments suggests
that, at least in the foreseeable future, the chances of revolution,
whether from below or above, are as poor in ACG as the chances
of achieving a 'western-type' integrative mode of political inclusion of the masses. If these two unlikely prospects form the outer
political limits, what can be said about the political arrangements
and slow-changing structures that are typical within this limiting
framework? What, on the level of political organisations, are the
most typical forms of political inclusion of the lower classes in
ACG politics?
3.3.2 Vertical incorporative modes of inclusion
The short answer to the above questions is that in ACG the lower
classes entered politics in a more vertical, heteronomous manner.
Given the pre-independence patrimonial legacy, the relatively
early expansion of the state and its repressive apparatuses, the
127
type of transition to post-oligarchic politics and the type of

industrialisation already discussed, the rural and urban lower
classes were brought into the political system in such a way that
their organisations neither managed to put a decisive check on the
incorporative/paternalistic tendencies of the state, nor could they
eliminate (or rather render peripheral) the all-pervasive particularistic, personalistic features of the polity that had characterised
the oligarchic period.
The large-scale persistence of personalistic/particularistic politics (and the extensive political corruption associated with it)
leads, of course, to incorporation rather than integration of new
strata into the political arena. This incorporative process took a
variety of forms in ACG, such as:
(a) the persistence or reactivation under novel and more centralised forms of clientelistic political organisations which cut
across and weakened more horizontal forms of political
representation;
(b) the vertical/populistic mobilisation of the lower classes by
charismatic leaders;
(c) the manipulation and control of parts of the trade-union
movement by the state;
(d) direct state control of parts of the mass media, of higher
education and of other, voluntary organisations, and so on.
Whenever such vertical modes of incorporation are found to be
insufficient, they are supplemented by the selective exclusion from
politics of specific groups or political parties. 102 So for example at
certain periods of their post-oligarchic parliamentary development
all three ACG countries introduced laws excluding certain political
parties from parliamentary representation.
While the above is true in over-all terms, both the degree and
mode of lower-class incorporation has of course varied considerably in the three ACG countries. In Greece, until the establishment of military rule in 1967, the major bourgeois political parties
retained their marked clientelistic/personalistic character, trade
unions were to a large extent state-controlled and, after the bloody
civil war (1944-49), the Communist Party was legally banned and
fellow-travellers and left-wing sympathisers systematically persecuted through a variety of legal and illegal means. 103
In Argentina the party-political structures, characterised by a
mixture of clientelistic and populistic elements, portrayed a parti-
128
cularism quite similar to those of Greece. Trade unions on the

other hand, after the advent of Per6n, acquired a much more
massive and militant character. However, one should keep in mind
that the growing strength of Argentinian trade unionism was
initially based on Per6n's large-scale unionisation from above of
workers newly arrived from rural areas. As already mentioned, for
a long time these newly-founded unions were a pliable instrument
for Per6n to consolidate his power not only against the oligarchic
forces but also against the older, socialist and communist
unions. 104
In Chile, given the strength of her party system, incorporative
mechanisms of inclusion were less accentuated in the postoligarchic period. Even so, while at the national level Chilean
politics focussed on issues rather than personalities, on the local
level (and especially in the rural areas) clientelism and the various
particularistic practices associated with it played such a central role
that the way in which the national political centre articulated with
the periphery resembled the Greek or Argentinian rather than the
French or English model. In fact, even Chile's socialist and
communist parties, in their attempt to enlarge their social bases in
a society of late and restricted capitalist development, were often
obliged to resort to particularistic modes of recruitment and
organisation. 105 Similar personalistic/particularistic characteristics
are to be found in Chile's public administration system. 106 Furthermore, the highly paternalistic and restrictive manner of Chile's
labour legislation (which made trade unions dependent on political
parties for financial support) 107 is another element pointing to the
incorporative character of Chilean politics.
Despite the considerable variations in relations between state
and civil society in ACG, none of these countries has the type of
civil society which has, for its backbone, strong trade unions and
political parties which, although closely linked, are relatively
independent of the state and each other. On the other hand, as
noted earlier, compared with more peripheral capitalist formations, civil society in ACG is quite strong. In that sense these three
countries are semi-peripheral in not only economic, but also
political, terms (that is, their civil society and parliamentary
institutions are closer to those of the West than to those of other
third-world capitalist countries).
129
3.3.3 The major political contradiction

After analysing the incorporative/exclusionist ways in which the
lower classes were brought into the political arena, we can now
formulate the basic political antinomy of post-oligarchic, post-war
ACG: the contradiction between, on the one hand, high levels of
political participation/mobilisation and, on the other, the prevailing incorporative/exclusionist modes of political control, the latter
being incapable of coping in any stable manner either with the
successive waves of new entrants into politics, or with the higher
levels of activation/radicalisation of those already in the political
arena.
A. It is of course only the integrative, more autonomous mode of
inclusion that, other things being equal, can on a long-term
irreversible basis accommodate the massive entrance of new
participants intq the political game without reinforcing tendencies
towards a breakdown of the parliamentary institutions and the
imposition of dictatorial solutions. It is only with an integrative
system that the new entrants, given the horizontal/nonpersonalistic mechanisms of inclusion, can reinforce the strength
and autonomy of existing collective organisations. Only then, on
the level of collective action, can the distribution of political power
be organised in such a way that extreme polarisation between
rulers and ruled is avoided and civil society is strengthened by
becoming more resilient to state manipulation; moreover, this
type of strengthening, as the English model of political development has shown, presents no threat to the bourgeois order but, on
the contrary, further legitimises it by making it more hegemonic.
An over-all incorporative context on the other hand is less
capable of a stable long-term assimilation of the masses in active
politics and, therefore, of any effective solution to the powerdistribution problem. The more accentuated the economic and
political processes that lead to a state of mass politics, the greater
the inequalities in the distribution of power, and the more
inadequate the incorporative mechanisms of inclusion to cope with
the ensuing conflict between politically dominant and dominated
groups. For example, as the politicisation/radicalisation of the
masses increases, clientelistic mechanisms of inclusion become
130
more fragile and precarious, not only in the urban centres but even
in the rural areas. 108 As argued earlier, if it is clear that restricted
and uneven capitalist development does not irreversibly destroy
but merely transforms clientelistic networks, it is just as clear that
in this process clientelism not only changes its complexion but also
becomes more fragile and precarious. While patronage networks
can indeed survive in the rapid development of urbanisation,
commercialisation and industrialisation, they are constantly disrupted by the emergence and growing radicalisation of nonclientelistic organisations (whether populist or not) which can
seriously challenge the political status quo. 109
My major argument therefore is that, as the masses in ACG
tend to become fully politicised, the mixture of incorporative/
exclusionist controls which developed in the immediate postoligarchic period is gradually being undermined. The increasing
contradiction between high levels of political participation on the
one hand, and persisting incorporative/exclusionist control
mechanisms on the other, tends to generate a growing polarisation
as excluded or incorporated groups become more radicalised, and
as groups more or less directly involved in the maintenance of the
prevailing relations of domination become more reactionary. This
is to say that the increase in systemic contradictions between high
participation and mechanisms of incorporative controls tends, on
the level of collective action, to exacerbate political struggles 110 the prevailing relations of domination being challenged from
below by organisations and movements trying to establish a degree
of autonomy that is incompatible with these control mechanisms.
Such a challenge to the prevailing relations of domination often
means a challenge to the military as one of the major monitors of
the incorporative/exclusionist system of controls; it means, in
other words, a challenge to the power position of the armed forces
within the state.
It has been pointed out repeatedly that the major economic
contradiction between accumulation and distribution becomes
particularly marked in the rapidly industrialising post-war capitalist countries. In fact, given the late and relatively exogenous
character of their industrial development, and the greater heterogeneity between advanced and backward sectors in their economies, any really serious attempts at the reduction of inequalities
and the development of welfare create tendencies towards flights
131
of capital, investment strikes, and over-all economic paralysis.

In the light of the above one can argue that in societies that are
'late-late' developers and have adopted parliamentary forms of
rule relatively early (that is, ACG), the accumulation v. distribution contradiction is coupled with an analogous contradiction on
the political level: between high mobilisation and incorporative/
exclusionist controls. 111 Given the timing and mode of development of parliamentary institutions in the semi-periphery (that is,
the fact that they preceded industrialisation, and that they kept
their particularistic character even in the post-oligarchic period),
any attempt to broaden the distribution of effective political rights
among the masses threatens the over-all political system.
It is not, therefore, only that the economic system in ACG is not
flexible enough to acccommodate a wider distribution of wealth
and the development of the welfare state; it is also that the
political system lacks the capacity to absorb, without severe
disruptions of parliamentary rule, new entrants into politics or
higher levels of political activation/participation. In other words,
late-late capitalist developers with early parliamentary institutions
face severe limitations in the effective distribution of both socioeconomic and political rights. 112 Moreover, unlike western Europe
where (as already mentioned) the problems of economic and
political distribution were tackled consecutively (the distribution of
political rights coming after the 'hump' period of industrialisation,
that is, after a certain stabilisation of economic inequalities), 113 in
ACG the problems of economic and political distribution became
acute more or less simultaneously, creating increasingly severe
strains on the over-all social formation.
B. Before closing this short section, a digression might be useful
in order to specify the above historically oriented centre-semiperiphery comparison a little further, particularly as far as the
'centre' side of the comparison is concerned.
In terms of economic and political-distribution problems, the
'English' pattern of socio-political development is quite different
from the 'Franco-German' one. In the former case the transition
from oligarchic to mass politics occurred in such a way that the
new entrants into politics (particularly, if not exclusively, the new
industrial classes) provided a further strengthening of an already
strong and autonomous civil society, that is, they managed to
132
constitute themselves into well-organised interest groups relatively

autonomous from the state. This mode of integration of the masses
into politics meant a less authoritarian and more equal distribution
of power among rulers and ruled, a type of distribution which,
seeing it was combined with an already achieved stabilisation of
economic inequalities, 114 enhanced bourgeois hegemony and parliamentary/democratic forms of rule.
On the continent of Europe, political developments in the
nineteenth century were not uniform: Belgium, Switzerland and
the Scandinavian countries followed the English pattern, but
France and Germany did not. It may not have been coincidental
that France and Prussia, both of them considered as models of
'political modernisation' during the rise of absolutism, should have
had the greatest difficulties in coping with the problem of political
participation during a later and different phase of modernisation.
Having effectuated the expansion of absolutism and the destruction of feudal localism in a fairly thorough manner, France and
Prussia were to find that their very success and the ensuing
centralist/authoritarian character of the state became a serious
obstacle to a liberal/democratic solution of the mass-participation
problem.
The economic-distribution issue in France and Germany was
tackled by the dominant classes in a manner similar to the English.
Bismarck's welfare policies, for instance, and particularly the
social-insurance schemes he initiated, were the most advanced in
all of Europe. However, his famous social-insurance laws for
sickness, accidents and old age had been preceded by an 'antisocialist law' (1878) designed to curb the political power of the
working-class movement. So in view of the late and authoritarian
manner of German unification and the rapid industrialisation
from above which created a state-subservient bourgeoisie, it is
not surprising that, even as late as the First World War, real power
resided outside parliament. 115
In France, where the restrictions of parliamentary power had
weakened after 1870, the bloody events of the commune resulted in
a polarisation of the political system which seriously inhibited
political reforms. Whereas the extreme Left and extreme Right
clung to their ideological purism, the Centre coalitions that
continuously held power during the Third Republic were not
pressured as the English were to change their organisational
133
structures and to transform themselves into modern mass parties.

This is one of the causes of the marked chasm in French ThirdRepublic politics between the non-ideological, immobile, 'corrupt'
Centre and the 'purist' extremes, which latter presented no serious
electoral challenge to the weakly organised Centre forces. 116
There are, of course, many other reasons to explain the
cheque red career of bourgeois parliamentary institutions in France
and Germany, reasons which for lack of space and expertise
cannot be discussed in depth here. However, enough has been said
to show the contrast between the Anglo-Saxon and FrancoGerman patterns of political development. This indicates clearly
that, although the successful resolution of the economicdistribution problem is a necessary condition for the consolidation
of a strong civil society and of parliamentary democracy, it is not
sufficient by itself. Another major ingredient is the way in which
political power is distributed during the transition from oligarchic
to mass politics. If this process is to take place in a nonauthoritarian, liberal/democratic manner, then the crucial factor
seems to be not only the mode of political organisation of the
working classes, but also the degree of state authoritarianism and
the willingness and ability of the ruling groups to accept the
electoral challenge of the working-class movement.
The Anglo-Saxon road to development is characterised by a
(from the point of view of the consolidation of bourgeois parliamentary democracy) successful resolution of the economic and
the political distribution problems, this double success leading to
the formation of a type of civil society which operates both as a
bulwark against state arbitrariness, and as a strong source of
legitimation of bourgeois rule. The Franco-German road is
marked by the failure to deal effectively with the problem of
political participation, this failure leading to a civil society whose
autonomy vis-a-vis the state (an autonomy based on a massive
working-class movement) is at moments of severe crisis jeopardised by state authoritarianism.
It is one of the main theses of this book that the societies under
consideration here, partly due to their modes of expansion and
timing of industrial capitalism, have failed to tackle successfully
either economic or political distribution. This double failure is
relevant for understanding the long-term disruption of parliamentary rule in ACG in the post-war period.
134
3.4 On the post-war rise of military dictatorial regimes

It is in the light of the considerations formulated in the previous
section that the post-war military dictatorships in our three

societies will now be examined (1966-73 in Argentina, 1967-74 in
Greece, and since 1973 in Chile).
In all three countries significant levels of capitalist industrialisation had aggravated economic and political inequalities and led to
the rapid increase in the politicisation and radicalisation of the
popular masses in both the towns and the countryside. The
loosening of conservative controls in the latter, combined as
it was with growing socio-economic contradictions, created
unprecedented levels of political radicalisation. Such developments posed a serious threat to the status quo, a threat that the
post-oligarchic system of incorporative/exclusionist controls, refurbished and accentuated in the cold-war era after the Second
World War, was hardly able to cope with.
The growing contradiction between high levels of politicisation/
radicalisation and incorporative/exclusionist controls generated
on the level of concrete collective actors a growing polarisation
between groups that were excluded from the political arena
or included in an incorporative/vertical manner and those politically dominant groups, among them the military, that had a
vested interest in upholding the political status. In fact, the threat
from below was felt very strongly by the military who, particularly
in Greece and Argentina, were occupying a powerful position
within the polity in the early post-oligarchic period. Needless to
say, both the threat from below and the way in which the military
perceived and reacted to it has taken different forms in each
specific country.
In considering each case in some detail, I shall concentrate first
on the structure of the post-war system of incorporative/exclusionist controls; then deal with the processes which gradually undermined and posed a serious threat to it; and finally examine
developments within the military to explain their reaction to this
threat.
3.4.1 Greece
As far as post-war Greece is concerned, as soon as the

right-wing forces in 1949 had emerged the victors in the protracted
civil war, they instituted a regime of guided democracy which
A.
135
methodically excluded from power all who had been on the losing
side. The Communist Party was outlawed, and all left-wing
sympathisers kept out of the state apparatus and out of public life in
general through an intricate system of legal and illegal means of
exclusion.
The Americans, who in March 1947 had taken over from the
British the task of defending Greece against communism, played a
decisive role in shaping Greece's post-war system of guided
democracy. Given the large amount of economic and military aid
which in the context of the Truman Doctrine they poured into the
country both during and after the civil war, 117 their influence on
forming the Greek polity in the late 1940s and early 1950s was very
considerable. During the early years of their involvement, American officials, sensitive to American public opinion (which was
hostile to the granting of aid to anti-democratic/ultra-reactionary
forces), advocated moderation and supported the centrist political
forces. 118 So from 1949 to 1952 Greece was governed by a series of
centre or centre-right coalitions, some of them attempting to heal
the wounds of the civil war and to dispel the Right-Left polarisation by such measures as lifting martial law, reviewing the death
sentences for treason imposed by military courts, granting an
amnesty to political prisoners, and so on.
Exasperated, however, by the chronic fragmentation of the
Centre and the ensuing political instability, the Americans (especially after the outbreak of hostilities in Korea in 1950) adopted a
tougher, cold-war attitude, shifting their support to Marshal
Papagos and his newly-founded right-wing Greek Rally. Papagos's
party, which in both its title and its strategy deliberately imitated
de Gaulle's Rassemblement du Peuple Fran~ais, managed in the
elections of September 1951 to win 36 per cent of the votes and 114
(out of 250) seats in parliament. At the same time the left-wing
EDA party (United Democratic Left), which had strong links with
the banned Communist Party, succeeded in obtaining ten seats in
parliament. As Papagos was unable to form a government (he
refused to collaborate with other political parties in a coalition
government), he became the main opposition leader in parliament. He lost no time in launching an attack on the rickety
Centre/Centre-Left coalition government of General Plastiras,
and focussed especially on Plastiras's Pacification Bill which
granted amnesty to a large number of political prisoners. A few
months later, under strong American pressure, a new electoral law
136
adopted the principle of majority representation, and with the

help of this law the Greek Rally in the November 1952 elections
secured 49.2 per cent of the total vote and 247 of the 300 seats in
parliament. EDA on the other hand, although winning 10 per cent
of the vote, was not allotted a single parliamentary seat.
The Greek Rally coming to power ended the tradition of
unstable coalition governments which had run Greece both during
and after the civil war and began an eleven-year stretch of
uninterrupted conservative rule. By 1952 the system of exclusionist/incorporative controls which had been developing since the end
of the Second World War had achieved its full form. Apart from
the ban on the Communist Party and the persecution of its
followers, other major features of this system were:
(a) The revival of the inter-war clientelistic networks which on
the whole worked in favour of the conservatives - especially
in the rural areas where local political brokers were getting
considerable help from the gendarmerie and the various
other state and para-state agencies. So during elections
the gendarmerie could let certain individuals know that if
they did not co-operate, they would be refused a passport
to emigrate, or a hunting permit, an agricultural loan or a
certificate of 'civil reliability'. 119
(The certificate of 'civil reliability', which guaranteed that its
holder has no communist connections, was an efficient,
all-pervasive device with the help of which communist and
left-wing sympathisers were systematically treated as secondclass citizens.)
(b) The re-establishment of firm state tutelage over the tradeunion movement, achieved by such incorporative techniques
as:
(i) the creation of rubber-stamp unions to assure comfortable majorities for the Right;
(ii) the granting of representation rights to 'reliable' unions
only;
(iii) the governmental control of union finances;
(iv) the institutionalisation of electoral malpractices such as
ballot-rigging, blackmailing of independent-minded unionists, and so on. 120
137
(c) The maintenance of a huge army 121 which, purged of all

republican and/or left-wing officers and organised by the
Americans, was emerging as the major force within the
post-war Greek state. This military apparatus, through its
extensive intelligence and security services, its considerable
hold on the mass media (particularly radio) and its control of
quasi-military or even para-military organisations was exercising enormous power in post-war Greece. To give just
one example, the army was in control of the notorious
Battalions of National Defence (TEA) which, headed by
right-wing officers, were often terrorising political dissenters
in the countryside. And it goes without saying, the army
authorities were also monitoring the soldiers' vote. By moving, during elections, whole regiments into marginal constituencies, they could use the 'military vote' in order to help
the 'right' candidates to get elected. 122
As far as the relation between the army and the throne is
concerned, although it is quite true that the fiercely anticommunist officers who held key positions in the army were
royalists, they were by no means under the firm control of
the palace. Having been bled in the protracted civil war and
having emerged victorious, those in control of the armed
forces were not prepared to become the instruments of either
king or parliament. The meteoric political rise of Marshal
Papagos, who enjoyed enormous prestige among officers and
who was intensely disliked and mistrusted by the King, 123
was a clear indication of where the centre of power lay in
post-war Greece.
B. This mixture of exclusionist/incorporative controls was gradually undermined by the extensive structural rearrangements of
Greek society, resulting in part from the swift post-war foreign-led
industrialisation. The huge disruptions and inequalities of
Greece's rapid but highly unbalanced growth, whether seen in
terms of the massive exodus of the rural population, the fastgrowing income inequalities or in terms of regional imbalances,
created strong social unrest which, in combination with mounting
political tension, steadily undermined the prevailing relations of
domination.
138
To illustrate this point let us take the huge rural exodus of the
1950s and 1960s. In the course of these two decades, one and a half
million of the total population of nine million had to leave the
countryside. Since Greece's industry was unable to absorb this
labour force, those of the rural unemployed who could not find
parasitic jobs in the tertiary sector were virtually forced to
emigrate to the industrial countries of the West. On the one hand
this massive emigration operated as a safety valve by keeping
down urban unemployment, and through the migrants' financial
remittances helped both their families at home and the country's
balance of payments. On the negative side it disrupted thousands
of families and so fostered discontent, both among the migrants
themselves who felt pushed into a kind of exile, and among their
families whom they had to leave behind. Besides, emigration
accentuated an already emerging geographical mobility that was
eroding the villagers' mental horizons and making the growing
social inequalities both more visible and less tolerable. All of these
changes, in concert with a series of political developments to be
discussed below, steadily sapped the political controls established
after the civil war.
With respect to the trade-union movement for instance, the hold
that the pro-government right-wing forces had exercised over it
immediately after the civil war began to weaken, especially among
those unions that operated in economic contexts of large units with
relatively high concentrations of employees (banking, publicutility corporations, the construction industry and some manufacturing sectors). It was from the mid-1950s onwards that opposition
parties intensified their trade-union activities. In 1956 for example,
the Left founded the Democratic Trade Union Movement (DSK),
and not long afterwards the Democratic Trade Union Change
(DSA) brought together trade unions close to the political forces
of the Centre.
These and similar moves precipitated an intensification of strike
activities. Whereas during the 1950s the working days lost per 1000
employees had been fewer in Greece than in most EEC countries,
the situation was reversed during the 1960s. In 1959 the number of
working days lost per 1000 workers was 48, in 1960 it more than
doubled, in 1963 it went up to 271, and in 1966 it reached 519
days.t24
139
If increasing strike action was an indication of popular mobilisation and discontent linked more or less directly to the growing
inequalities of Greek capitalist development, the electoral decline
of the Right, both in the towns and the countryside, was an
indication of political mobilisation more directly linked to such
developments as rapid urbanisation and the drastic decline of rural
constituencies. In fact, while the spectacular growth of the
Athens-Piraeus conurbation and the massive exodus of the labour
force weakened the clientelistic networks of the Right in the
countryside, 125 a series of significant political events, combined
with the above favourable circumstances, contributed to the
strengthening of the opposition forces.
In 1958 the non-clientelistic left-wing EDA party was able to
register quite astonishing election gains and, with the continuing
fragmentation of the Centre parties, became for a while the main
opposition party in parliament. EDA was also able to give a boost
and partly control the student movement which developed in a
very spectacular way in the early 1960s. These developments
sounded the alert for the repressive state apparatus, particularly
for those die-hard anti-communist officers who occupied key positions
in the army and the state's Central Information Service (the
Greek CIA). These officers participated in the formulation of the
notorious Pericles Contingency Plan which, devised for the purpose of neutralising the communists in case of war, was used
instead to 'monitor' the voting process and to achieve victory in
the 1961 elections. 126 These elections were a major turning-point,
because those in control of the means of domination and coercion
did not limit their intimidation tactics to EDA followers but
extended them to the centrist political forces as well. This indiscriminate attack against all of the opposition, and the blatancy of the
para-state's intervention in the electoral process, was the start of
anendotos ('the unyielding struggle'), the famous campaign by
veteran liberal leader George Papandreou against the fraudulent/
repressive politics of the conservatives.
Given these circumstances George Papandreou managed to
unite the various centrist parties under the banner of the Centre
Union, and in the 1963 elections he triumphed over the traditional
right-wing forces, consolidating his gains in the 1964 elections with
an unprecedented 53 per cent of the vote. Meanwhile his son
140
Andreas emerged as the leader of a strong left wing in his father's

Centre Union. The successful left-wing anti-American, antiNATO populist ideology of this faction (the majority of whose
members formed the core of Andreas Papandreou's postdictatorial PASOK party) enjoyed a rising popularity and, despite
the hesitation and apprehensions of George Papandreou and of
the party's traditional/clientelistic old guard, it started moving the
Centre Union away from purely clientelistic orientations and
principles of organisation.
When George Papandreou became prime minister, pressures
from his party's left wing and the general state of mobilisation and
popular unrest forced him to tackle some of the economic inequalities by loosening wage controls and by allocating larger state
funds to social welfare and education. Concerning the latter,
considerable funds were used to introduce a series of measures
(such as extending compulsory schooling from six to nine years,
establishing free education at all levels, founding new provincial
universities, and so on) which compared favourably with
Venizelos's inter-war attempts to reform the post-oligarchic educational system. George Papandreou also tried to liberalise the
regime by loosening the repressive system of controls still operating in the countryside. The repressive role of the gendarmerie in
the villages was attenuated and the certificates of 'civil reliability',
although not actually abolished, were required less extensively
than before. Moreover, the Left was allowed to organise more
freely and this led to a spectacular increase in demonstrations,
'peace marches', and so on. These actions and, more seriously, the
prime minister's subsequent attempt to bring the armed forces
under firmer civilian control (see below), posed a direct
threat to the army's dominant position within the state. In general,
notwithstanding the moderate character of George Papandreou's
measures, his policies and the growing political unrest from which
they stemmed were sufficient to imperil the prevalent relations of
domination.
To sum up what has been said so far: by the mid-1960s economic
and political developments in Greece had resulted in conditions
that were steadily undermining the existing system of political
domination. Growing inequalities and the massive population shift
from the countryside to the towns and abroad had created a level
of discontent which, articulated and channelled in a political
141
direction, exacerbated specifically political contradictions and

presented a serious threat to the system of guided democracy. To
put it differently: the combination of a highly unequal distribution
of both wealth and political power, at a time of large-scale
lower-class politicisation, created a level of discontent which
ominously challenged the post-civil war mechanisms of political
control. These mechanisms (that is, the mixture of exclusion and
incorporation of the lower classes) could no longer cope effectively
with this threat from below. To meet the danger, the system had
either to be reinforced by the total abolition of parliamentary rule,
or thoroughly rearticulated and opened up.
The 'open' solution implied that, through pressure from below,
the parliamentary forces would establish a less subordinate role for
themselves, and that therefore the army's and, to a lesser extent,
the throne's dominant position within the state would be
weakened. This was indeed quite acceptable to an important
section of the bourgeoisie which, despite its apprehension at the
growing number of strikes and George Papandreou's liberalisation
policies, was not sufficiently alarmed to opt for a dictatorial
solution. A significant part of the Right under the leadership of
Kanellopoulos therefore accepted the risk of an electoral confrontation set for April1967, and came to a secret agreement with
Papandreou on electoral arrangements.
The reaction of the army, potentially the main loser from an
eventual opening-up of the system, was very different. If the
formation of a more open parliamentary regime posed no substantial danger for the bourgeoisie, it was likely to bring a noticeable
curtailment of the army's dominant power position within the
state, and particularly of the power of all who were holding key
posts in the repressive police, army and secret service apparatuses.
In view of the degree of popular support for the parliamentary
forces of the Left and Centre, the army had no hope of retaining
its hold over parliament by intervening on the hustings. If it wished
to ensure its continued dominance, it was obliged to intervene
unilaterally and directly.
C. A more detailed look at post-war developments within the
Greek armed forces will help to give a better understanding of the
form the military's reaction took to the threat from below.
During the German occupation of Greece, the bulk of the
142
armed forces, together with the government in exile, were in

Egypt. There, the internal disputes between republican and
monarchist officers reached such a pitch that after a failed 'mutiny'
of left-wing officers, the British (under whose high command the
Greek forces were serving) intervened and purged all left-wing
and republican officers. It was this ideologically 'cleansed' army
which in December 1944 played a decisive role alongside the
British forces in the Battle of Athens (the first round of the civil
war). It was the same army that managed in 1949, with American
help this time, to deal the final blow to the communist forces and
establish an exclusionist quasi-parliamentary regime. Within the
army itself, considerable influence was wielded by a group of
hardline anti-communist officers who as far back as 1944 had
formed the secret organisation IDEA (The Sacred Bond of Greek
Officers). 127 During the civil war and its aftermath these officers sabotaged all attempts to reintegrate republican officers
into the army128 and made sure that the same exclusionist,
anti-communist spirit prevailed in all the other branches of the
state apparatus. 129 Moreover, in May 1951, after Marshal Papagos's resignation from his army post, the IDEA officers staged an
unsuccessful coup with the ultimate aim of establishing a military
dictatorship under Papagos's leadership. However, given that the
latter had decided to gain political power by electoral means, he
persuaded the officers to go back to their barracks and saw to it
that their short-lived insubordination was not severely punished.
After the electoral success of Papagos's Greek Rally and the
consolidation of the guided-democracy model, the IDEA group
went through a period of quiescence. It was reactivated after the
1958 electoral success of EDA; and at about the same time, within
the IDEA organisation, a core of activist officers under the
leadership of the future dictator George Papadopoulos, formed a
more tightly knit subgroup called EENA (Union of Young Greek
Officers). 130 This group participated actively in the rigging of the
1961 election results. As a matter of fact, the secretary of the
secret committee set up by the general staff to organise the
electoral fraud was George Papadopoulos.
When George Papandreou had become prime minister he
included in his liberalisation policies the removal of Papadopoulos
and some of his associates from both their key posts in the Central
Information Service and from strategic military units near the
143
capital. He posted them to frontier areas or to Cyprus, and at the

same time tried to staff the Central Information Service with
officers of a more liberal/democratic orientation. These mild
measures were quite insufficient to dislocate the IDEA group,
however. The displaced officers counter-attacked by using their
networks within the army to create a climate of crisis and
anti-communist hysteria. So, for instance, in June 1965 George
Papadopoulos announced the discovery of an alleged communist
plot to put a certain number of tanks in his frontier unit out of
action. Although the supposed sabotage was eventually proved a
false alarm, Papadopoulos was not punished. A more serious
anti-communist fear-raising campaign had begun in May 1965
when a number of anti-IDEA officers were accused of having
formed a subversive group called ASPIDA, which was claimed to
be led by the prime minister's son Andreas. 131
Since these developments within the armed forces were constantly jeopardising the standing of the government, premier
George Papandreou finally decided to replace the chief of the
defence staff, General Ghennimatas, a man close to the palace and
suspected of being involved in, or at least aware of, the military's
anti-government machinations. Defence minister Garoufalias, a
close friend of the King's, refused to replace Ghennimatas. At this
point the prime minister decided to take over the defence portfolio
himself, the King refused to let him do so, and a long constitutional crisis ensued which was eventually to lead to the 1967 military
take-over. Given the impasse created by the King's refusal,
George Papandreou resigned in July 1965, expecting that the King
would either give way or dissolve parliament and call new elections. Instead, the King decided to form a new government from
the existing parliament by persuading (mainly through bribery) a
number of conservatively-oriented and/or disgruntled Centre Union MPs to abandon their leader and join the opposition. After
repeated attempts the operation succeeded and a new government
was formed, headed by ex-Centre Union politician Stephanopoulos and supported by the conservatives and members of
parliament who had abandoned George Papandreou, the socalled apostates (renegades).
As far as the army was concerned, the new Stephanopoulos
government hastened to re-man the Central Intelligence Service
with more 'reliable' right-wing officers as well as to reactivate the
144
ASPIDA affair. However in late 1966 (as mentioned already),

after a series of complicated manoeuvres, George Papandreou and
the moderate conservative leader Kanellopoulos (on whose parliamentary support the Stephanopoulos government was based)
came to a secret agreement on the organisation of new elections
set for May 1967. But while the major civilian leaders turned their
energies towards the forthcoming electoral contest, the military,
alarmed by the growing popular mobilisation and the Right's poor
election prospects, were united in their determination to prevent
George Papandreou's return to power. On the question of tactics
they were seriously divided. While a group of senior officers who
were close to the palace (the so-called Big junta) were vacillating
about the timing of the coup and about whether to proceed even
without the King's consent 132 another group of lower-ranking
officers (the Little junta), headed by George Papadopoulos decided to present the King and their military superiors with a fait
accompli and staged a coup on 21 April 1967.
This division between two groups of officers motivated by the
same goal had its roots in the army's post-war promotion structure. Having to expand quickly to turn out sufficient numbers of
officers for the civil war, the army had been obliged to lower its
standards for cadet recruits; after the civil war was over, the
resulting abundance of officers spelled a serious bottleneck in
promotions. (A reliable computation gives 2000 captains in the
Greek armed forces prior to the 1967 coup. Since only 1()(}..150
could be promoted each year, those with the lowest seniority
would have to wait 15 years before it was their turn. The problem
was similar if not quite so acute for the higher ranks. In these
circumstances 200 captains had formed an association for the
advancement of their professional interests.) Aside from the
promotional difficulties creating ill-feeling in the armed forces,
there was also a distinct social gulf now between the older,
higher-ranking officers and the younger lower-ranking officers. 133
In this general atmosphere of discontent and frustration the
existence of two juntas side by side becomes rather more understandable, as does the ease with which George Papadopoulos
could manipulate the grievances of the junior officers and make
himself their leader in a conspiracy that ended with the establishment of a seven-year long military dictatorship.
145
Looking finally at the various interests behind the coup that

were responsible for the situation in which a group of colonels
could abolish parliamentary democracy, these consisted not only
of domestic groups (such as the extreme Right, part of the
conservative press, certain financial interests, and so on), but also
included the Americans who had not only built up the system of
guided democracy but had also maintained strong links with the
Greek armed forces. It is beyond question for instance that there
was a strong connection between the IDEA officers and the CIA,
whose local organisation had set up and trained the Greek
Intelligence Service, a service in which, as already mentioned,
IDEA officers held key posts. Moreover, the IDEA officers,
through training courses in the United States, had been amply
exposed to the new anti-insurgency ideology of 'internal security
and national development', an ideology that had been developed
after the Cuban revolution and fashioned to sensitize the military,
especially in Latin America but also elsewhere, to new 'professional' values and new ways of combatting the growing communist
threat (see below). 134
The above notwithstanding, it must be emphasised that it was
neither domestic civilian groups nor external agencies that played
the major role in bringing about the 1967 coup in Greece. As I
have pointed out already, the country's economically privileged
groups, although displeased, had no reason to be alarmed by
George Papandreou's reformist socio-economic policies; neither
had the Americans, despite their annoyance with Andreas Papandreou's anti NATO, anti-American rhetoric, any interest in taking
the enormous risk of withdrawing their traditional support from
the palace and the army's top brass in favour of a clique of obscure
lower-ranking officers. To repeat, the major impetus for the
establishment of the military dictatorship came from within the
army and was directly linked to the threat that the growing
political mobilisation posed to its dominance within the prevailing
exclusionist/incorporative relations of domination.
3.4.2 Argentina
A. In Argentina the re-establishment of a system of guided
democracy with the usual incorporative/exclusionist control
146
mechanisms occurred after Peron's fall in 1955. The first provisional president to succeed Peron was retired General Lonardi, leader
of the 1951 unsuccessful anti-Peronist conspiracy. Lonardi
adopted the motto 'Neither victors nor vanquished' and his
strategy, like General Plastiras's conciliatory policies vis-a-vis the
communists in Greece in the early 1950s, was to adopt a moderate
attitude towards the Peronists, in the hope of eventually overcoming the deep Peronist/anti-Peronist split within both the army and
civil society. So Lonardi for instance purged several Peronist
generals and admirals, but refused to extend the purge to the
lower-ranking officers. His moderate policies were opposed,
however, by a number of groups within the military who were in
the ascendancy in the early post-Peron period. These constituted
the so-called Faction of 1951, (consisting of those die-hard antiPeronist officers who had been persecuted because of their involvement in the failed insurrection of 1951), a group of very
influential navy officers who played a crucial role in the overthrow
of Peron, and a large number of junior officers who saw antiPeronism as a means of climbing up in the army hierarchy. 135
These forces, in association with civilian groups, were determined
to eradicate Peronism utterly from Argentina's public life. They
demanded a much more drastic purge of Peronists than Lonardi
was willing to make and, using his illness as a pretext, succeeded in
replacing him by the less moderate General Aramburu.
General Aramburu pursued the de-Peronisation course vigorously. This led to both of the Peronist parties being outlawed, to
the persecution of Peronist trade unionists, and to the massive
purge of Peronists from the armed forces, the civil service and the
judiciary. 136 At the same time the anti-Peronist Aramburu, just
like the anti-communist Papagos in Greece a little earlier, was for
the establishment of a 'guided' parliamentary regime rather than
for a straightforward dictatorship. 137 In this respect his position
was strengthened after the suppression of a Peronist officers'
insurrection in June 1956 which resulted in the execution of 27
rebel leaders. This unprecedentedly harsh treatment of rebellious
officers helped Aramburu overcome the opposition not only from
the moderates but also from those extreme anti-Peronist officers
(the so-called gorillas) who were hostile to any form of parliamentary rule.
147
Having consolidated his position, Aramburu proceeded to

announce (in October 1956) the election of a constituent assembly
in July 1957. Given the earlier re-establishment of 'restricted' civil
liberties, non-Peronist parties were allowed to organise and compete relatively freely. Those of the parties which accepted Aramburu's plan for a return to quasi-parliamentary rule obtained a
clear majority in the July 1957 elections. The ensuing constitutent
assembly abrogated Peron's constitution and ratified an earlier
decree by the provisional government which re-established the
1853 constitution. The road was now open for the organisation of
the 1958 national elections which were to put an end to Aramburu's rule.
With respect to economic policies now, Aramburu adopted
Raul Prebisch's austerity proposals for economic recovery. 138
These included the removal of price and exchange controls, a
drastic reduction in state expenditures, the de-nationalisation of
major public enterprises, affiliation to the World Bank and the
International Monetary Fund, and the encouragement of foreign
investments and wage restraints.
Finally the Aramburu administration took steps that, after
Peron's radical opposition in the 1940s to the United States'
hemispheric policies, 139 brought Argentina fully back into the
'inter-American family' by promoting close co-operation with the
Americans in both the political and military fields. So in April
1956, after eight years of delay, Argentina finally ratified the
Bogota Charter of the Organisation of American States, while at
the same time she hastened to strengthen her links with the US
military. In fact, during this period
Argentina displayed an unprecedented willingness to cooperate with the United States Armed Forces. In quick succession it signed an agreement for an Air Force mission to assist its
aeronautical development, sellt a ninety-man Army mission to
study American techniques, and authorised its Navy to conduct
joint anti-submarine exercises with an American task force. 140
If Argentina had been out of line with Chile and the other Latin
American countries vis-a-vis the United States' cold-war policies

during Peron's early rule, she partly caught up with the others
148
when Peron was forced to change his policies in the early 1950s,
and was fully re-integrated into the inter-American security system
during the Lonari-Aramburu governments. By the end of Aramburu's provisional administration the guided-democracy model,
complete with its strong American connections, was in full operation in the Argentine republic.
B. It was in this context of exclusionist/incorporative
controls that the first presidential elections of the post-Peron
era took place in May 1958. They were won by Arturo Frondizi,
the leader of the Radical Intransigents (UCRI), after a secret
agreement with the exiled Peron who urged his followers to
support the UCRI.
Frondizi had plans for an ambitious industrialisation programme
based on the creation of favourable conditions for attracting
massive foreign capital. Such a project would, it was hoped,
transcend both the agrarian oligarchy's anti-industrialisation 'export-oriented' model, and the anti-foreign capital bias of Peronist
industrialisation. However, once in office Frondizi was faced by a
severe economic crisis which had begun even before his inauguration and which forced him to opt, at least during the first years of
his presidency, for stability and austerity rather than growth. At
the same time he tried to adopt a less harsh, more conciliatory
stance towards Peronists. He granted an amnesty to exiled or
imprisoned followers of Peron, he allowed a number of them to
return to their trade-union and civil-service posts, and he even
recruited some neo-Peronists as advisers.
These timid steps towards the gradual reintegration of Peronists
into political life were enough to alarm the hard-core anti-Peronist
officers (the 'gorillas') who at that moment dominated the army.
Accordingly they embarked on a policy of close surveillance and
control of the president's every move. Frondizi's loss of support of
the working classes through his austerity policies had made him
increasingly dependent on the military, and the latter now imposed its will on the government by a stream of ultimatums that
quickly stripped the civil government of what legitimacy it had
enjoyed. 141
It must be remembered that these developments took place in
an ideological cold-war climate which, after the Cuban revolution,
deliberately encouraged the fear of communist subversion and
149
tried to equate Peronism with communism in the people's minds.

Obsessive fear of an omnipotent and omniscient 'Peronocommunism' did in fact become widespread in both military and upperclass circles. 142 It was in this context that Frondizi, partly in order
to fulfil some of his pledges to Peron, and partly in a desperate
attempt to legitimise his rule and break the political impasse in
which he found himself, decided to legalise the Peronist parties
and to allow them to participate in the provincial elections of
March 1962. Upsetting all his calculations, the Peronist forces,
permitted to canvass for the first time since 1955, won a resounding victory by gaining one-third of the votes and the majority of
the contested seats in the Chamber of Deputies. 143 These election
results speedily cut short Frondizi's political career. He was
deposed by a coup led by General Poggi.
The spectacular victory of the Peronist forces in their first
electoral contest since their leader's fall directly underlined the
exceptional fragility of the incorporative/exclusionist system of
controls that the so-called Liberating Revolution had established
in post-war Argentina. Contrary to the other cases of post-war
guided democracy examined above (where repressive parliamentarism functioned relatively smoothly for fairly long periods), in
Argentina the post-war incorporative/exclusionist system of controls began to crumble very early on. There were two reasons for
this. For one thing, Argentina had experienced high levels of
political participation/mobilisation much earlier than Chile or
Greece; 144 for another, Peron had managed during his rule to
create a large working-class movement which, despite serious
internal divisions, remained loyal to him even during his long
exile.
Against such a background, and in a climate of mass politics, the
attempt by the military and the dominant classes to bring back a
pluralistic system which could exclude such a sizeable and militant
political force from the political game was quite unrealistic. This
was particularly so in view of the fact that the model of foreign-led
industrialisation and development pursued more or less consistently since 1955 had further estranged the working classes in both the
towns and the countryside from their civilian and military rulers.
Neither the austere anti-inflationist policies implemented intermittently by Aramburu and Frondizi, nor the frequent devaluations of
the peso (which favoured exports and transferred resources to the
150
agrarian oligarchy), nor finally the de-nationalisations and opening-up of Argentina's economy to multinational capital did much
to shift working-class loyalties away from Peron. On the contrary:
the huge disruptions and growing inequalities typically associated
with the neo-liberal model of capital accumulation, 145 in combination with widespread measures of political repression, only radicalised the lower classes further and enhanced Peron's position, to
such an extent that all the abuses of the Peronist era were erased
from popular memory and the years of Peron's rule became
idealised as the workers' lost paradise. As Alain Rouquie puts it:
in the early 1960s
all kinds of virtues were ascribed to the Peronist era. In the
popular classes, the myth took the form of a judicialist Golden
Age, and this myth was reinforced by the upper classes who saw
the era as the period when 'workers could permit themselves
everything'. The Liberating Revolution, instead of dePeronising the workers, actually re-Peronised large popular
sectors which had been disappointed with Peron's second
presidency. 146
This process of re-Peronisation or of mounting radicalisation/
mobilisation in the late 1950s and 1960s can be seen quite clearly in
both the growing militancy of the trade-union movement and in a
definite shift of Peronism's political wing leftward: left-wing,
Marxist intellectuals were increasingly joining the movement, and
new cadres and young followers were determined to pursue a more
revolutionary path, engaging even in armed struggle against the
government and its military tutors.
Let us take the trade-union movement first. After Peron's
overthrow the unions were strictly under state tutelage and the
Peronist leadership was ousted. A few months later however,
when union elections were allowed to take place, the Peronists
under new leadership 147 managed to get control of the majority of
unions. In August 1957 the General Confederation of Labour
(CGT) was split into two main blocks: the '62 organisations'
controlled by Peronists, and the '32 organisations' controlled by
radicals, socialists and independents. 148 Despite this split the
Peronists managed easily during the Frondizi years to keep the
CGT under their control and to orient the trade-union movement
151
towards a highly politicised confrontational course. For instance,

in 1959, after some serious incidents which involved a strike and
the occupation by workers of a nationalised meat-packing plant
(the Lisandro de la Torre plant), the violent suppression of the
strike by the government and the declaration of a general strike by
the leaders of the '62 organisations', the latter were accused by
hardline Per6nist unionists of being too conciliatory vis-a-vis the
authorities. The executive council of the '62 organisations' was
therefore dissolved and replaced by more militant leaders. This
leftward shift by the trade-union leadership became clear in 1962
when the CGT national conference adopted a programme that
advocated, among other things, the nationalisation of the main
economic sectors, workers' control of industry, and the large-scale
expropriation without compensation of big landed property. 149
The radicalisation of the trade-union movement was reinforced
by and went hand in hand with similar developments in Per6nism's
political wing. In fact, soon after Peron's fall, part of the Per6nist
political movement began to orient itself towards the organisation
of underground, clandestine activities. In January 1956 the
National Command was formed, which tried to bring under the
same political banner all Per6nist groups that had survived the
1955 purges. (This National Command, led by radical/revolutionary Per6nists, played a crucial role in the 1959 general strike and
the subsequent radicalistion of the trade-union movement.) The
year 1958 saw the emergence of the Per6nist youth organisation,
the cadres of which were to play a major role in the organisation of
insurrectionary activities that gained momentum in the years
ahead. However, even during Frondizi's rule the first guerilla
training camps were established and the first guerilla group
launched in the Tucuman province, in imitation of the Cuban
insurrectionary model. 150
The radicalisation of the trade-union and political wings of
Per6nism, even if it did not always advance in linear fashion, was
very marked not only during the Frondizi period but acquired
further momentum during Arturo Illia's presidency. Although the
growing radicalisation of Per6nism and the alarming 1962 election
results had persuaded the military to oust Frondizi, given the
'legalists" ascendance within the army (see below), the armed
forces decided not to abandon their experiment in guided democracy just yet. In the eighteen months of Jose Maria Guido's
152
presidency (following Frondizi's overthrow) a series of decrees

dictated by the military imposed strict controls on the propaganda
activities of Per6nist parties, excluded the Union Popular (the
largest Per6nist party) from presidential and gobernatorial elections, and prohibited the collaboration of Per6nist candidates with
other political parties. 151 These severe restrictions, which led to
the Per6nists casting blank votes in protest, enabled the Radical
Populists to win the national elections of July 1963, and to get their
candidate Dr Illia into the presidential palace.
During the early years of his presidency Illia implemented a
more nationalist policy than his radical predecessor. He annulled
the petroleum contracts with foreign firms signed by Frondizi,
established exchange controls, and tried to redistribute income
from foreign to domestic investors. He also endeavoured to
expand production through the growth of money and credit supply
to the public sector, while at the same time increasing wages and
imposing price controls on consumer essentials. 152
These Per6n-like policies encouraged the more moderate wing
of the Per6nist trade-union movement to press for a more conciliatory, 'economistic' attitude towards the government. The hardline
militant trade unionists, however, with the active support of Peron
and of the highly radicalised political wing of Per6nism, managed
to prevail and to impose a programme of political confrontation
aimed at Illia's downfall. So in 1964 the CGT's executive committee, following the famous Plan de Lucha (formulated at the CGT's
1963 convention) launched a wave of strikes and factory occupations that seriously paralysed the country's economy. Between
May 21 and June 24 for instance, 11000 plants were occupied and
more than three million workers were actively drawn into this vast
political-mobilisation campaign. During the same year, the successor to the underground National Command was organised, the
Per6nist Revolutionary Movement (MRP), and adopted a programme of armed struggle 'for social liberation and the replacement
of capitalism by a socialist order'. 153
The growing radicalisation and popular appeal of Per6nism was
unequivocally shown in the March 1965 elections for 99 seats in the
Chamber of Deputies: the Per6nist Union Popular achieved a
spectacular victory by winning 36 seats. Together with the neoPer6nist parties, which achieved impressive results in the more
rural areas, 154 it controlled 52 votes in the Chamber and emerged
153
as the major opposition to the Popular Radicals. These results

clearly indicated that the anti-government left-wing forces had
regrouped under the banner of Peronism. By so doing they had
put an end to all hopes that a progressve/leftist non-Peronist party
could eventually draw Per6nists away from their charismatic
leader or his institutionalised movement. 'Quite the contrary, the
election showed that the Peronists were absorbing the Argentine
Left.>~ss
It is true of course that at this time a strong faction of Peronism,

under the leadership of the trade-union leader Augusto Vandor,
tried to detach the movement from Peron's personalistic, longdistance control. 156 Peron, however, managed to squash his lieutenant's rebelliousness when, in the April 1965 partial elections in
Mendoza, he undermined the Vandor-supported Peronist candidate, Seni Garcia. In a bold move two days before the election,
Peron urged his supporters not to vote for Garcia but for a
Peronist outsider Carvahin Nanclares; the latter eventually won
more votes than Vandor's favourite. This show of strength reestablished Peron's authority and cut short Vandor's attempts at
isolating Peron from his movement. Of course, the struggle
between orthodox Peronists and Vandoristas was not over after
the Mendoza events. In February 1966 for instance, Vandor, still
in control of the '62 organisations', managed to capture the CGT's
secretariat. The orthodox Peronists under the leadership of Jose
Alonso retaliated by creating the '62 unions loyal to Peron'.
These internal developments in Peronism aside, the important
point here is that Peron succeeded, from his exile, in keeping his
movement's various tendencies under his control; he also succeeded in mobilising the radicalised popular classes in a way which
created a direct threat not only to Illia's government but also to the
army's dominance. As far as Illia was concerned, his position
became increasingly precarious not only because of the March
1965 national election results, but also because of the growing
economic difficulties he had to contend with towards the middle
and at the end of 1965. Like Peron's in the early 1950s, Illia's
nationalist strategy for economic growth and social welfare began
to go wrong when soaring inflation dramatically reduced wage and
salary levels, and growing balance-of-payment difficulties rendered nationalist postures vis-a-vis foreign lenders untenable.
Unlike Peron, however, Illia could not impose severe austerity
154
measures in these difficult circumstances, and was unable to

restrain effectively wage-increase demands by the working class.
Unable to secure popular support or the co-operation of the
trade-union movement, he attempted to curb trade-union political
activities and to control their finances. However, IIlia's toughening
attitude towards the unions did not prevent high wage-claim
settlements, and did not succeed in controlling inflation. lilia not
only failed to win working-class support, he also alienated the
large salaried middle classes whose incomes were being undermined by the rising inftation. 157 If one adds to this the big
landowners' and industrialists' enduring hostility to lilia's nationalist policies, one obtains some idea of the degree of isolation in
which the government found itself at the beginning of 1966. In
these circumstances, and given the fact that lilia was as unsuccessful as Frondizi in 'taming' Per6nism, the road lay wide open for the
establishment of a new regime which would try to preserve the
military's dominance and to cope with the growing threat from
below by replacing the mixture of incorporative/exclusionist controls of the guided-democracy model with a straightforward exclusionist military dictatorship.
C. Having examined the form and trajectory of Argentina's
mobilisation/radicalisation process, let us now take a closer look at
the military and the way in which they dealt with the growing
threat from below.
After Frondizi's ascent to the presidency in 1958, the 'gorillas'
(or 'interventionists' as they came to be called), having failed to
obstruct a return to civilian rule, were quite determined to impose
severe limits on the president's powers. Their attempts were not
unsuccessful, especially after Frondizi started losing popular support towards the end of 1958. For example, from the end of 1958
until the end of 1959 there were acts of insubordination in all three
branches of the armed forces by anti-Per6nist officers who managed to impose their demands and to obtain the resignation of the
pro-Frondizi ministers of the air force, of war, and of the navy. 158
The zenith of the interventionists' power was reached in
September 1959 when General Toranzo Montero incited an army
rebellion which the then newly-appointed war minister General
Anaya, after gathering some loyalist troops, was determined to
suppress. Frondizi, however, wished to avoid an armed confronta-
155
tion and gave in to Montero by accepting the resignation of his war

minister. 159 After this clear anti-Per6nist victory, General Montero overreached himself by formulating a doctrine of 'total
military vigilance' over the civilian government, and by compiling
a list of demands that Frondizi found himself unwilling or unable
to meet in its entirety. By that time opposition within the armed
forces to the interventionists' constant planteos had begun to grow.
This 'legalist' opposition forced General Montero to concede
partial defeat in the matter of his 'list of demands'. Moreover, a
few months later, after Montero (enraged by Frondizi's Cuban
policy) had demanded the president's resignation, the legalists
managed to curb the General's power and forced him into
retirement. Montero's fall was a turning point, marking the
gradual rise to dominance of those officers who, although antiPer6nist, were against extreme factionalism and politicisation of
the officers' corps.
It must be emphasised, however, that the decline of die-hard
anti-Per6nists, and the rise to prominence of the 'legalists' in the
early 1960s, did not mean that the army was becoming less
anti-communist or more willing to reintegrate Per6nism into the
parliamentary system. In fact the legalists' rise coincided with the
marked radicalisation of the Per6nist movement and, on the
international level, with the intensification of the USA's efforts to
draw the Latin American military into an anti-Cuba and anticommunist crusade.
It is well known how, after the Cuban revolution and the
gradual spread of guerrilla activities in several Latin American
countries, the United States made systematic attempts to integrate
the Latin American armies more closely into an inter-American
system which would stress not merely the principle of collective
defence against external enemies (as did the 1947 Rio de Janeiro
Treaty), but also the principle of defence against internal subversion. The basic idea behind this was that, in addition to reforms
advocated by President Kennedy's scheme 'Alliance for Progress',
an effective way of fighting communist subversion in Latin America would be to reorient military training in these countries by
propagating a new type of military professionalism which would
focus more on anti-subversive skills and on the need for the armed
forces to be more directly involved in 'civic action' and in the
over-all process of 'national development'. 160
156
To realise this new strategy, the United States administration

not only stepped up material and financial assistance to the Latin
American military, 161 but also restructured and intensified courses
(offered in the United States and elsewhere) geared to training
officers from 'friendly' countries in the new anti-subversion
skills. 162 As far as Argentina is concerned, the growing radicalisation of the early 1960s was parallelled by a considerable increase in
the number of Argentine officers trained in the United States, and
in the spectacular expansion of US military assistance. 163 Since the
Argentine armed forces had been less insulated since 1930 than
their Chilean counterparts, they embraced the doctrine of the 'new
professionalism' with great enthusiasm, as well as the US Pentagon's attempts for the creation of a hemisphere-wide antisubversion network.
In view of the above it can be understood why the ascending
'legalist' faction in the Argentine army was not concerned merely
with the defence of legality and the maintenance of professional
discipline. If these officers were hostile to the decline of army
discipline and to the anti-Per6nist fanaticism of the interventionists, in the new circumstances they also became ardent anti-Cuba
crusaders, combining fanatical anti-communism with faith in technocratic rationality, professional autonomy, and western Christian
values.
These new attitudes had crystallised as early as 1962 when all
three branches of the military, infuriated with Frondizi's abstention from a resolution to exclude Cuba from the Organisation of
American States, had forced him to sever relations with Cuba. It
was a month later, when the Per6nists had registered huge gains in
the March 1962 elections, that after some manoeuvering the
armed forces proceeded to the overthrow of Frondizi. Yet despite
the fact that the interventionists played a crucial role in the ousting
of Frondizi, the legalists managed to avoid the imposition of a
military dictatorship and led the country to the July 1963 elections
which, with only one quarter of the votes cast, brought Dr Illia
into power.
Frondizi's fall did not terminate the conflict between legalists
and interventionists, or azules and colorados as they were now
called, and the battle continued unabated. After the government's
failure in August 1962 to discipline colorado officers demanding
the replacement of the moderate minister of war (General Loza),
September 1962 saw an armed confrontation between azules and
157
colorados. The former prevailed, and with the successful suppression a few months later of a navy-based colorado revolt, azul
dominance was fully consolidated.
This consolidation of azul power gave a large impetus to the
doctrine of 'new professionalism'. In fact, this doctrine was seen
by the all-powerful azul leader General Juan Carlos Ongania as an
effective means for overcoming the army's extreme factionalism
and for strengthening vertical lines of command. He therefore
introduced new methods of military training, with the emphasis on
both modern technology and human engineering. In 1964 an
agreement was signed with the United States on a programme of
military assistance, aimed at the modernisation of not only military
equipment but also army attitudes and organisation. Through
numerous US military missions in Argentina, systematic efforts
were made to gear the army organisation to the new requirements
of 'internal security and national development'. 164
At the same time, on the level of foreign policy the Argentine
military were increasingly oriented towards an internationalist,
anti-communist direction, viewing themselves and their Brazilian
counterparts as the gendarmerie of the subcontinent. Particularly
after the establishment of the 1964 military dictatorship in Brazil,
the Argentine military sought close co-operation with their northern colleagues. In August 1965, for example, General Ongania
made statements in Rio de Janeiro about the necessity of an
alliance between the Brazilian and Argentinian armed forces in an
effort to jointly combat communist subversion in South
America. 165
These developments clashed badly with lilia's nationalist/neutralist foreign-policy orientations. For example, the military were
furious with lilia's rejection of their proposal to send Argentinian
troops to join the US peace force in the Dominican Republic
during the 1965 crisis. 166 If one adds to this the military's exasperation with the government's incapacity to effectively crush Per6nism, one must acknowledge the growing incompatibility between
the military's anti-communist visions and the president's policies.
It is not surprising therefore that the electoral victory of the
Per6nist forces in 1965
sent tremors through the Argentine military. Judging from their

success in March 1965, the Per6nists might well capture the
major governorships, as well as more congressional strength in
158
the elections of 1967. Also the possibility of Peron as president

in 1969 could not be ruled out. For the military this was a
nightmare possibility. 167
In consequence General Ongania, alarmed by the election results
and by Illia's reluctance to support United States proposals for
setting up an inter-American peace force against subversion,
resigned over a minor matter: for not having been consulted about
the appointment of Brigadier General Castro. A few months later
Ongania staged a successful coup, Illia was ousted, and a longterm military dictatorship was established in June 1966.
Although civilian groups did very little to prevent the military
intervention, 168 there was no pre-arranged civil-military alliance
before the military proceeded to abolish parliamentarism: such an
arrangement would seriously have limited their room for manoeuvre in the post-coup period. It was rather, as in the case of
Greece, that although many groups contributed in various ways to
the dictatorial outcome, the major momentum for the intervention
and for the establishment of authoritarian rule came directly from
within the armed forces.
3.4.3 Chile
A. In Chile the contemporary foundations for the basic incorporative controls of the country's social and political system were
put down in the early 1930s, after the turbulent years (1924--32)
following the demise of oligarchic parliamentarism. Alessandri,
who came back to power in 1932 (this time without Ibanez's
tutelage), decided against rural unionisation and against the
application of Ibanez's labour code to agriculture. This crucial
decision laid the basis for four long decades of parliamentary
stability, a stability based on leaving the countryside to its oligarchic masters and playing radical politics in the cities and mining
centres. The prevention of rural unionisation actually constituted
an effective and long-lasting political bargain between the big
landlords and the urban-industrialised classes at the expense of the
peasantry. As B. Loveman puts it, the stability of Chilean
democracy was ultimately based
on the continued dominance of the landowners over the votes
and the political activity of their farm work force. This domi-
159
nance in turn depended upon maintenance of the hacienda

system through the prevention of rural unionisation and the
exclusion of outside political influences. 169
Not only radicals but even socialists and communists, despite
spasmodic attempts to change the 'social pact', for the greater part
supported such arrangements. For instance, during the Popular
Front period (1938-41), after a short time when socialists and
communists tried to mobilise the urban and rural population, all
popular front parties (including the communists) agreed to
'suspend' 170 legislation for rural unionisation. The situation was to
become much more restrictive still after the Second World War,
during Gabriel Gonzalez Videla's presidency (194&-52).
During the early post-war period Chile, like most other Latin
American countries, was strongly affected by the cold-war climate
generated by the growing tension between the two super-powers.
Although the United States' attention after the declaration of the
Truman Doctrine in 1947 focussed on Europe, there were also
systematic attempts to reinforce inter-American co-operation and
integration, politically as well as militarily, as a means of countering external communist aggression. As early as 1945 at the
Chapultepec Conference, an Act of inter-American mutual assistance was agreed, and this Act was converted into a full treaty in
Rio de Janeiro in 1947. Chile promptly signed the Rio treaty. Five
years later, in 1952, she also signed a military agreement with the
United States which inaugurated a long-term programme of
military assistance and co-operation. (In the 1950s a:nd early 1960s
Chile, together with Brazil, received more military aid per capita
of their populations than all other Latin American countries.) 171
On the internal front,
cold war intrigue made post-World War II Chilean politics a
confrontation zone for 'Communism' and the 'Free World'.
American policymakers considered the presence of communist
ministers in the Gonzalez Videla government to be dangerous,
and so allied with the Chilean Right in an active campaign to
weaken, then destroy, Marxist political parties and the labor
movement. 172
In implementing this strategy the Americans provided financial
assistance to the socialist factions of Chile's Labour Confederation
160
(the CTCH) which were trying to stop the growing communist

influence in the trade-union movement. Moreover, they made
badly needed financial assistance to the Chilean government
conditional on measures for combatting communism. Accordingly, Gonzalez Videla, prompted by US pressures, and alarmed by
communist electoral successes in the 1947 municipal elections as
well as by the growing communist-instigated agitation in the rural
areas and the mines, decided to expel communist ministers from
his cabinet. With mounting labour agitation, the president did not
hesitate to use the army and police to put down the labour unrest
in the mines. At the same time he gave himself extraordinary
powers through new legislation to leave him a free hand for
dealing with subversion. This enabled him to arrest and intern
thousands of communists in various camps all over the country.
(Pinochet, the future dictator, was commander of one of these
camps in the North.) Videla went one step further with the 'law for
the permanent defence of democracy' which outlawed the Communist Party. 173 This law, together with new anti-rural unionisation legislation also introduced during Videla's presidency, destroyed quite effectively the rural unions that had begun to emerge
in 1946 and 1947, and so restored the Chilean conservatives' hold
over the rural population.
All these developments contributed to the construction of
Chile's model of post-war guided democracy, a system of exclusionist/incorporative controls which, although not as accentuated
as that operating in post-civil war Greece, was going to function
quite successfully all through the 1950s to be challenged, as in
Greece, in the early 1960s.
In the Greek case, of course, the inter-war agrarian reforms and
the overwhelming prevalence of independent smallholders meant
that there was no threat to the existing relations of production in
the countryside. The major concern of the ruling groups in
post-civil war Greece was to keep those defeated in the civil war in
their subaltern place as second-class citizens. In Chile on the other
hand, the persistence and high concentration of big landed property, as well as the considerable fusion of landed, financial and
industrial interests, 174 meant that one major function of the
exclusion/incorporation mechanisms was to prevent the collective
organisation of the rural labour force. It was feared that such an
organisation could lead to a rural-urban working-class alliance
161
which (in view of the exceptionally strong political organisation of

the urban working classes) might threaten not merely the incorporative relations of domination, but the system of bourgeois rule
in its entirety.
Such obvious differences aside, what both control systems had
in common was the preoccupation with preventing an integrative
type of inclusion of the lower classes (especially in the rural areas)
in politics. Also, as a result of the increasingly higher levels of
industrialisation and urbanisation leading to rural radicalisation,
neither system was able after a certain point to maintain its
effectiveness without abolishing parliamentary rule altogether,
that is, without replacing its incorporative/exclusionist mixture of
controls by a straightforward, long-term dictatorial-exclusionist
system of rule.
B. In Chile, as in Greece, post-war economic and political
developments were such that they gradually undermined the
'social pact' and the related exclusionist/incorporative controls on
which the stability of the country's political system rested. So
during the second Ibanez presidency (1952-8) the forces of the
Left managed to strengthen their position by such significant
developments as the relative unification of the trade-union movement, the 1957 merger of the two socialist parties into the Partido
Socialista de Chile, and the consolidation of the socialistcommunist political alliance that had originally been formed
around Salvador Allende's 1952 presidential candidacy. 175
Moreover, towards the end of his second presidency Ibanez,
despite his generally repressive orientation, had to adopt certain
measures which were to weaken the over-all control system
considerably. Although he had extensively applied Videla's law
'for the permanent defence of democracy' to suppress workingclass agitation in the rural and urban areas, he was obliged to
abolish it in 1958. At the same time he helped to bring about
electoral reforms which, through such innovations as the imposition of the secret ballot, the introduction of penalties against
electoral fraud, and so on, greatly enfeebled the clientelistic hold
of landlords over their labour force. 176 Ibanez's electoral reforms
actually led to a situation where, for the first time in twentiethcentury Chile, the combined conservatives and liberals failed to
win one-third of the total vote; whereas the left alliance of the
162
Popular Action Front (FRAP) and Frei's newly created Christian

Democratic Party filled the vacuum by having attracted large
sections of the rapidly growing electorate.
From 1958 onwards, the well-known policies of Alessandri
(1958--64), Frei (1964-70) and Allende (1970-3), occurring in a
context of sluggish economic growth, continued relentlessly to
weaken the 'social pact' and the exclusionist/incorporative system
of political controls on which the social fabric of Chile had been
based, and dramatically increased processes of political participation and mobilisation. 177 Alessandri, under pressure from Kennedy's 'Alliance for Progress' programme, despite his conservative, anti-inflation policies introduced the land-reform law of 1962
which, in addition to its (limited) redistributive effects, further
liberated rural voters from landlord control. 178 Frei accentuated
these processes with more agrarian reform and redistributive
measures. The years of Frei's rule also saw extensive mobilisation
of the rural and urban populations as the Christian Democrats
tried to create a popular mass base. Frei not only encouraged town
dwellers and particularly the marginalised poor to set up neighbourhood councils and various self-help groups, he also urged
small landholders to form co-operatives (resulting in a proliferation of rural co-operatives, especially in the expropriated lands),
and he actively supported rural unionisation. Whereas in 1964 the
total trade-union membership of 270 000 included only 1600
members of rural unions, in 1970 there were 167 000 rural militants
and total union membership had exceeded the half million
mark. 179
As a striking result of such changes in the countryside, the
customary controls the conservative parties had been exercising
over the rural population declined drastically. A clear indication of
this is that after 1967 most rural municipalities had come to be
controlled by the Christian Democratic Party. The Christian
Democrats who were sitting on the communal councils were
peasants and agricultural workers. In consequence the peasantry
became much more militant. So the number of rural strikes
increased from 693 in 1967, to 1119 in 1969, and land seizures
jumped from a mere 9 in 1967, to 148 in 1969, and 456 in 1970. 180
Considering in addition the extremely intense urban agitation
among trade unionists and students makes one realise the degree
163
to which the traditional incorporatist/exclusionist controls were

being eroded during Frei's presidency.
All these attempts at redistribution and political liberalisation
were, however, accompanied by a slow-down in production and
high levels of inflation, and this cancelled most of the economic
benefits the working classes were receiving from Frei's 'revolution
in liberty'. 181 Contrary to Frei's expectation therefore, his endeavour to mobilise the masses in a populist-reformist direction
led less to a weakening of the left-wing forces than to a serious split
within the conservatives, that is, between those favouring redistribution and those favouring greater repression as a means for
coping with the rising tide of mobilisation and social unrest. It was
this split that was a major reason for Allende's Popular Unity
Alliance (UP) coming into power. 182
C. Allende's ascent to power was followed by a complex process
of escalating polarisation that brought mass mobilisation to a peak
and eventually led to the army's imposition of a brutal military
dictatorship in September 1973. Given Chile's previous record of
relatively stable parliamentary politics, it is worth looking at this
process more closely.
Allende inherited from Frei a high rate of inflation and a
slow-growth economy undergoing a recessionary phase. To revive
the economy and at the same time direct income redistribution
towards the underprivileged, he greatly increased wages and
salaries while imposing controls on prices. These policies increased
demand and, in view of industry's considerable unused capacity
and the existence of foreign-currency reserves, gave a major boost
to the flagging economy. This early economic success enabled
Allende's Popular Unity Alliance to perform well at the municipal
elections of April 1971. With this relative renewal of the popular
mandate Allende went on to implement the UP's more radical
policies for a major structural transformation of the economy.
These policies consisted of the acceleration and partial restructuring of Frei's agrarian-reform programme, and of the creation of a
relatively large social-property sector through the nationalisation
of copper, banking, and a number of other key enterprises. These
policies of structural transformation, implemented haphazardly
and in the context of the propertied classes' wholehearted sabot-
164
age, while at the same time the ultra-Left was encouraging the
workers to go beyond the government's programme by seizing
factories and farms directly, not unnaturally led to a steep drop in
production. Simultaneously, the UP's continuation of its highwages policies at a time when the over-all production of goods and
services was shrinking resulted in very high levels of inflation.
With the deterioration of the over-all economic conditions goods
became scarce, extensive black-market networks emerged, while
the United States' 'invisible blockade' was creating a desperate
balance-of-payments problem. 183
On the political level the situation was equally grim because of
the deep and paralysing cleavages within the Popular Unity
Alliance. The communists in the UP were advocating a more
cautious policy so as to consolidate the socialist advances, and
wished to build a broader base for the government with the
co-operation of the Christian Democratic Party (PDC). The more
radically oriented socialists on the other hand were pressing for a
deepening and acceleration of structural reforms, and for the
adoption of an intransigent attitude vis-a-vis all 'bourgeois' parties. Allende was in favour of the communist position, but he was
as unable or as unwilling to impose his views on the rapidly-rising
left wing of his party as he was to discipline the extreme-Left
forces operating outside the UP 184 This was the reason that,
although at the UP's meeting at Lo Curro in May 1972 the
communist orientation prevailed, the further deterioration of the
economy and the continuing sabotage of the governmental programme by the Right reinforced the socialists' intransigent position.
By the middle of 1972 talks between the Alliance and the
Christian Democrats broke down; as the latter started collaborating more closely with the right-wing National Party the government was rapidly losing the support of the middle and lowermiddle classes and the conservative offensive was gaining momentum. In fact, whereas during the first eighteen months of the UP
administration the political opposition had boycotted Allende's
policies primarily through congress and the judiciary, 185 afterwards it broadened its attacks by directly mobilising the disaffected lower-middle and middle classes. The first clear sign of
this new strategy was the famous 'empty pots' demonstration,
when middle-class housewives came out into the streets to protest
against rising prices and the deteriorating economic situation (this
165
took place during Fidel Castro's visit to Chile in December 1971).

A more serious instance of the conservative opposition's new
confrontational tactics was the prolonged lorry owners' strike of
October 1972 which severely disrupted the country's distribution
network.
This more aggressive right-wing posture naturally provoked
only more strongly intransigent tendencies in the Left, leading to
such measures as the rapid take-over of numerous factories, the
government placing the economy on a war footing and announcing
its intention of controlling the distribution of all agricultural
produce. At grass-roots level, left-wing radicalisation took
the form of the spontaneous emergence of large numbers of
popular organisations outside the control of the existing political
parties or trade-union organisations. Not only were organisations
initially set up by the UP revitalised, but a plethora of new
self-help groups sprang up as people were trying to cope with the
immediate problems of distribution and production in a situation
of seriously disrupted government services. While this type of local
organisation was proliferating, action committees emerged to form
these groups into larger units. For example, the famous industrial
cordons were co-ordinating the activities of local factory committees, and community commands were bringing together factory
and neighbourhood committees in the same geographical area.
The industrial cordons, which were the major form assumed by
'people's power', were
organisations which had developed outside the framework of
the CUT and which, on the whole, operated independently of
its political leadership. Their leaders were elected from the
shop-floor - not appointed by the government - and could be
changed very quickly if their political line did not please the
base - simply by a decision of the workers' assembly in any
factory to change its delegate to the cordon. Meetings of the
cordon itself were completely open. Anyone who wished to
could attend them and speak, although only delegates could
vote. 186
Towards the end of 1972 this escalation of confrontational politics
slowed down temporarily as government and opposition began
preparing for the elections. The March 1973 congressional elec-
166
tions, in which the UP increased its percentage of the total vote

from its 1970 figure of 36.2 to 43.4 per cent, destroyed the
opposition's hopes for gaining a two-thirds majority in congress
- a situation which would have enabled them to overthrow
Allende's government by constitutional means. After these
results the conservative opposition reactivated its sabotage policies
on all fronts. To the resurgence of strikes and the legislative
blocking of all governmental measures were added direct calls
from both major opposition parties for the army to step in and oust
the government. 187 On the other side of the political spectrum,
left-wing forces beyond Allende's control intensified the process of
factory seizures and at the same time incited the population to
armed resistance. It was when the polarisation reached its peak in
September 1973 that the military (after an abortive 'dress rehearsal' coup in June) stepped in to put a brutal end to Allende's road
to socialism.
D. Having examined how the relations of production and
domination were threatened by the process of increasing mobilisation/radicalisation, it remains to focus a little more closely on
developments in the military sphere which explain the armed
forces' reaction to this threat from below.
Chile's military (as mentioned in section 3.1 of this chapter),
while not occupying as dominant a position within the early
post-oligarchic polity as did Argentina's, frequently intervened in
politics through back-stage intrigues and a series of abortive,
predominantly anti-communist coups. 188 The post-oligarchic anticommunist tradition among the Chilean military was reactivated
after the Cuban revolution and later with Frei's victory in the 1964
elections. The rise of the extreme left-wing MIR organisation
(Movimiento de Izquierdo Revolucionaria) and the increasing
opposition from landowners to Frei's agrarian-reform programme
brought growing unrest among right-wing officers. This explains
several incidents during this period of brutal suppression of
strikers by the armed forces (operating with, and in some cases
without, governmental consent). For example, in November 1967
army and air-force patrols killed several people, including children, during a general strike in Santiago. Responsibility for these
events was ultimately watered down, and after the
167
... harassment of journalists and medical personnel, the whole

case ended quietly ... Rumours at this time held that the armed
forces had again operated very much on their own and were not
responsive to civilian authorities. 189
A clearer indication of unrest within the armed forces and their
involvement in politics was the Viaux incident in October 1969.
On this occasion, men and officers of the Tacna regiment, under
the leadership of General Roberto Viaux (who was connected with
the Linea Recta affair in the 1950s), barricaded themselves in at
their headquarters and took control of the main arsenal, the
non-commissioned officers' school, and the main recruiting centre.
Although their demands were primarily concerned with salaries
and the acquisition of military equipment, there is no doubt that
this type of 'military strike' was directly related to a more general
discontent among officers about Frei's reforms and the ensuing
political mobilisation/radicalisation of the masses. Frei's reaction
to this incident was to declare a state of emergency and to mobilise
units loyal to the government. He was immediately supported by
the Communist Party and the CUT, the latter calling for a general
strike and the occupation of factories and farms. The important
point here is that, despite Viaux's and his followers' failure, their
demands were eventually met by the government. Shortly after the
end of the affair the defence minister Marambio was made to
resign and very substantial increases were granted in the military's
salaries. 190
The last incident in civil-military relations during the Frei
administration was the abortive kidnapping of commander-inchief General Rene Schneider a few days before Allende's inauguration as president. The kidnapping, which it was discovered
later was engineered by extreme right-wing elements (it was
significant that Viaux was involved) in collaboration with foreign
agencies, was intended to create an explosive situation that would
eventually have pushed the military leadership to step in and stop
Allende's inauguration. However, the operation did not go
according to plan. General Schneider resisted, was shot, and died
a few days later. Subsequently, as the right-wing origins of the
undertaking became known, the anti-constitutionalist elements
within the army were put on the defensive and a phase of relative
168
calm in civil-military relations followed during the first year of

Allende's presidency. 191
Moreover, as soon as Allende took over he adopted a very
conciliatory policy vis-a-vis the military. For instance, he dropped
the UP's original electoral platform in respect of the armed forces
which had advocated important changes in their internal organisation (for example, a more open system of promotions, granting
NCOs and serving soldiers the right to vote, restrictions on the
participation of the armed forces 'in activities of interest to foreign
powers', 192 and so on). Allende now tried to interfere as little as
possible in the military's much cherished 'professional autonomy';
at the same time he tried to infuse the armed forces with a
developmentalist ideology.
Prior to Allende the Chilean officers, unlike their Argentinian
counterparts, were not in any marked way involved in the management of civilian organisations. 193 This is not to say they were not
exposed to the 'new professionalism' ideology and to the general
anti-Cuban campaign mounted by the US Pentagon in the 1960s. If
anything, the Chilean armed forces, despite their smaller size, in
the 1960s received (in absolute terms) more military aid from the
United States and had many more officers trained abroad than the
Argentine army. 194 In general, the United States' fear that Chile,
with her strong working-class movement, might develop into a
second Cuba led to a sharp acceleration of all types of aid
programmes both military and civilian, official and unofficial. 195
To give just a few examples, the US spent large sums of money to
fund Frei's electoral campaign during the 1962--64 period. During
Frei's presidency, more than $30 million were granted as military
aid, some of it going to the training and equipping of the rapidly
enlarged national police force and black-beret counter-insurgency
army units. 196
Nevertheless, until Allende's presidency Chilean officers did not
occupy important positions in civilian institutions on any large
scale. It was only under Allende that this pattern changed. In fact,
he not only started appointing retired officers to key state agencies
(for example, mining enterprises, the Chilean Development Corporation, the Agrarian Reform Project); he also at critical moments during his rule brought the military into his cabinet. For
example in November 1972 after the lorry owners' strike, he
formed a cabinet that included three military officers; and in June
169
1973 after Colonel Roberto Souper's abortive military coup he

again invited the military to participate in his administration, but
this time they refused to join his cabinet.
These moves, which Allende saw as a means to strengthen his
regime, were interpreted by his left-wing allies as an
embourgeoisement of the government; military officers on the
other hand saw them as a means of co-optation and of destroying
the army's professional neutrality. It was particularly when the
military realised that, neither its members in the cabinet, nor
Allende himself, could control the left-wing elements in the UP
and the extreme-Left forces outside it, that they started to regard
the close co-operation between the government and the top army
leadership with great suspicion. They interpreted it as a serious
threat to the army's autonomy, and as an attempt by the Marxist
parties to implicate military officers in responsibility for developments (for example, the factory seizures) they were not in fact
allowed to control.
The military's perception of a threat from above was coupled, of
course, with the perception of a still more alarming threat from
below. As the polarisation between the government and the
opposition reached an extreme, and as political organisations were
emerging that were beyond trade-union or government control, it
is not surprising that the majority of the officers saw these
developments as a direct threat to their power position in the state.
This becomes even more obvious if one remembers that 'people's
power' did not simply involve neighbourhood and factory committees but also para-military organisations aiming to create 'dual
power'. It is quite true that the distribution of arms to civilians was
not on as large a scale as the opposition was claiming
vociferously. 197 But the increasing political violence perpetrated
by both extreme Left and extreme Right during the last months of
Allende's rule and the alarmist cries of the conservative press were
enough to create the impression in the military that there were
para-military forces strong enough to challenge their monopoly
control of the means of coercion. This impression could only be
reinforced by left-wing attempts at 'consciousness-raising' among
the NCOs and the army's rank and file. 198
All these developments led to a situation where the 'institutionalist', pro-coup officers managed to draw into their ranks the
large number of undecided officers, so isolating their 'constitu-
170
tionalist' opponents. 199 The constitutionalist top leadership was

then forced to resign and this resignation allowed the golpistas to
take over without any formal breach in the hierarchical principle
of military command. 200 In consequence the situation in Chile
showed none of the sudden disruption of the bureaucratic chain of
command that the colonels' coup created in Greece.
On the other hand, although it was in all three ACG countries
that right-wing forces helped to bring about conditions favourable
to the establishment of military rule, it was only in Chile that their
contribution was both massive as well as decisive. This is not to say
that the Chilean coup was a joint enterprise based on a definite
military-civilian alliance. As noted earlier, the conservative political forces (especially those around the National Party and the
Fatherland and Liberty organisation), although linked in various
ways with anti~communist officers, did not establish the type of
partnership with the military which would have allowed them to
demand participation in the post-coup army-dictatorial government. As a matter of fact most conservative politicians were
hoping and expecting that, after establishing 'law and order', the
military would quickly return the reins of power into their hands.
The fact that they did not is another indication of the relative
autonomy of the armed forces vis-a-vis civilian economic and
political interests. The Chilean armed forces therefore, as those of
Greece and Argentina, acted as a relatively independent force,
unfettered by binding commitments to civilian politicians.
3.4.4 Conclusion
The above analysis of developments leading to the establishment of military dictatorships in Greece, Argentina and Chile,
although rather sketchy, does make clear certain common features
in the political trajectories of these countries. In all three societies
two basic long-term phases can be clearly identified:
(a) In the initial phase there is the construction and/or consolidation of a system of guided democracy, consisting of a mixture
of exclusionist/incorporative controls aimed at preventing the
horizontal/autonomous integration of the popular classes into
the rapidly expanding political arena. In all three countries
this system of controls acquired its major characteristics
during the broadening of political participation in the interA.
171
war, early post-oligarchic period, and was reconstituted in

more accentuated form in the post-war years, in the late 1940s
for Greece and Chile, and the mid-1950s for Argentina.
(b) In the subsequent phase such control mechanisms were undermined by rapid processes of industrialisation and urbanisation
which, by creating huge economic and political inequalities,
aggravated economic and political contradictions and, on the
level of collective action, accentuated the polarisation between dominant and dominated groups. Of course the radicalisation of the latter, which became very marked in the 1960s,
took different forms in each of the three countries. In Greece
it mainly manifested itself through a decisive strengthening of
the centre, centre-left political forces as clientelistic/exclusionist controls began to weaken, especially so in the countryside;
in Chile this radicalisation was expressed in the rapid unionisation of rural workers and urban marginals, a development that
was related to the dramatic electoral decline of the conservative, liberal and even radical parties. In Argentina, radicalisation meant the reactivation and reorientation of Per6nism. In
all three societies these developments indicated a decisive shift
to the left of the political spectrum, a shift that posed a serious
threat to the incorporative/exclusionist character of the prevailing relations of domination and in Chile endangered also
the relations of production.
This growing threat was more or less directly related to military
intervention and the imposition of long-term dictatorial rule. This
is quite obvious from the army's role in the ACG post-oligarchic
political systems. As outlined earlier, in Argentina and Greece the
post-war military had a key role in establishing and maintaining
the incorporative/exclusionist character of the relations of domination, and not merely in the general sense of any military apparatus
within a capitalist formation being the ultimate guarantor of
bourgeois order, but in the sense of directly monitoring the
parliamentary system and frequently interfering (through veto
power and otherwise) in the political decision-making process. In
Chile, where army dominance was less pronounced, the popular
threat to the army's power position took a more indirect form.
Because the economically dominant classes, and particularly those
conservative political forces that no longer had a realistic chance of
maintaining or regaining electoral control of the countryside, were
172
the most directly affected by the popular mobilisation, they

frantically and ultimately successfully endeavoured to create a
situation of economic chaos which the military perceived as
jeopardising not only their own power position but also the whole
fabric of bourgeois society.
It should be stressed, however, that notwithstanding the considerable variation in the military's power position in ACG (more
pronounced in Argentina, less so in Chile and Greece- see section
3.1.2 earlier in this chapter), the establishment of long-term
military dictatorships in all three countries becomes most understandable when seen in the context of (a) the growing contradiction between high levels of political radicalisation and incorporative/exclusionist mechanisms of political control, and (b) in terms
of the impact of this growing contradiction on the position of the
military within the prevailing mode of domination.
To repeat: a popular threat to the existing relations of domination was a more or less direct threat to the army's position within
the state. It is also important to stress that in all three cases the
army's intervention to stem the growing threat from below was not
merely or even primarily in order to safeguard capitalist interests,
nor because military officers were of middle-class origin; the army
intervened in order to safeguard its own interests which transcended concerns over pay or military expenditures and were
primarily power interests. As F. Nunn has put it in discussing the
Chilean case:
In lieu of considerations based on democratic traditions, the
military tradition of apolitical behavior, or the military as an
agent of class interests, we face the task (however unsavory it
may appear to some) of considering military professional
interests per se. Where tangible evidence of extra-professional
interests is lacking, we need to consider that political action can
be based on very narrowly conceived, highly militaristic values
and interests. 201
Such interests are threatened in several ways by high levels of
popular mobilisation. Not only can popular mobilisation endanger
the position of the armed forces' top personnel but, more importantly, it can also challenge the tutelary functions of the military
institution as such. For instance, as outlined earlier, in Greece the
173
rising liberal forces of George Papandreou's party directly challenged the highly interventionist role played by the monarchy and
the army in Greek politics, and that challenge was directly related
to the 1967 coup. Even more seriously, in Argentina
army officers of all political lines perceived Castro's avowal of
Marxist-Leninism, his acceptance of Soviet assistance and the
support expressed for Fidel in Argentine student, intellectual
and worker circles, as a threat not only to national security, but
to the very existence of the armed forces. The destruction of the
pre-Castro Cuban officer corps at the hands of Fidel's firing
squads did not pass unnoticed. 202
Finally in Chile it was not only that the Left was trying in various
ways to infiltrate the armed forces and to politicise the rank and
file, but as the economic and political crisis was reaching its climax,
elements within the Popular Unity party were calling for the
formation of workers' militias. Naturally the military perceived
such developments as a direct threat to its power position in the
state, and this led to a situation where the 'constitutionalist'
officers lost the support they had previously enjoyed among their
colleagues.
Since such threats to the army's power position are, at the
post-war stage of political participation, linked to structural rather
than merely conjunctural developments, short military interventions
of the inter-war type are no longer effective. It is only by the
establishment of more permanent and institutionalised exclusionist controls (that is, by the establishment of authoritarian dictatorial regimes) that at least some factions within the army can hope
to hold on to the type of privileged power position they had
acquired in the early post-oligarchic period.
B. A final point that should be emphasised is that how the
military perceived the threat and how they reacted to it cannot be
explained exclusively in terms of internal political developments and
contradictions. Although these latter were indeed crucial, two
other factors were also highly relevant: the state of the economy
and, on the international level, the strong organisational and
ideological links that the United States had established with the
ACG military.
174
To begin with the latter, the ideology of the 'new professionalism' to which the Argentinian, Chilean, and to a lesser extent the
Greek military were exposed through US military missions and
training programmes affected both the way in which they saw the
threat from below and the manner in which they sought to cope
with it and defend their power interests. Especially those aspects
of the new ideology that emphasised the need for 'civic action' as a
fundamental prerequisite for dealing with the 'internal enemy' are
highly relevant for an understanding of the military's new determination not to follow up their intervention with a quick return to
barracks but to impose directly their own long-term solution to the
problems of 'internal security and national development'.
I have mentioned already that the new ideology deliberately
inculcated the need for the military to go beyond their traditional
concern with martial skills and know-how and to move into the
larger sphere of over-all socio-economic development. Since the
acquisition of expertise in such fields necessarily breaks the
'old-fashioned' distinction between the military's professionalisation
and political involvement, from the perspective of the new ideology, professional expertise in civil action and national development
leads directly and unavoidably to politics. If the politicians are
seen to hinder national development through the pursuit of
demagogic/inflationary policies, they are undermining the foundations of both internal security and external defence and, from this
point of view, the business of governing cannot be left to their
'corrupt' and self-interested machinations. Of course, the notion
of politicians being corrupt and incapable of ruling in the national
interest had always been popular among the military in ACG.
What the 'new professionalism' did was to give such ideas a new
form and impetus by resolving the dilemma of professional
neutrality v. political involvement and by emphasising the need for
the military to acquire new skills in the fields of development and
human engineering. This new training enhances the military's
confidence in their capacity to rule. The doctrine of the 'new
professionalism' teaches not only that, free from clientelistic ties
the military officers are in a better position to impose the 'correct'
policies, but also that their new broader training and the managerial ethos associated with it enables them to tackle the complex
problems involved in 'national' development.
175
It is not difficult to see how this new technocratic and developmentalist mentality that the ACG military acquired from their
dealings with the North Americans will have looked to them as
making complete sense in the context of the import-substitution
difficulties they were facing in the late 1950s and early 1960s,
particularly so in the Latin American southern cone countries
(see 3.2 above).
It is quite true that in southern Latin America the importsubstitution crisis of the 1950s was greatly exacerbating the
contradictions between growth and distribution. During the 'easy'
period of import-substitution industrialisation between the wars,
neither popular demands for higher wages nor high inflation rates
seemed seriously to block industrial growth - hence the blossoming in several Latin American countries (including those under
consideration here) of populist formulas which, with some degree
of short-term success, promised economic growth combined with
social justice.Z03 Towards the end of the 1950s however, such
formulas, although they did not disappear, had fewer chances of
even short-term success. Even Peron (as already pointed out) was
forced by the difficult economic circumstances to a sudden volte
face, sacrificing redistribution and his anti-foreign, nationalistic
attitudes on the altar of economic recovery and further industrial
growth.
In other words, the easy inter-war phase of industrialisation
provided an economic setting that, without 'creating' or 'causing'
populism, prepared the soil for its eventual flourishing. This
changed, however, with the emergence of the import-substitution
difficulties of the 1950s. The slow-down in manufacturing investments on the one hand, and the more massive and aggressive
popular demands for redistribution on the other made private
investors increasingly reluctant to involve themselves in technologically more complex and economically less certain industrial
ventures.
These growing contradictions were an obvious object lesson for
the military for applying what they had learnt abroad to the
analysis of their economy's predicament at home. Their new
technological, managerial ethos made them sensitive to the 'irrationality' of giving in to popular pressures for higher wages and
more welfare. In fact, with respect to southern Latin America
176
there is evidence that the military were not immune to the

economic ideologies of the day, either before or after the establishment of dictatorial rule. As A. Hirschman has clearly
shown, 204 there was widespread disillusionment in Latin America
in the early 1960s with the ultra-Keynesian ideology of the ECLA
school. When at the end of the easy phase of import-substitution
industrialisation economic contradictions became more acute,
there was a growing ideological shift, encouraged by Chicagotrained social scientists, towards stricter monetarist policies of
economic growth. It is certain that the military did not remain
unaffected by such ideological currents.
In Greece neither the 'new professionalism' among military
officers nor the import-substitution difficulties were as marked as
in the Latin American southern cone countries. Although most of
the officers involved in the 1967 coup had received training in the
United States, and although there is empirical evidence of a shift
in the officers' attitudes towards a more managerial/technocratic
orientation, 205 the army's preoccupation with 'civic action' and
'national development' was not as intense or as obvious to the
general public as in Argentina for instance. This might be due
partly to the fact that the difficulties associated with the end of the
easy import-substitution phase had not yet, at least not up to the
influx of foreign capital in the early 1960s, taken acute form.
Moreover, the fiercely anti-communist officers who played a key
role in the coup did not need to project a new technocratic/
developmentalist ethos in order to justify their intervention. Given
the civil-war legacy in Greece, whipping up old fears of an
imminent communist threat seemed to be quite sufficient to the
officers involved.
Despite these differences between Greece and the Latin
American cone countries, I would like to stress once more that
what all three countries had in common was a growing
popular mobilisation that, more or less directly, threatened the
military's position within the state. I consider this common element
more important than the 'new professionalism' ideology and
the import-substitution difficulties for understanding the rise of
military dictatorships in ACG. As regards, for instance, the
military's new professional orientations, it goes without saying that
the Argentinian, Chilean and Greek military were not mere
puppets totally brainwashed or blindly obeying orders from the
177
Pentagon or State Department. Neither were the ACG military,

particularly in Greece and Argentina, pushed towards imposition
of military rule by powerful capitalist pressure groups or in order
to safeguard the 'expanded reproduction of capital'. As I have
emphasised repeatedly, the basic reason for the rise of the ACG
dictatorial regimes lay primarily in the contradiction between high
levels of political mobilisation and the incorporative/exclusionist
controls, a contradiction leading to a type of polarisation that in
the end threatened the military's power position within the polity.
It is from the perspective of this fundamental contradiction that
one should view both the varying degrees of attraction for the
military of the new anti-insurgency ideologies developed in the
United States, as well as their growing sensitivity to the need for
structural transformation of the economy through the disciplining
of labour, a discipline which would better induce the collaboration
of foreign and indigenous capital. These ideologies gave the
military, especially in Latin America, a broader and more acceptable justification for their intervention; and above all they provided clues on how to go about defending their power interests on
a long-term basis, that is, by establishing more permanent exclusionist controls in the name of higher economic rationality and
greater national security.
How the above interpretation relates to other theories on the
rise of post-war military dictatorships will be discussed in the next
chapter.
3.5 Postscript: a short note on the structure and dynamics of the
post-war military dictatorships in Argentina, Chile and Greece
Although it goes beyond the scope of this study to discuss the
structural characteristics and long-term trajectories of the post-war
military regimes in ACG, I would like in this postscript to offer a
few brief remarks on this problem area.
A. Concerning first the basic structural features of these regimes:
unlike military interventions in ACG during the inter-war years,
which were meant and indeed tended to be of a short duration,
post-war military rule not only lasts longer but is much more
institutionalised. Once in direct governmental control, the military
178
attempt to drastically rearrange a whole set of institutions in civil

society. For instance, a complex system of repressive mechanisms
regulates the press and other mass media, the courts, trade unions,
schools, and so on, in ways which far surpass in sophistication and
systematic thoroughness the more haphazard repressive arrangements of the inter-war military interventions.
A major feature of such military regimes is the prominent role
given to various 'neutrally oriented experts' or 'technicos' for
rendering operational and translating into precise institutional
practices the broad exclusionist goals of their military patrons.
Their expertise is applied to areas which vary from economic
policy and 'national development' to the improvement and systematisation of torture or counter-insurgency techniques, hence the
term bureaucratic authoritarianism often used in the literature to
characterise the post-war military dictatorships, not only in Argentina and Chile but in several other semi-peripheral societies in
Latin America and elsewhere. 206 This term correctly designates
the formal rationality (systematicity, predictability, and so on)
which characterises the various repressive controls, a rationality
that clearly distinguishes post-war military dictatorships from
'sultanistic' types of authoritarian rule. 207
Although the setting up of such complex and all-pervasive
repressive institutional structures differentiate post-war military
dictatorships from pre-war attempts at direct rule by the armed
forces, one should be careful not to identify them with fascist or
even quasi-fascist regimes. In fact, none of the post-war military
dictatorships in ACG managed to build up organisations for mass
mobilisation and support on the pattern or scale of Nazi Germany
or even Mussolini's Italy. In those two regimes (and especially so
in Germany) the state managed to penetrate civil society and to
mobilise the masses to a degree previously unknown in bourgeois
societies. Such a vast mobilisation presupposes not only that
fractions of the former ruling classes give their active support to
the regime; it also presupposes large-scale popular support in both
towns and countryside, with the help of which the fascist leaders
are able to squash most centres of opposition within civil society.
Nowhere in the parliamentary semi-periphery have post-war
military dictatorships achieved this type of quasi-totalitarian control. In that sense Juan Linz is quite correct when he speaks about
a certain degree of social (not political) pluralism as one of the
179
distinguishing marks of bureaucratic-authoritarian regimes. 208

Another and related feature of such regimes is their lack of a
coherent ideology which could be used as a basis for mass
mobilisation. There are two explanations for this lack: (i) the
existence of limited social pluralism forces these regimes to dilute
their ideological orientations in order to appeal to the largest
possible audiences (at least at the level of rhetorical
pronouncements); 209 and (ii) the mainly defensive and reactionary
reasons for military intervention (for example, the maintenance of
the army's privileged power position within the state) do not lend
themselves to the elaboration of ideological themes capable of
rousing the masses. Moreover, the heightened need for attracting
foreign capital diminishes the possibility of successfully developing
the anti-imperialism, anti-USA themes so skilfully manipulated
by Latin American populist leaders in the past few decades.
Finally, in parallel with the military's loose, partly defensive
anti-communist ideologies failing to provide the preconditions for
the development of a solid mass base, their over-all pro-capital or
orthodox economic policies - which presuppose for their success a
strict control of wage demands and the creation of a favourable
climate for private investments - are equally unsuccessful in this
respect.
These brief considerations of some salient features of post-war
military regimes bring us to the politically vital issue of their
prospects for long-term survival and institutionalisation.
B. If quasi-parliamentary forms of politics in ACG are unstable,
the dictatorial-military forms of political rule are too, at least at
the present stage of mass politics. This becomes obvious if one
considers what has been said above about these countries' (by
third-world standards) relatively strong civil societies and the
inability of the military, despite their relative dominance within
the state, to consolidate their position irreversibly.
In fact, other things being equal, the long-term institutionalisation of military dictatorial regimes is fairly easy in social formations which have not yet reached high levels of political participation/mobilisation, and/or where civil society is very weak (as for
instance in several African countries). At the stage of mass politics
on the other hand, and in a context of relatively strong and well
organised urban middle classes, the long-term institutionalisation
180
of military rule, even when supported by large fractions of the

bourgeoisie, is a much more difficult enterprise. Such a long-term
consolidation of the army's dominance over civilian forces would
require going beyond the type of control that the mere use of
repressive state apparatuses can give. It would require either the
creation of some sort of organic/corporatistic framework within
which various institutional interests are allowed some very limited
representation under strict state 'guidance'; and/or the building up
of a mass party through which the people are mobilised and
brought into politics in a fascist or quasi-fascist manner. 210
However, neither way of going beyond the typical military
bureaucratic authoritarianism has much chance of success in ACG
at their present stage of capital accumulation and mass involvement. The external obstacles to such solutions are related to the
fact that, after the defeat of the Axis forces, fascist and organic/
corporatistic models have been discredited as an alternative to free
party representation; and to the fact that the present-day heavy
reliance of these countries on foreign capital makes it much more
difficult to drum up popular support and legitimation by means of
xenophobic, chauvinistic nationalism or by adopting isolationist
policies of capitalist growth.Z 11 Moreover, internally the necessity
of controlling inflation and keeping down wages does not make it
any easier to gain the support of large sections of the rural or
urban lower classes, or to build up mass organisations providing
grass-roots support for authoritarian rule. Yet without such mass
organisations the military cannot consolidate their dominance
vis-a-vis recalcitrant civilian forces (trade unions, left-wing parties,
those political representatives of the economically dominant classes who resent their displacement by the military and their loss of
political power, and so on). The only way to shatter such forces is
through a mass party, which, however, the military seem unable
and/or unwilling to build up.
In addition, the relatively long parliamentary/democratic traditions and their strong ideological roots among large sections of the
population in ACG oblige the military to legitimise their intervention as an attempt at redressing democracy and establishing a
strong and 'healthy' parliamentary system. The more they prolong
their dictatorial rule, therefore, the more they are subjected to
pressures, internally as well as from the international community,
for adhering to their initial proclamations and for withdrawing
181
from the direct exercise of governmental power. As pressures

from below and abroad keep mounting, cleavages develop within
the ruling military groups: between those who favour a withdrawal
or a limited liberalisation as a means of safeguarding army unity
and prestige, and those who want to maintain or reinforce
dictatorial controls. As such intra-junta rifts widen, those political
groups which had been linked with parliamentary institutions and
which were displaced by the junta will try to aggravate the split
and do their best to mobilise the population in an effort to win
back their previous dominant positions. This type of structurally
generated process becomes sharply accentuated in a crisis,
whether the cause is economic, a question of 'succession, a
military defeat or whatever. Such a crisis invariably exacerbates
fractional divisions and strengthens popular pressures for democratic change.
The above remarks should not, of course, lead one to underestimate the importance of variations in the structure and dynamics of
military dictatorships in ACG. In Greece, for instance, given that
military intervention was primarily in reaction to a strictly political
threat (the relations of productions being in no serious danger, and
economically dominant classes' support of the junta only lukewarm), the organisational foundations of the dictatorship were
more fragile than in Chile; hence its sudden demise with the
Cyprus fiasco in 1974. 212 In Chile on the other hand, the fact that
the establishment of the dictatorship was in reaction to a more
general threat meant that the dominant classes did not merely
acquiesce in dictatorial rule but were actively involved in bringing
it about; hence the greater consolidation of the Chilean dictatorship and its systematic and large-scale application of violent
repression. 213 Finally in Argentina, the formidable force of the
Per6nist movement on the one hand, and the exceptionally
dominant position of the army within the state on the other,
explain the post-war alternations of quasi-parliamentary and military-dictatorial forms of rule as the military and civilian forces each
fail to establish a more permanent modus vivendi.
C. In terms of the over-all trajectory of parliamentary democracy in ACG, the pattern these three countries have in common is an
initial, relatively stable 214 phase of oligarchic parliamentary rule
(second half of the nineteenth century) followed (in the first half of
182
the twentieth) by a more unstable parliamentary phase of broader

political participation; this phase was punctuated by frequent
covert or overt army interventions of short duration, interventions
which did not aim at or succeed in seriously altering the dynamic
and basic parameters of the parliamentary game. Finally in the
post-war period, given a further dramatic broadening of the
political system (partly due to the radicalisation of the masses,
particularly in the rural areas), the long parliamentary history of
these three countries was much more seriously interrupted - this
time by the imposition of fairly long military dictatorships.
In sum, as one moves from the nineteenth-century restrictive
system of oligarchic parliamentarism through the limited inter-war
broadening of political participation to the mass politics of the
post-war period, the functioning of parliamentary democracy in
ACG becomes more problematic and the breakdowns more
disruptive and traumatic. This is not because the massive entrance
of new groups into politics ipso facto makes parliamentary democracy more fragile and precarious. As I have argued in several
parts of this book, when the new entrants are brought into a
political arena in circumstances where the problems of distribution
of economic rewards and political rights have been tackled relatively successfully, then mass participation consolidates democratic pluralism rather than endangering it. If on the other hand
there has been a relative failure in solving either or both the
economic and political distribution problems, then parliamentarydemocratic institutions do become precarious at the stage of mass
politics.
As I argued in Chapter 1, in the semi-periphery the inter-war
broadening of political participation, by incorporating rather than
integrating new groups into politics, led to a highly unequal
distribution of power between leaders and led. It was against this
background of a failed distribution of political rights that these
societies had to cope in the post-war years with the further
broadening of political participation, in a context moreover of
foreign-led industrialisation which exacerbated the existing economic inequalities.
Nevertheless, one must not belittle the importance and resilience of parliamentary institutions in ACG. Unlike most thirdworld capitalist societies where parliamentary democracy is either
absent or plays a merely decorative role, parliamentary institutions in these countries were not, and (in post-dictatorial Greece
183
and Argentina) are not, only decorative. Despite the system of

exclusionist/incorporative controls and the army's role within it,
parliamentary institutions in ACG create a degree of political
pluralism and of political freedom that is by no means negligible.
Moreover, despite the growing difficulties that mass participation
and increasing inequalities have created for the functioning of
parliamentary rule in ACG, these countries' long democratic
traditions and the relative strength of their civilian forces make the
expectation of a long-term decline and eventual disappearance of
parliamentary democracy anything but a certainty.
There is one last point worth making in relation to future
political developments in ACG. The fact that neither a military
dictatorship of the bureaucratic-authoritarian type, nor a fully
open parliamentary-democratic regime of the western European
type has high chances of long-term institutionalisation does not
mean that these societies are bound to experience a constant
alternation between open/democratic and closed/dictatorial regime forms. Although, as the Argentinian case shows, this is a
distinct possibility, there is no reason to exclude the alternative of
some or all of them (including Argentina) eventually establishing
more permanent political arrangements on a point somewhere
between the fully 'open' and fully 'closed' solutions.
Such a stable solution might, for instance, involve the maintenance of a pluralistic parliamentary regime with a simultaneous
increase in the authoritarian tendencies of the state. 215 This
increase in state authoritarianism might take a number of forms:
for example a further strengthening of the army's regulatory role,
the serious deepening of incorporative controls by civilian dominant groups, the dramatic expansion of the executive's regulatory
powers at the expense of those of the legislative and/or the
judiciary, and so on. It is not possible at this stage to give a full
blueprint of the possible political outcomes, nor is it the purpose of
this work to make predictions. Moreover, the generalisations
formulated in this chapter are of course again in no way intended
as 'iron laws', but simply as indications of certain structural
tendencies which may be neutralised by a host of more or less
conjunctural developments (wars, super-power politics, and so on).
Finally, one should never underestimate the reflexive/voluntaristic
dimension of social life and the fact that, in certain conditions,
collective awareness of 'structural tendencies' may constitute a
serious step towards their reversal or transcendence.
Chapter4
Praetorianism, Bureaucratic
Authoritarianism and Problems of
Reductionism in Marxist Political
Theory
Since the military's intervention in politics has been a major focus

of attention in both Chapters 1 and 3, this second theoretical
chapter will begin by placing my substantive conclusions into the
context of theories on the high frequency of army interventions in
the polities of peripheral and semi-peripheral capitalist societies.
After that more general discussion I shall focus specifically on
theories which try to account for the post-war rise of military,
bureaucratic-authoritarian regimes in Latin America and elsewhere.
4.1 Military interventions and theories of political modernisation

During both the nineteenth and twentieth century the parliamentary/democratic regimes under consideration here (with perhaps
the partial exception of Chile) have been seriously disrupted by
frequent, overt or covert endeavours by civilian as well as military
leaders to bring the armed forces directly into the political
decision-making process. If one searches for general attempts to
explain the reasons for the markedly praetorian tendencies in
these and most other late-late industrialising capitalist societies,
leaving aside theories that simply present an ad hoc list of 'factors'
or what they call correlations between democratic institutions and
184
185
various variables (literacy, communications, and so on) 1 , 1t IS

mainly political scientists working within the Parsonianfunctionalist tradition who have provided theoretically coherent
accounts.
Within this tradition, the current trend is to reject the naive
neo-evolutionism of early modernisation theories which optimistically linked modernisation/industrialisation with parliamentary
democracy, and stable civilian rule, as the 'developing' societies,
by means of the world diffusion of western culture and technology,
moved up the evolutionary ladder. In view of the repeated failure
of democratic-parliamentary institutions to take strong roots in
several rapidly industrialising countries, the present mood is much
more sombre and pessimistic. 2 The main reason for the failure of
democracy is seen to be the incompatibility, the lack of fit,
between imported western political institutions and the preexisting indigenous ones. 3
Another related theme, and one which is extensively elaborated
by S. Huntington (the major writer in this field), explains the
fragility and frequent disruption of parliamentary rule in modernising polities in terms of the weak institutionalisation of
political controls at a time of growing political mobilisation and
participation of the masses in politics. He argues that it is precisely
this which accounts for the praetorian tendencies of modernising
polities, that is, for the frequent intervention of the military and
their use of force to impose the kind of order that weak political
institutions are unable to provide. 4
The systemic incompatibility between weak political institutionalisation and high levels of political participation is surely
indisputable. The problem is how to explain the emergence of
such incompatibilities, as well as the intricate ways in which
these contradictions or incompatibilities are linked with military
interventions and frequent regime alternations. It is exactly in this
area that present theories have serious drawbacks. Since the
Parsonian functionalist framework does not lend itself to a full
consideration of the relations of production and of classes, 5 these
incompatibilities are either left unaccounted for, or explained in
purely cultural terms. 6 If economic factors are brought into the
analysis at all, they are dealt with in so systemic and neutral a
manner that the really fundamental dimensions of the problem are
given insufficient attention.
A.
186
I would like to substantiate these critical remarks by a closer

look at S. Huntington's theory on modernising polities, the most
theoretically coherent and comprehensive on the subject.
B. Trying to account for the violence and instability that
characterise the modernising societies, Huntington sees this as
the product of rapid social change and the rapid mobilisation of
new groups into politics, coupled with the slow development of
political institutions ... The political instability of Asia, Africa
and Latin America derives precisely from the failure to meet
this condition: equality of political participation is growing
much more rapidly than 'the art of association' together. 7
If one asks the crucial question why the 'art of association' is weak
in modernising societies, Huntington, by refusing to distinguish

capitalist from non-capitalist modernising societies, can give only a
highly unconvincing answer. As is the case with most functionalist
theorists, his main focus is on cultural explanations. So we learn
that the political cultures of most modernising societies are
marked by suspicion, jealousy and latent and actual hostility
towards everyone who is not a member of the family ... These
characteristics are found in many cultures, their most extensive
manifestations perhaps being in the Arab world and in Latin
America. 8
According to Huntington, this cultural mistrust leads to an
incapacity to build strong institutions and organisations on the
political level. 9 If this were really so, one wonders why noncapitalist modernising societies (for example, North Korea,
Vietnam) have invariably managed to overcome this so-called
cultural mistrust and achieved strong political institutionalisation.
Have these societies been exceptional in that they never had a
political culture of mistrust in the first place?
Huntington is no more convincing when he puts forward other,
less cultural explanations for the weak development of the 'art of
association' in developing countries. For instance, he argues that
another fundamental reason is that in the developing countries the
speed of economic change and of modernisation generally is much
higher than it was in western Europe, and that therefore political
187
institutionalisation has been lagging behind. But again, it might be

asked, why is there no such lagging behind in the modernising
collectivist non-capitalist regimes which, as is well known, have
equally high rates of economic change and modernisation? Why
for instance did post-war Bulgaria, whose industrialisation was
both abrupt and quite spectacular, have a more stable political
regime than her southern neighbour? Speaking more generally,
why has not a single post-war communist regime been overthrown
by the military from within?
The answers to such questions are so obvious that only the
systematic bias of an inadequate conceptual framework can fail to
see them. The major reason that certain modernising polities have
stable political institutions and others do not has less to do with the
type of political culture, the rate of economic change, or the
degree of development of the 'art of association', and so on, than
with the relations of production and domination in these societies.
With respect to the relations of production, for instance, I have
tried to show extensively how in the parliamentary semi-periphery
both the timing and the structure of capitalist industrialisation are
of paramount importance for explaining the regime instability and
military interventions in these societies. With respect to the
relations of domination, I have tried to show how their incorporative character made them incapable of coping with the rising levels
of political participation and radicalisation, an incapacity that is
the result less of their weak institutionalisation than of their
restrictive character. In other words, 'weak political institutionalisation' is not so much due to lack of political skills or insufficient
political socialisation, as it is due to the incorporative/exclusionist
character of the political control mechanisms. These mechanisms,
operating in a parliamentary context, become increasingly unacceptable and unstable as the lower classes are brought into the
political arena.
It seems to me that any attempt to brush aside these two
fundamental dimensions of the problem (which connect it with the
relations of production and domination), any attempt to indiscriminately lump together capitalist and non-capitalist, nontotalitarian and quasi-totalitarian societies can only lead to highly
dubious conclusions.
Huntington does, of course, accept that the communist states
have shown remarkable regime stability, and did so even before
their full industrialisation. Yet as far as he is concerned, relations
188
of production or even of domination have little relevance to

differences in regime stability. He speaks extensively about the
fact that communist regimes have 'a stronger political development', that communist leaders have developed greater organisational skills, and so on, 10 but nowhere does he say anything about
the relations of production or the relations of domination as
playing a role in explaining such differences.
C. It is not surprising therefore that Huntington's remedies for
the political instability to be found in modernising polities consist
of closing the gap between rapid growth and sluggish political
development by further strengthening the political institutions,
and by a concerted effort (presumably by political scientists and
other 'experts') to teach the politically less developed countries
new skills in the 'art of association' and political leadership. This
kind of policy, and the more general theoretical analysis that
underlies it, misses the fundamental point that, however much
skill is developed in the 'art of association', and however ingenious
the tactics and strategies of the political leadership concerned, firm
civilian control of the military on the pattern of some western
European parliamentary democracies is highly unlikely in a social
formation characterised by the late and restricted development of
capitalism, and by incorporative relations of domination. On the
other hand, the combination in communist regimes of drastic
economic levelling with the imposition of quasi-totalitarian political controls leads to relative regime stability/immobility, regardless
of the 'arts of association' or the political skills that national
leaders may or may not have. (This is not to deny of course that, in
the case of the eastern European satellites, the fear of the Red
Army's intervention provides another fundamental reason for
regime stability.)
What needs stressing here is that this non-perception of what is
essential and what is peripheral in analysing the political instability
of capitalist third-world societies is due not merely to the author's
political conservatism. It is largely the result of a conceptual
framework (that is, the Parsonian one) 11 which systematically
avoids an examination of those institutionalised relations which
indicate differential control of the means of production, domination, coercion, and so on; a conceptual framework which invariably conducts the analysis of social processes from the point of
189
view of the 'system' and its functional requirements, rather than

from the point of view of collective actors and their antagonistic
interests and struggles over scarce economic and political resources. This conceptual framework focusses on the production of
wealth and power, rather than on its distribution among collective
agencies; it focusses on how and to what extent resources are
mobilised for the achievement of 'system goals', it does not ask
who controls such resources and whose interests lie behind societal
or system goals. 12
In view of the above, the use of systemic 'neutral' concepts by
functionalist political sociologists like Huntington is only to be
expected. It is not at all surprising that he constantly talks about
modernisation or industrialisation in general, rather than about
capitalist or non-capitalist industrialisation; that he refers to
'industrial' or 'modern' societies without much emphasis on
whether these societies are capitalist or collectivist in their economic organisation; that he talks about the polity's goals or requirements without seriously considering the goals or projects of
specific interest groups (especially the dominated ones). So political order/stability as a systemic goal is regarded as good in itself,
irrespective of the price that ordinary citizens may have to pay for
it. In this way systemic goals are more or less automatically
identified, in Hegelian fashion, with the general good, despite the
awareness that behind these systemic or societal goals lie the
power interests of the ruling elite. Huntington is actually quite
explicit on this point:
The public interest of the Soviet Union is approximated by the
institutional interests of the top organs of the Communist Party:
'What's good for the Presidium is good for the Soviet Union'. 13
4.2 The rise of bureaucratic-authoritarian regimes: notes on the
relevant literature
We shall now briefly examine some of the theoretical attempts that

have been made to explain the post-war emergence of military,
bureaucratic-authoritarian regimes. Although most of these
attempts focus on Latin American societies, the arguments they
develop are also directly relevant to the Greek case.
190
In the review in chapter 3 of the post-war rise of military regimes

in Argentina, Chile and Greece, particular emphasis was laid on
the contradiction between high levels of political mobilisation/
radicalisation and the incorporative/exclusionist mechanisms of
control that crystallised in the post-oligarchic period. It was argued
that in the 1960s unprecedented levels of popular mobilisation
threatened the above system of controls and hence more or less
directly the power position of the military within the state, and
that this threat is one of the major reasons for the emergence of
bureaucratic-authoritarian regimes in these countries.
Now in the literature which tries to explain the post-war rise of
bureaucratic-authoritarian military regimes, those writings which
go beyond a mere historiographical account of events typically
stress the economic contradictions and struggles which are supposed to be at their root. So if the army is not crudely seen as a
passive instrument in the hands of the indigenous or foreign
bourgeoisie, 14 explanations of the establishment of post-war military dictatorships put the stress on the predominantly middle-class
origin of the officers or the changing functional requirements of
capitalist industrialisation.
A. J. Nun's analysis, for instance, exemplifies the first
approach. 15 For him, the relatively early professionalisation of the
Latin American military meant the large-scale admission of the
middle classes into the military profession, and hence the constitution of the army into a distinct stronghold of middle-class rule:
This early professionalisation had two important social consequences: first, as has been indicated above, the middle class
was admitted to the career of arms through the creation of
military academies; and secondly, in contrast to its organisational weakness, this class was now allied to a sector with a
remarkable degree of institutional cohesion and articulation. In
other words, the armed forces became one of the few important
institutions controlled by the middle class. 16
Moreover, since the Latin American middle classes lacked hegemony, they had to make use of the army to achieve their political
goals, that is, first to break the oligarchy's monopoly of power, and
then, when they were in power, to safeguard middle-class rule
191
from the popular threats arising from the mass mobilisation of the
post-war era.
In as far as Nun does not show precisely how the middle class
exercises its control over the military, he turns this class into a
crude anthropomorphic entity pulling the strings behind the backs
of the military. Yet it has been emphasised repeatedly - and
rightly so - that the social origins of the personnel of both military
and non-military state apparatuses cannot in themselves explain
the over-all policies of the state. 17 To give a very obvious example:
the lower middle-class origins of the colonels who instituted the
seven-year military dictatorship in Greece certainly do not explain
the dictators' active support of big indigenous and foreign capital
at the expense of not only the working classes but the lower middle
classes as well. 18 In other words, of much greater relevance than
the officers' class origin is the over-all socio-political context in
which the military find themselves at a certain historical conjuncture. Where the officers' own interests do have relevance to the
explanation of military intervention in politics, such interests are
not merely derived from an awareness of their father's or their
family's class position. They derive much more from the actual
position of the army within the state, from the divisions within the
military organisation and, more generally from the complex
relationship between those who control the means of production
and those who control the means of coercion and/or administration.
B. If Nun's explanation tends to adopt what might be called an
economistic-'instrumentalist' view of the army apparatus, O'Donnel's theory of the establishment of bureaucratic-authoritarian
military regimes 19 comes close to an equally unsatisfactory economistic-structuraVfunctional view, a view which links in more or less
direct manner the establishment of long-term army rule with
primarily the changing functional requirements of the more modernised Latin American capitalist economies. Very briefly, in his
view, given the exhaustion of the relatively easy importsubstitution phase and the incapacity of the state and the indigenous bourgeoisie to carry the process any farther, the 'deepening'
of the industrialisation process demands a massive attraction of
foreign capital and a drastic curtailment of popular demands for
higher wages and welfare. Bureaucratic-authoritarian regimes, he
192
holds, can thus be seen as a response to such changing structural

requirements of the economy. As the 'popular sector' (which was
politically activated during the preceding import-substitution/
populist phase) continues to articulate demands which the economic system can no longer cope with, a coalition of technocratic
elements from within and outside the armed forces is formed
which, with the support of entrepreneurs and other middle-class
fractions, imposes an exclusionary bureaucratic-authoritarian regime. By suppressing popular demands in the interests of technocratic rationality, such a dictatorial regime hopes to create the
proper conditions for the attractions of foreign capital and the
deepening of the industrialisation process. 20
O'Donnel has been criticized extensively, most of his critics
focussing on his theory's functionalist/teleological character. They
argue that by explaining military and political developments in
terms of the changing functional requirements of industrial capitalism, his analysis is clearly teleological. To avoid this charge
O'Donnel should have shown, not only that these countries
actually did require in-depth industrial development, but also that
the military had perceived the necessity for it. There is ample
evidence, according to the critics, that this was not the case; that in
several of the countries O'Donnel examines there was no awareness either before or after the imposition of dictatorial rule of any
need to deepen the industrialisation process; or conversely, that
some societies (for example, Brazil) had initiated a process of
deepening industrialisation much before the imposition of
bureaucratic-authoritarian rule. 21
It seems to me, however, that O'Donnel could defend himself
against the charge of teleology by arguing (as many theorists both
Marxist and non-Marxist have already done) 22 that the functionalist approach is not necessarily teleological; and that it is legitimate
to talk about functional requirements and their constraining
impact on society even if actors are not clearly aware of such
constraints. For instance, the concept of latent functions and
dysfunctions, as elaborated long ago by Merton, points precisely
to the possibility that certain constraints or 'requirements' of the
social system can be very real, even when social participants are
unaware of their real nature and the way in which they evolve. 23
Concerning the specific example we are concerned with here,
O'Donnel could argue (and to some extent he has done so in his
attempt to cope with the above criticism) that his theory does not
193
try to establish a direct, causal relationship between changing

economic requirements and bureaucratic-authoritarian regimes;
that it simply postulates an 'elective affinity' between the recent
multinational phase of capitalist industrialisation and the type of
exclusionist military dictatorships that have sprung up in several
post-war semi-peripheral societies.
If it is agreed that the contradiction between economic accumulation and distribution becomes more acute after the easy phase
of import-substitution; and if it is further agreed that growing
political mobilisation leads, sooner or later, to the articulation of
economic demands which those who hold power cannot meet
without creating investment strikes and paralysing levels of inflation, then it follows that inclusionary authoritarian attempts
(of the populist type, for instance) would have fewer chances
of success than exclusionary types of authoritarianism (see
Chapter 3.4.4). In circumstances where pre-coup attempts to
combine growth with distribution have failed (for example,
Allende's in Chile), and given the army's greater capacity, once in
power, to disregard distribution and opt one-sidedly for rapid
growth, the chances for a successful military intervention and of
the imposition of long-term military rule are very high. From this
point of view the military might well succeed for a lengthy period
in holding on to dictatorial power, even if they are not much
concerned with considerations of 'deepening industrialisation'.
These 'deepening' requirements are still relevant, however, in that
a lack of deepening, or insufficient deepening after the exhaustion
of the easy import-substitution phase may create a crisis situation,
and this in turn may enhance the power position of those groups
that, for whatever reason, are in favour of an authoritarian
solution to the crisis.
In other words, this watered-down version of O'Donnel's theory
simply points out that these regimes are compatible with the
demands of the late, more difficult phase of import-substitution
industrialisation; it does not imply that they emerged because of
this functional compatibility. The assumption of 'elective affinity'
between further industrialisation and dictatorial controls can still
be questioned, of course, but the theory can no longer be labelled
teleological.
In the light of the analysis developed in the previous chapter, I
think that a more effective criticism of O'Donnel's theory should
focus not on his functionalism but on his economism. The links he
194
tries to establish between changing economic requirements and

the rise of bureaucratic-authoritarian regimes are too direct and
unmediated. This is not to say that one cannot provide any
structural explanation of the emergence of the post-war military
regimes in the semi-periphery, but rather that one must look for
structural explanations primarily within the politico-military sphere.
For it is not only the economic system that has reproduction
requirements relevant to a structural explanation of long-term
political developments. The polity too has its proper reproduction
requirements, 24 and it is precisely these which are more directly
relevant to the establishment of bureaucratic authoritarianism in
ACG. As argued above, it was rising levels of political mobilisation/activation that undermined the system of incorporative/exclusionist controls established in the post-oligarchic period, that is,
growing political forces that could not be accommodated within
the existing system of domination.
The ACG countries did not, therefore, simply have an importsubstitution crisis, they also had a quite distinct political crisis. If
the economic crisis was pointing to the need/requirement to
deepen the industrialisation process, the political crisis was pointing to the need to tighten politico-military controls as a means of
maintaining the existing relations of domination. In other words, a
more adequate explanation of the rise of military regimes in ACG
must take into account the changing reproduction requirements of
both the capitalist economy and the incorporative polity. In that
sense O'Donnel's theory is not wrong but incomplete. As I see it,
this incompleteness, this neglect of the specifically political dynamic, is particularly damaging since, at least in the case of Argentina and Greece, the major focus for understanding the rise of the
post-war dictatorships should be on the polity rather than the
economy. Although there is always a strong interconnection
between the economic and political aspects of the crisis, I would
argue that the crisis which led to the establishment of dictatorial
controls was primarily political. If this is acknowledged, then one
would certainly have had strong authoritarian tendencies in ACG
even without a crisis in these countries' economies- that is, even if
the 'deepening requirements' had not been imperative. 25
C. Three qualifications are in order here with reference to the
above arguments.
195
First, while criticising Nun's and O'Donnel's theories, I do not

wish to belittle the crucial importance of economic factors for
explaining the emergence of post-war military regimes. It is quite
clear that both class struggles and economic constraints/requirements have a great deal to do with the post-war military interventions in the semi-periphery. My objection to the above theories
concerns rather the type of linkages their authors have established
between the economy and the politico-military spheres. Being too
direct and economistic, these linkages distract from a proper
consideration of the strictly political determinations which mediate
between classes or modes of production and military-political
developments.
As a matter of fact, the explanation adopted in this study by no
means excludes economic determinations; it simply introduces
them into the analysis in a different manner. If it is accepted that
explaining the post-war military interventions must seriously take
into account the army's reaction to the growing political mobilisation that is threatening the incorporative/exclusionist character of
the relations of domination (and hence its own privileged power
position within the state), there are several ways in which economic forces enter into the analysis:
(a) The incorporative nature of the state-civil society relationship
in the semi-periphery is partly related to both the timing of
capitalist industrialisation (see Chapter 1.1) and its more
restrictive and unequal character (Chapter 3.2).
(b) The massive post-war mobilisation/radicalisation and the
further broadening of the political system are also strongly,
although not exclusively, related to the rapid and highly
unbalanced capitalist growth in these societies.
(c) While the 'deepening requirements' are not linked in a direct
cause-to-effect relationship with the establishment and consolidation of military-dictatorial regimes, one can still quite
legitimately view the economic difficulties associated with the
end of the easy import-substitution phase as providing a
ground favourable to the emergence of bureaucraticauthoritarian regimes (see Chapter 3.4.4).
Second, I would like to emphasise once more that pointing to an
'elective affinity' between bureaucratic-authoritarianism and functional/reproduction requirements (whether economic or political)
by no means provides a full explanation of the rise of military
196
regimes in post-war ACG. It merely indicates a set of structural

constraints which put more or less strict limits to actors' strategies
and collective projects. Given these limits, certain solutions for the
post-war politico-economic crises have more chances of success
than others - certain actors are in a more favourable position to
defend and promote their interests than others. If the problem is
stated in such presumptive terms, it becomes fairly clear that the
authoritarian solution to the crisis was by no means inevitable;
neither will it, when and if it occurs, have to take a dictatorial
form. In fact one could easily imagine a different scenario of
reaction in which the dominant groups handle the growing contradiction between high levels of radicalisation and incorporative/
exclusionist controls by authoritarian measures that do not abolish
parliamentary rule (for instance, by strengthening the role of the
executive or the president, by 'deepening' incorporative controls,
by reintroducing or extending further selective exclusionist measures, and so on).
It follows that to explain why the authoritarian reaction in ACG
took the bureaucratic-authoritarian form one has to go beyond the
basic political contradiction and look at such things as the receptivity of officers to the 'new professionalism' ideology, the specific
nature of the army's internal cleavages, the specific socioeconomic situation in which the military found themselves during
the period preceding the coup, and so on.
This brief enumeration of some secondary factors related to the
rise of dictatorial rule brings me to the third and final qualification/
methodological remark I would like to make. This concerns the
fact that it is all too easy to draw up long lists of factors at different
levels of abstraction. Historians, for instance, can point to circumstances which were unique to each specific country (for
example, the civil-war legacy in Greece); Latin Americanists can
point to factors specific to their hemisphere (for example, upperclass fears of a Cuban-type revolution); and political scientists can
make a long list of factors of a more general/universal nature (for
example, the inept behaviour of civilian leaders before the coups)
and so on. 26 The basic problem with such considerations is to show
how all these relevant factors or causes are connected with one
another. There are two main ways of establishing such connections, one atomistic, one holistic.
197
(a) One can adopt a highly positivistic methodology and establish

correlations between variables in an attempt to find the
specific weight of each factor involved. In this type of atomistic, context-less exercise the long-term historical development,
as well as the over-all socio-political structures in which the
variables are embedded, tends to be neglected and the variables are conceptualised as discrete billiard balls which maintain their shape and colour whatever the context in which they
are found. Although such an approach can sometimes provide
useful statistical information, most often, and especially when
it attempts to build a body of 'universal laws', it leads to
generalisations which are either inconclusive, in the sense that
they hold true only in certain conditions which, their social
context being neglected can never be determined; or, while
perfectly conclusive, they are platitudinous. 27
(b) The second way of connecting the various items on the
'laundry list' is, as this study has suggested, a long-term
historical-comparative approach which takes more seriously
into account the over-all context in which the variables are
lodged. Within such a holistic approach, specific political
outcomes (for example, the establishment of a military dictatorship) are not merely correlated in positivistic manner with
isolated variables; they are placed within an over-all
framework of economic and political contradictions and struggles, which facilitates the assessment of what is central and
what is peripheral to the argument, what is related to longterm structural trends, and what is accidental or the outcome
of conjunctural developments. As argued in Chapter 3 for
instance, in so far as the 'new professionalism' doctrine was
widely accepted among officers, particularly in Argentina and
Chile, this was not merely due to the clever training schemes
of the US military, but primarily to the changing structural
conditions in the countries involved (that is, the high level of
political mobilisation threatening the army's power position);
such conditions provided fertile ground for the propagation of
the ideology of the 'new professionalism' among certain
fractions of the military.
198
The same is true, of course, about the import-substitution

difficulties of the late 1950s and early 1960s. They too must be
seen in the context of those basic political contradictions and
conflicts the long-term historical trajectory of which this book
book has tried to trace. As noted already (see previous
chapter), in Greece import-substitution difficulties were only
loosely and indirectly linked with the basic political contradictions and conflicts that are crucial for understanding the rise of
the dictatorial regime, whereas these linkages were closer and
more direct in the Latin American cone countries. In more
general terms I have argued that in the Greek case, where the
threat from below was strictly political, economic contradictions and conflicts were only indirectly linked with the politico-military dynamic which led to the imposition of dictatorial
rule. At the other extreme, in Chile, the more global threat
from below meant a more direct link between economic and
political contradictions, the latter, however, still being central
for an understanding of the demise of parliamentary politics in
post-war Chile.
As the above examples indicate, the basic conceptual
framework put forward in this book, with its focus on the
historical evolution of basic political contradictions and conflicts, is useful for establishing adequate links between various
relevant factors, as well as for assessing the relative importance of each of them. Without such a framework, factors
become 'variables' to be correlated in context-less, positivistic
manner; or they become mere items in a laundry list of
disconnected 'causes'.
A full discussion and demonstration of the utility of a
non-atomistic, more holistic approach to the explanation of
the post-war military dictatorships would, of course, require
not a section or a chapter but a whole book to itself. Even so,
the brief substantive and methodological remarks made in this
section do suggest, for instance, that considerations like the
new 'professionalism', CIA or foreign-capital machinations,
and so on, though relevant, are definitely secondary: strong
trends towards military bureaucratic-authoritarian regimes
would have emerged in ACG even in the absence of these
factors. At the most these factors can explain the precise
199
moment or the specific form that a given military intervention

has taken.
To sum up, the thesis I have tried to develop here is that the
post-war military bureaucratic-authoritarian regimes in
Argentina, Chile and Greece have not been random phenomena; they have a primarily structural rather than conjunctural basis, they can be more or less directly linked to certain
common long-term historical developments of their economy
and, even more so, of their polity. 28
4.3 On the relationship between the economic and the political:
problems of reductionism in Marxist political theory
An underlying constant concern of this book has been to examine

economic and political developments in a noncompartmentalised,
holistic manner - while at the same time avoiding the reductionist
risk that such a holistic approach often entails. Given that the
problem of the relationship between the economic and the political constitutes a central preoccupation of Marxism, I would like to
discuss briefly those aspects of the ongoing debate that are
particularly relevant to the issues and themes discussed in this
work. At the risk of overgeneralisation I would argue that
present-day Marxist theories put forward two equally unsatisfactory views of the economy-polity relationship.
A. The first of these consists of a straightforward, old-fashioned
reductionist approach, whereby political phenomena are explained in terms of either the reproduction requirements of capital
or the interests and projects of the economically dominant
classes. 29 In the former, the analysis necessarily results in some
kind of teleology as political institutions and developments are
explained in terms of the changing systemic needs or requirements
of the capitalist mode of production. 30 Where, on the other hand,
the approach focusses on the economically dominant classes, the
analysis usually jumps straight from classes and class struggles
(nebulously defined) to forms of state or regime, for example, the
'liberal', the 'interventionist', the 'bonapartist' state, 'normal' or
'exceptional' bourgeois regimes, and so on. This totally neglects to
200
give due consideration to the complex organisational and institutional realities which lie between classes and the state (such
phenomena as political parties, pressure groups, clientelistic networks, and so on.) 31 However, neglecting this crucial area (which
mediates between, and gives substance to, the concepts of both
class struggle and state form) not only leads to the portrayal of
classes (especially dominant classes) as omniscient and omnipotent
anthropomorphic entities mysteriously regulating everything on
the political scene; it also deprives the researcher of the only
conceptual tools which would enable him to arrive at a nonteleological, non-reductionist conception of the capitalist state and
politics.
This underconceptualisation by Marxist theory of organisational
structures is particularly disastrous given that political parties (as
distinct from political factions or clubs of notables) are one of the
few institutional forms that are unique to modern politics: preindustrial societies have parliaments, elections and state bureaucracies, but not mass parties. 32 It is also highly deficient with respect
to our concerns, since the kind of party structure that prevails is a
central consideration for understanding a given polity's type of
political stability or instability. More generally, the attempt to
analyse political developments in the semi-periphery/periphery in
exclusively economic categories (class, requirements of capital) is
especially absurd in view of the saliency of political structures and
the crucial role the state has played in generating capitalist
industrialisation in such societies.
B. The second approach attempts to avoid the crude reductionism of the first by stressing the 'relative autonomy' of the political
sphere, and the fact that political forms cannot be automatically
derived from economic determinations. In this approach the
economy is not supposed to determine political developments
directly, but merely to delineate what is possible on the level of the
political superstructure. What, however, according to this view, is
going to emerge within these set limits depends on the political
conjuncture, and this leaves no more room for a theorisation of
specifically political structures and contradictions. 33
This approach subjects the political sphere to a subtle and
sophisticated downgrading. While it is conceded that economic
constraints or forces can no longer be regarded as the direct
201
determinants of politics, it is proposed that political phenomena,

although relatively autonomous, are not amenable to the same
kind of analysis and theorisation as economic ones. The latter can
in part be accounted for in terms of the structural tendencies of the
capitalist mode of production; for political phenomena, however,
with their fluid and transient character, structural analysis must be
entirely replaced by a study of the political conjuncture. In other
words, this theoretical position boils down to ascribing different
ontological status to the economic and political realms; presumably, because of its greater 'materiality', the economic sphere is
accorded superior explanatory status.
This is not the place for a full criticism of this subtle form of
neo-Marxist reductionism. The only point I would like to make
here is that, if the political sphere seems to some Marxists less
amenable to theorisation (once one attempts to avoid economic
reductionism), then I think this has to do less with the highly
dubious proposition that the former is less 'material' than the
latter, and much more with the fact that Marxist analysis is
extremely poor in conceptual tools specific to the sphere of politics.
In view of the tendency of Marxism to explain politics exclusively
in terms of economic categories, very few efforts have been made
by Marxist theorists to construct political concepts which seriously
take into account that political institutions often have their own
histories and their own logic; that if political institutions can be
conceptualised as the conditions of existence of economic structures, then economic institutions can equally well be conceptualised as the conditions of existence of political structures. It is not,
therefore, surprising that a non-reductionist, 'regional' Marxist
theory of politics in general, and of third-world politics in particular, simply does not exist.
To give a relevant example: Marxist theory has very little to say
about the crucial problem of political clientelism in peripheral
capitalist formations, except to dismiss it cavalierly as an epiphenomenon that merely 'reflects' class antagonisms. 34 While functionalist anthropologists and sociologists fail to give due weight to
classes in their examinations of clientelism, Marxists neglect
clientelism altogether in favour of a class reductionist approach
which attempts to set up direct one-to-one linkages between class
locations and political practices. By doing so, Marxists eliminate
the all-important problem of the intricate and shifting rela-
202
tionships between vertical/clientelistic and horizontal/nonpersonalistic organisations in the course of capitalist development.35
If Marxist political theory makes no serious attempt to theorise
the phenomenon of clientelistic politics, neither has it much to say
about, for instance, ethnic or racial cleavages, about different
types of political mobilisation, about types of political parties and
party systems, about the nature of state administrative systems in
the periphery. The theoretical elaboration and discussion of such
concepts is left to functionalist political sociologists. Any attempt
at breaking out of the reductionist straitjacket and trying to
incorporate such terms as clientelism within a Marxist discourse is
usually denounced as eclectic, as an unacceptable 'contamination'
of Marxism by 'bourgeois' concepts.
In this state of affairs it is no wonder that Marxists, sensitive to
the reductionist pitfalls of the first approach, have come to the
conclusion that political phenomena are not amenable to the same
type of analysis as economic ones. In the absence of conceptual
tools specific to the political level, any attempt to avoid reductionism necessarily leads to an empiricist, ad hoc treatment of the
political realm as an area of social life lending itself to only
conjunctural analysis. In consequence, this particular theoretical
endeavour to avoid the crude reductionism that has plagued
Marxist theory from its very inception can only result in atheoretical treatments of political phenomena. While it removes an
unsatisfactory conceptual framework, it puts nothing useful in its
place.
My own position on this particular theoretical issue is that the
political sphere is as 'material' or as 'immaterial' as the economic
one (however materiality may be defined). As far as the highly
differentiated capitalist societies are concerned, both their political and economic spheres exhibit a complex of institutional structures setting limits to collective action, and constituting arenas
within which collective agents can succeed or fail in the realisation
of their (often antagonistic) projects. In that sense there is no
ground whatsoever, ontological or heuristic, for arguing that
political phenomena are not amenable to the same type of analysis
as economic ones. What is needed is structural as well as conjunctural analysis, whether one focusses on the economic level or the
political or both.
203
If this position is accepted, and coming back now to the specific

concerns of this study, the point is that it is not enough (in terms of
an examination of the development of industrial capitalism in the
semi-periphery) to simply show the limits set by evolving economic
structures to political developments, and to leave the rest to
'political conjuncture'. If the idea of the relative autonomy of the
political sphere is taken seriously, political structures, and specifically political contradictions, must be studied on their own terms
- presented neither as mere expressions of economic needs or
conditions or existence of the economy, nor dismissed altogether
by relegating all political phenomena to the sphere of conjunctural
analysis.
C. It was with such preoccupations in mind that I tried to

conceptualise populism and clientelism as typical modes of political inclusion in the parliamentary semi-periphery, as well as to
formulate the basic political contradiction between mass political
mobilisation and incorporative/exclusionist means of political controls as relevant to understanding the rise of bureaucraticauthoritarian regimes. I am aware, of course, that such a preliminary, limited conceptualisation is neither very rigorous nor does it
fill the huge lacunae in Marxist political theory. Moreover, given
the lack of adequate concepts of empirical research based on such
concepts, I have not always managed in my own empirical analysis
to avoid completely reductionist simplifications when dealing with
the relationship between economically and politically dominant
groups. Despite all this, I think that the attempt in this book to
conceptualise political structures and contradictions in a noneconomistic manner makes it quite clear that there is an urgent
need to create new concepts within the tradition of Marxistoriented approaches to the study of politics, new concepts which
give serious attention to the specificity of political structures and
contradictions of capitalist formations.
Referring back to the earlier discussion of reductionism, I think
it has now become amply clear that there is no way of overcoming
the economic reductionism v. empiricism dilemma other than by
going beyond the ritualistic invocation of the relative autonomy of
the political sphere, and by creating conceptual tools which:
(a) on the one hand try to deal with political phenomena in a way
that does not build into their very definition (and therefore
204
excludes from empirical investigation) the type of relationship

they are supposed to have with the economy; (b) on the other
hand try to avoid reductionism without abandoning the 'political
economy' holistic approach, that is, without falling into the type of
compartmentalisation of the political and economic spheres which
is to be found in neo-classical economics and in non-Marxist
political science.
A thorough and systematic discussion of how such concepts of a
non-reductionist, holistic approach are to be developed cannot be
given here. Their full theoretical elaboration would necessitate not
only the critical examination of already existing attempts in that
direction from Marx onwards, 36 but also a detailed demonstration
of how the newly-created concepts are more useful than their
predecessors. Since this goes far beyond the scope of the present
work, I shall restrict myself to a few unsystematic suggestions
which, I hope, will indicate that such an endeavour is indeed
possible.
To start with, there is no reason why one should not speak of a
mode of political domination in ways which are pretty similar or
isomorphic to those used for the analysis of a mode of production.
So if a dominant mode of production designates the major
productive forces or technologies 37 , and the manner of their
control (relations of production), a mode of political domination
can designate the major political technologies of domination
(types of means of administration, types of party political apparatuses, techniques of political indoctrination or propaganda, techniques of political coercion, and so on), and the main institutionalised ways in which such political technologies are controlled (that
is, the relations of domination). Moreover, in the same way that
relations of production are the basis for the formation of exploiting
and exploited classes, relations of dominations similarly constitute
the structural basis for the formation of politically dominant and
dominated groups, of rulers and ruled.
It seems to me that political technologies (forces of domination)
as well as their manner of control (relations of domination)
deserve as much attention, specific conceptualisation and analysis
as the forces and relations of production. After all, a mode of
domination refers to a specific type of production: the production
of political power. Just as the production of economic goods can
be differentiated according to the types of technology used and the
205
manner of their appropriation/control, so can the production of

political power (as a resource) be differentiated by the political
technologies used and the manner of their control. 38
If this is accepted, then a mode of domination can be examined
in ways quite similar to those used for a mode of production. For
instance, one can study the 'genealogy' of a mode of domination
(by tracing the particular history of its basic institutional features),
its basic structural tendencies and contradictions, the type of
political struggles that are possible on the basis of such contradictions, and so on.
To move the analogy to a more concrete level, in the same way
that western capitalism can be distinguished from peripheral/
semi-peripheral capitalism (the latter being more restrictive and
unequal), so two sub-types can be discerned within what may be
called the liberal-democratic mode of political domination:
the western mode of parliamentary domination, characterised by
integrative relations of domination; and the semi-peripheral/
peripheral mode of domination, characterised by incorporative
relations of domination. The different ways in which the ruled
majority is brought into the political arena in the centre and the
periphery/semi-periphery respectively, indicate the different ways
in which political technologies are being controlled. In the western
case, political subjects, although, as Michels has argued, 39 they do
not exercise direct democratic control over the means of administration, do participate in this control in a more autonomous
manner. In the periphery/semi-periphery the dominated groups as
collective political subjects participate much less in the effective
control of the means of administration, and one is therefore
justified in speaking of different relations of domination.
Before closing these very sketchy notes on possible ways of
conceptualising the political sphere, I want to stress just one final
point. The attempt to create concepts specific to the political level
must by no means be taken as implying a denial of the importance
of the economic phenomena, or of the necessity of a holistic
approach to the complex interrelationships between the economic
and the political spheres. Its purpose is to prevent the kind of
a-priorism so characteristic of Marxist analyses, which consists of
defining political phenomena in such a way that certain crucial
problems are 'solved' in advance by being left out of the field of
empirical investigation altogether.
206
For example, if the state in capitalist formations is defined as an

instrument of the economically dominant classes, or even as
performing the functions of capital, this automatically rules out of
court the investigation of cases, quite frequent in the periphery and
semi-periphery, where the economically dominant groups are the
passive creatures of policies initiated by the politically or militarily
dominant groups, or cases where state policies hinder rather than
advance the enlarged reproduction of the capitalist mode of
production. 40 In other words, the creation of conceptual tools
specific to the political level not only does not deny the study of the
inter-relationships between the economy and polity, but on the
contrary provides the fundamental precondition for opening up
the study of such relationships to empirical investigation. Furthermore, it does this by avoiding simplistic reductionist generalisations, and without falling into an ad hoc or compartmentalised
analysis of political developments.
4.4 Relations of production and relations of domination: the
relevance of the distinction for the study of political developments
in thesemi-periphery
The suggestion made in the previous section for the badly needed
creation of new tools within Marxism for the study of politics may
appear far removed from the problems of political transition in the
Balkans and Latin America. However, given that most attempts at
going beyond a merely descriptive/historiographic account of such
transitions are based on some sort of Marxist analysis, I feel that a
greater awareness of theoretical and methodological issues of the
type just discussed could dispel a lot of confusion and lead to more
fruitful empirical research.
I shall illustrate this by taking first an example from Greek
historiography. Marxist historians have conventionally explained
the 1909 military coup as a 'bourgeois revolution' marking the
decline of-the landowning classes and the rise of the bourgeoisie.
More recently, historians and social scientists have pointed out
that the above interpretation transposes in mechanistic and illegitimate fashion explanatory schemes developed for the analysis of
western European developments to a country which had a radicalA.
207
ly different type of historical trajectory. They rightly point out, for

instance, that the oligarchic families who controlled the
nineteenth-century Greek state (the famous tzakia) did not base
their political dominance on large landownership. As already
mentioned in Chapter 1, the formation and consolidation of big
landed property encountered severe obstacles in nineteenthcentury Greece. In the immediate post-independence period the
formerly Turkish estates were not taken over by Greek landlords
but became national lands under the direct ownership of the
Greek state. These lands were then distributed to peasant families
from as early as 1871. The tzakia families, whose origin can be
traced back to the military chieftains and notables (proestoi) of the
pre-independence era, initially derived their authority from the
leading role they played during the War of Independence Given
that the landowning avenue was relatively blocked, they turned
to the professions (to preserve and perpetuate their power)
and, above all, to the state itself as politicians, state officials,
academics, lawyers, and tax farmers. Moreover, clientelism
proved a remarkably effective strategy, which turned universal
suffrage to their lasting advantage. Ultimately, the state budget
itself became their principal economic base- hence the designation 'state bourgeoisie'. 41
Although this analysis of the tzakia families is rather more
accurate than the previous one, the term 'state bourgeoisie' (which
is widely used by Marxists in the analysis of third-world societies)
is highly confusing. Since the tzakia families did not base their
political power on the ownership/control of the means of production or distribution, in what sense are they bourgeois? At best the
expression 'state bourgeoisie' is a contradiction in terms. At worst
it implies an a priori linkage or close affinity between state
bureaucrats and various bourgeois elements (merchants, industrialists, and so on); it could also imply that state bureaucrats play
a role which in normal circumstances the bourgeoisie is or should
be playing; it could finally imply that those in control of the state
apparatus are part and parcel of the bourgeois class. Whatever
the interpretation, there is no doubt that the attempt to view the
tzakia families as 'state bourgeoisie' confuses rather than clarifies
matters. It conflates in a-prioristic manner the economic and the
208
political. Rather than facilitating, it prevents the empirical investigation of the actual relationship between economically and
politically dominant groups. It serves no purpose other than
complying with the conventional Marxist view that the holders of
state power must be defined in economic terms, even when it is
patently obvious (as in this case) that both the source of their
power and the nature of their activities are predominantly political.
It seems to me that it is the absence in Marxist political theory of
non-economistic conceptualisations of political structures which
inevitably leads to this type of terminological confusion, to this
systematic tendency to fuse a priori, through misleading definitions, economic and political power. This confusion can be easily
avoided if the mode of domination concept (or something equivalent) is introduced into the analysis.
So, for instance, if the tzakia families are viewed as the
holders/controllers of the means of domination within a specific
mode of domination, leading to specific cleavages between politically dominant and dominated groups, the interesting problem can
then be investigated of the actual relationship between holders of
the means of production and holders of the means of domination.
What was this relationship in the immediate post-independence
period? How did it evolve during the nineteenth century, how did
it change after 1909 or during the post-1929 period of rapid
industrialisation? By introducing the mode-of-domination concept
this relationship can be studied in terms of not only the degree of
interpenetration between economically and politically dominant
groups, but also in the equally interesting terms of the degree of
compatibility between the reproduction requirements of the economy and polity. For example, in certain cases and within certain
limits, the possibility cannot be excluded that the reproduction
requirements of a certain mode of domination might be relatively
incongruous with the reproduction requirements of the dominant
mode of production. In this case, those who control the means of
domination, even if closely related to economically dominant
groups, might adopt policies which favour the consolidation of
their political power rather than the growth of the economy or the
interests of those who control the means of production. This is to
say, that it is by no means certain, especially in the societies under
study here, that there will always be coincidence of interests
between those who control the means of domination and those
209
who control the means of production - even in cases where there is

a lot of overlapping (in terms of ideology or class origin) between
these two groups. Needless to say, the above problems cannot be
formulated at all if one conceptualises the state as the instrument
of the economically dominant classes, or if one defines a priori the
holders of the means of domination as state bourgeoisie. 42
B. In nineteenth-century southern Latin American (see above)
the separation between the holders of the means of production and those of the means of domination were not as clearcut as in nineteenth-century Greece. But even in these cases,
despite a considerable overlapping of political and economic
power, there was certainly no complete fusion. For instance, in
Chile the political oligarchy was not entirely identifiable with
landowning or export interests:
By mid-century (19th century) the state was not simply a tool of
economic elites but had in fact attained a considerable degree of
autonomy. An entirely new profession of urban-based government officials and politicians had appeared on the political scene.
Like President Manuel Montt himself, they relied on the state for
their positions and had a real stake in the expansion of government authority. 43 (my italics)
If non-landowning groups formed part of the oligarchic political
establishment, it is also true that landowning families were often
part of the anti-oligarchic opposition. In Argentina for example,
despite the implementation of the Saenz Peiia electoral-reform law
that had broadened political participation and brought the Radicals into power in 1916, not only was Yrigoyen himself a landowner, but (as already mentioned) several members of his cabinet
were influential members of the Sociedad Rural. Moreover, the
Radical government's policies often promoted agrarian/export
interests. 44
Taking the above into proper consideration makes one realise
that the demise of oligarchic parliamentarism and the rise of a
broader system of political participation in Latin America's southern cone as in Greece, did not directly correlate with the decline of
an economically dominant class and the rise of another; and that
therefore the transition cannot be fully explained by an exclusive
210
emphasis on the concept of class. Marxists usually try to overcome

the above difficulties by using the more flexible concept of class
fractions. But in as far as certain political struggles, at a particular
stage of their unfolding, have their primary structural base within
the polity rather than the economy, the concept of class fractions
cannot deal adequately with the situation.
To revert to the previous example: since landowners are found
both in the oligarchic camp and in the anti-oligarchic political
forces, one should not a priori exclude the possibility that these
groups may be differentiated along political rather than economic
lines, that is, they take different sides in this political confrontation, not so much because they have a different economic base,
but because of their differential access to political power (in which
case an analysis exclusively in terms of class fractions would not go
very far). Moreover, even supposing this not to be the case (that
is, even if the underlying cleavage between landowners belonging
to different political camps were economic), there should still be
concepts available allowing the researcher to pose the alternative
as a possibility, as a hypothesis to be investigated. At the present
time the idea prevails in the Marxist tradition that all political
conflicts, in so far as they are amenable to structural explanations,
can be explained adequately in terms of economic categories
(class, class fractions, reproductive requirements of capital, and so
on). Therefore a situation in which social actors belonging to the
same class or class fraction are structurally divided because of their
differential access to the means of domination is theoretically
unthinkable.
At this point Marxists can argue of course that within Marxism
the concept of class does not refer to a mere economic category.
Nicos Poulantzas, for example, conceptualises class practices45 as
not merely determined by the economy, but by the over-all
'structural matrix' which consists of a complex articulation of
economic, political and ideological structures. It is precisely
because political and ideological structures, as well as economic
ones, have 'pertinent effects' on class practices that there can be
no one-to-one correspondence between objective class locations/
places (determined by the capitalist division of labour) and class
practices/struggles. 46
Stretching the concept of class in this manner results, however,
in not being able to distinguish class from not-class practices, and
211
every collective action, every social struggle, then becomes a class

struggle. The concept of class practice/struggle then operates as a
blanket term preventing the study of the complex interrelationships between economic, political and ideological struggles.
Poulantzas could object to this by arguing that within his
formulation it is possible to distinguish class practices and struggles
from other types of struggle, in that in the former case economic
factors, although always related to political and ideological ones,
are determining 'in the last instance'. Even introducing this
qualification, and leaving aside the structuralist teleology implied
by 'in the last instance' concept, still leaves one with the task of
theorising practices/struggles, the primary determination of which
derives not from the economy but from the political or ideologicaV
cultural spheres (for example, racial, ethnic struggles). This is
precisely what neither Poulantzas nor any other Marxist theorist
has tackled seriously.
C. On a less theoretical level, Marxists have tried to deal with
the difficulties arising out of the inadequacy of a strict class
analysis for explaining the political transition to post-oligarchic
politics by viewing it as a 'failed bourgeois revolution'. To take the
southern Latin American societies as an example again, in this
view the rising middle classes in these countries, unlike the
'conquering' western European bourgeoisies, did not succeed in
building an autonomous industrial base and in revolutionising
relations of production in the towns and countryside. Instead,
industry remained subjugated to the requirements of the agrarian/
export economy, and the weak industrial bourgeoisie failed to
proceed to a thorough bourgeois transformation of state and civil
society. 47
This concept of 'failure', however, is highly unsatisfactory. It
assumes that, as in western Europe a century or so ago, objective
possibilities existed in the early twentieth-century semi-periphery
for an autonomous type of capitalist industrialisation, objective
possibilities of which the rising middle classes failed to take
advantage. This is a very dubious assumption, based less on a
serious assessment of the situation and more on a teleological
attempt to read the history of southern Latin America in terms of
western European industrialising experiences. At best, even if the
212
'failed revolution' theme can be put forward in non-teleological

fashion, it simply tells us what has not happened, not what actually
did take place in the economy and polity of early twentiethcentury Argentina and Chile.
It seems to me that a clear distinction between relations of
production and relations of domination is rather more helpful for
understanding the nature of the oligarchic-post-oligarchic transition. Thus the 'revolutions' of 1916 in Argentina and 1924 in Chile
point to profound changes, not so much in the sphere of economic
production but in the sphere of political domination. They point to
a radical change in the control of the means of domination, in the
sense that the considerable overlapping of economic and political
power that had existed in the oligarchic period was drastically
reduced. This did not mean that the economically dominant
classes lost all political power, of course. As argued already, they
managed to retain a great amount of direct political control in
Chile (where landlord control of the countryside declined only in
the 1960s), and indirect political control in Argentina (through
powerful pressure groups). All the same, the type of political
monopoly they had enjoyed in the oligarchic period was lost
irretrievably and the post-oligarchic reforms and ensuing broadening of political participation meant both a spectacular growth of
the state bureaucracy and a clear restructuring of political leadership. From then on the old political oligarchy had increasingly to
share power with 'new men', that is, men who during the
oligarchic period had been political outsiders. These new men
could be merchants, lawyers, landlords, industrialists, bureaucrats, and so on - what matters about them in this context is not
their class origin or occupation so much as their position within the
polity, their relation to the means of domination both before and
after the demise of oligarchic parliamentarism.
From this point of view the relative loss of power of the
landed/export interests and the 'rise of the middle classes' did not
mean the passage from an 'upper-class' to a 'middle-class' state.
The amorphous/heterogenous middle classes could not make the
state an instrument of their hazily defined class interests, similar to
the way in which the upper classes had managed to gear the state
to their narrow interests in the oligarchic period. It was rather that
with the demise of oligarchic parliamentarism the state became
relatively autonomous vis-a-vis both upper- and middle-class in-
213
terests. True, having to operate within structural constraints

imposed by national and international capitalism, the state often
had to adopt policies favouring the economically dominant groups,
as it also had to take increasingly into account middle- and
working-class demands. But what is important is that, compared to
the oligarchic era, the holders of the means of domination were
now less of an 'instrument' for serving the landed/export interests,
but without at the same time having become an instrument of any
new rising class. Instead, the holders of the means of domination
enhanced their capacity for relatively insulating themselves from
specific class pressures, either by favouring different class interests
at different conjunctures, or by playing off one interest group
against another as occasion required. 48 In other words, those in
control of the means of the huge and proliferating state apparatuses were increasingly capable of shaping class interests rather
than being shaped by them. This is perfectly understandable when
one remembers the degree of expansion/penetration of the semiperipheral state, its enormous resources, and the advanced technology at its disposal, a technology that could be used both for
repression and for the shaping of 'public opinion'. (This applies to
the state of the capitalist centre as well as to the state of the
semi-periphery, but particularly so to the latter because of the
relative dependence and permeability of its civil society.)
In view of the above, it is not difficult to understand why and
how the post-oligarchic holders of the means of domination
enhanced their capacity to promote their political interests, even
when such interests might clash with upper-class economic interests. To put it in systemic/structural terms, the post-oligarchic
relative autonomy of the state means that instead of the 'logic of
capital' shaping state structures it is often the 'logic of domination'
which profoundly distorts the logic of the market. Although this
may sound shockingly heretical to orthodox Marxists, it is a truism
to anyone aware of the extent to which primarily political determinations shape state economic policy in the countries under
consideration.
Finally, as regards the issue of relative state autonomy, it is
worth comparing the situation in Latin America's southern cone
with that in Greece. In the latter case the holders of the means of
domination, not only in the post-oligarchic era but even during the
period of oligarchic parliamentarism, enjoyed a considerable
214
degree of autonomy in relation to the holders of the means of

production. One reason (already given) for this was that, unlike
their opposites in Latin America, the oligarchic families ruling
nineteenth-century Greece did not primarily base their dominance
on large-scale landownership. This early autonomy of the political
sphere in Greece explains to some extent why the two political
transitions examined in this book (1909 and 1967) were less
directly linked with economic issues than were the equivalent
transitions in the southern cone countries (compare on this point
Chapter 1, conclusion, and Chapter 3.4.1).
D. This extraordinary predominance of the political, both in
Greece and in the post-oligarchic societies of southern Latin
America, has led some writers to formulate the theory of the rise
to dominance of a new bureaucratic-military class. It has been
argued with reference to southern Latin America that those who
have been in control of the state apparatus since the rise of the
post-war bureaucratic-authoritarian regimes have imposed their
hegemony over the old dominant classes of landowners, merchants
and industrialists. 49 This adaptation of the 'managerial revolution'
theme to the Latin American semi-periphery is as unsuccessful for
the analysis of political developments in these countries as James
Burnham's theory was incapable of analysing the state-economy
relations in the more developed world. It is quite clear that before,
during, and after the imposition of military dictatorships in the
semi-periphery, the politically dominant bureaucratic-military
group did not displace landed/industrial capitalist interests in the
economic sphere. Although once they had taken over they promoted both their own economic interests and those of the armed
forces as a whole, they did not destroy the economic hegemony of
capitalist interests; on the contrary, both in Argentina and Chile
they promoted the interests of big capital at the expense of middleand working-class demands. In fact (as already mentioned) this
was so much so that the military and their bureaucratictechnocratic allies were seen by many Marxist-oriented writers as
the instrument par excellence of big capital.
However, both the theory which sees the military-bureaucratic
groups as a new dominant class, as well as its opposite which sees
them as capital's puppets, are misleading. Both theories conftate
the economic and political spheres in a highly confusing manner.
215
The former obliterates the distinction between economic and

political power by viewing the holders of the means of domination
as a new class which displaces the old dominant classes (here the
political absorbs, so to speak, the economic); the latter, following
a more conventional Marxist analysis, views the state as a mere
reflection of capitalist struggles and interests (here the political is
an epiphenomenon of the economic). 50
In my own view a necessary, although not sufficient, precondition for reducing the growing confusion that reigns within the
Marxist-oriented literature on the post-oligarchic state in the
parliamentary semi-periphery is to spell out as fully as possible the
distinction between modes of production and modes of domination. Once this distinction is theoretically formulated and its
practicaVheuristic implications rigorously examined, then it becomes fairly obvious that the holders of the means of domination
in the post-oligarchic semi-peripheral state are neither a new
dominant class nor a passive instrument of capital.
The distinction between modes of production and domination
suggests that, given the smaller degree of overlapping between
economic and political power in the post-oligarchic semiperiphery, the holders of the means of domination have an
enhanced capacity for defending and, within certain limits, further
promoting their own primarily political interests. This they systematically do by the continual expansion of state functions and
resources, and by the systematic defence of their control of the
means of coercion/domination against threats from groups previously excluded from the political arena or groups occupying a
peripheral position within it.
As I have already explained, however, (see Chapter 3.3.1), in
Argentina and Chile as well as in Greece the relative autonomy of
those holding political power vis-a-vis those who control the
private means of production does not allow the former to displace
the latter in the economic field and hence to abolish the distinction
between the economic and political spheres - a distinction which is
the very foundation of capitalist social formations.
Another important point that should be stressed in relation to
the mode-of-domination concept is, that like that of the mode of
production, it can lead to an analysis of social processes from the
point of view of both action and system. Concerning for instance
the state's greater post-oligarchic autonomy vis-a-vis class in-
216
terests, in terms of action the politically dominant groups (that is,

the holders of the means of domination/coercion) are relatively
independent from class or other civil society interests if they can
successfully impose policies that may go against the interests of the
economically dominant groups (for example, the pursuit of ultranationalistic, militaristic projects which could endanger the longterm expansion of capitalist/export interests). In systemic terms
the state, as a system of complex institutional structures, is
relatively autonomous vis-a-vis the economic system if, in case of
functional incompatibility between the reproduction requirements
of the mode of production and the mode of domination, the latter
systematically prevails over the former (for instance, political
institutional arrangements might be structured in such a way that
they systematically lead to state policies that encourage the growth
of low-productivity bureaucratic jobs and discourage productive
investments in industry or agriculture). 51
Two final remarks concerning the theoretical argument about
modes of domination and modes of production. First, I would like
to stress once more that the rough conceptual framework suggested above is not based on the assumption of the 'primacy of the
political' over the economic or vice versa. My argument is rather
that, irrespective of which institutional sphere is dominant, the
economy-polity relationship in capitalist societies cannot be studied properly without a non-economistic conceptualisation of
political institutions and actors. This is so, especially when, as in
the cases under investigation, the state is all-encompassing.
Second, despite its economic reductionism and its underconceptualisation of political phenomena, Marxism more than any other
paradigm in the social sciences can suggest very fruitful ways of
studying the polity from the point of view of both action and
system; both as a configuration of actors struggling over scarce
resources, and as a systemic whole with specific reproduction
requirements the institutionalised sub-systems or parts of which
can contradict one another.
As Lockwood pointed out long ago, Marxism combines a system
and a social integration view of society. It encourages the examination of contradictions between systemic parts or institutions (for
example, that between forces and relations of production), as well
as the ways in which such contradictions lead/fail to lead to the
development of class conflict. 52 Unlike action theories for inst-
217
ance, collective actors (that is, classes) in Marxist theory do not

operate in an institutional vacuum; their strategies have to be seen
within existing structural/systemic constraints and contradictions constraints and contradictions which they may or may not be
aware of. In contrast to Parsonian functionalism at the other
extreme, Marxism (Althusser's structuralism apart), does not
conceptualise actors as mere puppets of the system; it shows them
as both producers and products of their social world. Hence
Marxist concepts are free of the exclusively systemic, 'neutral'
character with which they are endowed in the Parsonian scheme.
They are constructed in such a manner that those who use them
are forced to ask 'who?' questions. For example it is not possible
within the Marxist tradition to talk about societal values and their
neutral control over action: instead, the accent is on hegemony, on
dominant ideologies and the degree to which they succeed in
concealing systemic contradictions.
It is true, of course, that in Marxism the only methodical use of
the powerful action-system synthesis is in the analysis of economic
production systems. There is no such synthesis nor any analogous
conceptual tools for the analysis of political systems, hence the
poverty of Marxism as far as a 'regional' theory of the political is
concerned. For some critics this poverty is innate to Marxism. If
Marxist theory ceased to be economistic/reductionist, it would lose
its distinctiveness and specificity. In other words, a nonreductionist Marxist theory of politics would be a contradiction in
terms. 53 Contrary to this point of view I have tried to suggest in
this section that a non-economistic theory of the political might be
possible within the Marxist tradition. This would require less
emphasis on the 'primacy of the economic' theme and more
systematic attempts towards a conceptualisation of the polity
analogous to that used for the analysis of the economy, a conceptualisation, that is, in terms of systemic contradictions between
institutional parts leading (in certain favourable conditions) to the
development of political struggles over the distribution and control
of power between dominant and dominated groups.
As will have become clear by now, what I am advocating is a
rapprochement between Marx's political-economy approach and
Weber's political sociology (especially his insights on types of
domination, on the crucial importance of not only the means of
production, but also the means of administration, coercion, and so
218
on). By such a rapprochement I do not mean of course an eclectic

ad hoc mixture of their ideas; I mean the serious elaboration of
new conceptual tools which can prove their utility in solving the
type of theoretical and methodological difficulties outlined above.
Yet such a solution remains anathema to many Marxists. For my
part, provided the rapprochement is done in a theoretically
rigorous manner, I see it as an essential precondition for breaking
out of the general theoretical impasse that hampers both Marxist
and non-Marxist sociology today. It is also an essential precondition for a more effective empirical study of the long-term politicoeconomic transformations in the countries of the parliamentary
semi-periphery.
General Conclusion
This book presents a comparative examination of some fundamental political developments and transitions in what I have
called countries of the parliamentary semi-periphery, that is, in
certain Latin American and Balkan countries which share not only
fairly comparable experiences of integration into the world market
and of belated industrialisation, but which also have similar
historical experiences of an (by third-world standards) early
political independence and long parliamentary traditions.
More specifically, I have looked at two crucial turning points in
these countries' long parliamentary history: the transition prior to
the Second World War from oligarchic to broader forms of
political participation, and the post-war emergence of military
dictatorial regimes which endeavoured to restrict political participation and to exclude the masses from active politics. In Chapter
1 it is argued that in the Balkans and in southern Latin America in contrast to the western pattern- processes of state, market, and
urban expansion, as well as the early adoption of parliamentary
institutions, undermined oligarchic politics and led to broader
forms of political participation before the development of industrial capitalism. This meant that in the parliamentary semiperiphery the transition from oligarchic to mass politics occurred
in a context where the industrial classes were relatively weak
and/or dependent, and where an already over-developed state
found very little resistance to its incorporative tendencies incorporative tendencies that can be traced back to the preindependence despotic patrimonial empires of which Balkan and
Latin American societies formed a part.
This fundamental point was developed by examining the predominantly vertical, clientelistic and/or populistic mechanisms of
political incorporation through which the lower and middle classes
were brought into the post-oligarchic political arena. Three specific patterns of incorporation were identified. In the South
219
220
General Conclusion
American cone, where the landowning classes were all-powerful

and urbanisation very advanced, the demise of oligarchic parliamentarism led to the development of populist movements in the
towns. In the northern Balkan countries on the other hand, where
landlords were weak and where, due to relatively low urbanisation
rates, the urban middle classes were neither numerous nor effectively organised on the political level, the transition was marked by
the development of strong peasant populist parties. Finally in
Greece, where landowners were as weak as their Balkan counterparts but where urbanisation was quite advanced, the demise of
oligarchic parliamentarism and the transition to post-oligarchic
politics took place without any large-scale urban or rural populist
mobilisation, by the mere transformation/extension of pre-existing
clientelistic forms of political organisation.
What all three types of transition have in common is the
dependent/incorporative type of lower-class inclusion. This was
further reinforced by the co-optive/repressive manner in which the
state tried to control working-class organisations during the acceleration of the industrialisation process in the 1930s and 1940s. In
consequence, the transition to post-oligarchic politics, as well as
the subsequent post-1929 shift in emphasis from export orientation
to import-substitution industrialisation, resulted in a political
system marked by the persistence of particularistic/personalistic
politics and by the accentuation of the state's authoritarian tendencies.
After this long comparative essay, the shorter and more theoretical Chapter 2 (which completes Part I) assessed the relevance of
the preceding empirical analysis for certain key concepts in the
political sociology literature: those of populism, clientelism, and
political incorporation.
Part II, Chapter 3, by focussing only on Greece, Chile and
Argentina, strove to identify certain common long-term structural
developments that are relevant for understanding the post-war
establishment of military bureaucratic-authoritarian regimes in
these three countries. It was argued that both the inter-war
transition to post-oligarchic politics, and the post-war restricted
and uneven development of foreign-led industrialisation, to some
extent explain the relative dependence of these countries' civil
society, the incorporative/paternalistic character of their state, and
the important political role played by the army (especially in
General Conclusion
221
Greece and Argentina) in the post-oligarchic political system.

Particular attention was paid to the increasing contradiction in
these polities between rapidly growing levels of political mobilisation/radicalisation and the mixture of incorporative/exclusionist
mechanisms of control within which the military, more or less
directly, assumed a guardian role.
These mechanisms of political control (which, as is argued in
Part I, took on definite shape in the inter-war, post-oligarchic
period) were reinforced in all three countries, with varying support
and guidance from the United States, in the early post-war period:
in the late 1940s in Greece and Chile, and in the mid-1950s in
Argentina. They were severely undermined in the 1960s, however,
by the mass mobilisation/radicalisation of both urban and rural
populations. The type of linkages that can be established between
the above state of affairs and the rise of military regimes in
Argentina, Chile and Greece is discussed systematically. A postscript to Chapter 3 takes a brief look at the nature and dynamics of
these military regimes, as well as at the over-all trajectories of
parliamentary institutions in these three countries.
Finally in Chapter 4 I have tried to show the relevance of the
preceding comparative analysis for the more theoretical debates
on the praetorian features of third-world politics in general, and
on the post-war rise of bureaucratic-authoritarian military regimes
in some of them in particular. Given the general focus of this book
on the interrelationship between economic and political developments, Chapter 4 also examines the reductionist ways in which
Marxist theory conceptualises the relationship between the economic and the political, and the difficulties that this conceptualisation
creates for the investigation of the issues dealt with in this book.
Tentative suggestions have been made on ways to overcome these
difficulties.
I would like to end this work by stressing once more that in
bringing together, for the purposes of comparison, countries that
are so far apart in terms of geography and culture, my purpose was
not to disregard their significant differences. It was rather to show
that, given a number of similarities (related to the timing and
structure of their industrialisation, their early adoption of parliamentary institutions, and so on), the comparative examination
of certain systematic differences in their social structure and
historical development can generate insights or suggest hypotheses
222
General Conclusion
to enrich or supplement those proposed by area studies (Latin

American, South European, Mediterranean studies).
For example, I think that students concerned with Latin American urban populist movements can learn a great deal about such
movements by comparing them with the strikingly different interwar Balkan peasant ones. This is not only because of the populist
ideological and organisational elements held in common by both,
but also because both of them can be conceptualised as modes of
transition from oligarchic to mass parliamentary politics. Alternatively, and on a more concrete level, students of Balkan peasantism or Mediterranean clientelism can learn a great deal from
comparing, say, Venizelos's ineffective liberal associations with
Yrigoyen's more effective Radical committees or with Stamboliiski's druzhbi. These different types of local party organisation,
which at first sight seem so disparate, become comparable once
one realises that they appeared at approximately the same time
and that they were all organisational means of incorporating new
strata into the post-oligarchic parliamentary arena.
Moreover, such a cross-regional approach is particularly useful
in the field of South European studies, where Greece is often
treated on a par with Italy and Spain without due emphasis on the
vast gap between Greece and the rest of southern Europe - a gap
which is due to Greece's patrimonial legacy from Ottoman rule,
and to the fact that, unlike northern Italy or Spain, she did not
experience any indigenous industrialisation in the nineteenth
century. These two fundamental differences, which are the result
of centuries of strikingly different historical trajectories, cannot
always be minimised by mere geographical proximity or the recent
integration of Greece into the European Common Market. If such
differences are taken into account, then for certain research
purposes it makes more sense to compare Greece with Chile than
with Italy or Spain.
These arguments do not, of course, dismiss the usefulness of
area studies; they simply emphasise the need to complement them
with comparative, cross-regional studies. The latter can be particularly fruitful if they succeed in striking the correct balance between
totally context-bound approaches which refuse comparisons across
geographical regions or cultures, and at the other extreme crosscultural studies of the ahistorical, positivistic kind which set out to
establish correlations between 'variables' without taking into
General Conclusion
223
account long-term historical developments and the over-all social

context in which such variables are embedded. This work has
attempted to strike such a balance, and in that sense it is a plea for
the type of historically-oriented comparative analysis which argues
that certain structurally generated similarities can provide a sound
basis for comparing societies which are geographically distant and
culturally dissimilar.
1. The countries comprising the Balkan peninsula are Greece, Bulgaria, most of
Yugoslavia and Albania. Tiny Albania, who only gained her independence in the
twentieth century, is not included in this study. Nor is Rumania which, although
often considered a Balkan country, strictly speaking lies beyond the northern
boundary of the Balkan peninsula (which runs along the river Danube). (Cf.
Encyclopaedia Britannica, Balkan Peninsula, vol. 3, p. 5.)
2. The term 'southern cone' as used here will refer only to the major southern
Latin American countries: Argentina, Chile and Brazil. Although Brazil, given her
huge size and her more heterogeneous social structure, is not strictly comparable
with the other two countries, southern Brazil is quite similar to Chile and Argentina
in terms of levels of urbanisation, industrialisation, and so on. Moreover, in terms
of the dominant economic and political role played by the southern states in
twentieth-century Brazil, certain basic patterns of overall political transition are
also comparable to those of Chile and Argentina. (Cf. on this point G. O'Donnel,
Modernisation and Bureaucratic Authoritarianism, University of California, Berkeley, 1973). Although therefore, as far as Latin America is concerned, the major
focus in this study is on Argentina and Chile, there will be certain systematic
references to Brazil also.
3. On the specificity of western European absolutism cf. Perry Anderson,
Lineages of the Absolutist State (London: New Left Publications, 1974) pp. 397ff.
4. The 'late-late' label, which is used extensively in development theory, aims at
distinguishing the (compared to England) relatively late European industrialisers
(Germany, France) from those semi-peripheral societies which only experienced
large-scale industrialisation in the post-1929 period. (Cf. on this point A. Hirschman, A Bias for Hope (Yale University Press, New Haven 1970) ch. 3.
5. Cf. particularly Gunder Frank's early study, Capitalism and Underdevelopment in Latin America (New York: Monthly Review Press, 1969) and E. Wallerstein, The Modern World System (New York and London: Academic Press, 1974
and 1980) vols. 1, 2.
6. To be more specific, I disagree first of all with the Frank/Wallerstein definition
of capitalism which more or less equates capitalist development with the commercialisation of an economy. In this paper, a narrower definition of capitalism has
been adopted: capitalism here refers to a mode of production characterised by the
large-scale use of wage labour and the private ownership of the major means of
production. Thus the integration of an economy into the world market and/or the
commercialisation of some of its sectors do not automatically make it capitalist.
(Cf. on this point M. Dobb, Studies in the Development of Capitalism (New York:
International Publishers, 1968) and E. Laclau, 'Feudalism and capitalism in Latin
America', New Left Review, May-June 1971.)
Apart from the concept of capitalism, another problematic dimension of the
centre-periphery distinction as used in the Frank/Wallerstein school of thought is
224
225
the idea of a world economic system which plays a crucial role in shaping political
and economic developments in individual nation/states. From that point of view it
is only by seeing nation-states as parts occupying positions of domination/
dependence within the international capitalist system that one can make sense of
events and developments on the national level.
The emphasis on a world economy is a healthy reaction to the neo-evolutionist,
functionalist tendency of conceptualising societies as isolated units placed along a
continuum according to their degree of proximity to the fully 'modernised' western
European bourgeois democracies. The world-economy approach demands that
serious consideration is given to the complex network of economic and political
relationships between and among nation-states as this has developed historically
through the rise and fall of western colonialism, the emergence of neo-colonialism,
of multinational capitalism, and so on. On the other hand, theorists like Frank and
Wallerstein put so much emphasis on the international context that the relative
autonomy of nation-states is underemphasised or neglected altogether. At
best, national processes and forces - if considered at all - are portrayed as passive
outcomes of world-economic processes and contradictions.
7. As far as Latin America is concerned, many writers have stressed the negative
impact of the Iberian patrimonial/despotic legacy on the functioning of parliamentary institutions in these societies. (Cf. for example R.M. Morse, 'The heritage of
Latin America: the distinct tradition', in H.J. Wiarda (ed.), Politics and Social
Change in Latin America (Amherst: Massachussetts University Press, 1973); M.
Sagatti, Spanish Bureaucratic Patrimonialism in America Politics of Modernisation
Series no. 1 (University of California, 1966); C. Veliz, The Centralist Tradition in
Latin America (Princeton University Press, 1980) and H.J. Wiarda, 'Toward a
framework for the study of political change in the Iberia-Latin tradition: The
corporative model', World Politics, January 1973. Some of the above studies
emphasise more cultural/ideational aspects of Iberian patrimonialism, while others
focus rather on more structural features - such as the incorporative character of
state controls.)
An even stronger case can be made out for the effect that sultanic rule, and the
Ottoman legacy generally, had on the malfunctioning of subsequent parliamentary
democracy in the Balkans. It is not by chance that Weber coined the term sultanism
when analysing those extreme forms of patrimonial domination where officials are
totally subjected to the ruler's absolute will. (Cf. Max Weber, Economy and
Society, eds G. Roth and C. Wittich, University of California Press, 1978, pp.
231-2.)
In the literature on both Latin American and Balkan societies, arguments about
historical legacy are usually associated with a culturalist approach to the study of
politics: the authoritarian nature of contemporary political institutions is explained
in terms of a distinct/unique cultural tradition (for example, Hispanic, Ottoman)
which has persisted from the sixteenth century to the present day. However, a
comparative study, by establishing similarities in political structures between such
culturally disparate cases as the Balkans and the Latin American regions, makes it
clear that the factor they had in common in their pre-independence past is not a
unique set of 'core values', but a type of political framework which inhibited the
development of relatively autonomous corps intermediaires between rulers and
ruled.
For instance, the Ottoman patrimonial system of power was organised with a
view to preventing the creation of a strong landed aristocracy which could have
challenged the absolute authority of the sultan. In contrast to western European
attitudes to land ownership, the Ottoman empire saw all land as belonging, in
theory at least, to Allah and his representative on earth, the sultan. Despite the de
226
facto existence of private lands, generally speaking all cultivated land came under
the timar landholding system- a system resembling the Carolinginan benefice more
than the medieval fief. Timar holders (the spahis) held no ownership rights over the
land; they simply had a non-hereditary right to a portion of the produce, in

exchange for which they were obliged to provide both administrative and military
services to the Porte. During periods of decline or weak central rule, officials would
of course manage to acquire de facto ownership of their timars, but such ownership
was never legalised.
When the state was strong therefore, sultanic despotism kept in check not only
the aristocracy's aspirations to landownership, but equally so any designs it might
have had on controlling the state apparatus. The appointment of slaves, eunuchs,
Jews, Christians, Greeks, and so on (instead of nobles) to key administrative posts
ensured total subservience to the sultan's wishes.
Iberian rule in Latin America set no such restrictions on landownership, and this
difference is relevant for understanding the post-independence developments in the
two cases. However, imposing the traditions of absolutist rule developed in the
motherland in a colonial context where political feudalism was absent eventually
resulted in a type of state centralism which could not easily tolerate the development of autonomous groups and associations. As Claudio Veliz says (in The
Centralist Tradition, p. 10):
In the Indies the fledgling Renaissance State of the crown of Castille was able to
bring into being a centralised political structure without the hindrance of feudal
traditions or the opposition of a baronial periphery.
Another point that needs to be made briefly a propos the historical legacy
arguments is that pre-independence structures, although relevant, cannot by
themselves explain the present organisation and functioning of the semi-peripheral
state. Beyong trotting out purely culturalist explanations it is necessary to show
how and why authoritarian elements inherited from the past could persist, albeit in
changing forms, until the present day. For this the political structures in these
societies must be studied to see how they evolved, and particularly how they relate
to the development of industrial capitalism.
8. For the connection between this type of balance and the development of
democratic institutions in the West, cf. Otto Hintze, 'The preconditions of
representative government in the context of world history', in F. Gilbert (ed.), The
Historical Essays of Otto Hintze (New York: Oxford University Press, 1975). Cf. also
M. Weber, The City (London: Macmillan Press, 1958)
Chapter!
1. Oligarchic parliamentarism naturally took different forms in the various

countries we are concerned with. For instance in Chile from 1891 to 1924 and in
Greece from 1864 (particularly 1875) to 1909, parliament played a more important
role in the oligarchic polity than in Brazil during the Old Republic (1889-1930) or
in Argentina from 1853 to 1912. In Bulgaria, King Alexander as the country's first
royal master attempted to rule dictatorially for a short while, although the country
had adopted highly democratic institutions from the very beginning of her
227
independence (February 1878). He was obliged though to change course in 1883.

What matters is that in all the cases under consideration parliamentary institutions
were a significant feature of the political system and did provide a restricted arena
within which the upper classes could compete for political power.
2. As noted already, capitalism in this book refers to a mode of production
characterised by the large-scale use of wage labour and the private ownership of the major means of production. For a debate on the meaning of
capitalism and the relevance of diverging definitions in explaining the transition
from feudalism to Cl\pitalism in western Europe cf. R. Hilton (ed.), The Transition
from Feudalism to Capitalism (London: New Left Books, 1976).
3. Cf. Sydney Verba, The Civic Culture (Princeton, New Jersey: Princeton
University Press, 1963).
4. Cf. R. Bendix, Nation-building and Citizenship (New York: Action Books,
1969); and P. Nett!, Political Mobilisation (London: Faber, 1967).
5. I am fully aware of the risks of overgeneralisation in global, undifferentiated
comparisons of developments in the 'West' and the 'semi-periphery'. However,
despite the obvious pitfalls of bringing countries with quite dissimilar socioeconomic trajectories under the same label I think that such a macro-historical
comparative approach is useful and even indispensable - provided one always
remembers (a) the heuristic character of the exercise (the aim being to show certain
crucial differences between developments in the centre and the parliamentary
semi-periphery in the sharpest possible contrast); (b) that such generalisations are
in no way meant as 'iron laws' but merely as tendencies; and (c) that whether or not
it is appropriate to talk about 'western' patterns of development in general depends
on the type of the problem being dealt with and the level of abstraction it imposes
on the researcher. In other words, in order to assess just how crucial certain
differences or similarities between countries are, one cannot merely look at 'the
facts'. To determine what these facts are and what weight they should be given
demands serious consideration of the nature of the problematic, the level of
analysis adopted, and so on.
Finally, it must also be kept in mind that the overcautious approach of certain
historians who categorically refuse to 'lump together' countries which portray
differences in their socio-economic development leads- when taken to its extreme
-to a denial of the comparative method tout court. (For an attempt to differentiate
broad patterns of sociopolitical developments in western Europe, see Chapter 3,
section 3.3.3B.)
6. Cf. C. Furtado, Economic Development of Latin America (Cambridge
University Press, 1970) 2nd edn, p. 108; also C.F. Diaz-Aiejandro, Essays on the
Economic History of Argentina (New Haven and London: Yale University Press,
1970) pp. 209ff.
7. Cf. N. Mouzelis, Modern Greece: Facets of Underdevelopment (London:
Macmillan Press, 1978) p. 120.
8. For a presentation of the neo-evolutionist thesis, cf. W.W. Rostow, The
Stages of Economic Growth (Cambridge: Cambridge University Press, 1962). For
an early critique of neo-evolutionist theories cf. A. Gunder Frank, 'Sociology of
development and the underdevelopment of sociology', Catalyst, No.3, University
of Buffalo, 1967; also H. Bernstein, 'Modernisation theory and the sociological
study of development', Journal of Development Studies, vol. 7, 1971.
9. Some representative works of the neo-Marxist school are P. Baran, The
Political Economy of Growth (New York: Monthly Review Press, 1957); A.
Gunder Frank, Capitalism and Underdevelopment in Latin America: Historical
Studies of Chile and Brazil (New York: Monthly Review Press, 1969); S. Amin,
228
L'Accumulation a /'chelle Mondiale (Paris: Anthropos, 1970). Of course, the

general label of neo-Marxism covers a variety of theoretical orientations and
sub-schools, and it is not possible to deal with such differentiations here. For an
article which does so for Latin America cf. P. O'Brien, 'A critique of Latin
American theories of dependency', in I. Oxaal et al. Beyond the Sociology of
Development (London: Routledge, 1975) pp. 7-27.
10. For such a distinction cf. P. Schneider eta/. 'Modernisation and development', in Comparative Studies in Society and History (June 1972) p. 340.
11. It is not only Marxist-oriented writers who tend to underemphasise the
importance of parliamentary institutions in the periphery and semi-periphery.
Non-Marxist political scientists, especially those who focus on political corruption
and the clientelistic aspects of peripheral politics, tend also to look at parliamentary
structures as superficial 'forms' and consider the 'substance' of politics as residing in
the extensive clientelistic networks and practices which continue regardless of
regime forms and changes. Cf. for instance K. Legg, Politics in Modern Greece
(Stanford: Stanford University Press, 1969).
12. In all these countries the turbulent post-independence period of constant
civil wars was relatively short, the central state managing quite early to discipline
local elites and to impose a minimum of national unity and constitutional order
(this happened around the 1830s in Chile, in the 1840s in Brazil, and in the 1850s
and 1860s in Argentina).
13. Cf. Shlomo Avineri, Karl Marx on Colonialism and Modernisation (New
York: Anchor Books, 1969).
14. On this cf. V. Kremmidas, Introduction to the History of Modern Greek
Society (in Greek) (Athens, 1975) p. 201.
15. Cf. N. Mouzelis, Modern Greece, 1978, pp. 17ff.
16. For an analysis of western economic penetration of the Balkans cf. L.
Stavrianos, The Balkans since 1453 (New York: Holt, Rinehart & Winston, 1958).
17. Cf. C. Furtado, Economic Development, 1970, p. 58. On the role of foreign
capital in the southern cone countries of Latin America cf. J.F. Rippy, British
Investments in Latin America 1822-1949 (Minneapolis, 1959); A. G. Ford, Britishowned Railways in Argentina (Austin, Texas, 1974); R. Graham, Britain and the
Onset of Modernisation in Brazil 1850-1914 (Cambridge, 1968); B. Albert,
South America and the World Economy (London: Macmillan Press, 1983).
18. Cf. A.S. Banks, Cross-polity Time-series Data (Cambridge, Mass.: The MIT
Press, 1971) p. 48 for English figures, and p. 69 for German figures. For more
detailed statistics focussing exclusively on Europe cf. B.R. Mitchell, European
Historical Statistics 1750-1970 (London: Macmillan Press, 1975).
19. A.S. Banks, Cross-polity, 1971, p. 62 (Chile) and p. 59 (Argentina).
20. Ibid., p. 59.
21. Cf. G. O'Donnel, Modernisation and Bureaucratic Authoritarianism: Studies
in Southern American Politics, (Berkeley: Institute of International Studies,
University of California, 1973).
22. A.S. Banks, Cross-polity, 1971, p. 72.
23. Ibid., p. 60.
24. Ibid, p. 90.
25. For an analysis of the linkages between the growth of the export sectors and
rapid urbanisation cf. R. M. Morse, Las ciudades Latinoamericanas (Mexico D. F.,
1973).
26. For a discussion of the linkages between the growth of the export sectors and
the development of state institutions cf. Oscar Oszlak, 'The historical formation of
the state in Latin America', Latin American Research Review, vol. 16, no. 2, 1981.
27. Since in the semi-periphery capitalism developed in a more abrupt and
229
dependent manner than in the centre, it is not surprising that semi-peripheral

capitalist economies are less self-regulating than those of the centre. To give a
simple and obvious example: the weaker development of a capital market in
semi-peripheral economies means that the state is forced to assume financial/credit
functions which in a more developed capitalist economy are performed by the stock
exchange- in which case the state intervenes indirectly to regulate stock-exchange
operations, rather than directly to provide substitute solutions for their weak
development.
28. Cf Steven Topik, 'The evolution of the economic role of the Brazilian state
1889-1930', Journal of Latin American Studies, November 1979, pp. 332ff.
29. Ibid., p. 327.
30. Cf. G. Dertilis, Social Change and Military Intervention in Politics: Greece
1881-1928, Ph.D. thesis, University of Sheffield, 1976, Table xiv.
31. Cf. C. Tsoukalas, 'The reforms ofTrikoupis', in History of the Greek Nation:
Modern Hellenism 1881-1913 (in Greek) (Athens: Ekdotiki Athinon, 1977) p. 13.
32. The rate of increase in public employment during the period was 32 per cent
and 100 per cent above the figures for industry and mining/agriculture respectively.
Cf. B. Loveman, Chile (New York: Oxford University Press, 1979) p. 259.
33. L. Stavrianos, Balkans, 1958, p. 607.
34. In calculations of this sort the reader should never forget that the comparison
is rather problematic: the pre-industrial western European state was not only
quantitatively but also qualitatively different from the pre-industrial semiperipheral state. Statistical comparisons which neglect the specific historical
trajectories as well as the over-all national and international context around the
phenomena concerned may well lead to fallacious conclusions. However, the
historical-comparative data put forward in this chapter have a very modest and
limited purpose: they are simply meant to indicate that in the parliamentary
semi-periphery (contrary to the capitalist centre) the state, the market, and the city
had developed considerably before the dominance of industrial capitalism.
35. Cf. League of Nations, Statistical Yearbook 1927 (Geneva, 1928) Table 7.
36. A.S. Banks, Cross-polity, 1971, segment 3. Concerning the margin of error
in converting national currencies into dollars, Banks writes (on p. xviii):
Revenue and expenditure data, particularly when expressed in U.S. dollar
equivalents, are particularly susceptible to both random and systematic error ...
Since the British pound sterling was the principal unit of international exchange
in the pre-World War I period, most data for the period were assembled
accordingly and converted to dollar equivalents at the rate of 4.87 dollars per
pound. Some data for 1919-1939 and most data for 1946-1966 were assembled
by means of direct conversion to dollar equivalents.
37. As far as state personnel is concerned one gets some idea of the difference if
one considers that as late as 1871 the English civil service was only employing
54000 people, at a time when England's total population was approximately 22.7
million (cf. M.G. Mulhall, Dictionary of Statistics (London: George Routledge &
Sons, 1884) p. 335). Yet in Argentina in 1920, out of a population of approximately
9 million, 107000 were employees in the state bureaucracy (cf. League of Nations,
Statistical Yearbook 1927 (Geneva, 1928) Table 7). For central government
revenue and expenditure figures in eighteenth and nineteenth-century Europe, cf.
B.R. Mitchell, European Statistics 1975, pp. 370ff.
38. A.S. Banks, Cross-polity, 1971, p. 217 and p. 233.
39. Ibid., pp. 208ff.
40. Ibid., p. 217
41. Ibid., pp. 212ff.
230
42. Cf. A. Rouquie, L'etat militaire en Amerique Latine (Paris: Seuil, 1982) pp.
120ff.
43. In writing this section I have relied extensively on the works of A.A. Boron,
The Formation and Crisis of the Liberal State in Argentina 1880-1930, Ph.D. thesis,
Harvard University, 1976; and D. Rock, Politics in Argentina 1890-1930: The Rise
and Fall of Radicalism(Cambridge: Cambridge university Press, 1975).
44. For instance, between 1881 and 1911land values went up 218 per cent in the
cereal-growing provinces. Cf. James Scobie, Revolution on the Pampas: A Social
History of Argentine Wheat 1860-1910 (Austin: University of Texas Press, 1964) p.
51.
45. Cf. Jacinto Oddone, La Burguesia Terrateniente Argentina (Buenos Aires:
La Vanguardia, 1967) pp. 185-6, quoted in A.A. Boron, Formation and Crisis,
1976, p. 230.
46. A.A. Boron, ibid., p. 231.
47. In the first national census of 1869 the number of citizens listed as eligible to
vote was 333 725. Of these not more than ten per cent voted in the national
elections (cf. R.J. Walter, The Socialist Party of Argentina (Austin: Institute of
Latin American Studies, University of Texas, 1977) p. 7).
48. A.A. Boron, Formation and Crisis, 1976, p. 334.
49. B. Albert, South America and the World Economy, 1983.
50. D. Rock, Politics, 1975, p. 267.
51. Cf. A. Rouquie, Pouvoir militaire et societe politique en Ia Republique
Argentine (Paris: Presses de Ia Fondation nationale des sciences politiques, 1978) pp.
700ff. Other historians ascribe only minor importance to army agitation as a factor
leading to the initial opening-up of Argentina's oligarchic parliamentary system.
52. For data indicating the considerable broadening of political participation since
the 1916 election cf. G. Germani, Politico y Sociedad en Una Epoca de Transicion
(Buenos Aires: Raigal, 1955) pp. 225ff.
53. Cf. Peter Smith, Politics and Beef in Argentina (New York: Columbia
54. Cf. C. Solberg, 'Rural unrest and agrarian policy in Argentina 1912-1930',
Journal of Inter-American Studies, January 1971, pp. 52ff.
55. On this point cf. D. Rock, Politics, 1975.
56. Only in 1924 was the party able to elect deputies (three of them) outside the
capital. But even such a modest advance was lost again by 1928. For a general
account of Argentina's Socialist Party cf. R.J. Walter, Socialist Party, 1977.
57. Cf. D. Rock, Politics, 1975, pp. 57-8.
58. Thirty-four radical interventions took place during the fourteen years of
radical rule. A.A. Boron,Formation and Crisis, 1976, p. 581.
59. D. Rock, Politics, 1975, p. 242.
60. Ibid., p. 103.
61. For a development of this point cf. Chapter 2.
62. Cf. Anne Potter, 'The failure of democracy in Argentina 1916-1930: An
institutional perspective', Journal of Latin American Studies, vol. 13, May 1981.
63. There are various conflicting interpretations of the 1930 military coup and
the advent of 'fraudulent parliamentarism'. Cf. J.J. Johnson,Political Change in
Latin America: The Emergence of the Middle Classes (Stanford: Stanford University Press, 1958) pp. 94-127; D. Rock, Politics, 1975; Peter Smith, Politics and Beef
1969; and Anne Potter, 'The failure of democracy', 1971.
64. Cf. on this point A.A. Boron, Formation and Crisis, 1976, pp. 616ff.
65. Peron also received support from the urban marginals, the rural poor, and
from women- that is, from people who were previously politically disenfranchised.
231
Cf. Walter Little, 'Electoral aspects of Per6nism 1946-1954', Journal of Inter-
American Studies, August 1973.
66. Cf. M.L. Connif, 'The tenentes in power: A new perspective of the Brazilian
revolutionof 1930', Journal of Latin American Studies, May 1978, p. 82.
67, Cf. P. Flynn, Brazil: A Political Analysis (London: Westview Press, 1978} p.
59.
68. Cf. M.L. Connif, Urban Politics in Brazil: The Rise of Populism 1925-1945
(University of Pittsburg Press, 1981)
69. According to P. Drake, quoting from Socialism and Populism in Chile
1932-52 (Chicago/London: University of Illinois Press, 1958) p. 51, the ideologies
and organisations of the various labour organisations
paled before the appeal of Alessandri. The workers, even from FOCH,
responded more to Alessandri's oratory and to his emotional identification with
the common man. In extreme cases, some workers believed that Alessandri
personally would ease their toil and feed their families, others knelt to kiss his
hand and brought sick children to be cured by his touch. For many workers, the
messianic Alessandri represented a political awakening, for others, a distraction, but for nearly all he appeared as the most tangible, viable and exciting
alternative in 1920.
70. As in all similar circumstances, aside from the problem of social legislation
the army had its own specific grievances which pushed it into social intervention.
Cf. B. Loveman, Chile, 1979, pp. 240ff.
71. On the broadening of political participation and the entrance of new groups
into the political game cf. F.M. Nunn, The Military in Chilean History (University
of New Mexico Press, 1976) pp. 164ff.
72. P. Drake, Socialism and Populism, p. 3.
73. Ibid., chapter 10.
74. The above interpretation of Latin American inter-war populism as a means
of consolidating post-oligarchic rule does not mean that the actors involved were
always conscious of such goals. The analysis is based more on historical hindsight
and the concept of unintended consequences than on conscious goal elaborations at
the time the events occurred.
75. This is a chance other eastern European peasants never had since they never
experienced an equivalent power vacuum (cf. P. Anderson, Lineages of the
Absolutist State (London: New Left Publications, 1974), and neither of course did
the Latin American peasantry.
76. Cf. W.E. Moore, Economic Demography of Eastern and Southern Europe
(Geneva: League of Nations, 1945) pp. 77-91 and 230--52.
In the Hapsburg parts of Yugoslavia the considerable importance of big landed
property declined after the inter-war agrarian reforms. The same happened in
Rumania which, due to the indirect Turkish rule of her southern territories, had a
relatively strong landowning class. Only tiny southern Albania retained its pattern
of big landed property until the end of the inter-war period.
77. For the inter-war agrarian reforms in the Balkans cf. D. Mitrany, Marx
against the Peasant (London: Weidenfeld & Nicolson, 1951); C. Evelpides, Les
Etats Balkaniques, Paris, 1930. For a short survey of land tenure systems in various
Balkan countries during the inter-war years cf. W.E. Moore, Economic Demography 1945, pp. 210--67.
.
78. For instance, as late as 1930 Bulgaria had 80 per cent of its labour force in
agriculture and only 8 per cent in industry. In Rumania at the same time the figures
were 78 per cent and 7 per cent, and in Yugoslavia 79 per cent and 11 per cent. (Cf.
232
LT. Berend and G. Ranki, Economic Development in East-Central Europe in the

19th and 20th Centuries, (New York and London: Columbia University Press, 1974)
p. 306). In Greece in 1920, the percentage of the labour force employed in
agriculture was 49.6 and in industry 15.7 (cf. League of Nations, Statistical
Yearbook 1927, p. 38). In Chile, the equivalent figures are 36.6 per cent and 24.3
per cent (League of Nations, Statistical Yearbook 1926, p. 24). In Argentina, the
proportion of labour employed in agriculture was even smaller of course.
79. For instance, around 1938 industry's contribution to the GNP was approximately 17 per cent in Bulgaria and 22 per cent in Yugoslavia (cf. Berend and
Ranki, Economic Development, 1974, p. 307), and approximately 21 per cent in
Greece (cf. B.R. Mitchell, European Statistics, 1975, p. 809).
80. For statistics on this point see section 1.3.1.
81. On the zadruga family system cf. P. E. Mosely, 'The peasant family: The
zadruga or communal joint family in the Balkans and its recent evolution', in C.F.
Ware (ed.), The Cultural Approach to History (New York, 1940). For a contrast of
Greek and Bulgarian family institutions in the inter-war period cf. K.D. Karavidas,
Agrotika (in Greek) (Athens, 1931) pp. 201-68.
82. Cf. D. Daniilidis, Modern Greek Society and Economy (in Greek) (Athens,
1934) pp. 126--42.
83. D. Mitrany, Marx, 1951, pp. 118-19.
84. As regards modes of political mobilisation, it is quite clear that 'egalitarianism' is highly relevant to an explanation of Bulgaria's agrarian populism. The fact
that absence of great social inequalities can, in certain circumstances, favour
peasant solidarity and the propensity for collective action is a point frequently
stressed in studies of peasant politics. (Cf. for instance S.M. Lipset, Agrarian
Socialism (New York, 1950); or E. Wolf, 'On peasant rebellions', in T. Shanin
(ed.), Peasants and Peasant Societies (Harmondsworth: Penguin, 1971) pp. 26474.) The connection between egalitarianism and agrarian collective action is also
evident in Greece. As argued below, the Greek agrarian party, although much less
powerful than the Bulgarian one, was most successful in the northern province of
Greece where, after the agrarian reform, the villages had a more egalitarian social
structure.
85. Cf. on this point N. Pasic, 'Factors in the formation of nations in the Balkans
and among the South Slavs', International Social Sciences Journal, no. 3, 1971.
86. On the foreign origin of the Bulgarian bourgeoisie cf. T. Stoianovich, 'The
conquering Balkan Orthodox merchant', Journal of Economic History, 1960; cf.
also N. Pasic, 'Factors', 1971, pp. 419ff.
87. Cf. K.D. Karavidas, Agrotika, 1931, pp. 125-6.
88. A. Angelopoulos, 'Les finances publiques d'etats Balkaniques', Les Balkans, vol. 2, September 1933, pp. 674-9.
89. Cf. on this point J. Petropoulos, Politics and Statecraft in the Kingdom of
Greece 1833-1843 (Princeton: 1968). For the early struggles between the state and
various local elites cf. N. Diamandouros, Political Modernisation: Social Conflict
and Cultural Cleavage in the Formation of the Modern Greek State 1821-28, Ph.D.
thesis, Columbia University, 1972.
90. E. Dicey, The Peasant State (London: 1894) p. 145.
91. For instance, the first two railway lines in the Balkans (from Varna to
Ruschuk, and from Constantsa to Cernadova) were constructed after the Crimean
War in conformity with British interests to tap the commerce of the Danube area;
and proposals for trans-Balkan railways were promoted or boycotted according to
the strategic interests of the powers chiefly concerned. Cf. L. Stavrianos, Balkans,
1958, p. 417; also A.J. May, 'Trans-Balkan railway schemes', Journal of Modern
History, December 1936.
233
92. On inter-war debts of the Balkan states cf. Royal Institute of International
Affairs, The Balkan States, London, 1936.
93. Cf. W.E. Moore, Economic Demography, 1945, pp. 17-28
94. Ibid., pp. 63 and 71-72. For statistics on number of people per 100 hectares
of cultivated land, cf. J. Tomasevich, Peasants, Politics and Economic Development
in Yugoslavia (Stanford: Stanford University Press, 1955) p. 309.
95. For an analysis of Balkan peasant ideology cf. Branko Peselz, Peasant
Movements in South-Eastern Europe, unpubl. Ph.D. thesis, Georgetown University, 1950; G.M. Dimitrov, 'Agrarianism', in F. Gross, European Ideologies (New
York, 1948); and D. Mitrany, Marx, 1951.
96. The 'elite' intelligentsia, consisting of state bureaucrats and politicians, was
known in Bulgaria as the 'partisan' or 'parasitic' intelligentsia- and it was quite
distinct from the group of teachers who started the peasant movement. Cf. J.D.
Bell, Peasants in Power: Alexander Stamboliiski and the Bulgarian National Union
1899-1928 (Princeton: Princeton University Press, 1977) p. 17.
97. For a brief overview of Russian populist ideas and movements cf. M.
Canovan, Populism (London: Junction Books, 1981) ch. 2.
98. J. Bell, Peasants, 1977, p. 77.
99. On the tendency to land concentration towards the end of the nineteenth
century in Bulgaria cf. Petro Kunin, The Agrarian and Peasant Problem in Bulgaria
(in Bulgarian) (Sofia: 1971) ch. l.
100. Cf. L. Stavrianos, Balkans, 1958, p. 647.
101. J. Bell, Peasants, 1977, p. 182.
102. Stamboliiski tried to curb the monarchy's power position, his ultimate aim
being to reduce the king into a figurehead.
103. The dictatorial regimes in Bulgaria and Yugoslavia cannot, of course, be
explained exclusively or even largely in terms of the 'agrarian threat'. The
Depression, and the rise of authoritarianism in Italy and Germany, for instance,
are also relevant for understanding the royal dictatorships of the late inter-war
period. Cf. L. Stavrianos, Balkans, 1958, pp. 593ff.
104. The national lands amounted to 35 per cent of all cultivated land. Cf. C.
Vergopoulos, The Agrarian Problem in Greece (in Greek) (Athens: Exantas, 1975)
p. 82.
105. Cf. J. Petropoulos, Politics and Statecraft, 1968, pp. 236-8.
106. Cf. C. Vergopoulos, Agrarian Problem, 1975, pp. 80ff.
107. In this first land reform 662500 acres were distributed in 357217 individual
lots. At this time the agricultural population of Greece numbered 254 000 families.
(Cf. N. Vernicos, L'Evolution et les Structures de Ia Production Agricole en Grece
(Dosier de Recherche, Universite de Paris VIII, 1973).) It should be noted that
even before the 1871 land distribution a large portion of the cultivated land had
come into the hands of small peasant cultivators. Aside from the 1835 law, another
means of bringing this about was that those working on the national lands had the
right to claim the ownership of their plot if they had carried out permanent
improvements. It is not known, however, to what extent this right was actually
invoked.
108. Thessaly and Arta were incorporated in 1881, Macedonia in 1913, and
Thrace in 1923. Cf. D. Dakin, The Unification of Greece 1770-1923 (London:
Ernest Benn, 1972).
109. Cf. on this point Chapter 4, last section (4.4).
110. In 1922 Greece was defeated by Turkey in Asia Minor. As a consequence of
this more than one million Asia Minor Greeks poured into mainland Greece as
refugees.
111. Cf. D. Stephanidis, Agrarian Policy (in Greek) (Athens: 1948), p. 107.
234
112. As mentioned already, Greece's urbanisation was not as high as Argentina's or Chile's but much higher than in the northern Balkans.
113. Cf. T. Stoianovich, 'The conquering Balkan', 1960, pp. 296--73.
114. Bulgarian and other Slav merchants started to challenge the Greek
commercial supremacy in the eastern Balkans only after the Treaty of Adrianople
in 1829 which permitted the Danubian principalities to engage in international
trade.
115. Cf. T. Stoianovich, 'The conquering Balkan', 1960, p. 311.
116. Greece's cultural hegemony in the eighteenth-century Balkans was due not
only to its powerful merchant class. The position of the Greek Orthodox Church
was another contributing factor. Before the establishment of autocephalous Slav
national Churches, the Greek Patriarchate in Constantinople administered the
religious lives of all Orthodox Christians, Greek or non-Greek. All important
ecclesiastical positions were monopolised by Greeks, the Greek liturgy was
imposed in all churches, and Greek was the language of instruction in schools. Cf.
L. Stavrianos, Balkans, 1958, pp. 368ff.
117. The Greek War of Independence (1821-7) brought a temporary halt to the
commercial activities of the Greek merchant class. But this setback- at least as far
as the Mediterranean trade was concerned - was overcome and trade flourished
again after independence. As far as overland inter-Balkan trade was concerned, the
emergence of Balkan nationalism, and other unfavourable factors, led to its more
or less permanent decline at the beginning of the nineteenth century. (Cf. T.
Stoianovich, 'The conquering Balkan', 1960, p. 312.)
118. Cf. C. Tsoukalas, Dependence et Reproduction: Le role de l'Appareil
Scolaire dans une Formation Transterritoriale Ph.D. thesis, University of Paris,
1975, Part I, ch. 2; also G. Dertilis, Social Change, 1976.
119. With respect to Greece, cf. Great Encyclopaedia, volume 'Greece', p. 234.
With respect to Bulgarian immigration cf. J. Roucek, 'Les Boulgars d'Amerique',
Balkans, vol. 9, pp. 55-70.
120. J. Campbell and P. Sherrard, Modern Greece (London: Ernest Benn, 1968)
p. 98.
121. The economic crisis had actually started earlier, with the collapse of the
international market for Greek currants in 1893, but it was greatly aggravated by
the war. Towards the end of the decade Greece had to be declared bankrupt. An
international financial commission was appointed on Greek territory which controlled the main Greek sources of revenue and had a major say in currency issues, new
loans, and in the entire range of state finances. Cf. E. J. Tsouderos, Le Relevement
Economique de Ia Grece (Paris: Berger-Levrault, 1920).
122. For an accurate historical account and explanation of the 1909 military coup
cf. V. Papacosmas, The Greek Military Revolt of 1909, Ph.D. thesis, Indiana
University, 1970; and G. Dertilis, Social Change 1976.
123. Cf. on this point T. Veremis, The Greek Army in Politics Ph.D. thesis,
Oxford University, 1974, p. 29.
124. Such a disillusionment was very clearly reflected in the press of the period,
Cf. V. Papacosmas, Greek Revolt, 1970, pp. 55ff.
125. For statistics on the changing composition of the class origin of members of
parliament after 1909 cf. D. Kitsikis, 'L'evolution de !'elite politique Greque', in
M.B. Kiray (ed.), Social Stratification and Development in the Mediterranean basin
(Hague Institute of Social Studies Publications, 1973) and K. Legg, Politics in
Modern Greece, 1969, ch. 5.
126. Cf. N. Mouzelis, Modern Greece, 1978, pp. lOlff.
127. Cf. Llewellyn-Smith, The Ionian Vision (London: 1975).
128. For more details on these developments cf. N. Mouzelis, 'Capitalism and
235
the development of the Greek state', in R. Sease (ed.), The State in Western Europe
(London: Croom Helm, 1980) pp. 248ff.
129. For a development of these points cf. N. Mouzelis, 'Class and clientelistic
politics: The case of Greece', Sociological Review, November 1978.
130. Cf. on this point George Mavrogordatos, Stillborn Republic: Social Coalitions and Party Strategies in Greece 1922-1936 (Berkeley, Los Angeles and
London: University of California Press, 1983) p. 69.
131. The fact that this fundamental issue was personalised - in the sense that
voters were personalistically attached to Venizelos and King Constantine, the two
major charismatic figures representing the two sides of the conflict - does not
invalidate the point that ideological considerations were increasingly playing a large
role and setting stricter limits to clientelistic manoeuvres.
132. For all the above points cf. G. Mavrogordatos, Stillborn Republic, 1983, ch.
3.
133. Ibid., pp. 83ff.
134. In using the term 'necessitated' here I do not imply, of course, any causal
connection between 'social needs' teleologically conceived and political outcomes.
I simply mean that the existence of a very powerful landowning class on the one
hand, and of a rapidly expanding urban population on the other, in conjuncture
with the already mentioned other factors undermining the oligarchic system,
provided favourable conditions for the development of urban populism.
135. Cf. B. Moore, The Social Origins of Dictatorship and Democracy (Harmondsworth: Penguin, 1969) pp. 475-6.
136. J. M. Paige for instance emphasises the crucial importance of upper-class
controls for understanding the degree and type of peasant mobilisation in his
Agrarian Revolution: Social Movements and Export Agriculture in the Underdeveloped World (New York: Free Press, 1975).
137. The necessity for political leadership and organisation for the radical
mobilisation ofthe peasantry has been emphasised by, among others, J. S. Migdal in
his Peasants, Politics and Revolution (Princeton: Princeton University Press, 1974).
138. It is pertinent in this context to note some interesting similarities between
the political systems of Greece and Uruguay. Like the Greek one, Uruguay's
two-party system is notorious for its marked clientelistic features which persisted
until the imposition of the bureaucratic military dictatorship in 1974. (Cf. M. H. J.
Finch, A Political Economy of Uruguay since 1870 (New York: StMartin's Press,
1981) ch. 1; cf. also Francisco Panizza, 'Accumulation and consensus in post-war
Uruguay', mimeographed paper, University of Essex, 1983. For the enormous
proportions and the central role of state patronage in Uruguay cf. M. Weinstein,
Uruguay: The Politics of Failure (Westport: Greenwood Press, 1975) pp. 69ff; G.
Lindhol, Uruguay's New Path (Stockholm: Institute of lbero-American Studies,
1962) pp. 166ff.)
Moreover, Uruguay's landowning classes- like those of Greece but unlike those
of Argentina, Brazil and Chile - did not, for a variety of historical reasons, have a
strong political hold on the state in the nineteenth century. In consequence, and
given the country's rapid urbanisation and the rise of the middle classes, Battle y
Ordonez managed to come to power and introduce the socio-political reforms that
the rest of the southern cone countries were to introduce later, without any
intensive populist mobilisation before or during his rule.
The situation in Uruguay was in striking contrast with Argentina's. Yrigoyen was
swept into power by a broad popular movement in a struggle against a politically
entrenched landowning oligarchy; Battle was elected according to the 1830
constitution by the House of Congress, without any popular participation in that
process. Battle's election was the result of subtle clientelistic manipulations and of
236
oligarchic agreements. (Cf. on this point M.I. Vanger, The Model Country
(Hanover: Brandeis University Press, 1980) pp. 16-69.)
In view of this it is not surprising that both the early welfare reforms and the
spread of political rights in Uruguay were imposed from above and originated form
within the two major parties. Since Uruguay's landowning classes were politically
weak, the oligarchic system was broadened and social reforms were implemented
without any large-scale populistic disruption of the two-party clientelistic system. In
fact the transition occurred as 'smoothly' as in Greece, through the gradual
transformation and centralisation of already existing clientelistic networks. One
might argue that Uruguay's transition was even less abrupt than Greece's, in the
sense that for the latter, Venizelos brought about the post-oligarchic reforms
through the rapid creation of a new, more centralised clientelistic organisation (his
Liberal Party), whereas Battle introduced his reforms without disrupting the
two-party structure of the oligarchic political system.
139. For the origins of industrialistion during the nineteenth century cf. for
Chile, Luis Ortega, 'Acerca de los orfgenes de Ia industrializacion chilena
1860-1879', in Nueva Historia, (1981); for Argentina, cf. Adolfo Dorfman,
Evoluci6n Industrial Argentina (Buenos Aires: Editorial Losada, 1942); for
Greece, George Anastasopoulos, The History of Greek Industry (Athens: Greek
Publishing Company, 1947) 2 vols.
140. According to Dfaz-Alejandro the war, by reducing Argentina's capacity of
importing machinery and coal, slowed down the industrialisation process: 'On
balance, the shaky data for this period show a drop in manufacturing output during
1914-17. The common opinion that the war boosted industrialisation appears to be
at best a very partial view.' (Quoted from Essays on the Economic History of the
Argentine Republic (New Haven and London: Yale University Press, 1970) p. 218.)
According to Di Tella and Zymelman, the larger and more modern firms profited
from the war, whereas artisanal units (especially those linked with the construction
industry) suffered from the recession that marked the early war period and by the
drop in investments for construction activities; cf. their 'El desarrollo industrial
argentino durante Ia primera guerra mundial', Revista de Ciencias Econ6micas.
Buenos Aires, April-June, 1959.
141. Cf. C. Furtado, Economic Development of Latin America, 1970, pp.
100-106. It should also be mentioned that, especially in Argentina, there was a
small sector of industry capable of producing durable goods before 1929. Cf. R.
Cortes, 'Problemas del crecimiento industrial de Ia Argentina 1870-1914', in
Desarrollo Econ6mico, vol. 3, 1963, pp. 143-71. For the crucial importance of the
import-substitution process in Chile during and after the First World War, cf.
Gabriel Palma, 'External disequilibrium and internal industrialisation: Chile
1914-1935', paper presented to the conference on The effects of the 1929 Depression on Latin America, St Antony's College, Oxford University, Septmber 1981.
According to Palma, the decisive shift from export orientation to inward-looking
development occurred a decade before 1929, and from this point of view
the two decades following the outbreak of the First World War must be seen as
a single unit, whose principal characteristic was the instability of the external
sector and, in response to it, an attempt to carry through a radical transformation of the economy, in order to create a greater degree of local productive
autonomy. (p. 20).
Cf. also M. 1. Mamalakis, The Growth and the Structure of the Chilean Economy:
From Independence to Allende (New Haven and London: Yale University Press,
1976) chs 3-5; and H. W. Kirsch, Industrial Development in a Traditional Society:
237
The Conflict of Entrepreneurship and Modernisation in Chile (Gainesville: 1977)

chs 2-5; as well as F. H. Cardoso and E. Faletto, Dependency and Development in
Latin America (Berkeley and London: University of California Press, 1979).
142. I. T. Berend and G. Ranki, Economic Development in East-Central
Europe, 1974, pp. 171ff.
143. Cf. A. Sideris, 'Economic history of Greece', in Economic and Accounting
Encyclopaedia (in Greek), vol. 4, pp. 148ff.
144. C. Furtado, Economic Development of Latin America, 1970, pp. 47-9.
145. For example, in 1900 the railroad mileage per square mile of country was 95
in Argentina, 97 in Chile, and 29 in Brazil. In the Balkans at the same time rail
mileage was 192 in Bulgaria, 261 in Greece, and 186 in Serbia (cf. A. S. Banks,
Cross-polity, 1971, segment 3). Of course, one must take into account.the smaller
size of the Balkan countries as well as the fact that the development of railway
networks, strongly linked as it was to the growing commercialisation of the Balkans'
agrarian economies, was due also to the rival strategies of Great Power politics. On
the other hand, as will become clearer below, the Balkan economies were more
advanced in terms of infrastructural and industrial development than the relatively
slow decline of their agrarian labour force would suggest.
146. Berend and Ranki, Economic Development in East-Central Europe, 1974,
p. 248.
147. C. Furtado, Economic Development of Latin America, 1970, pp. 108ff.
148. In several of the semi-peripheral societies, protectionist policies were
operating even before the 1929 crisis, despite the generally free-market rhetoric of
their ruling oligarchies. (Cf. on this point C. F. Dfaz-Alejandro, Essays on
Economic History, 1970, p. 217.) But after 1929 protectionist policies were both
accentuated and systematised.
149. Cf. on this point Dfaz-Alejandro, Essays on Economic History, 1970, pp.
lOlff.
150. For a survey of Argentine economic development cf. Aldo Ferrer, The
Argentine Economy (Berkeley and Los Angeles: University of California Press,
1967); and Dfaz-Alejandro, Essays on Economic History, 1970.
151. For a survey of Chile's economic development cf. M. J. Mamalakis,
Chilean Economy, 1976.
152. Cf. Economic Commission for Latin America, The Process of Industrial
Development in Latin America (New York: United Nations, 1966) pp. 9ff.
153. Cf. C. Vergopoulos, 'The Greek economy from 1926 to 1935', (in Greek),
in History of the Greek Nation: The New Hellenism 1913-19 (Athens: Ekdotiki
Athinon, 1977).
154. Cf. T. Deldycke et al., The Working Population and its Structure (Brussels:
1908).
155. In 1948 Yugoslavia's labour distribution was 77.8 per cent (primary sector),
9.6 per cent (industrial) and 9.4 per cent (services) and 3.3 per cent (activities not
adequately described), according toT. Deldycke et al., Working Population, 1968,
p. 120.
156. Ibid., pp. 45ff. In the figures I have given, the primary sector includes
mining, and the industrial sector includes manufacturing, construction, electricity,
gas, water and sanitation services. In addition to the service sector, Deldycke's
statistics include a category labelled 'not adequately designed activities'. In the case
of Argentina (1947) this last category accounted for 10.1 per cent of the active
population.
157. Ibid., p. 49.
158. Ibid., p. 101.
159. C. Furtado, Economic Development of Latin America, 1970, p. 111.
160. B. R. Mitchell, European Satisfies, 1975, pp. 806ff.
238
161. For instance in 1946, the figures for the per-capita GOP originating from
industrial activity were $52 for Bulgaria, but only $18 for Bolivia and $21 for Peru.
Cf. A. S. Banks, Cross-polity, 1971, segment 8.
162. Berend and Ranki, Economic Development in East-Central Europe, 1974,
p. 307.
163. Underemployment in the Balkan countryside was aggravated by the
spectacular population growth in the peninsula - a trend that had started at the
beginning of the nineteenth century and accelerated during the inter-war years.
This resulted in the extreme fragmentation of peasant holdings. According to
W. E. Moore, approximately half of the rural population in the inter-war Balkans
was redundant (Economic Demography, 1945, pp. 63 and 71-72). For a more
optimistic calculation of labour surplus cf. N. Spulber, The Economics of Communist Eastern Europe (Cambridge, Mass.: MIT, 1957) pp. 275-6. For statistics on
numbers of people per 100 hectares of cultivated land, cf. J. Tomasevich, Peasants
and Politics, 1955, p. 309.
164. J. Rothschild, The Communist Party of Bulgaria 1883-1936 (New York:
Columbia University Press, 1959, p. 55).
165. In 1934 the communist-controlled Independent Trade Union Federation
claimed 11000 members, whereas the socialist-controlled trade unions claimed
30000 in 1929. Cf. J. Rothschild, Communist Party 1959, p. 268.
166. For reasons of convenience I shall use the designation 'Yugoslav' for both
the pre-1929 and the post-1929 periods.
167. Cf. on this point R. V. Burks, The Dynamics of Communism in Eastern
Europe (Princeton: Princeton University Press, 1961).
168. D. G. Jackson, Comintern and Peasant in Eastern Europe (New York:
Columbia University Press, 1966).
169. J. Rothschild, Communist Party, 1959, p. 276. For the dynamics of
Bulgarian communism in the post-1934 period cf. N. Oren, Bulgarian Communism
I934-I944 (New York and London: Columbia University Press, 1971).
170. J. Rothschild, Communist Party, 1959, p. 268.
171. For the history of the Greek Communist Party during the inter-war years cf.
A. Elefantis, The Promise of the Impossible Revolution: The Greek Communist
Party in the Inter-War Period (in Greek) (Athens: Olkos, 1976).
172. Cf. D. Dertouzos, The Greek Labour Movement, Ph.D. thesis, Rutgers
University 1962, pp. 116ff.
173. For instance, in 1922 trade-union membership in Bulgaria amounted to
54000 whereas in Greece it was 170000. Cf. American Labor Year Book, 1925, p.
293.
174. Cf. T. Katsanevas, The Industrial Relations System in Greece: Historical
Development and Present Structure, Ph.D. thesis, London School of Economics,
1980, pp. 73ff.
175. G. B. Leon, The Greek Socialist Movement and the First World War: The
Road to Unity (New York and Guildford, Surrey: Columbia University Press, 1976)
p. 102.
176. Ibid., ch. 7.
177. Cf. D. Dertouzos, Greek Labour Movement, 1962, chs 5 and 6.
178. Alan Angell, Politics and the Labour Movement in Chile (London: Royal
Institute of International Affairs, Oxford University Press, 1972) p. 41.
179. Union acceptance of the labour code became widespread only after the
communists changed their policy during the popular-front period. Cf. Angell, ibid.,
p. 58.
180. In addition to FOCH and the CNSL there were two other, less significant
labour confederations: one was controlled by anarcho-syndicalists and the other by
Trotskyists.
239
181. Cf: R. Alexander, Organised Labour in Latin America (New York: Free
Press, 1965) chapter on Chile.
182. A. Angell, Politics in Chile, 1972, p. 40.
183. Ibid., pp. 57ff.
184. Ibid., p. 20.
185. Cf. R. Alexander, Organised Labour in Latin America, 1965, p. 19.
186. Cf. H. Spalding, Organised Labour in Latin America (New York: Harper,
1977) pp. 181ff; cf. also K. P. Erickson, 'Corporation and labour in development',
in H. J. Rosenbaum and W. G. Tyler (eds), Contemporary Brazil: Issues in
Economic and Political Development (New York: 1972).
187. Cf. R. Alexander, Organised Labour in Latin America, 1965, p. 78.
188. R. Alexander, Communism in Latin America (New Brunswick, New
Jersey: Rutgers University Press, 1957) p. 130.
189. R. Alexander, Organised Labor in Latin America, 1965, p. 84.
190. The pro-Moscow Partido Socialista lnternacionalista was founded in 1918
after a split in the Socialist Party.
191. For a history of the Socialist Party cf. R. J. Walter, Socialist Party, 1977.
192. H. Spalding, Organised Labour in Latin America, 1977, pp. 158ff; cf. also
Victor Alba, Politics and the Labour Movement in Latin America (Stanford:
Stanford University Press, 1%8) pp. 217-40; and R. J. Alexander, Labor Relations
in Argentina, Brazil and Chile (New York: McGraw Hill, 1962); for a general
account of trade unionism during the first Per6nist period cf. S. L. Bailey, Labor,
Nationalism and Politics in Argentina (New Brunswick, New Jersey: 1%7).
193. Since statements about dependence and autonomy make sense only from a
macro-historical, comparative point of view, it is worth referring briefly to the
contrasting development of the working-class movement in the capitalist centre.
In fact, in several western European societies their working-class movements,
appearing at a time when advanced industrialisation had resulted in a massive
industrial labour force, could much better resist state attempts at manipulation and
repression. Moreover, they contributed in various ways to a certain 'rationalisation'
of politics. To take England as an example, the fact that the Labour Party's
strength was from the start based on the large number of its followers rather than
on their wealth or social prestige resulted in its highly centralised and bureaucratised structure - one that was indeed suitable for the organisation of large numbers
of people in a 'horizontal', class manner. That such a mode of integration
discourages particularistidpersonalistic elements in the political process is a point
well established in the relevant literature. (Cf. R. Alford, Party and Society
(London: John Murray, 1964) pp. 33-4.) Moreover, the fact that the Labour Party
did not develop out of the pre-existing parliamentary groups but was created by
extra-parliamentary forces (that is, trade unions) meant that it had a better chance
for distancing itself from the corrupt practices and the particularistic character of
parties created 'internally'. (Cf. M. Duverger, Les Parties Politiques (Paris:
Armand Colin, 1951) pp. 21lff.)
The Labour Party's mode of creation, as well as the development of massive
trade unions several decades earlier, not only enhanced the autonomy of the
working-class movement vis-a-vis the state, these facts also contributed to the
weakening of the parochial, localistic and particularistic elements of the other
parties: the latter were obliged to adopt similar mass methods of organisation if
they were to survive and compete effectively in the broadening political arena.
Contrary to their French counterparts which were constantly flirting and experimenting with authoritarian/dictatorial solutions, the British conservatives
accepted the challenge presented by the labour movement and began to adopt
more bureaucratiduniversalistic principles of organisation - that is, they started
quite early to transform themselves into a modern mass party. Moreover, not only
240
did they change their mode of political operation, they were also forced to change
their goals and become more receptive to ideas of popular welfare, collective
bargaining, and so on. (Cf. 0. Kircheimer, 'The transformation of European
political parties', in J. Lapalombara and M. Weiner (eds), Political Parties and
Political Development (Princeton: Princeton University Press, 1966.) The emergence of the working-class movement was not, of course, the only factor responsible
for the demise of particularistic/clientelistic politics in England. The introduction of
the secret ballot, large constituencies and stiffer penalties against corrupt political
practices such as vote buying, as well as broader developments such as educational
reform, the spread of literacy and so on- all these helped to bring about a change in
public attitudes and the emergence of a new and more democratically oriented
public opinion.
194. Concerning the role of the western industrial bourgeoisie it should be
remembered that, whereas merchant capital, even in its developed forms, could
easily be accommodated within feudalism or within the post-feudal institutions of
absolutist Europe, this was by no means the case with the expanding industrial
capital. (Cf. M. Dobb, Studies in the Development of Capitalism, 1968, ch. 1). By
its very structure and logic of development, the entrance of capital into the sphere
of production (threatening to some extent the dominant economic and political
interests of the old order) had far-reaching consequences. So, for instance, in
pre-industrial Europe there was no clear differentiation at the top of the economic
hierarchy between banking, merchant, and industrial functions. Industrial activities
were usually undertaken by wealthy merchants diversifying into manufacturing as a
secondary concern. It was only with the Industrial Revolution that big industrialists
emerged as a relatively specialised occupational category. (Cf. F. Braude!,
Civilisation Materielle, Economie et Capitalisme (Paris: Armand Colin, 1979) vol.
2, pp. 331ff.) Aside from a certain antagonism between industrialists and landed
interests, on the level of state institutions those middle-class groups that had
emerged from the Industrial Revolution but remained excluded from the oligarchic
political game (professionals, newly-rich merchants, industrialists) were pushing
for reforms which contributed to a limited broadening of political participation as
well as to the institutionalisation of a more liberal state administration.
For instance during the early phase of the Industrial Revolution textile manufacturers, being less dependent on the state and on state-controlled banking institutions for financing their technologically simple enterprises, became the champions
of liberal ideas, not only in England but also in France, Belgium, Switzerland, and
so on. (Cf. W.L. Langel, Political and Social Upheaval 1832-1852 (New York:
Harper and Row, 1969).) If in the latter cases they sometimes pressed for the
protection of 'infant industries' from foreign competition, as far as internal policies
are concerned they were fighting against the restrictive, particularistic practices of
the ancien regime state.
These manufacturers did not want the dynastic authoritarian state of the past,
and they did not need the technocratic authoritarian state of the future. They
did not want the internal tariffs, the consumer taxes and the tedious regulations
of the absolutist monarchies, which prevented the manufacturers from selling
their goods in a nationwide market. And they did not want the local guild
monopolies and local welfare systems, which also prevented them from drawing
their labour from a nationwide market.
(Quoted from J. R. Kurth, 'Industrial change and political change: A European
perspective', in D. Collier (ed.), The New Authoritarianism in Latin America
(Princeton: Princeton University Press, 1979).
241
It is true of course that the industrial bourgeoisie was not the sole force behind
administrative reforms. Not only did the process of state rationalisation and the
transition from patrimonial to bureaucratic principles of organisation start long
before the Industrial Revolution, but - even for the nineteenth-century reforms various other social strata and interest groups must be taken into account in
addition to industrial interests. It is difficult in fact to establish direct linkages
between industrial capitalist interests and specific reforms.
Despite these reservations, however, there is no doubt that the rising middle
classes, and among them particularly the industrial bourgeoisie, managed through a
variety of indirect mechanisms (for example, influencing or shaping public opinion,
bringing about new economic conditions which required new forms of state
intervention, and so on) to create a context conducive to the initiation and
implementation of universalistic principles of public administration. In other
words, far more important than whether or not the industrial capitalists acted as a
highly cohesive collective agency consciously promoting collective goals, is the fact
of their very existence (in a position of relative autonomy vis-a-vis the state) and
their indirect impact on the polity.
195. It should also be stressed that in many cases the dividing line between
landowners, financiers and industrialists was very thin or non-existent. The fusion
of landed, industrial, and financial interests was and still is particularly marked in
Chile. Cf. M. Zeitlin and R. A. Ratcliff, 'Research methods for the analysis of the
internal structure of the dominant classes: The case of landlords and capitalists in
Chile', Latin American Research Review, no. 3, 1975.
196. For Latin America cf. J. R. Kurth, 'Industrial change and political change',
1979; for the Balkans cf. Hugh Seton-Watson, The Eastern European Revolution
(London: 1950).
197. Cf. Claudio Veliz, The Centralist Tradition in Latin America (Princeton:
Princeton University Press, 1980) p. 263.
198. In western European industrialisation - given the much simpler industrial
technology at that time and therefore the lower requirements of know-how and
skills - Marx's "way I" transition to capitalism (from small artisanal to capitalist
production) was much more feasible than it is for those societies that launched their
industrialisation a century later. For them the greater technological complexity and
cheaper modern transport costs mean that it makes more economic sense to import
technology than to produce it indigenously. For an extensive elaboration of this
crucial point cf. P. Bairoch, Le Tiers-Monde dans /'Impasse (Paris: Gallimard,
1971).
199. Needless to say, the greater dependence of industrial capital on the state
does not mean that the state in the semi-periphery is omnipotent, that it can for
instance take systematic action against the interests of capital. In its role as the
general co-ordinator of the whole capitalist economy the semi-peripheral state,
especially in the post-1929 era, has to provide a favourable institutional framework
for the enlarged reproduction of capital; that is, it has to use its resources in ways
that safeguard and promote bourgeois interests, even if such interests were
originally created by the state and are poorly organised. This statement has of course
to be qualified (cf. below Chapter 4, section 4).
Relative state autonomy vis-a-vis organised class interests can therefore coincide
with the state having to operate within severe structural constraints (for example,
when it has to adopt policies which do not frighten off capitalist investors). Given
that such constraints are particularly severe in the semi-periphery, capitalist
interests can be safeguarded and promoted even without being strongly organised
on the political level. Norma Hamilton's useful distinction between instrumental
and structural autonomy of the state helps one to understand the above problem
better. Cf. her 'State autonomy and dependent capitalism in Latin America',
242
British Journal of Sociology, September 1981. For a discussion along similar lines
concerning the problem of the relative autonomy of the Greek state cf. N.
Mouzelis, 'Capitalism and the development of the Greek state', 1980, section IV.
200. For the persistence of the particularistic, personalistic character of Latin
American public bureaucracies cf. Roberto de Oliveira Campos, 'Public administration in Latin America', in Nimrod Raphaeli (ed.), Readings in Comparative
Public Administration (Boston: 1967); Robert E. Scott, 'The government bureaucrat and political change in Latin America', Journal of International Affairs, vol. 20,
pp. 289-303. Cf. also J. W. Hopkins, 'Contemporary research on public administration and bureaucracies in Latin America', Latin American Research Review,
Spring 1974. For the persistence of the clientelistic, particularistic features of
politics even in a country like Chile- where working-class organisations are strong
and relatively autonomous - see chapter 3. For the all-pervasive character of
clientelism in the Greek post-oligarchic system cf. K. Legg, Politics in Modern
Greece, 1969.
Chapter2
1. To explain fully the type of state-civil society relationship that prevailed in the
parliamentary semi-periphery, one must of course take into account many more
factors than are mentioned here. (For instance the violent nature of the wars of
independence and the severe dislocations they produced - cf. for example Tulio
Halperin Donghi, The Aftermath of Revolution in Spanish America (New York:
1973) pp. 1-82.) However, it seems to me that the most fundamental factors still
are: the pre-independence legacy, the late timing of industrialisation in relation to
the early expansion of the state, and the restricted and unequal character of the
industrialisation process itself.
2. The concept of the 'overdeveloped' state, or the 'relative autonomy' of the
state in the capitalist semi-periphery, has received extensive discussion in the
literature on the 'post-colonial' state. (Cf. for instance H. Alavi, 'The state in
post-colonial societies: Pakistan and Bangladesh', New Left Review, July-August
1972; Colin Leys, 'The "overdeveloped" post-colonial state: A re-evaluation',
Review of African Political Economy, January-April1976; John Saul, 'The state in
post-colonial societies', Socialist Register, 1974; W. Ziemann and M. Lanzendorfer,
'The state in peripheral societies', Socialist Register, 1977; J. Saul, The State and
Revolution in Eastern Africa (London: Heinemann, 1979); H. Goulbourne (ed.),
Politics and the State in the Third World (London: Macmillan, 1979); R. Muck,
'State and capital in dependent social formations: The Brazilian case', Capital and
Class, no. 8, Summer 1979.)
In view of the weak organisation of class interests, and the facility with which
formal organisations can be manipulated from above, it is justifiable to see state
institutions as relatively free of the pressures emanating from organised interest
groups. As noted earlier however, this relative autonomy of the state does not
imply omnipotence.
3. P. Schmitter, 'Still the century of corporatism', The Review of Politics, vol.
36, no. 1, January 1974.
4. Cf. J. Malloy (ed.), Authoritarianism and Populism in Latin America
(University of Pittsburg Press, 1977) p. 4.
5. Cf. for instance J.D. Powell, 'Peasant society and clientelistic politics',
American Political Science Review, vol. 64, 1970. For more specific accounts of the
243
long-term transformation of clientelistic politics in the Mediterranean area, cf.

articles by M. Attalides, S. Khalaf and K. Brown, in E. Gellner and J. Waterbury
(eds), Patrons and Clients in Mediterranean Societies (London: Duckworth, 1977).
6. Cf. A. Weingrod, 'Patrons, patronage and parties', Comparative Studies in
Society and History, vol. 10, 1%8, pp. 377-400.
7. For a case illustrating this point cf. N. Mouzelis, Modern Greece: Facets of
Underdevelopment (London: Macmillan Press, 1978) pp. 14-17, and 79-104.
8. For the link between populism and the relatively abrupt entrance of the
masses into politics, cf. Di Tella, 'Populism and reform in Latin America', in
Claudio Veliz, Obstacles to Change in Latin America (London: Oxford University
Press, 1965).
9. Considering populism as a mode of broadening political participation during
or soon after the demise of oligarchic parliamentarism does not imply that populist
movements occur only during such a transition, or that they always lead to a
broadening of political participation.
10. At the generalised level of this study, populism and clientelism will, like the
integrative mode of inclusion, be presented as highly abstract analytical constructs
never found in their pure form in any actual social formation. Any specific political
system or organisation will always contain elements of all three modes, albeit
articulated in such a way that one of them is dominant. When it is argued
therefore that in the West the horizontal/non-populist mode of political inclusion of
the lower classes has become dominant in irreversible fashion, this does not mean
that clientelistic or populistic elements have completely disappeared from the West
European political parties; it simply means that they have become peripheral.
11. On the antagonistic character of pepulist ideologies cf. E. Laclau, Politics
and Ideology in Marxist Theory (London: New Left Books, 1977).
12. In its extreme form the Megali Idea (the Great Idea) ideology advocated the
recapture of Constantinople and some sort of resuscitation of the Byzantine
empire.
13. Cf. J. D. Bell, Peasants in Power: Alexander Stamboliiski and the Bulgarian
National Union 1899-1928 (Princeton: Princeton University Press, 1977). For a
more general analysis of Balkan peasant ideologies cf. Branko Peselz, Peasant
Movements in South-Eastern Europe, unpubl. Ph.D. thesis, Georgetown University, 1950; and G. Dimitrov, 'Agrarianism', in F. Gross, European Ideologies (New
York: Philosophical Library, 1948).
14. Cf. G. Germani, Authoritarianism, Fascism and National Populism (New
Brunswick: Transaction Books, 1978) pp. 174ff.
15. For numerous examples of this type of clientelistic control over voters in
post-war Greece cf. Maria Comninos, The Development of the Patronage System in
Aitolo-Akharnania and Kava/a, Ph.D. thesis, London School of Economics, 1984.
16. Cf. G. Mavrogordatos, Stillborn Republic: Social Coalitions and Party
Strategies in Greece 1925-1936 (Berkeley, Los Angeles and London: University of
California Press, 1983) p. 82.
17. Ibid., p. 87.
18. K. Legg, Politics in Modern Greece (Stanford, California: Stanford University Press, 1969) pp. 134-5.
19. For instance, deputies from large urban areas where clientelistic networks
were very weak, or new deputies who did not yet have the time to consolidate their
positions in their constituencies. (Cf. K. Legg, ibid., p. 137.)
20. M. L. Connif, Urban Politics in Brazil: The Rise of Populism 1922-1945
(University of Pittsburg Press, 1981).
21. Ibid., pp. 68--9.
22. Ibid., pp. 71ff.
244

24. Ibid., p. 101.
25. Cf. Max Weber, Economy and Society, ed. G. Roth and C. Wittich (Los
Angeles: University of California Press, 1978) vol. 2, chs 12 and 13.
26. On this point cf. A. Stewart, 'Populism: The social roots', in E. Gellner and
G. Ionescu (eds), Populism (London: 1969).
27. Cf. A. E. Van Niekerk, Populism and Political Development in Latin
America (Rotterdam University Press, 1974); also A. Hennessy, 'Latin America',
in Gellner and Ionescu (eds), Populism, 1969.
28. On the changing definitions of the concept of populism cf. J. B. Allcock,
'Populism: A brief biography', Sociology, September 1971; cf. also M. Canovan,
Populism (London: Junction Books, 1981) pp. 7ff.
29. E. Laclau, Politics and Ideology in Marxist Theory, 1977, ch. 3.
30. M. Canovan, Populism, 1981, p. 294.
31. For a development of this point cf. my critique of Laclau's theory of
populism, 'Ideology and class politics: A critique of Erneso Laclau', New Left
Review, November-December 1978.
32. Cf. P. Drake, Socialism and Populism in Chile 1932-52 (Chicago/London:
University of Illinois Press, 1958).
33. S. M. Lipset, Agrarian Socialism (New York: 1950).
34. For typical definitions of clientelism cf. for instance John Duncan Powell,
Peasant Society, 1970; and James G. Scott, 'Patron-client politics and political
change in south-east Asia', American Political Science Review, March 1972.
35. In the light of what I have already said about the structural similarities
between clientelism/populism on the one hand and feudalism/patrimonialism on
the other, my point about the necessity of seeing the patron-client relationship
within the broader system of domination is similar to the Weberian argument that
feudalism should not be conceptualised merely in terms of the serf-lord exploitative
relationship; that it implies also a certain type of autonomy of the lord vis-a-vis
those higher up in the feudal hierarchy, an autonomy which is non-existent in the
'pure' patrimonial case.
Chapter3
1. On the Latin American military cf. J. J. Johnson, The Military and Society in
Latin America (Stanford, California: Stanford University Press, 1964); E.
Lieuwen, Arms and Politics in Latin America (New York: Praeger, 1961); A. F.
Lowenthal, Armies and Politics in Latin America, New York and London: Holmes
and Meier, 1976). On the Greek military cf. T. Veremis, The Greek Army in
Politics, Ph.D. thesis, Trinity College, Oxford, 1974.
2. Cf. J. Nun, The middle-class military coup', in C. Veliz (ed.), The Politics of
Conformity in Latin America (London: Oxford University Press, 1967).
3. Cf. on these points Chapter 1.2.
4. To give an obvious example, all civilian politicians who played a decisive role
during the demise of oligarchic parliamentarism (Vargas in Brazil, Yrigoyen in
Argentina, Alessandri in Chile, Venizelos in Greece) have, every one of them, at
certain points in their career, tried to draw the military into politics and to use them
for their own political purposes.
5. Cf. A. Stepan, The Military in Politics: Changing Patterns in Brazil (Princeton: Princeton University Press, 1971) pp. 64ff.
245
6. Cf. Alain Rouquie, Pouvoir Militaire et Societe Politique au Republique

Argentine (Paris: Presses de Ia Fondation Nationale des Sciences Politiques, 1978)
p. 122.
7. Cf. A. Rouquie, ibid., 1978, pp. 182ff; on the divisions within the military
who fabricated the coup cf. Jose Maria Sarobe, Memorias Sobre Ia Revolucion del6
Septiembre de 1930 (Buenos Aires: Ediciones Gure, 1957).
8. Cf. M. Goldwert, Democracy, Militarism and Nationalism in Argentina
1930-1966: An interpretation (Austin and London: University of Texas Press,
1972), pp. 68ff. An important point concerning the oligarchic restoration in
Argentina is the diffusion of anti-liberal, anti-democratic and anti-communist ideas
which had a profound impact on certain groups of officers. Cf. on this L. Allub, 'El
colapso de Ia democracia liberal y origines del fascismo colonial en Argentina',
Revista Mexicana de Sociologia, July-September, 1980.
9. Cf. M. Goldwert, Democracy, Militarism and Nationalism, 1972, pp. 35ff.
10. It should however be stressed that it was not only the conservatives who
appealed to the military for support:
The Radical Party, however, confident of its popularity, wanted the Army to
serve as the instrument for guaranteeing honest elections. When this failed in
1937, and after the brief flickering of hope for electoral honesty under President
Ortiz turned into despair with his withdrawal from office, some radical leaders
again turned to the military. The cumulative effect of these mutually hostile
efforts, Radical and Nationalist, to involve the Army in political action was
probably to convince the officers that they alone could save the nation.
From R. Potash, The Army and Politics in Argentina 1928-1945 (Stanford,
California: Stanford University Press, 1968) p. 284.
11. It is not, therefore, surprising to read (A. Rouquie, Pouvoir Militaire, 1978,
p. 4, and cf. pp. 77ff) that
in Argentina since 1930, the hegemony of military power through the most
varied of governmental forms seems to have become the norm ... since 1930, no
freely elected, constitutional president has fulfilled his six-year mandate without
pressure or veto from the army ... Out of the sixteen presidents that Argentina
has had since 1930, eleven were military men.
13. Cf. F. M. Nunn, The Military in Chilean History: Essays on Civil-Military
Relations 1810-1973 (Albuquerque: University of New Mexico Press, 1976) pp. 3ff.
14. Ibid., pp. 115-16.
15. On the conspiratorial activities of the Chilean military cf. Jorge Nef, 'The
politics of repression: The social pathology of the Chilean military', Latin American
Perspectives, vol. 1(2), pp. 58-77.
16. F. M. Nunn, The Military in Chilean History, 1976, p. 150.
17. The numerous military plots and interventions dunng this period seriously
discredited the military. There were calls even from their own ranks for an
immediate return to the barracks and Ibanez himself in his farewell message made a
strong plea for such a return. Furthermore, in 1932 the right-wing parties created
the Milicia Republicana, a very well-armed para-military organisation, with the
declared purpose of keeping the military out of politics.
18. As F. M. Nunn put it (The Military in Chilean History, 1976, p. 236):
The strength of the party system and governmental institutions and the ability of
socioeconomic interest groups - syndicates for example - to work with and
246
within the system precluded an institutional stance by the armed forces on issues
of national import.
19. The Chilean conservative classes had also made concessions in other areas.
Of the greatest importance was the introduction of price controls for agricultural
products. Cf. on this point Thomas C. Wright, Landowners and Reform, The
Sociedad Nacional de Agricultura 1919-1940 (University of Illinois, 1982) ch. 3. For
a comparison of civil-military relations in Argentina and Chile, cf. Lisa North,
Civil-Military Relations in Argentina, Chile and Peru (Berkeley: Institute of
International Relations, University of California, 1967).
20. Cf. on this point Carlos Nunez, Chile: ;.Ia Ultima Opci6n Electoral?
(Santiago: Politica Latinoamericana Nueva, 1970) p. 60.
21. J. Nef, 'The politics of repression', p. 59.
22. F. M. Nunn, The Military in Chilean History, 1976, pp. 235ff.
23. Cf. J. Nef, 'The politics of repression', also H. E. Bichens, 'Antiparliamentary themes in Chilean history, in K. Medhurst (ed.), Allende's Chile
(London: Hart Davis MacGibbon, 1972).
25. Although general conscription was introduced in 1880, it was only during the
Balkan Wars that numbers increased spectacularly. For statistics on tb.is point cf.
D. Dakin, The Unification of Greece 1770-1924 (London: Ernest Benn, 1974) p.
316.
26. In 1917 tuition fees for the Evelpidon School (the top Greek military
academy) were abolished. Before then only wealthy students could afford this type
of military education, and in the ninteteenth century familial property or a good
marriage were necessary for maintaining an officer's lifestyle: 'the old respectable
families stopped sending their children to the Military Academy once the institution became less exclusive and lost its social prestige.' From T. Veremis, The Greek
Army in Politics, 1974, p. 78. Cf. also Veremis, 'The officer corps in Greece
1912-1936', Byzantine and Modern Greek Studies, vol. 2, 1976, pp. 113-14.
27. Cf. on this T. Veremis, 'Some observations on the Greek military in the
inter-war period 1918--1935', Armed Forces and Society, May 1978, p. 357.
28. For instance, during the first six months of 1936 there were as many as 344
strikes in various places. This strike wave culminated in the events of 9-10 May in
Salonica when 12 demonstrators were killed.
29. For a historical analysis of the Metaxas coup cf. K. Koliopoulos, Britain and
Greece 1935-1941, Ph.D. thesis, London School of Economics, 1972.
30. Cf. on this point Chapter 1.2.1.
31. The first national census of 1896listed 333,725 citizens eligible to vote, but it
is doubtful that national elections ever attracted more than 10 per cent of that
number to the polls. Oligarchical control of the electoral system discouraged
higher voter turnout.
From R. J. Walter, The Socialist Party of Argentina 1890-1930 (Austin: Institute
of Latin American Studies, University of Texas, 1977) p. 7.
32. Agriculturists in Argentina, colonists and tenant farmers alike, did not as a
rule live in villages and go out each day to cultivate the surrounding fields.
Because of extensive agriculture where a great deal of land was superficially
tilled, farm homes were spread out at considerable distances from each other.
When eighty acres was the basic unit of wheat cultivation, it nevertheless
required much walking or riding to reach a neighbor; with five hundred acres,
the distance was nearly tripled, and the possibility of social institutions reaching
247
the farm decreased in proportion. Weddings, funerals, and special church

holidays might warrant the long trip to town or city. But the priest, clergyman,
or rabbi could not minister to widely scattered families that lacked the resources
or interest to form a congregation.
From James Scobie, Revolution on the Pampas; A Social History of Argentine
Wheat 1860-1910 (Austin: University of Texas Press, 1964) p. 63.
33. The tenant farmer was relentlessly pushed from plot to plot because the
estanciero wanted the land for raising cattle or sheep. Pastoral interests
accepted agriculture as the initial step in planting alfalfa pastures but leased
land to farmers only long enough to break up the sod and prepare the land for
pasturage. The landowner also preferred the short lease that permitted periodic
increases in rents. In surroundings that favored the large unit, the small
farm-owner found himself at the mercy of equally restrictive forces - petty
officials, discriminatory railroad rates, and powerful market interests. The land
remained something to be exploited in order to secure the maximum return in
the shortest time without regard for the consequences. As a result the
immigrant rarely sank roots in the land, and the city rather than the farm
increasingly became the purveyor of rapid and easy advancement for the newly
arrived.
From James Scobie, ibid., p. 31.

34. Cf. Peter G. Snow, Political Forces in Argentina (London and New York:
Praeger, 1979) p. 42.
35. A. Rouquie, Pouvoir Militaire, 1978, p. 698.
36. D. C. Lambert and J. M. Martin, L'Amerique Latine: Economie et Societe
(Paris: Armand Colin, 1971) Table 21.
37. Cf. on this point Arnold Bauer, Chilean Rural Society: From the Spanish
Conquest to 1930 (Cambridge and London: Cambridge University Press, 1975) pp.
159ff.
38. Ibid., p. 159.
39. Ibid., p. 151.
40. Ibid., p. 233.
41. Cf. Cesar Caviedes, The Politics of Chile: A Socia- Geographical Assessment
(Boulder, Colorado: Westview Press, 1979) p. 138.
42. Cristobal Kay, 'Transformaciones de los relaciones de dominacion y dependencia entre terratenientes y campesinos en Chile', Revista Mexicana de Sociologia,
April-June 1980.
43. C. Caviedes, The Politics of Chile, 1979, p. 53.
44. See section 3.4.3.
45. Cf. Paul W. Drake, 'The political responses of the Chilean upper class to the
Great Depression and the threat of socialism', in F. C. Jaber (ed.), The Rich, the
Wellborn and the Powerful (Secaucus, N.J.: Citadel Press, 1975); cf. also A.
Valenzuela, 'The origins of democracy: Reflections on the Chilean case', paper
presented at a colloquium at the Woodrow Wilson Centre for Scholars, Washington D.C., July 1982.
46. Cf. on this point Chapter 1.1.2.
47. Due to the peculiarities of the electoral law at that time, the conservatives
gained 260 out of a total of 370 seats in parliament, despite the fact that the
Venizelists had obtained 52 per cent of the total vote.
48. Cf. on this point George Mavrogordatos, The Stillborn Republic: Social
Coalitions and Party Strategies in Greece 1922-1936 (Berkeley, Los Angeles and
London: University of California Press, 1983).
248
49. For industrial growth rates of various Latin American economies in the
1960s cf. D. C. Lambert and J. M. Martin, L'Amerique Latine, 1971, p. 315.
50. According to C. F. Diaz-Alejandro, Peron's inward-looking policies would
have been more appropriate in the 1930s than in the 1940s. In the later decade with
its more favourable conditions for exports, export-oriented policies and a more
encouraging attitude towards foreign capital might have meant not only a higher
over-all growth rate but also a more rapid and successful development of
Argentina's manufacturing sector. Cf. his Essays on the Economic History of the
Argentine Republic (New Haven and London: Yale University Press, 1970) pp.
106ff.
51. Cf. National Accounts of Greece I948-70, pp. 120-21. For a detailed analysis
of the growth ofthe post-war Greek economy, cf. N. Vernicos, L'Economie de Ia
Grece I95{}-I970, Ph.D. thesis, University of Paris VIII, 1974, vol. 1.
52. For instance as far as Greece is concerned, the state through its powerful
Monetary Commission regulates in great detail the credit policy of all banks, so
setting strict limits to the manner and extent of their operations.
53. Chile was a precursor of that type of state control. As early as 1939 the
Popular Front government decided to promote the growth of the economy more
systematically by creating CORFO (Chilean National Development Corporation),
and in the post-war period many Latin American countries followed suit. It should
also be mentioned that in 1928 Chile had an Instituto de Credito Industrial which,
together with other organisations, channelled public funds into productive activities.
54. For the extraordinary importance of the state as a producer in late-late
industrialising economies cf. C. Vai:tsos, 'Order principally denotes power: Elements and effects of relative bargaining power between state enterprises an~
transnationals', Report to the United Nations, Centre of Transnational Corporations, Athens 1981, p. 3ff.
55. For a recent analysis comparing Greek and western European agricultural
productivity cf. D. Delis, 'Labour productivity in Greek and western European
agriculture', Agrotiki no. 5, 1978 (in Greek); cf. also K. Thompson, Land
Fragmentation in Greece (Athens: Centre of Planning and Economic Research,
1963); OECD, Agricultural Statistics I953-I969 (Paris: 1969) pp. 52-3; and OECD,
Agricultural Development in Southern Europe (Paris: 1969) pp. 22ff.
56. C. F. Diaz-Alejandro, Essays on the Economic History of the Argentine
Republic (New Haven and London: Yale University Press, 1970) pp. 158ff. For an
analysis of the over-all decline of Argentinian agriculture after 1930 cf. ibid., pp.
142-207; cf. also A. Ferrer, The Argentine Economy, (Berkeley and Los Angeles:
University of California Press, 1967) pp. 170ff; and Gilbert Merkx, 'Sectoral
clashes and political change: The Argentine experience', Latin American Research
Review, vol. 4, no. 3, 1969.
57. For an over-all view of the development of Chilean agriculture and its
marked decline after 1930 cf. M. J. Mamalakis, The Growth and Structure of the
Chilean Economy: From Independence to Allende (New Haven and London: Yale
University Press, 1976); cf. also M. Mamalakis and C. W. Reynolds, Essays on the
Chilean Economy (Homewood, Ill.: Irwin, 1965), particularly pp. 122ff.
58. Cf. D. C. Lambert and J. M. Martin, L'Amerique Latine, 1971, pp. 49ff.
59. Cf. on this point B. Stallings, Class Conflict and Economic Development in
Chile I958-73 (Stanford, California: Stanford University Press, 1978) pp. 86ff.
60. An indication of this reluctance to invest in industry is the fact that the two
biggest banking establishments in Greece often failed to dispose of the 15 per cent
of the funds they were obliged to advance for the development of the industrial
249
sector. Cf. D. Psilos, Capital Market in Greece (Athens: Centre of Economic

Research, 1964) ch. 4.
61. Cf. A. Hirschman, A Bias for Hope (New Haven and London: Yale
62. For comparable statistics on Chile and Argentina cf. B. Kadar, Problems of
Economic Growth in Latin America (New York: StMartin's Press, 1980) p. 91; for
Greece cf. G. Giannaros, 'Foreign capital in the Greek economy', in E. Eliou et al.,
Multinational Monopolies (in Greek) (Athens: 1973).
63. Cf. C. F. Diaz-Alejandro, Economic History of the Argentine Republic,
1970, p. 266.
64. Cf. B. Stallings, Class Conflict and Economic Development, 1978, p. 162.
65. Taken from G. Giannaros, 'Foreign capital...', 1973, p. 404.
66. For instance, by the end of 1973, foreign capital invested in Greece had risen
to a total of approx. $725 million, an amount not so very impressive if one considers
that in a single year (1969) $2,504 million went to the gross formation of fixed
capital in the Greek economy. Cf. N. Vernicos, L'tconomie de Ia Grece, 1974, vol.
1, p. 372.
67. Cf. C. F. Diaz-Alejandro, Economic History of the Argentine Republic,
1970, pp. 269ff; and M. J. Mamalakis and C. W. Reynolds, Essays on the Chilean
Economy, 1965.
68. Cf. B. Nefeloudis, Demythisation with Numbers (in Greek) (Athens: 1973)
p. 146.
69. Cf. World Bank, World Development Report (Oxford University Press,
1980) Table 115.
70. While at the beginning of the 1960s approximately half of Greece's labour
force was still in agriculture, by the end of the decade this figure had dropped to
about 40 per cent (cf. International Labour Office Year Book 1976, p. 136), and by
the beginning of the 1980s to 25 per cent.
71. This extremely lopsided type of urbanisation is glaringly obvious in Greece
for example, where at present approximately one-third of the country's entire
population lives in the great and sprawling Athens-Piraeus conurbation. (Cf.
Statistical Yearbook of Greece (Athens: National Statistics Service, 1978) p. 22). In
Argentina as early as 1960 as much as 33.7 per cent of the total population was
concentrated in the capital; in Chile at the same time the percentage was 25.9. (Cf.
D. C. Lambert and J. M. Martin, L'Amerique Latine, 1971, pp. 139-40.)
72. An ECLA report (The Process of Industrial Development in Latin America,
New York: United Nations, 1966) makes a very clear distinction between
industrialisation in depth and in breadth:
In development in depth, most of the annual reinvestment by entrepreneurs is
made in their own enterprises, in the form of progressive modernization and
consequent reduction in costs. The proportion of products manufactured in the
country does not increase rapidly from year to year, but the efficiency of the
existing activities does.
fn development in breadth, the most usual type in Latin American industry,
profits are usually reinvested in new activities, the production of new items
which can replace those formerly imported, while the existing activities remain
at a standstill as regards average productivity. Development in breadth appears
to offer more advantages to the private entrepreneur, because in new lines of
production, at least for a few years, competition is very limited and the
producer may have a virtual monopoly, whereas reinvestment in the same field
continually increases competition, and also because it is easier to establish a
250
wholly new production unit than to keep modernizing and improving existing
plants, where routine is strongly entrenched. However, widespread development in breadth tends to increase and perpetuate situations of monopoly or
restricted competition, and the stagnation of traditional industries. This is
apparently one reason why such industries in Latin America are now faced with
an urgent need to renew the obsolete equipment they have accumulated, and
why their levels of organizational and operational efficiency are so low.
73. For the concept of disarticulation cf. S. Amin, Le Developpement lnegal
(Paris: Minuit, 1973).
74. Given the book's attempt to make as explicit as possible the centresemiperiphery comparative perspective, it is pertinent to show more precisely in
what sense capitalist development in western Europe was less 'disarticulated' and
unbalanced than in the ACG countries.
At the risk of overgeneralisation, one can argue that the development and
eventual dominance of industrial capitalism in the western European economies in
the late eighteenth and nineteenth centuries was characterised by the large-scale
destruction of non-capitalist modes of production as capitalist relations of production expanded very widely in both agriculture and industry; and by the establishment of strong complementarity between different sectors of the economy.
It was in the industrial sector, however, where capitalist expansion was the most
spectacular and, from our point of view, the most relevant. Towards the end of the
nineteenth century, industry not only in England but also in the north-western part
of continental Europe came to supersede agriculture as the major sector of the
economy (in terms of its contribution to the national product); concomitantly,
within industry as a whole it was the sector of heavy industry which acquired a
dominant position. In the area of light industry, new inventions (such as the electric
and gasoline motors) contributed to the rapid mechanisation of many branches
(clothing, food processing, shoes, and so on), and to the partial displacement of
non-capitalist, craft forms of work organisation by big capitalist enterprises.
Since mechanisation ipso facto does not destroy artisanal production, some of the
new mechanical inventions could be adopted by small, simple-commodity producers, and in fact the creation of new industries frequently stimulated further artisanal
production. Nevertheless, there is no doubt that the spectacular development of big
industry, rapid mechanisation, and the spread of modern forms of organisation did
displace artisanal forms and made capitalist relations of production dominant in
most branches of industry. From 1870 onwards the artisan class as a whole started
to decline even in Germany and France, and by the end of the century half of the
labour forces in Germany, Britain and Belgium had become employees in firms of
more than twenty workers. (Even in France, despite large-scale persistence of
simple-commodity production in both industry and agriculture, industrial capitalism took a fairly extended form compared to the semi-peripheral economies. (Cf.
P. N. Stearns, European Society in Upheaval, (London: Macmillan Press, 1966) pp.
199ff.
Concerning now the complementarity between sectors, the general point that
needs to be made is that western capitalist development, being a relatively
indigenous and gradual process in social formations which were politically and
economically strong in the world system, managed to articulate with the persisting
non-capitalist sectors in such a way that, obvious disruptions and huge inequalities
notwithstanding, the benefits of technological improvements and the high productivity initiated in the capitalist sectors could rapidly spread to the rest of the
economy. (According to G. Therborn, the extensive development of the forces of
production and the spectacular growth of productivity in the West meant that
251
capital managed to both increase the relative exploitation of labour and improve
the workers' standard of living. Cf. 'The role of capital and the rise of democracy',
New Left Review, no. 103.)
Admittedly, whether to characterise eighteenth and nineteenth-century Western
European capitalist development as an instance of balanced or unbalanced growth
is a highly controversial issue among economists and economic historians. (For a
review of the debate cf. B. Higgins, Economic Development (London: Constable,
1968) pp. 327-42.) But in terms of a broad, macro-level comparison with the
formations of the industrialised semi-periphery, it is quite clear that in the West the
entrance of capital into the sphere of industrial production was much less abrupt.
Even when one considers such European powers as France and Germany which
tried to 'catch up' with England, the process of their capitalist industrialisation can,
in terms of abruptness and imbalances, by no means be compared with that of
societies which industrialised a century or more later. For these latter (as
mentioned already) it was much easier, given the level of technological complexity
and the relatively low costs of international transport, to transfer western technology in toto, rather than to start building a technological base of their own by
mobilising indigenous resources and skills. In the case of European industrialisers
on the other hand, given the relative simplicity of techniques and prohibitive
transport costs, it was possible, by simply copying British inventions and by using
local craftsmanship, to build, with state help, a strong and relatively independent
technological infrastructure. (Cf. Paul Bairoch, Revolution lndustrielle et Sousdeveloppement (Paris: Mouton, 1974); cf. also his Diagnostic de /'Evolution
Economique du Tiers Monde 1900-1968 (Paris: Gauthier Villars, 1969).) Such a
solution was of course facilitated by the fact that western late industrialisers were
not in so dependent a position vis-a-vis England as Balkan and Latin American
late-late industrialisers were and still are vis-a-vis the West.
These obvious differences accepted, it is not difficult to see why, in broad
comparative terms, capitalist growth in the West can be described as having been
less 'disarticulated', more balanced. This is not to say that all sectors of the
economy grew at a similar pace, or in such a way as to eliminate bottlenecks and
disruptions. Since capitalist growth always implies leading sectors that operate as
spearheads disrupting and revolutionising established techniques and modes of
work organisation, the crux of the matter is rather what linkages exist between
these leading sectors and the rest of the economy. A more balanced growth, such as
was experienced in the West, implies the ability of the more dynamic sectors to
spread their dynamism and bring the backward sectors into the developmental
process so that eventually, in terms of technology and productivity, the gap
between the backward and the dynamic sectors is drastically reduced. This type of
growth leads not only to a greater homogeneity of the productive structures (and
hence to stronger inter-sectoral complementarity), but also to a wider diffusion of
the wealth generated by capitalist growth, a diffusion which enhances the development of national markets and the further expansion of production. (Cf. P. Leon,
Economies et Societes Pre-industrielles (Paris: Armand Colin, 1970) vol. 2, pp.
73ff.)
In fact one of the most striking features of the Industrial Revolution, especially in
comparison with periods of growth in ancien regime Europe or other pre-capitalist
formations, was that for the first time the beneficial effects of economic expansion
eventually spread to the lower social strata. Previous to the massive entrance of
capital into the sphere of production, economic expansion had usually meant the
growth of merchant capital and enormous enrichment at the top of the social
pyramid at the expense of the labouring classes. For instance, the extraordinary
economic upsurge experienced by Europe in the 'long' sixteenth century only
252
spelled greater wretchedness for those at the bottom of the social scale, both in
rural and urban areas.
There is no denying that the early stages of the Industrial Revolution, both in
England and on the continent, aggravated social inequalities, partially destroyed
certain artisanal strata, and created inhumane working conditions for those who
had left the countryside to sell their labour power to the capitalist factory owners.
Even if the debate about working conditions during the early period of European
industrialisation has not yet arrived at a definite conclusion, it is quite clear that the
first generations of industrial workers, poorly organised and without any real state
protection, had to pay a terrible price for the spectacular economic advances of the
initial 'take-off period.
However, it is also clear that frem the middle of the nineteenth century onwards,
that is, even before the strong development of industrial unionism, the continuing
economic growth did not result in the further impoverishment of the masses but in a
gradual if timid improvement of the general well-being. (For all the above points cf.
F. Braude!, Civilisation Materiel/e, Economie et Capitalism (Paris: Armand Colin,
1979) vol. 3, pp. 511ff.) If this elevation of the standard of living did not
immediately bring a drastic reduction of income inequalities, it is quite probable
that such inequalities reached a plateau during the last quarter of the nineteenth
century. Finally, despite the paucity of data and the lack of unanimity among
researchers, there is sufficient evidence to support the thesis that with the
development of the working-class movement and the initiation of large-scale
welfare measures at least income inequalities started on a slight downward curve
during the inter-war period. (Cf. S. Kuznets, 'Economic growth and income
inequality', American Economic Review, vol. 49, March 1955.) As I shall argue
below, the trajectory of inequalities was quite different in ACG.
75. If one looks at Argentina for instance as industrially more advanced than
either Chile or Greece, a shift in industry towards consumer durables in the 1960s
and '70s made the Argentine economy more dependent on the importation of
capital and intermediate goods. Cf. C. F. Diaz-Alejandro, Essays on the Economic
History of the Argentine Republic, 1970, p. 517.
76. Looking at this problem from a historical-comparative perspective, the ACG
situation is in sharp contrast with the economic policies of the so-called late
European industrialisers who instead concentrated their early efforts on the
production of capital rather than consumer goods. Gerschenkron's generalisation
that the more economically backward a country when it starts the industrialisation
race the greater its emphasis on capital-goods production is quite true for western
Europe's latecomers, but not true at all for Latin America's and the Balkan
late-late comers. (Cf. A. Gerschenkron, Economic Backwardness in Historical
Perspective (London: Praeger, 1962).) The difficulties of greater depth, particularly
in the sense of industry establishing 'backward' linkages with capital-goods sectors,
are not simply due to short-sighted state policies or merely entrepreneurial
preferences for quick and easy profits (although this is certainly no negligible
factor). It is rather that, given the timing of the semi-periphery's industrialisation,
the deepening process after a certain point encounters obstacles of a more
recalcitrant nature.
77. By capitalist enterprises proper I mean units employing a relatively large
number of wage labourers (say a minimum often). For Marx (Capital (New York:
International Publishers, 1967) vol. 1, p. 322),
capitalist production only then really begins when each individual capitalist
employs simultaneously a comparatively large number of labourers; when
consequently the labour process is carried on on an extensive scale and yields
253
relatively large quantities of products. A greater number of labourers working

together, at the same time, in one place (or, if you will, in the same field of
labour), in order to produce the same sort of commodity under the mastership
of one capitalist, constitutes both historically and logically the starting-point of
capitalist production.
Lenin adopts a similar definition although he specifies more precisely what he
means by 'large numbers' (The Development of Capitalism in Russia (London:
Lawrence and Wishart, 1%4) Collected Works, vol. 3, p. 351. 'When the
handicraftman turns into a real capitalist employing from I5 to 30 wage workers, the
part played by family labour in his workshop declines and becomes quite
insignificant.' (my italics). For a definition of capitalism as adopted in this book cf.
Chapter 1, note 2.
78. ECLA report, The Process of Industrial Development, 1966, pp. 36ff. It
should be noted that the West European and Latin American figures are not strictly
comparable: for Europe, industrial employment includes construction as well as
such services as electricity, gas and water; for Latin America, the figures reflect
employment in manufacturing. This difference does not, however, appreciably
change the over-all picture. Fully aware of such differences, the authors of the
ECLA report nevertheless conclude: 'For Latin America as a whole, the industrial
sector managed to absorb only 5 million out of 25 million persons added to the
labour force during the 1925-1960 period.'
79. From 1955 to 1976 approximately 1.25 million workers emigrated to western
Europe, Australia and Canada. Cf. Statistical Yearbook of Greece I977 (Athens:
National Statistics Service, 1978) p. 49.
80. Cf. D. C. Lambert and J. M. Martin, L'Amerique Latine, 1971, pp. 326ff.
Cf. Also C. F. Diaz-Alejandro, Essays on the Economic History of the Argentine
Republic, 1970, pp. 243ff for Argentina; M. Mamalakis, Growth and Structure of
the Chilean Economy, 1976; and H. Ellis et al., Industrial Capital in the Development of the Greek Economy (in Greek) (Athens: Centre of Economic Research,
1965).
81. Dualistic evolutionary theories see third-world capitalist countries in terms
of a dichotomy between a modern capitalist sector and a traditional-technologically
backward one, with very few links between them. According to such theories, the
further development of capitalism will strengthen the inter-sectoral links as the
values and technology of the modern sector are gradually diffused to the traditional
one, the whole society moving up the evolutionary ladder towards full industrialisation and development. (The classical work on dualism is J. H. Boeke's Economics
and Economic Policy of Dual Societies (New York: Willink, 1953). For a critique of
such theories cf. R. Stavenhagen, 'Seven erroneous theses about Latin America', in
L. H. Horowitz eta/., Latin American Radicalism (New York: Vintage Books,
1969) pp. 102-17.
82. D. C. Lambert and J. M. Martin, L'Amerique Latine, 1971, p. 325.
83. For instance, in 1961 the state issued 4088 permits for the establishment of
new enterprises. Of these, only 29 were of any financial importance (that is,
consisted of investments of more than $30000). Cf. H. Ellis et a/., Industrial
Capital, 1965, pp. 197-204. For some of the reasons which can explain the inability
of small enterprises to grow cf. E. Kartakis, Le Developpement Industriel de Ia
Grece (Lausanne: 1970) pp. 20ff. In a recent article A. Doxiadis has argued that in
Greece the degree of concentration and centralisation of industrial capital is not
very accentuated - and in that sense Greek industry does not portray a marked
dualistic cleavage between a big monopoly capital, and an artisanal sector. ('Is
there a monopoly capital in Greek industry?' (in Greek) Sihrona Themata,
254
July-September 1984). This argument about the degree of monopolisation of

industrial capital is quite distinct from the argument developed here about the way
in which industrial labour is distributed between large and small firms.
Another way of looking at the relatively restricted character of the capitalist
mode of production in industry from a more comparative point of view this time, is
to measure the numerical strength of industrial wage earners in relation to the total
active population in the centre and ACG. So in 1952, industrial wage earners in
Chile numbered approximately 229000 out of a total labour force of 2187000 (a
roughly 1:9 ratio). Even adding those employed in the mining industry (85000 in
1952) to the industrial work force does not change this ratio significantly. In
Argentina, a country with economic structures most nearly resembling those of the
West, there were only 900000 industrial wage earners in 1947 in a total labour force
of 7.5 million (a roughly 1:7 ratio)). The contrast with western Europe is
considerable. Belgium, for instance, in 1961 had around 1 million industrial wage
earners in a total labour force of 3.5 million (roughly I :3.5). In the United
Kingdom in 1951 there were roughly 7 million wage earners among 23 million total
(1:3.2). (Cf. International Labour Office Yearbook I964, Table 5. The ILO does
not provide this type of information for subsequent years.)
84. If, asS. Kuznets has pointed out, ('Economic growth'. 1955) higher growth
rates tend to accentuate inequalities, this is especially so when growing economies
have heterogeneous structures of production. Cf. on this point Michael Don Ward,
The Political Economy of Distribution: Equality versus Inequality (New York:
Elsevier, 1978). Cf. also ECLA, Income Distribution in Latin America (New York:
United Nations, 1971) and A. Pinto and A. di Filippo, 'Notas sobre Ia estrategia de
Ia distribucion del ingreso en America Latina', Trimestre Econ6mico, vol. 41, 1974,
pp. 357-75.
85. In Chile (1968) the top 20 per cent of the population received 56.8 per cent of
the national income, in Argentina (1961) 52 per cent. (Cf. M. S. Ahluwalia,
'Inequality, poverty and development', Journal of Development Economics, 3
December 1976.) As far as Greece is concerned (where reliable data on income
inequalities are lacking), a 1973 estimate gives to the 17 per cent in the top income
brackets 58 per cent of the national income, after deduction of taxes and social
benefits. Cf. D. Karageorgas, 'The distribution of tax burdens by income groups in
Greece', Economic Journal, June 1973.
86. By contrast, in the UK (1964) the top 5 per cent received 19 per cent of the
national income, in the USA (1969) 14.8 per cent, and in Denmark (1963) 16.9 per
cent. Cf. F. Paukert, 'Income distribution at different levels of development: A
survey of evidence', International Labour Review, August-September 1973.
87. Ibid.
88. On this point cf. C. Zuvekas, Jr., Income Distribution in Latin America: A
Survey of Recent Research (Milwaukee: University of Wisconsin, Centre for Latin
America) Essay no. 6, July 1976, p. 12.
89. Cf. on this point K. Schweinitz, Industrialisation and Democracy (London:
Collier-Macmillan, 1914). On the long-term pattern of inequalities in western
Europe cf. also the classic article by S. Kuznets, 'Economic growth ... ', 1955.
90. For evidence concerning Greece, cf. N. Vernicos, Greece Facing the Eighties
(in Greek) (Athens, 1975) p. 116; and M. Malios, The Present Phase of Capitalist
Development in Greece (in Greek) (Athens: 1975) pp. 139, 141. For Argentina cf.
C. F. Diaz-Aiejandro, Essays on the Economic History of the Argentine Republic
1970, pp. 129-30 and ECLA, Economic Development and Income Distribution in
Argentina (New York: United Nations, 1969) pp. 184-5 .. In Chile, as already
mentioned, foreign capital flowed in massively during Frei's administration (196470). Given Frei's systematic attempts at redistribution, inequalities in Chile did not
255
grow as dramatically as in Greece and Argentina during their periods of intensification of foreign-capital investments (in the early 1960s). Thus whereas during the
Alessandri period (1958-64) the percentage of the national income accruing to big
capital increased considerably, during the Frei administration it remained more or
less constant. On the other hand income concentration (within the bourgeoisie as
well as within both blue- and white-collar workers) increased significantly during the
1964-70 period. (Cf. Barbara Stallings, Class Conflict and Economic Development,
1978, pp. 216ff.)
91. Like any other 'universal' generalisation in the social sciences, the correlation of certain types of inequality with political radicalisation can only, of course,
be a tentative one. In certain contexts the combination of inequalities, with the
absence of starvation levels of poverty, may not lead to political radicalisation, for
instance, if the majority of the population is kept outside the political arena, or if
totalitarian controls succeed in keeping radical mobilisation to a minimum.
Nevertheless I believe that in ACG there did exist a number of conditions
facilitating the linkage between inequalities and radicalisation: the post-war
inclusion of large sections of the population in active politics, high levels of literacy,
the relative weakening of clientelistic controls in the urban centres and, to a lesser
extent, in the countryside, and so on.
92. For the concept of the distribution of economic, civil and political rights cf.
T. H. Marshall, Class, Citizenship and Social Development (New York: Doubleday, 1964).
93. For a development of this point in relation to the Latin American tradeunion movement cf. H. A. Landsberger, 'The labor elite: Is it revolutionary?', in
S.M. Lipset and A. Solari (eds), Elites in Latin America (New York: Oxford
94. Cf. Theda Scocpol, States and Social Revolution: A Comparative Analysis of
France, Russia and China (London and New York: Cambridge University Press,
1979) pp. 51-66, 81-98.
95. Cf. on this point E. Wolf, Peasant Wars of the Twentieth Century (London:
Faber, 1971); and Ellen K. Trim berger, Revolution from Above (New Brunswick,
N.J.:Transition Books, 1978) pp. 41ff.
96. Cf. on this pointS. Huntingdon, Political Order in Changing Societies (New
Haven and London: Yale University Press, 1968) p. 199.
97. For empirical evidence on the conservative attitudes of the industrial
bourgeoisie cf. D.L. Johnson, 'The national and progressive bourgeoisie in Chile',
in J. Cockroft et al. (eds), Dependence and Underdevelopment: Latin America's
Political Economy (New York: Doubleday, 1972); as far as Argentina is concerned,
cf. J. Petras and T. Cook, 'Dependency and the industrial bourgeoisie', in J. Petras
(ed.), Latin America: From Dependence to Revolution (New York: John Wiley,
1973).
98. For a development of this point with reference to the political organisations
of the 'marginalised' urban lower classes cf. J.M. Kuhl, 'Urbanization and political
demand measurement in Latin America', Latin American Research Review, no. 1,
vol. 24, 1979.
99. Not only the Chilean but even the Argentinian parties, despite their
particularism, were playing a much more important role in the political system than
parties in countries like Nicaragua or Guatemala. (Cf. R. H. McDonald, Party
Systems and Elections in Latin America (Chicago: Markham, 1971) pp. 2ff.
100. For the southern cone countries cf. F. H. Cardoso and E. Faletto,
Dependency and Development in Latin America (Berkeley: University of California
Press, 1979) pp. 29-74. It is true, of course, that contrary to Argentina and Greece,
Chile's two main export sectors (nitrates until the 1920s and copper) were almost
256
completely in foreign hands. However, unlike the 'banana republics', the Chilean
state was strong enough to tax these foreign interests and to use some of these
revenues (especially in the 1930s and afterwards) for the development of the
national economy.
101. Cf. E.K. Trimberger, Revolution from Above, 1978.
102. Needless to say, the mixture of incorporative/exclusionist controls mentioned here is not the only way in which the lower classes are 'kept in their place' in
the parliamentary semi-periphery. Also important are predominantly economic
mechanisms of control (for example, the type of control the traditional landlord
exercises over his peasants through his advancement of personal loans), as well as
predominantly ideological ones (for example, through specific types of primary or
secondary socialisation).
103. For a detailed account of the various incorporative and exclusionist
mechanisms of control cf. G. Katiphoris, The Barbarians' Legislation (in Greek)
(Athens: 1975).
104. Cf. Ch. 1.3.3.
105. It has been pointed out- rightly, I believe- that the Chilean party system
has a dual, two-tier structure. At the national centre, the system is
characterised by its highly ideological nature and its pragmatic orientation. This
contrasted sharply with the local arena of neighborhoods, small towns and rural
areas in which payoffs and political favors were more important, although
ideology did play a role. Indeed, much of the Chilean style of electoral
campaigning depended on face-to-face contacts and the delivery of particularistic favors. In small constituencies, candidates related directly to the voters; in
larger ones, they communicated through an array of local brokers. This two-tier
system, with a national ideological arena and a local electoral one, was bound
together by vertical party networks of an essentially clientelistic nature. In
exchange for votes, a congressman interceded with the bureaucracy to obtain
necessary goods and services for individuals and communities. The existence of
local clientele systems indicated that an enormous amount of the legislator's
effort and time had to be expended in performing small favors. In a coalition
government such as Allende's, elements of the coalition had to divide up the
spoils for their electoral clientele. Rivalries resulted not only over top ministerial appointments, but also over patronage at all levels of the bureaucracy, and
over the scarce resources available for new programming.
Quoted from A. Valenzuela, 'Political constraints in the establishment of
socialism in Chile', in A. and J. S. Valenzuela (eds), Chile: Politics and Society
(New Brunswick, N. J.: Transaction Books, 1976) p. 13. For the development
of clientelistic relations in the Chilean countryside cf. Crist6bol Kay,
'Transformaciones ... ', 1980. On some of the organisational weaknesses of the
Chilean socialist party cf. B. Pollack, 'The Chilean Socialist Party: Prolegomenon
to its ideology and organisation', Journal of Latin American Studies, May 1978.
106. Cf. C. J. Parrish, 'Bureaucracy, democracy and development: Some considerations based on the Chilean case', in C. E. Thurber and L. S. Graham (eds),
Development Administration in Latin America, (Durham: Comparative Administration Group Series, 1973) cf. also J. Petras, Politics and Social forces in Chilean
Development (Berkeley: University of California Press, 1969) ch. 8.
107. Cf. A. Angell, Politics and the Labour Movement in Chile (London: Oxford
University Press, 1972); cf. also J. 0. Morris, Elites, Intellectuals and Consensus: A
Study of the Social Question and the Industrial Relations System in Chile (Ithaca,
N.Y.: State School of Industrial and Labor Relations, 1961).
257
108. For example, the considerable development of rural organisation and the
mobilisation/radicalisation of the peasantry in Chile during recent decades clearly
indicates the weakening of the landlords' clientelistic controls over the peasants.
Cf. for example J. Petras, Politics and Social Forces in Chilean Development, 1969;
R. R. Kaufman, The Politics of Land Reform in Chile (Cambridge, Mass.: 1972);
and C. Kay, 'Transformaciones ... ', 1980. A similar weakening of clientelistic
controls has occurred in the Greek countryside during the last two decades. Cf. M.
Comninos, The Development of the Patronage System in Aitolo-Akharnania and
Kava/a, Ph.D. thesis, London School of Economics, 1984; and G. Mavrogordatos,
The Stillborn Republic, 1983.
109. For the development of this point cf. N. Mouzelis, 'Class and clientelistic
politics: The case of Greece', Sociological Review, November 1978.
110. As will be quite obvious by now, I look at the socio-political processes
leading to the dictatorial regimes both from a systemic perspective (that
is, in terms of contradictions/incompatibilities between institutionalised parts of the
political system) and from a collective action perspective (that is, in terms of the
major actors' political strategies, projects and struggles). For a theoretical elaboration of the fundamental distinction between system contradictions and social
conflict cf. D. Lockwood, 'Social integration and system integration', in G. K.
Zollschan and W. Hirsch (eds), Explorations in Social Change (London: Routledge, 1964); see also N. Mouzelis, 'Social and system integration: Some reflections
on a fundamental distinction', British Journal of Sociology, no. 4, 1974.
111. The distinction between economic and political contradictions is of course
an analytical one. In actual cases these two aspects are so closely interrelated that
they fuse, leading to an acute crisis of the whole system. However, despite the
difficulty of separating the economic from the political aspects of a crisis in a
concrete situation, the distinction is a very useful one: the political dimension, for
instance, in some cases (for example, Greece) is more important than the economic
one.
112. For the concept of the distribution of different types of rights cf. T. H.
Marshall, Class, Citizenship and Social Development, 1964, pp. 71-2.
113. On this point cf. R. Grew, 'The crises and the sequences', in R. Grew (ed.),
Crises of Political Development in Europe and the United States (Princeton:
Princeton University Press, 1978).
114. According to Kuznets (the main theorist and researcher on problems of
growth and inequality), all western European societies experienced a growth of
income inequalities during the initial stages of industrialisation. Basing his conclusions on long-term time series, he tentatively sets the growth period of inequalities
from 1780 to 1850 for Britain, and from 1840 to 1890 for Germany and the United
States. The major reasons behind such unequal growth were (a) that the increasing
importance of the urban sectors accentuated over-all inequalities, given that the
average per capita income of the rural population tends to be lower and income
differentials narrower than for the urban population; and (b) as a country moves
away from a subsistence economy, the ensuing surplus tends to be appropriated by
the economically dominant rather than the lower classes. Kuznets argues that the
initial phase of growing income differentials eventually attains a stage where
inequalities reach a plateau and are stabilised. In a subsequent third period (which
for most of western Europe came after the First World War), income inequalities
tend to decrease again. Cf. S. Kuznets, 'Economic growth and income inequality',
American Economic Review, vol. 49, March, 1955; cf. also his 'Quantitative aspects
of the economic growth of nations', Economic Development and Cultural Change,
January 1963; and his Modern Economic Growth: Rate, Structure and Spread (New
Haven: 1966).
258
Most theorists accept Kuznet's broad generalisations about initial growth,

stabilisation, and decrease in income inequalities in western capitalist societies, but
there are also some dissenting voices. Cf. for instance L. Soltow, 'Long-run
changes in British income inequality', Economic History Review, April 1968.
115. Cf. R. Dahrendorf, Society and Democracy in Germany (London: Weidenfeld & Nicholson, 1965).
116. Cf. M. Duverger, 'The eternal morass: French centrism,' in M. Dogan and
R. Rose (eds), European Politics: A Reader (London: Macmillan Press, 1971). Cf.
also C. Micaud, 'French political parties: Ideological myths and social realities', in
S. Neumann, Modern Political Parties (Chicago: University of Chicago Press,
1956).
117. From 1947 to 1956 the American aid given to Greece amounted to $2.565
million, the highest per capita aid advanced to any third-world country during this
period. Cf. T. Couloumbis, Greek Political Reaction to American and NATO
Influences (New Haven and London: Yale University Press, 1966) p. 28.
118. Before the start of American aid to Greece, US Ambassador to Greece,
MacVeagh, met King George II and set the conditions for such aid grants:
He felt that the American public should be convinced by deeds that the Greek
government was not oligarchic and reactionary, that democratic institutions
were not stifled and that all non-Communist Greeks were united. Therefore he
recommended such measures as broadening the Greek government, moderation towards the Centre and Left ... '
From T. Couloumbis, ibid., 1966, p. 27.
119. 'The Athenian', Inside the Colonels' Greece (London: Chatto & Windus,
1972) p. 36 (translated by Richard Clegg).
120. T. Katsanevas, The Industrial-Relation System in Greece: Historical Development and Present Structure, Ph.D. thesis, London School of Economics, 1980,
pp. 189ff.
121. As late as 1966 in Greece 3. 7 per cent of the GNP was going to military
expenses. In Argentina at the same time the equivalent was 1.5 per cent and in
Chile 1.9 per cent (Cf. A. Rouquie, L'Etat Militaire en Amerique Latine (Paris:
Seuil, 1982), p. 186.
122. On all these points cf. V. Kapetanyiannis, Army and Politics in Post-War
Greece, Ph.D. in progress, Birkbeck College, University of London.
123. Cf. on this point S. N. Gregoriadis, The History of Modern Greece (in
Greek) (Athens: Kapopoulos, 1974) pp. 45ff. Seen in long-term historical perspective, the power position of the throne started weakening even before the civil war
and the post-war political developments. As early as 1875 the establishment of
dedilomeni (a parliamentary practice which did not allow the king the right to
appoint minority governments) to some extent reduced blatant royal interference
in parliamentary affairs. The demise of oligarchic parliamentarism and the
broadening of political participation brought a further decline in the power position
of the throne, the new political forces being less loyal to the king, or rather less
prepared to leave military and foreign affairs as the exclusive preserve of the
monarch. (Cf. on this point D. Kitsikis, 'L'evolution de I'elite politique Greque', in
M. B. Kiray (ed.), Social Stratification and Development in the Mediterranean
Basin (Hague Institute of Social Studies Publications, 1973). Moreover, the advent
of Venizelos, although, of course, it did not bring the emergence of modern mass
parties, saw the development and strengthening of national party organisations (for
example, the Liberal Party established offices in various provincial towns). This
259
meant that political leaders began to have more control over their members and
greater bargaining power vis-a-vis the king. With increasing party discipline, the
ability of the king to manipulate politicians and elections decreased. (I am speaking
in relative terms, of course.) Such an ability did not entirely disappear, but royal
manipulation was not as easy in the twentieth century as it had been in the
nineteenth. So it was very simple for King George in 1868 to dismiss premier
Koumoundouros (who, despite his overwhelming majority in parliament, gave in
without protest) over a disagreement on the Cretan question, but much less easy
for King Constantine to get rid of Venizelos in 1915.
124. Cf. T. Katsanevas, The Industrial-Relation System in Greece, 1980, pp. 96,
99.
125. Empirical evidence on this point is given in M. Comninos, The Development of the Patronage System, 1984.
126. Cf. S. N. Gregoriadis, The History of the Dictatorship (in Greek) (Athens:
Kapopoulos, 1975) vol. 1, p. 14. The Pericles Plan should not be confused with the
Prometheus Plan used by the same team to take power in April 1964.
127. IDEA was founded in Athens in 1944 by the merging of two pre-existing
right-wing officers' groups. The most interesting insider's account of IDEA is given
by G. Karayannis, 1940-1952: The Greek Drama (in Greek) (Athens) n.d.
128. For instance, in February 1945 a precarious agreement was concluded
between Left and Right, the so-called Varkiza Agreement. The IDEA officers
sought to ensure that articles of the Varzika Agreement, which provided for the
purge from the army, police, and other state agencies of all those who had
collaborated with the Nazis, would remain a dead letter. In the army at least they
were very successful. Republican officers appointed immediately after the Varkiza
Agreement were soon dismissed again from the forces; officers who had participated in the Security Battalions (right-wing organisations collaborating openly with
the Germans) on the other hand were readmitted to the service. By 1946 the aims
of IDEA were fully achieved: the Greek army was totally purged of 'unhealthy'
elements and the IDEA officers were firmly established in its key positions.
129. For a detailed account of the construction of the anti-communist state cf. N.
Alivizatos, Les 1nsitutions Politiques de Ia Grece 1922-1979, (Paris: Pichon, 1979);
cf. also G. Katiphoris, The Barbarians Legislation, 1975.
130. Cf. on this point, V. Kapetanyiannis, Army and Politics in post-war Greece.
131. Such a group did exist within the army but, as it was established later, it
consisted of a rather insignificant gathering of junior officers keen to defend their
professional interests. There was no conclusive evidence of 'subversive' activities
and of Andreas Papandreou's involvement.
132. To be more precise, among the 'Big junta' generals there were disagreements about whether the coup should be staged before, during or after the elections
- as well as disagreements about whether one should intervene with or without
royal approval. Cf. on this point V. Kapetanyiannis, Army and Politics in Post-War
Greece.
133. Concerning the captains' association cf. General Panourgia's report on the
events that led to the coup, presented to Karamanlis in June 1967 (newspaper
Acropolis, issue 20 August 1974). Concerning the social gulf between high- and
low-ranking officers, one should take into account that the civil war requirements
lowered the Military Academy's standards of recruitment, and for the first time
established a system of free education (Law 577/22-9-1945). As people of poorer
backgrounds could now study at the Academy, there was a distinct difference in
class origins between officers who had graduated before and after the war. The top
leadership at the time of the coup belonged to the former cohort, while the
majority of the middle- and low-ranking officers belonged to the latter. For
260
statistics on the Greek officer class origins cf. D. Smokovitis, A Special Social
Group: The Greek Armed Forces, Ph.D. thesis, University of Salonika, 1975.
134. For detailed information on the organisational and cultural links between
the Greek and the United States armed forces cf. V. Kapetanyiannis, Army and
Politics in Post-War Greece.
136. Ibid., pp. 162ff.
137. During the Aramburu administration the army was divided into three
factions: the quedantistas who wanted a continuance of military rule until Per6nism
should be completely eradicated; the continuistas who were in favour of elections
provided they did not go against the military's goals; and those who wanted honest
elections- but without permitting Per6nists to nominate candidates. By 1962 these
three factions had evolved into the gorillas or colorados, and the legalistas or
azules. Cf. P. G. Snow, Political Forces in Argentina (New York and London:
Praeger, 1979).
138. The austerity policy was enhanced when later on the military forced the
president to replace his economic team with a new one headed by Alvaro
Alsogaray, a former air force officer and a hardline free-marketeer. Cf. Gary W.
Wynia, Argentina in the Post-War Era: Politics and Economic Policy-Making in a
Divided Society (Albuquerque: University of New Mexico Press, 1978) pp. 90ff.
139. Per6n's opposition to the United States lessened after 1949 with the failure
of his nationalist economic policies. However, even after that date Per6nist
Argentina often dragged her feet in matters of inter-American co-operation and
security. Cf. on this point H. F. Peterson, Argentina and the United States
1810-1960 (New York: State University of New York, 1964) pp. 465ff.
140. Ibid., p. 505.
141. A. Rouquie, Pouvoir Militaire, 1978, pp. 481ff.
142. Ibid., p. 470.
143. The Per6nists acquired 45 of the 86 contested seats in the Chamber of
Deputies, and 10 of the 14 governorships
144. In Argentina the voting population was increased both comparatively early
and relatively abruptly. For instance, with the Saenz Peiia 1912 electoral-reform
Jaw the number of voters jumped from 9 to 30 per cent of the population. It
doubled again in 1949 when women were given the vote. In Chile on the other
hand, as late as 1949 only 20 per cent of the population fulfilled the conditions for
inscription on the electoral registers. Moreover in 1955, when voters in Brazil
accounted for 15.6 per cent of the total population, the figure in Argentina was 41.8
per cent. Cf. A. Rouquie, Pouvoir Militaire, 1978, pp. 698-9.
145. On the growing inequalities and the relative deterioration of the economic
position of the lower classes in the post-Per6n era, cf. ECLA, Economic Development, 1969. In addition to the over-all trend of rising inequalities at that period one
can see growing inequalities within the middle and working-class sectors. Cf. on this
point G. A. O'Donnel, Modernisation and Bureaucratic Authoritarianism (Berkeley: Institute of International Studies, University of California) pp. 138ff.
146. On the radicalisation/transformation of the Per6nist movement after 1955
cf. D. Rock, 'The survival and restoration of Per6nism', in D. Rock (ed.),
Argentina in the Twentieth Century (London: Duckworth, 1975).
147. The old Per6nist trade-union leadership was not allowed to hold office.
148. There was also a group of 19 organisations controlled by communists. Cf.
P. G. Snow, Political Forces in Argentina, 1979, p. 89.
149. Cf. D. C. Hodges, Argentina 1943-1976: The National Revolution and
Resistance (Albuquerque: University of New Mexico Press, 1976) p. 40.
150. Ibid., pp. 32ff.
261
151. Under the decree of June 1963, any party that was found to run candidates
connected to the Uni6n Popular would have had its entire list of candidates
banned.
152. G. W. Wynia, Argentina in the Post-War Era, 1978, p. 117.
153. Cf. D. C. Hodges, Argentina 1943-1976, 1976, pp. 45--6.
154. Cf. P. Ranis, 'Peronismo without Per6n: The years after the fall, 19551965', Journal of Inter-American Studies, vol. 8, 1966, p. 122.
155. Cf. M. Goldwert, Democracy, Militarism and Nationalism, 1972, p. 201.
156. For instance Vandor and his lieutenants made a pre-electoral deal with the
Frondizistas without consulting Per6n. (Cf. A. Rouquie, Pouvoir Militaire, 1978, p.
554).
157. Cf. G. W. Wynia, Argentina in the Post-War Era, 1978, pp. 120ff.
159. Cf. R. Potash, The Army and Politics in Argentina 1945-62 (London: The
Athlone Press, 1980) pp. 313ff.
160. Cf. on this point Alfred Stepan, 'The new professionalism of internal
warfare and military role expansion', in A. F. Lowenthal, Armies and Politics,
1976; cf. also Philippe C. Schmitter (ed.), Military Rule in Latin America (Beverly
Hills and London: Sage, 1973).
161. For figures on the increase of military assistance cf. A. Rouquie, L' Etat
Militaire, 1982, pp. 168-9.
162. The most famous school providing such courses outside the United States
was the Army School of the Americas (USARSA) at Fort Gulick in the Panama
Canal zone. In 1963 the school was reorganised and henceforth all courses were
given in Spanish or Portuguese. By 1974 more than 30000 students had gone
through USARSA's various training programmes. (Cf. A. Rouquie, ibid., p. 170.)
163. Between 1950 and 1965 the number of Argentine military who were trained
in the United States amounted to 1375; during the much shorter period 196.5-70
more than 1000 officers went to the US for training. Concerning US military aid to
Argentina, this rose from $3 million in the 1953-61 period to $5.6 million between
1962 and 1965. (Cf. A. Rouquie, ibid., pp. 168-9.)
164. Cf. G. A. O'Donnel, 'Modernisation and military coups: The Argentine
case', in A. F. Lowenthal (ed.), Armies and Politics, 1976, pp. 206-7.
165. Cf. A. Rouquie, Pouvoir Militaire, 1978, pp. 547-8.
166. The participation of Latin Americans in the peace force was minimal. To
the 21 000 US troops, all of the other central American states together provided 450
men and Brazil 1150. (Cf. J. W. Pratt et al., A History of United States Foreign
Policy (Englewood Cliffs, N.J.: Prentice-Hall, 1980, p. 516.)
167. M. Goldwert, Democracy, Militarism and Nationalism, 1972, p. 201.
168. In fact, in the period immediately preceding the coup, everybody was not
only expecting the coup but most groups (including several Per6nist groups and
Frondizi's supporters) were in favour of military intervention. (Cf. C. A. Astiz,
'The Argentine armed forces: Their role and political involvement', Western
Political Quarterly, December 1969, pp. 873ff.) Of course, most of them wanted a
short intervention, merely for the purpose of replacing lilia, but given the
prevailing balance of power they did not have much say in how things were going to
develop after the coup.
169. B. Loveman, Chile, (New York: Oxford University Press, 1977) p. 257.
170. They did so mainly under pressure from the National Agricultural Society,
a powerful lobby promoting landlord interests.
171. Cf. M.D. Wolpin, Cuban Foreign Policy and Chilean Politics (Lexington,
Mass.: Lexington Books, 1972) p. 89.
172. Cf. B. Loveman, Chile, 1977, p. 287.
262
173. Cf. E. Halperin, Nationalism and Communism in Chile (Cambridge, Mass.:

MIT Press, 1965) pp. 52-8.
174. Cf. J. Petras, Politics and Social Forces in Chilean Development, 1969, pp.
51ff.
175. For a discussion of these developments cf. R. Alexander, The Tragedy of
Chile, (Westport, Conn.: Greenwood Press, 1978) pp. 39ff.
176. Cf. on this point C. Kay, 'Transformaciones ... ', 1980, pp. 777ff.
177. Statistics on voter registration and participation show this increase very
clearly. So in the 1946 presidential elections only 12 per cent of the population was
registered to vote, and 9 per cent actually went to the polls. In the 1958 presidential
elections the two percentages were 21 and 18, and in the 1964 ones they had gone
up to 36 and 32 respectively. Cf. R. H. McDonald, Party Systems and Elections,
1971.
178. Cf. W. Thiesenhusen, Chile's Experiment in Agrarian Reform (Madison,
Wise.: 1966); and J. Swift, Agrarian Reform in Chile (Lexington, Mass.: 1971).
179. To this figure one should add 250000 public-sector employees and workers
whose associations, although technically not considered as trade unions, very much
operated as such. Cf. Jorge Tapia Videla, 'The difficult road to socialism', in F. G.
Gil, eta/. (eds), Chile at the Turning Point (Philadelphia: Institute for the study of
human issues, 1979) p 137.
180. I. Roxborough eta/., Chile: The State and Revolution (London: Macmillan
Press, 1977) p. 137.
181. Cf. B. Stallings, Class, Conflict and Economic Development in Chile, 1978.
182. Cf. M. J. Cavarozzi and J. Petras, 'Chile', in R. H. Chilcote and J.
Edelstein (eds), Latin America: The Struggle with Dependency and Beyond (New
York and London: John Wiley, 1974) pp. 546ff.
183. Cf. on the above points S. Bitar, 'The inter-relationship between economics
and politics', in F. G. Gil eta/. (eds), Chile at the Turning Point (Philadelphia:
Institute for the Study of Human Issues, 1979).
184. Cf. R. J. Alexander, The Tragedy of Chile (London: Greenwood Press,
1978) pp. 251ff.
185. Cf. on this point Luis Maira, 'The strategy and tactics of the Chilean
counter-revolution in the area of political institutions', in F. G. Gil eta/. (eds),
Chile at the Turning Point, 1979.
186. Cf. I. Roxborough et al., Chile: The State and Revolution, 1977, p. 183.
187. Ibid., pp. 218ff.
188. For the long tradition of hostility to Marxism among the Chilean military cf.
F. M. Nunn, 'New thoughts on military intervention in Latin American politics:
The Chilean case', Journal of Latin American Studies, November 1975.
189. Cf. J. Nef, 'The politics of repression: The social pathology of the Chilean
military', Latin American Perspectives, vol. 1, no. 3, p. 61.
190. Ibid., pp.62-3.
191. Ibid., p. 64.
192. Cf. I. Roxborough et a/. Chile: The State and Revolution, 1977, p. 192.
193. F. M. Nunn, The Military in Chilean History, 1976, pp. 220ff.
194. For instance for the period 1962-5 the Chilean army received $169.8 million
in military aid - approximately seven million more than the much larger Argentine
army. (Cf. A. Rouquie, L'Etat Militaire, 1982, p. 168.) Concerning the training of
officers between 1965 and 1970, Chile sent 1761 officers abroad for this purpose,
against only 1117 from Argentina's larger armed forces. (Ibid., p. 169).
195. For a detailed exposition of the various types of aid cf. M. D. Wolpin,
Cuban Foreign Policy, 1972, pp. 61-96.
196. Ibid., p. 90.
263
197. Cf. I. Roxborough eta/., Chile: The State and Revolution, 1977, p. 275.
199. Cf. General Carlos Prats, Una Vida par Ia Legalidad (Mexico: Fondo de
Cultura Econ6mica, 1976) pp. 76-80.
200. Joan Garces, Allende y Ia Experiencia Chilena: Las Armas de Ia Politica
(Barcelona: Edit. Ariel, 1976) pp. 295-7.
201. F. M. Nunn, 'New thoughts on military intervention in Latin American
politics: The Chilean case, Journal of Latin American Studies, November 1975.
202. Cf. R. Potash, The Army and Politics in Argentina, 1980, p. 338.
203. Cf. on this point R. R. Kaufman, 'Industrial change and authoritarian rule
in Latin America: A concrete review of the bureaucratic-authoritarian model', in
D. Collier (ed.), The New Authoritarianism in Latin America (Princeton: Princeton
University Press, 1979) pp. 214ff.
204. Cf. A. Hirschman, 'The turn to authoritarianism in Latin America and the
search for its economic determinants', in D. Collier, ibid., 1979, pp. 76-7.
205. Cf. on this point G. E. Kourvertaris, 'Professional self-images and political
perspectives in the Greek military', American Sociological Review, vol. 36, 1971.
Cf. also Kapetanyiannis, Army and Politics in Post-War Greece.
206. For an extensive analysis of the bureaucratic-authoritarianism term see G.
O'Donnel, Modernisation and Bureaucratic Authoritarianism, 1973. For a critique
of the concept cf. F. H. Cardoso, 'On the characterisation of authoritarian regimes
in Latin America', in D. Collier (ed.), The New Authoritarianism, 1979. F6r a
critique of O'Donnel's theory see ch. 4.
207. On the concept of sultanism cf. Max Weber, Economy and Society, (eds)
G. Roth and C. Wittich, University of California Press 1978, vol. 1, pp. 231-2; cf.
also J. Linz, 'Totalitarian and authoritarian regimes', in F. Greenstine and N.
Polsby (eds), Handbook of Political Science (Reading, Mass.: 1975) vol. 3.
208. Cf. J. Linz, 'Notes towards a typology of authoritarian regimes', paper
delivered at the 1972 Annual meeting of the American Political Science Association.
209. J. Linz uses the term 'mentalities' rather than ideology to characterise the
Iegitimising attempts of authoritarian regimes. Cf. his 'Notes towards a
typology ... ', ibid., 1972.
210. A third possible solution, nearer in spirit to parliamentary democracy,
would be the establishment of a dominant one-party political system which would
maintain some degree of both social and political pluralism, such as, for instance,
the Mexican party system. With respect to this 'Mexican solution', one should
remember that the PRJ was able to achieve its exceptional position within Mexico's
polity both because of the legitimacy bestowed upon it by the 1910 revolution, and
because of its relatively early institutionalisation (that is, before the large-scale
expansion of industrial capitalism and the mass mobilisation always associated with
it). A similar solution in other Latin American societies seems rather unlikely at the
present level of industrialisation and political mobilisation, and in the absence of a
legitimising revolutionary tradition. (Cf. B. Anderson and J. D. Cockroft, 'Control
and co-optation in Mexican politics', International Journal of Comparative Sociology, no. 1, 1966; and P. Gonzalez Casanova, Democracy in Mexico (New York:
Oxford University Press, 1978).)
211. As F. H. Cardoso puts it ('On the characteristics of authoritarian regimes
in Latin America', Centre of Latin American Studies, University of Cambridge
Working Papers, Series no. 30):
Economic expansion requires technicians, competence and a certain cultural
sophistication. When this is found in the context of a dependent economy, the
society is necessarily open because a current of information, people and
264
attitudes accompanies the flow of merchandise. It is difficult to culturally 'close'

a society of this type. There are pressures which are uncontrollable because of
this fact.
Cf. also on this point N. Diamandouros, 'The 1974 transition from authoritarian to
democratic rule in Greece: A southern European perspective', paper delivered at
the conference organised by the Centro de lnvestigationes Sociol6gicas, Madrid,
1980.
212. In Greece it was not only the support of the bourgeoisie being no more than
lukewarm which explains the weak consolidation of the dictatorial regime. Another
fundamental factor (mentioned already) was the severe split in the military
establishment itself that had occurred before the 21 April1967 coup. So it was not
just that the Greek military dictatorship failed to gain large-scale popular support,
it also alienated the post-war political and military establishment. Unsupported
from either below or above, and faced with hostility from international opinion, its
position became very precarious after the severe economic crisis of 1973.
213. This does not mean that the military will be able to maintain their power
indefinitely. Despite repression and despite the (relative to Greece) greater support
from the upper classes, they have not succeeded in effectively suppressing
widespread active opposition to military rule. On the military's failure to completely destroy Chile's relatively strong pre-dictatorial party system cf. Arturo and J.
Samuel Valenzuela, 'Party opposition under the Chilean authoritarian regime',
paper presented at the World Congress of the International Political Science
Association, Rio de Janeiro, Brazil, August 1982.
214. By stable parliamentary rule I mean of course regime rather than governmental stability.
215. For a useful discussion of the distinction between 'regime' and 'state'
authoritarianism cf. F. H. Cardoso, 'On the characterisations of authoritarian
regimes in Latin America', in D. Collier, The New Authoritarianism, 1979.
Chapter4
1. Cf. for instance P. Coulter, Social Mobilisation and Liberal Democracy

(Toronto and London: Lexington Books, 1975); S. M. Lipset, 'Some social
requisites of democracy', American Political Science Review, March 1959; J.
Coleman and G. Almond (eds), The Politics of the Development Areas (Princeton:
Princeton University Press, 1960) pp. 542-76; P. Cutright, 'National political
development: Social and economic correlates', inN. W. Nelson eta!. (eds), Politics
and Social Life (Boston: Houghton Mifflin, 1963).
2. Cf. D. Cruise O'Brien, 'Modernisation, order and erosion of a democratic
ideal', in D. Lehman (ed.), Development Theory: Four Critical Essays (London:
Frank Cass, 1979).
3. For a representative work of this kind cf. F. W. Riggs, Administration in
Developing Countries: The Theory of Prismatic Society (Boston: Houghton Mifflin,
1964).
4. Cf. S. Huntington, Political Order in Changing Societies (New Haven: Yale
University Press, 1968). Another, more recent work that again emphasises the
weak institutionalisation of political institutions leading to permanent regime
instability is D. Chalmers, 'The politicised state in Latin America', in James M.
265
Malloy (ed.), Authoritarianism and Corporatism in Latin America (Pittsburgh:

University of Pittsburgh Press, 1977).
5. For a development of this point cf. D. Lockwood, 'Social Integration and
System Integration' in G. K. Zollshcau and W. Hirsch (eds), Explorations in social
change (London: Routledge, 1964).
6. Cf. for instance R. M. Moore, 'The heritage of Latin America', in H. J.
Wiarda (ed.), Politics and Social Change in Latin America: The Distinct Tradition
(Amherst: Massachussetts University Press, 1973)
7. S. Huntington, Political Order, pp. 4-5.
8. Ibid., p. 28.
9. Ibid., p. 32.
11. Huntington has tried to differentiate his approach from that of Parsons (cf.
his 'The change to change: modernisation, development and politics', Comparative
Politics, April 1971) but, as I shall argue below, in certain respects his approach
closely follows the Parsonian functionalist tradition.
12. On this point cf. N. Mouzelis, 'System and social integration: A reconsideration of a fundamental distinction', British Journal of Sociology, December 1974.
13. S. Huntington, Political order, 1968, p. 28
14. Cf. for instance J. C. Portantiero, 'Dominant classes and political crisis',
Latin American Perspectives, vol. 1, no. 3, Fall 1974. For the Greek case cf. M.
Nikolinakos, Resistance and Opposition 1967-1974 (in Greek) (Athens: 1975).
15. Cf. J. Nun, 'The middle-class military coup', in C. Veliz (ed.), The Politics of
Conformity in Latin America (London: Oxford University Press, 1967), and his
'The middle-class coup revisited', in A. F. Lowenthal (ed.), Armies and Politics in
Latin America (New York and London: Holmes and Meier, 1976).
16. J. Nun, 'The middle-class coup revisited', 1976, p. 55.
17. Cf. on this the well-known Miliband-Poulantzas debate: N. Poulantzas, 'The
problem of the capitalist state', New Left Review, no. 58; R. Miliband, 'The
capitalist state: A reply to Nicos Poulantzas', New Left Review, no. 59; N.
Poulantzas, 'The capitalist state: A reply to Miliband and Laclau', New Left
Review, no. 95; and E. Laclau, 'The specificity of the political: The PoulantzasMiliband debate', Economy and Society, vol. 4, no. 1.
18. Cf. N. Mouzelis, Modern Greece: Facets of Underdevelopment (London:
Macmillan Press, 1978).
19. G. O'Donnel, Modernisation and Bureaucratic authoritarianism; Studies in
Southern American Politics (Berkeley: Institute of International Studies, University of California, 1973). Although O'Donnel does not use a strictly Marxist
terminology, the basic structure of his argument comes very close to the work of the
Marxist 'logic of capital' school. For an exposition and presentation of some
representative articles in this intellectual tradition cf. J. Holloway and S. Picciotto
(eds), State and Capital: A Marxist Debate (London: Edward Arnold, 1978).
Alternatively, O'Donnel's work can be seen as a systematic synthesis of already
existing ideas on the relationship between dependent capitalist development and
authoritarianism - ideas developed by such writers as F. H. Cardoso, E. Faletto,
A. D. Hirschman, P. C. Schmitter, and so on.
20. In addition to his book Modernisation and Bureaucratic authoritarianism,
1963, O'Donnel has restated or reformulated his basic arguments in numerous
articles. Cf. for instance his 'Reflections on the patterns of change in the
bureaucratic-authoritarian state', Latin American Research Review, vol. 13, no. 1,
1978; 'Corporatism and the question of the state', in James M. Malloy (ed.),
266
Authoritarianism, 1977; 'Modernisation and military interventions', in A. Lowenthal (ed.), 1976. O'Donnel's work has generated a great deal of research and
debate: cf. for example various articles by F. H. Cardoso, J. Cotler, A. Hirschman,
R. R. Kaufman and others in D. Collier (ed.), The New Authoritarianism in Latin
America (Princeton: Princeton University Press), 1979); Jose Luis Reyna and
Richard Weinert (eds), Authoritarianism in Mexico (Philadelphia: Institute for the
Study of Human Issues, 1977); Kenneth P. Erickson and P. V. Peppe, 'Dependent
capitalist development, U.S. foreign policy and repression of the working class in
Chile and Brazil', Latin American Perspectives, no. 1, 1976.
21. For an exposition and detailed critiques of O'Donnel's thesis cf. the various
articles in D. Collier (ed.), The New Authoritarianism in Latin America, 1979. (Cf.
particularly the article by Jose Serra, 'Three mistaken theses regarding the
connection between industrialisation and authoritarian regimes', pp. 11lff.)
22. Cf. for instance Gerry Cohen, Karl Marx's Theory of History: A Defense
(Oxford and Princeton: Clarendon Press, 1978); and R. K. Merton, Social Theory
and Social Structure (Glencoe, Ill.: Social Press, 1963).
23. R.K. Merton, ibid., 1963, pp. 60ff.
24. For an extensive discussion of this point cf. below, sections 3 and 4.
25. This was clearly the case in Greece where the massive attraction of foreign
capital and its direction into industrial sectors requiring a complex technology had
started before the 1967 coup, before the establishment of the seven-year military
dictatorship. When the military intervened it was neither in order to initiate the
deepening process nor to carry it forward or to safeguard it from hyper-inflation or
paralysing strike activities. As argued already, the reasons for the coup were above
all political. Cf. N. Mouzelis, Modern Greece: Facets of Underdevelopment
(London: Macmillan Press, 1978) pp. 131ff.
26. For a review of the literature which draws attention to all these factors cf.
R. C. Rankin, 'The expanding institutional concerns of the Latin American
military establishments: A review article', Latin American Research Review, Spring
1974.
27. The literature on military interventions is full of this type of generalisation.
Cf. for instance S. Finer, The Man on Horseback: The Role of the Military in
Politics (New York: Praeger, 1962). Even a writer like Juan Linz, who has made
significant contributions to the study of authoritarian regimes, often has his
enormous erudition worsted by his determination to follow what he calls Weber's
'methodological individualism'. For instance in his The breakdown of democratic
regimes (Baltimore and London: John Hopkins University Press, 1978) he formulates generalisations pertaining to the breakdown of all types of democratic regimes
(in the centre as well as in the periphery), and quite predictably ends up with
statements that are either banal - for example, 'In every society there are those
who deny the legitimacy to any government and those who believe in alternative
legitimacy formulae. Regimes vary widely in the amount and intensity of citizen
belief in their legitimacy' (p. 17) - or inclusive/wrong, for example, 'It seems
unlikely that military leaders would turn their arms against the government unless
they felt that a significant segment of the society shared their belief and that others
were at least indifferent to the conflicting claims of allegiance' (p. 17).
28. Of course this is not true for all types of military intervention in ACG. For
instance, in inter-war Greek politics certain military interventions cannot be
accounted for in terms of either economic contradictions/class struggles, or
structural trends within the polity. (Cf. N. Mouzelis, Modern Greece, 1978, p. 110.)
29. For a systematic criticism of this type of literature .cf. my article 'Types of
reductionism in Marxist theory', Telos, Fa111980.
267
30. For a typical illustration of this type of analysis cf. S. Brunhoff, Etat et
Capital (Grenoble: Presses Universitaires de Grenoble, 1978); also J. Holloway
and S. Picciotto (eds), State and Capital, 1978.
31. For a book on Marxist political theory which, although in certain respects it
tries to overcome class reductionism, neglects the organisational aspects of class
and class struggles, cf. E. Laclau, Politics and Ideology in Marxist Theory (London:
New Left Books, 1977). On Marxism's failure to deal with the organisational and
institutional aspects of advanced capitalist societies cf. T. Johnson, 'What is to be
done? The structural determination of social class', Economy and Society, vol. 6,
no. 2, May 1977.
32. S. Huntington, Political Order in Changing Societies, 1968, p. 80.
33. For the adoption of such a theoretical position in the analysis of third-world
formations, cf. John Taylor, From Modernisation to Mode of Production: A
Critique of the Sociologies of Development and Underdevelopment (London:
Macmillan Press, 1979), pp. 132ff. For a critique of Taylor's book, cf. N. Mouzelis,
'Modernisation, underdevelopment, uneven development, Journal of Peasant
Studies, April 1980. For a more specific application to Latin America, cf. J. C.
Portantiero, 'Dominant classes', 1974, p. 94.
34. Cf. for instance Li Causi, 'Anthropology and ideology: The case of
patronage in Mediterranean societies,' Radical Science Journal, no. 1, 1975.
35. For a further development of this point cf. N. Mouzelis, 'Class and
clientelistic politics: 'The case of Greece', Sociological Review, November 1978.
36. One finds such attempts in the work of Antonio Gramsci, of Galvano della
Volpe, Lucio Colletti, and so on. For more recent writings cf. U. Cerroni,
'Democracy and socialism', Economy and Society, vol. 7, no. 3, August 1978; and
E. Laclau, Politics and Ideology, 1977; as well as his 'Democratic antagonism and
the capitalist state', mimeo.
37. The term technologies as used here refers of course not only to material tools
but also to organisational forms, to types of knowledge used in production etc.
38. In fact this is so much so that one could equally well speak of a political mode
of production, instead of a mode of domination.
39. R. Michels, Political Parties (New York: Collier, 1962).
40. For instance, the kind of large-scale political patronage on which big
business depends in peripheral and semi-peripheral capitalist formations may
benefit specific capitalists but not necessarily the development of capitalism in
general. Given that the criteria for assisting economic agents are often not even
based on the 'general interest of capital' but on the particularistic interests of
political favourites, it is quite misleading to build into the definition of the capitalist
state (as many structuralist Marxists do) the idea that the state, even when
relatively autonomous from the pressures of economically dominant groups, is
structurally bound to serve their over-all interests. For instance, there are
numerous examples in Greek parliamentary history of laws tailored to suit
particular and politically influential capitalists, and of cases where huge state
resources are channelled into the area where the political clientele of the relevant
minister happens to be located. (For a recent study which shows the extent to which
the state has prevented the extended reproduction of capitalism in Greece cf. M.
Petmesidou-Tsouloudi, Social Reproduction and cultural transmission within the
family: the case of middle class of Salonika, Ph.D., Oxford University, 1984; cf.
also A. Doxiadis, "Is there monopoly capital in Greek industry?", Sihrona
Themata, July-September 1984). Of course, in defence of the above criticism it can
be argued that in so far as capitalist relations of production remain dominant, the
state- irrespective of its specific economic policies- by definition contributes to the
268
maintenance of capitalism and in that sense helps the general interests of capital. If
such a watered-down version is accepted, then it is perfectly true that all states
which do not set about destroying capitalist relations of production are ipso facto
helping capital. But this is a banality which adds absolutely nothing to an
understanding of the nature of the state, either in the capitalist centre or in the
periphery.
41. Cf. George Mavrogordatos, Stillborn Republic: Social Coalitions and Party
Strategies in Greece 1922-1936 (Berkeley, Los Angeles and London: University of
California Press, 1983) p. 123. For an extensive application of the concept of state
bourgeoisie in the Greek case cf. C. Tsoukalas, 'The problem of political
clientelism in nineteenth-century Greece', in G. Kontogiorgis (ed.), Social and
Political Forces in Greece (in Greek) (Athens: Exantas, 1977); cf. also his otherwise
excellent work, Dependence et Reproduction:Le Role de l'Appareil Scolaire dans
une Formation Transterritoriale, Ph.D. thesis, University of Paris, 1975.
42. For an extensive discussion of some of these points cf. N. Mouzelis, 'On the
crisis of Marxist theory', British Journal of Sociology, March 1984.
43. A. Valenzuela and J. S. Valenzuela (eds), Chile: Politics and Society (New
Brunswick, N.J.: Transaction Books, 1976) p. 238.
44. Peter Smith, Politics and Beef in Argentina (New York: Columbia University
Press, 1969).
45. N. Poulantzas distinguishes class places from class practices. The first refers
to the 'objective' locations of agents in the technical and social division of labour;
the second term refers to such things as the political organisations or agencies
representing a class and the strategies/policies it formulates in a concrete conjuncture. Cf. his Les Classes Sociales dans le Capitalisme d'Aujourd'hui (Paris: Seuil,
1974) pp. lOff.
47. Cf. for instance A. A. Bor6n, The Formation and Crisis of the Liberal State
in Argentina 188~1930, Ph.D. thesis, Harvard University. 1976. One of course
finds similar interpretations for the Greek case. Cf. for example V. Filias, Society
and Power in Greece 183~1909 (in Greek) (Athens: Makriniotis, 1974).
48. As A. Rouquie has very aptly put it in discussing the Argentinian case:
In Argentina the army is neither the party of the middle classes, nor the
protector of the industrial bourgeoisie, the agrarian bourgeoisie or the multinationals. Its interventions modify the direction of sectoral transfers and inverse
social trends. (my translation)
Quoted from L'Etat Militaire en Amerique Latine (Paris: Seuil, 1982) p. 351.
49. Cf. for instance Louis Mercier-Vega, La Revolution par l'Etat: Une Nouvelle
Classe Dirigeante en Amerique Latine (Paris: Payot, 1978).
50. A third theoretical position that, within the Marxist tradition, tries to
overcome the deficiencies of the above theories can be found in the late writings of
Nicos Poulantzas (cf. his State, Power, Socialism, London: New Left Books, 1978).
For a discussion of the ways in which Poulantzas's more recent formulation of his
theory of the state differs from his previous positions cf. B. Jessop, For Poulantzas
(London: New Left Books, 1984) ch. 7. Although his theory is of more general
relevance, it is also meant to apply to the semi-peripheral capitalist state (he used it
to analyse the rise and fall of the Greek military dictatorship: cf. La Crise des
Dictatures: Portugal, Grece, Espagne (Paris: Seuil, 1975). According to Poulantzas,
the contemporary capitalist state is neither a mere instrument of the capitalist
classes nor has it become its subject/master. If the focus of analysis is shifted from
the level of actors and their strategies to the level of the capitalist system and its
269
institutionalised arrangements which constrain social and political action, then it is

possible to view the state as an institutional arena which reflects in a mediated,
condensed or distorted manner class struggles and contradictions. (For instance, in
the case of Greece, such fundamental developments within the political arena as
the rise and fall of the post-war military dictatorship were directly linked with
antagonisms between different fractions of capital.) However, this shift of focus
from action to system analysis does not help to overcome the reductionist/
economistic impasse. The state, despite its alleged autonomy, is still conceptualised
in economic categories, that is, in terms of struggles and contradictions that have
their origins basically in the economic division of labour. The political institutional
arena, although it does not passively mirror these struggles and contradictions, is
primarily defined in their terms: it reflects them (albeit in condensed and/or
refracted manner), while its ultimate function is always the extended reproduction
of the capitalist mode of production, that is, the maintenance of a specific type of
class domination. Thus the idea that the political arena or system itself might have
reproduction requirements and contradictions which cannot be easily translated
into those of capital is not seriously considered. What is even less considered is the
possibility of the reproduction requirements of the prevailing mode of domination
taking precedence over those of capital's extended reproduction whenever the two
are found to clash, that is, that political structural constraints might be more
decisive than economic ones in the formation of state policies.
51. According to C. F. Diaz-Alejandro, Essays on the Economic History of
Argentina (New Haven and London: Yale University Press, 1970), pp. 106ff, this is
precisely what happened during the decade of Peron's first rule - especially in the
late 1940s.
52. Cf. D. Lockwood, 'Social integration and system integration', in G. K.
Zollschan and W. Hirsch (eds), Explorations in Social Change (London: Routledge, 1964). Cf. also N. Mouzelis, 'Social and system integration: Some reflections
on a fundamental distinction', British Journal of Sociology, no. 4, 1974.
53. Cf. on this point N. Mouzelis, 'Types of reductionism in Marxist theory',
1980; and his 'On the crisis of Marxist theory', 1984.
Index
Subject references appear in roman type, author references in italic.

absolutism
western European xv-xvi, 224 3
Agrarian Party (Greece) 44
Agrarian Peasant Union (Bulgaria) 4,
35-8,91
agrarian reform
Balkans 231 77
Bulgaria 37
Chile 162, 163
Agrarian Reform Project 168
agriculture
commercialisation 3, 31-2
decline 248 5657
inefficiency 115
labourforce 231,24970
population 233 78
price control 246 19
production 24856
superseded by industry 250 74
Aguirre Cerda, Pedro 103
aid
to Greece 258 117
Alavi, H. 242 2
Alba, Victor 239 192
Albania 224 1
land ownershi~ 231 76
Albert, B. 228 1 , 23049
Alem, Leandro 19
Alessandri, Arturo 28, 63, 102, 158,
231 69 ,244 4
Alessandri, Jorge 116, 117, 162, 2559<1
Alexander, King of Bulgaria 226 1
Alexander, R.I. 239181.185.187.188.189,
262175.184
Alford, R. 239 193
Alivizatos, N. 259 129
Allanca de Liberacao Nacional 66
Allcock,J.B. 244 28
Allende, Salvador 161, 162, 163, 164,
166,167-9
Alliance for Progress 155, 162

Allub, L. 245 8
Almond, G. 264 1
Alonso, Jose 153
Alsogaray, Alvaro 260 138
Althusserian structuralism 217
Alvear, Marcelo 25
American Labor Yearbook 238 173
Amin, S. 2279,25073
anarcho-syndicalists
Brazil 66
Anastasopoulos, George 236 139
Anderson, B. 263 210
Anderson, Perry 224 3,231 75
Angell, Alan 64,238 178 ,239 182 ,256 107
Angelopoulos, A. 232 88
anti-elitism
populism as 88-9,90
anti-Venezelists
class support 47
Aramburu, General 146-8, 149,
260 137
Argentina 222,224 1
conscription 13
economicproblems 113-14
immigrants 19
industrialisation 5, 51, 53, 54, 11314
landowners 16-19,21
lower class incorporation 126-7
military coups 25-6,68,99,23063
military dictatorship xvii, 145-58
military interventions 99-101,
2458.10.11' 261168
military links with USA 147-8, 1578
national unity 228 12
oligarchic parliamentarism 16-19,
226 1
public employment 12, 22937
270
Index
trade union movement 67-9
urban populism 16--27
urbanisation 9
Argentine Rural Society 18, 21,209
aristocracy
Ottoman empire 225-67
army
see military
Army School ofthe Americas 261 162
Army Society for Regeneration 102
Arriagada, General 103
artisan class
decline-of 25074
artisanal units
family operated 120
Asia Minor
refugees 44
ASPIDA 143, 144
Astiz, C.A. 261 168
Athenian, The 258 119
Athens-Piraeus conurbation 249 71
Attalides, M. 2435
authoritarianism
bureaucratic 178-9,180,189-99,
220-1, 263206
Ottoman and Iberian empires xvxvi
state 74, 132-3
authority relationships
in populism 87-8,90-1
A vineri, Shlomo 228 13
azules 156--7, 260
Bairoch, Paul 241 198 ,251 74
Balkan Wars 24625
Balkans 224 1
agrarian reform 231 76
communism 56-8
industrialisation 52, 53-5
landownership 29-30
middleclasses 31
Ottoman rule xiii, 225-67
parliamentary institutions xiii-xiv
peasantconditions 33-5
rural underemployment 238
trade unions 70
western economic penetration 34,
228 16
working-class movement 55-62
Banks, A.S. 228 18 1922 ,
22936,38,39,40,41, 238161
BANU 4, 35-8, 91
Baran, P. 22~
271
Battalions of National Defence 137

Battle of Athens 142
Battle y Ordonez 235-6 138
Bauer, Arnold 109, 24737 38 39 40
Bell,!. D. 23396 101 , 243 13
Bendix, R. 22r
Berend, I.T. 23278 79 , 237 142 146 , 238 162
Bernstein, H. 2278
Bichens, H. E. 24623
Bismarck, Otto von 132
Bitar, S. 262 183
Boeke, J.H. 253 81
BogotaCharter 147
Bolivia 54, 238161
Boron, A.A. 23Q4346 48 64 , 26847
bourgeois revolution
1909 military coup as 206
failed 211-12
bourgeoisie
Bulgarian 23286
industrial 240-1 194 ,25597
state 207-8, 26841
see also middle classes
Braude[, F. 240194 ,25274
Brazil 2242
clientelism 83-6
industrialisation 52, 53
military dictatorship 157
oligarchic parliamentarism 278,
226 1
state intervention 10
urban populism 27-8
urbanisation 9
Brown, K. 243 5
Brunhoff, S. 267 30
Bulgaria 2241
agrarian structure 30-2
communist movement 56-8
immigration 234 119
industrialisation 54, 187
oligarchic parliamentarism 4, 32-3,
226 1
public employment 11
social structure 32
subsistence agriculture 31
trade unions 56
urbanisation 9-10
Bulgarian Agrasian Union 4, 35-8,91
Bulgarian Communist Party 58
bureaucracies
Latin America 242200
272
Index
bureaucratic authoritarianism
178-9,
180, 189-99,220,221,263206
bureaucratic-milita?' class 214-15
Burks, R. V. 238 16
Burnham, James 214
cabo eleitoral 84
caciques 18-19,23,24
Campbell, J. 234 120
Canavan, Margaret 89, 90, 92,233 97 ,
24428.30
capital goods industries 118, 252 76
capital investment 228 16 17
Balkans 34
direct foreign 116-17
Greece 8, 26625
needfor 179,191,192
capitalism
defined 2246, 2272
reproduction requirements 199
state role 267-840
capitalist development
western Europe 250--2 74
capitalist economies
semi-periphery 228-9 27
capitalist enterprises
defined 252-3 77
capitalist system
analysis 268-950
Cardoso, F. H. 237 141 , 255 180 , 263 211 ,
264 215 ,265 19 ,26620
Castro, Brigadier General 158
Castro,Fidel 165,173
Cavarozzi, M.J. 262 182
Caviedes, Cesar 24741 43
Celman,Juarez 24
Centrallnformation Service 139,142,
143
Centre coalitions (France) 132, 133
Centre coalitions (Greece) 135
Centre Union (Greece) 82-3, 139-40
centre-right coalitions (Greece) 135
cereal growing provinces
Argentina 230
Cerroni, U. 26736
CGT 152,153
Chalmers, D. 2644
Chapultepec Conference 160
charismaticleadership 80, 90, 126
Grove 28-9
Peron 26-7
Yrigoyen 24-5
chefe politico 84
chiflik estates 39, 50
Chile 222,224 2
agriculture labour force 232 78
clientelism 128, 242 199 , 256 105
conscription 13
conservatives 108-11
industrialisation 51, 53,54
lower-class incorporation 128
military dictatorship xvii, 158-70
military intervention 101--4
oligarchic parliamentarism 4, 28-9,
226 1
public employment 11,12

urban populism 28-9
urbanisation 9
Chilean national development
Corporation 168,24853
Christian Democratic Party
(Chile) 162, 164
CIA 145
civil reliability, certificate of 136
civil service
employment 11-12, 22932 37
civil society
relationship with state 195
strong and autonomous 73--4, 131
civil war, Greek 134
civil wars 228 12
class fractions 210
class places 26845
class practices 210--11,268 46
class support
anti-Venezelists 47
Greek Liberal Party 47
clientelism 219, 222, 228n, 243 10 ,
257108
73-83, 129-30,
203
Brazil 83-6
Chile 128 242 199 256 105
definitions ' 244 34 '
Greece 39-48,81-3,112,220,
236 138 ,242200 ' 243 15
Greek Liberal Party and 45-8
in Marxist theory 201-2
theory of 92--4
transition to populism .83-8
Uruguay 235-6 138
CNSL 63
Co-operative Commonwealth
Federation 91
Cockroft, J.D. 263 210
Cohen, Gerry 26622
as political inclusion
Index
Coleman J. 264 1
collective bargaining
restricted 65
Colletti, Lucio 267 36
colorados 156-7, 260 137
Comintern 56, 57, 60, 62,63
commercialisation
agriculture 8, 31-2
communism
Balkans 56-8
Chile 160-1
Communist Party (Brazil) 67
CommunistParty(Chile) 161,167
Communist Party (Greece) 58-9, 601,105,126,135,136,238171
communist regimes
political controls 188
stability 187-8
communist subversion
fear of 148-9
communists
Brazil 66
Chile 63-5
Comminos, Maria 243 15 ,257 108 ,
259 125
compulsory labour service 37
Confederaci6n de Trabajadores de
Chile 63
Confederaci6n Nacional de Sindicatos
Legales 63
Confederat;oa dos Trabalhadores do
Brasil 67
Confederat;oa General dos Trabhaldores
doBrasil 66
Connif, M.L. 83,85--6,231 66.68 ,243 20
Conservative Party (UK) 239--40 192
conservatives
Argentina 106-8
Chile 108-9
Greece 111-12
military dominance and 105-12
Constantine I, King of Greece 46,
235 131
constitutionalists
Chilean officers 169-70
consumer goods
production 252 75
continuistas 260 137
Cook, T. 255 97
copper
Chile 255--6 100
CORFO 248
corporatism 74-5
Cortes, R. 236 141
273
Cotler, J. 26620
Couloumbis, T. 258 117 118
Coulter, P. 264 1
counter-insurgency units
Chile 168
Croatia
peasant populism 38
Cruise O'Brien, D. 264 2
ern 67
Cuba 125
cultural explanations
of political institutions 186
cultural hegemony
Greece 234 116
CUT 167
Cutright, P. 264 1
Cyprus 181
Dahrendorf, R. 258 115
Dakin, D. 233 108
Daniilidis, D. 232 82
debts
Balkan states 233 92
Deldycke, T. 237 154,155,156,157,158
Delis, D. 248 55
democratic regimes
breakdown 26627
Depression 1929 25, 51,53
Dertilis, G. 2293(',234 118 122
Dertouzos, D. 238 172 , 177
development 5-6
comparative studies 227 5
dualist theories 253 81
neo-evolutionist theories 2278
neo-Marxist theories 227-89
di Filippo, A. 254 84
Di Tella 236 140 , 2438
Diamandouros, N. 23289 ,264 211
diaspora
Greek 41-2
Diaz-Alejandro, C.F. 227 6,236 140 ,
237148,149,150' 24850,56' 24963,67
Dicey, E. 33,23290 ,25275 ,25380 ,
25490 26951
Dimitrov, G.M. 233 95 ,243 13
disarticulation 118, 25073 74
Dobb, M. 2246,2272,240 194
Dorfman, Adolfo 236 139
Dotation Law 1835 39
Doxiadis, A. 253 83 ,26740
Drake, PaulW. 231 69 72 73 ,24432 ,
24745
druzhbi 36, 222
274
Index
dualism
development theory 253 81
Dutra, Eurico 67
Duverger, M. 239 193 ,258 116
Economic Commission for Latin
America 237 152 249 72 253 78
25484,260145
'
'
'
economic explanations
of bureaucratic
authoritarianism 195
economic growth
Greece 114,137,24851
Industrial Revolution 251-274
economic ineQualities 121-2,130-1,
171, 260(45
economic penetration
Balkans 34,228 16
economic policies
Aramburugovernment 147
Peron 113--4
economic recession
Chile 163
economically dominant classes
explaining political
phenomena 199--200
economism 193--4
economy
changing structural
requirements 191-4
state intervention 114
EDA 135, 136, 139
egalitarianism
peasant populism and 32, 232 84
elections
oligarchiccontrol 18-19
electoral reform
Argentina 4, 20-1,99,106,209,
260144
Ibanezgovernment 161-2
electorate
Argentina 23047 , 260 144
Chile 262 177
Elefantis, A. 238m
Ellis, H. 253 80
emigration
Greece 42, 138, 253 79
employment
in industry 119--20
Encyclopaedia Britannica 224 1
England
political institutions 131-3
ERE 82-3
Erickson, Kenneth P. 239 186 , 266 20
Ernesto, Pedro 85
Estado Novo 27-8, 66-7
Evelpides, C. 23C7
Evelpidon School 24626
exports
Argentina 248 50
Chile 255-6 100
LatinAmerica 22826
semi-periphery 53
Faction of 1951 146
factory occupations
Per6nist 151, 152
Faletto, E. 237 141 ,255 100 ,265 19
fascist regimes 178
Fatherland and Liberty organisation
(Chile) 170
Federaci6n Obrera de Ia Republica
Argentina 67
Ferdinand I, King of Bulgaria 79
Ferrer, A/do 237 150 ,248 56
feudalism 86
transition to capitalism 227 2
Figuero Alcorta 24
Filias, V. 26847
Finch, M.H.J. 235 138
Finer, S. 26627
Flynn, P. 231 67
FOCH 62-3
FORA 67
Ford,A.G. 228 17
foreign capital
Argentina 24850
direct investment 116-17
Greece 266 25
need for 179, 191, 192
role in Latin America 228 17
France 12, 13
Frank, Gunder xiv, 224-5 56, 2278
FRAP 162
Frei 116,117,162,162-3,166, 167,
168,254-5 90
Frondizi, Arturo 116,148-9,149, 151,
52, 154, 261 156 168
functionalism 217,265 11
functionalist explanations
of bureaucratic
authoritarianism 191-4
ofmodernisation 185-9
Furtado, C. 227 6,228 17 ,236 141 ,237 147
Garcia, Seru 153
Garces, J6an 263 200
Index
Garoufalias 143
GCT 150-1
GDP
industrial sector contribution 54
General Confederation of Greek
Labour 60-1
General Confederation of
Labour 150-1,152,153
General Trades Union Federation 58
George I, King of Greece 259 123
George II, King of Greece 258 118
Germani, G. 23052 ,243 14
Germany 9, 12, 13
Gerschenkron, A. 25276
Ghennimatas, General 143
Giannaros, G. 24962
Goldwert, M. 24589 12 , 260 135 136 ,
261155,158.167
Gonzalez Casanova, P. 263 210
Gonzalez Videla, Gabriel 159-60, 161
gorillas 148, 154-5,260
Goulbourne, H. 2422
government employment 21-2
government expenditure 229
government measures
for industry 53
governors, provincial 18
Graham, R. 228 17
Gramsci, Antonio 267 36
Grand Federaci6n Obrera de Chile 623
Great Encyclopaedia 234 119
Great Idea 79, 243 12
Greco-TurkishWar1897 14,43
Greece 222,224 1
agriculture 31-2,23278
capital inflow 8
clientelism 39-48,81-3, 112,220,
236 138 ,242200 ,243 15
conservatives 111-12
diaspora 41-2
industrialisation 51-2,53--4, 114
landownership 39-40, 207
lower-class incorporation 126
military coups 14, 43, 104, 144,206,
234 122 , 26625
military dictatorship xvii, 134-45,
264212 , 26625 , 268 50 , 26g5
military intervention 104-5
military links with USA 135, 145,
260134
oligarchic parliamentarism 4 42-5,
226 1
275
political structure xiii-xiv, 32-3

urban structure 42
urbanisation 9
US aid 258117
Greek
language of instruction 234 116
Greek Agrarian Party 44
Greek Civil War 134
Greek Communist Party 58-9, 60-1,
238 171
Greek diaspora bourgeoisie 41-2
Greek liturgy 234 116
Greek merchant class 40-2, 234 116
Greek nationalism
role of merchant class 41
Greek Orthodox Church 234 116
Greek Patriarchate 234 116
Greek Rally 135, 136, 142
Greek refugees
from Asia Minor 40, 44,51
Greek War oflndependence 40, 207,
234 116
green international 79
Gregoriadis, S.N. 258 123 ,259 126
Grew, R. 257 113
gross domestic product
industrial sector contribution 54
Grove, Marmaduke 28-9,63,91,102
GSEE 60-1
Guatemala 255
guerillas
Per6nist 151
Guido, Jose Maria 151
hacendados
Chile 108-9
Halperin, E. 262173
Halperin Donghi, Tulio 242 1
Hamilton, Norma 241 199
heavy industry 250
Hennessy, A. 24427
Herzen, Alexander 35
Higgins, B. 251 74
Hilton, R. 227 2
Hintze, Otto 2268
Hirschman,A.D. 176,2244,24962 ,
263 204 , 265 19 , 26620
Hodges, D. C. 260149,150' 261153
Holloway, J. 265 19 ,26730
Hopkins, J. W. 242200
Huntington, S. 185-9,25596 , 2644,
2657,8,9,10,11,13, 26732
Index
276
Ibanez 28,63,64,102,103-4,158,
161,245 17
Iberian Empire xiii, xv, 225 7
IDEA 142-3, 145, 259 127 128
ideological controls 256 102
Iilia, Arturo 151, 152, 153-4, 157, 158,
261 168
immigrants
Argentina 19
Bulgaria 234
Greece 40
import-substitution 53
Chile 236
crisis 175,188
import-substitution
industrialisation 10, 30, 51, 55,
113-14,113-16,175,176,220
income distribution 121-2, 25274 ,2578114
Argentina 17, 25485

Chile 254-5 85 90
Greece 25486
incorporation 73-6
incorporative relations of
domination 205
Independent Trades Union
Federation 58,238 1~
Independent Workers of Yugoslavia
Party 57
industrial bourgeoisie 70
Balkans 31
Chile 25597
western 240-1 194
industrial capital
dependence on state 241 199
semi-periphery 71-2
industrial capitalism
influence on semi-periphery 7-8
western Europe 250-274
industrial cordons 165
industrial employment 119-20
Industrial Revolution 240 19\ 241 194 ,
251-274
industrial sector
contribution to GDP 54
Latin America 253
industrial wage earners 25483
industrialisation 4-5,113-22,171, 191
Argentina 5, 51, 54,236 139
Balkans xiv, 30-1
Brazil 52, 53
Bulgaria 54, 187
Chile 51,53,236 139
Greece 51-2,53-4,114,137,236139
import-substitution 10, 30, 51, 55,

113-14,113-16,175,176,220
in depth and breadth 249-5072
income inequalities during 257-8 114
late-late 2244
restricted and uneven 119-21, 242 1
semi-periphery 51-5,211-12
timing and structure 187
western Europe 241 198
industrialists 24019\241 195
industry 23278 79 , 236 140141 , 24849 ,
25074
foreign capitalinvestment 116-17
inequality
and political radicalisation 25591
economic 121-2,130-1,171,260145
inflation 115-16
Argentina 153-4
Chile 163, 164
infrastructures
for industrialisation 52
inquilinaje 10&--9
institutionalists
Chileanofficers 169-70
lnstituto de Credito Industrial 24853
instrumental autonomy
ofstate 241
integrative relations of
domination 205
interest groups 73-4
autonomous 132
International Labour Office 24970 ,
25483
International Monetary Fund 147
interventionists 154-5, 156-7
investment
foreign capital 116-17
industrial 24849
Ivanchov 35
Jackson, D. G. 238 168
Jessop, B. 26850
Johnson, D.L. 25597
Johnson, J.J. 23()63, 244 1
Johnson, T. 26731
Junta Militar 102
justicialism 101
Justo, General 100
Kadar, B. 24962
Kanellopoulos 141
Kapetanyiannis, V. 258122 , 259 130 132 ,
260134 263205
Karageorgds, D. 25485
Index
Karamanlis, Constantine 116,259 133
Karavidas, K.D. 23297
Karayannis, G. 259 127
Kartakis, E. 25383
Katiphoris, G. 256 103 , 259 129
Katsanevas, T. 238 174,258 120,259 124
Kaufman, R. R. 263203 , 26620
Kay, Cristobal 24742 ,256 105 , 257u18 ,
262176
Kennedy, President J.P. 155,162
Khalaf, S. 2435
Kirsch, H. W. 236 141
Kitsikis, D. 234 125 ,258 123
Koliopoulos, K. 24629
Koumoundouros 259
Kourvertaris, G.E. 263
Kremmidas, V. 228 14
Kuhl, J.M. 255 98
Kunin, Petro 233 99
Kurth, J.R. 240 194 ,241 196
Kuznets, S. 25274 , 25484 89 , 257 114
LabourCode1924 63,64-5,238 179
Labour Confederation, Chile 160-1
labour exploitation
service-tenancy system 108-9
labour force
industrial sector 54
Yugoslavia 237 155
labcur legislation 60
Argentina 68-9
Brazil 65-6
Labour Party (Argentina) 80
Labour Party (UK) 239-40 193
Labour Party of Bulgaria 58
labour service
compulsory 37
Lac/au, Ernesto 88,89,90,2246,
243 11 , 24429 26731
Lambert, D.C. 24736 , 24849 58 ,24971,
253 80
land concentration
land reform
Greece 40, 233 107
land tenure
Balkans 231 76
land values
Argentina 23W4
landowners 241 195
Argentina 11)-18,107
Chile 108-11,158-9
Uruguay 235 138
Yrigoyen's policy towards 21
277
landownership
Albania 231 76
Argentina 16-18
Balkans 29--30
Greece 207
Ottoman empire 225-67
Rumania 231
Yugoslavia 231 76
Landsberger, H.A. 25593
Langel, W.L. 240194
Lanzendorfer, M. 242 2
Latin America
foreigncapital 288 17
Iberian rule xiii, 2257, 2267
industrial growth rates 248511
middle classes 211-13
parliamentary institutions xiii-xiv
trade unions 70
Lavrov 35
Law on Professional Associations 689
leadership style
Grove 28-9
Peron 26-7
Yrigoyen 24-5
League ofNations 54,22935 ,23278
legalists 155,156-7,260137
Legg, K. 82,83,228 11 , 234 125 ,2422m,
243 18
Lenin, V.I. 25377
Leon, G.B. 238 175
Leon, P. 251 74
Leys, Colin 24]2
LiCausi 26734
Liberal Alliance (Brazil) 27
liberal ideas
Industrial Revolution and 240 194
LiberalParty(Greece) 4,43-5,47,
81-3,111,236138 ,258 123
LiberatingRevolution 100,149,150
Lieuwen, E. 244 1
Liga Militar 101
light industry 25074
Lindhol, G. 235 138
Linea Recta, La 104
Linz, Juan 178, 263 207 208 209 , 26627
Lipset, S.M. 91,23284 ,24433 ,264 1
Little, Walter 231 65
Little junta 144
Llewellyn-Smith 234127
Lockwood, D. 216,257110 ,2655,26952
Lonardi, General 146
Loveman, B. 159,22g32,231 70 ,
261169,172
278
Index
Lowenthal, A.F. 244

lower classes
benefits from economic
growth 251-2 74
exclusion from politics 3
Loza, General 156
McDonald, R.H. 255 99 ,262 177
MacVeagh 258
Maira, Luis 262 188
Malios, M. 25490
Malloy,]. 2424
Mamalakis, M.J. 236 141 ,237 151 ,248 57 ,
249 67 ' 253 80
managerial revolution 214
Marshall, T.H. 255, 92 ,257 112
Martin, J.M. 24738 , 24849 58 , 249 71 ,
253 80
Marx, Karl 204,252-3 77
Marxism
action-system synthesis 216-17
attitudes of Chilean military 262 188
development and 5-6, 227-8
reductionism in 199-206,217,26731
Mavrogordatos, George 235 130 132 133,
243 16 17 , 24748 257 108 ' 26841
May, A.J. 232 91
Megali Idea 79,243 12
Menendez, General 101
merchant class
Greek 4(}..2, 234 116
Mercier- Vega, Louis 26849
Merkx, Gilbert, 24856
Merton, R.K. 192, 26622 23
Metaxas, John 61, 105, 246 29
methodological individualism 26627
Mexico
party system 263 210
Micaud, C. 258 116
Michailovsky 35
Michels, R. 205,267 39
middle classes
Argentina 21-2
Balkans 31
Greece 42, 43
Industrial Revolution 240
Latin America 211-13
semi-periphery 124-6
see also bourgeoisie
Migdal, J.S. 235 137
Miliband, R. 265 17
Milicia Republicana 245 17
militarism
Bulgarian 79
military
class origins 19(}..1, 259 133
in civilian institutions 168-9
links with USA 135, 145, 155-6,
157,160,168,173-5,260 134 ,
261161.162.164,262194
new professionalism 155-6, 157,
174,196,197
power position 172-3,177,197
professionalisation 98, 101, 104,
190
transition to post-oligarchic
politics 99-105
military academies 98, 190
Greece, 104,24626 ,259 133
military conscription 13-14, 246 25
military coups
Argentina 25-6, 68, 99, 23063
Bulgaria 38
Chile 166
Greece 14,43, 104,144,206,234 122 ,
26625
military dictatorships xvii, 219,221
Argentina 145-58
Brazil 157
Chile 158-70, 264 213
Greece 134-45, 264 213 214 , 26850 ,
26950
institutional structures 178-9
long-term 171,172,177,178-9,182
Uruguay 235 138
military dominance
conservatism and 105-12
military expenditure
Argentina 258l2l
Chile 258l2l
Greece 258 121
military interventions 182, 266 27 28
Argentina 99-101, 245 10 u. 261 168
Brazil,27
Chile 28, 101-4,166-8,245 17
Greece 104-5, 26625 28
Military League 38
minifundia
Chile 110
MIR 166
Mitchell, B.R. 228 18 ,22937 ,23279 ,
237 160
Mitrany, D. 31,231 77 ,232 83 ,233 95
modernisation 5-6, 76
theories of 184-9
Venizelist 79
monarchical dictatorships
Bulgaria 38,233 103
Index
Greece 105
Yugoslavia 38, 233 103
monarchy
Bulgaria 38, 233 102
Greece 44,46, 105,137,173,235 131 ,
258-9
Yugoslavia 38
Monetary Commission
Greece 24852
Montero, General Toranzo 154-5
Moore, B. 235 135
Moore, R.M. 265 6
Moore, W.E. 231 76 77 ,2339 \238 163
Morris,J.O. 256 107
Morse, R. M. 225 7, 22825
Mouzelis,N. 2277,228 15 ,234 126 ,
235 129 ,242 199 ,2437,244 31 ,
257109,110,26512,18,26625,28,
26733 35 , 26842 , 26953
Movimento de Izquierda
Revo/ucionaria 166
MRP 152
Muck, R. 2422
Mulhall, M.G. 22937
Nanclares, Carvalan 153
National Accounts of Greece 194870 24851
National Agricultural Society 261 170
NationalBankofGreece 47
National Command
Peronistgroups 151
national currencies
conversion to dollars 22~
national lands
Greece 39,233 107
national party
BANUas 36
NationalParty(Chile) 164,170
national unity
Latin American countries 228 12
nationalisation
Allende government 163
nationalism
Balkan 234 117
Bulgarian 79
Nationalist Party (Argentina) 245 10
Nef,Jorge 101, 102,245 1\24621 23 ,
262189
Nefeloudis, B. 24968
neo-evolutionist theories
of development 5,185,2278
neo-Marxist development theories 5,
22~
279
Nett/, P. 2274
Nicaragua 255 99
nitrates
Chile 14,255--6100
North, Lisa 246 19
North Korea 186
Nun, A.J. 190-1,2442,265 15 16
Nunez, Carlos 24620
Nunn, F.M. 172,231 71 , 245 13 17 18 ,
262188,193, 263198,201
0' Brien, P. 2289
Oddone, Jacinto 23045
0' Donne/, G.A. 191-4, 2242, 22821 ,
260145 ,261 164 ,263206 , 265 19 20 ,
26621
OECD 248
Old Republic
Brazil 2261
oligarchic parliamentarism xvi-xvii, 3,
209-10,219,226-7 1
Argentina 16-19,226\230,245
Chile 28-9,2261
defined 3
Greece 4,42-5,226 1
Oliveira Campos, Roberto de 2422{)(1
Ongania, General Juan Carlos 157,
158
Organisation of American States 147
Organization for Economic Cooperation and Develofment 24855
Ortega, Luis 236 13
Ortiz, President 245 10
Ottoman Empire xiii, xv, 30, 40, 22567
Ozslak, Oscar 22826
Pacification Bill 135

PADF 85
Paige, J.M. 235 136
pa/eokommatikoi 79, 111
pa/eokommatismos 43
Palma, Gabriel 236 141
pan-Hellenism 79,243 12
Pangalos 60,61
Panizza, Francisco 235 138
Panourgias, General 259 133
Papacosmas, V. 234 122
Papadopoulos, George 142,143,144
Papagos, Field Marshal
Alexander 135,137,142, 146
Papandreou,Andreas 92,143,145,
259 131
280
Index
Papandreou, George 139--40,141,

142,143,144,145,173
para-military organisations
Chile 169
parliamentary institutions xii-xiv, 6-7,
181,182-3,225-67 ,228 11
parliamentary semi-periphery
defined xiv-xv
Parrish, C.J. 256 106
Parsonian functionalism 217,265 11
modernisation theory 185-9
Parsons, T. 265 11
Partido Autonomista Nacional 20
Partido Laborista 80
Partido Socialista de Chile 161
Partido Socialista
lnternacionalista 239 190
Partido Socialista Obrero 62
Partido Unico 80, 82,91
partisan intelligentsia 233 96
party managers 85-6
party system
Chile 256 105
Pasic, N. 23285
paternalism
Yrigoyenism 22-4
patrimonialism xv-xvi, 86-7, 225-67
patron-client relationship 24435
patronage 45
business 26740
changed nature 76-7
Radical Party 22-4
Uruguay 235 138
patrons
autonomy 93
Paukert, F. 25486
PDC 164
peasant conditions
Balkans 33--5
Peasant Party (Croatia) 38
peasant populism 29-38,49,57,79,
220, 222, 23284
peasantry 231 75
Petmesidou- Tsouloudi, M. 26740
Peppe, P. V. 26620
Pericles Plan 259
Per6n, Eva 101
Per6n,Juan 25,26,68-9,80-1,82,87,
91,100-1,113--14,127,146,147,
148,149,150,153,158,175,23065 ,
248 50 , 260 139 , 261 156
Per6nism 80-1,100-1,171
Per6nist Revolutionary
Movement 152
Per6nists 146, 149,150,152,157-8,

181, 260 137 146 , 261 168
Peni 54, 126,238161
Peselz, Branko 2339 5,243 13
Peterson, H. F. 260 139
Petras, J. 255 97 ,256 106 ,257 108 ,262 174
Petropoulos,]. 23289 ,233 105
Picciotto, S. 265 19 , 2673!J
Pinochet 161
Pinto, A. 25484
Plan de Lucha 152
Plastiras, General 135, 146
plebiscitarian leadership 79,90-1
pluralism
social and political 75
Poggi, General 149
political analysis
conceptual tools for 201
political controls
communist regimes 188
incorporative/exclusionist 187
weak institutionalisation 185
political corruption 228 11
political development xvii-xix
European 131-3
political domination 204-18
political inclusion 243 10
clientelism and populism as 76-83, 203
integrative 129
vertical incorporative 126-9
political incorporation 73-6, 219-20
political participation and 129-131
political modernisation
theories of 184-9
political participation 182,185,219,
209-10,23052 ,231 71
exclusionist/incorporative
controls 129-33,170-1
political parties 200
Chile 256 105
exclusion 126
role in political system 255 99
political patronage
business 26740
political sociology 217
political sphere
relative autonomy 200-3
political systems
analysis of 217
one-party 263210
political theory, Marxist 217
reductionismin 199-206
politicians
Bulgaria 233 96
Index
Pollack, B. 256w 5
Popular Action Front (Chile) 162
Popular Front (Chile) 103, 248 53
Popular Party (Greece) 111
Popular Unity Alliance (Chile) 163--6,
173
populism xvi, 175,219,243910 11
Argentina 8~ 1
as political inclusion 77-8, 203
Brazil 83--6
definitions 88--9, 244 28
ideology 87
incorporation 126
Latin America 23 C4
political inclusion 78--83
Russia 233 97
theories of 88--94
transition to 83--3
see also urban populism, peasant
populism
Populist Party (Greece) 44
Portantiero,J. C. 265 1\267 33
Potash, R. 245 10 ,261 159 ,263 202
Potter, Anne 23062
Poulantzas, Nicos 21~11,265 17 ,
26845.46.50
Powell, John Duncan 252 5, 244 34
power position
military 172-3,177,197
Prats, General Carlos 263 199
Pratt, G. W. 261 166
Prebisch, Raul 147
Prestes, Luis Carlos 66, 67
PRI 263 210
price control
Allende government 163
agricultural products 246 19
primary school registration 12-13
production, relations of
and political domination 206-18
professionalisation
military 98,101,104
Prometheus Plan 259 126
provincial governors
Argentina 18
Prussia 132
Psilos, D. 24960
PSO 62
PUMA 103
quedantistas 260
Radical Intransigents (Argentina)
148
281
RadicalParty(Arentina) 4,19-25,
99,209,245 1
Radical Populists (Argentina) 152,
153
Radical Union (Greece) 82-3
radicalisation
dominated groups 171
inequality and 255 91
lower classes l23--4
Radich brothers 38
Radoslavov 35
railways 232,237
Ranis P 261 154
Ranki,
232 78 ,237 142 146 ,238 162
Rankin, R. C. 26626
Ratcliff, R.A. 241 195
Recabarren, Luis Emilio 62
Red International of Labour
Unions 62
Red Peasant International 57
reductionism
in Marxist political theory 199-206,
267 31
refugees
from Asia Minor 44, 51, 105,233 110
reproduction
capitalism 199
revolution
potential for 124-6
Reyna, Jose Luis 26620
Reynolds, C. W. 248 57 ,24967
Riggs, F. W. 264 3
Rio de Janeiro treaty 1947 155,160
Rippy, J.F. 228 17
Rivadavia, Bernardino 17
Roca, Julio Anx,entino 16, 20, 24
Rock, D. 230 3.50.57,59.63, 260 146
Rosas, Juan Manuel de 16, 17, 24
Ross Santamaria, Gustavo 103
Rostow, W. W. 2278
Rothschild, J. 238164.165,169,170
Roucek, J. 234ll9
Rouquie, Alain 150, 23042 51 , 245 67 11 ,
24735 258l2l 260141,142,144
26116i.162,163\65
,
Roxborouf!h, I. 262180.186,187,192,
263 19'7
Royal Institute of International
Affairs 233 92
royalists
Greece 44
Rumania 224,231 76
rural co-operatives
Chile 162
G.
282
Index
rural emigration
Greece 138
rural immigrants
Argentina 26
Brazil 28
rural unionisation 162, 171
Chile 161-2
prevention 158-9
Saenz Pefia 4, 20, 99, 106,209,260 144
Sagatti, M. 225 7
Sarobe, Jose Maria 245 7
Saul,John 242 2
Schmitter, P. C. 242 3, 261 160 , 265 19
Schneider, General Rene 167
Schneider, P. 228 10
school registration 12-13
Schweinitz, K. 25489
Scobie, James 23044 , 247 32 33
Scocpol, Theda 255 94
Scott, James G. 244 34
Scott, Robert E. 242 2"'
Second International 56, 60
Security Battalions 259 128
Serra,Jose 266 21
service sector
inflated 115
service-tenancy system
labourexploitation 108-9
Seton-Watson, Hugh 241 196
Sherrard, P. 234 120
Sideris, A. 237 143
Smith, Peter 23053 , 26844
Smokovitis, D. 260 133
Snow, Peter G. 24734 ,260 137
social pluralism 75
bureaucratic authoritarian
regimes 178-9
Socialist Labour Party of Greece 60
Socialist Party (Argentina) 22
Socialist Party (Chile) 28-9,91,
239 190 ' 256 105
Socialist Workers' Party of
Yugoslavia 56-7
Sociedad Rural de Argentina 21,209
societal corporatism 75
Solberg, C. 230545
So/tow, L. 258 114
spahis 226
Spalding, H. 239 186 192
Spulber, N. 238 163
Stallings, Barbara 248, 59 , 2559(1, 262 181
Stamboliiski, Alexander 4, 35-8,57,
58,79,91,222,233 102
standard of living
post-Industrial Revolution 25274
state
as producer 24854
authoritarianism 74, 132-3, 183
autonomy 213-14,241-2 199
capitalism and 267-8411
centralised 18, 27
expansion of 10-13
overdeveloped 242 2
relationship with civil society 195
state bourgeoisie 26841
tzakia families as 207-8
state bureaucracy 115,23396
employment 229 37
state corporatism 75, 76
state employment
growth 11-12
state institutions
development 228 26
state intervention
in economy 10,114
state patronage
Uruguay 235 138
Statistical Yearbook of Greece 249 71 ,
253 79
Stavenhagen, R. 235 81
Stavrianos, L. 11,228 16 ,22933 ,23291 ,
233100.103' 234116
Stearns, P.N. 250 74
Stepan, Alfred 2445, 261 160
Stephanidis, D. 233 111
Stephanopoulas 143
Steward, A. 244 26
Stoianovich, T 232 86 , 234 113 115 "117
strikes
Argentina 150-1
Greece 24628
illegal 66
Per6nist 151, 152
rural 162
suppression 166-7
structural autonomy
of state 241 199
structuralism, Althusserian 217
susbsistence agriculture
Bulgaria 31
sultanic despotism 225-67
sultanism 225 7, 263 207
Swift, J. 262 178
Tapia Vide/a, Jorge 262 179
Taylor, John 26733
TEA 137
Index
tenant farmers
Argentina 246-7 32 33
tenente movement 84-5
textile manufacturers
Industrial Revolution 240 104
Therborn, G. 25074
Thiesenhusen, W. 262 178
Thompson, K. 248 55
timar landholding system 2267
Toilers' Bloc (Bulgaria) 58
Tomasevich, J. 233 94 ,238 163
Topik, Steven 229 28
trade unions 123
Argentina 67-9, 80
Balkans 70
Bulgaria 56, 238 165 173
Chile 62-5, 161, 162, 262 179 180
control 60, 63,64-5,66,68-9, 126
corporatist regulation 75
Greece 59-61, 238 173
Latin America 70
militancy 150-1
Peronist 127,239 192 , 260 147 148
relationship with communist
movement 57-8
rural 171
state control 136
UK 239
TreatyofNeuilly 37,38
Trikoupis, Harilaos 78
Trimberger, Ellen K. 255 95 , 256""
Truman Doctrine 135,160
Tsouderos, E.J. 234 121
Tsoukalas, C. 22931 ,234 118,26841
Turkey 126
tzakiafami!ies 40,45,50, 79,207-8
UCRI 148
underemployment
Balkan countryside 238 163
Union Civica 19
Union Civica Anti-personalista 25
Union Civica Radical 19
Union Popular 152, 261 151
Unitary Confederation of Labour 61
United Democratic Left
(Greece) 135,136,139
United Kingdom 9, 12
UP 163--6, 168, 169
urban middle classes
Balkans 31
urban population
expanding 235 134
283
urban populism xvi, 49, 220, 222

Argentina 16-27
Brazil 27-8
Chile 28-9
favourable conditions for 235 134
see also peasant populism, populism
urbanisation 8, 9--10, 10-11, 171
Argentina 34971
Balkans 30-1
Chile 24971
Greece 42,234 112 ,24971
Uriburu, General 100
Uruguay
political system 235-6 138
USA
Bulgarian immigrants 234 119
influence in Greece 135, 258 117 118 ,
260134
military aid and influence 147-8,
155-6,168,173--5,260 134 ,262 104
policy towards Chile 160-1, 164
USARSA 261
Vai"tsos, C. 24854
Valenzuela, A. 24745 ,256 105 ,264 213 ,
26843
VanNiekerk, A.E. 244 27
Vandor,Augusto 153,261 156
Vanger, M.I. 236 138
Vargas, Getulio 27-8,65,66, 67, 87,
244 4
Varkiza Agreement 259 128
Veliz, Claudio 71,225 7,226 7,241 197
Venizelists 247 47
Venezuela 125
Venizelos, Eleutherios 4, 40,43--5,59,
78-9,81-3,104-5,111,22 2,235 131 ,
236 138 ,2444,258 123 ' 259 123
Verba, Sydney 227 3
Veremis, T. 234 123 , 244 1, 24626 27
Vergopoulos, C. 233 104 106 ,237 153
Vernicos, N. 233 107 ,248 51 ,25490
Viaux, General Roberto 167
Vietnam 186
villages
Bulgaria 32
Volpe, Galvano della 267 36
voting population
Argentina 260 144
Chile 262 177
voting turnout
Argentina 24631
284
Index
wages policy
Allende government 163, 164
Wallerstein, Emmanuel xiv, 224-5 6
Walter, R.J. 23047 56 , 239 191 ,246 31
War oflndependence, Greek 40, 207,
234117
Ward, Michael Don 25484
warsofindependence 242 1
Weber,Max 86,217,225 7,2268 ,24435 ,
263207' 26627
Weinert, Richard 266 20
Weingrod, A. 2436
Weinstein, M. 235 137
welfare policies
Bismarck 132
western capitalism
in Balkans 34
westernisation 5-6
Wiarda, H.J. 225 7
Wolf, E. 23284 ,255 95
Wolpin, M.D. 261 171 , 262 195 196
working-class movements 69-70,
25274
Argentina 67-9
Bulgaria 55-8
Chile 29, 63-7
Greece 58-62
Western Europe 239-40 193
WorldBank 147,24969
world economic system 225 6
World Warl 37,51
Wright Thomas C. 246 19
Wynia, GaryW. 260 138 ,261 152
Yrigoyen,Hipolito 4, 19-26,67,68,
99-100,209,222,2444
Yugoslavia 54,224 1,231 76 ,237 155
communist movement 56--8
peasant populism 38
zadruga family system 31,23281

Zeitlin, M. 241 195
Ziemann, W. 242 2
Zuvekas, C. 254 88
Zymelman 236 140 ,243 8

(New Studies in Sociology) Nicos P. Mouzelis (Auth.) - Politics in The Semi-Periphery - Early Parliamentarism and Late Industrialization in The Balkans and Latin America-Macmillan Education UK (1986)

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

(New Studies in Sociology) Nicos P. Mouzelis (Auth.) - Politics in The Semi-Periphery - Early Parliamentarism and Late Industrialization in The Balkans and Latin America-Macmillan Education UK (1986)

Uploaded by

Copyright:

Available Formats

Politics in the Semi-Periphery

Nicos P. Mouzelis 1986

Softcover reprint of the hardcover 1st edition 1986 978-0-333-34933-5

All rights reserved. For information, write:

To the memory of my father

Modes of Transition to Post-oligarchic Politics:

Theoretical Implications (1):

2.1 On the concept of incorporation

Routes to Military Dictatorship:

Theoretical Implications (2):

I would like to thank the foilowing persons who, at one time or

While it may be conceded that there are systematic differences

any great theoretical pretensions, nor is it the product of a

laristic/personalistic features of parliamentary politics; neither did

illustrate some more general trends. For instance, in discussing (in

As already mentioned in the General Introduction, a distinctive

Politics in the Semi-Periphery

restrictive form of parliamentary government began to weaken

Modes of Transition to Post-oligarchic Politics

transition from political 'clubs' to parties occurred before these

Politics in the Semi-Periphery

Modes of Transition to Post-oligarchic Politics

Argentina, Chile, and to a lesser extent Brazil, the relatively stable

A. It was not, of course, merely the adoption of parliamentary

Politics in the Semi-Periphery

monetisation of at least certain sectors of the economy, the

Modes of Transition to Post-oligarchic Politics

pean societies. I shall try to illustrate this fundamental point by

Politics in the Semi-Periphery

This becomes obvious if Bulgaria is compared with a European

Modes of Transition to Post-oligarchic Politics

minimum of municipal facilities (transport, sanitation, and so on),

Politics in the Semi-Periphery

of the labour force employed in the public administrations of the

Modes of Transition to Post-oligarchic Politics

8.88 per cent, and in Bulgaria 13.47 per cent. 39

Politics in the Semi-Periphery

Modes of Transition to Post-oligarchic Politics

bourgeoisie. Whereas in western Europe the demise of oligarchic

1.2.1 Urban populism

Politics in the Semi-Periphery

Modes of Transition to Post-oligarchic Politics

As early as 1826 the Bernardino Rivadavia government in

Politics in the Semi-Periphery

Modes of Transition to Post-oligarchic Politics

approval it was impossible to win any elections. The system

Politics in the Semi-Periphery

Modes of Transition to Post-oligarchic Politics

immigrants, it did have a dramatic impact on Argentina's political

Politics in the Semi-Periphery

Modes of Transition to Post-oligarchic Politics

tees, or in the case of the city of Buenos Aires the committee of

Politics in the Semi-Periphery

educated, professional people possessing very different skills from

Modes of Transition to Post-oligarchic Politics

Politics in the Semi-Periphery

Modes of Transition to Post-oligarchic Politics

leadership to an unprecedented level, he gave the coup de

Politics in the Semi-Periphery

attempted to continue to rule in a parliamentary context by

As an anti-status quo party, the Chilean Socialists broke into

Modes of Transition to Post-oligarchic Politics

behind a charismatic caudillo from the armed forces ... While