Analyzing The Third World Essays From Comparative Politics (Norman W. Provizer)
Analyzing The Third World Essays From Comparative Politics (Norman W. Provizer)
Analyzing The Third World Essays From Comparative Politics (Norman W. Provizer)
All rights reserved. No part of this book may be reprinted or reproduced or utilised in any form or by
any electronic, mechanical, or other means, now known or hereafter invented, including
photocopying and recording, or in any information storage or retrieval system, without permission in
writing from the publishers.
Notice:
Product or corporate names may be trademarks or registered trademarks, and are used only for
identification and explanation without intent to infringe.
APPENDIX
CONTRIBUTORS
The list of persons who, in some way, contributed to this volume would
rival the technical credits for Star Wars (the movie).
A special thanks, however, must be extended to: Patricia Hughes White
and Margaret Bayldon, who, as Managing Editors of Comparative Politics,
participated in this project from its very inception; Helen Kryka, Juanel Vo-
taw, and Glenzetta Walker, who typed order out of chaos; Alfred
Schenkman, who kept the faith; and to the twenty co-authors of this book,
who made it all possible.
Needless to say, all errors of fact and vision are solely the responsiblity
of the editor.
Norman W. Provizer, a native of Chelsea, Massachusetts, is currently an
Assistant Professor of Political Science at the Louisiana State University,
Shreveport. He received his A.B. from Lafayette College and his Ph.D.
from the University of Pennsylvania; he has held research grants from the
Anspach Institute of International Affairs and Diplomacy at the University
of Pennsylvania, from the National Endowment for the Humanities, and
from the LSU Foundation. His articles have appeared in a number of
journals, including Comparative Politics and the Journal of African Studies.
*The conclusion of each contributor sketch gives the reference to the issue of Comparative Politics
in which the author’s article originally appeared.
In recent years, the phrase “the Third World” has become part and parcel of
common parlance in the United States. Fueled by the politics of energy and
the growing concern over a number of scarce resources, interest in the Third
World has expanded beyond the walls of academia into the arena of
everyday life.
Whether or not the dire prophecies of an earth depleted, overpopulated,
and saturated with pollutants are accurate, it takes but little imagination to
see that the future will, in no small way, be shaped by events occurring in le
Tiers Monde. In a world ever contracting, les damnés de la terre (the
wretched of the earth1) will not be without a voice; the global “silent
majority,” despite protestations to the contrary, will and should be heard.
Yet the usage of a phrase, even if it be ubiquitous, does not preclude the
existence of multiple meanings. The fact that words are expressed does not
necessarily signal that there is a single, common definition accepted by all.
So it is with “the Third World.”
Emerging with the era of decolonization and the rise in East-West
tensions that followed the conclusion of World War II, the Third World
represented those nations, new and old alike, which rested between the First
World of the industrial West and the Second World of the Communist
“bloc.” In this sense, Third World countries were distinguishable from the
states most intimately involved in the system of East-West conflict. From
the perspective of bipolarity, Third World nations were the uncommitted of
the world’s regions.
The view of the Third World as the nonaligned retains a certain validity
in the current context of global affairs, but time and the flow of events have
directed us to a somewhat different perspective on the nature of the Third
World. Given that one of the earliest and most symbolically important of
Third World meetings, the 1955 Bandung Conference, “was a gathering not
of uncommitted but of Afro-Asian states regionally defined and including
Asian members of SEATO as well as the Communist states of China and
North Vietnam,”2 such a changing perspective is neither surprising nor
lacking in past roots.
As the number of Third World states dramatically increased, the primary
locus of concern shifted from a fear of being destroyed in the battle between
“two elephants”3 to anxiety over a hydra-like “developed” world standing
in the way of a redress of global imbalances. Cutting across the horizontal
line of East-West conflict, the vertical division of the world into a
“developed” North and an “underdeveloped” South has grown in
importance.
This almost Manichaean vision of the global order was based on the
perception of several interrelated dualities which dominate the hearts as
well as the minds of both leaders and led within the Third World.4 From
this perspective, the earth is a place of rich and poor; of masters and
servants; of colonizers and colonized; of strong and weak; of urban and
rural; and of white and non-white. We thus moved toward the congruence
of the idea of the Third World with that class of countries labeled less
developed, and states such as Yugoslavia no longer remained at the center
of gravity.
“We are the Underdeveloped,” sings the Third World.5 Our past, present,
and future separate us from them and provide the commonalities which bind
us together despite our diversity. As such, the Third World connotes “a self-
defined and self-conscious association of nation-states,”6 whose
membership is closely, but not exclusively, linked to a range of empirical
socio-economic referents (of which per capita Gross National Product is a
prime example), and whose members have and continue to struggle for
liberation from “colonial” rule and for development and modernity.7
Needless to say, caution must be exercised in the use of the terms
development and modernity in order to avoid the excesses of the unilinear
evolutionary syndrome. To state that the Third World countries share a
common passion to develop and to enter the “modern” world does not mean
that they must or will be carbon copies of what has come before; nor does it
necessarily posit a mythology of stasis and final end-states in the
organization of human affairs. This passion to develop, to modernize, is not
without its ambiguities and contradictions; the question of the appropriate
level of technology and the issue of perpetual underdevelopment as
satellites, if Third World nations are incorporated into the global “capitalist”
system, are indeed very real.8 Nonetheless, the existence of such debated
problem areas does not extirpate the desire of the Third World to improve
its political and material position vis à vis the “developed” nations.
Specifics may be in doubt, but the generality remains valid. And whether
it be for good or for ill, the countries of the Third World thus see themselves
not only as the “underdeveloped,” but also as the “developing.” “The winds
of change” are blowing; while their exact direction is not always known,
their reality is certain.
Importantly, this view of the Third World from a global perspective is
intimately interwoven with the dynamics of intrasystem politics; to speak of
development in splendid isolation is not enough. The Third World’s shared
search for a future, at the least, implies a degree of internal transformation
and multidimensional discontinuities which is related to the theme of
nationalism and the vexatious processes of state and nation-building.9 In
short, the Third World’s vision of its future is inextricably linked to the
forging of nations and the establishment of viable state frameworks. It is on
these most difficult of tasks that this volume focuses.
Our approach is to present essays from Comparative Politics which
analyze the Third World in terms of commonally shared intra-state
dynamics. The first section of the volume, “Comparative Politics,
Modernity, and Change,” provides a review of and a basic foundation in the
concepts most widely utilized in the study of countries associated with le
Tiers Monde. The second section, “Politics and the Social Order,” examines
the interrelationships of individuals, society, and politics from what is,
primarily, a behavioral perspective. The middle section, “The Organization
of Politics,” deals more directly with the structure of Third World political
systems, and it is followed by a section on “Politics and the Men on
Horseback” which considers the critical role played by the military in these
systems. The final section, “Leadership and Public Policy,” looks at those
who hold “the royal spear” in the states of the Third World and probes the
policies sounded by their “drums of rule.”10
Throughout the volume, attention is drawn to the internal attitudinal,
behavioral, and organizational problems and transitions which characterize
countries of the Third World. Given the multi-dimensionality of the
discontinuities of change (i.e., the psycho-cultural dimension, the socio-
cultural dimension, the ecological-economic dimension, and the political-
ideological dimension) and the plethora of components existent within each
dimension, numerous conceptual, as well as methodological difficulties are
inevitable. Such difficulties, while noted in the appropriate sections of the
volume, will not be resolved with any false sense of finality in the pages of
this collection. Analyzing the Third World is too involved and complex for
self-righteous and often self-serving answers in black and white.
Simplifications are valid within their sphere and are useful as tools or
means of understanding, but they are not in themselves ends.
With this warning about oversimplification firmly implanted, we turn to a
consideration of three broad issue areas that are related to the preceding
discussion. The first of these is related to the idea that while the focus of
this volume is on the Third World, the world outside is not ignored. It is
true that many of our perceptions of the internal dynamics of Third World
states have been warped by a Western-centric form of determinism, yet
significant distortion also develops when we embark on a “neo-parochial
concentration”11 on the distinctiveness of these states. Joseph Gusfield, for
example, has pointed out that:
[T]o be ‘modern’ appears in many new nations as an aspiration toward which certain groups seek
to move society. ‘Modern’ becomes a perceived state of things functioning as a criterion against
which to judge specific actions and a program of actions to guide policy.12
Thus there is within the states of the Third World a kind of “teleological
insight” at work, through which the future of “developing” states is seen in
part from the perspective of the experiential patterns of the “developed”
states.13 The past may well be a questionable guide for the future, but it is
not without substantial import when the vision of the future is rooted in the
images of the past.
Clearly, this is not a call for the return to crude determinism; the use of
“teleological insight” is one of *’... choice among possibilities, not a fixed
and self-evident set of propositions.”14 Rather, it is merely a recognition
that the selection of forms implies a range of limits and imperatives. It is
this recognition which provides the comparative rationale for the inclusion
of a number of essays in this volume, including those by Sherrill, Kautsky,
and McKinlay and Cohan.
The second issue area that draws our attention centers on the
convergence of national and international systems, on what has been termed
“linkage politics.” The Third World, as already noted, is defined both by the
factors of global hierarchy and conditions and problems of internal order.
Intrasystem dynamics provide the focal point for this volume, but it is
obvious that no neat boundary exists between the internal and external
dimensions of politics. Third World states are affected by and react to
numerous sequences of behavior that originate in the “developed” world.15
Irving Louis Horowitz, following the path of Andre Gunder Frank, has,
in this vein, argued that:
Too rarely is it appreciated that underdeveloped conditions may be the consequence of the
developed sectors. Underdeveloped, in contrast to simply undeveloped, economies are in the
condition they are in large measure because of the intervention and interference of fully developed
nations.16
With this thought in mind, we turn to our analysis of the Third World, the
states of Africa, Asia, the Middle East, and Latin America which rank’ last
in a world of win, place and show.24
NOTES
1. The Third World is from the French le Tiers Monde. Irving Louis Horowitz, in Three Worlds of
Development (New York: Oxford University Press, 1966), p. 5, notes that Frantz Fanon’s “book on
Algeria [Les Damnes de la terre] is probably the first to use [le Tiers Monde] as a colloquial
expression for the newly emergent nations.”
2. Laurence Martin, “Introduction: The Emergence of the New States,” in Martin (ed.), Neutralism
and Nonalignment (New York: Frederick Praeger, 1963), p. xiv. The Bandung Conference was
attended by some 29 nations, at the Colombo Conference held in August 1976, some 85 states were
represented; for a brief overview see William Borders, “Nonaligned Nations Have Come a Long Way
Since Bandung,” The New York Times, August 22, 1976, p. E-3.
3. From the African proverb: “When two elephants fight, it is the grass that suffers.” See Cecil
Crabb, Jr., The Elephants and the Grass: A Study of Nonalignment (New York: Frederick Praeger,
1965).
4. For a different viewpoint, see Kwame Nkrumah, “The Myth of the Third World’ “ in Nkrumah,
Revolutionary Path (New York: International Publishers, 1973), pp. 435-438. There Nkrumah argues
that: “Today then, the Third World’ is neither a practical political concept nor a reality” (p. 438).
5. “A few years ago a new song was heard during the carnival at Rio de Janeiro. Its refrain began
with the words, ‘We are the Underdeveloped!’ It was a very cheerful song, with a touch of irony,
perhaps of definance.” From Peter Berger, Pyramids of Sacrifice (Garden City: Doubleday, 1976), p.
7.
6. Horowitz, Three Worlds of Development, Second Edition. (New York: Oxford University Press,
1972), p. 17.
7. See Alex Inkeles and David Smith, Becoming Modern (Cambridge: Harvard University Press,
1974), p. 3.
8. On the lower level technology issue, see E. F. Schumacher, “Industrialization through
‘Intermediate Technology,’” in Ronald Robinson (ed.), Developing the Third World: The Experience
of the Nineteen Sixties (Cambridge: Cambridge University Press, 1971). On the question of
“perpetual underdevelopment,” see Andre Gunder Frank, “The Sociology of Development and the
Underdevelopment of Sociology,” in Frank, Latin America: Underdevelopment or Revolution? (New
York: Monthly Review Press, 1969).
9. Nationalism is connected to both the attitudinal-behavioral formation of a national identity (i.e.,
nation-building) and the structural construction of a state organization (i.e., state-building).
10. Quotes are from “The Spear, Bead and Bean Story,” in Taban Io Liyong, Eating Chiefs: Lwo
Culture from Lolwe to Malkal (London: Heineman, 1970), p. 3.
11. Richard Rose, “Modern Nations and the Study of Political Development,” in Stein Rokkan (ed.),
Comparative Research across Cultures and Nations (The Hague: Mouton, 1968), p. 121. In this
regard, see the latest volume in the studies in Political Development series sponsored by the
Committee on Comparative Politics of the Social Science Research Council — Charles Tilly (ed.),
The Formation of National States in Western Europe (Princeton: Princeton University Press, 1975).
12. Joseph Gusfield “Tradition and Modernity: Misplaced Polarities in the Study of Social Change,”
in Jason Finkle and Richard Gable (eds.), Political Development and Social Change, Second Edition
(New York: John Wiley, 1971), p. 22.
13. Ibid. The phrase “teleological insight” is from Robert Scalapino, “Ideology and Modernization
—The Japanese Case,” in David Apter (ed.), Ideology and Discontent (New York: The Free Press,
1964), p. 106. There Scalapino states that:
Teleological insight is the capacity or, more accurately, the assumed capacity — to discern the
future of one’s own society by projecting it in accordance with conditions and trends in the
‘advanced’ world. The impact of teleological insight upon creativity, political-social interrelation,
and timing in the ideological development of the ‘emergent’ world can scarcely be exaggerated.
From this perspective, the very acceptance of the nation or territorial-state framework by the Third
World is of enormous significance.
14. Gusfield, “Tradition and Modernity,” p. 22, emphasis added.
15. See James Rosenau, “The Concept of Linkage,” in Rosenau (ed.), Linkage Politics, (New York:
The Free Press, 1969), p. 45, for his sequence of behaviour and response definition of linkage.
16. Irving Louis Horowitz, “The Search for a Development Ideal: Models and Their Utopian
Implications,” in Wilard Beling and George Totten (eds.), Developing Nations: Quest for a Model
(New York: Van Nostrand Reinhold, 1970), p. 92. Also see Andre Gunder Frank, Capitalism and
Underdevelopment in Latin America (New York: Monthly Review Press, 1967).
17. Here too we face a situation in which there is no single accepted definition. Quotation is from
Helen Low and James Howe, “Focus on the Fourth World,” in Howe, (ed.), The U. S. and World
Development: Agenda for Action 1975 (New York: Praeger, 1975), p. 35. Their basic distinction for
the Fourth World is:
By early 1970’s a sizable group of developing countries had achieved sustained economic
growth at an annual rate above 6 per cent. At the same time, however, it became apparent that
other developing countries were not participating in the general trend, and that these nations in
effect constituted a ‘Fourth World’ still esentially trapped by poverty. A Time survey article
(December 22, 1975), on the other hand, divided the less developed countries (the LDCs) into a
Third, a Fourth, and a Fifth World and noted that the Fifth World is “perhaps doomed to remain on
a permanent dole” (p. 35). Adding further complications is the fact that the Fourth World is also
used in another sense, that is to refer to “victims of group oppression;” see, for example, Ben
Whitaker (ed.), The Fourth World: Victims of Group Oppression (New York: Schocken Books,
1973).
18. The Group of 77 name continues to be utilized although the “membership” has now risen to more
than 110 states.
19. Ranji Kothari, Footsteps into the Future (New York: The Free Press, 1975), p. 30.
20. See S. E. Finer, Comparative Government (Harmondsworth, England: Penguin Books, 1974), p.
98.
21. Kothari, Footsteps into the Future, p. 30.
22. This is the title of Denis Goulet’s work, The Cruel Choice: A New Concept in the Theory of
Development (New York: Atheneum, 1971).
23. Ibid, p. 326.
24. Consider Gavin Kennedy’s comment in The Military in the Third World (New York: Charles
Scribner’s Sons, 1974), p. 2:
Other terminology, such as ‘backward’ countries, ‘underdeveloped countries,’ ‘developing
countries,’ ‘less developed countries’ and so on, has implications which can produce objections to
its use in all circumstance. Third World’ is relatively neutral in this respect — though still
suggests a rank ordering.
Also note Robert Packenham’s comment, in Liberal America and the Third World, (Princeton:
Princeton University Press, 1973), p. 3, that the Third World:
Refers to all the underdeveloped countries of Asia, Africa, and Latin America, not just to those
which are neutral in the Cold war. ‘Underdeveloped’ is a term of convenience, not a judgement
about the quality of all aspects of life in Third World countries. It is a synonym for ‘Third World’
and one that is commonly used even among people from those countries.
Lastly, Wolf, in United States Policy and the Third World (Boston: Little Brown, 1967), pp. vii-viii,
makes the point that the usage of the label “the Third World,” rather than “underdeveloped,” moves
us from a predominantly economic perspective to one in which political, social, and military factors
are more clearly brought into focus.
PART I
Introduction
Huntington fills out this sequence by first examining the general theory
of modernization, its intellectual history, and the characteristics widely
assigned to the process. Following this overview, Huntington discusses
modernization revisionism—an approach which called into question the
zero-sum perspective so often taken toward the “Great Dichotomy.”
He then explores the concept of political development/modernization and
the major paths followed in examining the political
development/modernization question. Here special attention is paid to the
theoretical work of Almond and Powell, the synthesis produced by Pye, and
Huntington’s own dissent which emphasized the distinction between
political development and political modernization.
This valuable resume of the conceptual and methodological orientations
which affect our perceptions of the Third World concludes by concentrating
on the growing “concern for the formulation of more general theories of
political change”—a formulation which recognizes that “a political system
can be thought of as an aggregate of components [e.g., culture, structure,
groups, leadership, and policies] all changing, some at rapid rates, some at
slower ones”—and thus investigates the questions: “What types of change
in one component tend to be related to similar changes or the absence of
change in other components?” and “What are the consequences of different
combinations of componential changes for the system as a whole?”
With these perspectives in mind, we turn in the third essay in this section
back to the issue of modernity and modernization. For despite the numerous
difficulties found in these terms it is nonetheless rather apparent that they
cannot be ignored within the current context of the Third World. While
caution must be employed, Lerner’s parable of the grocer and the chief in
Balgat, Turkey retains its power and significance.3
If the goal of Third World societies is to create a modern state and if a
modern state requires “participating citizens ... who exercise their rights and
perform their duties as members of a community larger than that of the
kinship network and the immediate geographical locality,” then there is “no
more relevant and challenging task . . . than to explain the processes
whereby people move from being traditional to becoming modern
personalities.”4
This task is one of enormous complexity and one surrounded by
vexatious problems. In “The Attitudes of Modernity,” Kenneth Sherrill aims
“to distill and operationalize some of the significant aspects of theories of
political modernization.” Interestingly, this attitudinal analysis takes place
“within the context of the study of American politics,” and is based on the
proposition that an examination of “the way in which the United States has
developed and maintained these characteristics of modernity” may allow us
to make “certain prescriptive remarks about strategies to follow in the
modernization process.”
After looking at a number of hypotheses about “the politically modern
man” (ranging from “the politically modern man identifies with the national
political community” to “the politically modern man is characterized by a
general faith in the government”), Sherrill concludes that good reason
exists, based on data gathered in the U.S., to accept most of these
hypothetical statements as being “accurate descriptions of the political
orientations of modern men and to believe that these attitudes form a
syndrome, or latent ideology, of modernity.”5 Clearly, such conclusions are
not without importance for our analysis of the Third World.
This brings us to the final essay in this section, Rupert Emerson’s “The
Problem of Identity, Self-hood, and Image in the New Nations.” While
Emerson concentrates on the situation in Africa, the basic principles
explored in the article are applicable to much of the Third World. For, in
many ways, the search for national-identity lies at the core of our overall
concern with development, modernization, and change.
To a considerable extent, nationalism in the Third World was born in
reaction to colonialism. John Kautsky, for example, has argued that:
In the absence of a common language, culture, religion, or race, what is it, then, that provides
the focus for the unity among politically conscious elements from all strata of the population that
is characteristic of nationalist movements in underdeveloped countries as of European nationalist
movements? Speaking of underdeveloped countries in general, there would seem to be no positive
factor at all, but rather the dislike of a common enemy, the colonial power.6
Yet, over time, negative feelings are no substitute for positive impulses,
especially where the idea of nation and the territoriality of the state are not
coterminous.
Herein lies the crux of the problem; self-determination was paramount,
but who was to constitute the self? Relative to this question, Emerson notes
that broad acceptance was given to “three basic propositions: [l] that all
colonial peoples want to get out from under colonialism as speedily as
possible, [2] that each colonial territory as established by the imperial
powers constitutes a nation whose aspirations to become an independent
state have unchallengeable validity, and [3] that the political independence
and territorial integrity of the states thus created must be safeguarded
against attack from within or without.”
The first proposition presents us with little difficulty, for who could deny
the existence of a wave of anticolonialism, a wave now rushing against the
few remaining bastions of “colonial” rule? The second and third
propositions however, while widely accepted, raise more serious questions.
The maintenance of states with severe internal fragmentation along the lines
of “cultural sub-nationalism”7 can be both a challenging and costly affair.
The Nigerian Civil War, mentioned by Emerson, which raged from 1967
through 1970 and which witnessed the “creation” and the “death” of Biafra
was but one grim reminder of this fact. “The enduring challenge of
diversity” linked to “the politics of cultural pluralism”8 within the Third
World will not soon fade into oblivion.
In this respect, Emerson points out that: “To a greater degree than was
expected ... the effect of independence and modernization in some cases has
not been mobilization into the national community but ... a strengthening of
the more parochial tribal affiliation.” Like the state, ethnicity has not
withered away, but, rather, has ofttimes undergone a renaissance in newly
politicized contexts.9
We are thus led to two final considerations. The first of these is that we
must never discount the appearance of the “uninvited guest”, who embodies
the unanticipated consequences of actions. The second of these is that
human beings are multi-leveled creatures, who can simultaneously hold
views which seem to be contradictory. Sub-nationalism, state-nationalism,
and supra-na-tionalism (e.g., Pan-Africanism, Pan-Arabism) can all exist
within the “world view” of single individuals and group clusters. While
there are those who have difficulty in grasping this point, it is, nonetheless,
a necessary first step in analyzing the Third World.
NOTES
1. See David Apter and Charles Andrain, “Comparative Government: Developing New Nations,” in
Apter, Political Change (London: Frank Cass, 1973); Roy Macridis, The Study of Comparative
Government (New York: Random House, 1955); and Robert Holt and John Richardson, Jr.,
“Competing Paradigms in Comparative Politics,” in Holt and John Turner (eds.), The Methodology of
Comparative Research (New York: The Free Press, 1970).
2. Also note Jorgen Rasmussen’s, “ ‘Once You’ve Made a Revolution, Everything’s the Same’:
Comparative Politics,” in George Graham, Jr. and George Carey (eds.), The Post-Behavioral Era
(New York: David McKay, 1972).
3. Daniel Lerner, The Passing of Traditional Society (New York: The Free Press, 1964), pp. 19-42;
also see S. N. Eisenstadt, “Post-Traditional Societies and the Continuity and Reconstruction of
Tradition,” Daedalus, Vol. CII, No. 1 (Winter 1973).
4. Alex Inkeles and David Smith, Becoming Modern (Cambridge: Harvard University Press, 1974),
pp. 4 and 5.
5. See Ibid, pp. 293–295, for a discussion of “a general modernity syndrome.”
6. John Kautsky, “The Politics of Underdevelopment and Industrialization,” in Kautsky (ed.),
Political Change in Underdeveloped Countries (New York: John Wiley, 1962), p. 38.
7. This phrase is from Victor Olorunsola (ed.), The Politics of Cultural Sub-Nationalism in Africa
(Garden City: Doubleday, 1972).
8. See Crawford Young, The Politics of Cultural Pluralism (Madison: The University of Wisconsin
Press, 1976), Chapter 1.
9. On the importance of the “politicization of ethnicity,” see Joan Vincent, “Anthropology and
Political Development,” in Colin Leys (ed.), Politics and Change in Developing Countries
(Cambridge: Cambridge University Press, 1969), p. 52; Burton Benedict, “Pluralism and
Stratification,” in Leonard Plotincov and Arthur Tuden (eds.), Essays in Comparative Social
Stratification (Pittsburgh: University of Pittsburgh Press, 1970), p. 40; and Benedict, Mauritius:
Problems of a Plural Society (London: Pall Mall Press, 1965), pp. 43-67.
Comparative Politics and the Study of
Government
The Search for Focus
Roy C. Macridis
Years ago, it seems now decades ago, I outlined in a little book some of the
most widespread dissatisfactions with what was at the time the study of
comparative government—the way it was taught, the kinds of
preoccupations and research it inspired, and more generally its place in the
discipline.1 I concluded, not unjustifiably it seems to me in retrospect, that
the traditional approach was essentially parochial, monographic,
descriptive, bound to the West and particularly to Western Europe,
excessively formalistic and legalistic, and insensitive to theory-building and
theory-testing. I suggested at the time a crude conceptual outline in terms of
which individual systems could be studied and compared. It comprised the
following three categories: interests and interest configuration, ideology,
and governmental structures. The first corresponded to what are generally
referred to today as the “input” factors. I defined interest in a broad sense. It
encompassed primarily manifested and articulated interests rather than
“latent” interests. My definition had, therefore, a concrete and direGt
relevance to the political process. Ideology was a loose term I gave to all
the relevant political attitudes as they manifest themselves and as they have
crystallized in various political systems over a period of time. I think it
corresponds to what some call today the political culture. Finally, by
“government” I understood the structures through which public officials,
selected in one manner or another, make decisions. I viewed a political
system in terms of its capacity to translate interests and aspirations into
policy and to resolve conflicts2 by transforming both interests and
aspirations into decisions that are widely accepted. In this view, stability
and consensus correlate directly with performance and responsiveness.
I did not go beyond this crude formulation. Perhaps I lacked the
appropriate theoretical sophistication. But I also felt, and continue to feel,
that given the state of our discipline, an attempt to develop a well-knit and
broad-gauge theoretical scheme was, and remains, not only premature but
downright unproductive. I felt, and continue to feel, that the major task of
comparative politics was, and remains, that of raising political questions,
illuminating through parallel studies aspects of political behavior and
decision-making, providing us gradually with a body of experience and data
and sharpening our evaluation of governmental structures and policies.
Only in the long run could hypotheses be developed and tested, and only
then could a scientific outlook in the proper sense of the term be discerned
at least as a distant promise. Finally, I pointed out that one of the major
functions of the study of comparative government was to broaden the
horizons of the students of American politics and help them shed their
parochialism. I argued the obvious: that the study of foreign governmental
forms, policies, and political predicaments helps us step out of our system
and look at it from outside. Plato’s allegory of the cave may be used to
illustrate the plight of the scholars of American politics who have
consistently and persistently failed to take this obvious advice.3
Much has happened since my little book appeared and the study of
comparative politics—rather than of comparative governments—has been
greatly modified. Many of my suggestions have now become part of the
field—not, of course, because I made them, but because many factors have
converged to fashion a new outlook and a new approach. The “behavioral
revolution” or the “successful protest” to which Robert Dahl wrote an
elegant epitaph came down upon us with a crash.4 It is my purpose here not
to judge this revolution as a whole,5 but rather to evaluate some of its most
serious contributions—not all of them beneficial—to the study of
comparative politics.
I
Behavioralism provided, to begin with, a salutary emphasis upon political
factors other than the governmental forms. Although their discovery was
not entirely original, it opened up the study of what we may call the
contextual factors within which political structures and forms develop and
political roles flourish. Borrowing in great part from sociology and
anthropology, behavioralism put a stress upon careful definitions of the
empirical problems to be investigated and upon the formulation of
hypotheses and their testing. It sharpened the tools of our analysis by
introducing new techniques—surveys, interviewing, the compilation of
aggregate data—in an effort to provide correlations between various
socioeconomic and psychological factors and political behavior. Weighing,
measuring, and correlating are among the most positive aspects of the
behavioral revolution in politics. When applied to a political scene,
American or foreign, of which the observers had adequate knowledge, the
emphasis was most beneficial.6 The students were picking dark or shadowy
areas and throwing light upon them. Their findings, or at least their
observations, added to the picture we had and helped us refine it. The
findings, in other words, made sense and produced new data about the
system being studied because we already had a great deal of information
about the system.
Where the behavioral revolution went wrong was at its two extremes—
in its efforts to build “grand theory” or “system theory,” at the one extreme,
and in its study of what may well be called political trivia, at the other. In
between the two lay a fertile field for study and exploration. But it is to the
extremes that most of the work was directed, generally with disappointing
results. This will become abundantly clear, I hope, when I discuss the
following points that exemplify the state of the discipline today: its failure
to establish criteria of relevance and its gross neglect of the study of
governmental structures and forms.
II
The search for relevance and the criteria of relevance has bedeviled political
thought and inquiry. It is an issue that cannot be easily resolved. Society as
an overall system, i.e., a set of interrelations, roles, and structures, consists
of a number of subsystems for which no hard and fast boundaries can be
drawn. In a sense, all that is social is also political, firmly rooted in history.
What is social can be broken down analytically into subsystems, but again,
the lower we move in identifying subsystems, the harder it becomes to set
boundaries. The “web” is there.
Conceivably then every manifestation, every attitude or relationship,
every motivation or idea in society has a relevance to politics. It may
engender aspirations; it may shape interests; it may evoke demands; it may
call for decisions; it may lead to conflicts about values and interests which
necessitate arbitration. Child-rearing, the school curriculum, modes of
entertainment, sex relations, to say nothing of economic interests and
activities, are all potentially related to politics. Yet what is potential is not
actual in empirical terms. In most cases and for most of the time, the great
host of social, economic, and interpersonal relations has no actual relevance
to politics and therefore to the discipline. Yet each and all may at a given
time and place, and under a set of conditions that is impossible to foresee,
assume a political relevance, only to subside again into an apolitical stance.
The dilemma is obvious. Should we study everything that is potentially
political?7 Should we narrow our definition and, if so, how? Behavioralism
provides the worst possible answer—study everything. It postulates that
every aspect of political behavior relates to every aspect of social behavior.
Hence we may finish by studying manifestations and attitudes and
relationships that have no discernible political relevance.
To be sure, there are no a priori grounds on the strength of which one can
discard this holistic approach. The element of potentiality is ever present,
and our inability to develop any rules about the intricate phenomenon that
accounts for the actualization of what is potential makes it impossible to
condemn potentiality as a criterion of relevance. Only two closely related
grounds for its rejection can be suggested. The first is what Professor La-
Palombara has called the rule of parsimony,8 and the second is what I call
the concern with focus. Parsimony suggests the choice of those categories
and concepts on the basis of which we are as sure as we can be that what we
are studying is politically relevant. Concern with focus simply suggests the
most direct way to get at politically relevant phenomena.
Relevance and focus: a set of priorities First and above all, I think it is
our obligation to study all those organized manifestations, attitudes, and
movements that press directly for state action or oppose state action. No
matter what terms we use—decision-making, authoritative allocation of
values, regulation, adjudication, enforcement—we are concerned with the
same old thing, the state. What is it asked to do? And what is it that people
in a community do not want to see it do? To deny this pervasive empirical
phenomenon, in the name of a given theory, is to deny our art or, for those
who prefer, our science. The demand for state action or the demand that a
given action cease is the very guts of politics. No science of politics—or for
that matter, no science—can be built upon concepts and theories that
disregard or avoid empirical phenomena. Why do the French farmers throw
their peaches in the river and their beets on the highway? Why do the
American students leave their comfortable homes to demonstrate in the
streets? Why have American workers patterned their political demands in
one way, but French workers in another? Obviously, to control, to influence,
or to oppose state action.
Thus, my second priority also relates to what I have called the state,
resurrecting what may appear to many graduate students to be an ancient
term. I mean by it, of course, what we have always understood the term to
mean, stripped of all its metaphysical trimmings. It means all the structures
and organizations that make decisions and resolve conflicts with the
expectation that their decisions will be obeyed: the civil service, the
legislature, the executive, the judiciary, the host of public or semipublic
corporations and organizations that are called upon to resolve differences
and to make decisions. I include also the agencies whose function is to
study facts, to deliberate about them, to identify areas of conflict, and to
suggest policy decisions. The most relevant issue here is not the one that
David Easton discusses, or rather suggests, i.e., a theory of likely problems
and predicaments —especially when the theory is pitched at a very high
level of generalization.9 I think that what is instead important is to study the
preparedness of the state to discern predicaments or problems. Potential
problems can be theorized about. The actual political phenomenon,
however, is the existing machinery through which problems are perceived
—the agencies, the research, the flow of information, the manner in which
individual values and constituency considerations enter into the minds of
the men and women who work for the state—and finally includes that
happy or fatal moment when the state copes with, ignores, or is simply
unable to perceive the problem. The state can also, while perceiving the
problem, either alleviate the predicament or suggest solutions utterly
unrelated to it. It was not only the Queen of France who showed a gross
lack of discernment when she suggested that they eat cakes. In what is
reputed to be the most enlightened and modern political system, the
President and the Congress acted in the best tradition of Marie Antoinette
by declaring, “Let them have open housing. . . .”
It is this second priority—the study of the state and all state agencies;
their organization and performance; the scope of their decision-making; the
attitudes of the men and women who perform within their structures the
roles of informing, studying, consulting, and deciding; and the major
constituencies they serve—that has been so sadly neglected in the last
decade or so. Few are the studies that focus on the state as an agency of
deliberation, problem identification, and problem-solving. Few are the
studies of the institutions of the state in the modern developed systems.10
This is no accident at all. After the state was ostracized from the vocabulary
of politics, we found it far more fashionable to study the systems in which
there was no state, i.e., the so-called developing, emerging, or new
systems.11 The result was to eschew the urgent and nagging empirical
situations in the modern and highly industrialized societies where our fate is
to be decided —in order to study political phenomena and especially
political development in the societies where there was no state. No wonder
Professor Huntington began to despair of studying the process of
development in any terms other than “institutionalization,” i.e., the building
of institutions with authority and legitimacy, such as the state and the
party.12
My third priority is the study of political attitudes—the “civic culture,” as
Professor Almond puts it, or what Professor Beer calls “the structure of
norms and beliefs,”13 and what others have very loosely called ideology.
But whatever name we give to them, the phenomena to be studied must
point directly to the beliefs, norms, and orientations about the state
(authority, scope of action, legitimacy, sense of participation, and
involvement). If we are to remain strictly within the confines of relevance
we must narrow our scope to those manifestations and attitudes that directly
link the personal, economic, or psychological phenomena with the political.
The linkage between “micro” and “macro” so well developed by Professor
Almond in his Civic Culture in order to identify meaningful political
orientations needs to be carried a step forward. This can be done only when
we reintroduce the state and its agencies and link them directly to political
orientations. Unless we take this step, we shall remain at the “micro” level.
We shall not link attitudes to structures and forms, to decisions and policies.
The specifics about governmental decisions and performance will elude us.
Finally, my fourth priority—which in a real sense is no priority at all—
relates to the study of what may be called the infrastructure of the political
world: attitudes and ideas; social, economic, and cultural institutions; norms
and values that are prevalent in any given society, national or international.
There is no reason why we shouldn’t study child-rearing, the patterns of
socialization, the degree of concentration of economic power, the
identification of personality types and traits, family life patterns, small
groups and private associations, religious attitudes, and so on. All of these,
as I indicated, may have a relevance to politics. In a number of cases—and
they are the ones that count—the relevance is only too clear. It suggests
itself by the very nature of the empirical phenomenon we are studying. It
links a given organized political manifestation with a contextual factor that
may explain it. It would be difficult to understand the role of the French
military up until the Dreyfus case without knowing something about the
education they received in Jesuit schools. But in this case we study
education because we begin with the army as a political force operating
within the government and the state. We go deeper into contextual factors in
order to find an explanation of a manifest political phenomenon.
What I am trying to suggest by these priorities, then, is primarily a
change of focus. My concern is simply to pinpoint what is political. We
begin with the political; we catch it, so to speak, in its most visible, open,
and raw manifestation; we begin with the top of the iceberg before we go
deep to search for its submerged base. We focus on the state and its
agencies, on its types of action or inaction, and on all those organized
manifestations that call for action or inaction on its part. We study the forms
of decision-making and analyze and evaluate its substance. We explore its
impact upon groups, interests, and power elites within the system; we study
in turn their reactions to state actions and their counterdemands as they are
manifested through various media from political parties down to voting.
The fallacy of inputism Two terms that have gained wide currency in the
last decade are “input” and “output.” The system converts demands into
decisions. Through the feedback mechanism, output factors influence the
input side. Emphasis is placed upon the input factors, but the state is given a
degree of autonomy and independence, and through a process that is by no
means clear, it can influence supports and demands.14 The difficulty comes
with the selection of the input factors, i.e., with the same problem of
relevance that we have discussed. Do we study again all societal
manifestations on the assumption that they all make inputs? Do we consider
attitudinal data, aggregate data, hard and soft data ranging from the number
of hospital beds to child-father relations? Where and in what manner do we
define the subject matter for study, and what is our cutoff point? Political
scientists are often like thirsty men who go looking for water in a contextual
Sahara when more often than not it is right there fresh from the spring—or
at least from a well-chlorinated reservoir. Their search in the contextual
wasteland brings only further difficulties upon them, for there is no theory,
no conceptual scheme that links—in any form that is testable —the amassed
socioeconomic and psychological data with the political. In fact, emphasis
upon the input factors very often not only neglects the political but
sometimes explicitly avoids it.
I am inclined to define what I call “inputism” as the study of society by
political scientists without a political focus and very often without a
political question. The job is enticingly easy. All that is needed is a
questionnaire, interviewers, a pool of respondents, the UN Statistical
Yearbook, and a counter-sorter or, even better, a computer. I wish to
reemphasize the phrase “without a political focus and very often without a
political question.” When the empirical political situation and the empirical
political phenomenon we are investigating make it necessary—as often
happens—to study the socioeconomic or psychological factors on the input
side, then such study is focused and relevant, for the input factors are
analyzed to “explain” the situation we are investigating. We hypothesize
that the attitude of the French military with regard to a series of political
decisions or with regard to the process of decision-making was, among
other things, shaped by the education they received at the Jesuit schools.
The linkage between the two, I believe, can be made. It will not fully
explain the attitude of the military. But I think it may provide one of the
first steps leading to explanation. We start with the concrete political
problem. Inputism would reverse the priorities, advocating the study of the
socialization of the elite groups in the French educational system, with the
unwarranted expectation that such study would clarify “politics” in general
and help us explain political behavior. What behavior? With regard to what
problem? At what time?
Inputism tends to lead to three fallacies: that of determinism, that of
scientism, and that of superficiality. All three fallacies are related.
According to determinism, it is the input factors that shape political
action. The political phenomenon is almost invariably reduced either to a
number of nonpolitical determinants, in which case we have a multiple kind
of reductionism, or to one factor, in which case we have a single-factor
reductionism. In either case, the state can play virtually no independent
problem-solving or attitude-forming role. It is only through a process of
feedback—not clearly understood and not easily demonstrated—that
governmental action may influence the determinants that then in turn will
act upon the governmental decision-making process. Politics constantly
remains a dependent variable. It is, to use Bentley’s expression, the
parallelogram of interest action and interaction, that is, the parallelogram of
all socioeconomic and psychological determinants that will shape the
decisionmaking machinery and will determine its output. The famous
“black box,” as the graduate students have come to know the government,
is at its best a filter mechanism through which interests express themselves
and at its worst a simple transmission mechanism. The role of the state is
reduced to the narrow confines of an organization that channels, reflects,
and expresses commands and instructions that come from “elsewhere.” The
hint to political scientists is obvious: study everything but the machinery of
the state and its organizational structures; study the “elsewhere.”
Scientism constitutes the effort to measure as accurately as possible the
weight, scope, and persistence of the input factors, on the purely gratuitous
assumption that they are or can be linked causally to political phenomena.
The assumption is gratuitous because we have failed as yet to establish any
such causal links and because it is doubtful that we ever will. The
assumption is confounded when the political phenomena with which the
input factors are to be linked are not clearly stated. Sometimes system
theory suggests the broadest possible relationship among “concepts,” rather
than variables—consensus, stability, performance; sometimes it offers a
very narrow-gauge hypothesis linking some political manifestation to one
non-political variable (for example, voting with income or race). The first
attempt obviously bogs down into analytical exercise rather than empirical
testing, while the second one will never attain the level of testability on the
basis of which higher-level propositions can be made. Even when system
theory establishes clear-cut concepts linked to empirical phenomena from
which testable propositions can emerge, it is impossible to move back from
the propositions to the concepts and to the overall theory through a series of
verifiable tests that exclude all alternative propositions, concepts, and
theories. Scientism therefore leads us from hyperfactualism to hyper-
theorizing—the latter becoming progressively an exercise (often brilliant)
in intellectual virtuosity. It lacks, however, the only thing that really counts
—empirical relevance.
Since no theory as yet offered has shown its worth in causally linking
determinants with political phenomena, our efforts very often end with the
superficial juxtaposition of a given determinant with a given political
phenomenon—that is to say, with correlations. Since no adequate theory
has been offered, however, and since we therefore have no explanation,
correlational findings are a somewhat more sophisticated version of the
“sun spot” theory. Martin Lipset’s book Political Man is an illustration of
this.15 At the end of this excellent study, the reader is not sure whether open
democratic societies are affluent because they are open and democratic or
whether it is the other way round.
III
Determinism, scientism, and correlational studies that have a distinct trait of
superficiality typify the state of a discipline that has consistently eschewed
the hard and persistent empirical phenomena that ought to concern it in the
name of theory-building and theory-testing. Structures and processes and
the manifestly political institutions through which decisions are made have
been relegated to the level of epiphenomena. The examination and
evaluation of policies have been handed over to the journalists and
politicians, and the formulation of a judgment in the name of knowledge is
considered incompatible with the canons of a self-imposed scientific
objectivity. It is these trends that account for the state of the discipline as a
whole, and they affect particularly the study of comparative politics, or the
comparative study of politics. The state of the discipline can be summed up
in one phrase: the gradual disappearance of the political. To repeat, if
government is viewed as the reflection of the parallelogram of
socioeconomic and psychological and other determinants, the prescription
for political science becomes a proscription of the study of government.
Yet the behavioral revolution has also had, as I have noted, beneficial
effects. We shall never return exclusively to normative speculation, and we
shall never be satisfied with judgments about political phenomena without
the benefit of careful measurement. We shall continue to distinguish sharply
between “facts” and “values,” and we shall subject our postulates to a
critical examination, demanding always clarity of definitions and terms.
Where the propositions about behavior can be tested, we shall test them
under all the canons of controlled inquiry that the social sciences have
developed. We shall continue to seek to build theory, i.e., a set of
interrelated and interconnected propositions, each of which has direct
empirical meaning and relevance, and we shall continue to develop narrow
hypotheses that can be tested, i.e., invariably falsified. We shall use the
many tools of empirical inquiry available to us—survey opinion data,
attitudinal data, aggregate data—and in both the construction of our
research and our search for the explanation of political phenomena, we shall
feel free to borrow, when the occasion demands, from the theoretical
sophistication of many disciplines—sociology, anthropology, economics,
and psychology—and, of course, to use the empirical data that history
provides. But the time has come to qualify and reconsider our quest for a
science of politics in the full sense of the term. But in the last analysis this
may be a contradiction in terms. There can be no science where the element
of human will and purpose predominates. Politics is a problem-solving
mechanism; its study must deal with it and not with the laws surrounding
behavior. The ultimate irony is that even if laws could be discovered, then
our discipline would be primarily concerned with an effort to explain why
they are not obeyed—why the laws are really non-laws. Natural sciences
began by addressing themselves to empirical phenomena in order to
understand them, explain them, and control them. The ultimate goal of the
natural sciences has been to control nature. The higher the level of
generalization that subsumes a number of measurable relationships, the
higher the potentiality of control. It is the other way around with politics.
The study of politics explicitly divorces knowledge from action and
understanding from control. The laws that we constantly seek will tell us
little about our political problems and what to do about them. Our concern
becomes scholastic.16
I therefore suggest that we reconcile ourselves to the fact that while we
can have an understanding of some political phenomena, a history of
politics and political movements, an understanding of the functioning of
governmental forms and structures, a concern and indeed a focus on major
concepts such as power, decision-making, interest, organization, control,
political norms and beliefs, obedience, equality, development, consensus,
performance, and the like, we do not and cannot have a science of politics.
We can have, at most, an art. Second, and this is the sign of the art, we may
manage to arrive at some inductive generalizations based upon fragmentary
empirical evidence. An inductive generalization is at best a statement about
behavior. It can be derived from identical action and interaction under
generally similar conditions over a long period of time in as many different
contexts as possible. The behavior is not explained, but the weight of
evidence allows us to anticipate and often predict it. A series of solidly
supported inductive generalizations may in the last analysis be the most
fruitful way to move gradually to a scientific approach, as it provides us
with a rudimentary form of behavioral patterns. Our knowledge of politics
is then at most an understanding of our accumulated experience. It is in this
area that comparative politics has an important role to play. By carefully
identifying a given behavior or structure or movement and by attempting to
study it in as many settings as possible and over as long a period of time as
possible, we can provide generalizations backed by evidence.
If we view our discipline as an art and if we limit its goals to inductive
generalizations about politics, i.e., a well-ordered and catalogued table or
listing of accumulated experience, then two or three imperatives for
research emerge, providing focus and satisfying the need for parsimony.
First, we must study the practitioners of the art, the political leaders who
hold office and, more generally, the governing elites that aspire to or
possess political power. Second, we must study the structures and
organizations and mechanisms through which the elites gain political power
and exercise it, that is the parties and other political associations. Third, we
must be concerned with the governmental institutions through which
demands are channeled or, just as often, by which they are generated. These
imperatives do not exhaust our immediate task, but give us a starting point.
In studying the governmental elites and the institutions through which
they gain and exercise power, we ought to consider the art of government as
a problem-solving and goal-oriented activity. This kind of activity
characterizes any art. The task of government is to identify problems (or to
anticipate them) and provide for solutions. Our study then is constantly to
ask ourselves, How well is the art performed? Who within the government
listens, who foresees, who advises and suggests policy? What are the skills
of the practitioners, and what are their objective capabilities? Finally, what
is the impact of a decision upon the problem or the predicament it was
designed to alleviate or to remove? The practitioner is not strictly bound by
determinants. Communal life suggests and often sets goals of performance
and achievement that become more than normative goals. They become in a
way the “operative goals” that give direction to political action. The
governing elite plays an independent role in seeking out the goals and in
implementing them. Shonfield in his book on planning refers to the French
planning as the result of an “elitist conspiracy.”17 More often than not
decision-making is an elitist conspiracy whose study and assessment would
be far more rewarding than the survey and elaboration of all the input
factors or the nonpolitical determinants.
But in the last analysis government is an act of will that can shuffle and
reshuffle many of the determinants. Government involves choice, and the
parameters are often wider than we are inclined to think. Any government
will begin by surveying the conditions that appear to indicate the limits of
freedom and choice; a government must always ask “what it has.” But any
government must also be in a position to assess what it wills. To say this is
not to return to metaphysical speculation about the “will” of the state or the
government. It is simply to reintroduce as integral parts of our discipline the
state’s performance and choices and the institutions through which they are
implemented.
IV
The central focus of politics, therefore, and of the study of comparative
politics is, in my opinion, the governmental institutions and political elites,
their role, their levels of performance and nonperformance. Stating this in
such blunt terms appears to be utterly naive. Shall we return then to the
descriptive study of governmental institutions? Far from it. What I am
suggesting is a starting point and a focus of investigation. Any such
investigation, we know today, will inevitably lead us, as it should, far and
wide in search of the contextual factors (rather than determinants) within
the framework of which a government operates and to which its action, its
performance, and its policies may often be attributed. We shall have to
probe the infrastructure, but without losing sight of either our focus or the
relevant question we began our investigation with.
I can think, in the manner of Machiavelli and Montesquieu and Tocque-
ville, of a number of relevant questions, all of which we tend to evade either
because they are “difficult” or because they are not amenable to “scientific
inquiry” or because they involve “value judgments”: What accounts for a
well organized civil service? What is the impact of large-scale organizations
—parties, bureaucracy, and so on—upon the citizen? Under what conditions
does public opinion exercise its influence on the government? What
accounts for political instability? Is an executive who is responsible to the
people more restrained than one responsible to the legislature? How and
under what conditions does representation degenerate into an expression of
particular interests? Under what conditions do the young people maintain
political attitudes different from those of their fathers, and at what point do
they revolt? Why does a ruling class become amenable to reform? Under
what conditions do ruling groups become responsive to popular demands?
I can multiply these questions, but I think these illustrate my point. None
of them can lead to hard hypotheses and proof (or disproof). Some cannot
be easily answered. But this is not too important unless we are to accept that
only those questions that can lead to testing in the rigorous, and therefore
impossible, meaning of the term have the freedom of the market. In fact, the
questions I suggest lead to a comparative survey, both historical and
contemporaneous, of some of the most crucial political phenomena:
responsiveness, performance, change, development, and a host of others.
Such a survey will inevitably produce inductive generalizations, perhaps in
the manner of Machiavelli, but with far more sophisticated tools and greater
access to data than was ever the case before. It will inevitably help us to
qualify our questions and to reformulate them as hypotheses that will
suggest other qualifications—new variables, if you wish—and lead to
further investigation—testing if you wish—and the reformulation of the
questions —the gradual development of theory, if you like.
Through the study of governmental institutions and political elites, we
shall be concerned with fundamental problems of politics. The first
symptoms and indicators of all pervasive political phenomena—revolution,
authority and stability, legitimacy, participation—are registered in the
composition, organization, and performance of the government. I suggest
that we begin with these and broaden our horizon as our study goes on. The
policies pursued, their consistency, their congruence with the existing social
and economic problems, the governors’ awareness of such problems—all
constitute good indicators of stability or instability.
My second suggestion, therefore, for the study of comparative politics is
to concentrate on the policies pursued by various governments in differing
political systems—to highlight the conditions of performance or
nonperformance, as the case may be. Here one of the cardinal tasks of
political science is not to study all the integrative factors, as has been the
case, but to identify problems that may lead to conflict—not when conflict
erupts, but when the society is at the threshold, so to speak, of a conflict
situation. The study of the impact of the decisions made by the government
can provide us with an excellent laboratory for comparative analysis.
Conflict-policies-decisions-consequences, this is the heart of the political
life. If the political scientists and particularly the students of comparative
politics were to concentrate on these central manifestations, they would
liberate themselves from the shackles that bind them today, according to
which the “determinants” come first and the political questions second.
Problems and questions about economic planning, the decay of the
representative institutions fashioned in the nineteenth century, the impact of
the expanding population upon individual freedoms or of “bigness” upon
the individual citizen as a participant member of the body politic, the
handling of the new instruments of violence and the growth of the power
and status of those who command them, the military, these are the basic
problems that affect us. To evade them in the name of science is to abandon
our vocation to understand, to explain, to point to dangers, and to draw
from our accumulated experience both suggestions and warnings.
Last, we must be prepared in the light of the experience we have studied
and accumulated to move forward and offer policies. In doing so we do not
enter the forbidden territory of “political action” nor, as it is so naively
argued so often, do we simply leave our scientific hat in the office to don
the activist hat. We do not move from “fact” to “value”—to repeat the facile
distinction made by Max Weber, as if the dichotomy between the two did
not involve a value judgment. We remain at the level of problem
identification and problem-solving, and we suggest remedies to the
policymakers. The more detached our suggestions, the better based upon
political experience (in the broad sense), the more plausible they are likely
to be. And if we evaluate the same problem differently and suggest different
solutions and reach different conclusions, this is only an indication,
provided all canons of objectivity and reasoning are respected, that ours is
an art. The body of accumulated experience must still be refined and
studied if we are to gain an understanding that we can all share alike.
1. The Study of Comparative Government (New York, 1955). For a discussion of the evolution and
later state of the field, see Harry Eckstein and David E. Apter, eds. Comparative Politics: A Reader
(New York, 1963), particularly the excellent general introduction by Harry Eckstein. See also, Roy C.
Macridis and Bernard E. Brown, eds. Comparative Politics: Notes and Readings, 3rd ed.
(Homewood, 1968).
2. The term “authoritative allocation of values,” suggested by David Easton to define a political
system, is misleading unless the words convey only what the author understands. There are many
“values” that are not “authoritatively allocated” in a social system.
3. Students of American political institutions, after paying tribute to the “genius,” the “pragmatism,”
and the “consensual nature” of the polity, are now beginning to ask questions they should have raised
long ago.
4. Robert A. Dahl, “The Behavioral Approach in Political Science: Epitaph for a Monument to a
Successful Protest,” American Political Science Review, LV (December 1961), 763-772.
5. See Heinz Eulau, The Behavioral Persuasion in Politics (New York, 1963); and Eulau, ed.
Political Behavior in America: New Directions (New York, 1966).
6. For instance, Angus Campbell, Philip E. Converse, Warren E. Miller, and Donald E. Stokes, The
American Voter (New York, 1960). Also, E. Deutsch, D. Lindon, and P. Weill, Les Families
politiques aujourd’hui en France (Paris, 1966).
7. The pitfalls are obvious in David Truman’s The Governmental Process (New York, 1951). For a
criticism of the latent-group theory, see my “Interest Groups in Comparative Analysis,” Journal of
Politics, XXIII (February 1961), 25-45.
8. Joseph LaPalombara, “Macrotheories and Microapplications in Comparative Politics: A Widening
Chasm,” in this issue of Comparative Politics.
9. A Systems Analysis of Political Life (New York, 1955), esp. Chs. 4, 14. and 15.
10. See, however, among others, Kenneth Waltz, Foreign Policy and Democratic Politics (Boston,
1967); Andrew Shonfield, Modern Capitalism (London, 1965); Richard F. Fenno, The President’s
Cabinet (Cambridge, Mass., 1959).
11. The emphasis was given by the SSRC Committee on Comparative Politics, whose contributions
to the field, however, can receive only the highest possible praise.
12. Samuel P. Huntington, “Political Development and Political Decay,” World Politics, XVII (April
1965), 386-430.
13. Gabriel A. Almond and Sidney Verba, The Civic Culture (Boston, 1965); Samuel H. Beer and
Adam B. Ulam, eds. Patterns of Government, 2nd ed. (New York, 1962), Introduction by Samuel H.
Beer.
14. David Easton, “An Approach to the Analysis of Political Systems,” World Politics, IX (April
1957), 383-400.
15. Seymour Martin Lipset, Political Man (New York, 1960).
16. Barrington Moore, Jr. “The New Scholasticism and the Study of Politics,” World Politics, VI
(October 1953), 122-138.
17. Modern Capitalism.
The Change to Change
Modernization, Development, and Politics
Samuel P. Huntington*
1. Predominance of universal-istic,
1. Predominance of specific, and achievement norms
ascriptive, particularistic, 2. High degree of social mobility (in a
diffuse patterns general—not necessarily “vertical”—
2. Stable local groups and sense)
limited spatial mobility 3. Well-developed occupational system,
3. Relatively simple and insulated from other social structures
stable “occupational” 4. “Egalitarian” class system based on
differentiation generalized patterns of occupational
4. A “deferential” achievement
stratification system of 5. Prevalence of “associations,” i.e.,
diffuse impact functionally specific, non-ascriptive
structures
3. Modernization is a systemic process. Changes in one factor are related to and affect changes in
the other factors. Modernization, as Daniel Lerner has expressed it in an oft-quoted phrase, is “a
process with some distinctive quality of its own, which would explain why modernity is felt as a
consistent whole among people who live by its rules.” The various elements of modernization
have been highly associated together “because, in some historic sense, they had to go together.”11
5. Modernization is a lengthy process. The totality of the changes which modernization involves
can only be worked out through time. Consequently, while modernization is revolutionary in the
extent of the changes it brings about in traditional society, it is evolutionary in the amount of time
required to bring about those changes. Western societies required several centuries to modernize.
The contemporary modernizing societies will do it in less time. Rates of modernization are, in this
sense, accelerating, but the time required to move from tradition to modernity will still be
measured in generations.
9. Modernization is a progressive process. The traumas of modernization are many and profound,
but in the long run modernization is not only inevitable, it is also desirable. The costs and the
pains of the period of transition, particularly its early phases, are great, but the achievement of a
modern social, political, and economic order is worth them. Modernization in the long run
enhances human well-being, culturally and materially.
(a) culture, that is, the values, attitudes, orientations, myths, and
beliefs relevant to politics and dominant in the society;
(b) structure, that is, the formal organizations through which the
society makes authoritative decisions, such as political parties,
legislatures, executives, and bureaucracies;
(c) groups, that is, the social and economic formations, formal and
informal, which participate in politics and make demands on the
political structures;
(d) leadership, that is, the individuals in political institutions and
groups who exercise more influence than others on the allocation of
values;
(e) policies, that is, the patterns of governmental activity which are
consciously designed to affect the distribution of benefits and
penalties within the society.
The study of political change can fruitfully start with the analysis of
changes in these five components and the relations between change in one
component and change in another. How is change in the dominant values in
a system related to change in its structures? What is the relation between
mobilization of new groups into politics and institutional evolution? How is
turnover in leadership related to changes in policy? The starting assumption
would be that, in any political system, all five components are always
changing, but that the rate, scope, and direction of change in the
components vary greatly within a system and between systems. In some
instances, the rate of change of a component may approach zero. The
absence of change is simply one extreme rate of change, a rate rarely if ever
approximated in practice. Each component, moreover, is itself an aggregate
of various elements. The political culture, for instance, may include many
subcultures; the political structures may represent a variety of institutions
and procedures. Political change may be analyzed both in terms of changes
among components and in terms of changes among the elements of each
component.
Components and elements are the objects of change. But it is still
necessary to indicate what types of changes in these are significant to the
study of political change. One type of change which is obviously relevant is
change in the power of a component or element. Indeed, some might argue
that changes in power are the only changes with which political analysis
should be concerned. But to focus on power alone is to take the meaning
out of politics. Political analysis is concerned with the power of ideologies,
institutions, groups, leaders, and policies. But it is also concerned with the
content of these components and with the interrelation between changes in
content and changes in power. “Power” here may have the usual meaning
assigned to it in political analysis.54 The “content,” on the other hand, has to
be defined somewhat differently for each component. The content of a
political culture is the substance of the ideas, values, attitudes, and
expectations dominant in the society. The content of the political
institutions of the society, on the other hand, consists of the patterns of
interaction which characterize them and the interests and values associated
with them. The content of political groups refers to their interests and
purposes and the substance of the claims which they make on the political
system. The content of the leadership refers to the social-economic-
psychological characteristics of the leaders and the goals which they
attempt to realize. And the content of policies, of course, involves the
substance of the policies, their prescriptions of benefits and penalties.
The analysis of political change may in the first instance be directed to
simple changes in the power of components and elements of the political
system. More important, however, is the relation between changes in the
power of individual components and elements and changes in their content.
If political analysis were limited to changes in power, it could never come
to grips with their causes and consequences. The recurring problems of
politics involve the trade offs of power and content. To what extent do
changes in the power of a political ideology (measured by the number of
people who adhere to it and the intensity of their adherence) involve
changes in the substance of the ideology? Under what circumstances do
rapid changes in the power of political leaders require changes in their
purposes and goals (the “moderating” effects of power) and under what
circumstances may the power of leaders be enhanced without significant
changes in their purposes? History suggests, for instance, that professional
military officers can acquire political power in liberal-, socialist, or
totalitarian societies only at the expense of abandoning or modifying the
conservative military values.55 In most systems, the enhancement of the
power of an ideology, institution, group, leader, or policy is bought at the
price of some modification of its content. But this is by no means an
invariable rule, and a variety of propositions will be necessary to specify the
trade offs between power and content for different components in different
situations. One important distinction among political systems may indeed
be the prices which must be paid in content for significant increases in the
power of elements. Presumably the more highly institutionalized a political
system is, the higher the price it exacts for power.
Political change may thus be analyzed at three levels. The rate, scope,
and direction of change in one component may be compared with the rate,
scope, and direction of change in other components. Such comparisons can
shed light on the patterns of stability and instability in a political system and
on the extent to which change in one component depends upon or is related
to change or the absence of change in other components. The culture and
institutions of a political system, for instance, may be thought of as more
fundamental to the system than its groups, leaders, and policies.
Consequently, stability might be defined as a particular set of relationships
in which all components are changing gradually, but with the rates of
change in culture and institutions slower than those in other components.
Political stagnation, in turn, could be defined as a situation in which there is
little or no change in the political culture and institutions but rapid changes
in leadership and policies. Political instability may be a situation in which
culture and institutions change more rapidly than leaders and policies, while
political revolution involves simultaneous rapid change in all five
components of the system.
As a second level of analysis, changes in the power and content of one
element of one component of the system may be compared with changes in
the power and content of other elements of the same component. This
would involve, for instance, analysis of the rise and fall of ideologies and
beliefs, of institutions and groups, and leaders and policies, and the changes
in the content of these elements associated with their changing power
relationships. Finally, at the most specific level of analysis, attention might
be focused upon the relation between changes in power and changes in
content for any one element, in an effort to identify the equations defining
the price of power in terms of purposes, interests, and values.
A relatively simple set of assumptions and categories like this could be a
starting point either for the comparative analysis of the more general
problems of change found in many societies or for the analysis in depth of
the change patterns of one particular society. It could furnish a way of
bringing together the contributions which studies of attitudes, institutions,
participation, groups, elites, and policies could make to the understanding
of political change.
A somewhat different approach, suggested separately by both Gabriel
Almond and Dankwart Rustow, focused on crisis change and also provided
a general framework for analyzing political dynamics. Earlier theories of
comparative politics and development, Almond argued, could be classified
in terms of two dimensions.56 To what extent did they involve an
equilibrium or developmental models? To what extent were they predicated
upon determinacy or choice? Reviewing many of the writers on these
problems, Almond came up with the following classification:
Approaches to Comparative
Politics
Equilibrium Developmental
III Deutsch Moore
Determinacy I Parsons Easton
Lipset
II Downs Dahl
Choice IV Harsanyi Leiserson
Riker
1. Talcott Parsons, The Social System (Glencoe, 1951), p. 486; Don Martindale, “Introduction,” in
George K. Zollschan and Walter Hirsch, eds. Explorations in Social Change (Boston, 1964), p. xii;
Alvin Boskoff, “Functional Analysis as a Source of a Theoretical Repertory and Research Tasks in
the Study of Social Change,” ibid., p. 213; Robin Williams, American Society (New York, 1960), p.
568. By 1969, however, Williams felt a little more optimistic about the prospects for a breakthrough
in the sociological study of change. See Robin Williams, Sociology and Social Change in the United
States (St. Louis, Washington University Social Science Institute, Studies in Comparative
International Development, vol. 4, no. 7, 1968-69).
2. David Easton, The Political System (New York, 1953), p. 42.
3. Talcott Parsons, The Social System (Glencoe, 1951), pp. 480ff.
4. Frank X. Sutton, “Social Theory and Comparative Politics,” in Harry Eckstein and David Apter,
eds. Comparative Politics: A Reader (New York, 1963), pp. 67 ff.
5. Marion Levy, Modernization and the Structure of Societies (Princeton, 1966), 1:11.
6. Cyril E. Black, The Dynamics of Modernization (New York, 1966), p. 7.
7. Dankwart A. Rustow, A World of Nations (Washington, 1967), p. 3.
8. Dankwart A. Rustow and Robert E. Ward, “Introduction,” in Ward and Rus-tow, eds. Political
Modernization in Japan and Turkey (Princeton, 1964), pp. 6-7.
9. See Samuel P. Huntington, Political Order in Changing Societies (New Haven, 1968), pp. 32-37.
10. Black, Modernization, pp. 1-5; Reinhard Bendix, ‘Tradition and Modernity Reconsidered,”
Comparative Studies in Society and History, IX (April 1967), 29293.
11. Daniel Lerner, The Passing of Traditional Society (Glencoe, 1958), p. 438.
12. Black, Dynamics of Modernization, pp. 155, 174.
13. Reinhold Niebuhr, Christian Realism and Political Problems (New York, 1953), p. 5; T. S. Eliot,
The Idea of a Christian Society (New York, 1940), p. 64.
14. Karl Mannheim, Man and Society in the Age of Reconstruction (London, 1940); Hannah Arendt,
The Origins of Totalitarianism (New York, 1951).
15. Harold D. Lasswell, ‘The Universal Peril: Perpetual Crisis and the Garrison-Prison State,” in
Lyman Bryson, Louis Finkelstein, and R. M. Maclver, eds. Perspectives on a Troubled Decade:
Science, Philosophy, and Religion, 1939-1949 (New York, 1950), p. 323; and Walter Lippmann, The
Public Philosophy (Boston, 1955), pp. 3-8.
16. Sir Henry Maine, Ancient Law: Its Connection with the Early History of Society and Its Relation
to Modern Ideas (London, 1861); Ferdinand Tonnies, Gemeinschaft und Gesellschaft (Leipzig,
1887); Max Weber, Wirtschaft und Gesellschaft, Part I (Tubingen. 1922).
17. The late 1960s saw the emergence of “postmodern” theorizing, the leading scholars of which,
however, had not been primarily involved in the analysis of the transition from tradition to modernity.
These theories arose out of concern with the impact of technology on modern rather than traditional
society. See Daniel Bell, -Notes on the Post-Industrial Society,” The Public Interest, VI (Winter
1967), 2435 and VII (Spring 1967), 102-18, and Zbigniew Brzezinski, Between Two Ages: Americas
Role in the Technetmnic Era (New York, 1970). Both Brzezinski and Bell would probably assign
many of the nine characteristics of modernization mentioned above to the transition from modernity
to what follows. Both stand generally in the optimistic stream and in that sense share more with the
modernization theorists than they do with the early twentieth century pessimists. More than the
modernization theorists, however, both have been criticized by other writers who view with alarm the
prospect of a postindustrial or technetmnic society. Political scientists have yet to probe very deeply
the political implications of this new historical transition.
18. See Joseph R. Gusfield, “Tradition and Modernity: Misplaced Polarities in the Study of Social
Change,” American Journal of Sociology, LXX1I (January 1966), 351-62; Reinhard Bendix,
“Tradition and Modernity Reconsidered,” Comparative Studies in Society and History, IX (April
1967), 293-346; Lloyd and Suzanne Rudolph, The Modernity of Tradition (Chicago, 1967); S. N.
Eisenstadt, “Breakdowns of Modernization,” Economic Development and Cultural Change, XII (July
1964), 345-67, and “Tradition, Change, and Modernity,” Eliézer Kaplan School of Economic and
Social Sciences, Hebrew University; J. C: Heesterman, “Tradition in Modern India,” Bijdragen Tot
de Taal-, Land- en Volkenkunde, Deel 119 (1963), 237-53; Milton Singer, ed. Traditional India:
Structure and Chance (Philadelphia, 1959); Rajni Kothari, “Tradition and Modernity Revisited,”
Government and Opposition, III (Summer 1968), 273-93; C. S. Whitaker, Jr., The Politics of
Tradition: Continuity and Change in Northern Nigeria, 1946-1966 (Princeton, 1970).
19. Rustow, World of Nations, p. 12.
20. See Samuel P. Huntington, “Civilian Control of the Military: A Theoretical Statement/’ in Heinz
Eulau, Samuel J. Eldersveld, and Morris Janowitz, eds. Political Behavior: A Reader in Theory and
Research (Glencoe, 1956), 380-85 and “Civil-Military Relations,” International Encyclopedia of the
Social Sciences (New York, 1968), II: 487.
21. See, especially, Singer, ed. Traditional India, pp. x-xvii and Heesterman, ‘Tradition in Modern
India,” pp. 242-43.
22. Gusfield, “Misplaced Polarities,” p. 352.
23. See Hideo Kishimoto, “Modernization versus Westernization in the East,” Cahiers d’Histoire
Mondiale, VII (1963), 871-74, and also Heesterman, “Tradition in Modern India,” 238.
24. Bendix, “Tradition and Modernity,” p. 326, and also Whitaker, Politics of Tradition, pp. 3-15.
25. Gusfield, “Misplaced Polarities,” p. 352; Heesterman. ‘Tradition in Modern India,” p. 243; Lloyd
I. and Suzanne Hoeber Rudolph, ‘The Political Role of India’s Caste Associations,” Pacific Affairs,
XXXIII (March 1960). 5-22.
26. See Wilbert Moore, “Social Change and Comparative Studies,” international Social Science
Journal, XV (1963), 523; J. A. Ponsioen, The Analysis of Social Change Reconsidered (The Hague,
1962), pp. 23-25.
27. Bendix, “Tradition and Modernity,” pp. 313–14; Gusfield, “Misplaced Polarities,” pp. 352–53.
28. Bendix, pp. 315, 329; Eisenstadt. “Tradition. Chance, and Modernity.” pp. 27-28.
29. Gabriel Almond, “Introduction: A Functional Approach to Comparative Politics,” in Almond and
James S. Coleman, eds. The Politics of the Developing Areas (Princeton, 1960), pp. 23-24.
30. ibid., p. 25.
31. Gabriel A. Almond and G. Bingham Powell, Jr., Comparative Politics: A Developmental
Approach (Boston, 1966), p. 13.
32. Lucían W. Pye, Aspects of Political Development (Boston, 1966), pp. 31-48.
33. Samuel P. Huntington, “Political Development and Political Decay,” World Politics, XVII (April
1965), 387-88.
34. John D. Montgomery, “The Quest for Political Development,” Comparative Politics, I (January
1969), 285-95.
35. Huntington, “Political Development and Political Decay:’ pp. 389-93.
36. Lucían W. Pye, “Introduction,” in Pye, ed. Communications and Political Development
(Princeton, 1963), p. 16.
37. See Pye, Aspects, pp. 45-48; Almond and Powell, Comparative Politics, pp. 299 ff.; Ward and
Rustow, Japan and Turkey, pp. 6-7; Rupert Emerson, Political Modernization: The Single-Party
System (Denver, 1963), pp. 7-8; S. N. Eisenstadt, “Bureaucracy and Political Development,” in
Joseph LaPalombara, ed. Bureaucracy and Political Development (Princeton, 1963), p. 99.
38. Pye, Aspects, p. 47; Almond and Powell, Comparative Politics, p. 306. For an intriguing analysis
of some of these problems, see Fred W. Riggs, “The Dialectics of Developmental Conflict,”
Comparative Political Studies, I (July 1968), 197 if.
39. Alfred Diamant, “The Nature of Political Development,” in Jason L. Finkle and Richard W.
Gable, eds. Political Development and Social Change (New York, 1966), p. 92.
40. Huntington, “Political Development and Political Decay,” p. 390.
41. Dankwart A. Rustow, “Change as the Theme of Political Science (Paper delivered at
International Political Science Association Round Table, Torino, September 1969), pp. 1-2.
42. Some people may say that people in glass houses should not throw stones on the grounds that I
did, after all, argue that political development should be defined as political institutionalization in my
1965 article on “Political Development and Political Decay.” My answer would be: true enough. But
I do not mind performing a useful function by throwing stones and thus encouraging others to move
out of their glass houses, once I have moved out of mine. In my 1968 book, Political Order in
Changing Societies, which otherwise builds extensively on the 1965 article, the concept of political
development was quietly dropped. I focus instead on what I conceive to be the critical relationship
between political participation and political institutionalization without worrying, about the issue of
which should be labeled “political development.”
43. See Levy, Modernization and the Structure of Societies, and Leonard Binder, Iran: Political
Development in a Changing Society (Berkeley and Los Anéeles, 1962).
44. Easton, Political System, pp. 266-67, 272 ft\; Harold Lasswell and Abraham Kaplan, Power and
Society (New Haven, 1950), p. xiv.
45. Moore, “Social Change and Comparative Studies.” pp. 524-25.
46. Karl W. Deutsch, “Social Mobilization and Political Development,” American Political Science
Review, LV (September 1961), 493-514. For a suggestive effort to relate Almond, Huntington, and
SSRC Comparative Politics Committee theories of political development to the available quantitative
data, see Raymond F. Hopkins, “Aggregate Data and the Study of Political Development,” Journal of
Politics, XXXI (February 1969), 71-94.
47. Ibid.; James C. Davies, ‘Toward a Theory of Revolution,” American Sociological Review, XXVII
(February 1962), 5 ff.; Ivo Κ. and Rosalind L. Feiera-bend, “Aggressive Behaviors within Polities,
1948-1962: A Cross-National Study,” Journal of Conflict Resolution, X (September 1966), 253-54;
Ted Gurr, Why Men Rebel (Princeton, 1970).
48. Barrington Moore, Jr., Social Origins of Dictatorship and Democracy (Boston, 1966); Cyril E.
Black, The Dynamics of Modernization: A Study in Comparative History (New York, 1966).
49. Rustow, World of Nations, p. 36. For a thoughtful discussion of sequences in political
development, see Eric A. Nordlinger, “Political Development: Time Sequences and Rates of
Change,” World Politics, XX (April 1968), 494-520.
50. See Pye. Aspects, pp. 62-67.
51. Lasswell and Kaplan, Power and Society, p. xv.
52. Huntington, Politicai Order, p. 55.
53. William C. Mitchell, The American Polity (New York, 1962), pp. 369-70.
54. Major contributions to the analysis of power by contemporary social scientists include: Lasswell
and Kaplan, Power and Society, pp. 74 ff.; Herbert Simon, “Notes on the Observation and
Measurement of Political Power,” Journal of Politics, XV (November 1953), 500-16; James G.
March, “An Introduction to the Theory and Measurement of Influence,” American Political Science
Review, XLIX (June 1955), 431-51; Robert A. Dahl. “The Concept of Power,” Behavioral Science, II
(July 1957), 201-15; Carl J. Friedrich, Man and His Government (New York, 1963), pp. 159-79;
Talcott Parsons, “On the Concept of Influence,” Public Opinion Quarterly, XXVI (Spring 1963).
55. See Samuel P. Huntington, The Soldier and the State (Cambridge, 1957), pp. 80-97 and passim.
56. Gabriel A. Almond, “Determinacy-Choice, Stability-Change: Some Thoughts on a Contemporary
Polemic in Political Theory,” (Center for Advanced Study in the Behavioral Sciences, Stanford
University, August 1969).
57. For an initial application of the Almond model, see Wayne A. Cornelius, Jr., “Crisis, Coalition-
Building, and Political Entrepreneurship in the Mexican Revolution: The Politics of Social Reform
under Lázaro Cárdenas,” (Project on Historical Crises and Political Development, Department of
Political Science, Stanford University, July 1969).
58. Dankwart A. Rustow, “Change as the Theme of Political Science,” pp. 6-8. See also his
“Communism and Change,” in Chalmers Johnson, ed. Change in Communist Systems (Stanford,
1970), pp. 343-58, and “Transitions to Democracy: Toward a Dynamic Model,” Comparative
Politics, II (April 1970), 337-63.
59. Ronald D. Brunner and Garry D. Brewer, Organized Complexity: Empirical Theories of Political
Development (New York, 1971).
60. Dankwart A. Rustow, “Modernization and Comparative Politics: Prospects in Research and
Theory,” Comparative Politics, I (October 1968), 51.
*This essay was written while I was a Fellow of the Center for Advanced Study in the Behavioral
Sciences in Palo Alto. It will subsequently appear in a volume edited by Daniel Bell, Theories of
Social Change, copyright by The Russell Sage Foundation. I am indebted to Ronald D. Brunner and
Raymond F. Hopkins for helpful comments.
The Attitudes of Modernity
Kenneth S. Sherrill*
The aim of this article is to distill and operationalize some of the significant
aspects of theories of political modernization within the context of the study
of American politics. These fields of study are applicable to one another in
two areas. First, operating under the assumption that the United States is a
modern nation, we might isolate those salient features of American politics
that are absent from political systems that are, by common consent, not yet
modern systems. We might then study the way in which the United States
has developed and maintained these characteristics of modernity, and we
might then make certain prescriptive remarks about strategies to be followed
in the modernization process. This is a highly ethnocentric approach to the
study of modernization, but this type of ethnocentrism, which is explicit and
conscious, may be less wrong than the latent ethnocentrism that pervades
much of the work in the field of political modernization.1
The assumption that the United States is modern permits another approach
to the analysis of the problems of political modernization. Were we to derive
a description of a modern polity from the areas of consensus in the political
development literature, we might then test the accuracy with which the
description is mirrored in American politics. If parts (or all) of the
descriptions were inaccurate pictures of the United States, we might then
reject them as descriptions of a generalized modern political system.
The second area in which the study of political modernization is
applicable to the study of American politics is a more explicitly normative
one. Instead of testing the accuracy of a description of “the modern” by
seeing whether or not it is an accurate description of the United States, we
might test the proposition that the United States is modern by comparing it
with our model of modernity. If we believe that modernity is a desirable state
of being, then we might argue for certain changes in the nature of American
politics.
Thus, it becomes clear that the way in which we go about defining
modernity determines the types of questions we ask and, hence, the types of
conclusions we reach about modernity. Furthermore, the way in which we
define the modern will raise ethical issues concerning competing end states.
Many definitions of modernity have been offered in the past decade, and the
editors of a recent anthology were forced to write: “In this volume ... we
have not tried to adhere to a rigid definition of political development. ... The
authors of the papers in this volume thus tend to reflect the view that there is,
in a generic sense, a phenomenon of political development and
modernization.”2
If political scientists continue to operate on the assumption that our major
dependent variable exists in nothing more precise than a generic sense, we
may end up exacerbating the problems that have piqued our curiosity.
Without engaging in a lengthy review of definitions of political development
or modernization, let me suggest that we develop a parsimonious scheme for
arraying these definitions in terms of two basic dimensions.
First, definitions of political modernity tend to deal with one or both of
two classes of phenomena. Phenomena in the first class are structural. They
include such things as literacy rates or the existence of a rational
bureaucracy and a party system within the framework of institutional
stability and efficiency. Phenomena in the second class are attitudinal. We
are here concerned with the orientations of the mass public as well as of the
elites toward their role in politics, toward themselves, and toward their
fellow men.
The second dimension that serves to distinguish conceptions of the
modern is the way in which the concept is given a formal definition. The
traditional approach had been to assign “ideal type” definitions to those
characteristics that serve to distinguish the modern societies from the
traditional ones. Max Weber writes, “The construction of a purely rational
course of action in such cases serves the sociologist as a type (‘ideal type’)
which has the merit of clear understandability and the lack of ambiguity.”3
This kind of definition introduces a large normative element. An actor’s
efforts can be evaluated in terms of how well he conforms to the ideal
pattern of action. Similarly we should be able to evaluate a nation in terms of
how well it conforms to our ideal pattern of political modernity.
Although compliance with an ideal pattern may be possible in an ideal
world, political scientists frequently find insurmountable barriers existing in
the real world. This has led to an alternative approach to defining our major
variable. Writing in the context of democratic theory, Robert A. Dahl sets
forth “the descriptive method.” This “is to consider as a single class of
phenomena all those nation states and social organizations that are
commonly called democratic by political scientists, and by examining the
members of this class to discover, first, the distinguishing characteristics
they have in common, and second, the necessary and sufficient conditions
for social organizations possessing these characteristics.”4
The descriptive approach will lead us to the observation that not all people
in a modern society conform to the ideal type of the modern citizen. Are we
then to reject the ideal-type model as being unrealistic or undesirable? Or,
worse, are we to confuse the two approaches and declare that the real is the
ideal? If we were to fall into either of these methodological traps we would
lose our criteria for evaluating modern societies. The ultimate absurdity
would be to ask, How much does a nation measure up to itself?
We may now codify our classificatory scheme for definitions of the
modern. First, we ask if the definition is concerned with one or both of our
major classes—structural and attitudinal. If it is concerned with only one of
them, which one is it, and what are the implications of concentration on only
one of them? Second, is the definition of each variable ideal-type or
descriptive?
Max Weber, for example, is concerned with both the structural and
attitudinal dimensions, and he provides ideal-type definitions of both sets.5
Lucian Pye utilizes an ideal-type definition of the structural characteristics of
a modern political system and gives its attitudinal dimension a descriptive
definition.6 Almond and Verba emphasize the attitudinal components of the
civic culture, but provide descriptive definitions of both major dimensions.7
Peter Snow measures political development exclusively in terms of the
emergence of a rational institutional framework for politics.8 Leonard Doob,
at the other extreme, focuses on the attitudinal changes that come with
becoming more civilized.9 This article will be concerned almost exclusively
with the attitudinal dimension, and the political attitudes of modern men will
be defined in terms of ideal types.
The purpose of developing this classificatory scheme is not to argue that
one type of definition is more or less valid than another, but rather to
demonstrate the range of variation in our concerns and to enable us to begin
to distinguish the consequences for research and policy of these differing
types of definitions. Certainly David Apter’s use of ideal types results in
conclusions markedly different from those of Almond and Verba. While
many have challenged Almond and Verba’s work as having a marked
conservative bias,10 this charge can scarcely be leveled at Apter, who writes:
“. .. How can the West serve as the model for a highly industrialized society
if it . . . ignores many of its obligations in the fields of civil rights, poverty,
education, and foreign affairs? ... The future of democratic society will
depend on its ability to find new and effective ways to secure personal
identity through liberty and solidarity through knowledge. This has always
been the basis of the democratic ideal. These are the ultimate standards by
which we evaluate both ourselves and others.”11
The descriptive approach reveals that no society is purely modern and that
all societies combine the modern and the traditional. Changes in the mixture
of components frequently occur; what is modern to one generation will
appear traditional to their children. Sidney Verba notes that “what seem
today to be fundamental sets of beliefs may be quickly set aside.”12
This should make us recognize that confounding the real with the ideal is
a most dangerous procedure when studying modernization. The study of
political modernization is inevitably filled with normative issues. To put the
matter bluntly, there is general agreement that being modern is a good thing.
Fred Riggs writes, “To be modern is to be up-to-date, contemporary.
Normatively speaking, it is ‘better’ to be modern than out of date. Hence,
‘modernization’ conveys the notion of moving toward a preferred
condition.”13 A descriptive model inevitably places its imprimatur on the
nations on which the model is based, just as Almond and Verba place theirs
on the United States and the United Kingdom.
The remainder of this article will be devoted to outlining the political
attitudes of modern men in a democratic political system. It will be
specifically concerned with those attitudes believed to be supportive of a
democratic system. It will not contend that only democracies are modern,
but rather than democracy is the type of modern system we prefer and
toward which we are interested in developing. It focuses on the attitudes of
modern men rather than on structural variables, recognizing that both sets of
variables are significant and probably highly interrelated. There is ample
justification for emphasizing the attitudinal component. As one scholar has
put it, “The land, not the people, is the constant factor.”14 Almond and Verba
concur: “What is to be learned about democracy is a matter of affect and
feeling....”15
We generally assume that there are certain types of attitudes that can be
used to distinguish among individual societies, as well as among larger
classes of societies, such as the modern and the traditional or the democratic
and the authoritarian. In describing a society, we generally assume that all of
its members share these central attitudes. We invariably find that a society is
characterized by a distribution of attitudes that can be described in terms of
its mode (or a number of its modes) and the degree of dispersion about the
mode.16
Modern people share a peculiar world view. Their political attitudes are
integrated into a more general set of orientations toward the outside world
and toward their role in it to form a latent political ideology17 unlike a
forensic ideology such as that described by the authors of The American
Voter,18 which is the base of political debate within a society. A latent
political ideology characterizes the beliefs of society as a whole.
We shall now turn to an exposition of what many political scientists
believe to be the component of this latent ideology. A summary of the
political modernization literature reveals great consensus about these
attitudinal components.
As an initial disclaimer, it should be emphasized that we are engaged in a
review of the literature for the purpose of developing a consensus list of
hypotheses that might then be subjected to verification in the current state of
American politics. The fact that we state a hypothesis in no way means that
we necessarily agree with it or are unwilling to reject it when faced with the
weight of evidence. Whenever possible, each of these hypotheses will be
followed by some recently collected survey data that should serve as a
preliminary check on the validity of the statement.
1. The politically modern man identifies with the national political community.
There are two basic reasons why this proposition is necessary. First, it is a
guaranty of honesty and equality in politics. As long as authoritative
decisions reflect friendship or kinship patterns instead of rational
applications of specific criteria, politics does not meet the expectations of a
modern society.
The second reason for the maintenance of the distinction between personal
and political relationships is that this distinction lowers the intensity of
political involvement and therefore facilitates the processes of change.
When, as in traditional societies, political relationships are largely
determined by social and political relationships, “the political struggle tends
to revolve around issues of prestige, influence and personalities, and not
around questions of alternative courses of policy action. . . . Any change in
political identification generally requires a change in one’s social and
personal relationships.”25 In the modern society, on the other hand, it is
assumed that such considerations are not made. Political disagreement is
tolerated because the people expect distinctions to be drawn between
institutions such as politics, the family, religion, and economics. This is a
consequence not only of expecting functional specificity, but also of
becoming increasingly skillful in abstracting from the immediate situation to
a more general case.26
As was the case with the more general problem of national identification,
this is a particularly difficult concept to measure. If Pye’s hypothesis is
correct, modern repondents will justify their political preferences in terms of
some sort of forensic ideology, in terms of positions on certain issues that the
respondents deem to be significant, in terms of economic motivation, or in
terms of a Downsian loyalty to one or the other of the political parties. They
will not justify their political choices on the basis that one of the alternatives
is better for, or traditionally allied with, an ethnic, religious, geographic, or
kinship unit.
The data presented in Table 2 provide no support for this hypothesis.
Overwhelmingly, Americans use both personal and political criteria when
evaluating candidates for the Presidency. Furthermore, the ability to
distinguish between the personal and the political in evaluating political
objects is not significantly correlated with any of the other attitudes we shall
identify with modernity. If we have any faith in our measures, we must reject
this hypothesis.
The notion that the politically modern man has, and expresses, political
opinions should follow directly from the first two hypotheses. Having no
opinion may well be a consequence of not having an appropriate reference
group in a given area.32 Not having political opinions can therefore follow
from the absence of national identity. The willingness to express opinions
also indicates a strong ego—an image of the self as worthy of expressing
opinions and an image of society as respecting this worthiness.33
People living in traditional societies lack the skills needed to have
political opinions, just as they lack the other psychic prerequisites for being
opinionated. When faced with an issue that has never before confronted
them in their everyday life, they lack the knowledge needed to make a
decision, even if they feel worthy of expressing opinions. Traditional man
believes that his opinions are undesired or irrelevant. Daniel Lerner finds
that “persons who are urban, literate, participant, and empathic differ from
persons who lack any of these attributes”—and they differ on a significant
personal trait distinctive of the modern style. Such a trait is “having
opinions” on public matter. Traditional man has habitually regarded public
matters as none of his business. For the modern man in a participant society,
on the contrary, such matters are fraught with interest and importance. A
public opinion grows out of the expectation that many people will have
views on public issues; it is sustained by the reciprocal view that what
people think about such issues will make a difference in their solution.34
We can make two major inferences from this. First, the way of life in
traditional society prevents the development of political opinions among a
wide sector of the population, people who either believe that it is not right
for them to hold opinions or are incapable of holding or expressing opinions
because they lack the skills—or the motivation to obtain the skills —
involved in being opinionated. Second, when only a few people are involved
in the exchange of political ideas, they can monopolize the political process.
Oligarchic rule becomes inevitable when the mass of the people cannot
express their ideas about the major issues the country is facing. If democracy
is a system of government responsive to the will of the people being ruled,
democratic government is only possible in systems in which the people have
opinions, that is, in modern systems. (This should not be interpreted to mean
that all modern systems are democratic systems. No single variable is a
sufficient cause of democratic government. The point is that having an
opinionated public is a necessary condition of democracy. If having opinions
serves to distinguish modern people from traditional, then being modern is a
prerequisite to being truly democratic.)
We have devised a simple measure of opinionatedness by counting the
total number of political opinions that a respondent can voice if he answers
every question he is asked in a survey. (Unless otherwise noted, all data are
drawn from the 1964 Survey Research Center Election Study. See footnote
31.) We then subtract from that number the total number of times that a
respondent was unwilling or unable to voice a political opinion. We then
divide the maximum number of possible opinions by the remainder. The
quotient is our index of opinionatedness, which equals unity if the
respondent expresses the maximum number of opinions and equals zero if he
expresses no opinions. That is,
Our findings in this area are most gratifying and are presented in Table 4.
The idea that the modern political man has opinions is related to a fourth
hypothesis, essentially stating that he has the wherewithal to have opinions,
and that this trait is significant in and of itself.
4. The politically modern man is empathic.
“Are there any problems that the government in Washington has gotten into that you think it should
stay out of?”
Marginal distributions, however, do not tell the entire story. We find that
our measure of empathy is strongly related to other measures of politically
modernity, such as feelings of political efficacy and political involvement,
thus supporting Lerner’s emphasis on the importance of empathy.
Furthermore, there is a strong relationship betwen race and income and
the ability to name issues, i.e., to be politically empathic. The average white
respondent can name 2.5 issues; the average Negro respondent can name 1.5
issues. (Difference in means test is significant beyond the .0005 level.)
When family income falls below $3,000 a year, a respondent has great
difficulty naming more than two issues; respondents with family incomes of
over $15,000 a year average 3.2 issues. We now have more evidence in
support of the argument for a “syndrome” of being modern, together with
preliminary evidence that the characteristics of being politically modern are
not randomly distributed throughout the population. Rather than having
slack resources, Negroes and poor people are cumulatively disadvantaged. If
people are not aware of major issues, they cannot mobilize to protect their
interests. (It is interesting in this context that only 57.5 percent of the Negro
respondents could name an issue other than race relations—and this is about
as salient as an issue can get—and 29.6 percent could name no issue at all.
Only 11.6 percent of the whites were unable to name even one.) Obviously,
the people who are most empathic are at an advantage. These people are the
well-to-do and the white.
In order to imagine other people in other places, a person has to be
exposed to information. There are two basic sources of information:
interpersonal communication, which is a universal phenomenon, and the
mass media, which are more characteristic of the modern world than of
traditional societies. We must therefore state two corollaries to the
hypothesis that the politically modern man is empathic:
4a. The politically modern man is highly exposed to the mass media.
As a consequence:
The mass media expose people to a vast, and vicarious, universe that
would otherwise be beyond the experience of the average man. Furthermore,
the media provide enough homogeneity in this exposure to permit discussion
about this world of vicarious experiences. In the absence of mass
communication, public opinion becomes segmented. National politics
becomes difficult, and central mobilization to meet national crises is
frustrated.
The media perform a “status conferral” function: by attributing to persons,
events, and ideas the quality of being legitimate centers of attention,38 the
media in the new nations can legitimize new norms and focus public
attention on the goals of modernization by seeming to be a voice of
authority.39 Thus the media can, in the hands of skilled leaders, develop a
sense of political identity among the villagers, propel new leaders to
prominence, and endow ideas with the blessings of the nation and the leaders
as well as of the media.
The use of the mass media as a source of prestige suggestion facilities a
switch from a particularistic world view to a more universalistic one. The
traditional man learns to respect and value the opinions of a “generalized
other,” or public opinion. Respecting the views of other people lays the
foundation for a system in which public opinion, rather than traditional
leaders and institutions, determines the goals to be pursued and the rules of
the game. This, it is argued, gives the system greater flexibility in
responding to changed situations and in formulating new demands. It also
provides the basis for “market morality” in which the same price buys the
same quantity of material of the same quality for all prospective consumers.
What holds for economics should also hold in the political realm: in the
“market place of ideas, market morality implies equality and popular
sovereignty.”40
Thus, media exposure has two major consequences for the modernization
process. First, it gives people information and teaches them “to have
opinions on matters which are, strictly speaking, none of [their] business.”41
Second, it encourages “other-directedness,” which means that “the link
between the past of the individual and any subsequent behavior is attenuated
if not altogether obliterated.”42
Little need be said about the importance of information to the politically
modern man. In order to participate in politics, people must have at least
rudimentary information about what is going on. They must come to believe
they comprehend the workings of society and must have some knowledge
about the alternatives to traditional life if they are to bring about change.
Almond and Verba argue that “democratic competence is closely related to
having valid information about political issues and processes and to the
ability to use information in the analysis of issues and the devising of
alternative strategies.”43
The first part of this hypothesis receives ready confirmation. Over 95
percent of the 1964 SRC sample reported following the election campaign in
one or more of the mass media (with over half naming television as being
the most important, and about a fifth choosing the newspapers). Slightly
over half reported attending to three or four of the media.
A simple information test can be constructed from the 1964 Election
Study. The respondents were asked to identify the home states and religions
of Goldwater and Johnson, the party controlling Congress, the local
candidates for Congress, and the political party controlling Cuba and China.
Although these questions were not difficult, Table 6 shows a reasonable
range in information scores.
The politically modern man holds an optimistic attitude toward his own
potentialities and toward the characteristics of his fellow men. He is
concerned with the need to achieve, and he is characterized by faith in
people. In his pathbreaking book The Achieving Society, David McClelland
finds that the achievement motive (n Achievement), or concern with the
need to achieve, is highly correlated with subsequent economic growth, but
largely unrelated to previous levels of development. Thus, he concludes, the
stress placed on achievement in children’s readers, folk tales, and popular
literature probably represents national aspirations, or the tendency of the
public at large to think about achievement for the next generation. In fact,
motivational change precedes economic change by thirty to fifty years.44
There is general agreement that the modern man is more sanguine about
the future than the traditional man is. Kurt Back measures “modernism” by
asking his respondents to compare the present generation of youth to
previous generations. He finds that optimistic responses are related to self-
confidence, belief in the future, faith in people, belief in the worth of
education and of personal effort, and creativity and ambition.45 Everett E.
Hagen contends that innovative personalities are characterized by strong
concerns with the needs to achieve, to be autonomous, and to have warm
interpersonal relationships (when the need to be dependent on others is
low).46 Robert E. Ward believes that a modern society is characterized by the
ability to control and influence its circumstances and by optimism about the
possible use of this potential.47 In general, optimism is taken to be
supportive of participation in political dialogue. The authors of The
American Voter write: “Persons with a relatively pessimistic outlook tend to
be persons who feel ineffective in their own political activity, they
experience a lack of competence in guiding their own lives, and they attempt
to avoid conflict or disagreement in their relationships with other people.”48
Attempts to measure levels of optimism have not been highly successful.
However, optimism is no doubt implied by the corollary trust in mankind.
5a. The politically modern man is characterized by a general faith in people.
While some may contend that the questions were too rigorous to be an
accurate measure of efficacy feelings, we cannot help but notice that over
two-thirds of the respondents felt that politics and government sometimes
were too complicated for “a person like me” to understand what’s going on.
Over a third felt that public officials don’t “care much what people like me
think.” That these people are among the less politically modern Americans is
indicated by the fact that those scoring 3 or more on the measure of empathy
also scored above the mean on the efficacy index.
10. The politically modern man is characterized by a general faith in the government.
There is also some evidence that belief in electoral control is related to our
measure of empathy. More than half of the respondents who rank lowest on
the faith-in-elections scale cannot name more than one issue that the new
government should take care of. Once again, we have some evidence that the
political attitudes attributed to modern men are systematically interrelated in
a syndrome, or latent ideology, of being modern.
1. Leading sources of the “anti-ethnocentric” critique are Fred R. von der Mehden, Politics of the
Developing Nations (Englewood Cliffs, 1964), p. 3; Fred W. Riggs, Administration in the Developing
Countries: The Theory of Prismatic Society (Boston, 1964), p. 35; Ann Ruth Willner, “The
Underdeveloped Study of Political Development,” World Politics, XVI (April 1964), 471; Robert A.
Packenham, “Political Development Doctrines in the American Foreign Aid Program,” ibid., XVIII
(January 1966), 194-195; J. Roland Pennock, “Political Development, Political Systems, and Political
Goods,” ibid. (April 1966), 415-434; Gabriel A. Almond and James S. Coleman, eds. The Politics of
the Developing Areas (Princeton, 1960), Introduction, p. 3; Lucian W. Pye, Aspects of Political
Development (Boston, 1966), p. 34.
2. Lucían W. Pye and Sidney Verba, eds. Political Culture and Political Development (Princeton,
1965), pp. 12-13.
3. The Theory of Social and Economic Organization, ed. and trans. Talcott Parsons (New York, 1964),
p. 92.
4. A Preface to Democratic Theory (Chicago, 1956), p. 63.
5. Pages 328-331 ff.
6. Pye, ed. Communications and Political Development (Princeton, 1963), p. 18.
7. Gabriel A. Almond and Sidney Verba, The Civic Culture (Princeton, 1963), pp. 31 ff.
8. “A Scalogram Analysis of Political Development,” American Behavioral Scientist, IX (March
1966), 33-36.
9. Becoming More Civilized: A Psychological Exploration (New Haven, 1960).
10. See, for example, Peter Bachrach, The Theory of Democratic Elitism: A Critique (Boston, 1960);
Christian Bay, “Politics and Pseudopolitics: A Critical Evaluation,” American Political Science
Review, LIX (March 1965), 39-51; Lewis Lipsitz, “If, as Verba Says, the State Functions as a Religion,
What Are We To Do Then To Save Our Souls?” ibid., LXII (June 1968), 527-535; and Kenneth S.
Sherrill, “Political Modernization as an Attitudinal Syndrome,” a paper presented at the Annual
Meeting, American Political Science Association, September 1966.
11. The Politics of Modernization (Chicago, 1965), p. 463.
12. “Conclusion: Comparative Political Culture,” in Pye and Verba, p. 521.
13. P. 38.
14. Hermann Weilenmann, “The Interlocking of Nation and Personality Structure,” in Karl W.
Deutsch and William Foltz, eds. Nation-building (New York, 1963), pp. 33-55, at p. 49.
15. P. 5. For other support of this position, see Everett E. Hagen, On the Theory of Social Change
(Homewood, 1962), p. 86; Daniel Lerner “Introduction [to Attitude Research in Modernizing Areas],”
Public Opinion Quarterly, XXII (Fall 1958), 217-222, at 217; Kurt W. Back, “The Change-prone
Person in Puerto Rico,” ibid , 330-340, at 330; Doob, p. 5.
16. This approach has been widely used in recent years. See, for example, Almond and Verba, pp. 12-
36; V. O. Key, Jr. Public Opinion and American Democracy (New York, 1961), pp. 27-93; Alex
Inkeles and Daniel Levinson, “National Character: The Study of Modal Personality and Social-
cultural Systems,” in Gardner Lindsey, ed. Handbook of Social Psychology, Vol. II (Cambridge, Mass.,
1954); and Robert E. Lane and David V. Sears, Public Opinion (Englewood Cliffs, 1964), pp. 83-113.
17. Robert E. Lane, Political Ideology (New York, 1962), p. 3. This notion of latent ideology has
much in common with Gabriel Almond’s original conception of political culture: “Every political
system is embedded in a particular pattern of orientations toward political action. I have found it
useful to refer to this as political culture.” Almond, “Comparative Political Systems,” Journal of
Politics, XVIII (August 1956), 391409, at 396. Philip E. Converse introduces a similar concept in
“The Nature of Belief Systems in Mass Publics,” in David Apter, ed. Ideology and Discontent (New
York, 1964), pp. 206-261.
18. Angus Campbell et al. The American Voter (New York, 1960), pp. 192-193.
19. Lucian Pye is most frequently associated with the concept of national identity in recent years. See
his Politics, Personality, and Nation-building: Burma’s Search for Identity (New Haven, 1962), pp. 4-
6. Pye is by no means alone in this position. Other major sources are Sidney Verba, in Pye and Verba,
pp. 529-532; Rupert Emerson, From Empire to Nation (Boston, 1962), pp. 102 ff.; Richard Butwell,
“Individual and Collective Identity and Nation-building,” World Politics, XV (April 1963), 488-494;
Max F. Millikan and Donald L. M. Blackmer, eds. The Emerging Nations (Boston, 1961), p. 76;
Robert E. Lane, “The Tense Citizen and the Casual Patriot: Role Confusion in American Politics,”
Journal of Politics, XXVII (November 1965), 735-760; Lane, Political Ideology, pp. Ill ff.; Samuel P!
Huntington, “Political Development and Political Decay,” World Politics, XVII (April 1965), 386-430,
at 387; James S. Coleman, “Conclusion: The Political Systems of Developing Areas,” in Almond and
Coleman, p. 532; Daniel Lerner, “A Note on Nationalism and Political Identity,” Pubh’c Opinion
Quarterly, XX (Spring 1956), 289-292; Leonard Doob, “South Tyrol: An Introduction to the
Psychological Syndrome of Nationalism,” ibid., XXVI (Summer 1962), 172-184. On the relationship
between primary groups and national identity in the United States, see Morton Grodzins, The Loyal
and the Disloyal (Chicago, 1956).
20. Karl W. Deutsch, “Social Mobilization and Political Development,” American Political Science
Review, LV (September 1961), 493-514, at 500.
21. Politics, Personality, and Nation-building, pp. 255-266.
22. In addition to Pye, Politics, Personality, and Nation-building, see Lane, Political Ideology, p. 389;
Erik Erikson, “The Problem of Ego Identity,” Journal of the American Psychoanalytic Association, IV
(1956); Campbell et al., pp. 120-145; Theodore M. New-comb, “Attitude Development as a Function
of Reference Groups: The Bennington Study,” in Eleanor Maccoby et al., eds. Readings in Social
Psychology, 3d ed. (New York, 1958), pp. 265-275.
23. An analogous argument is made by Sidney Verba in his very interesting article “The Kennedy
Assassination and the Nature of Political Commitment,” in Bradley S. Greenberg and Ε. B. Parker,
eds. The Kennedy Assassination and the American Public (Stanford, 1965), pp. 348-360.
24. Karl W. Deutsch, “Integration and the Social System,” in Philip E. Jacob and James V. Toscano,
The Integration of Political Communities (Philadelphia, 1964), p. 202; Pye, Politics, Personality, and
Nation-building, pp. 16-17; Millikan and Blackmer, p. 19; Verba, in Pye and Verba, p. 549; Doob, pp.
34, 187; David C. McClelland, The Achieving Society (Princeton, 1961), p. 173; Bert F. Hoselitz,
Sociological Aspects of Economic Growth (New York, 1959), p. 33.
25. Pye, Politics, Personality, and Nation-building, pp. 16-17.
26. Doob, pp. 34, 87. In a content analysis of children’s stories, McClelland finds (p. 173) that in
rapidly developing societies, contractual relationships and material and impersonal cooperation
pressures are more frequent than in less rapidly developing societies, and that the opposite is true for
traditional institutional interaction pressure.
27. The major exponent of this position has been Harold D. Lasswell. See his Power and Personality
(New York, 1948), pp. 162 ff., and his “Democratic Character,” in The Political Writings of Harold D.
Lasswell (New York, 1951), pp. 465-525, esp. pp. 480 ff. Lasswell traces this idea back to Aristotle
and Plato (ibid., p. 468). See also Lane, Political Ideology, pp. 43 ff.; Morris Rosenberg, “Self-Esteem
and Concern With Public Affairs,” Public Opinion Quarterly, XXVI (Summer 1962), 201-211.
28. Political Ideology, p. 54.
29. “Democratic Character,” p. 521.
30. Everett Hagen, in On the Theory of Social Change, pp. 92 ff., develops this idea to a much greater
extent. See also Kurt Back, “The Change-prone Person.”
31. The questions in these and subsequent scales are not included for lack of space. I will make the
data for these scales available upon request. Utilizing the system of trichotomizing variables
demonstrated by Allen L. Edwards, Techniques of Attitude Scale Construction (New York, 1956), pp.
184-197, the CR (coefficient of reproducibility) for the first scale is .849 (minimum marginal
reproducibility = .620), and for the second scale it is .829 (MMR = .607). The normal CR of .900 for
Guttman scales refers to dichotomized variables rather than the trichotomizations used here, which
save information.
The data used for these scales, which were provided by the Inter-University Consortium for
Political Research, are from the 1964 Election Study conducted by the University of Michigan. I
would like to thank the ICPR, the University of Michigan Survey Research Center, and the
Department of Political Science at the University of North Carolina for making these data available. I
would also like to thank Mr. Neil Ludlam of the UNC Computation Center and Mr. Peter B. Harkins
of the UNC Department of Political Science for their invaluable aid with some of the massive
computational aspects of the larger project from which this article is drawn. They are, of course, in no
way responsible for the quality of the analysis of the data.
32. Eugene Hartley, ‘The Social Psychology of Opinion Formation/’ Public Opinion Quarterly, XIV,
No. 4 (1950), 673.
33. Lane, Political Ideology, p. 387.
34. Daniel Lerner, Passing of Traditional Society, pp. 70-71, 99.
35. Passing of Traditional Society, pp. 49-50.
36. Ibid., pp. 49 ff.
37. Ibid., pp. 96-100.
38. Robert K. Merton and Paul F. Lazarsfeld, “Mass Communication, Popular Taste, and Organized
Social Action,” in Bernard Rosenberg and D. M. White, eds. Mass Culture (Glencoe, 1957), pp. 461-
462.
39. McClelland, Achieving Society, p. 193. This position is widely held. See Lerner, Passing of
Traditional Society, pp. 64-68; Millikan and Blackmer, p. 20; Wilbur Schramm, “Communication
Development in the Development Process,” in Pye. Communications and Political Development, pp.
30-57, esp. pp. 38-53. In general, see this Pye volume and Schramm’s Mass Media and National
Development (Stanford, 1964), passim.
40. McClelland, pp. 193-195. See also Lane and Sears, pp. 84-93; Everett M. Rogers, “Mass Media
Exposure and Modernization Among Colombian Peasants,” Public Opinion Quarterly, XXIX (Winter
1965), 614-625.
41. Lerner, “Introduction,” 221.
42. Lasswell, “Democratic Character,” p. 487. The more positive value placed on other-directedness
may come as a slight surprise to readers of David Riesman’s The Lonely Crowd (New Haven, 1950).
43. P. 95. Many authors stress the importance of information to modernization. See Schramm, Mass
Media, p. 246; Doob, Becoming Civilized, p. 29; and McClelland, p. 399. Υ. B. Damle uses information
levels as his operational measure of modernization in “Communication on Modern Ideas and
Knowledge in Indian Villages,” Public Opinion Quarterly, XX (Spring 1956), 257-270.
44. p. 138. For other evidence in support of the argument presented thus far, see pp. 67 ff.
45. «The Change-prone Person.”
46. -How Economic Growth Begins,” Public Opinion Quarterly, XXII (Fall 1958), 375390, at 377.
47. “Political Modernization and Political Culture in Japan,” World Politics, XV (July 1963), 569-596,
at 570.
48. Campbell et al., p. 396. They also find (p. 397) that in the United States, Northerners are more
optimistic than Southerners, and that optimistic Southerners participate more than do pessimistic
Southerners.
49. P. 192. Once again, taking the United States as a model of modernity, we might generalize from
Florence Kluckhohn’s discussion of the American “core culture” (which includes the belief that man is
capable of “rational mastery over nature” and that human nature is “evil, but perfectable”) to the
position that modern man is characterized by optimistic attitudes toward other men. “Dominant and
Substitute Profiles of Cultural Orientations,” Social Forces, XXVIII (1950), 376-393, at 382.
50. “Democratic Character,” p. 495. This position is very similar to the one adopted by Lane in
Political Ideology, pp. 405-406. Lane cites G. M. Gilbert, The Psychology of Dictatorship (New York,
1950), pp. 278-280, in support of the position that the “constricted empathy” of the Germans enabled
them to endure their oppression of others without feeling much guilt. Thus, empathic feelings and
faith in man may be much more characteristic of open systems than they are of modern systems—or
else Germany is simply a deviant case.
51. Morris Rosenberg, “Misanthropy and Political Ideology,” American Sociological Review, XXI
(December 1956), 690-695.
52. See fn. 31. CR = .844; MMR = .581.
53. McClelland, p. 160.
54. Ibid., pp. 169-170. Furthermore, when η Affiliation is high, high η Power does not lead to
authoritarianism. McClelland cites the United States in 1950 as an example of such a society (p. 170).
One of the problems with content analysis is that it works with arbitrarily determined boundaries. We
therefore have difficulty in making inferences from the characteristics of the aggregate to those of the
individual. We can only wonder about the motivational patterns of McCarthyites.
55. P. 4. Among those emphasizing the importance of participation in modernization, see Frederick W.
Frey, ‘‘Political Development, Power, and Communications in Turkey,” in Pye, Communications and
Political Development, p. 301; Lasswell, “Democratic Character,” pp. 473 ff.; Lerner, “Introduction,”
220; Lerner, Passing of Traditional Society, p. 50; Millikan and Blackmer, pp. 41-42; Deutsch, “Social
Mobilization,” 500; Edward Shils, Political Development in the New States (The Hague, 1962), p. 16,
cited in Willner, “The Underdeveloped Study,” 471; Huntington, “Political Development and Political
Decay,” 388; Herbert Hyman, “Mass Media and Political Socialization,” in Pye, Communications and
Political Development, p. 132; Lasswell, “The Policy Sciences^ of Development,” World Politics,
XVII (January 1965), 290; S. N. Eisenstadt, “Modernization and the Conditions of Sustained Growth,”
World Politics, XVI (July 1964), 576-594, at 590; Karl W. Deutsch, Nationalism and Social
Communication, 2d ed. (Cambridge, Mass., 1966), p. 81.
56. Passing of Traditional Society, p. 68.
57. Hyman, “Mass Media,” p. 128.
58. See Millikan and Blackmer, pp. 41-42.
59. Verba, in Pye and Verba, pp. 557-558.
60. Almond and Verba, p. 247.
61. Lane, Political Life (New York, 1959), p. 344.
62. See D. R. Matthews and J. W. Prothro, Negroes and the New Southern Politics (New York, 1967).
The CR for Table 9 is .986.
63. Campbell et al, pp. 96-107.
64. Susanne Hoeber Rudolph, “Consensus and Conflict in Indian Politics,” World Politics, XIII (April
1961), 385-399, at 385, 389.
65. “Modernization and the Conditions of Sustained Growth,” p. 584.
66. “Democratic Character,” pp. 497-498.
67. See Kenneth S. Sherrill, “Political Modernization and the United States,” unpubl. diss., University
of North Carolina, 1967, pp. 129-131, for a discussion of the measurement and distribution of political
concern.
68. Civic Culture, p. 123.
69. Ibid., p. 132.
70. Ibid., pp. 151-152.
71. Politics, Personality, and Nation-building, p. 17. This position is probably highly inaccurate. See
Robert A. Dahl, ed. Political Oppositions in Western Democracies (New Haven, 1966), passim.
72. This position is developed at some length in Lane, Political Life, pp. 149 ff. See also Key, Public
Opinion, pp. 192-195.
73. Lerner, Passing of Traditional Society, p. 100.
74. p. 490.
75. Pye, Aspects of Political Development, pp. 63-64; Pye, in Pye and Verba, p. 22.
76. CR = .742;MMR = .445.
*I would like to thank Robert L. Hardgrave, Timothy M. Hennessey, and Nancy A. Shilling for their
comments on earlier versions of this manuscript. Of course, all blame for error is mine.
The Problem of Identity, Selfhood,
and Image in the New Nations
The Situation in Africa
Rupert Emerson*
Thirty nine states with more to come, and far more than thirty-nine peoples
—all in search of an identity in Africa.1
The continent in the last seven or eight decades has undergone such
revolutionary turns of fortune, almost wholly thrust upon it from outside,
that it would be an utter miracle if the processes of change and adaptation
were not still in full swing, with little more than question marks available as
to the outcome in a variety of important spheres. Even though it is possible
to say that much traditional life still goes on in rural Africa and that even in
the urban centers more of the old survives than the outsider can be aware of,
fundamental transformations are under way throughout the continent. The
point of no return has long since been reached. To cite a single statistic: in
1955 the entire continent contained only five independent states, and even
at the beginning of 1960 only five more colonies had joined the ranks of the
free.
The sharp massive jolt of colonialism was followed at a speed which
almost no one expected by the global wave of anticolonialism. Freedom
came in some instances to peoples who had barely demonstrated an active
interest in independence and in many instances to peoples ill-prepared for it
in terms of achieving an equal place in the highly complex contemporary
world. The result inevitably has been that for many Africans both their
sense of identity and their image of the society in which they live are
uncertain and unstable. It may well be that they are endowed, for good or
ill, with a wider and more open choice of identities than has in any normal
circumstance been available to other peoples; but then the circumstances of
Africa are far from normal.
Self-Determination
For the Third World and its partisans, self-determination has become the
bright and shining sword of our day, freeing peoples from the bondage of
empire. Yet the sword must be recognized as double-sided—sharp and
cutting on one side of its blade but blunt and unserviceable on the other.
Two intimately interrelated questions habitually appear as soon as the
issue of self-determination is posed: Who constitutes the self to be
determined, and by what means and by whom is the determination to be
undertaken? The range of possibilities under both headings is large, and the
choices made may drastically affect the outcome. At the extremes, self-
determination may either encompass the whole of an assumed
macro-”national” territory, giving no heed to minorities which may have
quite different alignments in mind, or it may operate at the level of small
and separable local entities such as villages and urban precincts, always
considering that the end result must be a reasonably coherent territory or
territories. The means range from the assumption, not left open to question
or empirical testing, that the populace of the entire territory wants what its
leaders say it wants, to provision for a plebiscite that will pose whatever
questions seem in order in the given situation—the fate of the whole
territory as a single indivisible block, its division among adjoining
territories, or the secession of a piece or pieces of it either as independent
states or as parts of neighboring states whose people may be ethnically
related.
Whatever political theory may say about a properly democratic society,
the trend since World War I has been to eliminate the plebiscite and either
to take for granted the “national” allegiance and commitment of all the
people involved or to assume that they have been adequately consulted in a
general election which, directly or indirectly, raised the question of self-
determination as one of the issues confronting the electorate. The disfavor
with which plebiscites tend to be viewed derives not only from the
disorders and political tensions which they are likely to arouse, but also
from the conviction of nationalist leaders that no consultation of the people
is necessary or even appropriate. To consult the people is to cast doubt upon
the valid claim of the leaders to represent the general will and, in a different
vein, to invite dissident groups to spread subversive and probably tribalist
doctrines that are inspired, so it may be charged, by the neocolonialists.
In the recent political evolution of Africa in the era of decolonization, a
simple formula has been worked out to apply in all but a few special cases,
notably those involving trust territories of the United Nations. This formula
consists of three basic propositions: (1) that all colonial peoples want to get
out from under colonialism as speedily as possible, (2) that each colonial
territory as established by the imperial powers constitutes a nation whose
aspirations to become an independent state have unchallengeable validity,
and (3) that the political independence and territorial integrity of the states
thus created must be safeguarded against attack from within or without.
The first of these propositions has a sufficient ring of authenticity to
make dubious any serious critical examination, although the footnotes that
might appropriately be added are by no means without interest. Certainly it
is beyond argument that in the main colonial peoples have repudiated
colonialism and welcomed independence; but it is more open to question
how deeply the African masses had internalized the demand for Uhuru or
Freedom Now by the time the colonial regime had withdrawn. The answer
is perhaps to be found less in the existence of an all-embracing positive
demand for freedom than in the readiness of the masses to follow their own
nationalist leadership when it overtly clashed with the expatriate colonial
authorities. In a number of the African colonies, and most strikingly in the
case of several of the French territories, even this readiness was not
effectively tested, since independence was granted far less because of a
successful local mobilization of nationalist forces than because of a general
withdrawal from colonial overlordship.2
Among those Africans who opposed independence or sought to have its
coming delayed, three major categories can be distinguished. The category
presumably of least consequence was composed of the few who had made a
profitable peace with colonialism. Often displayed by the colonial regime as
representatives of the people—under British auspices, for example,
Africans were named as unofficial members of Legislative Councils—they
were almost swept into the discard as the rising nationalist forces took over.
The two other categories, far more important both in numbers and in
political weight, consisted of tribes or other ethnic groups which felt
themselves seriously disadvantaged by the coming of independence at the
time and under the circumstances which were proposed. One of these was
made up of the smaller peoples, or those less favored politically, such as the
Ashanti and the people of the Northern Territories in Ghana and a number
of the tribes in Kenya and Congo (Kinshasa). The other, not wholly
distinguishable from the first, comprised peoples who found that they had
fallen behind in the process of modernization and were therefore less able
than their rivals to hold their own in the management of a modern society.
Here the most striking example is that of the Northern Region of Nigeria
where the Muslim emirates, imbued with a deep sense of their own
superiority, found themselves no match for the southerners, particularly the
Ibos, who had beaten them to a mastery of the ways and skills of the
modern world. Although a different order of issues was involved, we might
also mention the determined resistance to an independence in Kenya and
Southern Rhodesia that implied the subjection of the African majority to the
white settler minority: here continued control by the Colonial Office
seemed obviously preferable to domination by the local Europeans.
The second of the three propositions composing the African or, more
broadly, the anticolonial formula for self-determination requires no
considerable commentary here, both because it is a familiar theme and
because several phases of it are considered elsewhere in this article. The
heart of the matter is that, whatever the shortcomings of the colonial
boundaries that were laid down in the scramble for Africa, and however
little the people they enclosed constituted nations in any real sense, if
Africans were to seek the political kingdom, the only instruments readily at
hand were the existing colonial regimes which could be taken over
substantially intact; indeed, in the first round, by far the easiest answer was
that Africans should replace Europeans in the top management of the
colonies turned into independent states. Here was where political power
rested and here was in principle at least the embryonic paraphernalia needed
to maintain a functioning modern society. The problems of nation-building,
of establishing or reestablishing a variety of regional organizations, and, for
some the most significant aim, the achievement of some form of pan-
Africanism—all these could wait until the colonial regimes had been
replaced by African rule. The right of the people composing these colonial-
states to self-determination, interpreted as independence, was the cardinal
tenet of the doctrine, wholly beyond dispute.
The third and last of the propositions, attempting to secure agreement
that the existing boundaries were somehow sacred and must be preserved
by common consent and action, raises points of evident difficulty. The
sword of self-determination is sharp when severing the colony from its
metropole; however, its reverse side is blunt and unavailable when
minorities within the former colony seek either their own independence or
union with more desirable brethren across the frontier. Secession from
empire is not only praiseworthy but virtually obligatory, since colonialism
has been declared intolerable and illegitimate; secession from a newly
established state is banned as a breach of the fundamental proprieties on
which the African polities rest. To act otherwise, it is held, would be to
destroy nations whose identity and right to survive are, for all practical
purposes, established by the fact of their existence. The centrality of this
third proposition is evidenced by the provisions of the Charter of the
Organization of African Unity which, in the Preamble, the Purposes of
Article II, and the Principles of Article III, call for the safeguarding, the
defense of, and respect for the sovereignty and territorial integrity of its
member states.
I have contended elsewhere,3 and it deserves prompt and emphatic
repetition here, that this approach to self-determination in no significant
way distinguishes the African position from that universally adopted by
states. It is difficult, and indeed wholly impossible, to conceive any set of
circumstances under which states would allow their domains to be
fragmented by conceding a general principle that any random segment of
their population had an operative right to secede, taking with it any territory
it regarded as its homeland. In the shaping of the United Nations Charter, it
was specified that the principle of self-determination “conformed to the
purposes of the Charter only insofar as it implied the right of self-
government of peoples and not the right of secession.”4 Under special
conditions a limited and quite precisely defined right of self-determination
may be temporarily accepted, as in the present drive to rid the world of
colonialism; but even here the continued existence of British, Portuguese,
and American colonies, presumably still within the law, makes it more than
dubious that any operative right to secede from empire has been entered on
the international statute books.
The one element that makes the African position somewhat distinctive is
that, to a substantially greater degree than in other major parts of the world,
its states have not yet achieved an adequate degree of integration and inner
coherence. The threat of potential breakup on ethnic lines that would signal
a resurgence of tribalism and irrevocably shatter the existing scheme of
things is a more real and present danger than is normally the case
elsewhere. The issue was succinctly put by President Nyerere of Tanzania
in remarking that, despite differences in language, size, and resources, the
founding conference of the OAU in 1963 could still agree that “our
boundaries are so absurd that they must be regarded as sacrosanct.”5
Boundaries in other continents may be less absurd, but they are no less
guarded against the dangers of secession under cover of the symbols of self-
determination.
As I have suggested earlier, the only outstanding breaches of the
principle that each colony should “nationally” self-determine itself into
independent statehood occurred, for reasons that are at least partly obscure,
in the case of UN trust territories. Thus British Togoland joined Ghana; the
Northern Cameroons joined Northern Nigeria, while the remainder of the
British and French Cameroons joined to form the federal state of
Cameroun; Rwanda and Burundi, to the dismay of some anticolonialists in
the UN, separated from each other to form two independent states; British
Somaliland and the trust territory of Somalia joined to form the Somali
Republic; and, at a slightly later stage. Zanzibar joined the ex-trust territory
of Tanganyika to form Tanzania. In this latter case particularly, the
connection with the UN appears quite fortuitous; but the entire array (which
might also include the UN decision that Eritrea should join Ethiopia) raises
the question of whether, had the UN played a greater role in Africa’s
decolonization, the outcome might have produced more changes in the
political map of the continent than actually took place.
The key issues involved in the translation of the doctrine of self-
determination into African political reality find no more dramatic
illustration than the tragic decline of Nigeria into, first, political turmoil and
military coups, followed by large-scale bloodshed and, finally, a disastrous
civil war. It is typical of the structure of the African state system that this
most populous African country owes its existence to decisions that were
essentially arbitrary. The British authorities might equally have followed
quite different courses with very different results. If, for example, they had
either drawn Nigeria into much closer association with their other West
African holdings or had divided it into three or eight or a dozen separate
colonies— as the French did with the considerably smaller populations of
West and Equatorial Africa—neither policy would have seemed unusual or
in any way outrageous. Indeed, some critics contend that the root cause of
Nigeria’s troubles today was the misguided British effort to combine
markedly disparate peoples within the same political framework; others
take the reverse position that Britain encouraged Nigeria’s disintegration
both by following divergent policies in the North and in the South and by
introducing a federal rather than a unitary system of government.
In its later phases, the political evolution of colonial Nigeria centered on
the development of self-government in each of the three Regions, with the
North lagging behind the East and West, and in 1960 came the grant of
independence to the country as a whole. In 1966 the military took over in
two successive coups, each of which had strong tribal-political implications.
Following the slaughter of thousands of Ibos in the North, the flight of
perhaps two million more, and lengthy negotiations that brought no
mutually acceptable solution, the Eastern Region, under Ibo leadership, on
May 30, 1967, seceded from Nigeria to become the independent state of
Biafra and thereby opened the door to civil war.
Whatever one’s judgment as to the rights and wrongs of the case and the
assignment of responsibility for the breach, one matter which is
indisputably clear is that the Ibos, who unquestionably constitute a
“people,” have by force of arms been denied the right of self-determination
which they claimed for themselves.6 The ability of the OAU to mediate the
conflict and help bring it to an end was impaired by its commitment to the
territorial status quo. The dragging on of the war and the growing loss of
life finally impelled Tanzania and, shortly thereafter, three other African
states to break the ban on Biafra by extending recognition to it. In an
eloquent statement explaining this action, Nyerere reasserted his country’s
devotion to African unity but insisted that such unity must rest on freely
given consent: if the Zanzibaris were to repudiate their union with
Tanganyika, that union would then cease to exist and he could not advocate
that the dissidents be bombed into submission. The Biafrans, he held,
claimed the right to govern themselves when they were unable to achieve a
satisfactory agreement ensuring their personal security:
The Biafrans are not claiming the right to govern anyone else. They have not said that they must
govern the Federation as the only way of protecting themselves. They have simply withdrawn
their consent to the system under which they used to be governed.
Biafra is not now operating under the control of a democratic government any more than
Nigeria is. But the mass support for the establishment and defense of Biafra is obvious. This is not
a case of a few leaders declaring secession for their own glory.…
It seemed to us that by refusing to recognize the existence of Biafra we were tacitly supporting
a war against the people of Eastern Nigeria—and a war conducted in the name of unity. We could
not continue doing this any longer.7
Leaving aside the basic political issue raised by Nyerere, a lesser but by
no means insignificant point concerns his comment that the Biafrans are not
claiming the right to govern anyone else. Here again is one of the central
problems of self-determination: Who constitutes the self? It might equally
be said that the Nigerians are not claiming to govern anyone but Nigerians;
yet who are the Nigerians for this purpose? If they are all the inhabitants of
the country, making up a single nation, then Biafrans have no right to be
heard as Biafrans. Precisely the same proposition applies to Biafra itself,
where there are very substantial minority tribes (some five million non-Ibos
as against seven million Ibos) whose allegiance to the dominant Ibos
certainly cannot be taken for granted—and who, incidentally, inhabit the
area where the most valuable oil resources have been found. If the whole
former Eastern Region, restyled Biafra by the Ibo secessionists, is the
proper unit for self-determination, then the minorities may properly be
subordinated to the majority and swept along in its wake. If, on the other
hand, these minorities are assumed to constitute “peoples,” are their claims
to be heard less valid than those of the Ibos? The hope of the minorities for
consideration as separate communities was markedly enhanced by the
Federal decree of May 27, 1967, that divided the Eastern Region into three
states, one for the Ibos and two for the other peoples of the Region8—a
decree that led immediately to Biafra’s declaration of independence. The
Federal government asserts that it “has a moral responsibility to the
5,000,000 non-Ibo speaking minorities in the Rivers State and the Calabar-
Ogoja State whose areas cannot be handed back to the secessionist leaders.
These minorities deserve a place of their own under the Nigerian sun.”9
Should it be established that minority Efiks, Ibibios, Ijaws, and others
have withdrawn (or perhaps never gave) their consent to being part of an
Ibo-ruled Biafra, are they then also entitled to secede on the basis of
Nyerere’s admirable principle of consent freely given, which he applied to
Biafra as a single entity? Need it be added that undoubtedly the two new
states, subtracted from Biafra in the interest of these minorities, also contain
minorities who would be pleased to have autonomous or even independent
jurisdiction over their own affairs? It is clear that at some point even the
most ardent enthusiast for a permanent and continuing right of self-
determination must call a halt to the process of fragmentation.
This general moral, so familiar to African statesmen, was pointed out by
Chief Enahoro in his opening address, on behalf of the Federal authorities,
to the abortive peace talks held at Kampala in May 1968: “Many African
countries which today appear to condone or rejoice at the misfortune of
Nigerians may tomorrow find themselves fighting their fellow countrymen
and the forces of secession in their midst.”10
1. This article deals not with all thirty-nine African states but primarily with those that make up
independent black Africa.
2. John A. Ballard asserts that prior to the forcing of the independence issue by the Mali Federation
in December 1959, the four colonies that made up French Equatorial Africa had presented no request
for independence, and even after that event their leaders gave no clear indication that independence
was either necessary or desirable; but as other countries came to independence, including Cameroun
and the Belgian Congo, “the AEF governments were swept along.” “Four Equatorial States,” in
Gwendolen M. Carter, ed. National Unity and Regionalism in Eight African States (Ithaca, 1966), p.
240.
3. See my “Self-Determination Revisited in the Era of Decolonization,” Occasional Paper No. 9,
Center for International Affairs, Harvard University, December 1964.
4. Documents of the United Nations Conference on International Organization, V, (United Nations
Information Office, New York, 1945), 296.
5. The Dag Hammarsiqold Memorial Lecture, January 23, 1964, p. 7.
6. After a visit to Biafra, Conor Cruise O’Brien reported in The Observer (London) of October 8,
1967: “It is possible that the state of Biafra may be crushed out of existence by the numerically
superior and better-armed Federal troops, with their limitless access to outside support . . . What is
certain, however, is that a nation has been born and will in some form endure.” For an examination of
a number of the issues raised by the Nigerian conflict, see S. K. Panter-Brick, “The Right to Self-
Determination: Its Application to Nigeria,” International Avoirs, XLIV (April 1968), 254-266.
7. The Los Angeles Times, May 5, 1968.
8. At the same time, the previously dominant Northern Region was divided into six states, making a
total of twelve for the whole of Nigeria.
9. See the two-page advertisement of the Federal government in The New York Times, July 8, 1968.
10. The New York Times, July 8, 1968.
11. “African Problems and the Cold War,” Foreign Affairs, XL (October, 1961), 51. As one of a
number of speakers on the same theme at the OAU founding conference in Addis Ababa in 1963,
President Modibo Keita of Mali held that those who ardently desire African unity must take Africa as
it is: “African unity demands of each one of us complete respect for the legacy that we have received
from the colonial system, that is to say: maintenance of the present frontiers of our respective states.”
Proceedings of the Summit Conference of Independent African States, I (Addis Ababa, May 1963),
Sec. 2.
12. John Day, “Democracy in Africa,” Parliamentary Affairs, XVII (Spring 1964), 168.
13. For an account which generally exonerates the French from having acted with evil intent, see
James Carl Akins, ‘Trench Policy in Afrique Noire, 1955-1960” (unpub. Ph.D. diss., Harvard
University, 1966).
14. Africa: The Politics of Independence (New York, 1961), p. 85.
15. On Heroes and Uhuru-Worship (London, 1967), p. 70.
16. “Congo and Uganda: A Comparative Assessment,” Cahiers économiques et sociaux, V (October
1967), 381.
“Almost everywhere, the transfer of power from a colonial regime to a locally based regime has
intensified the parochial or tribal basis of already existing groups, or has activated latent
parochialisms and led to the formation and proliferation of new com-munalistic political
associations.” James S. Coleman and Carl G. Rosberg, Jr. Political Parties and National Integration
in Tropical Africa (Berkeley and Los Angeles, 1964), p. 689.
17. See Paul Amber, “Modernization and Political Disintegration: Nigeria and the Ibos,” Journal of
Modern African Studies, V, No. 2 (1967), 68-169.
18. “Nationalism in Africa, 1945-1965,” The Review of Politics, XXVIII (April 1966), 145. Among
others, Professor Curtin lists Ashanti, Mossi, Dahomey, Buganda, Ankole, Bunyoro, Burundi,
Rwanda, Bakongo, Barotse, Lunda, Zulu, Sotho, and Swazi nations.
19. See Panter-Brick, “The Right to Self-Determination,” 261.
20. The word Africa does not figure in any early Bantu vocabulary; it is perhaps the most important
article of export which Europe ever sent south of the Sahara.” L. H. Gann and Peter Duignan, Burden
of Empire (New York, 1966), p. 88.
21. Cited in Victor C. Ferkiss, Africa’s Search for Identity (New York, 1966), p. 130. For a skeptical
and entertaining examination of questions of African identity, see Ali A. Mazrui, Towards a Pax
Africana (London, 1967), esp. Chs. 1-3.
22. “The Idea of Freedom in African National Movements,” in David Bidney, ed. The Concept of
Freedom in Anthropoiogy (New York, 1963), pp. 183-184.
23. Kofi A. Busia, The Challenge of Africa (London, 1962), p. 139.
24. See Immanuel Wallerstein, Africa: The Politics of Unity (New York, 1967).
25. African Nationalism (Capetown, 1959), p. 98.
*This article was prepared for delivery at the Annual Meeting of The American Political Science
Association, Washington, D.C., September 1968. I am much indebted to the Center for International
Affairs, Harvard University, for research support and aid in its preparation.
PART II
Introduction
In analyzing Third World States, it is a basic and obvious point that the
pattern of politics and the structure and functioning of society are deeply, if
not extricably interwoven. Not surprisingly, the most widely discussed
factor of the social order with broad implications for Third World politics is
the existence of cultural pluralism, the severe fragmentation of a polity
along sub-national “identity” lines.
While the fragmentation issue (examined in the preceding section) is a
recurrent one throughout this volume and is of unquestionable importance,
it is not the only aspect of the interconnection of politics and the social
order which need be explored. In short, there remain additional dimensions
of the behavioral linkage between politics and the social order in Third
World countries which draw our attention.
In part, this linkage can be investigated through the perspective of
“political culture.” This perspective, closely connected to the “behavioral
persuasion” in political science, is based on the observation that “every
political system is embedded in a particular pattern of orientation to
political actions.”1 A political culture, therefore:
[I]s the set of attitudes, beliefs, and sentiments which give order and meaning to a political
process and which provide the underlying assumptions and rules that govern behavior in the
political system .... [It is] the manifestation in aggregate form of the psychological and subjective
dimension of politics.2
Although the political culture approach implies that the realm of the
political is to some degree distinguishable from the general culture,3 it also
evokes a series of relationships which move from the psychological basis of
a social order to the spillover of that order into the political arena. It is
several aspects of this serial relationship, rather than the general and elusive
concept of political culture, that provide the focal points for this section.
Our first area of inquiry revolves around the question of clientelism; a
question of great consequence for the predominately rural societies of the
Third World. Lemarchand and Legg, in the opening article of this section,
speak of political clientelism as a “personalized, affective, and reciprocal
relationship between actors, or sets of actors, commanding unequal
resources and involving mutually beneficial transactions that have political
ramifications beyond the immediate sphere of dyadic relationships.”
Lemarchand and Legg regard cli-entelism “as a generic trait” of political
entities and their typology of clientage systems, which includes the
normative orientations of actors as well as the penetration and scope of
political structures, indicates the interactions existent among individual
orientations, the ordering of society, and the pattern of politics within a
particular reference frame.
Arguing that the study of clientage (or patron-client) relations has tended
to ignore political implications, Lemarchand and Legg attempt to articulate
the role of clientelism within the context of “political development” and the
concern with modernization, political integration, and system equilibrium.
Of particular importance, from our perspective, is Lemarchand and Legg’s
analysis of the impact of clientelism in “patrimonial-modernizing regimes”
(viz, the countries of the Third World), including their view that the very
successes of a clientelist model in promoting economic and social
developments ultimately leads to the disintegration of that model. And
though they conclude on a note of pessimism for the future of clientelism in
many developing situations Lemarchand and Legg leave open the
possibility that clientelism can “contribute to breakthroughs in political
modernization ... by ushering in a breakdown of the preexistent system.”
The significance of clientelist politics and micro-level perspectives in
societies that are primarily agrarian is also touched upon by the second
essay in this section. In “Participation and Efficacy: Aspects of Peasant
Involvement in Political Mobilization,” Mathiason and Powell explore “one
face of the individual consequences of political change” through the use of
attitudinal survey data. This study of the peasantry4 in Colombia and
Venezuela focuses on the question “to what extent does participation affect
an individual orientation to government?” In other words, the authors ask if
participation (defined by Huntington and Nelson as “activity by private
citizens designed to influence governmental decision-making”5) by
peasants has, as one of its consequences, the development of a sense of
national political efficacy which “defines an individual’s orientation to
institutions in the political process.”
The analysis of this question by Mathiason and Powell and their
conclusions contain a number of valuable insights which range from the
recognition of type and subtype distinctions in the nature of efficacy, to the
“individual attitudinal consequences of being caught up in a political
revolution” and the relationship of ecological/economic factors to
politically relevant attitudes. Even without commenting on all of these
issues, we can see that Mathiason and Powell’s article offers a perceptive
exploration of the complex connections existent between the social order as
an action system and politics.
At this juncture, two points emerge for our consideration, the first of
which is related to our use of the phrase “action system”.6 Leaving aside the
numerous Parsonian intricacies of this idea, the action construct implies
“the structures and processes by which human beings form meaningful
intentions and, more or less successfully, implement them in concrete
situations.”7 Thus the social order as a term is used to signify the broad
field of patterned inter actions aggregated within the societal unit of the
state. In this sense, we can view the ‘“multivariate relationship”8 existing
between the social order and the analytically distinguishable realm of
politics in the following way: SOCIAL ORDER/ SO1 (including factors such
as personality, values, resources, and social and ecological structure)→-
POLITICS/P1 (embracing the structures and processes of decision-making,
policy-making, and resource and value dis-tribution)→SOCL4L
ORDER/SO2 (the effects of SO1↔P1) . . .
This leads us to our second point, which is that the SOPSO frame of
reference involves a dual directionality in the flow of influence between the
social order (including its personality dimension) and politics. While the
pattern of politics is affected by the overall social order, that pattern also
establishes its own effects on the ordering of society.
Both of the articles discussed provide illustrations of this “multivariate
relationship.” Lemarchand and Legg, for example, note that “[t]he forms
which [clientelism] takes depend to a considerable extent on the structure of
the society and on the political system in which it operates”; and Mathiason
and Powell point out that “[a] η explanation must be sought for the process
by which a person develops an orientation such as efficacy, and of the
system of behaviors which maintains, changes, or otherwise influences his
belief in his ability to influence his government.”
Returning to our earlier consideration of political culture, we can see that
while a political culture has importance for political actions, it is no less
true that political actions themselves have an impact on the nature of a
political culture. The serial relationship that we spoke of is neither
unidirectional nor static, but rather is characterized by dynamic and multi-
faceted interactions.
The concluding essay in this section, Weiner and Fields’s “India’s Urban
Constituencies” contributes an additional perspective to the examination of
the relationship between politics and the social order by exploring behavior
in the political realm in terms of an ecological dimension of the social
order. Although the Third World remains overwhelmingly rural, it
nonetheless contains more than one-third of the urban population of the
globe.9 The urban explosion in le Tiers Monde is expected to continue, and
with this explosion come the strains of change and transformation.
Like other complex phenomena that we have concerned ourselves with,
the exact meaning and significance of urbanization is subject to various
interpretations, each containing differing views on the impact of
urbanization on people and political actions. Weiner and Field, for example,
begin their article by outlining two views of “the political characteristics of
cities in low-income countries.” The first of these is a model of political
uniformity which suggests “a progression from politicization to
radicalization to polarization in urban political life, together with a tendency
for the city to be a diffusion center shaping the political character of the
rural region in which is it located.”
The second model, rather than “emphasizing the political imperatives
imposed by the attributes ... of urban life,” stresses “the common features of
cities” in a given context “with their rural surroundings.” This second
model implies “urban political diversity” wherein “political culture,
political organi zation, and processes of social change shape the city as they
do the rural environment.”
With these models in mind, the authors examine the voting behavior of
India’s large urban constituencies since 1952. The central theme that
emerges from their analysis is a bimodal one in which India’s large urban
constituencies are seen to be “similar to the rural constituencies of the states
in which they are located;” while, at the same time, the urban centers are
found to have “much in common with one another irrespective of their
location” in the country. Therefore, Weiner and Field conclude that the
“electoral trends from 1952 to 1972 suggests that the urban constituencies
are becoming more distinctive insofar as electoral protest and electoral
polarization are concerned, but less distinctive with regard to electoral
participation (if only because the rural areas have caught up).” In other
words, the appropriateness of either of the models discussed depends on the
form of behavior one wishes to examine. Moreover, they hold that: “If the
polarization and radicalization trends continue, it seems likely that urban
constituencies will command even more of the attention of the resources of
party and government leaders than they do at present.”
In this case, it may be argued that the 1975 Proclamation of Emergency
by the government of Prime Minister Gandhi rendered this analysis
meaningless. Such shortsightedness is not warranted on two counts. The
first of these, as Weiner and Field point out, is that the developments in
1975 do not stand in contradistinction to the implications underscored by
their article, but rather “have tended to confirm the trends described in this
study.” The second of these is that elections returned to India in 1977,
producing the first defeat of the Congress Party and the fall of Prime
Minister Gandhi. Thus insights into voting behavior as part of the complex
connections between politics and the social order not only help us in
explaining the past, but also provide assistance in that continual search for
an understanding of the present and the future.
NOTES
1. Gabriel Almond, “Comparative Political Systems,” in Almond, Political Development (Boston:
Little, Brown, 1970), p. 35.
2. Lucían Pye, “Political Culture,” International Encyclopedia of the Social Sciences, Vol. 12 (New
York: Macmillan, 1961), p. 218. Also see Lucían Pye and Sidney Verba (eds.) Political Culture and
Political Development (Princeton: Princeton University Press, 1963); and Samuel Beer, “Political
Culture,” in Beer, et. al., Patterns of Government (New York: Random House, 1973).
3. Lucían Pye, “Culture and Political Science: Problems in the Evaluation of the Concept of Political
Culture,” in Louis Schneider and Charles Bonjean (eds.), The Idea of Culture in the Social Sciences
(Cambridge: Cambridge University Press, 1973), p. 68.
4. On peasants, see Eric Wolf, Peasant Wars of the Twentieth Century (New York: Harper and Row,
1969), p. xiv.
5. Samuel Huntington and Joan Nelson, No Easy Choice: Political Participation in Developing
Countries (Cambridge: Harvard University Press, 1976), p. 4.
6. See Talcott Parsons and Edward Shills (eds.), Toward a General Theory of Action (Cambridge:
Harvard University Press, 1951).
7. Talcott Parson, Societies (Englewood Cliffs: Prentice-Hall, 1966), p. 5.
8. Richard Braungart, “Introduction,” in Braungart (ed.), Society and Politics (Englewood Cliffs:
Prentice-Hall, 1976), p. 8.
9. See Raanan Weitz (ed.), Urbanization and the Developing Countries (New York: Praeger, 1973).
Political Clientelism And Development
A Preliminary Analysis
All contacts among men rest on the schema of giving and returning the equivalence. . . . Social
equilibrium and cohesion do not exist without the reciprocity of service and return service. . . . Beyond
its first origin, all sociation rests on a relationship’s effect which survives the emergence of this
relationship.
Long before Simmel, classical historians have drawn our attention to the
persistence of “the schema of giving and returning the equivalence” at
various levels of human interaction. In one form or another, evidences of
personalized reciprocal relationships have been uncovered in a variety of
contemporary and historical settings, both Western and non-Western, and in
many different guises. Whether termed “patronage,” “machine pontics,” or
“political clientelism”—as it is here designated—this type of relationship
must indeed be regarded as a generic trait of political systems regardless of
their stages of development.
That the phenomenon should rarely, if ever, have been analyzed by
political scientists on a cross-national basis is therefore all the more
remarkable. Certainly, little systematic effort has been made to establish its
relevance to an understanding of the processes of change associated with the
growth—and decline—of political institutions.1 Our purpose in this article is
to place the concept of political clientelism within the mainstream of the
literature on developmental politics. Specifically, an attempt will be made to
incorporate the notion of clientelism into current theories of development in
the hope that it may provide a useful theoretical connection between micro-
and macro-level or state-centered analyses. With this end in view we shall
try, first, to provide a conceptual clarification of the notion of political
clientelism; second, to distinguish among different types of clientelism; and,
third, to discuss their relationship to processes of modernization and
integration in both developed and less developed polities.
Whether by default or misappropriation, the field of clientage relations
has tended to become the exclusive preserve of sociologists and
anthropologists concerned with small communities. In general, however, the
approaches and methods of sociologists investigating community power
have obscured the political implications, if not the very existence, of
clientage phenomena.2 If such charges can scarcely be leveled against
anthropologists, however much they may disagree among themselves over
definitions,3 professional boundaries continue to raise major obstacles in the
way of meaningful interdisciplinary dialogue. Between those who insist
upon equating clientelism with machine politics and those who use the term
to describe a “particular type of relationship ... not within a formal system of
government authority,”4 there is room for almost endless discussion and
disagreement.
Although differences in definition and terminology have undoubtedly
obscured the similarity of the phenomenon observed by scholars in the
various disciplines, the uncritical transposition of definitions from one
discipline to the other can easily lead to misinterpretations of the locus and
ramifications of clientelism in any given political system. In a recent
discussion of “models of politics,” Samuel Barnes asserts that the “clien-
telistic model” should be treated as an entity distinct from the “authoritarian,
democratic, and totalitarian species”—noting, however, that the paucity of
data available would make a general analysis of this fourth type “impossibly
difficult and incomplete.”5 There is, in fact, little evidence on which to base
this distinction, but if much of the literature on clientelism is admittedly
narrow-gauged and anthropologically oriented, the available data is
nonetheless sufficiently abundant and diversified to permit an interim
assessment of the place and role of clientelism in the context of comparative
politics.6
Patrimonial-Clientage Systems
Unlike feudalism, patrimonialism as a political system operates through a
specialized administrative officialdom appointed by and responsible to the
ruler.34 There is, nonetheless, an obvious parallel in the nature of the
relationships that linked state officials to the patrimonial ruler and the lord-
vassal relationship. As Weber’s discussion of the rise of patrimonialism
makes clear, both types of relationships were clientelistic in character: “The
community was transformed into a stratum of aids to the ruler and depended
upon him for maintenance through usufructs of land, office fees, income in
kind, salaries, and hence through prebends. The staff derived its legitimate
power in greatly varying stages of appropriation, infeudation, conferment
and appointment. As a rule this meant that princely prerogatives became
patrimonial in character....” 35 A major point of contrast between the
patrimonial patron-client relationship and its feudal counterpart is that the
former never acquired the same degree of formalization. Particularly
instructive in this connection is J. Russel Major’s discussion of the “new
feudalism” or “bastard feudalism” in Renaissance France: “The new
feudalism differed from the old in that the client did not render homage to
the patron and the patron did not provide the client with a fief or in most
instances with money payments at regular intervals. It was similar to the old
feudalism in that it was an honorable relation based on mutual loyalty and
interests.” 36
Perhaps an even more meaningful distinction is that which hinges on the
relationships between clientelism and political office accompanied by
prospects of prebends. In the feudal system the vassal-lord relationship was
both the expression and the cause of the political relationship arising
therefrom; in the patrimonial system appointment to office becomes the
essential precondition of the patron-client relationship. This in turn focuses
attention on yet a third distinguishing feature, namely, that the patrimonial
relationship is comparatively less personalized and hence more precarious
than that arising out of feudal homage.37 In one case fidelity of the client to
his patron is contingent upon conferment of office; in the other it is implicit
in the very nature of the clientelistic institution.
Patrimonialism has two possible forms. In one, the traditionalistic variant,
the patron-client relationship permeates the entire political system. The
infeudation of the staff to the ruler expresses itself in the form of reciprocal
loyalties based on expectations of mutual benefit. The differentiation of
official roles at the center does not imply differentiation of system
boundaries. Nor does the greater penetration and scope of government
structures (as compared with the feudal situation) really affect the
clientelistic nature of individual relations at the local level. The limited
distributive capacity of the system and the fact that resources are distributed
or squandered for purposes of political recruitment positively encourage the
maintenance of clientelistic relations both at the center and at the periphery
of the political system.
In the second, the modernizing patrimonial system, the greater rate and
extent of social mobilization lead to potential discontinuities in the net of
patron-client relationships, not only between different levels of
governmental authority but between different sectors as well. This situation
is typical of that category of systems somewhat inaccurately characterized
by Gabriel Almond and G. Bingham Powell, Jr., as “pre-mobilized,” in
which “large numbers of individuals have been urbanized, have become
literate, and have been exposed to differentiated economic enterprises.” 38
Precisely because change is restricted to specific sectors, mobilization is
uneven and a hiatus tends to develop between the clientelistic structures
inherited from the previous model and incipient “legal-rational” orientations
of leading members in the mobilized sector. Despite the growing
differentiation of the social, economic, and political spheres, the expansion
of bureaucratic structures, the emphasis on welfare values and secularization
—all reflective of responses to social mobilization-social change at the mass
level tends to lag far behind the structural changes’ effect at the center. Thus,
in its actual operation, the system constantly tends to swing back into the
clientelistic mold of its predecessor. The result is a hybrid situation in which
clientelism resuscitates itself in the traditionalistic interstices of the
modernizing polity.39
Traditionalistic and modernizing types can be distinguished along several
dimensions. Where the former tends to embrace the entire political system
with a consequent congruence of political relationships and patron-client ties
at every level of the government hierarchy, the latter is generally more
segmented and shifting. In the traditionalistic type, dependency is likely to
radiate from a single point in the political system (i.e., the ruler); in the
modernized variant, on the other hand, dependency relationships can
originate in a variety of sectors—the party, bureaucracy, trade unions, and
even from the more traditional oligarchs. In this case, the shifting character
of dependency relationships frequently reflects the existence of competition
among various sectors. Where the traditionalistic type involves the allocation
of political office in return for political service or allegiance, in the
modernizing variant the transactions are by no means confined to the
political sector.40 As the bases of patronage become more diversified, so
does the position of the patrons in society. This group comes to include not
only government officials but new men of status such as the “professionals”
and businessmen. More often than not, elections under these conditions are
Utile more than devices through which clienteles are given the opportunity
to register their loyalty to competing patrons through the vote. The
fundamental distinction, however, is between state structures which coincide
with, and replicate, the client-patron relationship with the state acting as the
major source of “prebends,” and state structures which are superimposed
upon patronage networks, with the latter acting as intermediary links
between rulers and ruled.
The dominant trend in modernizing-patrimonial systems is clearly in the
direction of a greater variety of clientage networks, a diversification of
transactions, and a greater precariousness in the incumbency of patron-client
roles. At the point where demands previously met through personalized
clientage networks are met predominantly through the economic system or
by universal governmental action, the stage of the industrial polity has been
reached.
1. Qualified exceptions to this include Sidney Tarrows discussion of clientelismo in the Italian
Mezzogiorno in his Peasant Communism in Southern Italy (New Haven, 1967), and John D. Powell,
“Peasant Society and Clientelist Politics,” American Political Science Review, LXIV (June 1970), 411-
25. The most ambitious attempt so far to analyze the political implications of clientelism has come
from an anthropologist; see Alex Weingrod, “Patrons, Patronage, and Political Parties,” Comparative
Studies in Society and History, X (July 1969), 376-400. For an illuminating discussion of selected
aspects of clientelism in the African context, see Elizabeth Colson, “Competence and Incompetence in
the Context of Independence,” Current Anthropology, VIII (February-April 1967), 92-111.
2. By and large students of community power have focused on policy decisions rather than on routine
relationships between individuals and power holders. See Robert Agger, Daniel Goldrich, and Bert
Swanson, The Rulers and the Ruled (New York, 1964), chap. 2.
3. Anthropologists have been less than unanimous in their definition of clientelism. Either the very
breadth of the definition offered precludes its operationalization— see, for example, Jeremy
Boissevain, “Patronage in Sicily,” Man, I (March 1966), 18, who characterizes patronage as “the
complex of relations between those who use their influence, social position, or some other attribute to
assist and protect others, and those whom they so help and protect”—or else its restrictiveness tends to
distort the nature of the phenomena involved, as in the phrase “contract of pastoral servitude” to
describe the clientage institution in prerevolutionary Rwanda. See Jacques J. Maquet, The Premise of
Inequality in Rwanda (London, 1961).
4. Weingrod, pp. 378-79.
5. Samuel Η. Barnes, “Mobilization and Political Conflict: A Theoretical Enquiry into the Structural
Bases of Consensus and Dissent” (Paper delivered at the Seventh World Congress of the International
Political Science Association, Brussels, 1967), p. 7.
6. In addition to the works mentioned above, passing reference must be made to the following
sources: Frederick G. Bailey, Politics and Social Change in Orissa in 1959 (Berkeley and Los
Angeles, 1963); Lloyd Fallers, Bantu Bureaucracy (Cambridge, 1958); George M. Foster, “The
Dyadic Contract in Tzintzunt-zan: Patron Client Relations,” American Anthropologist, LXV
(December 1963), 1280-94; Jean Grossholtz, Politics in the Philippines (Boston, 1966); Arnold
Hottinger, “Zuama in Historical Perspective,” in Leonard Binder, ed. Politics in Lebanon (New York,
1966), pp. 85-105; Keith Legg, Politics in Modern Greece (Stanford, 1969), chap. 3; Rene
Lemarchand, “Les Relations de Clientele comme moyen de contestation: Le cas du Rwanda,”
Civilisations, XVIII (4/1968), 55373; Lucy Mair, “Clientship in East Africa,” Cahiers d’Etudes
Africaines, Π (6/1961), 315-26; Lucy Mair, Primitive Government (Baltimore, 1965); Edward I.
Steinhart, “Vassal and Fief in Three Interlacustrine Kingdoms,” Cahiers tfÊtudes Africaines, VII
(4/1967), 606-24; Eric Wolf, “Kinship, Friendship and Patron-Client Relations,” in Michael Banton,
ed. The Social Anthropology of Complex Societies (New York, 1966).
7. Wolf, p. 16.
8. See Adrian Mayer, “Quasi-Groups in the Study of Complex Societies,” in Michael Banton, ed. The
Social Anthropology of Complex Societies (New York, 1966), p. 100.
9. One is also reminded of the concept of “gate-keeping” as the process by which wants are injected
into the political system and converted into demands. See David Easton, A Systems Analysis of
Political Life (New York, 1965), pp. 86 fl.
10. Jeremy Boissevain, “Patrons as Brokers” (Paper presented to the Conference on Patronage, Dutch
Sociological Association, Amersfoort, The Netherlands, June 6 and 7, 1969).
11. Wolf, p. 18.
12. Ibid.
13. Fredrik Barth, Models of Social Organization, Royal Anthropological Institute, Occasional Paper
No. 2 (1966), cited in Burton Benedict, “Family Firms and Economic Development,” Southwestern
Journal of Anthropology, XXIV (Spring 1968), 3.
14. John K. Campbell, Honour, Family and Patronage (Oxford, 1963), p. 260.
15. Centering his analysis on the notions of “capital” and “credit” (i.e., the actual communication
channels [a broker] controls, and what others think his capital to be), Boissevain argues that a broker’s
“ability to bring about communication rapidly and effectively is in large measure determined by credit,
and credit is usually dependent upon ability to bring about communication. It is a cumulative process.
... It is this flexibility in transaction which gives the broker his high credit: people can be led to think
all sorts of things about the nature of his network and his ability to manipulate it. With the patron they
come to know his limits more quickly.” Boissevain, “Patrons as Brokers,” pp. 5-6.
16. We do not suggest that all clientelistic situations will necessarily lend themselves to this type of
“rural problem-solving” device. In tropical Africa the exchange processes described by Powell operate
within a much more limiting set of parameters where ethnic, religious, and social factors combine to
circumscribe and define patron-client relationships. In these conditions the range of options available
to the elites at the center to foment change on the periphery is correspondingly restricted. See Rene
Lemarchand, “Political Clientelism and Ethnicity in Tropical Africa: Competing Solidarities in
Nation-Building” (Paper presented to the Harvard- MIT Joint Faculty Seminar on Political
Development, February 1970).
17. Alvin Gouldner, “The Norm of Reciprocity: A Preliminary Statement,” American Sociological
Review, XXV (April 1960), 175.
18. Mair, “Clientship in East Africa,” p. 315; see also her Primitive Government, p. 166.
19. Mair, “Clientship in East Africa,” p. 325.
20. Ibid.
21. Ibid.
22. Jacques J. Maquet, “Une hypothèse pour l’étude des féodalitiés africaines,” Cahiers d’Êtudes
Africaines, Π (6/1961), 292-315.
23. Karl Deutsch, “Social Mobilization and Political Development,’ American Political Science
Review, LV (September 1961), 494.
24. Gabriel Almond and G. Bingham Powell, Jr., Comparative Politics: A Developmental Approach
(Boston, 1966), p. 198. For a further elaboration of the concept of “distributive performance,” see
Gabriel Almond, “Political Development: Analytical and Normative Perspectives,” Comparative
Political Studies, I (January 1969), 463.
25. Easton, pp. 27-56. In this discussion the term “input” would include expectations, public opinion,
motivations, ideology, interests, and preferences, all of which are excluded by Easton from both
“inputs” and “demands.”
26. See Frederick G. Bailey, Stratagems and Spoils (New York, 1969), intro.
27. Italy provides examples of “mixed” clientelistic structure. For an analysis of the socioeconomic
roots of the geographical variants of the padrone-mezzadro relationship, see Sydel F. Silverman,
“Agricultural Organization, Social Structure, and Values in Italy: Amoral Familism Reconsidered,”
American Anthropologist, LXX (February 1968), 120. Silverman’s article suggests that the ethos of
“amoral familism” is a consequence rather than the basis of the social characteristics. For industrial
clientelism see Joseph LaPalombara, Interest Groups in Italian Politics (Princeton, 1964).
28. Rushton Coulborn, ed. Feudalism in History (Princeton, 1956), p. 4.
29. Max Weber, General Economic History, trans. Frank H. Knight (Glencoe, 1927), pp. 51-57.
30. Coulborn, p. 189.
31. Mare Bloch, Feudal Society (London, 1961), vol. 1, passim.
32. Indeed, if one is to subscribe to the argument advanced by Fallers and Lombard, feudal
relationships can only obtain among equals (i.e., among nobles). If so, only between serfs and vassals
can one discern something approximating the client-patron nexus; but then the question arises as to
whether the element of coercion does not tend to overshadow the force of contractual obligations
between them. In brief, if the lord-vassal relation might conceivably be regarded as too “equal” to fit
the requirements of a patron-client relation, that of vassal-serf would seem far too “unequal” to meet
these same requirements. See Fallers, Bantu Bureaucracy, and Jacques Lombard, “La vie politique
dans une ancienne société de type féodal: Les Bariba du Dahomey,” Cahiers d’Etudes Africaines I
(5/1960), 5-45.
33. See, for example, Steinhart, “Vassal and Fief in Three Interlacustrine Kingdoms,” and Bloch, vol.
1, passim.
34. On the characteristics of “patrimonialism” as distinct from both “feudalism” and “patriarchalism,”
see Almond and Powell, p. 44.
35. Cited in Hans H. Gerth and C. Wright Mills, eds. and trans., From Max Weber: Essays in
Sociology (New York, 1958), p. 297.
36. J. Russel Major, “The Crown and the Aristocracy in Renaissance France,” American Historical
Review, LX1X (April 1964), 635.
37. The point is undoubtedly overstressed by Maquet in his discussion of the client-chiefs of the
Busoga. See Maquet, “Une hypothèse pour l’étude des feudalities africaines,” p. 306.
38. Almond and Powell, p. 284.
39. See Fred Riggs, Administration in Developing Countries (Boston, 1964).
40. The modernizing type of dependency relationship is thus very similar to that which Tarrow
associates with clientelismo in the Italian Mezzogiorno. See Tarrow, p. 74.
41. This seems to be the point of Frederick G. Bailey, Strategems and Spoils.
42. Gerth and Mills, p. 299.
43. The model is elaborated in Almond and Powell, pp. 258-80.
44. See, for example, Ian Weinberg, The English Public Schools (New York, 1967), chaps. 2 and 6.
45. Chitoshi Yanaga, Big Business in Japanese Politics (New Haven, 1968), chap. 1.
46. See LaPalombara.
47. See, for example, Sidney Ploss, Conflict and Decision-Making in Soviet Russia (Princeton, 1965),
esp. appendix.
48. Quoted by James Scott, “Corruption, Machine Politics, and Political Change,” American Political
Science Review, LXIII (December 1969), 1156.
49. John P. Nettl, Political Mobilization (New York, 1967), p. 139.
50. See Joseph Schlesinger, Ambition and Politics (Chicago, 1966).
51. William Whyte, Street Corner Society (Chicago, 1955), pp. 210-11.
52. See T. Alexander Smith, “Toward a Comparative Theory of the Policy Process,” Comparative
Politics, I (July 1969), 498-515.
53. See Frank E. Myers, “Social Class and Political Change in Western Industrial Systems,”
Comparative Politics, II (April 1970), 405-12.
54. See Donald R. HalL Comparative Lobbying: The Power of Pressure (Tucson, 1969).
55. This, roughly, is the core of Fallers’ hypothesis that where patron-client ties exist in conjunction
with a hierarchical, centralized political structure, the incorporation of a Western-type civil service
structure occurs with less strain and instability than in segmentary ones. Fallers, Bantu Bureaucracy,
pp. 241-42.
56. Powell, “Peasant Society and Clientelist Politics,” p. 424.
57. Weingrod, p. 381.
58. Ibid., p. 382.
59. The role played by political machines in the United States in facilitating the integration of
immigrant populations into the political system is well known. See also Grossholtz, p. 124.
60. Maquet, The Premise of Inequality in Rwanda.
61. The influence of exogenous factors on the structure of clientelism is well illustrated in the
incorporation of immigrant political values into the structure of American local government. See, for
example, Richard Hofstader, The Age of Reform (New York, 1955), pp. 8-9. See also Edward Banfield
and James Q. Wilson, City Politics (Cambridge, 1963), p. 46. For a critical appraisal, see Raymond E.
Wolfinger and John Osgood Field, “Political Ethos and the Structure of City Government,” American
Political Science Review, LX (June 1966), 306-26.
62. “By itself clientage is unlikely to dissolve sectional boundaries wherever differences of culture,
religion or race are prominent.” Michael G. Smith, “Institutional and Political Conditions of
Pluralism,” in Leo Kuper and Michael G. Smith, eds. Pluralism in Africa (Berkeley and Los Angeles,
1969), p. 56.
63. See James S. Coleman, ed. Education and Political Development (Princeton, 1968), p. 15.
64. For a discussion of this phenomenon in the African context, see Rene Lemarchand, Rwanda and
Burundi (London, 1970).
65. Frank Sorauf, “The Silent Revolution in Patronage,” in Edward Banfield, ed. Urban Government
(New York, 1961), p. 309.
66. Ibid.
67. See Francine R. Frankel, “Democracy and Political Development: Perspectives from the Indian
Experience,” World Politics, XXI (April 1969), 448-69. One must note, parenthetically, that the
evidence adduced by Weingrod in support of his contention that “party-directed patronage” leads to
integration is largely contradicted by the findings of Francine Frankel and others.
68. Ibid., p. 456.
69. Ibid., p. 458.
70. Ibid., p. 464.
71. See Jesse F. Marquette, “Peasant Electoral Politics in the Philippines” (Masters thesis, Department
of Political Science, University of Florida, Gainesville, 1968), chap. 4.
72. For a further elaboration on this point, see Donal Cruise O’Brien, “The Murids of Senegal: The
Socio-Economic Structure of an Mamie Order” (Ph.D. diss., University of London, 1969).
73. See Legg.
74. Sidney Tarrow, “Political Dualism and Italian Communism,” American Political Science Review,
LXI (March 1967), 44.
75. Frankel, p. 468.
*Part of the information upon which this article is based has been drawn from a series of monographs
presented at the Colloquium on Clientelism organized by Professor Georges Balandier under the
auspices of the Groupe de Recherches en Anthropologic et Sociologie Politiques (GRASP) in Paris,
March 29 and 30, 1968.
Participation and Efficacy
Aspects of Peasant Involvement in Political Mobilization
Knowing that rural Colombia has in recent times been deeply penetrated
by various modernizing influences, such as roads, newspapers, and radios (in
our sample, 40 percent say they read a newspaper at least once a month, one-
half of those at least weekly, and 63 percent say they listen to the radio every
day), we might formulate another hypothesis, one congruent with much of
the literature on political modernization in traditional societies.25 Those
peasant respondents who scored positively on a measurement of political
efficacy toward national government are those who “participate” directly in
national affairs through exposure to the mass media. Once again, however,
none of these measurements of exposure to modern influences is related to
political efficacy to a statistically significant degree—although level of
education and newspaper reading approach the .05 level.
If political efficacy among our peasant respondents is related neither to
traditional and obvious forms of political participation nor to exposure to a
variety of modernizing influences, what is it related to? Who are our
efficacious peasants, and why? In brief, our efficacious peasants are the
smallholders in our sample, for land tenure status was the only independent
variable related to a sense of political efficacy toward national politics at the
.05 level or better.
We lack sufficient data to explain definitively why smallholders should be
disproportionately efficacious toward national government in Colombia.
What evidence we do have tends to support one of two rather
straightforward, common-sense hypotheses. The first might be called the
venal landlord hypothesis. Peasants who are smallholders living in
communities dominated by large landlords are more efficacious than tenants,
sharecroppers, day laborers, and squatters because they are relatively more
independent of the landlord and the agricultural resources he controls.26
They are better able to resist being drawn into the clientele system with its
attendant exactions, including involvement in large-scale violence on behalf
of landed elites in the name of party competition. Consistent with this
hypothesis, but indirect in its impact, is the finding (Table 4) that squatters
and day laborers are the least efficacious peasants; these categories of
peasants have historically been the most exploited—and the most resistant—
in their dealings with the landed elites.27
1. Oscar Lewis, Pedro Martinez: A Mexican Peasant and His Family (New York, 1967).
2. Gabriel A. Almond and Sidney Verba, The Civic Culture (Princeton, 1963), esp. pp. 414-28.
3. C. K. Yang, A Chinese Village in Early Communist Transition (Cambridge [Mass.], 1959).
4. Robert B. Lifton, Thought Reform and the Psychology of Totalism (New York, 1961); Paul Hiniker,
Chinese Attitudinal Reactions to Forced Compliance: A Cross Cultural Experiment in Cognitive
Dissonance (Cambridge [Mass.], Center for International Studies, Monograph No. C65/18, 1965);
Richard Solomon, “Communications Patterns and the Chinese Revolution,” The China Quarterly
XXXII (October/December 1967), 88-110.
5. The basic outlines of this study are reported in Frederick W. Frey, “Surveying Peasant Attitudes in
Turkey,” Public Opinion Quarterly, XXVII (Fall 1963), 335-55.
6. The most ambitious attempt to date is by Eric Wolf, Peasant Wars of the Twentieth Century (New
York, 1969). The data for this study are mostly drawn from anthropological studies.
7. This characterization of peasants is drawn from the mainstream of literature on the subject. Some of
the commentators include Charles Wagley and Marvin Harris, “A Typology of Latin American
Subcultures,” American Anthropologist, LVII (June 1955), 428-51; Charles Wagley, “The Peasant,” in
John J. Johnson, ed. Continuity and Change in Latin America (Stanford, 1964); Eric Wolf, “Types of
Latin American Peasantry,” American Anthropologist, LVII (June 1955), 452-71; Wolf, Peasants
(Englewood Cliffs, 1966); Everett Rogers, “Motivations, Values and Attitudes of Subsistence Farmers:
Toward a Subculture of Peasantry” (Paper delivered at the Agricultural Development Council Seminar
on Subsistence and Peasant Economics, Honolulu, March 1, 1965); Edward C. Banfield, The Moral
Basis of a Backward Society (Glencoe, 1958).
8. George Foster, “Peasant Society and the Image of Limited Good,” American Anthropologist,
LXVTI (April 1965), 293-315.
9. Cf. Lester B. Milbrath, Political Participation (Chicago, 1966), and Almond and Verba.
10. Political efficacy has been variously termed “subjective political competence” (by Almond and
Verba); “powerfulness” by Melvin Seeman, “Alienation, Membership, and Political Knowledge: A
Comparative Study,” Public Opinion Quarterly, XXX (Fall 1966), 353-67; and “political
effectiveness” by Robert Lane, Political Life: Why People Get Involved in Politics (Glencoe, 1959).
11. Angus Campbell, Gerald Gurin, and Warren Miller, The Voter Decides (Evanston, 1954), p. 187.
12. David Easton and Jack Dennis, “The Child’s Acquisition of Regime Norms: Political Efficacy,”
American Political Science Review, LXI (March 1967), esp. 25-27.
13. See Almond and Verba, esp. pp. 3-36.
14. We use only a few discreet measures of efficacy rather than the indices customarily used by other
scholars who have dealt with the attitude. We have what we consider to be compelling reasons for so
doing, though we are aware of the measurement problems which our procedure presents. For one
thing, there are difficulties in translating scale items across cultures. In the Colombia study we
attempted to translate the Campbell et al. efficacy index, but got extremely low inter-item correlations.
Similarly in Venezuela we attempted to use a variety of efficacy measures, again with an eye toward
index construction. Here, too, inter-item correlations were either extremely low or null, indicating that
our measures were not tapping the same attitude dimension.
Of all the measures attempted, that which we call “national efficacy” seemed to be the most
successful. First, the form of the measure is, “Suppose that the national government were considering
doing something which you considered harmful and unjust, what could you do to impede this?” This
is rather unambiguous in that it is almost a direct restatement of what is usually defined as efficacy.
Moreover, since the question involves a specific level of the political process, it may eliminate some
of the conceptual problems produced by broader definitions of efficacy. Unlike many of the items on
political efficacy indices in common use, the national efficacy measure is relatively less intrusive in
that it does not specify alternative responses and thus eliminates the difficulty that the question
response may not have previously occurred to the respondent. An efficacious response is simply
defined as any response which indicates some concrete action. These actions can be easily classified
as “direct” or “indirect” depending on the behavior suggested (e.g., writing one’s congressman is a
direct action, while talking to one’s precinct captain is indirect with respect to the object of the
problem).
With only very slight variations this specific measure has been used in a number of cultural
contexts. Almond and Verba have used it in the United States, United Kingdom, Germany, Italy, and
Mexico; Frey has used it in Turkey; and it was included in all of the Venezuelan surveys. There are
thus clear grounds for cross-national, as well as over-time comparisons. Finally, of all of the efficacy
items in our various surveys, the measure of “national efficacy” has the most consistent correlations
with other variables, especially those of the same type of form. These advantages outweigh any
disadvantage which might be imputed from the use of a single measure as our dependent variable.
15. Cf. German Guzman Campos, Orlando Fals Borda, and Eduardo Umaña Luna, La Violencia en
Colombia, 2 vols. (Bogotá, 1962, 1964); Vernon L. Flu-harty, Dance of the Millions (Pittsburgh,
1957); Richard S. Weinert, “Violence in Pre-Modern Societies: Rural Colombia,” American Political
Science Review, LX (June 1966), 340-47; and James L. Payne, Patterns of Conflict in Colombia (New
Haven, 1968).
16. John Duncan Powell, “Peasant Society and Clientelist Politics,” American Political Science
Review, LXTV (June 1970), 411-29.
17. The 1963 Venezuelan peasant sample survey was conducted as part of the “Study of Conflicts and
Consensus in Venezuela,” jointly sponsored by the Centro de Estudios del Desarrollo (CENDES) of
the Universidad Central de Venezuela and the Center for International Studies of the Massachusetts
Institute of Technology. This study is described in Frank Bonilla and José A. Silva Michelena, A
Strategy for Research on Social Policy: The Politics of Change in Venezuela, vol. 1 (Cambridge
[Mass.], 1967), and in John R. Mathiason, “Political Mobilization of the Venezuelan Campesino,”
(Ph.D. diss., Department of Political Science, Massachusetts Institute of Technology, 1968). The 1967
Venezuelan peasant sample survey and the 1966 peasant leader survey were conducted as part of a
large-scale study of the Venezuelan agrarian reform jointly sponsored by CENDES and the
Interamerican Agricultural Development Committee (CIDA) of the Organization of American States.
These samples are described in Mathiason. The 1969 Colombian peasant survey was conducted as part
of a program to assist the National Agrarian Federation of Colombia (FANAL) in its efforts to
organize peasants into unions and was jointly sponsored by FANAL and the Center for Rural
Development, of Cambridge, Massachusetts. While the Venezuelan surveys are national samples, for
financial reasons the Colombia survey is a sample of peasants in the Department of Cundinamarca
living in villages with less than 400 population. This sample is described in John Duncan Powell,
“Organizing Colombian Peasants” (Cambridge [Mass.], The Center for Rural Development, 1968),
mimeographed.
18. The striking nature of these participation figures may be seen by comparing them with the figures
obtained in five countries by Almond and Verba, pp. 302 and 308.
19. Weinert and Payne.
20. See T. Lynn Smith, Colombia: Social Structure and the Process of Development (Gainesville,
1967), for a detailed presentation of the rural sociology of Colombia, emphasizing the land problem.
21. For an analysis of the concept of peasant “following,” see Benno Galjart, “Class and ‘Following in
Rural Brazil,” America Latina, VII (1964), 3-24.
22. Guzman Campos et al., La Violencia, vol. 2.
23. Payne, Patterns of Conflict.
24. Not only did our interviews pick up much criticism of the influence of local large landowners, but
one can see from Table 3 that, while landlords are recognized as respected and influential, they are not
recognized for using this influence to help the peasant community.
25. See Daniel Lerner, The Passing of Traditional Society (Glencoe, 1958), for the most detailed early
statement of this hypothesis.
26. Concerning the crucial role of smallholders in peasant conflict with landlords, see Arthur
Stinchcombe, “Agricultural Enterprise and Rural Class Relations/’ in Jason L. Finkle and Richard W.
Gable, eds. Political Development and Social Change, (New York, 1966), p. 491.
27. Smith, Colombia, pp. 83-89.
28. The general outlines are as follows. After an unsuccessful student uprising in 1928 against the
dictator, Juan Vicente Gomez, Romulo Betancourt and other revolutionary leaders determined that the
most effective way to power was by mobilizing Venezuela’s disfranchised masses. In 1936, after
Gomez’ death, these leaders began organizing both the urban working class and the peasantry. Since
political parties per se were still proscribed, this was done through labor and peasant unions under a
liberal labor law passed as part of the post-Gomez liberalization. Considerable success was met,
especially in the urban areas and in the countryside surrounding Caracas, the national capital. The AD
party was formed officially in 1941 and came to power in a joint civil-military coup d’etat in 1945.
During 1945-48, AD was in power. Among its first acts was to extend suffrage to all adults, bringing
peasant and urban masses directly into politics for the first time. In three successive elections, AD
obtained 80 percent of the total vote, which meant that most peasants began their voting with AD. In
addition, through its peasant unions, the party began formally organizing peasants. By the time AD
was overthrown by another military coup in 1948, much of the peasantry in the states surrounding
Caracas and other central highlands cities had been organized and had benefited from a radical land
redistribution program. After another coup in January 1958 brought civilian government back, AD
recaptured the government in the 1958 elections, largely because of an almost total control of the rural
vote. One of the AD’s first postelection reforms was a comprehensive agrarian reform law which was
used as a further vehicle to organize the remaining peasants into unions. In the 1963 elections, AD
once more won the presidency, but with a bare 33 percent of the vote, mostly from the rural areas. For
detailed discussion on this general context of peasant mobilization, see John Duncan Powell, “The
Politics of Agrarian Reform in Venezuela” (Ph.D. diss., University of Wisconsin, Madison, 1966);
John Martz, Acción Democrática: Evolution of a Modern Political Party in Venezuela (Princeton,
1965); Robert J. Alexander, The Venezuelan Democratic Revolution (New Brunswick, 1964); and
Mathiason, “Political Mobilization.”
29. John Duncan Powell, “Venezuela: Agrarian Reform or Agrarian Revolution,” in Robert Kaufman
and Arpad von Lazar, eds. Reform or Revolution: headings in Latin American Politics (Boston, 1969).
30. For the Brazilian case see Galjart, “Class and ‘Following’.” Detailed evidence for the Venezuelan
case is presented by John Duncan Powell, Political Mobilization of Venezuelan Peasants (Cambridge
[Mass.], 1971).
31. These two factors combined to produce membership en masse in the local union. An organizer
would come and either almost everyone would join or no union would be founded. There was a certain
indirect compulsion for this, since agrarian reform benefits are distributed to unions of peasants rather
than to individuals making union membership on all-join-or-nothing basis a de facto requirement for
peasants to share government-provided benefits. This is reflected in the 1967 data in a high level of
correlation between the year a union was founded and the year most of its present members first joined
a union.
32. Domingo Alberto Rangel, Los Andinos En El Poder (Merida, 1964), pp. 7-37.
33. This interpretation seems to follow also from the fact that the temporal spread of peasant unions
was outward from Caracas, with the first unions being founded in the central states of Aragua,
Carabobo, and Miranda in the 1940s and in the remote states of Bolivar, Apure, and Barinas only in
the late 1950s and early 1960s.
34. The effectiveness score for each union in the sample was based generally on the assumption that
an “effective” union was one which maintained its internal organizational structure and had a wide
range of activities. The score which ranged from 0 to 9 was calculated on the following basis: a union
was given one point if it regularly meets at least twice a month; two points if 60 percent or more of its
members regularly attend meetings and one point if 40-59 percent regularly attend; one point if the
union maintains at least two of the three legally required record books; one point if it undertakes five
or more different types of activities in a given year (petitions, social events, e.g.); one point if it sent a
delegate to the most recent National Peasant Congress; one point if government or Peasant Federation
officials addressed a meeting during the last year; and two points if delegations were sent twelve or
more times to one agency and six to eleven times to three more. In the 1967 sample, the 1,260
respondents were distributed among 107 peasant unions. A score of four or less was classified as “low
effectiveness” and a score of five to nine as “high effectiveness.” Thirty-one percent of the unions
with 30 percent of the members had “low effectiveness” scores.
35. James S. Coleman, Introduction to Mathematical Sociology (New York, 1964), chap. 6. The
“effect parameter,” as Coleman terms the statistic, is the average percentage difference across all cells
in a multi-variate contingency table accounted for by a given independent variable with respect to the
dependent variable, holding all other independent variables in the table constant. The specific use of
this statistic in this research is described in detail in Mathiason, Appendix B. Another application of
this technique to similar problems is found in Robert R. Alford and Harry M. Scoble, “Sources of
Local Political Involvement,” American Political Science Review, LXII (December 1968), 1192-1206.
*This is a revised version of a paper presented to the panel on The Individual and Political and Social
Change at the Annual Meeting of the American Political Science Association, New York City,
September 4, 1969.
India’s Urban Constituencies
I Introduction
There are two widely held views concerning the political characteristics of
cities in low-income countries. One school holds that the urban experience in
developing countries makes individuals more politically aware, stimulates
citizens to expect more of their political system, generates frustrations as
governments fail to meet these expectations, and leads urban dwellers to
become increasingly radical and their politics to become polarized by
ideological cleavages. There is, or so this model suggests, a progression
from politicization to radicalization to polarization in urban political life,
together with a tendency for the city to be a diffusion center shaping the
political character of the rural region in which it is located.1
An alternative way of looking at urban areas in low-income parts of the
world is to note that they differ significantly from one developing country to
another, that they are affected by the rural areas around them in addition to
having an effect of their own on those areas, and that, even within a single
country, there are likely to be substantial variations from one city to another
as to the degree of politicization, radicalization, and polarization of politics.
These variations depend upon such factors as city size, migration, levels of
development, occupational complexion, and the political character of the
surrounding regions. This second model emphasizes interaction with the
countryside. It assumes that the city shares many of the political features of
the countryside; thus, political culture, political organization, and processes
of social change shape the city as they do the rural environment. Therefore,
we should not expect the cities of one developing country to resemble those
of another politically, nor should we assume that the cities in one region
within a developing country are necessarily like those of another.2
The first model is essentially attributional, emphasizing the political
imperatives imposed by the attributes or characteristics of urban life. The
second is more contextual, stressing the common features of cities with their
rural surroundings. The first model implies political uniformities; the second
suggests urban political diversity.
This study describes how India’s urban constituencies have voted since
1952. India is a useful case for exploring these alternative models. It has had
more than twenty years of elections, thereby providing uniform data on one
type of political activity over an extended period of time; it has competitive
political parties, thereby making it possible to find out the political
preferences of urban dwellers; and, despite being a preponderantly rural
country, it has a large number of cities, thereby making intercity
comparisons meaningful.
Figure 4 The Communist Vote in Urban and Rural Constituencies: All India
(1952-72)
*Denotes significance (difference of means) at .01; # at .05.
NOTES:
For 1952 and 1957, only single-member constituencies are tabulated.
Because the figures for 1972 were hand tabulated, no difference of means test was possible.
Figures represent the average of state percentages controlling for the number of constituencies in each
state, but not for constituency size. Jammu and Kashmir and Goa are excluded from the calculations;
data for Rajasthan were not available to us at the time of writing.
The recent increase in Communist support is somewhat greater when all elections between 1970
and 1972 are taken into account.
The difference between these states and the rest of India is less clear with
respect to urban constituencies because much Communist strength elsewhere
is concentrated in the urban areas; and on occasion it is mildly impressive.
Nevertheless, the distinction remains valid; and when the urban Communist
vote from West Bengal, Kerala, and Andhra Pradesh is compared to the
urban Communist vote in other states, we can see that the former is at least
three and usually four times as large as the latter (see Table 5). Moreover, a
solid majority of Communist victories in urban constituencies each year is in
these three states.23
Figure 5 The Jana Sangh Vote in Urban and Rural Constituencies: All India
(1952-72)
Table 7. The Average Jana Sangh Vote in the Six States of the Hindi Region
and the Rest of India, Showing Urban-Rural Differences for Each (1952-72)
The Jana Sangh’s vote in the six Hindi states has shown impressive
growth over the years, notwithstanding a slight downturn in 1969. The Jana
Sangh has been in the ascendancy throughout this area but especially in the
cities, where its vote has increased so much that it continues to be almost
double the party’s growing rural vote. The urban right-wing radicalization
suggested by these trends is quite spectacular, despite some recent reverses
in the Punjab and Haryana (Table 8).
The strong urban position of the Jana Sangh did not show itself in seats
held, however, until the 1967 elections, when the party won 48 out of 123
urban constituencies in the Hindi states (compared with 13 of 117 urban
constituencies in 1962) .30 In other words, the Jana Sangh won 39 percent of
the urban seats in 1967 (as against 17 percent of all the seats in the Hindi
states; or 245 seats out of 1,408). In 1962 they won 11 percent of the urban
seats (compared with 7.5 percent of the total, or 115 out of 1,352 seats).31
Neither the increased vote for the Jana Sangh through 1967 nor the mild
decline thereafter is a reflection simply of the number of Jana Sangh
candidates contesting. In city after city, Jana Sangh candidates won a larger
proportion of the vote in 1967 than in the same constituencies in 1962.
Moreover, although the Jana Sangh remained a party of the Hindi region in
1967 (forty-eight of its fifty-four urban victories were in the Hindi region), it
did make some modest inroads into the urban centers of Andhra Pradesh
(notably Hyderabad, Guntur, and Nellore) and Maharashtra (Bombay, Poona,
Nasik, and Akola). In 1969 and 1972 the Jana Sangh dropped out of very
few of the urban constituencies it had contested in 1967; it simply did less
well in them, especially outside the Hindi region.
Table 8. The Jana Sangh Vote: Urban Constituencies in the Six Hindi States
(1952-72)
The Jana Sangh has tended to secure a stronger vote in cities of 200,000 to
1 million than in either the great metropolitan centers or smaller cities. The
absence of appeal for the Jana Sangh in India’s largest cities, however, is
nothing more than a reflection of the party’s regional concentration of
strength. There is only one million-plus city in the Hindi region—Delhi—
and it was omitted from our analysis since it does not have a state legislative
assembly. However, in parliamentary and municipal elections the Jana Sangh
has been a powerful force in Delhi. (In 1967 it swept all but one of Delhi’s
parliamentary seats.) Similarly, while the Jana Sangh vote has tended to be
higher in larger than in smaller cities (those between 50,000 and 200,000),32
the patterns vary from state to state; within the Hindi states the votes are so
consistently small that city size hardly emerges as a significant factor in Jana
Sangh performance.
All in all, radical voting in India may be more urban than rural, but neither
in the case of the Jana Sangh nor the Communists can urban distinctiveness
be further specified according to city size.33
Congress and the “radical” protest vote This survey shows that the
Communist parties and the Jana Sangh have consistently won a larger
portion of India’s urban than its rural vote and that between 1952 and 1967
they grew stronger in the cities. By 1967 the Jana Sangh, CPI, and the CPM
together secured more than a quarter of the urban vote nationally (as against
a ninth in 1952), and about a sixth of the rural vote. It now remains to
determine at whose expense these parties have grown.
Surprisingly, support for the Congress party in both urban and rural India
remained remarkably constant in the first three general elections, with no
significant disparities between urban and rural voting for Congress.34 In
other words, the virtual doubling of the Jana Sangh and Communist vote
prior to 1967 was not accompanied by any comparable decrease in the
Congress vote. During this early period the Jana Sangh and the Communist
parties grew at the expense of independent candidates and other opposition
parties.35 Thus, between 1952 and 1962, the vote going to “other parties and
independents” declined by 10 percent in urban constituencies and by 6.7
percent in rural constituencies. The parties which lost the most, especially in
urban India, are the various socialist parties.36 Their combined nationwide
vote in the first three general elections was 9.7 percent, 10.1 percent, and 9.7
percent, respectively; but in the urban areas their decline was dramatic: 13.5,
9.9, and 5.9 percent in cities of 100,000 or more.37
Table 10. Distribution of Vote among the “Radical” Parties, Congress, and
Other Parties and Independents in Urban and Rural Constituencies: All India
(1952-72)
In short, from 1952 through 1967 the cities of India were increasingly
polarized between the radical left- and right-wing parties, on the one hand,
and Congress, on the other. In the cities of the Hindi region the clash was
between Congress and the Jana Sangh, while in many of the cities of West
Bengal, Andhra, and Kerala (plus a scattering of cities elsewhere in the
country), the polarization was between Congress and the Communists. From
1952 through 1962 Congress and the radical parties appear to have taken
their votes from independents and other opposition parties. In 1967, when
the smaller parties and independents improved their position in urban India,
the struggle was more sharply between Congress and candidates either of the
radical Left or Right.
Thus, the elections of 1967 were critical for the Congress party. Its
fortunes declined throughout India, but nowhere more than in the urban
centers, where the Jana Sangh in particular reaped major gains in both votes
and seats. Congress managed to win almost half (49 percent) of the rural
state assembly seats, down from 62 percent in 1962; but in the urban areas
Congress won only 43 percent of the seats, compared with 67 percent in
1962. Congress was in a beleaguered position everywhere; but now, for the
first time since independence, urban India gave Congress a smaller
proportion of votes and seats than rural India.38 Moreover, for the first time
the Congress party failed to win a clear majority of assembly seats. Indeed,
in half of the states the opposition parties were able to band together to form
coalition governments, and in several instances the urban seats provided the
non-Congress parties with the margin necessary to form a government.
The urban crisis for Congress was even more dramatic in the national
parliament. When the Congress ministers faced the opposition benches in
parliament after the 1967 elections, it was apparent that the M.P.’s from the
urban areas (parliamentary constituencies in cities with more than a quarter
million population) were predominantly with the opposition. Congress had
won only eighteen of the fifty-two urban parliamentary constituencies in
1967, even though victorious in a clear majority of the rural constituencies.
Although the decline of Congress in the urban areas in 1967 was an all-
India phenomenon, the losses were substantially greater in some states than
in others. The decline exceeded the national urban average in Bihar, Uttar
Pradesh, Madhya Pradesh, Punjab, and Haryana, states in which the Jana
Sangh’s urban vote had sharply risen. In city after city it was clear that the
decline in support for Congress was accompanied by comparable Jana Sangh
gains (rather than gains for the Communist parties). The 1967 elections thus
indicated that India’s increasingly radicalized urban population was moving
quite rapidly to the Right in the Hindi states, while elsewhere the movement
toward the radical Left had leveled off (except in Bihar). In West Bengal, for
example, which contains India’s most leftist urban population, the combined
urban Communist vote was 37.4 percent in 1967, not unlike the 35.7 percent
which the united CPI had received in 1962.
Although in percentage terms the decline in Congress votes was small in
1967, the margin of victory in previous elections had already become so
narrow that a slight shift in votes per constituency was decisive for many
Congress candidates. Cities that had consistently elected Congress
candidates in previous elections now elected Congress opponents. In the
Punjab, Congress lost all four assembly seats in Amritsar, the three seats in
Jullundur, and the seats in Patiala, Bhatinda, Hoshiarpur, and Chandigarh. In
Madhya Pradesh, Congress lost the three seats of Gwalior and three of the
four seats in Indore along with the seats including Burhanpur and Khandwa.
In Tamil Nadu, Congress lost all twelve seats in Madras City (as against six
in the previous elections), the three seats in Madurai and two in Salem, and
fifteen of the seventeen seats for cities with populations between 50,000 and
100,000, which had previously been won. In Uttar Pradesh, the Congress
was routed in all four seats of Lucknow, three in Benaras, two in Meerut,
and the constituencies in Gorakpur, Mathura, Aligarh, and Dehra Dun, plus
every one of the ten smaller cities it had won in 1962. Elsewhere in the
country Congress lost two of its three seats in Baroda, two seats in Mysore
City, and two in Ajmer.
Table 13. Regional Protest in Modern Punjab: 1962 Compared with Other
Election Years
Similarly, the rapid expansion of the DMK in Tamil Nadu from 1962
onward, reflecting the politicization of Tamil cultural consciousness, was as
substantial in the rural as in the urban areas, both with regard to votes
secured and seats won. Nor is city size a factor, although the DMK’s initial
urban victories were largely confined to Madras City (five of seven seats in
1962).
The agitation for Punjabi Suba is another instance of regional protest. The
Akali Dal, a Sikh party, campaigned for the creation of a Sikh-majority state
out of the Punjab in the 1962 elections. Surprisingly, the Akalis did not do as
well either in votes received or seats won in the regions of the state that were
eventually converted (in 1966) into a Punjabi Suba as they had previously
(1952) or were to do subsequently (1967 and 1969).43 Moreover, in none of
the five elections contested by the Akalis was their urban support as great as
their rural support, obviously because the Sikh population is more heavily
rural, while the cities of the Punjab have higher proportions of Hindus than
most rural constituencies. The Akali-Jana Sangh alliance in recent elections
has, in this sense, been a rural-urban alliance of anti-Congress political
forces.44
The three regional protest movements reviewed here were among the most
active opposition forces in the 1960s; and in two instances the regional
parties continue to have well-established institutionalized roots in their
respective states, Tamil Nadu and the Punjab. As we have seen, the major
regional parties in Bombay (in 1957) and Tamil Nadu (since 1962) won as
much electoral support in rural as in urban constituencies; and in the Punjab
the Akali movement is more solidly based in the rural areas in which the
Sikhs reside. In all three instances, therefore, regional protest takes the form
of solidarity movements based on linguistic, religious, and cultural affinities
that cut across urban-rural differences. Overall, regional protest differs from
“radical” protest in not being distinctively urban.
Electoral trends from 1952 to 1972 suggest that the urban constituencies
are becoming more distinctive insofar as electoral protest and electoral
polarization are concerned, but less distinctive with regard to electoral
participation (if only because the rural areas have caught up). To put it
another way, the attributional model is more appropriate for understanding
radicalization and polarization, while the contextual model is as appropriate
for understanding participation.
If the polarization and radicalization trends continue, it seems likely that
urban constituencies will command even more of the attention and resources
of party and government leaders than they do at present. Let us briefly
review, then, these two long-term trends, both so suggestive of future
developments.
First, elections are becoming more closely contested all over India. As the
margins of victory become small, relatively slight shifts in votes can mean a
substantial shift in the number of seats that a party wins or loses. The
massive defeat of Congress candidates in the urban constituencies in 1967
involved only a small decline in the Congress vote. To the extent that state
assembly or parliamentary elections become more closely contested, the
constituencies where victories are by the smallest margins will be given the
most attention by political parties-and by the government. In nearly three-
fourths of the state elections in 1967 the margins of victory were smaller in
the urban than in the rural areas; and in each successive election between
1952 and 1967 the urban constituencies became competitive more rapidly
than the rural constituencies. In Maharashtra, nearly one out of four voters
lives in an urban constituency; in West Bengal, one out of five; in Gujarat
and Tamil Nadu, more than one out of six; in Mysore and the Punjab, one
out of seven; and in Andhra, one out of eight. Closely contested elections in
these states are likely to make their urban areas the political battlegrounds
for competing parties.
Second, between 1952 and 1967, the “radical” vote-i.e., the vote for both
the Communist parties and the Jana Sangh-increased more rapidly in the
cities than in the countryside. This trend only slowed in the early 1970s,
when the governing Congress party itself assumed a more radical posture.
The protest vote was particularly acute in the 1967 elections, a time of
growing food shortages and rising prices, factors that are again appearing in
the mid-1970s as a concomitant of petroleum and fertilizer shortages.
While the lowest-income groups in the countryside suffer the most during
periods of food shortages and inflation, the urban middle class and the
organized working class have the political resources to harass the
government and threaten the regime. Certain urban occupational groups
occupy strategic nodal points in the economy, with the capacity to inflict
damage far beyond their own local constituencies: the port workers in
Calcutta and Bombay, railway workers in Hyderabad and Madras, state
government employees in Patna and Lucknow, engineers in electric power
stations, and so on. There appears to be a long-term trend toward greater
political militancy in India’s urban areas, a militancy that can be tapped by
parties either of the Left or Right if it is not tapped by the governing
Congress party.
Postscript
Recent developments in India have tended to confirm the trends described in
this study. Severe inflation in 1974 and 1975 (estimated at 30 percent) was
accompanied, to quote from a resolution of the Congress Working
Committee, by “organized strikes, go slow movements by government
employees, railway employees and industrial employees . . . acts of sabotage
paralyzing the railway and communication systems . . . student agitations
and indiscipline,”51 all urban-centered protest movements. Mass campaigns
against the government, first in Bihar, then in Gujarat, were followed by a
stunning Congress defeat at the hand of a coalition of parties in elections for
the Gujarat legislative assembly in mid-June 1975. To a significant degree
urban discontent seemed to be striking a responsive chord among rural
people as well. When Prime Minister Gandhi’s own legal difficulties
surfaced a few days later—the Allahabad High Court ruling that she had
violated the electoral law and that her election to parliament in 1971 was
therefore invalid, a mass public rally, organized by several leading
opposition parties, was held in New Delhi on June 25, calling for her
resignation and announcing plans for a nationwide protest movement.
On the very next day the Indian Government issued a Proclamation of
Emergency under the provisions of the Indian Constitution, arrested more
than a thousand members of the opposition (according to official reports),
amended the Maintenance of Internal Security Act so that no grounds had to
be given for detention, imposed censorship on the press, and banned twenty-
six political organizations, including the Rashtriya Swayam Sevak Sangh (a
militant group associated with the Jana Sangh), the Naxalites (pro-Peking
militant Communists), and a number of other Marxist groups. The
government subsequently announced that elections scheduled for the state of
Kerala in September would be postponed.
The opposition parties had clearly been able to mount organized urban
protest and to influence rural sentiment on a scale that was far more
alarming to the central government than any earlier movements. Whether the
government’s actions to suppress the opposition, combined with new
measures announced by the Prime Minister to “revive the health of the
economy,” will alter the situation or whether urban protest, with or without
rural sympathy, will persist in the form of extralegal movements and anti-
Congress voting (if free elections are again held) remains to be seen.
NOTES
1. Many studies have examined urban distinctiveness with reference to such political characteristics as
participation, radicalization, and polarization. The principal theoretical base for these studies is the
modernization-social mobilization paradigm common to much comparative analysis. See in particular
Daniel Lerner, The Passing of Traditional Society: Modernizing the Middle East (New York, 1958),
and Karl W. Deutsch, “Social Mobilization and Political Development,” American Political Science
Review, LV (September 1961), 493514. Additional interest has been stimulated by the presumably
combustible effects of large-scale rural-to-urban migration. For a critical review, see Joan M. Nelson,
“Migrants, Urban Poverty, and Instability in Developing Nations,” Monograph # 22 (Cambridge
[Mass.], Center for International Affairs, Harvard University, September 1969).
2. Recent empirical research has questioned the importance of urbanity insofar as popular political
dispositions and basic political dynamics are concerned. For example, see Norman H. Nie, G.
Bingham Powell, Jr., and Kenneth Prewitt, “Social Structure and Political Participation:
Developmental Relationships,” American Political Science Review, LXIII (June 1969 and September
1969), 361-78 and 808-32; and Alex Inkeles, “Participant Citizenship in Six Developing Countries,”
ibid., (December 1969), 1120-41. The apparent weakness of urbanity as an explanatory variable has
kindled renewed appreciation for contextual explanations of aggregate political behavior. Our analysis
treats context as political space, typically the state in which certain urban constituencies are located.
For a more varied, social treatment of context as applied to India, see David J. Elkins, “Regional
Contexts of Political Participation: Some Illustrations from South India,” Canadian Journal of
Political Science, V (June 1972), 167-89.
3. Small cities are rarely divided into two or more constituencies, while even a city of 50,000 typically
forms the dominant part of an assembly constituency in India. In 1967, when constituencies tended to
include more people than in earlier elections, the average number of eligible electors was 69,920 per
constituency. Instances where constituency boundaries combined portions of the urban area with a
substantial rural area are excluded from our analysis. Delhi is also eliminated because, as a union
territory, it has had infrequent state assembly elections. For methodological reasons we are
occasionally obliged to omit double-member constituencies in cities during the 1950s (none appear
after 1960); the substantive significance of doing so is typically minuscule.
4. The figures on turnout presented in this study pertain to the valid vote actually tabulated. In every
constituency a number of ballots are disqualified by officials of the Election Commission for one
reason or another. Usually the rate of invalidation is around 2-3 percent of the total votes cast,
although on occasion it has exceeded 10 percent. Invalidation is not a factor in our findings, but the
reader should keep in mind that we are referring to effective electoral participation, not total turnout.
All turnout figures have been rounded to the nearest percentage.
5. The 1972 assembly elections are not exactly comparable to earlier elections, since a number of
states (Uttar Pradesh, Orissa, Tamil Nadu, and Kerala) did not go in the polls that year.
6. Like the tables in this section, the data shown for 1952 and 1957 in Figure 2 pertain to single-
member constituencies only. Removing double-member constituencies is a methodological
convenience that has the effect of overstating the level of turnout, but never by enough to make a
meaningful difference.
7. Our analysis is based on the state divisions obtaining in 1967. For the sake of continuity and clarity,
we shall refer to earlier electoral patterns as if they occurred within the contemporary boundaries.
8. Kerala did not go to the polls in 1962. The spurt in urban turnout there took place in the special
elections of 1960.
9. Urban turnout in the Maharashtra region dropped from 63 to 60 percent in 1962, while rural turnout
increased from 51 to 55 percent. The same pattern occurred in Gujarat: urban turnout declined from 59
to 55 percent, as against rural figures of 52 to 54 percent for each year. The decline in urban turnout in
Gujarat is almost entirely a function of downswings in Ahmedabad, the capital city, and Baroda, a
major cultural-intellectual center where states’ reorganization sentiment had been strong in 1957. In
Maharashtra, the urban decline was greatest in Bombay and Poona, also the capital city and major
cultural-intellectual center.
10. The persisting urban-rural gap in such states as Madhya Pradesh, Bihar, and Orissa is partly a
function of the large numbers of low-voting tribal peoples in these states, most of whom are located in
specially reserved constituencies. However, even when tribal constituencies are removed from the
comparison, the urban-rural gap remains large. Low tribal turnout is therefore not the only reason for
the gap; it merely accentuates it. For a detailed analysis of electoral politics in tribal India, see Myron
Weiner and John Osgood Field, “How Tribal Constituencies in India Vote,” in Weiner and Field, eds.
Electoral Politics in the Indian States: Three Disadvantaged Sectors (New Delhi, 1975), vol. 2.
11. Urban turnout in Kerala has shown a steady decline since 1960, but it remains well above the
national urban average.
12. Operationally, urbanity is imputed to be a relevant attribute when a difference of means test fails
to reveal a significant (.01) difference between the state urban mean and the national urban mean for a
given election year. Context is said to be a relevant guide to urban turnout when the same test fails to
reveal a significant (.01) difference between urban and rural constituencies in the same state. In each
case significance is determined by a two-tailed test of Ζ scores produced by the conventional part-
whole formula
M1 −M0
Z =
σ0 /√N1 .
13. Both urbanity and context are said to be dependable guides to turnout when a difference of means
test fails to reveal a significant (.01) difference either in the urban-rural comparison or in the state
urban-national urban comparison. This occurred in nine states between 1952 and 1967, most
commonly in Mysore. The considerable overlap probably has something to do with the generally
small differences between urban and rural turnout in India as a whole (a maximum of nine percent)
and in many states over the past two decades.
14. Neither urbanity nor context is deemed an adequate guide to turnout when a significant (.01)
difference exists between a state’s urban constituencies and the national urban mean, on the one hand,
and the state’s rural constituencies, on the other. Resembling neither in the statistical sense, the urban
constituencies in question cannot be identified by either urbanity or context. This was the case for
urban constituencies in Saurashtra and West Bengal in 1952, Bihar and Madras in 1962, and Uttar
Pradesh in 1962 and 1967.
15. The low turnout of distinctly backward states may be due as much to low population density as to
their economic backwardness. R. Chandidas has argued and provided indirect evidence that the
distance people must travel in order to vote affects how many actually do vote (“Poll Participation
Slump,” Economic and Political Weekly, 15 July 1972, 1359-68; see Table 9 and the surrounding
discussion). Because low rural density and economic backwardness often go together, their effects are
difficult to sort out. At the state level the evidence on density is mixed. Some states with a low rural
turnout (e.g., Orissa, Madhya Pradesh, and Rajasthan) are below the all-India level in population
density. However, Gujarat, another low-density state, has a higher electoral turnout; and Bihar, which
has generally had a below-average rural turnout, is one of India’s most densely populated states.
16. Urbanity made a difference for both Bihar and India. Urban-rural differences were significant at a
.01 level of confidence or greater, as was the difference between the national urban and Bihar urban
figures.
17. As with urban constituencies, we have performed difference of means tests at the state level for all
constituencies.
18. The same basic patterns emerge when the comparison is made against all urban constituencies in a
state. The high-turnout states tend to have the largest proportion of urban constituencies which, by
national urban standards, are also characterized by high-turnout, whereas states that have the largest
proportion of low-turnout urban constituencies are not consistently low-turnout states. On the contrary,
they are almost as likely to be high-turnout states.
19. Nor are they substantial at the state level. The only instances in which turnout in small cities
revealed a distinctive bulge over that in larger cities and rural constituencies generally were in Madras
State in 1957 and 1967.
20. Several readers-of our manuscript quite legitimately objected to our use of the word “radical” to
characterize the Jana Sangh and Communist parties, preferring instead to describe them as “extremist”
or simply “right-wing” and “left-wing.” Of course, some people argue that the Jana Sangh is neither
radical nor extremist, although one or the other term might be appropriate for the recently banned
Rashtriya Swayam Sevak Sangh, a militant youth-oriented affiliate of the Jana Sangh. “Right-wing” is
appropriate only insofar as the Jana Sangh has a pro-Hindu orientation, since its economic platform
proclaims a socialist program. The Communist parties are equally difficult to characterize, for the pro-
Soviet CPI is no longer viewed by most Indians as “radical” or “extremist,” and some no longer regard
it as a party of the “Left” given its present collaboration with the “bourgeois” government of Prime
Minister Gandhi. In short, none of the nomenclature is adequate to describe the two major groups that
have for most of the elections analyzed here been the major nationally recognized ideological
opponents of the ruling Congress party. Two reasons underlay our selection of these parties for special
analysis and our designation of them as “radical”: they have an organizational base in many urban
constituencies, and they tend (especially in their youth wings) to attract those who are ideologically
opposed to the existing political order as well as those who are merely in opposition to the governing
party.
21. In Andhra Pradesh in 1972, when the CPI and CPM combined acquired only 9.2 percent of the
total vote.
22. The CPI received 17.7 percent of the vote in what is now Punjab in the 1957 elections, when it
provided the only real opposition to the ruling Congress (the Akali Dal having merged with Congress
that year). In 1969 the CPI secured 10.1 percent of the vote in Bihar, the CPM polling another 1.2
percent. Tripura gave the Communists (mostly the CPM) 29 percent of the vote in 1967, and 40.8
percent in 1972. These are the only instances in which the Communists have received as much as 10
percent of the vote in any state other than West Bengal, Kerala, and Andhra Pradesh.
23. Twelve of twenty in 1952 (seven in West Bengal alone, which has more urban constituencies than
Andhra and Kerala combined), twenty-one of thirty-two in 1957 (nineteen in West Bengal), twenty-
seven of thirty-eight in 1962 (twenty-three in West Bengal), twenty-eight of forty-six in 1967 (twenty-
one in West Bengal), and twelve of twenty-four in 1972 (nine in West Bengal; these figures include
Kerala elections in 1970). The 1952 figures are understated because the CPI contested under the
banner of the People’s Democratic Front in parts of what is now Andhra Pradesh and as independents
in Travancore-Cochin. Many of these candidacies were successful, including in urban constituencies.
The 1962 figures do not include Kerala, where assembly elections were not held that year.
24. Only 1 percent separated the two in 1972, reflecting the serious attrition of support for the CPM in
the Calcutta metropolitan area.
25. Fifteen percent of all constituencies were urban (50,000 +) in 1967.
26. This finding runs counter to findings from Colombia, where larger cities tend to be more radical.
See Lars Schoultz, “Urbanization and Changing Voting Patterns: Colombia, 19461970,” Political
Science Quarterly, LXXXVII (December 1972), 40. In our analysis the national urban data were
examined in the following size categories: 1 million plus; 200,000 to 1 million; 100,000 to 200,000;
and 50,000 to 100,000. At the state level the comparisons are almost always above as against below
the 100,000 threshold.
27. To be sure, most urban constituencies in West Bengal are in the greater Calcutta metropolitan area.
However, those that are not tend to be quite similar to those that are.
28. In 1972 the urban Communist vote declined by 2.3 percent in Bihar and by 10.6 percent in West
Bengal.
29. In Madhya Pradesh the Jana Sangh has tended to do almost as well in the countryside as in the
cities.
30. Elsewhere in the country the Jana Sangh failed to win a single urban seat in 1962. In 1967 it
managed to win six out of 258.
31. The year 1972 witnessed a general downturn in Jana Sangh victories in the face of the “Indira
Wave” which swept the country. In the Hindi region the Jana Sangh won seven of sixty urban seats
(excluding Rajasthan, for which we were unable to secure returns); elsewhere it won only one urban
constituency (in Poona).
32. For example, in 1967 the Jana Sangh won 20.4 percent of the vote in cities with 200,000 to 1
million people as against 13.9 percent in cities between 50,000 and 200,000. For 1952, the figures are
8.8 and 4.7 percent, respectively.
33. Difference of means tests yield almost no significant results, and regressions indicate that the
value added from introducing size categories is rarely significantly greater than the variation already
explained by the urban category alone. We might note that these same tests indicate constituencies
encompassing cities of 50,000 to 100,000 to be clearly “urban” in their support for the radical parties.
Wherever the urban threshold is politically, it is not above the 50,000 cutoff employed in this analysis.
34. The differences in 1952 and 1962 are statistically significant at .05, a somewhat loose standard
when the number of cases is so great (more than 2,000 constituencies each year). The absence of
highly visible disparities during the first decade, true in most states as well as the country at large,
means that urban areas were no different from rural in their support for Congress. It is possible that
urban centers influenced the rural areas around them (or were influenced by them), but the spatial
plotting of deviations from state means reveals such seemingly random distributions that proximity
and influence hypotheses appear to be inadequate. It would seem that through the 1962 elections
urbanity was neither distinctive nor influential in this regard.
35. Voting for nonparty independent candidates cannot easily be separated from the vote for smaller
political parties not recognized by the Election Commission. Typically, as much as one-fifth of India’s
voters will support such candidates, although somewhat less in urban than in rural India.
36. The socialist parties include the Socialist party and the Kisan Mazdoor Praja party in 1952, the
Praja Socialist party (PSP) in 1957, the PSP and Socialist party in 1962, the PSP and Samyukta
Socialist party in 1967, and the Socialist party in 1972.
37. The performance of the socialist parties has not been analyzed in the smaller cities of 50,000. We
have no reason to believe that their record should be any different there.
38. Not only was the balance in urban-rural support for Congress reversed in 1967, but for the first
time the difference was significant at .01.
39. This figure excludes Rajasthan, for which returns were not available to us at the time of writing,
along with Jammu and Kashmir and Coa.
40. Both the CPI (in alliance with Congress in a number of states) and the CPM actually improved
their urban vote in 1972, although marginally in each case.
41. See Marguerite Ross Barnett, “Cultural Nationalist Electoral Politics in Tamil Nadu” in Weiner
and Field, eds. Electoral Politics in the Indian States, vol. 4.
42. During the 1960s a similar dynamic emerged at the local level, particularly in Bombay, where the
Shiv Sena led a nativist reaction to the inmigration of South Indians. For a detailed analysis, see Mary
Fainsod Katzenstein, “Political Nativism: Shiv Sena in Bombay” (Ph.D. diss, in preparation,
Department of Political Science, Massachusetts Institute of Technology).
43. In 1972 the urban Akali vote declined to the 1962 level and no urban seats were won. This
comparatively poor performance mirrored a rural decline as well, both in part at least a function of the
factionalism gripping the Akali movement, which led three Akali parties to contest in 1972.
44. For a detailed analysis of the Akali movement and Punjabi politics generally prior to
reorganization of the state, see Baldev Raj Nayar, Minority Politics in the Punjab (Princeton, 1966).
See also Paul R. Brass, “Ethnic Cleavages and the Punjab Party System, 1952-1972,” in Weiner and
Field, Electoral Politics in the Indian States, vol. 4.
45. Table 10 provides summary data.
46. We are unable to report data for 1972. The returns that year were hand tabulated, and margins of
victory can only be derived from time-consuming calculations which we did not perform.
47. The vote may have been unusually close in urban Uttar Pradesh, but it was also quite fractionated,
as is typical of Uttar Pradesh generally.
48. For an empirical analysis of the relationship between polarization and participation, see Peter
McDonough, “Electoral Competition and Participation in India,” Comparative Politics, IV (October
1971), 77-87.
49. This is a reasonable assumption if fertility rates continue to decline while rural-urban migration
remains at present levels.
50. Most (84 percent) of these urban dwellers will live in cities larger than 100,000.
51. Overseas Hindustan Times, 24 July 1975.
* This article is a product of the M.I.T. Indian Election Data Project, funded by a grant from the
National Science Foundation. The results of the research will appear in a four-volume series, Studies
in Electoral Politics in the Indian States, edited by Myron Weiner and John Osgood Field (New Delhi:
Manohar Book Service; distributed in the United States by South Asia Books, Box 502, Columbia,
Mo. 65201). A somewhat longer version of this article, with statistical appendices and the results in
each urban constituency between 1952 and 1972, will be published in Volume 3, Electoral Politics in
the Indian States: The Impact of Modernization. We would like to express our appreciation to Priscilla
Battis, our programmer and statistical collaborator in the entire project; to Glynn Wood for his
painstaking assistance in confirming the urbanity of ostensibly urban constituencies by assessing the
magnitudes of rural population in them; and to Jessie Janjigian for research and editorial assistance.
PART III
THE ORGANIZATION OF
POLITICS
Introduction
“Seek ye first the political kingdom and all else shall come unto you…”1
This simply stated, biblically based aphorism of the late Kwame Nkrumah
lies at the center of a storm of controversy: a controversy as old as
organized political societies, and one which has not lost its vitality through
the passage of time.
For Third World states, this aphorism generates two immediate questions.
On the one hand there is the question of face validity, since the exposition
of a doctrine is not equal to a statement of a truth; while on the other hand
there is the question of the translatability of what may be an abstract truth
into concrete realities within a given time/condition context. In other words,
we need ask not only if the proposition is true, but also if it is achievable.
Obviously, such questions are not easily answered; yet, whether or not
there exists in Third World states a “primacy of politics,”2 the ways in
which the political realm is organized and the ends toward which the
organization of politics is directed are of clear and fundamental importance.
While the organization of politics itself does not exist in a vacuum, multi-
level inquiries dealing with the construction of a state framework provide a
critical orientation in any analysis of le Tiers Monde.
To nationalist movements, the attainment of independence was but the
first step in the organization of politics. The world of “the morning after”
brought with it the crises of organization and management, in which goals
needed to be formulated, resources mobilized, and patterns established.3
The issue, for many observers, was one of institutionalization — “the
process by which organizations and procedures acquire value and
stability.”4 The primary imperative in the Third World was to construct the
framework of the state, to build an institutionalized political order.
According to Samuel Huntington, the major spokesman for the
institutionalization perspective, the high level of instability and disorder in
many Third World countries was produced by the gap between increasing
social mobility and political participation and enfeebled political
organization and institution-alization. From this view, it is the strength of
political organizations and procedures which provide for the viability of the
political community existing within the boundaries of a state. And though
simplified institutionalism is not without its difficulties (e.g., what are we
trying to institutionalize? toward what ends? and at what costs?), it is
nonetheless based on the compelling logic that before a state can run, it
must stand.
The first essay in this section, Joungwon Alexander Kim’s “The Politics
of Predevelopment,” makes a valuable contribution by emphasizing that the
widely used term “political development”, in fact, refers to two very
different, if interrelated, concepts. In other words, “political development”
deals with both the creation and institutionalization of “a viable new
political system” and the adaptation of “an existing political system” to the
complexities of the modern world.
While Third World states often have to face both challenges
simultaneously, Kim suggests that there is utility in distinguishing one from
the other. The politics of regime formation need be separated from the
politics of regime functioning.5 Thus “development” politics are viewed in
terms of modernization and adaptation and “predevelopment” politics are
concerned with the building of a new framework of governance.
According to Kim, the “predevelopment” process revolves around
consolidation and institutionalization (wherein “consolidated control
becomes rou-tinized” and society resocialized). And it is in a
“predevelopment” system that the role of leadership is so important (see
Part V).
Holding that institutionalization is not static and “does not imply a
freezing of the system,” Kim points to that factor as the critical variable in
Third World politics. For “once institutionalized the system is no longer
within the realm of the politics of predevelopment. It is only then that it
may move into the phase where it can begin to develop effectively or to
modernize.”
Yet, whatever the significance attributed to the establishment of
institutionalized patterns, a meaningful discussion of the organization of
politics in Third World states must move beyond such imagery. The value
of a generalized map of institutional growth lies in its ties with more
specific details. Here we can see that interwoven throughout the political
consolidation process are questions dealing with intra-elite bonding, elite-
mass relationships, and centre-periphery linkages.6 The situational context
of each of these factors provides a critical environment within which the
organization of politics takes place.
These factors, in turn, are connected to the cross cutting issues of
participation (or departicipation7), legitimacy, and ideology in Third World
political systems. The complex interactions of all of these variables lead
one to a consideration of several very different perspectives on the
organization and purpose of politics in le Tiers Monde.
To a significant degree, our attention is drawn to the problems of political
consolidation in societies that exhibit considerable cultural fragmentation.
On the one hand, the consolidation of national institutions, as well as their
broad penetration into society, is thought to be necessary to hold the system
together in the face of strong centrifugal pulls. Yet, on the other hand, the
very efforts to consolidate may well intensify centrifugal forces and thereby
prevent achievement of the desired goal.
Even where cultural fragmentation itself is not originally a major factor,
the existence of severe political segmentation in many Third World
countries places these states in a similarly precarious position. This
situation has led to discussions of national governments in terms of political
machines and patri-monialism.
Interestingly, arguments over this question have come full circle. We
have gone from an emphasis on federal-accommodative policies to an
emphasis on unitary-consolidationist policies, and then have begun to work
back to the former view.8 In part, this is related to the reemphasis on micro
politics in Third World states. All too often, the symbols and expressions of
organized political power at the centre were devoid of substance, while the
actual influence of the periphery, of the villages, was ignored.
The examination of a primary organizational mechanism, the political
party, and discussions of one-party systems, for example, were very
different dependent upon whether one’s view was guided by the rhetoric
from top or the actual impact as seen from below. Yet, if a more micro-level
perspective is needed to counterbalance an unbridled macro-level vision of
societies that is not to say that macro-level orientations should be
abandoned. The struggle to establish a national political framework implies
some degree of shifting in the hierarchical ordering of subunits within a
society. As Teune and Ostrowski have noted: “The very idea of a local
political system . . . connotes membership in a larger political system, and
the importance of that membership must be examined.”9 The emperor may
well be wearing scanty and threadbare clothes, but he is not quite naked.10
The second essay in this section, “The National Electoral Process and
State Building,” brings into focus a number of the interactive factors
delineated above within the context of Ugandan politics. While recognizing
that elections in the Third World are viewed as an “activity which
exacerbates, rather than ameliorates system divisiveness,” Provizer also
notes the argument that elections can serve a positive purpose relative to the
mobilization of political consciousness. The political organization and
procedures, which produce the strength of the political community, are,
after all, themselves dependent not only on their level of
institutionalization, but also on the scope of support provided to them.11
Integrating a discussion of the clash of sub-central autonomy and
nationalist politics with an overview of party development, this essay traces
the movement toward a consolidationist system in Uganda and the
unfulfilled proposals for a new method of elections in that country. Aspects
of inter and intra-party rivalries, the role of ideology, and the impact of
organizational change come together in the discussion of this attempt to
maintain individual participation while promoting departicipation along
sub-national, ethnic lines.
Although these proposals were never translated into action (interrupted
by a military coup d’etat), they nonetheless provide us with some
interesting perspectives on the organization of politics in a Third World
state, including the numerous political dilemmas that cannot be wished
away.
This brings us to the final essay in this section, Hudson’s most interesting
study of “Democracy and Social Mobilization in Lebanese Politics” From
Hudson’s perspective:
Scholars and policy-makers frequently assume that the new states must develop stability before
they can hope to operate the complex and delicate institutions of liberal democracy. But in the
case of the Lebanese Republic this relationship is reversed: democratic institutions have brought
about and maintained stability in an unfavorable political environment.12
NOTES
1. Kwame Nkrumah, quoted in Bob Fitch and Mary Oppenheimer, Ghana: End of An Illusion (New
York: Monthly Review Press, 1966), p. 25.
2. See Herbert Spiro (ed.), Africa: The Primacy of Politics (New York: Random House, 1966) and
the discussion in Robert Gamer, The Developing Nations (Boston: Allyn and Bacon, 1976), pp. 330-
333.
3. See Gerald Heeger, The Politics of Underdevelopment (New York: St. Martin’s Press, 1974), p. 47.
4. Samuel Huntington, Political Order in Changing Societies (New Haven: Yale University Press,
1968), p. 12.
5. Also note Dankwart Rustow’s distinction between functional and genetic inquiry in “Transitions to
Democracy: Toward a Dynamic Model,” Comparative Politics, Vol. II, No. 3 (April 1970).
6. These issues overlap with questions of political integration. See, for example, Myron Weiner,
“Political Integration and Political Development,” The Annals of the American Academy of Political
and Social Science, Vol. CCCLVIII (March 1965).
7. Nelson Kasfir, The Shrinking Political Arena (Berkeley: University of California Press, 1976), pp.
3-27.
8. See Robert Melson and Howard Wolpe, “Modernization and the Politics of Communalism: A
Theoretical Perspective,” The American Political Science Review, Vol. LXIV, No. 4 (December 1970)
and Ali Mazrui, “Pluralism and National Integration,” in Leo Kuper and M. G. Smith (eds.).
Pluralism in Africa (Berkeley: University of California Press, 1969).
9. Henry Teune and Kryzstoff Ostrowski, “Local Political Systems and General Social Processes,”
mimeo, 1970 (prepared for the VIII World Congress of the International Political Science
Association).
10. From Rupert Emerson, “The Prospects for Democracy in Africa,” in Michael Lofchie (ed.), The
State of the Nations (Berkeley: University of California Press, 1971), pp. 249-250.
11. Huntington, Political Order in Changing Societies, p. 12.
12. For an overview on democracy, see Giovanni Sartori’s, Democratic Theory (New York: Praeger,
1965).
The Politics of Predevelopment
In recent years studies of political development have dealt with two very
different concepts: the problem of creating and institutionalizing a viable
new political system, on the one hand, and that of adapting an existing
political system to a complex modern world, on the other. Evaluation of
these phenomena has been complicated by the fact that both have been
termed “political development.” It is even more difficult because both tasks
have been thrust simultaneously upon many of the so-called “emerging” or
“third world” nations. In a practical as well as in a theoretical sense,
therefore, it has become almost impossible to separate the two phenomena.
As the third world nations have emerged from colonial rule, many have
had to deal with the necessity of creating entirely new political systems, in
the absence of existing organizational bases, and turning them into viable
political entities. They have also been confronted with the very different
task of adaptation to a complex postindustrial international setting in the era
of technology, with the perpetual changes and challenges which this world
presents to those who were not among the earliest to cross the industrial-
technological threshold.
Throughout history, long before the industrial or technological
revolutions made the world more complex, nations have faced the task of
constructing new political systems. Creating a viable new state has never
been a simple task. From the birth of the earliest Greek city-state or the first
Chinese state, to the founding of Rome, to the establishment of the
American republic, to the independence of the most recently acquired
member of the world of nations, societies have been faced with the task of
establishing new systems of political organization.
Manfred Halpern has suggested that even the creation of a new system is
different in the modern era because one must create an “open” system—i.e.,
a system which has an intrinsic capacity to generate and absorb further
transformation.1 While it is true that in order to be successful in
institutionalizing a new system which will be stable over time, given the
interaction in the international system in the modern world, the system will
have to be one which has developmental capacity (i.e., is open to change),
the means for introducing a new system (as opposed to the type of system
desirable) are not significantly different from the means required for
introducing new systems in earlier periods. The same generic types of
political tools must be utilized, though if the society now contains a much
broader politically relevant population, the degree to which political tools
must be mobilized to establish a new system successfully will obviously be
considerably escalated.
While political societies have faced the problem of creating new systems
throughout history, and not simply in the modern era, many na-tionsduring
the past century have grappled with the dilemma of adapting to the age of
perpetual change without having to build new political systems. These
nations have sought to expand the capacity of their political systems to
adapt to economic advances, industrialization, and the technological
revolution, without having at the same time to face the problems which
confront a nation whose entire system of political organization has vanished
overnight, and which must wrestle to fill the vacuum. There have indeed
been cases where the inability of the old political system to respond to
modern demands has itself led to the collapse of that system and generated
the need for introducing a new one. Nonetheless, the two challenges need
not come together, and they are not the same.
In order to distinguish between these two very different challenges, it
would be useful to give them specific names. Among those suggested for
the task of building a new political system are the terms, “nation-building”
and “state-building.” But these have been used in so many different ways
that they could create confusion rather than clarification if we adopted them
here. Some have used them interchangeably, while others have given them
separate meanings. Gabriel A. Almond and G. Bingham Powell, Jr. use
each with a more restrictive meaning than that intended here: state-building
is the name they give to the problem of penetration and integration of the
system into the society, and nation-building is used to refer to the problem
of creating loyalty and commitment to the system.2 For the purpose of this
article, the term “development” is applied to the process of adapting and
increasing structural complexities to meet the changes in the modern
industrial and technological world. This process is often referred to also as
political modernization. On the other hand, the period during which the
nation is undergoing the process of building a new political system is called
“predevel-opmental.” This term is used because it does not seem possible
for a political system to develop—i.e., to adapt and change to meet complex
extrinsic changes—until it has attained a viable pattern of political order.
Although guns and funds may come from external sources, organization
is difficult to supply externally. Nonetheless, groups may be organized and
trained in organizational skills outside the country, or foreign political
advisers may assist in setting up an organization within the country. A
group may be part of an international or foreign organization, as some local
communist parties were at one stage subordinates of the Comintern, and as
religious organizations sometimes cross national boundaries.
Persons may be motivated to work within political organizations for
various reasons—as a source of employment in a time of turmoil, for
prestige or promised rewards, to obtain protection from other groups, out of
a sense of sharing a common cause, for fear of retaliation for not joining,
out of hero worship for a charismatic leader, or in support of a symbolic or
programmatic appeal which the group represents.
Ideas are also used as political instruments. As tools of politics, ideas are
measured not by their theoretical validity—i.e., their conformity with truth
or with reality—but by their ability to influence the actions of people.
Indeed, a set of idea symbols may be used purposely by a leadership group
to bring about results contrary to those implied in the ideas themselves.
Although the effects of ideas advocated by the leadership group must be
measured by their impact on political support to determine how effective
they have been as political tools, the validity of the ideas actually espoused
by the leadership group (which may or may not be the same as those overtly
used as political tools) is important to the extent that these ideas may
ultimately shape political activity and have an impact upon the political
environment. A distinction must thus be made between the ideas seriously
held by a leadership group, and ideas simply used as political tools for the
purpose of securing and maintaining political support.
The distinction between ideas seriously held and ideas utilized as
political tools may be less obvious in a political system with a highly
sophisticated populace, for the leaders may be required by their following
(which is aware of the ultimate impact of the ideas on political reality) to
express valid concepts. Where the audience is poorly educated and
politically unsophisticated, however, the winner of a contest of ideas will be
the one who uses concepts meaningful to the audience. To a less
sophisticated audience, the general appeal of an attractive slogan, such as “a
chicken in every pot,” may generate more support than the appeals made by
leaders who come across as more coldly expert. To the sophisticated
audience, the sloganeering politician may seem crude and opportunistic and
the “expert” more capable of handling the needs of the system. In the first
instance, the leader must divorce his seriously held ideas from those he
must use as political tools; in the latter, the politician may be able to utilize
his seriously held ideas as tools of politics because they also have audience
appeal.
Political ideas may be the invention of the politicians themselves; or they
may come from trusted advisors or be borrowed from contemporary or
former native thinkers. They may also be taken from foreign thinkers.
When ideas are borrowed from abroad, the followers of the line of thought
are often sensitive to interpretations of these ideas as they are applied
abroad, and thus tend to be receptive to “authoritative” foreigners who
criticize or seek to influence their actions, using the framework of the
borrowed idea system. For this reason countries which have sought to
borrow entire idea systems from abroad have often been subject to
extensive external intervention, as foreign analysts have become the best
interpreters of what is right within the framework of the borrowed idea
system. Thus, for example, Soviet criticism of the “cult of personality” in
North Korea led to an effort by two factions of the North Korean party to
overthrow Kim II Sung, using Soviet reprimands as a justification.25
Similarly, American criticism of the Syngman Rhee government in South
Korea as undemocratic, reported widely in the South Korean press, was a
factor in the Student Revolution which brought about Rhee’s downfall.26 So
long as North Koreans saw Soviet government officials as experts on
Marxism, or South Koreans saw American analysts as experts on
democracy, criticism by foreigners could seriously affect the government’s
bases of support.
One fallacy in political analysis has been concentration upon one of the
four political tools alone as the primary determinant of the outcome of a
political contest. Economic determinists (in the Marxist tradition) would
attribute the results of all political activity to the locus of funds and their
manipulation. Others concentrate on the role of guns in politics, assuming
that the holder of guns will be ultimately responsible for the outcome of any
political contest. Organization also has its adherents, while many have seen
the battle of ideas as the arbiter of history. Chief Justice Oliver Wendell
Holmes referred to the American faith in the power of ideas when he said,
“When men have realized that time has upset many fighting faiths, they
may come to believe . . . that the ultimate good desired is better reached by
free trade in ideas—that the best test of truth is the power of the thought to
get itself accepted in the competition of the market. . . . That at any rate is
the theory of our Constitution.” 27 In fact, all of the political tools interact,
so that none alone will determine the outcome of a political contest.
Where one or more of the political tools can be neutralized, so that it will
not be available, or be equally available to either side in a political contest,
competition with the other tools takes on added significance. Thus, in a
political system where the role of guns in the normal political process has
been largely neutralized between political contenders, the other political
tools have taken on added importance. To read Theodore H. White’s The
Making of the President, 1960, for example, is to witness the operation of a
contest of ideas, of organization, and of funds in American politics.28 In a
political system where virtual agreement on ideas is also normal, the contest
may be in terms of which side can mobilize the most funds and have the
most efficient organization. (Arguably, if one genuinely wanted political
contests to become contests of ideas alone, neutralization of the role of
funds and organization, as well as of guns, would be necessary.)
The tools of politics are used by the leadership group to consolidate its
control, and to institutionalize a new system of political order. The sources
of these tools, however, are determined by the three variables described
below.
In the same way, the leader may draw on the old myth system to support his
own individual claim to leadership legitimacy.
In the predevelopmental system, many of the old sources of political
tools may have disintegrated or been discredited by the same circumstances
which brought the fall of the preexisting political system. In this case, the
political tools must be created anew. In such circumstances, external
assistance in creating these new tools may be sought by different leadership
contenders. And it is because of this that the role of external factors in the
predevelopmental system takes on such significance, and has had such a
great impact in the contemporary international arena.32
Because the sources of the tools of politics are not institutionalized in the
predevelopmental system, they may be easily supplied from external
sources. The system which has completed the predevelopment phase and
has institutionalized internal sources of these tools will be less likely to
desire or to tolerate intrusion of these tools from external sources. Because
of the temptation in the predevelopmental stage, and, indeed, sometimes the
necessity, of securing additional tools from external sources, the “political
vacuums” in these areas tend to become the focus of international conflict.
As Rustow notes, “During the formation of a new state . . . domestic and
foreign affairs are more closely intertwined than at any other time; the
whole task consists in drawing ... a boundary between the domestic and
foreign spheres.”33
While the interest of other countries in the outcome of the internal
political contest may well affect how the external actors behave, the
influence they actually exert upon the power struggle is restricted to the
concrete political tools they provide to, or withhold from, the various
political contenders.34 Frequently, external actors may be blinded by their
own myth systems to the extent that they are unable to recognize the
realities of the internal situation within the predevelopmental system to
which they are supplying external inputs. Whatever they may believe about
what is happening within the subject system, and however they conceive
the results of their assistance, their only effect upon the outcome of the
contest is through the political tools they supply. They cannot alter the
preexisting political behavior patterns or political mythology of the subject
system, nor can they prevent it from exhibiting the characteristics of a
predevelopmental system.
Because they are blinded or constricted by their own political myth
systems, the external actors frequently take steps which are either irrational
or simply not suitable within the framework of the predevelopmental
system. For example, an external actor supplying inputs may demand that
leaders alter the political behavior patterns of their followers or other
members of society, without first supplying sufficient tools to allow the
leadership to create a new institutional system which is a precondition for
enforcing such an alteration of behavior patterns. The external actors may
also supply one tool but withhold another—e.g., supply funds but withhold
guns—though both are necessary to the internal leadership group which
they are supporting. They may also require irrational concessions, such as
renunciation of the use of guns against an armed opponent in exchange for
supplying funds. The external supporter, seeing guns neutralized (at least
according to its own myths) within its own system, may reach the
conclusion that the use of guns is therefore inappropriate political behavior
on the part of its protégés in the predevelopmental system. Thus, while
external factors may be essential to the success of a group in a political
contest, where the internal sources are weak or held by another group, and
especially when other contenders are supported by external inputs from
other sources, these external factors may also introduce such inappropriate
or irrational elements as to frustrate totally the leadership group they
ostensibly support, and undermine its goals. Frequently, in the emerging
nations today, the result is an indefinite prolongation of the conditions of
predevelopment. Such unintentional extension of predevelopment as a
result of external inputs is what many frustrated leaders in the emerging
nations have termed “neocolonialism.”
The nonpolitical, or extrinsic, factors include those conditions in the
environment which, though extrinsic to politics, nonetheless determine the
effectiveness of the political tools.35 Education, technology,
communications facilities, and the like are all extrinsic to politics but will
influence the tools of politics available to the contestants.36
A nation may emerge from foreign domination, military defeat, or
revolution with weak internal resources—no organized army or police, no
organized political groups, little concentration of wealth, a fragmented
cluster of competing ideological systems—and yet possess nonpolitical
factors favorable (or unfavorable) to the quick formation of political tools.
An economically developed system is obviously more capable of
accumulating funds for political purposes than an economically backward
one. A literate populace which possesses various mass media may be more
readily indoctrinated with a unifying political ideology or idea system than
a populace fragmented by great impediments to communication. An
economically and technologically developed system may more readily
equip and train an armed force and an effective police force. Because of the
relation of these nonpolitical factors to the ability to create effective
political tools, these factors have often been used as measures of
development. They may well be such in the sense of modernization, but
they are not effective measures of the predevelopmental phase, for these
nonpolitical factors serve only to provide the raw materials of the tools of
politics. The growth of these factors may also, as some writers have pointed
out, be disintegrative in the predevelopmental phase, for the existence of
too many organizations, too many financially independent groups, and too
many competing idea systems, made possible by the expansion of economic
development, communications, and education, may make political
consolidation extremely difficult.37
Norman W. Provizer*
Like much of the “First Statement by the Soldiers,” this position though
based in fact, was clearly hyperbolic. Furthermore, the issue of elections was
not paramount in the military’s decision to act.3 Yet this discussion does
direct us to an extremely important area of comparative political
development thought—the relationship of the national electoral process to
state building.
In the eyes of many, this relationship is seen as an inverse one in which
elections are considered inimical, if not totally antithetical, to the
construction of a viable political framework. As one astute observer of
inchoate states has written: “In many, if not most, modernizing countries
elections serve only to enhance the power of disruptive and often reactionary
social forces and to tear down the structure of public authority.”4 From this
perspective, in which state building is viewed as the creation of
organizations that foster stability, unity, and economic development within
an environment of diversity and fragmentation, elections are perceived as an
activity which exacerbates, rather than ameliorates system divisiveness.
Substantive competition within these societies is severe enough without the
multiplier effect produced by legitimized procedural competition.
To a large degree, this position of the inverse relationship between the
electoral process and state building was reflected in the thought of the top
leadership stratum of Uganda, and thus the Obote regime could increasingly
be characterized as a “regime of development.”5 Understood in these terms,
it was hardly surprising that “... Obote failed to resort to the method of
legitimization [thought] most natural to the mechanism of democracy—
elections.”6
We would be guilty of unrealistic simplicity if we took this movement to
mean that no countertendencies existed within the state. At the same time
that national elections were recognized as having a negative impact on state
building, the regime also understood that “elections serve a useful purpose in
nation-building, in the mobilization of political consciousness and in the
control of leaders.”7
It was the interplay of these contradictory ideas that provided the context
from which the proposals for new methods of election in Uganda emerged.
These proposals were an attempt to reconcile the electoral process to the
perceived imperatives of state building, and thus have been aptly called “ . ..
the most original piece of constitutional theory to have emerged from
independent Uganda and one of the most challenging political experiments
to have been seriously considered anywhere in Africa.”8
With the fall of Obote and the rise of military rule in Uganda, the
proposals were never implemented, as the very prospect of a national
election itself was eliminated. This fact limits our discussion of the new
methods of election to what was proposed and what might have been
accomplished, but given the important implications of these proposals for
comparative political development thought, even such a conjectural analysis
is not without merit.
Since the proposals themselves did not emerge from a vacuum, the first
part ôf this study will outline: the national elections of 1961 and 1962; the
problems which these elections were thought to create for state building; and
the revisions in the electoral process which followed the constitutional crisis
of 1966. After establishing this base, we will turn to an examination of the
proposals for new methods of election and an evaluation of their meaning for
what Sidney Verba has termed “ . . . the most basic questions of politics—the
entire set of questions involved in the creation and maintenance of political
societies.”9
Diagram 1
This plan for “ ... a unique and complex experiment in multiple
constituency elections”63 was presented by Obote to the National Council of
UPC in mid-August 1970. During this meeting in Kampala, it became
apparent that, despite Uganda’s movement toward a consolidationist, de
facto one-party system, the ruling UPC remained far from a monolithic
entity. The National Council accepted the new method of election suggested
for National Assembly members, but rejected paragraphs 26-34 of the
“Proposals ...” which dealt with the selection of the republic’s president. In
taking exception to these paragraphs, the National Council stated that from
its perspective the president of the party must automatically be the president
of the state, and that the changes put forward in the “Proposals...” could only
be adopted by the Annual Delegates’ Conference of UPC.64
Thus Obote’s desire, as president of the country and of UPC, to have, for
both philosophical and pragmatic political reasons, the aura of popular
legitimacy surrounding the office of the president was thwarted by the
Council.65 To some observers this meeting was a subcutaneous maneuver
executed by people like Felix Onama (the minister of defense and secretary-
general of UPC) and intended to undermine Obote’s position within the
party.66 The president may have been the only leader that an independent
Uganda had known, but his role within the state and party was not beyond
challenge.
Two weeks after the National Council meeting, the Delegates’ Conference
of UPC assembled in Mbale to reach a “final’ ‘ decision on the proposed
method of election of the republic’s president. In a “Memorandum” dated
only four days before the conference, Obote discussed the position taken by
the National Council and reiterated his own views on the importance of
electoral legitimacy for the presidential candidate so as not to “ . . . alienate
the Party from people.”67 Obote expressed willingness to amend some
paragraphs of the “Proposals ...” and accepted the idea that the party
president should also be the republic’s president, but on the necessity of a
popular vote he remained firm.68
The Delegates’ Conference supported the National Council’s nonelectoral
position on the presidency, and despite the “Memorandum,” Obote’s
arguments were not heeded.69 Discussions of this crucial issue, however, had
not yet ended, and a compromise arrangement was accepted by the National
Council and the Delegates’ Conference held during December 1970 in
Mbale.70 The compromise arrangement shifted responsibility for the
nomination of presidential candidates from the single Delegates’ Conference
to the 96 Parliamentary Constituency Conferences. Thus any party candidate
for the presidency of the republic who received the support of 32 (i.e., one-
third) of these Parliamentary Constituency Conferences would become an
official party nominee for president of the country. If two or three candidates
were nominated in this manner, a popular vote would determine the
winner.71 In this way the compromise did allow for the option of popular
participation in the selection of the state’s president, although the delineation
of such specific circumstances under which citizen participation would take
place constituted a significant variance from the intent of the original
position set out in the “Proposals
By the end of 1970 the new system of elections was ready to be put into
practice. Lots had been drawn to link “national” constituencies to each
“basic’ ‘ constituency, and the Electoral Commission, chaired by Ateker
Ejalu, had filed its report on the makeup of the 96 parliamentary
constituencies. A Presidential Electoral Commission (whose members were
nominated by the president) was set up to provide the mechanism through
which the slate of candidates for each constituency would be prepared for
the voting public of Uganda; but the elections scheduled under this system
for April 1971 were not to be.72
Conclusion
The “Proposals. ..” developed within the context of the national election/
state-building controversy and utilized “... the written word for the purposes
of formulating new social directions and new political goals.”73 Since the
procedural competition of elections was viewed as an accelerator of the
substantive competition which fragmented the state, it was considered
necessary to alter the form of elections in order to further the process of state
building.
To some, such an alteration could only be meaningful if it resulted in a
total abjuration of procedural competition; while others held the position that
the system of national elections could be transformed so as to eliminate the
problems created for state building without extracting all aspects of popular
legitimacy. The plan proposed by Obote followed this latter manipulative
course, and offered a method of reconciling the imperatives of state building
with the perceived need for elections.
With the overthrow of the Obote regime, the words of document five in
the “move to the Left” were never translated into action. Even before the
military coup d’état occurred, there was some question as to whether the first
national election since independence would actually be held. Given the
volatile nature of African politics, no certain answer can be provided for this
question; but it does not seem unreasonable, in light of the branch and
constituency elections of 1970 and the large-scale debate over the “Proposals
...,” to assume that the elections for the National Assembly would have taken
place.74
Though the new method of elections obviously reflected an increasingly
consolidationist perspective and furnished numerous safeguards for the
Obote regime, it is not so clear that this plan completely “preordained” the
election and was merely a “trick ... to keep [Obote’s] henchmen in power.”75
In short, it appears that the national elections were intended to cohere
national leadership, revitalize the organization of UPC, offer limited
accommodation to Buganda, and legitimize the move toward a one-party
state.76 Under the 1 + 3 system, all of this was to be accomplished while the
actual parameters of procedural competition were narrowed.
Needless to say, the change in procedural competition spelled out by the
new electoral system was not without its critics. To many, the new method of
elections meant an overwhelming advantage to those candidates with wealth
and exposure; a dilution of the quality of representation due to the lack of
contact with a specific constituency; and an unhealthy increase in the
formation of pragmatic alliances among candidates.77 In questioning
whether the social gains produced by the “Proposals . . .” compensated for
the problems created by the document, commentators also pointed to the
language difficulties built into the system and the “error” of not weighting
the vote in the basic constituency over votes in the national constituencies.78
Beyond these real but not insurmountable problems, there remained even
more fundamental difficulties. Difficulties which were primarily related to
the persistent pattern of substantive competition which continued both
between identity groups and within the ruling party itself. Although the
“Proposals ...” presented a method of election which would have centralized
and “nationalized,” to a degree, the selection process for members of the
National Assembly, it was unlikely that the implementation of the
“Proposals. . .” would have been able, in itself, to eliminate either parochial
tendencies within the state or intraparty fractionalization. The very attempt
to eradicate the ethnic-tribal factor may itself have led to the resurgence of
other parochial predispositions (viz., religion).79 The translation of the
rhetoric of unity into reality would not be so easily achieved, especially
since the institutional power of the center exhibited far from total dominance
over the system as a whole.
Despite the existence of such problem areas the “Proposals. . .,” in its
recognition of the positive as well as the negative dimensions of the electoral
process, deserves to be described as an “original piece of constitutional
theory.” This is not to say that the structure of the single party, upon which
the “Proposals . . .” was built, is the “... only viable alternative to multi-party
anarchy or government by brute force,”80 or that the mechanism of the one-
party state is not perverted to support the practical goal of the retention of
power within a political society.81 What it is meant to imply is simply that
the “Proposals . . .,” both as a symbolic expression and as a pragmatic
policy, offered a viable approach to the continual search for that illusive
balance point which lies between freedom and order.
NOTES
1. Michael Twaddle, ‘ The Amin Coup, ‘ ‘ Journal of Commonwealth Political Studies, X (July 1972),
99-112; and David Martin, General Amin (London, 1975), pp. 44-45, 158-61. Also see Cynthia Enloe,
“Ethnicity and the Myth of the Military in African Development,’’ Ufahamu, IV (Summer 1973), 35-
36; and Ali Mazrui, “The Lumpen Proletariat and the Lumpen Militariat.” Political Studies, XXI
(March 1973), 1-4.
2. The Government of Uganda, The Birth of the Second Republic, (Entebbe, n.d.), p. 27.
3. See Padma Spinivasan, “Uganda’s Malaise: An Attempt at Coup Analysis,” Africa Quarterly ,
XIII (July-September 1973), 86-104; Michael Martin, “The Ugandan Military Coup of 1971 : A Study
of Protest,” Ufahamu, III (Winter 1972), 81-121; and Michael Lofchie, “The Political Origins of the
Uganda Coup,” Journal of African Studies, I (Winter 1974), 464-96.
4. Samuel Huntington, Political Order in Changing Societies (New Haven, 1968), p. 7. “The
problem,” in Huntington’s words, “is not to hold elections but to create organizations.”
5. Richard Lowenthal, “Government in the Developing Countries,” in Henry Ehrmann, ed.
Democracy in a Changing Society (New York, 1965), pp. 195, 204, 206, 208.
6. Spinivasan, 97.
7. From a statement made by Basil Bataringaya in November 1970, quoted in Colin Legum, ed. Africa
Contemporary Record 1970-71 (London, 1971), p. B191.
8. Mazrui, Cultural Engineering and Nation-Building in East Africa (Evanston, 1972), p. 70.
9. Sidney Verba quoted by B. Schaffer in “The Concept of Preparation: Some Questions about the
Transfer of Systems of Government,” World Politics, XVIII (October 1965), 42.
10. See Norman Provizer, Politics and the Political Economy of Penetration in Uganda (Ph.D. diss.,
1974), pp. 70-88.
11. Report of an Inquiry into African Local Government in the Protectorate of Uganda (Entebbe,
1953), pp. 14-15.
12. Report of the Constitutional Committee, 1959 (Entebbe, 1959), p. 46.
13. G.S.K. Ibingira, The Forging of an African Nation (New York, 1973), p. 186.
14. Joseph Nye, Jr., Pan-Africanism and East African Integration (Cambridge, 1965), p. 107.
15. See D. A. Low, Political Parties in Uganda, 1949-1962 (London, 1962), p. 24; A. Hughes, East
Africa: The Search for Unity (Baltimore, 1963), p. 176; and Terrance Hopkins, “Politics in Uganda:
The Buganda Question,” in Jeffrey Butler and A. Castagno, eds. Boston University Papers on Africa
(New York, 1967), p. 256.
16. David Apter, The Political Kingdom in Uganda (Princeton, 1967) p. 374-75.
17. Donald Rothchild and Michael Rogin, “Uganda,” in Gwendolen Carter, ed. National Unity and
Regionalism in Eight African States (Ithaca, 1966), pp. 353, 380.
18. Ibid., p. 353; and Audrey Richards, “Epilogue,” in Fallers, ed. The King’s Men: Leadership and
Status in Buganda on the Eve of Independence (London, 1964), p. 381.
19. See Martin Lowenkopf, “Uganda’s Prime Minister Obote,” Africa Report, VII (May 1962), II;
Low, Political Parties, pp. 30-31; Apter, Political Kingdom, pp. 347-48; and F. Welbourn, Religion
and Politics in Uganda, 1952-1962 (Nairobi, 1965), p. 21.
20. Robert Byrd, “Characteristics of Candidates for Election in a Country Approaching Independence:
The Case of Uganda,” Midwest Journal of Political Science, VII (February 1963), 4.
21. Ibid., p. 27. Outside of Buganda, the figures were 50.3 percent and 41.3 percent, respectively.
22. Low, p. 47.
23. C.J. Gertzel, “How the Kabaka Yekka Came to Be,” Africa Report, IX (October 1964), 9-10.
24. Report of the Uganda Relationships Committee, 1961 (Entebbe, n.d.), p. 55.
25. Nye, Pan-Africanism, pp. 107, 109; Welbourn, Religion and Politics, p. 27; and Fred Burke, Local
Government in Uganda (Syracuse, 1964), p. 19.
26. The UPC supported the idea of indirect elections in Buganda for this election. See the “Policy
Statement of the UPC,” mimeo (1962), p. 2. Consult Byrd, 23-24, on the terms of the alliance between
UPC and KY.
27. R.C. Peagram,/! Report on the General Elections to the National Assembly of Uganda Held on the
25th April, 1962 (Entebbe, n.d.), pp. 23-28.
28. Ibid., pp. 29-33; and Byrd, 27. The elections in two of the Toro constituencies took place a week
after the general election due to protests over federal status for the Kingdom of Toro.
29. Of the nine specially elected seats, three went to KY and six to UPC.
30. See the background provided in the Institute of Public Administration, “Evolution of Local
Administrations and District Councils,” mimeo (Kampala, 1967), pp. 9-10.
31. Quoted in Low, “Buganda and Uganda,” in Low, Buganda in Modern History (London, 1971), p.
240.
32. Gertzel, “Kabaka Yekka,” 12-13; and Mazrui and G.F. Engholm, “The Tensions of Crossing the
Floor in East Africa,” in Mazrui, Violence and Thought (New York, 1969), p. 126.
33. Rothchild and Rogin, “Uganda,” pp. 389-91; and Gertzel, “Report from Kampala,” Africa Report,
IX (October 1964), 5-6.
34. Colin Leys, Politicians and Policies: An Essay on Politics in Acholi, Uganda, 1962-65 (Nairobi,
1967), p. 18.
35. Hopkins, “Politics,” pp. 263, 279; and George Shepherd, Jr., “Modernization in Uganda: The
Struggle for Unity,” in Stanley Diamond and Burke, eds. The Transformation of East Africa (New
York, 1966), pp. 330-33.
36. Martin Staniland, “Single-Party Regimes and Political Changes,” in Leys, ed. Politics and Change
in Developing Countries (Cambridge, 1969), pp. 137-38.
37. See Emory Bundy, “Uganda’s New Constitution,” East Africa Journal, III, (June-July 1966), 26.
38. See A. Milton Obote, “The Footsteps of the Ugandan Revolution, “ East Africa Journal, V,
(October 1968), 12; and Mazrui and Engholm, “Tensions,” pp. 122-46.
39. According to the UPC-KY alliance, UPC was not to actively organize in Buganda, but by early
1963 UPC was opening branches in the Kingdom. Ibid., 127. On previous splits within the internal
structure of UPC-Buganda, see Apoio Nsibambi, “Political Integration in Uganda: Problems and
Prospects,” East Africa Journal, VI, (February 1969), 38.
40. W. Arthur Lewis, Politics in West Africa (New York, 1965), p. 60, argues that “Singlepar-tyism is
logically attainable if the party has a single ideology to which all of its members adhere.”
41. Institute of Public Administration, “Evolution of Local Administrations,” p. 11.
42. M. Davies, “Central/Local Government Relations,” Institute of Public Administration, mimeo
(Kampala, 1966), p. 2.
43. “The Development of Political Parties in Uganda,” mimeo (Kampala 1966), p. 13.
44. For a brief review of this crisis, see M. Crawford Young, “The Obote Revolution,” Africa Report,
XI (June 1966), 8-14. An interesting, though far less objective, portrait of the crisis is painted by
Akena Adoko, Uganda Crisis (Kampala, n.d.), pp. 23-112. Also useful is The Government of Uganda,
The Birth, pp. 6-15.
45. The commission consisted of: Sir Clement deLestang of the East Africa Court of Appeals; Justice
E. Miller from Kenya; and Justice A. Saidi from Tanzania. The report was not made public for several
years. Before the 1971 coup, Obote had finally decided to publish it and Amin allowed publication to
take place after he took power. See Martin, Amin, p. 89.
46. The five ministers arrested were G. Ibingira, E. Lumu, G. Mugezi, B. Kirya, and M. Ngubi.
Obote’s statement can be found in The Government of Uganda, The Birth, p. 7.
47. The KY members, led by A. Sempa, who refused to take the oath had their seats declared vacant.
The DP members, who had walked out, later took the oath and were seated.
48. SeeThe Constitution of Uganda, 15 April 1966 (Entebbe, n.d.), Chapter V, Part III, 73. (5).
49. Uganda Parliamentary Debates (Hansard), 2nd ser., LXXVII (September 1967) (Entebbe, n.d.), p.
1631; and The Government of Uganda, The Birth, p. 17.
50. J. Ochola, “Consultations between the Government and Local Authorities,” mimeo (July 1970), p.
6.
51. O.G. Ivorian Jones, “The Legal Framework ofthe District Administration in Uganda,” The
Institute of Public Administration, mimeo (Kampala, June 1970), p. 2.
52. See The Constitution of the Republic of Uganda (Entebbe, n.d.), chap. 4, pt. 1, 26. (1) (a)-(e).
53. Ibid., chap. 5, pt.l, 40. (1) (b), (2), (3). The original proposals for the Republican Constitution
offered a somewhat different system in which the nominated members would be selected by the
president and could number one-third of the total of elected members.
54. In Acholi, for example, in March 1968 the UPC secretary-general suspended several party
officials for making statements against the government. A few days later, the minister of regional
administrations dismissed them from their positions on the District Council.
55. Obote, The Common Man’s Charter (Entebbe, n.d.), Forward.
56. See the Uganda Parliamentary Debates (Hansard), 2d ser., XCVIII (December 1969) (Entebbe,
n.d.), esp. pp. 101-7; the vote was 60 to 0 in favor of the extension of the state of emergency.
However, this did not mean that all DP members joined the UPC; see the comments of Latim and
Okelo in the Uganda Argus, 17 July 1970, p. 1.
57. Documents two (“Proposals for National Service”), three (“Communication from the Chair of the
National Assembly”), and four (“Labour Day Speech”) can be found as appendices to Obote, Common
Man’s Charter, pp. 12-43. For document five, see Obote, Proposals for New Methods of Election of
Representatives of the People to Parliament (Kampala, n.d.).
58. Mazrui, Cultural Engineering, p. 70.
59. Obote, Proposals, p. 14, par. 28. The presidential candidate would either be the party’s president at
that time, or a person selected by the Delegates’ Conference of the party.
60. If the candidate was not elected, the National Assembly would select one of its members to be an
interim president for up to one year. See ibid., p. 18, par. 34 b.
61. The National Assembly (Prescription of Elected Members) Act, 1970, set the number of elected
MPs at 96, an increase of 14 over the then current number. The delineation of the 96 constituencies is
presented in The Government of Uganda, Report of the Electoral Commission (Entebbe, n.d.). The
electoral regions basically followed the already established regional division within the state.
However, in August 1970, Teso North-West, East Mengo North, and Bunyoro North were placed in
the Northern region to give all four regions 24 constituencies. See the Uganda Argus, 14 Aug. 1970, p.
1.
62. Obote, Proposals, p. 22, par. 40. See also Peter Willetts, “The Mathematics of Document No. 5,”
Uganda Argus, 12 Aug. 1970, pp. 5-6.
63. Special correspondent, “Uganda Under Military Rule,” Africa Today, XX (September 1973), 12.
64. See Obote, “Memorandum,” in his Proposals, pp. ii-iii, pars. 3-4; D. Cohen ana J. Parsons, “The
Uganda Peoples Congress Branch and Constituency Elections of 1970,” Journal of Commonwealth
Political Studies, XI (March 1973), 52; mdUganda Argus, 14Aug. 1970, p. 1.
65. Ibid.; and Garth Glentworth and Ian Hancock, “Obote and Amin: Change and Continuity in
Modern Ugandan Politics,” African Affairs, LXXII (July 1973), 245-46. For a brief discussion of the
act of voting as an integrating device, see A.J. Milnor, Elections and Political Stability (Boston, 1969),
p. 99.
66. Legum,4/nca 1970/1971, p. B187; and Legum, ed. Africa Contemporary Record 1971172
(London, 1972), p. B227.
67. Obote, “Memorandum,” p. xiii, par. 18. See also comments on the “Memorandum” in the Uganda
Argus, 22 Aug. 1970.
68. Obote, “Memorandum,” pp. vii-ix, par. 12.
69. Cohen and Parsons, 53, and the Uganda Argus, 29 Aug. 1970.
70. Both of these bodies had been “reconstituted” following the party elections of October-November
1970; see Cohen and Parsons, 53. Also see the Uganda Argus, 17 and 21 Dec. 1970.
71. Kampala Domestic Service, 15 December 1970, reported in Foreign Broadcast Information
Service, Daily Report: Middle East and Africa, 16 December 1970, p. U4.
72. See Obote, Proposals, p. 20, par. 37c; Cohen and Parsons, “Uganda Peoples,” 49; and Legum,
Africa 1970/1971, p. B191.
73. Mazrui, Cultural Engineering, p. 67.
74. This is also the opinion expressed by Cohen and Parsons, 54. An example of the contrary view can
be found in a press statement issued on 27 January 1971 by Idi Amin Dada, the head of Uganda’s
military government, and reprinted in Martin Minogue and Judith Molloy, eds. African Aims and
Attitudes (Cambridge, 1974), p. 363.
75. On the election as a “shame’ ‘ see Peter Gukiina, Uganda: A Case Study in African Political
Development (Notre Dame, 1972), pp. 170-72.
76. Legum, Africa 1970/71, p. B187, points to the election as a means for Obote to rid himself of
many of his comrades and potential rivals. Lofchie, “The Uganda Coup: Class Action by the Military,”
Journal of Modern African Studies, X (May 1972), 31, writes of the election in terms of legitimizing
the one-party system. And Legum, Africa 1971/1972, p. B226, discusses Bugandan participation in the
party elections of 1970.
77. See Cohen and Parsons 50-51 (and 52, where they note the negative aspects of exposure).
78. See the comments of Makerere professors Ryan and Gomboy in the Uganda Argus, 12 Aug. 1970,
pp. 1,4. According to Donald Morrison et al., Black Africa: A Comparative Handbook (New York,
1972), p. 27, Uganda ranks 27th out of 32 countries in terms of the “percentage of its population
speaking the dominant vernacular language” (which is Luganda). Unlike its neighbors Kenya and
Tanzania, Swahili as a lingua franca is not widespread in Uganda.
79. Mazrui, “Piety and Puritanism under a Military Theocracy: Uganda Soldiers as Apostolic
Successors,” in Catherine McArdle Kelleher, ed. Political Military- Systems: Comparative
Perspectives, Sage Research Series on War, Revolution, and Peacekeeping, IV (Beverly Hills, 1974),
p. 114.
80. See Benjamin Neuberger, “Has the Single-Party State Failed in Africa?” African Studies Review,
XVII (April 1974), 178. For a critical analysis of the one-party system, see S.E. Finer, Comparative
Government (Harmondsworth, 1970), pp. 506-73.
81. Ibid., p. 528.
*The author would like to express his appreciation to the Anspach Institute at the University of
Pennsylvania, the Makerere University Library, and the Makerere Institute of Social Research. He is
also grateful to Mr. James Ellis and Dr. Lucy C. Behrman for their assistance.
Democracy and Social Mobilization in
Lebanese Politics
Michael C. Hudson*
Scholars and policy-makers frequently assume that the new states must
develop stability before they can hope to operate the complex and delicate
institutions of liberal democracy. But in the case of the Lebanese Republic
this relationship is reversed: democratic institutions have brought about and
maintained stability in an unfavorable political environment. The Lebanese
case suggests that formal institutions, although neglected in behavioral
political science, deserve renewed attention as causal agents in the process of
political modernization.1 At the same time, it raises the question of whether
such institutions can supply enough systematic flexibility to meet the social
mobilization demands of a rapidly changing society.
Lebanon’s fragmented body politic—its traditional pluralism—has
necessitated a political system based upon the balance of power, in the
absence of positive legitimacy for the institutions of the state. In turn, the
balance of power has required institutions that promote democratic values.
Without the democratic institutions the balance of power would cease to be
stable, and without stability the state would cease to exist. Lebanon’s
representative institutions are an essential condition of its stability, not a
lucky byproduct. This relationship, of course, has not gone unnoticed by
Lebanon’s politicians. In his inaugural address of September 23,1964,
President Charles Helou said: “I believe that the democratic system is an
intrinsic necessity for our country.... It assures a balance between powers and
makes possible a fruitful meeting among the Lebanese spiritual families.
Thus their energies are activated within the democratic foundations, and
their needs are met within a framework of brotherly cooperation. Rule is
consultative, and the consultative system in Lebanon is one of the conditions
of cooperative and stable life.”2
This situation, however, is no cause for unrestrained optimism on the part
of advocates of moderate democracy. First, the Lebanese system is
democratic only in a limited sense. Second, modernization creates a dilemma
for this balance-of-power system because it burdens a weak governmental
apparatus with additional social demands. If the political system becomes
capable of handling these demands and more responsible to the people in
general, then it will threaten the existing democratic values that are a product
of the balance of power among autonomous traditional groups. Yet attempts
to maintain the status quo may run afoul of rising popular demands and the
demonstration effect of efficient administration in other countries.
Social mobilization
Deutsch’s concept of social mobilization is perhaps the most elaborate
indicator of the strength of social change.24 Comparisons with Deutsch’s
hypothetical model of a country undergoing rapid social mobilization
suggest that Lebanon’s population is shifting rapidly into the category of
socially mobilized. Lebanon scores within the ranges Deutsch postulates on
all indicators of changes (for which data are available) that make increased
demands on the political system: shifts into mass-media audience, increase
in voting participation, increase in literacy, and population growth.25 As for
that set of changes “related to the capabilities of the government for coping
with these burdens,” Lebanon again falls within the hypothesized ranges: the
occupational shift out of agriculture, the change from rural to urban
residence, and the income-growth and per capita income-growth indicators
leave no doubt that social mobilization is occurring.
Table 1 summarizes certain trends indicative of social mobilization in
Lebanon in the period of its independence. Even casual inspection of the
figures shows clearly that Lebanon is a society in the process of rapid
modernization. Taken together or separately, the data indicate the
development of a typically complex urban society: the trends are toward
urban density, a certain material advancement, and an increasingly
politically aware population. But what do these numbers signify about
political loads?26
Problems of modernization
If the political issues posed by social mobilization are serious, Lebanon’s
capacity for dealing with them appears meager indeed by all the usual
standards of political modernization. Lebanon ranks well up among the
upper half of all the countries in the world on selected indicators of social
mobilization, yet it falls into the lower half on “political development”
indicators such as political deaths per million, political party representation,
government expenditures as a percentage of GNP, voting participation, and
executive stability.31 Specific countries that rank near Lebanon on social
mobilization rank generally above it on the political indicators, while its
political “neighbors” exhibit generally lower levels of social mobilization.
Observers and practitioners of Lebanese politics alike condemn the
corruption and inefficiency of political life, although it is by no means clear
that Lebanese are any more “developed” in these qualities than, say, Italians
or Americans. However, two weaknesses in Lebanon’s performance as a
political system are particularly notable: the country has hardly been able to
develop and integrate nontraditional interest groups and parties; and its
vulnerability to sudden crises remains extremely high. The system has
succeeded in integrating, in a functional sense, only those groups and
organizations compatible with the requirements of the balance of power.
These include the pressure groups representing Lebanon’s overdeveloped
financial sector, which influence tariff policies, public works programs, and
general domestic policy, and the so-called parties, such as the Destour and
the National Bloc, which are in reality only the followings of certain
traditional figures. The new interests generated by economic growth have
been accommodated within the traditional framework through an
interlocking directorate of notables.
Centralized parties too must conform to the traditional requirements. For
example, the Kataeb (Phalanges Libanaises), established during the late
1930’s and the only significant nonclandestine party in operation today, is a
vehicle for maintaining Maronite interests in an independent Lebanon.32 Yet
even the Kataeb, in its most successful year (1960), placed only six men in a
parliament of ninety-nine members; and in 1964 it placed only four. It is
only occasionally represented in the cabinet. But perhaps the classic case is
Kamal Jumblat’s Progressive Socialist party. It is distinctly unsuccessful as a
party, as Jumblat himself has admitted, but it is effective as a regional-
sectarian grouping. The PSP is an ideological cover for Jumblat’s personal
following as the dominant leader of the Chouf district and the hereditary
chief of one of the two Druze factions. This party currently commands one
of the largest followings in parliament—ten deputies. But its programs33 are
too radical—at least in the Lebanese context—to command serious attention
from the establishment. The Lebanese are unlikely to permit the government
to nationalize industries, curtail freedom of trade, institute effective
progressive taxation, and “reorganize society on the basis of the family.”
Parties of a more radical cast have appeared, but have never attained
legitimacy because they deny the traditional requirements. The Social
Nationalist party (formerly the Syrian National party) demanded a
revolutionary adaption to modernization: secularism, social reform, a
“greater Syrian” nationalism, and income redistribution. It drew membership
from various minority sects, organized itself along paramilitary lines in the
fashion of the Fascist and Nazi parties, and was guided by intellectuals—
university professors, students, and lawyers. This party was never able to
operate within the traditional limits of the system. Only once did it place a
man in parliament, and never were its policy recommendations seriously
considered. The party was felt to be so dangerous to the political structure
that President Khoury’s regime dissolved it in 1949 and executed its leader,
and the Chehab regime tried to destroy it again in 1962 when it attempted a
coup d’état. The other nontraditional parties, the Communists and the
Baathists, are also beyond the pale and function as semiclandestine
organizations. In this category also fall the Arab nationalist clubs and
societies, although Nasserist sentiment is sufficiently widespread to assure
them of a broader measure of tolerance. Again, the membership appears to
come from the young, socially mobilized, discontented intellectuals of
Beirut, Tripoli, and the Mountain.
The second problem in Lebanon’s political modernization is also a
product of the growing politicization of its inhabitants. It is sometimes said,
with some justification, that Lebanon is like a powder keg: a sharp
disturbance may make it explode. While this quality undoubtedly deters
some of those who desire to tamper with the political system, it obviously
cannot guarantee complete deterrence, particularly in the turbulent Arab
world. A rock thrown at a curé, or an inflammatory radio speech, can and
does shock the system to its foundations. The Lebanese public is highly
exposed to various transnational ideologies that unavoidably compete with
allegiance to the Lebanese state. Public demonstrations, general strikes, and
attempted coups d’état weaken the balance and those democratic values that
flow from it. Political assassinations, while infrequent, present a severe
threat to the delicate balance when they do occur. Labor unions are
becoming politically significant as they develop their organizational power.
Twice during the 1950’s, institutional devices proved unable to resolve a
major crisis and the “solution” was spurred by a general strike. In 1952 the
Khoury regime found that it had alienated a strong coalition of notables, and
President Khoury was unable to find a prime minister: none of the eligible
Sunnites would serve. The threat of a general strike and disorders persuaded
the President to resign. In the impasse Khoury naturally called upon the
army to maintain order, but the commander, General Chehab, would not
commit his forces to a “political” struggle. As a result the regime fell. Owing
to the deterrent built into the traditional system, civil and confessional
disorder was avoided and the old institutions were preserved. But the 1958
crisis was more serious. Pan-Arabism was stronger than it had ever been.
The Chamoun regime seemed to be violating the neutrality principle of the
National Pact, denying the traditional representation of notables, and trying
to amend the constitution to permit Chamoun’s reelection. Chamoun became
the ruler of a country polarized by the pro-Western, Christian faction, which
controlled the disintegrating coercive instruments of the state, and the pro-
Nasser insurgents, consisting of many of the leading notables. Once again
General Chehab refused to use the army to save the regime. American
intervention helped bring to an end a summer of violence.
It would be naïve to underestimate the gravity of these problems, yet it is
an oversimplification to suppose that the Lebanese system is incapable of
responding to them. The fact that the state has weathered the crises of its
first two decades, showing remarkable persistence and durability, is surely
significant in itself and can hardly be attributed to the United States Sixth
Fleet. Why, if the political system is as anachronistic in its environment as
theory and data indicate, does it survive as well as it does? The political
scientist pondering this question is inevitably reminded of the visiting
economist who, when asked for advice, replied, “I don’t know what you are
doing, but whatever it is, keep it up.” No simple answers seem adequate, and
any analysis that takes account of the subtleties of the Lebanese situation is
likely to appear unduly hedged to some readers.
Despite its serious problems, the Lebanese system has two important
characteristics that mitigate to some extent its weaknesses. Both provide
limited political means for coping with a dynamic situation; but, as we shall
see, neither can be expected to resolve the fundamental dilemma of political
modernization in Lebanon. Although ideal political modernization is
virtually impossible under present conditions, these two positive factors
supply a certain capacity for muddling through. One of these factors may be
described as increasing institutional strength; the other is a degree of
democratic procedure.
The most important sign of growing institutionalization in Lebanon has
been the growth in power and capability of the executive branch. As we have
seen, the president of the Lebanese republic has considerable formal power.
During the brief period since independence this power has steadily
increased; yet, with the exception of the 1958 crisis, it has not increased
dramatically enough to set the traditional notables into open revolt. Bechara
al-Khoury, the father of Lebanese independence, succeeded in welding most
elements of the elite together with all the skills of Lincoln and Boss Tweed.
Despite his inglorious retirement in 1952, his was a considerable
accomplishment. Camille Chamoun, who had the misfortune to rule during
the turbulent period 1952-1958, was if anything too aggressive; and his bid
to reform the “feudal” system was singularly tactless. During the postcrisis
regime of General Fuad Chehab, Lebanon experienced its first taste of
planning for national development and a degree of military interference in
politics and administration. Expenditures of the Public Works Ministry
nearly tripled between 1961 and 1964, and the state began to engage in
deficit spending.34 The number of civil servants rose from around 18,000 in
1957 to 26,000 at the end of his regime. “Chéhabisme” became synonymous
with national development and represented perhaps the least controversial
position between the advocates of the “free economy” and the advocates of
“Arab socialism.” Chehab’s successor, Charles Helou, a compromise
candidate in the tense succession struggle, although less dynamic than his
predecessors, still managed to guide Lebanon through the Intra Bank
collapse of 1966 and the Arab-Israeli war of 1967—crises of the first
magnitude. That a relatively weak leader could weather such storms is due in
part to the growth in the effectiveness of the presidential office and the
resilience of the system as a whole.
Institutionalized representative processes have also shown a notable
expansive capacity, though perhaps a less spectacular one than the growth of
the executive branch. Lebanon’s growing politically conscious population is
reflected in the increase in voting participation. The ratio of voters to total
adult population has risen markedly since 1943, the most spectacular
increase occurring in 1953 after the granting of suffrage to women. In the
1960 parliamentary elections, some 522,000 people voted, and by 1964 the
number had increased by around 14 percent to 595.000.35
The composition of the Chamber of Deputies is also changing. Especially
notable is the declining percentage of great landowners. In the first
representative council under the Mandate (1922-1925), landowners
accounted for two-thirds of the membership; in the 1943 parliament the
percentage was 44; in the 1960 parliament it was 22; and in the 1964
parliament it was down to 15. While the lawyers maintained a fairly constant
level of representation (around 30 percent), the big businessmen—financiers
and the professionals (including teachers, engineers, former judges, and civil
servants) increased their representation.36 The change is also reflected in the
differing political atmospheres of the 1940’s and the 1950’s. During the
earlier period politics was a game played by traditional notables; during the
latter the game was opened up to some new players from the upper middle
class and the non-Christian sector, and several of the notables, especially
those of Beirut, adopted ideological means to consolidate and increase their
popular following. Although the new complexities added strains to the
system, the institutions proved capable of operating under them. Admittedly,
these trends are far less impressive than many observers might wish;
everybody has his anecdotes about corruption, and nobody, including the
major participants, has any illusions that the system works well. These
qualifications, however, do not obscure the evidence that the system is far
from static: even if traditional pluralism retards institutions such as a party
system and perhaps checks presidential power too much, it has not robbed
executive and legislature of limited capacities to develop.
The second factor ameliorating Lebanon’s theoretically underdeveloped
political condition is the same one discussed earlier as functional for
maintaining the internal balance of power. But it is easier to argue that
democratic procedures permit a stable balancing process than it is to contend
that they contribute to the development of the political capabilities that
social mobilization requires. Indeed, it can be argued that the kind of
democratic procedures found in Lebanon are actually dysfunctional in this
respect, in that they tend to paralyze the policy process.37 The issue boils
down to the question of whether Lebanon’s limited democratic procedures
serve only to lubricate the balance among sects and traditional factions.
During the Mandate and in the early years of independence Lebanese
democracy could generally be described in those limited terms. But 1952 —
the year of the Egyptian revolution and the deposition of President Khoury
—marked a turning point for Lebanon. From that time on the Lebanese
people and elite alike seem to have become conscious of the imperatives of
development, social justice, and national dignity which social mobilization
has precipitated. What was hitherto primarily sectarian democracy has begun
to acquire elements of populist democracy.
In domestic affairs these concerns are reflected in the persistent efforts at
administrative reform and, especially during the Chehab regime, in
government-directed development policies such as the social security
program, the Central Bank, the Social Development Office, and the
agricultural development program (“The Green Plan”). In foreign affairs
they may be seen in the abandonment of the traditionally close Western
relationship so carefully nurtured by the Maronites, as well as in the
acceptance of Lebanon as an Arab state whose destinies, for better or worse,
are inextricably tied to the Arab world. These policies are evidence that the
political system has become responsive to newly crystallized populist
demands that transcend the primordial, sectarian perspective of the early
democracy. The attempts to promote development, social justice, and closer
Arab solidarity have not arisen primarily out of sectarian pressures; they
have come out of peculiarly modern political conditions and they have been
transmitted, albeit imperfectly, into public policy through democratic
procedures. The events of the last decade suggest that while sectarianism at
the popular level shows no signs of disappearing, the political system can to
some extent transcend the confessional, corporate democracy.
That this “remedy” should have emerged may be explained by precisely
the same condition which gave rise to the old, sectarian democracy. This
condition is insecurity. Today, however, in the modernized and turbulent
Arab world, insecurity for the inhabitants of Lebanon is not confined to
historical sectarian and parochial fears. Modernization generates new and
broader insecurities: fears of the violence that can arise out of ignoring the
modern ideas now prevalent throughout the Arab world, in “progressive”
and “conservative” states alike. However imperfect Lebanon’s institutions
may be, its politicians cannot ignore new demands. For Lebanon more than
most states, survival today requires political responsiveness to broadly
populist demands, and this fact is recognized by Lebanese politicians.
Lebanon does not need a party system to make these realities apparent and a
determinant in policymaking. Indeed, the very vulnerability of the system
increases its responsiveness.
This is not to say, however, that Lebanon will overcome the persistent
challenges of modernization in the long run. Institutional growth and
democratic procedures are palliatives, not cures. In order to explain
Lebanon’s surprising durability it is necessary to point out these little-
noticed capacities for adaptation in a dynamic situation. But a question
remains: Are these capacities likely to be sufficient to respond to the
persistent and increasing challenges posed by social mobilization in Lebanon
and the Arab world? The further relevant questions are now whether a
Lebanese government can actually implement its social security scheme or
save the country from financial disasters; whether it can be sufficiently Arab
in its foreign policy without awakening sectarian fears; whether it can
integrate the modernizing radicals as it once integrated sectarian radicals;
whether, in short, it can innovate as well as adapt.
To those who value the present system the outlook cannot be encouraging.
Many Lebanese feel discontented, disenfranchised from active political
participation, perhaps by their sect or lack of money, with no means of
legitimate political expression. They see the country ruled by a clique of
wealthy landlords, bankers, and lawyers. In their view, parliament is an
obstacle on the road to effective power, and reform within the present system
is ephemeral. Here lies a dilemma. If democracy is Lebanon’s guarantor of
stability and if stability is threatened by these increasing pressures, then the
obvious need is more democracy. More democracy for Lebanon means a
government more successfully responsive to popular demands, a political
arena in which more people can be influential, and a system in which
popular majorities rule. If these were the only consequences then there
would be reason for optimism. But “more democracy” also will create new
power centers, which might displace, by one means or another, the
incumbent elite. If more democracy would assuage the discontent of those
whom the present system excludes and would produce effective
administration it would also challenge the traditional powers. The 1958
crisis amply demonstrated that such a confrontation can completely disrupt
civil order. On the other hand, if modernization were in fact eradicating the
old group structure as fast as it molded “participant personalities,” then the
players in the old balance-of-power game might just fade away peacefully, a
modern democratic state might appear, and Lebanese nationalism might
arise to invest the state with enough legitimacy to enable it to tackle the new
social problems effectively. But this idea, held by some critics of
confessionalism and the status quo, is unrealistic too: it is Utopian to
imagine that the traditional power-holders will disappear quickly or that
individuals in the “modernized” state will forsake their traditional ties for a
“rational” social contract that denies such ties. Hence the dilemma: too little
democracy may lead to social disorder and revolution; too much may bring
down upon the state the wrath of the traditional leaders.
This dilemma is reflected in discussions that took place in the summer of
1968 regarding the next presidential election, scheduled for 1970. The
political modernizers who are identified with former President Chehab, in a
candid assessment of developments during the Helou regime, arrived at
some pessimistic conclusions. Despite a decade of struggle since the 1958
crisis, they feel that “Chéhabisme” has not gained general acceptance;
indeed, the 1968 parliamentary elections appeared to repudiate the Cheha-
bists in favor of former President Chamoun and his supporters,
representatives of the traditional, conservative politics. Neither the 1958
events nor six years of Chehabist rule succeeded in supplanting the old
notables and cliques or reducing sectarian tension. The long-term effects of
the Intra Bank crash and the 1967 war, in their opinion, continue to
exacerbate the unresolved domestic tensions in the country and pose certain
dangers. Among these dangers are a possible duality of power between the
army and the regime, a paralysis of democratic procedures owing to a
resurgence of uncontrolled clique politics, and a further diminution of
political and social reform which could crystallize tendencies toward
revolution, civil disorder, or military dictatorship. Some seasoned observers
of the Lebanese scene see the present situation as potentially more
dangerous than that of 1956-1957 because the army is now regarded as
partisan (pro-Chehab) rather than impartial; because the threat of new Israeli
aggressive in the area (Lebanon included) is real, greatly complicating the
delicate foreign-policy compromise embodied in the National Pact; and
because the younger generation is dissatisfied with the traditional politics of
compromise.
“Like the states of seventeenth-century Europe,” Samuel P. Huntington
has written, “the non-Western countries of today can have political
modernization or they can have democratic pluralism, but they cannot
normally have both.”38 Lebanon, however, has been showing signs of
abnormality for over two decades; the balance-of-power situation has forced
it to enjoy a little of each. Lebanon’s skillful politicians, undaunted by grim
logic, have thus far avoided the dilemma by maintaining “just enough”
democracy. In the absence of positively legitimized central control, a kind of
democratic pluralism has been a necessity rather than a luxury. Similarly,
under the increasing pressures of modernization, the balance has been
altered to accommodate the moderate development of rationalized authority,
differentiated political structures, and mass participation. How much future
alteration will such a delicately balanced system sustain without collapsing?
How strong are the disintegrative tensions inherent in such a fragmented
political culture, shaken as it is by conflicting external and domestic
demands? These questions must be analyzed, as I have suggested, in a
setting of interrelated demands. Failure to handle the higher-order problems
of national development and social justice will have adverse repercussions
for the basic problem of stability, and by the same token, too much success
in those areas may be equally disastrous.
Postscript
Since the original publication of this paper, the worst forebodings expressed
in it unfortunately have come to pass, and with a degree of violence that has
shocked even the most pessimistic analysts of Lebanon’s future. An
unbearable convergence of domestic and external loads, coupled with a
relapse of the political system itself into a particularly atavistic
traditionalism, set the stage for one of the bloodiest civil wars the world has
known in the twentieth cen-ury.
Lebanon’s political deterioration during these years began with the Arab
defeat by Israel in 1967 and the subsequent meteoric rise of the Palestinian
resistance movement. This movement took root wherever there were
Palestinians, but especially in Lebanon where the concentration of
Palestinians—Lebanon had the biggest refugee camps—and the relative
freedom of political action led to active Palestinian forays into Israel from
bases in south Lebanon. Israel retaliated massively with air, artillery and
ground attacks which destroyed villages and crops. The Israeli actions
caused a substantial depopulation of south Lebanon and created enormous
political problems for the Lebanese state. Inasmuch as most of the Muslim,
Arab-oriented Lebanese strongly supported the Palestinians while the
conservative Maronite Christian elite opposed them, government fell into a
condition of quasi-paralysis.
In 1970 Suleiman Frangieh, a parochial mountain zcTim from north
Lebanon, was elected president by a one vote margin, thus bringing the
modernizing reformist impulse of Chehabism (flawed as it was) to a
definitive end. What followed was a return to traditional neo-feudal politics:
virtually no new social welfare programs were undertaken, and such reform
efforts as were tried (for example, in the education ministry), soon collapsed
in failure. The army (as forecast in the paper) did in fact become polarized—
not in a Chehafrst direction, however, but rather in the direction of the most
conservative Maronite tendencies. Against a background of increasingly
serious conflicts between the Palestinian organizations and the army in
collusion with the Maronite militias, the Lebanese Muslims began to
demand once again structural reforms giving them greater influence in
political life. They demanded in particular a greater representation in the key
commands of the army. But these demands went unheeded, and Lebanon
continued with its unchecked social mobilization and corruption.
Many Lebanese thought that the 1973 Arab-Israeli war, which Israel failed
to win, would set in motion a general settlement process for the entire
Middle East conflict, one in which the Palestinians would find a degree of
redress and Lebanon a degree of respite. But this did not happen, and
Secretary of State Henry Kissinger’s “step-by-step” diplomacy pointedly
sidestepped the Palestinian question. Thus Lebanon was left in turmoil as
conservative and militant forces among the Lebanese, each aided by outside
Arab and non-Arab regimes, continued to vie for position. The point of no
return may have been reached in April 1973 after an Israeli raid on Beirut in
which three Palestinian leaders were murdered and a violent but
inconclusive clash between Palestinian guerrillas and Lebanese army units,
in which the army failed to dislodge the Palestinians from certain refugee
camps near the Beirut airport. From then on it appears that the conservative
Maronite leaders, led by former president Camille Chamoun, began
preparing for all-out war with the Palestinians and their many Lebanese
leftist and Arab nationalist sympathizers. The two precipitating events of the
civil war took place in February and March of 1975. One was the ambushing
of a bus carrying Palestinians by Lebanese Phalangists; the other was a
violent clash at Saida between Lebanese and Palestinian fishermen, on the
one hand, and Lebanese army units, on the other. Nineteen months later, in
November 1976, the savage fighting and enormous destruction appeared to
have ended when a largely Syrian-manned Arab peacekeeping force moved
into Lebanon to restore order. Between 40 and 60 thousand Lebanese and
Palestinians, nearly all of them innocent civilians, were dead; Beirut’s port,
commercial center, and several suburbs were in ruins; and the economic and
political future of the country was at best problematical. While nearly all
prominent leaders on both sides publicly proclaimed their insistence on a
reunified Lebanon, some (especially conservative Maronites) spoke privately
of the desirability of partitioning the country on grounds that the hostility
between the two sides was now too deep for reconciliation.
The dilemma of the post civil war period lay in the fact that while the old
system of confessional democracy had proven itself to be totally and
disastrously inadequate, the only viable long-term alternative for Lebanon’s
political development—gradual secularization—seemed virtually impossible
to achieve in light of the effects of the civil war itself.
1. Samuel P. Huntington, “Political Development and Political Decay,” World Politics, XVII (April
1965), 386-430.
2. al-Anwar (Beirut), September 24, 1964, p. 4.
3. Tocqueville, Democracy in America, 2 vols. (New York, 1958), I, 433-436, and Chs. 16 and 17;
David Truman, The Governmental Process (New York, 1951), Ch. 3 and pp. 519-524.
4. The question arises of whether or not it is Lebanon’s high living standard that supports its
democracy. There is little doubt that there is a rough association between wealth and stable democracy
in the world. See, e.g., S. M. Lipsets correlations of wealth and stable democracy for fifty countries in
Political Man: The Social Bases of Politics (New York, 1960), pp. 51-54. I would argue that although
the correlation holds for Lebanon-Lebanon scores relatively high on the economic indices and can also
be classified as democratic—the correlation is not a causal relation. If there is a causal relation in
Lebanon it would—once again—seem to be the other way around. Wealth has not made Lebanon any
more pluralistic than it would otherwise be, nor has it invested the organs of the state with legitimacy.
Neither has it affected the balance of power, which in Lebanon would not become much more or much
less democratic than it now is if the per capita income were either half or twice its present level of
around $400 annually. For a contrary argument, see Charles Issawi, “The Economic and Social
Foundations of Democracy in the Middle East,” in Walter Z. Laqueur, ed. The Middle East in
Transition (New York, 1958), esp. pp. 48-49.
5. Confining himself to traditional group orientations, John Gulick, in a study of al-Munsif, a Greek
Orthodox village north of Beirut, suggests that the individual is involved in six concentric spheres of
association: the family, the lineage, the village, the nation, the sect, and the linguistic group. Social
Structure and Cultural Change in a Lebanese Village (New York, 1955), p. 163. Note that the nation
ranks only fourth. For similar findings, based upon a sample of undergraduates at the American
University of Beirut (not all of whom were Lebanese), see Levon Melikian and Lutfy N. Diab, “Group
Affiliations of University Students in the Arab Middle East,” Journal of Social Psychology, XLIX
(1959), 145-159. In general, the students ranked the family first, the ethnic group second, the religion
third, citizenship fourth, and the political party fifth.
6. Carleton Coon, Care van (New York, 1951), pp. 1-9 et passim; Albert H. Hourani, Syria and
Lebanon (London, 1946), p. 127.
7. See, e.g., Afif I. Tannous, “The Village in the National Life of Lebanon,” Middle East Journal, III
(April 1949), 151-164; Raphael Patai, “The Middle East as a Cultural Area,” Middle East Journal, VI,
No. 1 (1952), 1-21.
8. Gabriel Puaux, Deux Années au Levant: 1939-1940 (Paris, 1952), Ch. 7, “Le panthéon syrien.”
Puaux was High Commissioner of Lebanon and Syria at the time.
9. A thorough description of the power configuration before and during the 1958 crisis will be found
in Fahim I. Qubain, Crisis in Lebanon (Washington, 1961), esp. pp. 28-69. In the summer of 1958,
widespread disorders broke out, stemming (1) from President Chamoun’s attempt to wrest powers
from certain traditional notables and to renew his term of office and (2) from his relatively pro-
Western and anti-Nasser position.
10. For historical examples, see Kamal Salibi, The Modern History of Lebanon (London, 1965), Chs.
5, 8, and Part II. Also Puaux, Ch. 10, and Charles W. Thayer, Diplomat (New York, 1959), pp. 1-37.
11. Doxiadis Associates of Athens, Economic Data for the Ekistic Programme of Lebanon (LGHP-1,
R-LA-42/19-10-57), prepared for the US Operations Mission in Lebanon (Beirut, 1957), p. 32. The
Doxiadis investigators estimated that about 27,500 (or two-thirds) of the 41,570 “Businessmen and
Self-employed” were in the middle-income groups (i.e., with annual incomes from $500 to $3,000).
Of the total work force, 46 percent was in the middle-income group and 3 percent was in the $3,000
and over category.
12. Arnold Hottinger has described this phenomenon in two papers, “Zu’ama’ and Parties in the
Lebanese Crisis of 1958,” Middle East Journal, XV (Spring 1961), 127-140; and “Zu’ama’ in
Historical Perspective,” a paper delivered at the University of Chicago Conference on Democracy in
Lebanon, May 28-31, 1963, published in Leonard Binder, ed. Politics in Lebanon (New York, 1966),
pp. 85-105.
13. These sentiments were expressed by representatives of the various political tendencies in
interviews I conducted in Lebanon in the period 1962-1965.
14. Bechara al-Khoury, Haga’iq lubnãniyyah [Lebanese Truths], the memoirs of the former President,
3 vols. (Beirut, 1961), II, 289-300. See also the memoirs of President Camille Chamoun, Crise au
Moyen-Orient (Paris, 1963), pp. 118-119.
15. The Lebanese Constitution: A Reference Edition in English Translation, prepared by the
Department of Political Studies and Public Administration, American University of Beirut (Beirut,
1960), Article 58.
16. Ibid., Article 57.
17. Ibid., Article 55.
18. Ibid., Article 53.
19. For two descriptions see Salibi, Modern History, Part II, and Pierre Rondot, Les Institutions
politiques du Liban (Paris, 1947).
20. Article 95. The Constitution stipulates that this is only a provisional measure.
21. Clyde G. Hess, Jr., and H. L. Bodman, Jr. “Confessionalism and Feudality in Lebanese Politics,”
Middle East Journal, VIII,, No. 1 (1954), 10-26.
22. See also Malcolm H. Kerr, “Political Decision-making in a Confessional Democracy,” a paper
delivered at the University of Chicago Conference on Lebanese Democracy, May 28-31, 1963, and
published in Binder, Politics in Lebanon.
23. Hess and Bodman, 23.
24. Karl W. Deutsch, “Social Mobilization and Political Development,” American Political Science
Review, LV (September 1961), 493-514.
25. Ibid., 503, Table 1.
26. For a complete discussion of Lebanese modernization, see my study The Precarious Republic:
Political Modernization in Lebanon (New York, 1968), Chs. 2 and 3.
27. L’Orient (Beirut), October 22, 1958.
28. République Libanaise, Ministère du Plan, Besoins et possibilités de développement du Liban, a
report made by the Institut pour Recherche et de Formation en Vue de Développement (IRFED), 2
vols. (Beirut, 1960-1961), I, 93.
29. I, 81, Table 26.
30. “Social Mobilization,” 501.
31. Michael C. Hudson, “A Case of Political Underdevelopment,” Journal of Politics, XXIX
(November 1967), 821-837.
32. Kataeb Party, Declaration (Beirut, n.d), issued (in 1956?) in English, French, and Arabic,
containing inter alia a speech by Pierre Gemayel setting forth the party’s beliefs.
33. See Mithãq al-hizb al-taqaddumi al-ishtirãki [Covenant of the Progressive Socialist Party]
(Beirut, 1961).
34. See Sleiman M. Gemayel, Evolution de budget libanais (Paris, 1962), p. 70; and Le Commerce du
Levant (Beirut), March 3, 1965.
35. Voting statistics published in L’Orient (Beirut), April 7, 15, and 28 and May 3, 1964. See also
Michael C. Hudson, “The Electoral Process and Political Development in Lebanon,” Middle East
Journal, XX (Spring 1966), 173-186.
36. These data are from the Library of Parliament and the newspaper L’Orient. Cf. the generally
similar findings, but different conclusions, reported by Jacob Landau, “Elections in Lebanon,” Western
Political Quarterly, XIV (March 1961), 130. Classification is difficult since many deputies can
reasonably fit in more than one category.
37. See, e.g., Kerr, “Political Decision-making.”
38. “Political Modernization: America vs. Europe,” World Politics, XVIII (April 1966), 412.
* I gratefully acknowledge the support of the Harvard Center for International Affairs, the Brooklyn
College Research Center in Comparative Politics and Administration, the Yale World Data Analysis
Program, the American Philosophical Society, and the comments of Ahmad Haffar, Joseph S. Nye,
Leonard Binder, Manfred Halpern, and Samuel P. Huntington, none of whom necessarily agrees with
the views expressed.
PART IV
Introduction
But the unexpected has become the rule, and throughout the Third World
military involvement in politics has emerged as standard operating
procedure. Now, in few states of le Tiers Monde do the men on horseback2
sit as silent observers of the process of governance; and increasingly they
have ridden at full stride into the political arena to capture unmediated
control of the reins of state.
To students of Latin America, military involvement in politics is neither a
unique nor a new experience, yet even in these Western Hemisphere states
of the Third World “[t]he recent decades of rapid change and social crisis . .
. [have] brought the armed forces back into a position of political
prominence they had not held since the nineteenth century.” 3
Although the number of military coups in the world of the South has
been enormous, the forms and functions of military interventions remain
varied and the theoretical explanation for their occurrence stands subject to
debate. While it can be argued that “[n]o generalizations are likely to reach
a useful level of specificity and at the same time embrace all relevant
material,”4 there nonetheless are several general lines of inquiry into the
nature of military coups and regimes within the context of development in
the Third World which should be understood. Such inquiries focus on the
reason, the value, and the results of military intervention in politics.
Political activity by the men on horseback (or, to borrow Samuel
Decalo’s phrase, the colonels in the command car) within the Third World
is often explained in terms of the societal crises of legitimacy and stability.
From this broad perspective, it is both the fundamental weakness of the
civilian institutions of state and their inability to legitimize their rule which
provoke political action on the part of the military. Given the failure or, if
one prefers, the instability of civilian regimes, the men on horseback act to
fill the void.5 Underlying this position is the view that “... the most
important causes of military intervention in politics are not military but
political, and reflect not the social and organizational characteristics of the
military establishment but the political and institutional structure of the
society.”6 In effect, Third World countries are often seen as praetorian states
(an idea rooted in the historical role of the Roman Praetorian Guard)
wherein “private ambitions are rarely restrained by a sense of public
authority; [and] the role of power (i.e., wealth and force) is maximized.” 7
The first essay in this section, Perlmutter’s “The Praetorian State and the
Praetorian Army,” is a basic statement of this perspective. To Perlmutter, “a
modern praetorian government may develop when civilian institutions lack
legitimacy or are in a position to be dominated by the military.” Thus, a
modern praetorian state is a state in which the political processes at work
(including the civilian government’s inability to successfully pursue
“nationalist-modernist goals”) favor the emergence of the military as the
“core group” in the political arena.
While Perlmutter concentrates on the Middle East and North Africa, his
discussion of praetorianism as representative of “certain stages of
development” and his analysis of the social and political conditions
contributing to a praetorian state clearly have wide applicability throughout
the Third World.
However, despite both the utility and seductiveness of the position which
finds the reason for military intervention in the pull effect of systemic
weakness and the interrelated push effect of military organization (i.e., the
“characteristics of professionalism, nationalism, cohesion, and austerity”
that impel the armed forces “to move into the political arena and to rescue
the state . . .”8), the logic of that position is not without its difficulties and
its critics. Samuel Decalo, for example, writing on the African experience
has argued that: “It is both simplistic and empirically erroneous to relegate
coups in Africa to the status of a dependent variable, a function of the
political weakness and structural fragility of African states and the failings
of African civilian elites.”9
Without overemphasizing the “personal” element as the source of
military intervention, as does Decalo, Bienen’s essay in this section,
“Military and Society in East Africa,” raises several questions about the
overall validity of the praetorian mode of analysis for our understanding of
the action of Third World armed forces in the arena of politics. Primarily,
Bienen expresses the view that we must be extremely careful not to
“overhomogenize” Third World states and neglect what can be termed
idiosyncratic factors.
In short, then, the praetorian and idiosyncratic perspectives on the
reasons for military intervention in politics need not be viewed
dichotomously. While the focal point of each perspective is different, the
two views can be combined to complement one another. For example, S. E.
Finer, in a volume from which part of our section title is borrowed, notes
that the reasons for military intervention are found in two sets of forces:
“The capacity and propensity [or disposition] of the military to intervene,
and the conditions in the society in which it [the military] operates.”10 In
line with the former set of forces, Finer further states that the military’s
disposition to intervene can be explored in terms of five, mutually non-
exclusive motivations which range from the defense of the national interest
and class interests, to the pursuit or maintenance of corporate/military self-
interests, particularistic jethno-regional interests, and
personal/individualistic interests.11
to intervene based on the military’s brokerage involvement in politics, for
our purposes two major propositions stand out. The first is that within the
Third World there exists a variety of reasons for military intervention in
politics, with certain reasons being paramount in specific cases. The second
is that in all cases it is of the greatest importance to examine the position
and motivation of the armed forces, the nature of the total societal context,
and the linkages that exist between society and the men on horseback.
No less complex a situation surrounds our second line of inquiry into the
value of military activity in politics. The analysis of the military as leaders
in the Third World has been influenced by two conflicting images.12 The
first image, largely derived from the historical experience of Latin America,
is a negative one in which the military is viewed as a retrogressive force in
society. The second image, rooted in the optimism of the early 1960s, has a
positive orientation wherein the men on horseback are seen as a progressive
force for modernization and development. This latter image looks to the
military as a modern organization (and its officers as modern men) to act as
an agency of progressive change in the Third World.
From such a viewpoint, the military is considered to be the last chance
for the continuance of the processes of state and nation building within a
context of instability and fragmentation. The men on horseback not only
bring order, but they bring an order which pays homage to the gods of
development. Strongly influenced by the role of Kemal Ataturk in post
World War I Turkey, this image holds up the men on horseback not as
intruders into politics, but rather as the saviors of politics in the Third
World.
But if the 1960s witnessed a shift from the negative to the positive image
of the value of military intervention, the 1970s have seen the pendulum
swing back. It is true, as one of the essays in this section points out, that a
military government “... may lead a successful modernization effort,” but it
is equally true to note that more often than not military regimes have failed
in this effort. The “Kemalist achievement” has not been approximated
elsewhere; promise and performance have remained on two separate planes.
Inquiries into the value of military intervention clearly link up with
inquiries into the results of military rule; for it is the empirical study of
results which reinforces, or erodes our assumptions of value. While all of
the essays in this section address themselves somewhat to the issue of the
results of the military’s role and rule in Third World societies, it is the
article by McKinlay and Cohan, “A Comparative Analysis of the Political
and Economic Performance of Military and Civilian Regimes,” which most
clearly focuses on this question. Their cross-national aggregate study, which
includes the more developed as well as the less developed countries,
indicates “that the nature of military regimes cannot be adequately
understood in terms of assisting or inhibiting modernization or in terms of a
typology based on a simple distinction between military and civilian
regimes.”
Just as we have come to realize that despite their surface hierarchies,
individual military organizations are not monoliths, so too have we
increasingly recognized that as an aggregate, military regimes are marked
more by diversity than homogeneity. The military, acting either as an arbiter
or ruler, has no greater immunity to societal conditions and cleavages than
does a civilian regime. And often military intervention, rather than
establishing a firm foothold on the difficult path of socio-economic and
political transformation, leads to the further corrosion of already fragile
institutions and ideas.13
Yet whether we approach the political activity of the military from a
position of admonishment or approbation, such activity will continue to be
of great importance. Whatever its ideological cloak, military involvement in
politics will remain a critical factor in the politics of the Third World for
some time to come. For even if the military “shares” the reins of power with
civilians, the men on horseback will not soon retreat to the barracks. The
military coup may be “a method of change that changes little,”14 but it is
not a method of change whose end is in sight.
NOTES
1. Lucian Pye, “Armies in the Process of Political Modernization,” in John Johnson (ed.), The Role
of the Military in Underdeveloped Countries (Princeton: Princeton University Press, 1962), p. 69.
2. This phrase is borrowed from S. E. Finer, The Man on Horseback, Second Edition
(Harmondsworth, England: Penguin Books, 1976).
3. Edwin Lieuwen, “Militarism and Politics in Latin America,” in Johnson (ed.), The Role of the
Military, p. 131.
4. Robert Dowse, “The Military and Political Development,” in Colin Leys (ed.), Politics and
Change in Developing Countries (Cambridge: Cambridge University Press, 1969), p. 217.
5. See Claude Welch, Jr., “The Roots and Implications of Military Intervention,” in Welch (ed.),
Soldier and State in Africa (Evanston: Northwestern University Press, 1970); Claude Welch, Jr. and
Arthur Smith (eds.), Military Role and Rule (North Scituate, Massachusetts: Duxbury Press, 1974);
Morris Janowitz, The Military in the Political Development of New Nations (Chicago: The University
of Chicago Press, 1964), pp. 75-106; and Gavin Kennedy, The Military in the Third World (New
York: Charles Scribner’s Sons, 1974), pp. 1-154.
6. Samuel Huntington, Political Order in Changing Societies (New Haven: Yale University Press,
1968), p. 194.
7. David Rapoport, “A Comparative Theory of Military and Political Types,” in Samuel Huntington
(ed.) Changing Patterns of Military Politics (New York: The Free Press, 1962), p. 72.
8. Samuel Decalo, Coups and Army Rule in Africa (New Haven: Yale University Press, 1976), p. 12.
9. Ibid, p. 13.
10. Finer, The Man on Horseback, p. 224.
11. Ibid, pp. 229-231.
12. This analysis draws on the work of: Kenneth Grundy, Conflicting Images of the Military in Africa
(Nairobi: East Africa Publishing House, 1968); Finer, The Man on Horseback-, Pye “Armies and the
Process of Political Modernization;” Henry Bienen, “The Background to Contemporary Study of
Militaries and Modernization,” in Bienen (ed.), The Military and Modernization (Chicago: Aldine-
Atherton, 1971 ) ; and Decalo, Coups and Army Rule.
13. See Eric Nordlinger, “Soldiers in Mufti,” The American Political Science Review, Vol. LXIV, No.
4 (December 1970); Nordlinger, Soldiers and Politics (Englewood Cliffs: Prentice-Hall, 1977); and
Henry Bienen (ed.), The Military Intervenes (New York: Russell Sage, 1968).
14. Ruth First, Power in Africa (Harmondsworth, England: Penguin Books, 1972), p. 22.
The Praetorian State and the
Praetorian Army
Toward a Taxonomy of Civil-Military Relations in Developing
Polities
Amos Perlmutter*
1. Harold D. Lasswell, “The Garrison State,” American Journal of Sociology, XLV1 (January 1941),
455-468.
2. I am now engaged in relating the propositions to the cases of twelve states in depth (in four areas),
comparing the states along the continuum of praetorianism. (The countries are Mexico, Argentina,
Brazil, Turkey, Egypt, Syria, India, Burma, Indonesia, Tunisia, Ghana, and Nigeria.) The
propositions are divided into five clusters: (1) the military and society—particularly the
transformation from oligarchical to radical praetorianism; (2) military intervention—coups and
patterns of military intervention; (3) types of military roles, particularly along the continuum of
arbitrator-ruler; (4) the military and modernization-development—whether military rule has brought
about more or less stability than nonmilitary rule; (5) the relationship between military organization
and political skill. I have also begun quantitative analyses of qualitative data on military and
modernization in five Middle Eastern countries. These analyses will be based on cross-national data
(1950-1962) and on quantitative correlations between social, political, and economic indicators and
the level and pace of military, economic, and political performances.
3. I am grateful to my friend M. D. Feld of Harvard University for this analysis, which he has written
especially for the benefit of this article.
4. David C. Rapoport, “Praetorianism: Government Without Consensus” (unpub. Ph.D. diss.,
University of California, Berkeley, 1960), pp. 14-15, defines praetorianism as a constitutional form of
“government without consent.” Rapoport’s thesis provides an outstanding theoretical discussion of
praetorianism. Although the present essay closely follows Rapoport’s definition of praetorianism, it
emphasizes the descriptive aspects of the subject and foregoes discussions of constitutionalism,
consensus, and authority, which are discussed at length in Rapoport’s work.
5. This essay has been directly influenced by the works of Professor Samuel P. Huntington,
especially his seminal essay, “Political Development and Political Decay,” World Politics, XVIII
(April 1965), 386-430. For some time, we have carried on an intellectual dialogue which, I hope, has
resulted in a more positive approach to a theory of civil-military relations in developing polities.
Especially excellent is Huntington’s most persuasive chapter, “Praetorianism and Political Decay,” in
Political Order in Changing Societies (New Haven, 1968), pp. 192-263. He argues there that the
concept of praetorianism becomes a useful operational tool to explain the relationship between
political development and modernization. In our view Huntington’s analysis of the role of political
decay in modernizing polities becomes most crucial in the case of praetorianism.
6. Huntington’s central thesis (the Gap hypothesis) is that political modernization breeds both
political instability and praetorian political order. Political Order, pp. 32-39, 53-56. See Gino
Germani and Kalman H. Silvert, “Politics, Social Structure and Military Intervention in Latin
America,” Archives européennes de sociologie, II, No. 1 (1961), 62-81.
7. The most impressive analysis of breakdown in modernization is found in S. N. Eisenstadt’s
extensive studies (by the author’s admission, mostly of a preliminary nature) of the relationship
between traditionalism, modernization, and change. See especially his “Breakdowns of
Modernization,” Economic Development and Cultural Change, XII (July 1964), 345-367; “Political
Modernization: Some Comparative Notes,” International Journal of Comparative Sociology, V
(March 1964), 3-24; Modernization and Change (Englewood Cliffs, 1966); and “Some Observations
on the Dynamics of Tradition,” Working Paper, Ballagio, July 1968.
8. It is most interesting to observe that some of the historical types of praetorianism were also
associated with the disintegration of legitimate authority. After patriarchal power had declined,
patriarchal praetorianism no longer exercised authority without restraint. The military took over, but
only for a short period. In patrimonial (prebureau-cratic) political systems, the military became a
permanent establishment. Since the military was the most powerful and rational structure after the
patrimonial system, it took over when patrimonial legitimacy disintegrated. In the caudillismo case,
the union of personalism and violence seized the disintegrating patrimonial state apparatus. Under the
guise of Republicanism (in Latin America) and Liberalism (in nineteenth-century Spain), the
caudillismo became the driving force of the new nation. On personal patriarchal and patrimonial
types of domination, see Max Weber, Economy and Society, III, ed. G. Roth and K. C. Wittich (New
York, 1968).
9. See Nadav Safran, Egypt in Search of Political Community (Cambridge, Mass., 1961), p. 2, for an
analysis of the consequences of this lack of parallelism in the development of Egypt.
10. See Jean and Simmone Lacouture, Egypt in Transition (New York, 1958), pp. 367-387, for a
discussion of divisions of this type in the Egyptian working class.
11. Wyatt MacGaffey and Clifford R. Barnett, Cuba (New Haven, 1962), p. 144.
12. Morroe Berger, “The Middle Class in the Arab World/’ in Walter Lacqueur, ed. The Middle East
in Transition (New York, 1958), pp. 63-65.
13. For a discussion of the importance of resource accumulation and use by the political institutions,
see S. N. Eisenstadt, The Political Systems of Empires (London, 1963), esp. Chs. 6, 7. For Eisenstadt,
the promotion of “. . . free resources and . . . [the] freeing [of] resources from commitments to
particularistic-ascriptive groups . . .” is one of the conditions for the creation and maintenance of
autonomous political institutions. P. 119.
14. These ideas are borrowed from Edward Shils’ seminal studies on ideology and civility. See
“Primordial, Personal, Sacred and Civil Ties,” British Journal of Sociology, VIII (June 1957) 130-
146; “Ideology, Center and Periphery,” in The Logic of Personal Knowledge (Glencoe, 1961), pp.
117-130. A parallel interpretation of the role of primordial forces in the process of political
integration is given by Clifford Geertz, “The Integrative Revolution,” in Clifford Geertz, ed. Old
Societies and New States (Glencoe, 1963).
15. Shils, “Ideology, Center and Periphery,” p. 117.
16. See Aristide R. Zolberg, Creating Political Order (Chicago, 1966).
17. Huntington, “Political Development,” 394.
18. Ibid.
19. On Muslim Brotherhood-army relationships, see Richard P. Mitchell, ‘The Society of the Muslim
Brothers” (unpub. Ph.D. diss., Princeton University, 1960), pp. 61-250; Ishak Musa Husaini, The
Moslem Brethren (Beirut, 1956), pp. 125-130; Eleizer Beeri, “On the History of the Free Officers,”
The New East (Hamizrah Hehadash), XIII, No. 51 (1963), 247-268; Kamil Isma’il al-Sharif, al-
Ikhwan al-Muslimin Fi Harb-Filastin (The Muslim Brotherhood in the Palestine War) (Cairo, 1951);
J. and S. Lacouture, Egypt, pp. 131 ff. On relationships between Free Officers, Egypt, and the Axis,
see Lukasz Hirszowicz, The Third Reich and the Arab East (London, 1966), pp. 229-249.
20. José Nun, “A Latin American Phenomenon: The Middle Class Military Coup,” in Trends in
Social Science Research in Latin American Studies: A Conference Report (Berkeley, 1965); and Liisa
North, Civil-Military Relations in Argentina, Chile, and Peru (Berkeley, 1966). For a related
argument concerning the military and the “new” middle class in the Middle East, see Manfred
Halpern, “Middle Eastern Armies and the New Middle Class,” in John J. Johnson, ed. The Role of the
Military in Underdeveloped Countries (Princeton, 1962), pp. 277-315.
21. See Edward A. Shils and Morris Janowitz, “Cohesion and Disintegration in the Wehrmacht in
World War II,” Public Opinion Quarterly, XII (Summer 1948), 288-292. Shils and Janowitz
demonstrated that the defeat and disintegration of the Wehrmacht were due to the collapse of
primary-group cohesion.
22. North, p. 17.
23. R. Kitchen, The German Officer Corps, 1890-1914 (Oxford, 1968), pp. 28-32.
24. Majid Khadduri, independent Iraq: A Study in Iraqi Politics from 1932-1958, 2d ed. (London,
1960), pp. 124-125.
25. Uriel Dann, Iraq Under Qassem (Jerusalem, 1969).
26. See Irving Louis Horowitz, “The Military in Latin America,” in Seymour Martin Lipset and Aldo
Solari, eds. Elites in Latin America (London, 1966).
27. The transformation of army rebels of 1960-1961 into permanent senators only reiterates the
persistence of the Kemalist legacy in Turkey, at least as of 1967.
28. Charles Simmons, “The Rise of the Brazilian Military Class, 1840-1890,” Mid-America, XXXIX
(October 1957), 227-238.
29. Samuel P. Huntington, The Soldier and the State: The Theory and Politics of Civil-Military
Relations (New York, 1964), pp. 80-81, 93-94.
30. North, pp. 52-57.
31. Khadduri, pp. 78-80.
32. Kalman H. Silvert, “The Costs of Anti-Nationalism: Argentina,” in Kalman H. Silvert, ed.
Expectant Peoples (New York, 1963), pp. 366-369.
33. North, pp. 1-10.
34. Huntington, Political Order, pp. 219-237.
35. North, pp. 34-37.
36. Lacouture, pp. 144-145.
37. Eleizer Beeri, The Officer Class in Politics and Society of the Arab East (Ha-Ktzuna ve-hashilton
Ba-Olam ha-Aravi) (Israel, 1966), pp. 78-80.
38. Shimon Shamir, “Five Years of the Liberation Rally,” The New East (Hamizrah Hedadash), VIII,
No. 4 (1957), 274.
39. Khadduri, pp. 76-80,126 ff.
40. North, pp. 34-37.
41. Huntington, The Soldier, pp. 10-11.
42. Ibid., Ch. 3, pp. 59-79.
43. Dankwart A. Rustow, “The Military in Middle Eastern Society and Politics,” in Sydney Nettleton
Fisher, ed. The Military in the Middle East (Columbus, 1963), pp. 34-37.
44. See George Blanksten, Ecuador: Constitutions and Caudillos (Berkeley and Los Angeles, 1951),
pp. 51-54, for a discussion of this phenomenon. For Latin American cases of this type, see L. N.
McAlister, “Civil-Military Relations in Latin America,” Journal of Inter-American Studies (July 3,
1961), pp. 342-343.
45. Amos Perlmutter, “From Opposition to Rule: The Syrian Army and the Ba’th Party,” Western
Political Quarterly, forthcoming.
46. See Walter F. Weiker, The Turkish Revolution 1960-1961 (Washington, 1963), for a complete
discussion of this event.
47. Huntington, The Soldier, pp. 93-94, discusses briefly the conservatism of the professional officer,
albeit in nonpraetorian states.
48. See Richard Patch, “The Peruvian Elections of 1962 and Their Annulment,” American
Universities Field Staff Reports (West Coast South America Series), IX (September 1962), 6.
49. Huntington, The Soldier, pp. 83-85.
50. For an analysis on parallel lines, emphasizing the role of the army as the bearer of explicit
political norms and images, see Moshe Lissak, “Modernization and Role-Expansion of the Military
in Developing Countries: A Comparative Analysis,” Comparative Studies in Society and History, IX
(April 1967), 249-255.
51. Daniel S. Lev, “The Political Role of the Army in Indonesia,” Pacific Affairs, XXVI (Winter
1963-1964), 349-364.
52. For the most detailed description of Egypt’s officer-politicians, see Lacouture, Egypt, pp. 125-
129, and Beeri, The Officer Class, pp. 67-73; also, Anwar al-Sadat, Revolt on the Nile (London,
1957), p. 74. See also fn. 19.
53. Shamir, “Five Years,” 261-278.
54. Patrick Seale, The Struggle for Syria (London, 1965), pp. 124-131.
55. Khadduri, “The Role of the Military in Iraqi Society,” in Fisher, ed. The Military, p. 47.
56. I am indebted to Daniel S. Lev for his help in my understanding of the Indonesian army. For most
of the points made here on the Indonesian army, see Lev, “The Political Role,” 360-364. For the
interpretation of Lev’s analysis, only I am responsible.
57. Gamal Abdel Nasser, Egypt’s Liberation: The Philosophy of the Revolution (Washington, 1955),
pp. 32-33, 42-45.
58. Khadduri, Independent Iraq, pp. 200-206. See Col. Salah al-Din al-Sabbagh, Fursan al Uruba fi
al Iraq [The Knights of Arabhood in Iraq) (Damascus, 1956), pp. 29-30.
59. Huntington, Political Order, pp. 208-219.
60. David A. Wilson, “The Military in Thai Politics,” in Johnson, ed. The Role, p. 253.
61. Fred W. Riggs, Thailand: The Modernization of a Bureaucratic Polity (Honolulu, 1966).
62. Wilson, pp. 266-268.
63. Lucian W. Pye, “Armies in the Process of Political Modernization,” in Johnson, ed. The Role, p.
75.
64. ibid.
65. No penetrating and objective study yet exists concerning the army’s role in the economic
modernization of Egypt, where the army has demonstrated limitations in the management of large-
scale economic and industrial enterprises. One article on-that subject is James Heaphey’s
“Organization of Egypt: Inadequacies of a Non-political Model for Nation-building,” World Politics,
XVIII (January 1966), 177-193.
66. Ibid.; see also Patch, “Peruvian Elections,” 6.
67. On Ataturk’s political program and accomplishments, see (in addition to Rustow, fn. 43, and
Weiker, fn. 46) George Antonius, The Arab Awakening (New York, 1965); Rustow, “The Turkish
Army and the Founding of the Republic,” World Politics, XI (October 1958), 513-552; Kemal
Karpat, Turkey’s Politics (Princeton, 1959); and Daniel Lerner and Richard D. Robinson, “Swords
and Ploughshares*. The Turkish Army as a Modernizing Force,” WorJd Politics, XIII (October
1960), 19-44.
* The list of colleagues who have read various versions of this article is too long to include here. I
am especially grateful to Professors S. M. Lipset, S. P. Huntington, J. C. Harsanyi, E. Nordlinger, and
E. Rosenstein. and to M. D. Feld, L. North, and G. Orren for their great help in clarifying the
concepts of this article.
Military and Society In East Africa
Thinking Again about Praetorianism
Henry Bienen*
I The Problem
The army seized power in Uganda in January 1971.1 It had revolted in 1964,
along with the armies of Kenya and Tanzania, but civilian regimes were
restored and maintained in all three East African countries until the Uganda
coup.2 In the latter country, however, there was considerably more friction
between civilian leaders and armed forces and much more instability within
the army from 1964-71 than in Kenya and Tanzania. This article will
consider whether the concept of praetorianism, as developed in the works of
Samuel Huntington, helps explain the differences in the evolution of
relationships between armed forces and society in these three East African
countries.
Huntington’s fundamental proposition about military intervention states
that “... the most important causes of military intervention in politics are not
military but political, and reflect not the social and organizational
characteristics of the military establishment but the political and institutional
structure of the society.”3 The author will argue here that variations in the
social and organizational characteristics of armed forces do make a
difference with regard to the propensity of the military to intervene.
Moreover, while there is always some relationship between the
social/organizational characteristics of the military and the
political/institutional structure of the society, ways in which the evolution of
the social and organizational characteristics of the military can to some
extent be cut away from happenings in the wider society will be shown. In
other words, at a certain level of generality changes within the military do
not depend very closely upon changes within the larger society. Comparisons
will be made of both the political/institutional structures and the armed
forces of East Africa in order to defend this point of view.
It must be acknowledged in advance, however, that there is insufficient
information about the various East African military. Indeed, until we have
more detailed empirical studies of the African military, we cannot deal with
major questions concerning their modernity or how these military regimes
make decisions, who makes them, and what political goals govern their
decision making. So far, the argument that the military are the relatively
modern groups in society rests solely on assertions. No one has shown how
the military become rational, cohesive, universalistic, disciplined, and
industry oriented.4
The point is not that all military regimes have failed to be mod-ernizers.
But there is a risk in inferring qualities about armed force organizations
because they have chains of command, tables of organization, uniforms, and
weapons. This warning should have particular force in respect to African
armed forces, for many of these are still extremely small-scale organizations
with only a few battalions and some thousands of men in large countries
with populations in the millions. They are usually predominantly infantry
battalions with little firepower or mechanization.
In addition, information is inadequate as to how officer corps are
recruited, what norms govern officer behavior, how officers,
noncommissioned officers, and enlisted men relate to one another, and how
the latter categories operate.5
The fact remains that Kenya’s army did not break down in the way that
Uganda’s did. Although Kenya had a mutiny, the army was stabilized once
civilian authority was reasserted. This may be in part because Kenyatta had
more personal authority than Obote. Furthermore, Kenya’s ethnic splits had
not reached the point that Uganda’s had in 1964. Kenya was not without
ethnic struggle, however, especially Luo-Kikuyu conflict, which became
more intense later on.
Kenya had an advantage in that its army recruitment worked to insulate
the army from the major ethnic conflict. The Kamba had provided a
disproportionate number of recruits to the colonial King’s African Rifles and
they maintained this position. They were still the largest single tribe
represented in the officer corps in 1966, with 28 percent of the total
compared to their population share of 11 percent. The Kikuyu had 22.7
percent of the officers and 19.2 percent of the population, and the Luo
officer component was 10.3 percent although Luos made up 13.9 percent of
the total population. The Luyha were the most underrepresented large tribe,
with only .4 percent of the officers whereas they comprised 12.9 percent of
the people.66 Thus the Kamba, who were outside the main ethnic conflict,
were the most important group in the army.67
Kenya was not without its own tribal competition for positions of
authority in military and police forces, however. Indeed, Kenya’s political
leadership tried early on to master its own armed forces in a self-conscious
way. But the importance of the Kamba in the Kenyan army may have
insulated that army from Kikuyu-Luo tensions at a critical period. There has,
however, been a progressive Kikuyuiza-tion of the officer corps in the last
few years as there has of the civil service. The Kikuyu are the best educated
Africans in Kenya, and political sponsorship of Kikuyus in the government
service is a clear feature of Kenyan political life.
Lee reports that the Kikuyu had as many officers as the Kamba by 1967.
He sees the growth of Kikuyu officers and the relative diminution of officers
from small tribes such as the Nandi and Kipsigis, who had played some role
in the colonial army, as an attempt to make the army more broadly
representative.68 Yet the fact that the general service unit has increasing
numbers of Kikuyu officers, and that the army proper has been
commissioning more Kikuyus than any other tribe69 has led many to feel
that the Kikuyus are becoming overrepre-sented.
The issue of Kikuyuization is a sensitive one in Kenya; the insulation of
the army from ethnic struggle cannot be taken for granted in the future. But
there are factors which may continue to work for armed forces in Kenya
more disciplined and freer of ethnic conflict than those in Uganda. Uganda
and Tanzania moved in 1964 to end a British presence in their armed forces.
But in Kenya, British training missions continued; British officers remained
seconded to the Kenya armed forces; British units trained in Kenya, and had
contact with the Kenya armed forces even when the British base system in
Kenya was ended and facilities were turned over.70 Kenya continues to send
most of her naval cadets to the Royal Naval College at Dartmouth; her
airmen and signalmen learn their duties in England, and army personnel train
at Sandhurst, Mons, and the Imperial Defense College. The British impact
remains great and works in the direction of making the Kenyan armed forces
relatively well trained and effective.
The very factor of past success is important in the future evolution of the
armed forces. It has been suggested earlier in this article that Uganda’s army
was not up to the military tasks it was given. Kenya’s army, on the other
hand, has acquitted itself well. This keeps rather narrowly defined
professional norms viable as action within the scope of the norms is
perceived to be feasible. If the Kenyan army were called on continuously to
put down overt ethnic strife, as appeared possible after Tom Mboya’s
assassination, it would be increasingly difficult to keep the army insulated
from tribal conflict. The Kenyan army, like the Ugandan, might well be
unsuccessful in such endeavors. Furthermore, it cannot be assumed that
Kenya will avoid indefinitely the factional alliances between civilian and
military groups that have occupied Uganda’s army. Up to now, however, a
balancing of battalion commands and even an increasing Kikuyuization of
the officers corps have not led Kenyan civilian politicians to use the army or
parts of it as a political base for their own personal power.
Kenya does have nonmilitary bases for political power. The present
Kikuyu ruling group can rely on its civil service, on a growing commercial
and small business group, and on a more prosperous farming class. It can
appeal at large to Kikuyu sentiment. Unlike President Obote in Uganda, the
Kenyan political leadership is not forced to politicize the army. In Kenya,
then, factors intrinsic to the armed forces have interacted with wider societal
factors to create a military less engaged in politics in a day-to-day way and
not the sole arbiter of the regime. A particular crisis around the succession to
Kenyatta could conceivably trigger military intervention; but it does not
appear that the basic problems in Kenya attendant on phenomena of social
change will in the short run lead to military intervention on the praetorian
model. In other words, Kenya’s class evolution, the conjunction of tribe and
class, regionalism, and Kikuyuization do not seem to be leading to military
intervention along the lines set forth in Huntington’s concept of praetorian
society.
The argument here has been that we can compare Uganda’s and Kenya’s
societies, their armies, and the interaction of army and society. But there are
fundamental differences between the Ugandan and Kenyan armies that are
not entirely dependent on societal factors but themselves operate as
independent factors. It may be useful to look briefly at the way Tanzania as
well went about reconstructing its army in order to see if any factors unique
to the Tanzanian military can be isolated.
It was stated earlier that the then Prime Minister Obote in effect denied
that a mutiny had taken place in Uganda in 1964. In Kenya, Kenyatta noted
the gravity of the army action that year, but he did not condemn the entire
armed forces as such. Kenya reconstructed its army by continuing with a
process of professional training which utilized British officers.71 In
Tanzania, on the other hand, Nyerere condemned the army as a whole,72
disbanded it, and set about creating new armed forces.73 The aim was to
construct an army that would be politically loyal. Of course, that was the aim
in Kenya, too, but in Tanzania the pool for recruitment of the army was to be
TANU and TANU-affiliated organizations and commitment to party norms
was to be the basis for the initial formation of the new armed forces.74
Immediately after the landing of British troops in Tanzania on January 25,
1964, Nyerere called for a new army to be built around the TANU Youth
League (TYL). By February 12, however, Second Vice-President Rashidi
Kawawa said that, although the Youth League was “another nation-building
group,” it lacked leadership. A new group—National Servicemen—would be
instituted to provide the necessary leadership, and TYL members would be
recruited into special village schemes, each headed by National
Servicemen.75 The result was that Youth League members flocked to be
recruited into the National Service. Local TANU secretaries used the
promise of an army job to get people to take out TANU cards or to pay back
dues, though some of the people promised army positions by local
secretaries were patently physically unfit. Recruitment was carried out by
traveling teams, which moved from place to place reviewing men assembled
by the regional police officers and the regional commissioners.
Parliamentary secretaries (junior ministers) along with loyal
noncommissioned officers also helped recruit.
Although many TYL members were eventually recruited into the army,
the idea of using them to form a new army was not sustained. It soon gave
way to the more encompassing concept of the National Servicemen, whose
recruits had to have exhibited political loyalty, to be citizens aged eighteen
to twenty-five, and to be able to read and write Swahili. Infantry officers
were required to have completed Standard XII.
In May 1964, defense became the responsibility of Second Vice-President
Kawawa; both the National Service and youth sections and the headquarters
of the Tanzanian People’s Defense Forces are now part of his portfolio. A
regular army composed of four infantry battalions, one artillery battalion,
and one tank company was formed by 1970. An air wing was also created.
Initially, it was stated that all young men would be liable for national
service duty, but limited funds have restricted the number of national
servicemen. Recruitment begins with three months of national service
training, which includes taking part in such nation-building activities as, for
example, bush-clearing or road construction. Following this, recruits may
join the army, the police, or a specialist unit for six months of paramilitary
training. Those who do not enter the armed forces are to establish new
village settlements where they may settle if they like; but they are recallable
for military duty.
It has been estimated that about 2,000 of the 5,000 who had passed
through national service by the end of 1967 were absorbed into the security
forces as regular soldiers or policemen.76 The national service provided a
reserve which could be called up as well as a sizable number of recruits to
the regular army. It is important to note that the army’s recruitment was
selective, and it became standarized. Not all army recruits were from
national service.77 Moreover, those who entered the army became full-time
soldiers.
The government adopted a number of measures aimed at ensuring
political loyalty and integration of the armed forces with TANU, among
which was to invite all members of the military and the police to join TANU.
On June 24, 1964, Second Vice-President Kawawa told recruits that they
were citizens of the country and could participate fully in the politics of the
United Republic. The former practice of refusing soldiers the right to
political participation had been instituted by the colonialists, according to
Mr. Kawawa.78 (The independent Tanganyika government took three years
to change this rule.)
Police and soldiers have responded en masse to TANU’s invitation. In
fact, they enroll as whole units, and company commanders become heads of
the TANU committee, elected by the company. Officers are expected to do
party liaison work and to explain to the troops their role in Tanzania’s
development. President Nyerere granted an honorary commission in the
People’s Defense Forces, with rank of colonel, to S. J. Kitundu, the Coast
Regional Commissioner and a resident of Dar es Salaam. The President also
appointed Mr. Kitundu Political Commissar of the Tanzania Defense Forces
effective November 6, 1964. This post is not listed with the Ministry of
Defense; it is designed to give a TANU official a high army position. Mr.
Kitundu has, however, had no open operational direction of the armed
forces.
Before the mutiny, TYL members were often assigned to police duties at
meetings or formal roadblocks; they were instructed to prevent smuggling
and sometimes even to collect taxes. In certain places, TANU members did
constabulary work. But these informal and irregular procedures have now
been formalized.79 To prepare them to defend Tanzania in the event of an
attack, members of the nation’s polrce force, prisons service, national
service, and TANU Youth League were supposed to undergo full military
training with modern weapons. Yet Nyerere later (1968) promised a mass
rally in Dar es Salaam that trusted youths who proved to be hard working
would be given weapons training. Apparently, the original intentions of 1964
had not been implemented.80
Another intention was to expand the defense force beyond the limited
number of national servicemen and army by creating a national reserve. A
field force of militarized police was to be formed in each region, and special
village police, consisting of volunteers working under two regular
policemen, were to be posted to villages —particularly border areas and
other places where there has been occasional trouble.81 The people have
been told that they are obliged to help these policemen.
There is no information as to whether this national reserve has become an
effective force. Villages are, in fact, sparsely settled areas spread out for
miles and hard to police. And Tanzania has had one of the smallest salaried
police forces in the world in relation to the population and size of the
country.82 In response to this situation, a special constabulary of volunteers
has been established to train for police work; its members are recruited from
the TYL. Although most regions have such a force, it is best organized in
Dar es Salaam.
None of these measures guarantees a loyal and pliable security force,
however. For one thing, an army with a large element of politically aware
TYL youth may be difficult for the leadership to digest, for the TYL has
caused trouble in the past. Despite sporadic attempts to governmentalize it
by incorporating it into a Ministry of Culture and Youth, the TYL has
managed to retain its identity. Difficulties not related to the new recruits
have also sprung up in the army since 1964. When the Congo border became
very unstable, the government decided to allow some of the disgraced
mutineers back into the army as a stop-gap measure; shortly thereafter,
unexplained trouble broke out in the army.83 There have also been rumors to
the effect that a Zanzibari unit posted near Mozambique was very
undisciplined and had to be shipped back. In 1969-70 other army officers
were arrested and accused of plotting to overthrow Nyerere.
Conclusions
Certain inferences may be drawn about the three countries under discussion.
It does not appear that the Tanzanian People’s Defense Force has been
completely free of internal conflicts or that relations between the military
and civilian leaders have been without any tension, although the army has
posed no threat to the TANU government since 1964. Both the Kenyan and
the Tanzanian governments seem to have insulated their respective armies
from major political and ethnic issues during the last ten years. In both cases,
the personal position of the leader was stronger than in Uganda. In addition,
the Kenyan civil service and TANU provide institutional bases of support
not available in Uganda. At the same time, the Kenyans were able to build
on a military much less internally split and with better trained men than the
Ugandan army. The Tanzanians started afresh in 1964 with a much more
thorough housecleaning than the Ugandans undertook. In a more consensual
society, the Tanzanians appear to have had some success in welding the army
to the TANU government, although it is not clear that specific TANU
institutions have played a major role in this process. The Tanzanian army is
itself occupied with conflict on the southern border with Mozambique; and
such a preoccupation with external rather than internal conflicts acts to keep
an army out of constant intervention in domestic affairs. If the army handles
the external conflict creditably, it works to increase the possibilities of
building a disciplined force in the future.
It cannot be assumed that the Kenyan or Tanzanian armed forces will not
become increasingly active in domestic politics or that their military will
never make coups. But this article has tried to suggest that there is a much
greater probability of military intervention in Uganda because of the nature
of the society. Indeed, as General Amin was reported to have said after the
1964 coup, the Uganda army had been rather openly keeping the regime in
power. The Uganda coup may not have been well planned;84 it may have
been triggered by Amin’s own perception that he would be removed soon
and his awareness of the growing power of nonregular army forces.85 But
these were mere précipitants of the coup;86 basically, it was caused by
underlying conditions in Uganda that differed from those in Tanzania and
Kenya in important respects.
We can perhaps draw some conclusions about the prospects for military
rule from certain features shown by the intervention itself. The reason why a
military intervenes is important. Armed forces can make coups for rather
narrow professional and interest group reasons: they can claim that the army
is not being paid enough, that its honor is being sullied, or that it is being
destroyed by the politicians; or they can have personal fears of particular
officers. Alternatively, the military may intervene for broad political goals.
In Uganda, the military seems to have reacted to personal fears of the
officers, to growing discontent with the way the regular army was being
treated, and to unhappiness about the political situation, especially its ethnic
components. An important factor is that the military was internally split
before and during the coup. In effect, one part of the military fought against
another part that was more closely allied with Obote. Yet General Amin and
his colleagues moved slowly and cautiously after the coup. They do not
appear to have had clearly drawn political goals. The new cabinet that Amin
announced was composed largely of civil service personnel—it had two
soldiers, three politicians, and seventeen civil servants. Only three Baganda
were in the cabinet although the coup was at first enthusiastically received in
Baganda. If the military remains internally split and if the officers in Uganda
are not political animals, then the prospects for either successful or long-
term military rule are diminished.
This article has tried to show that we should not overhomogenize African
societies or their armed forces. We shall have to continue to explore the
interrelationships between armed forces and societies and our comparisons
are going to have to deal with interactions.
All three East African countries discussed in this article have problems of
political participation and political order. However, the army’s intervention
in politics cannot be deduced from a ratio of participation to
institutionalization. We must agree that military explanations do not explain
military interventions and that the general politicization of social forces and
institutions must be examined.87 At the same time, knowing about rates of
mobilization, degrees of social cohesion, and levels of political
institutionalization will not tell us all we want to know about the
potentialities of military intervention in a given society.
The discussion of East African armies and societies suggests that
institutionalization and performance of the armed forces are critical
factors.88 But analysis of the armed forces—the special conditions which
pertain to each individual military—must be linked to wider societal
factors.89 This means we should be studying the connections among
variables.
We examine the recruitment procedures of armed forces within the
context of analysis of class, ethnicity, and factional cleavages in society at
large. We examine attitudes of soldiers not in isolation, but by looking at the
various groups and social formations that espouse ideologies and programs.
Thus we look at the nature of the participation of armed forces in politics by
analyzing patterns of political participation throughout a society.
At the same time, in order to understand political participation in society
at large, we must examine the internal workings of armed forces
organizations.
1. Reports of the Uganda coup may be found in: Foreign Broadcast Information Service, Daily
Report, Middle East and Africa, 25 January 1971-1 February 1971; New York Times, 25 January-1
February 1971; Economist, 6 February 1971, p. 27, 30 January 1971, pp. 14-15; East African Standard
(Nairobi), 25 January 1971 ff., and Uganda Argus (Kampala), 25 January 1971 ff.
A recent analysis of the 1971 Uganda coup is in Michael F. Lofchie, “The Uganda Coup: Class
Action by the Military,” Journal of Modern African Studies, X (May 1972), 19-36. Two replies critical
of Lofchie’s arguments are: John D. Chick, “Class Conflict and Military Intervention in Uganda,”
Journal of Modern African Studies, X (December 1972), 634-37; and Irving Gershenberg, “A Further
Comment on the 1971 Uganda Coup,” in ibid., 638-40. Another major analysis of the Uganda coup is
Michael Twaddle, “The Amin Coup,” Journal of Commonwealth Political Studies, X (July 1972), pp.
112-28.
2. For a discussion of the revolts in 1964, see Henry Bienen, “Public Order and the Military in Africa:
Mutinies in Kenya, Uganda, and Tanganyika,” in Bienen, ed. The Military Intervenes: Case Studies in
Political Development (New York, 1968), pp. 35-70.
3. Samuel Huntington, Political Order in Changing Societies (New Haven, 1968), p. 194.
4. For a discussion of the modernizing capabilities of the military in developing countries, see Bienen,
ed. The Military and Modernization (New York, 1971).
5. Some interesting work has been done recently on compositions of officer corps in Africa. For a
general study, see J. M. Lee, African Armies and Civil Order (London, 1969). On Nigeria, see John
Colas, “Social and Career Correlates of Military Intervention in Nigeria: A Background Study of the
January 15th Coup Group” (Paper delivered at Annual Meeting, Inter-University Seminar on Armed
Forces and Society, Chicago, October 9-11, 1969). Also on Nigeria, see Robin Luckham, “Authority
and Conflict in the Nigerian Army, 1966: A Case Study in the Transfer of Military Institutions” (Paper
presented to Seventh World Congress of Sociology, Varna, Bulgaria, September 1970).
6. The ideas of the praetorian state and praetorian society have been developed in Huntington’s
Political Order and in Amos Perlmutter, “The Praetorian State and the Praetorian Army,”
Comparative Politics (April 1969), 382-404, and Perlmutter, “The Arab Military Elite,” World
Politics, XXII (January 1970), 269-300. Also see David Rapoport, “A Comparative Theory of
Military and Political Types,” in Huntington, ed. Changing Patterns of Military Politics (New York,
1962), pp. 71-101, especially pp. 71-74. This essay was based on Rapoport’s “Praetorianism:
Government without Consensus” (Ph.D. diss., University of California, Berkeley, 1960). Huntington’s
arguments will be used for discussion here.
7. Huntington, Political Order, p. 194.
8. Ibid., p. 196.
9. Ibid., pp. 196-97.
10. Ibid., pp. 197-98.
11. Ibid., p. 200.
12. Ibid., p. 212 and pp. 216-17.
13. Ibid., p. 200.
14. Ibid.
15. Ibid., p. 221.
16. In the discussion which follows, my aim is to consider Huntington’s idea of praetorianism for the
light it may throw on civil-military relations in East Africa. I am not here interested in reviewing
Political Order in Changing Societies.
17. For political analyses of East African countries, the following may be consulted: Stanley Diamond
and Fred G. Burke, eds. The Transformation of East Africa (New York, 1966); Bienen, Tanzania:
Party Transformation and Economic Development (Princeton, 1970); William Tordoff, Government
and Politics in Tanzania (Nairobi, 1967); G. Andrew Maguire, Toward “Uhuru” in Tanzania (New
York, 1969); Goran Hyden, Tanu Yajenga Nchi: Political Development in Rural Tanzania (Lund,
1968); Burke, Local Government and Politics in Uganda (Syracuse, 1964); David Apter, The Political
Kingdom in Uganda (Princeton, 1962); Julius Nyerere, Freedom and Socialism (Dar es Salaam, 1968);
Nyerere, Freedom and Unity (Dar es Salaam, 1967); M. P. K. Sorrenson, Land Reform in the Kikuyu
Country (Nairobi, 1967); John Nottingham and Carl Rosberg, Jr., The Myth of Mau Mau (New York,
1966); Oginga Odinga, Not Yet Uhuru (New York, 1967); Cherry Gertzel, The Politics of Independent
Kenya (Nairobi, 1969); Gertzel, Mauré Goldschmidt, and Don Rothchild, Government and Politics in
Kenya (Nairobi, 1969),
18. See M. Crawford Young, “The Obote Revolution,” Africa Report, XI (June 1966), 8-15.
19. These remarks are taken from Bienen, “What Does Political Development Mean in Africa?”,
World Politics, XX (October 1967), 128-41.
20. For a discussion of one-party systems in Africa, see Bienen, “The Ruling Party in the African
One-Party State: TANU in Tanzania,” Journal of Commonwealth Political Studies, V (November
1967), 214-30 and “One-Party Systems in Africa,” in Huntington and Clement H. Moore, eds.
Authoritarian Politics in Modern Society (New York, 1970), pp. 99-127. Huntington’s framework
allows for political development in poor countries; he singles out the Congress party in India and
TANU in Tanzania as vital parties which provided an institutional base to the political system.
21. Bienen, Tanzania.
22. See Bienen, “Public Order and the Military in Africa: Mutinies in Kenya, Uganda, and Tanzania,”
in Bienen, ed. The Military Intervenes, pp. 35-70.
23. For the district base of KANU politics, see John Okumu, “Charisma and Politics in Kenya,” East
Africa Journal, V (February 1968), 9-16.
24. There are certainly many “almost”parallels one could draw between KANU and the UPC.
Ugandan cabinet ministers were arrested under Obote, while in Kenya Odinga and other Luo M.P.’s
were incarcerated in late 1969 after they had left the government and were already in opposition.
Bildad Kaggia, a former assistant minister, had earlier been put in detention after he also had left the
government.
25. For a discussion of violence and ethnicity in Uganda, see Martin R. Doornboos, “Kumanyana and
Rwenzururu: Two Responses to Ethnic Inequality,” in Robert Rotberg and Ali A. Mazrui, eds. Protest
and Power in Black Africa (New York, 1970), pp. 1088-1138. Also see G. S. Engholm and Mazrui,
“Violent Constitutionalism in Uganda,” Government and Opposition, II (July-October 1967), 585-99;
Colin Leys, “Violence in Africa,” Transition, V (March-April 1965), 17-20; Young, “The Obote”; A
Special Correspondent, “The Uganda Army: Nexus of Power,” Africa Report, XI (December 1966),
37-39.
26. Huntington, Political Order, p. 1%.
27. While Dar es Salaam has recently been growing by about 10 percent annually, the urban
population as a whole grows at about 6 percent in Tanzania. In 1970, the urban population was around
5 percent of the total; there were 10 towns of 15,000 or more people. See Tanzania Second Five-Year
Plan for Economic and Social Development, 1st July 1969-30th June 1974, I (Dar es Salaam, 1969),
pp. 176-82.
28. Tanzania’s northwestern region has also had a stratified land system, called the nyarubanja, in the
Eastern Buhaya area, which was not so different from Buganda. See Hyden, p. 79.
29. Bienen, “An Ideology for Africa,” Foreign Affairs, XLVII (April 1969), 545-59.
30. Huntington, Political Order, p. 222.
31. I argued in The Military Intervenes, pp. 45-46, that the decisive factor in the army mutinies of
1964 was the fragility of East African institutions per se. I am now arguing that we must explain the
role of the armed forces in East Africa since 1964 in terms of the interaction of specific armed forces
with their environments. While the emphasis here is on domestic environments—that is, the societies
of each East African state—a full consideration would include the interactions of the East African
armies with the international environments. The 1964 mutinies, for example, must be related to the
Zanzibar Revolution. (See Bienen, “National Security in Tanzania after the Mutiny,” Transition, V
[April 1965], 39-46.) Explanations of subsequent civil-military relations must take into account the
liberation movements in southern Africa and Tanzania’s army, the Kenya-Somali conflict, the Congo
upheavals, and also refugee problems in southern and northern Uganda and the reaction of the Uganda
army to the refugees.
32. Colonel T. N. Dupey, ed., The Almanac of World Military Power (Harrisburg, 1970) gives armed
forces figures very close to Richard Booth, “The Armed Forces of African States, 1970” (See Table 1)
for Uganda, a smaller armed force number for Tanzania, and considerably larger figures for Kenya
because internal security forces are included in the latter. See pp. 208-9, 246-47, 252-53.
33. In 1967, armed forces were put at 3,000 for Tanzania, 4,775 for Kenya, and 5,960 for Uganda. See
Charles Stevenson, “African Armed Forces,” Military Review, XLVII (March 1967), 18-24, and David
Wood, “The Armed Forces of African States,” Adelphi Papers, No. 27 (April 1966).
34. Dupey, p. 248, puts the size of the Tanzanian air force at 400 men.
35. For a discussion of African police forces, see Christian Potholm, “The Multiple Roles of the
Police as Seen in the African Context,” Journal of Developing Areas, III (January 1969), 139-58.
36. See Bienen, “When Does Dissent Become Sedition?”, Africa Report, XIV (March/April 1969),
10-14. Also see his “Foreign Policy, the Military, and Development: Military Assistance and Political
Change in Africa,” in Richard Butwell, ed. Foreign Policy and the Developing Nation (Lexington,
1969).
37. From Lee, p. 44.
38. Ibid. At the time of independence, Kenya had 13.9 percent Africans in its gazetted police officers;
Uganda had 42.9 percent, and Tanganyika 10.4 percent. Since police played a large internal security
role in Kenya after Mau Mau, the relatively high percentage of Kenyan African officers in the army
may have been the other side of the coin of low percentage of Kenyan Africans in the police officer
corps.
39. Harvey Glickman, Some Observations on the Army and Political Unrest in Tanganyika, Duquesne
University, Institute of African Affairs, Paper No. 16 (Pittsburgh, 1964), p. 4.
40. Ibid.
41. Ibid.
42. Ibid.
43. Lee, pp. 75-76.
44. A Special Correspondent, p. 38.
45. Lee, p. 105.
46. From time to time the Government of the Sudan threatened to pursue Sudanese rebels into
Uganda.
47. Leys, “Violence in Africa,” Transition, V (April 1965), 18-19. Leys reports that in only nine
months of 1961, 253 people were killed in Karamoja.
48. For a discussion of the separatist movements in western Uganda, see Doornboos, esp. pp. 1088-
1136.
49. A Special Correspondent, p. 39.
50. Ibid.
51. A motion by an opposition M.P. to suspend Colonel Amin was passed in the National Assembly
with only one dissenting vote. (See Africa Report, XI [March 1966], 22.) Colonel Amin was later
cleared by a commission of inquiry and became head of the army as Obote consolidated his power
with Amin’s support. Also see Young, p. 12.
52. There were more than 18,000 Baganda exservicemen in the 1950s who were veterans of World
Wars I and II. See Eugene Schleh, “Post-Service Careers of World War Two Veterans: The Cases of
Gold Coast and Uganda” (Paper delivered at 1967 Annual Meeting, African Studies Association, New
York City, November 1-4), p. 9.
53. East African Standard, 27 January 1964, quoted in Mazrui and Rothchild, “The Soldier and the
State in East Africa: Some Theoretical Conclusions on the Army Mutinies of 1964,” Western Political
Quarterly, XX (March 1967), 83.
54. Bienen, “Public Order and the Military in Africa,” p. 50.
55. Ibid. Although the mutineers held the Minister for Defense captive, they did not move out from
Jinja, where they were barracked, to the main centers at Kampala and Entebbe.
56. As of 1967, the annual starting salaries of nonofficers were considerably higher in Uganda than in
Kenya. A Kenyan sergeant received about $1,000 a year compared to a Ugandan’s $1,500. A private
received about $350 in Kenya, $600 in Uganda. Officers had more nearly comparable salaries from
Second Lieutenant up. Lee, p. 94.
57. A Special Correspondent, p. 38.
58. Ibid.
59. Ibid.
60. Lee, pp. 42, 108-9.
61. Ibid., p. 108.
62. A Special Correspondent, p. 39.
63. Lee, p. 105, says that it may be the highest on the continent.
64. A Special Correspondent.
65. Lee, p. 106.
66. Ibid., p. 110. The Luyha had 16.5 percent of the police, while the Kikuyu had 11.2 percent, Luo
8.5 percent, and Kamba 9.8 percent. Lee’s figures for officers are for 1966, but the population shares
are based on the 1962 census. The police figures are also for 1962.
67. The Kamba areas also had given support to a Kamba leader’s own party—the African People’s
Party, in the 1963 elections, rather than to either KANU or KADU, the major parties. In the 1966
Little General Election, Kamba areas gave some support also to the opposition Kenya Peoples Union.
The Kamba, then, have been an important “swing” tribe in Kenya.
68. Lee.
69. The social pages of the East African Standard frequently announce the marriages of young Kikuyu
officers to daughters of prominent Kikuyus. A study of marriage and political/economic relationships
would be most interesting in this area.
70. A 1968 issue of Majeshi Yetu (Armed Forces journal) makes interesting reading. It is published
with English and Swahili pages, is well written and professional. By-lined articles can be found by C.
L. Galloway, Commander A. A. Pearse, and Major A. M. Tippett, among others.
71. See the statement by then Prime Minister Kenyatta of 25 January 1964, reprinted in Gertzel,
Goldschmidt, and Rothchild, eds. Government and Politics in Kenya (Nairobi, 1969), p. 562.
72. Kenneth Grundy, Conflicting Images of the Military in Africa (Nairobi, 1968), p. 29.
73. The bulk of the two battalions in Tabora and Dar es Salaam were dismissed and mutineers were
sent back to their home areas. They had to report periodically to area commissioners. Internal security
was provided first by British and then by Nigerian troops.
74. The following discussion is taken from Bienen, Tanzania: Party Transformation and Economic
Development (Princeton, 1970), pp. 374-80. I have also commented on the formation of new
Tanzanian armed forces in the previously cited “National Security in Tanzania after the Mutiny,” and
“Public Order and the Military in Africa: Mutinies in Kenya, Uganda, and Tanganyika.”
75. Tanganyika Standard, 12 February 1964, p. 5.
76. Lee, p. 150. I estimate that somewhat fewer than 5,000 men actually passed through the national
service by 1967.
77. Ibid.
78. Nationalist, 25 June 1964, p. 2.
79. The Reserve Forces Bill legalizes paramilitary training and functions of the TYL.
80. Africa Report, XIII (April 1968), 23-24.
81. Reporter (Nairobi), 12 February 1965; Nationalist, 2 February 1965, p. 1.
82. The police force grew by 11 percent per annum between 1961-67 (Lee, p. 105). This was the
highest rate of growth in East and Central Africa.
83. The 12 September 1964 headline of the Nationalist read: “U. R. [United Republic] Army Arrests.”
The text of a government announcement was printed: “In active pursuance of its duty to maintain the
integrity and safety of the United Republic the government yesterday found it necessary to arrest and
detain a small number of servants of the Republic. This number included Officers and other ranks of
the United Republic Army, who were of doubtful loyalty and guilty of insubordination by default.” It
was on this occasion that Elisha Kavonna, who had been nominated by the 1964 mutineers, was
arrested.
84. The Economist made this assertion in its issue of 6 February 1971, p. 27.
85. General Amin reportedly said that a secret meeting was held on January 11 at which a decision
was taken to murder him and others. Amin was said to have claimed that the president, then in
Singapore, was kept abreast of the plot throughout by communications with the Minister of
International Affairs. Foreign Broadcast Information Service, Middle East and Africa, January 1971,
p. U-6.
86. Twaddle also uses the language of “précipitants” and “preconditions” following the terms in Harry
Eckstein, “Introduction. Toward the Theoretical Study of Internal War,” in Eckstein, ed. Internal War
(Glencoe, 1964), pp. 1-32. Twaddle sees the Uganda coup not so much as an army mutiny as part of an
internal war situation. Lofchie (p. 19) argues that the Uganda army “can best be understood as a kind
of economic class, an elite stratum with a set of economic interests to protect.” Chick (p. 634) takes
Lofchie to task over the notion of stratum and class as applied to the Uganda army, and both he and
Gershenberg doubt whether the Obote government threatened the military’s interests.
87. Huntington, Political Order, p. 194.
88. I have greatly benefited from reading Abraham LowenthaTs “Armies and Politics in Latin
America: A Review of the Recent Literature,” World Politics, forthcoming. Lowenthal emphasizes the
need to examine institutionalization within armies and to analyze the comparative strength of civilian
and military organizations in specific countries.
89. Perlmutter, “The Arab Military Elite,” p. 276.
* This article was written under the auspices of the Center of International Studies, Princeton
University.
A Comparative Analysis of the
Political and Economic Performance
of Military and Civilian Regimes
A Cross-National Aggregate Study
The analysis of military intervention in the form of coups has achieved the
status of an institutionalized topic in comparative politics. In marked
contrast, the examination of the performance of military regimes established
as a consequence of coups has attracted little attention.1 This is particularly
unfortunate since it is the performance of a regime rather than its origin (i.e.,
whether it emerged as a consequence of an election or a coup), which is of
greater significance when assessing the regime’s effect on the processes of
economic and social change.
This article is the second of a series of reports on a cross-national
aggregate project designed explicitly to examine the performance of military
regimes. The main purpose of the piece is to evaluate the performance of
military and civilian regimes along a range of political, economic, and
military variables.
Research Design
It is clearly impossible to assess adequately the performance of military
regimes without comparing their performance to that of civilian regimes.
This requirement has led to the definition of four populations of regimes.
The first consists of military regimes (MR). A military regime is one in
which the armed forces have made a coup, established a government in
which the main executive post is held by a military person, and have stayed
in power for at least the major part of one year. If a country has experienced
several periods of military rule, these have been aggregated. The second
population consists of the periods of civilian rule in countries that have
experienced a military regime (CRM). The third population consists of all
other low-income countries which have experienced only civilian rule (CR
900-). The fourth population is that of the high-income systems (CR 900+)
and is included primarily for reference purposes.2 This article deals mainly
with the MR, the CRM, and the CR (900-).
The total population, constituted by the four subpopulations cited, consists
of all independent countries in the world with a population greater than one
million. The Communist countries were excluded due to data shortage and
data incomparability. The time span over which performance is examined is
the period 1951-70.
The comparison of the three main regime types attempts to look for
differences and similarities among the populations. In the exercise not only
can we use a variety of standard tests; but we can also rely on the designs of
the populations. The strongest part of the design relates to the comparison
between the MR and the CRM. Although the populations of the MR and the
CRM cannot be regarded as matched pairs, they are as close to matched
pairs as one is likely to find m macro cross-national aggregate research, and,
as such, control for a large number of intervening variables. Even this form
of comparison, however, is inadequate without an additional reference point;
hence, the inclusion of the CR (900-).
The advantages of this type of design may be seen from a brief survey of
the types of outcome that can be anticipated when these three populations
are compared on any variable. Four main types of outcome may be defined;
these can be summarized conveniently by an adaptation of simple set
theory.3 If we let X, Y, Z represent three sets, each of which corresponds to
the mean score of the distribution of the three populations MR, CRM, and
CR (900-) on a particular variable, then the four main types of outcome can
be defined as:
Cluster Analysis
Before the different types of regime are compared in detail across the range
of political, military, and economic variables, it is useful, given the large
number of cases and the large array of variables, to construct a summary
picture of the population. The most appropriate multivariate technique for
this purpose is cluster analysis. As a method of empirical typology
formation, cluster analysis permits the construction of profiles, thus
providing an excellent way of simplifying and categorizing the population.
Specifically, cluster analysis enables us to find out which regimes group
together, and which variables provide the defining characteristics of any
such groupings. Thus, in addition to providing an empirical clustering of
cases, and an empirical profile of each cluster in terms of the political,
military, and economic variables, it is possible to examine in general terms
whether the MR cluster separately from the CRM, and whether the CRM
cluster separately from the CR, and which variables, if any, make the MR
distinctive.
The cluster analysis was run on 115 cases, with 23 variables. An
interactive method—relocation—was used. This is superior to a number of
other hierarchical methods in that it permits relocation of cases once fusion
has begun. The similarity coefficient used was the error sum, which
basically seeks to minimize variance within clusters. One main problem of
cluster analysis is that clustering takes place even in the absence of empirical
justification for clustering. However, there are two ways in which the
accuracy of the clusters may be checked. First, the dendrogram gives a
measure of the error scores at which fusion takes place, and, second,
examination of the F ratio and T tests gives a measure of the degree of
homogeneity of each of the clusters. Using these criteria, the cluster level
chosen for the extraction of clusters was level six.10
Political Performance
The analysis of variance on each of the political indicators shows highly
significant variation (at .01 level) among the regime types. Difference testing
for each variable produces two main patterns. One indicates executive
change where the CR (900-) appear to be the deviant population; the other
represents all remaining political variables where MR have much higher
scores than both CRM and CR (900-)—which, in turn, have slightly higher
scores than the CR (900+).
The average tenure of the main executive is significantly shorter in the
MR than in the CRM (at the . 10 level). The average tenure in the CR (900-),
however, is approximately twice that of the CRM (with differences between
the CR [900-], with both the MR and the CRM significant at the .01 level.)
Two explanations are suggested for the difference. The greater circulation of
the main executive in the CRM may be responsible for encouraging military
coups, although this explanation does not seem to be satisfactory since the
average tenure of the main executive in the CRM closely correlates with that
in the CR (900+). The more likely explanation is that military coups
terminate civilian (CRM) regimes, and the shorter tenure is more a
consequence than a cause of coups.
The restrictions on the constitution, assemblies, and political parties
demonstrate a consistent and expected pattern. The MR are clearly
differentiated from the very similar CRM and CR (900-), both of which have
higher scores than the CR (900+).12 Using the CR (900-) as a reference
point, this general finding suggests two conclusions. First, CRM do not
maintain the types of restrictions imposed during the period of military rule.
Second, military regimes are not likely to occur in systems where levels of
political activity are generally low.
The restrictions on political activity under the MR are much higher than in
any of the other regime types, but they are not total. Assemblies are banned
for approximately 75 percent of the regime, but parties are banned for only
half the period of military rule. It is, however, obvious that one function of
political parties is terminated if assemblies are banned. This lower priority
on the part of military regimes to ban parties may be explained by several
factors: (1) any ban on assemblies must entail some restriction on parties; (2)
the practicalities of curtailing party activities are likely to be greater than
imposing and maintaining bans on the constitution or assemblies; and (3) the
military may find it advantageous to rely on an existing party or parties
either to recruit civilians with governmental expertise or to provide some
means for political mobilization through a party led by the military.
The view that military regimes are hostile toward communism is
convincingly supported by the fact that Communist parties are scarcely
tolerated at all.13 The scores for both the CRM and the CR (900-) indicate
that they also discriminate against the Communist party.
Political restrictions, measured by the percentage of cabinet posts held by
the military, again show the MR with a significantly higher score than either
the CRM or CR (900-), which in turn have significantly higher scores than
CR (900+).14 For the first time, however, there appears to be a significant
difference between the CRM and the CR (900-). These results are of interest
because it is clear that, while military regimes place little reliance on
assemblies, much more extensive use is made of civilians in terms of
composition of the main executive body. In addition, the absence of
significant differences between CRM and CR (900-) on the other political
variables indicates the lack of any continuity between military and civilian
periods of rule, at least in terms of political restrictions. The appearance of a
significant difference on this variable would seem to indicate some
continuity, but any continuity between military and civilian periods of rule
can be easily explained. Thus, a greater percentage of cabinet posts held by
military personnel in a civilian regime may predispose the military to initiate
a coup. On the other hand, with the termination of a military regime, the
military may wish to retain some degree of formal control through the
retention of certain cabinet posts. While the significant difference between
the CRM and the CR (900-) cannot be denied, it should be noted that the
scores for neither regime are very high, nor is the absolute difference
between them substantial. Therefore, it is suggested that the general finding
on the absence of continuity between MR and CRM in terms of political
restrictions is not weakened.15
A two-way analysis of variance of GNP and regime type on the political
variables shows, with few exceptions, the persistence of differences among
regime types and the absence of differences among GNP levels.16 In other
words, the patterns of restriction apparent in the aggregate findings are
attributable to regime type rather than to GNP. The first slight deviation
relates to executive tenure. While there is still the significant variation
between regime types, there is also an interaction effect (both at the .01
level). Closer examination shows this to be due to GNP level three, where
the score for the MR is almost twice that for the MR in the two other levels;
where the CRM actually have a lower score than the MR; and where the CR
have the lowest score.17 If there were any relation between civilian
institutionalization and GNP, it might have been expected that as GNP
increased, the military regimes would become more unstable. This is not the
case, however; and, in terms of tenure, military regimes are more stable at a
higher GNP level.
A second exception relates to the ban on the constitution—the only factor
which varies significantly with GNP. This variation is largely attributable to
the very low score of the military regimes in GNP level three. Since these
military regimes ban assemblies and parties to a degree similar to other
levels, however, little importance can be attached to this deviation.
A final point is that, although there is little variation and interaction effect
among GNP levels, the civilian regimes in GNP level three consistently have
the least restrictions. Since the scores of both the MR and CRM in this level
are no lower than the same type of regime in the other levels, the disparity
between countries that have experienced military regimes and those which
have not is most acute in the low-income systems with the highest level of
economic development.
The introduction of area controls does not lead to any important changes
in the aggregate findings.18 Irrespective of area, levels of political
restrictions are highest in military regimes, whereas little difference appears
between the two types of civilian regimes.
Military Performance
When all four populations are compared by analysis of variance, significant
variation at the .05 level appears on military size, but not on expenditure.
Closer examination, by difference of means tests, shows that, with reference
to size, the deviant category is the CR (900+). No significant differences
appear among the other three populations, but all have military forces
significantly smaller than the CR (900+) at the .01 level. While no
significant variations appear for military expenditure, the absence of
difference is undoubtedly due to the substantially higher GNP of the CR
(900+). The CR (900+) would again have been deviant if military
expenditure had been measured on a per capita basis.
From these findings four main conclusions may be drawn. First, the
absence of any significant differences between MR and CRM seems to
indicate that military rule does not lead to an increase in military expenditure
or in the size of the armed forces. If military rule periods precede civilian
rule periods and if there is the usual steady rate of growth over time in
military size and expenditure, then time may obscure an actual expenditure
increase by military regimes. (The distribution of military regimes, however,
is such that the majority are of more recent origin.) Considering positive
rates of growth, it is surprising that size and expenditure under military
regimes are not higher. No evidence is given to support the popular concept
that military regimes show a special predisposition to the armed forces in
terms of increased size and expenditure.21 Second, it may be argued that
military regimes occur in systems presently demonstrating large military size
or high expenditure, thereby obviating the need for military regimes to
increase size or expenditure; but the absence of any significant differences
among MR, CRM, and CR (900-) indicates that this is not the case.
Inversely, we may further conclude that noncoup systems avoid coups by
catering to the military; but again, the absence of any significant differences
between MR and CRM and CR (900-) seems to disprove this. A final
conclusion concerns the position of CRM vis-à-vis CR (900-) and MR.
Unlike the political variables, where significant differences between MR and
CRM appeared, with CRM lying closer to CR (900-) than to the MR
counterparts, there is no evidence that the type of regime is an adequate
predictor of military size and expenditure.
Controls for GNP fail to produce any significant differences, thereby
reinforcing the main finding that there are no differences in military size and
expenditure between the MR, CRM, and CR (900-). Two-way analysis of
variance of regime type by GNP on size shows no significant variation
between regime types controlling for GNP and no significant interaction
effect, while controlling for regime type shows significant variation among
GNP levels.22
Two-way analysis of variance on expenditure shows similar results. There
is no significant difference between types of regime controlling for GNP, but
there is a significant variation between GNP levels.23 Size and expenditure
seem to bear rather different relationships to GNP levels. Size doubles from
GNP level one to GNP level two and remains high, while expenditure
increases by some 30 percent from level one to level two, but then decreases
by some 50 percent. Another difference between size and expenditure is
found when one-way analyses of variance are run on size and expenditure
for each of the different GNP levels. A significant difference (at the .05
level) appears between the expenditure of the three regime types at GNP
level one. While this single finding can hardly disturb the observation that
GNP controls do not provide major differences, it does indicate that, in low-
level GNP systems, the countries with high military expenditure have
military regimes. Since there is no significant difference between MR and
CRM, however, there is no evidence that military regimes in low-level GNP
systems take advantage of military rule to increase expenditure. It is more
likely that any armed force must reach a certain level of development before
it is capable of establishing a regime.
Table 5 Military Size, by Regime Type and GNP Level
Table 10. Rate of Growth of GNP, by Regime Type and GNP Level
Conclusion
It is useful to compare our summary picture of military regimes with the
current literature in which three quite different views of military regimes are
represented. The first of these, most prevalent in the early sixties, proposes
that the military are agents of development and modernization.37 The second
view, generally accepted from the mid-sixties onward, sees the military as
antithetical to development.38 A major problem arises in our attempt to set
our findings in perspective with other studies if we support the view that the
concepts of development and modernization have little utility. Indeed, given
the criticisms that have been levelled against the concepts of development
and modernization from the mid-sixties, it is rather curious that writers still
persist in their endeavors to “explain” certain facets of military regimes by
reference to these concepts.39 Thus, there is no agreed set of criteria for
either development or modernization through which the performance of
military regimes can be evaluated.
If we examine the two opposing views of the military, some of the
differences are easily explained. One source of difference lies in the fact that
these studies often look at quite different phenomena. In this respect, it is not
surprising that their general pictures differ. This perfectly legitimate
divergence only becomes anomalous when findings are processed through
the concept of modernization. A second source of difference arises because
many studies are either lacking in empirical data or tend to draw
overgeneralized conclusions on the basis of their data. There is an interesting
similarity, however, between the two opposing views in that both sides tend
to see military regimes as an homogeneous group clearly differentiated from
the group of civilian regimes. On this basis we would diverge most strongly
from many studies of military regimes.
This point of divergence has given rise to the third view of military
regimes. Our study supports the more recent studies of the case or area
variety, strongly emphasizing the diversity of military regimes.40 At the most
general level, our findings suggest three main points: (1) military regimes do
not in aggregate form a distinctive regime type in terms of performance; (2)
there is a degree of diversity within military regimes, which is not dissimilar
to the diversity found within civilian regimes; and (3) the general degree of
similarity or dissimilarity between military and civilian regimes varies from
one variable or one category of variables to another.
In this respect, we have found the type of picture generated by the cluster
analysis the most useful general overview of military regimes. This has
shown that, while some military regimes are quite clearly distinct, as indeed
are some civilian regimes, a sizable proportion of military and civilian
regimes are indistinguishable in terms of performance. We would argue,
therefore, that the nature of military regimes cannot be adequately
understood either in terms of assisting or inhibiting modernization or in
terms of a typology based on a simple distinction between military and
civilian regimes. Rather, it is the overview generated by the cluster analysis
and complemented by the array of difference tests which more accurately
and profitably describes the diversity of military regimes.
NOTES
1. For a review of the literature and a bibliography, see Henry Bienen, “The Background to
Contemporary Study of Militaries and Modernization,” in Bienen, ed. The Military and Modernization
(Chicago, 1971), pp. 1-39. Most of the work on military regimes is of a case or area study nature, but
there is one cross-national aggregate study to which we would call attention, Eric A. Nordlinger,
“Soldiers in Mufti,” American Political Science Review, LXIV (December 1970), 1131-48. This study
is the one most analogous to ours, but it differs in that some of Nordlinger’s findings do not tally with
ours and the general tenor of Nordlinger’s paper is dissimilar. There are valid reasons why our study
differs from Nordlinger’s and why we consider our study a more accurate representation of the
performance of military regimes. (1) We have a different range of variables—Nordlinger uses seven,
while we have used twenty-one; and only two of ours are directly comparable to Nordlinger’s. (2) We
use a different time period. Our data have been collected annually over the period 1951-70, whereas
Nordlinger uses data from the study by Irma Adelmann and Cynthia T. Morris, Society, Politics, and
Economic Development (Baltimore, 1967), which generally covers the period 1957-61 and in some
cases, 1950-63. (3) We use a slightly different independent variable. Nordlinger uses a threefold
categorization of the political strength of the military (measured over 1957-62, even though this does
not correspond with those dependent variables measured over 1950-63), whereas we have focused
explicitly on regimes. (4) We use a rather different design which permits us to isolate the effects of
military regimes more systematically and rigorously. Since we have collected data annually, means or
rates of growth can be measured independently over periods of military or civilian rule. Thus, we have
specifically compared military rule against civilian rule in countries that have experienced military
rule and again compared both of these with civilian regimes in countries which have not experienced
military rule. (5) We have used a more comprehensive population than Nordlinger. (6) We consider
that Nordlinger rather overstates his use on the basis of the data he presents. The controls could have
been better handled and some of the variables are very loosely defined and operationalized.
2. The GNP division of the CR is made because the higher income category differs markedly in
economic terms from the rest of the world. The empirical justification for this will be seen in the next
section, which deals with a cluster analysis. Further, no country with a mean per capita GNP greater
than $900 (1951-70) experienced a military regime.
3. Set theory notation fits very well with difference testing. The null hypothesis and the alternative
hypothesis for a two-tailed test may be written as ΧΠΥ τ* φ and ΧΠΥ = φ. If two population means do
not differ significantly, their distributions would overlap in a way in which two sets may intersect in a
Venn diagram. The principle of significance can also be applied to the definition of a set when a set is
conventionally defined as consisting of a well-defined set of elements. If the mean score of a variable
set X differs from that of set Y, then we can argue that we have two sets, which are well defined in the
sense of being significantly different.
4. For example, if we let X, Y, and Z represent the population CRM, CR (900-), and MR, other things
being equal, outcome type 1 would tell us that each of the populations is distinct. If the scores were
arranged Y<X<Z, then we would have to inquire whether it is the deviation of X from Y which gives
rise to Z, or whether it is the occurrence of Z which gives rise to the deviation of X from Y. To
illustrate again, outcome type 3 (as written [X∩Y]~Z) would provide strong evidence to indicate that
military regimes differ substantially from civilian ones. Had the outcome been written [Y∩Z)~X, then
we would know that countries which experience military regimes are somehow different from those
which do not. We would then be obliged to investigate the case of whether the difference between Y
and X gives rise to Z and Z proves unable to redress the balance, or whether it is the occurrence of Z
which leads to the difference between Y and X. Had the outcome been [X∩Z]~Y, then we would have
some evidence indicating that, although military regimes are not distinct, the military regimes in
countries which experience military rule rather than the civilian regimes of these countries put them on
a par with other civilian regimes.
5. Some of the more recent standard data compilations such as Charles L. Taylor and Michael C.
Hudson, World Handbook of Political and Social Indicators (New Haven, 1972), and Arthur S. Banks,
Cross-Polity Time Series Data (Cambridge, [Mass.], 1971) mark significant advances on earlier
collections from the standpoint of the time span covered. But even these proved to be of limited value
in view of the scope of the variables utilized.
6. The main data sources were: Europa Yearbook (London, 1958-70); Statesman’s Yearbook (London,
1951-70); Political Handbook and Atlas of the World, ed. Richard P. Stebbins and Alma Amoia (New
York, 1970); Whitaker’s Almanac (London, 1951-70); and Keesing’s Contemporary Archives (London,
1950-70).
7. The main data sources were: The Institute for Strategic Studies, The Military Balance (London,
1960-72); Stockholm International Peace Research Institute, Yearbook of World Armaments and
Disarmament (Stockholm, 1968-72); Richard Booth, “The Armed Forces of African States,” Adelphi
Papers, no. 67 (London, 1970).
8. The main data sources were: United Nations, Statistical Yearbook (New York, 1951-72); United
Nations, Yearbook of International Trade Statistics (New York, 1950-72); United Nations, Yearbook of
National Account Statistics (New York, 1964-72); International Monetary Fund, International
Financial Statistics (Washington, 1964-72).
9. The main areas not covered are central government capital, and monetary transfers involving
transactions of deposit money banks and central banks.
10. The scores for each of the clusters are contained in Appendix 1.
11. Of the thirty-two cases of MR and CRM, there are twenty-nine pairs (Table 2). Three MR have no
CRM and three CRM have no MR. The absence of a counterpart may be attributable either to lack of
data or to the fact that a MR has occupied the whole period. Spain, Thailand, and Algeria have no
CRM; Libya, Cambodia, and Paraguay have no MR. Since none of the difference tests treat the MR
and CRM as matched pairs, these cases are retained for the majority of comparisons.
12. Analysis of variance across the variables of constitution, assembly and party are significant at the
.01 level. Means testing shows significant differences between MR and CRM/CR (900-) on each
variable at the .01 level. Tests for the CR (900+) are null and void on account of their zero variance.
13. Analysis of variance is significant at the .01 level. Means testing shows that the MR have
significantly higher scores than the CRM and CR (900-) and CR (900+) at the .01 level; and the CRM
and CR (900-) have higher scores than the CR (900+) again at the .01 level.
14. Analysis of variance is significant at the .01 level. The MR have a substantially higher score than
the CRM, the CR (900-) and the CR (900+) at the .01 level: the CRM and CR (900-) have
significantly higher scores than the CR (900+) at the .01 level; and the CRM have a much higher score
than the CR (900-) at the .05 level.
15. This finding can be used to test the hypothesis that civilian regimes may provide greater capacity
for the military in the main executive to avoid a coup. If this were the case, it would be expected that
the CR (900-) would have a high score. Since this is not the case, there is no evidence to indicate that
civilian regimes may avoid a coup by catering to the military in giving the military greater powers in
the main executive body.
16. Per capita GNP is divided into the following categories: $1-150, $151-400, $401-900, and $901+.
17. One-way analysis of variance for levels 1 and 2 shows significant variation at the .01 level, but
means testing shows no difference between the MR and CRM. Analysis of variance in level 3 shows
no significant variation:
18. The areas are: Central America, South America, the Middle East and North Africa, Sub-Saharan
Africa, and Central and Southeast Asia.
19. Scores for the CRM and CR on constitution restrictions are 20, 48; for assembly restrictions, 26,
29; for party restrictions, 39, 35. For military posts in the executive, however, the scores are 3.8 and
0.3.
20. The scores for the CRM of Central and South America are 11 and 19, compared to 1, 3, 4 in
Africa, Asia, and the Middle East.
21. These data conflict with those presented by Nordlinger, p. 1135. However, in footnote 1 of this
article we have outlined some reasons that may account for this. Like Nordlinger, however, we find
that countries that have had military regimes have higher levels of expenditure and larger armies than
countries which have not had military regimes, even though the differences are slight. It could be
argued that we place excessive faith in difference testing but, on the other hand, if we remove the
difference tests, we could only substitute quite arbitrary judgments. Furthermore, if we examine
growth rates of size and expenditure in countries that have and have not had military regimes, we find
for size they are 8.1 percent and 7.9 percent, and for expenditure, 8.8 percent and 11.7 percent. Thus,
while growth rates for size are similar, growth rates for expenditure are actually higher in countries
that have not had military regimes (difference is significant at .10 level), Thus, the lower level of
expenditure in countries that have not had military regimes seems to be countered by a higher rate of
growth.
22. The variation is significant at the .05 level, and the mean scores for levels 1,2, and 3 are 27, 65,
and 58, respectively.
23. The variation is significant at the .05 level, and the mean scores for levels 1, 2, and 3 are 2.70,
3.67, 1.96, respectively.
24. The scores for Central America, South America, Middle East, Sub-Saharan Africa, and Central
and Southeast Asia on size are 23, 40, 86,. 10, 90; on expenditure they are 1.17, 2.24, 5.83, 1.71,4.92.
25. The one-way analyses of variance on each of the variables show significant variation at the .01
level.
26. The GNP variations among regimes and among areas are significant at the .01 level; and for
primary product at the .05 and .01 levels.
27. The variation between areas is significant at the .01 level.
28. The variation is significant in Asia at the .01 level; in Africa, at the .10 level; in South America, at
the .05 level; and in Central America, at the .10 and .05 levels, for MR and CRM, respectively.
29. In Central America and the Middle East variation is significant at the .10 level; and in South
America at the .05 and .01 levels, for MR and CRM, respectively.
30. For both exports and imports, MR and CRM have significantly lower scores than the CR (900-) at
the .01 and .05 levels, respectively. On trade balance, the MR, CRM, and CR (900-) have lower scores
than the CR (900+) at the .01 level, and higher scores on the investment balance for MR and CRM at
the .05 level, and for CR (900-) at the .10 level. There are no differences on international liquidity.
31. The return on this capital is included in the exports of the trade balance and is one reason for the
stronger trade balance position of the high income systems.
32. An important problem is that area differences may simply reflect GNP differences. Unfortunately,
the number of cases is too small to combine regimes, GNP, and area into contingency tables, and there
would be too many dummy variables to use multiple regression. However, since we know the scores
on the international trade variables of regimes by GNP, and since we know the GNP location of
regimes by area, we also know what score would be predicted for regimes by area using scores taken
from GNP tabulations. The problem can be partially overcome, therefore, by comparing actual scores
of regimes in areas to those that would have been expected from GNP.
33. Our findings for rate of growth of GNP do not seem to differ markedly from Nordlinger’s; he
finds a correlation of .13 between political strength of the military and rate of growth of GNP. We
differ in the inferences drawn in that we would argue that military regimes certainly perform no worse
than civilian regimes and, if anything, slightly better. Nordlinger seems to turn the absence of a high
correlation into an assertion that military regimes do not promote— and even retard—change. Of
course, Nordlinger would have needed a significant negative correlation to support this claim.
34. The mean scores for GNP levels 1, 2, and 3 for exports are 3.37, 7.26, 8.79; for the food index,
2.29, 3.00, 1.29; and for primary education, 4.19, 3.52, 1.24.
35. These differences are: (1) in Africa the MR have higher rates of growth of GNP and (2) lower
primary education growth (at .10 levels) than the CRM; (3) the MR in Asia have lower primary
education growth than the CRM (at . 10 level); (4) the MR in Central America have higher primary
education growth than the CRM (at .05 level); and (5) the MR of South America have a higher export
growth rate (at .05 level).
36. In Africa the CR have a significantly higher GNP and a lower cost-of-living growth than the CRM
(at .05 levels) and a higher primary education growth rate than the MR (at the .10 level). These are
consistent with the GNP tabulations.
37. For examples of this view, see: Lucían W. Pye, “Armies in the Process of Political
Modernization,” and Edward A. Shils, “The Military in the Political Development of New States,” in
John H. Johnson, ed. The Role of the Military in Underdeveloped Countries (Princeton, 1962);
Manfred Halpern, The Politics of Social Change in the Middle East and North Africa (Princeton,
1962); Guy J. Pauker, “South East Asia as a Problem Area in the Next Decade,” World Politics, XI
(April 1959); Hans Daalder, The Role of the Military in the Emerging Countries (The Hague, 1962);
Marion J. Levy, Modernization and the Structure of Societies (Princeton, 1966), pp. 571-605.
38. For examples of this view, see: Bienen, ed. The Military Intervenes (New York, 1968); and
Bienen, ed. The Military and Modernization; Nordlinger, “Soldiers in Mufti”; and Edwin Lieu wen,
Generals v. Presidents (London, 1964).
39. See, for example, M. R. Doornbas, “Political Development: Search for Criteria,” Development and
Change, I (January 1969); Chong-Do Hah and Jeanne Schneider, “A Critique of Current Studies of
Political Development and Modernization,” Social Research, XXXV (Spring 1968); Samuel P.
Huntington, “Political Development and Political Decay,” World Politics, XVII (April 1965); J. P.
Nettl and Roland Robertson, “Industrialization, Modernization and Development,” British Journal of
Sociology, XVII (March 1966); Karl de Schweinitz, “Growth, Development and Political
Modernization,” World Politics, XXII (July 1970).
40. Manfred Kossok, “Changes in the Political and Social Functions of the Armed Forces in the
Developing Countries: The Case of Latin America,” and Philippe C. Schmitter, “Military Intervention,
Political Competitiveness and Public Policy in Latin America: 1950-1967,” in Morris Janowitz and
Jacques van Doom, eds. On Military Intervention (Rotterdam, 1971); E. Be’eri, The Army Officers in
Arab Politics and Society (London, 1970); Jacob C. Hurewitz, Middle East Politics: The Military
Dimension (London, 1969); Robert Pinkney, “The Theory and Practice of Military Government,”
Political Studies, XXI (February 1973); James A. Bill, “The Military and Modernization in the Middle
East,” Comparative Politics, II (October 1969); Edward Feit, “Pen, Sword, and People,” World
Politics, XXV (January 1973).
* This article is the second report on a project funded by the Social Science Research Council (U.K.).
The authors would like to thank Anthony Mughan, who was the research assistant.
Introduction
NOTES
1. F. LaMond Tullis, Politics and Social Change in the Third World (New York: John Wiley, 1973),
p. 79. Emphasis added.
2. Cited by Dankwart Rustow, A World of Nations (Washington: The Brookings Institution, 1967), p.
135.
3. Dankwart Rustow, “Introduction to ‘Philosophers and Kings: Studies in Leadership,’ “ Daedalus,
Vol. XVCII, No. 3 (Summer 1968), 689.
4. See the discussion in H. H. Gerth and C. Wright Mills (eds.), From Max Weber: Essays in
Sociology (New York: Oxford University Press, 1968), pp. 245-264, 295-299. Also note Peter Blau’s
“Critical Remarks on Weber’s Theory of Authority” and Reinhard Bendix’s “Reflections on
Charismatic Leadership” both reprinted in Denis Wrong (ed.), Max Weber (Englewood Cliffs:
Prentice-Hall, 1970).
5. Gerth and Mills (eds.), From Max Weber, p. 296.
6. Weber cited by Rustow, A World of Nations, p. 150.
7. Richard Wright, White Man Listen! (Garden City: Doubleday, 1964), p. 65.
8. W. Howard Wriggins, The Ruler’s Imperative (New York: Columbia University Press, 1969), p. 6.
9. See Okot p’Bitek, “Song of Prisoner,” in p’Bitek, Two Songs (Nairobi: East Africa Publishing
House, 1971) and Norman Provizer, “The Other Face of Protest: The Prisoner and the Politician in
Contemporary Africa,” Journal of African Studies, Vol. II, No. 3 (Fall 1975). Compare with Ann
Ruth Wiilner and Dorothy Wiilner, “The Rise and Role of Charismatic Leaders,” The Annals of the
American Academy of Political and Social Science, Vol. CCCLVM (March 1965).
10. Rupert Emerson, “The Fate of Human Rights in the Third World,” World Politics, Vol. XXVII,
No. 2 (January 1975), 203.
11. See the section on “Development Strategy and Leaders,” in Norman Uphoff and Warren Ilchman
(eds.), The Political Economy of Development (Berkeley: University of California Press, 1972 and
Uphoff and Ilchman’s The Political Economy of Change (Berkeley: University of California Press,
1969) -
12. Charles Anderson, “Comparative Policy Analysis: The Design of Measures.” Comparative
Politics, Vol. IV, No. 1 (October 1971), p. 121.
13. These are the concluding words of Otto Neurath’s, Foundations of the Social Sciences, Vol. II,
No. 1 of Foundations of the Unity of Science: Toward an International Encyclopedia of Unified
Science (Chicago: The University of Chicago Press, 1944).
The Comparative Analysis Of Political
Leadership
Lewis J. Edinger*
What is Leadership?
A fundamental source of confusion for comparative analysis is the many
definitions of leadership. This is certainly not a new problem; but it has been
compounded as political scientists have sought to broaden the conceptual
scope by utilizing definitions employed in other social sciences. “There are
almost as many definitions of leadership as there are leadership theories” in
psychology;5 and in other social sciences implicit and explicit definitions are
perhaps even more numerous than theories. The current literature is replete
with definitions that variously associate leadership with the exercise of
power, influence, command, authority, and control in ways that may suit the
purposes of some scholars, but not of others.6
Furthermore, whether and how such definitions can be applied to
comparative studies of leadership is a matter of widespread argument.
Apparently identical or similar conceptualizations are often translated into
very different operational criteria for empirical research; and seemingly
different ones are operationalized in much the same way.7 Then, again,
political leadership will be defined at a level of abstraction that may suit
formal deductive theories, but defies empirical analysis.
Still and all, there is a certain degree of coherence to this seeming anarchy
which allows us to delineate two major definitions—the positional and the
behavioral. For the purposes of comparative inquiry both of these meet three
conceptual requirements:
1. See Glenn Paige, ed. Political Leadership: Readings for an Emerging Field (New York, 1972).
2. This surge of interest in leadership analysis seems to reflect to some degree the overcoming of a
cultural lag as American political science has become increasingly internationalized, intellectually as
well as substantively. To some extent, too, it appears to be a response to contemporary developments
that have posed new normative issues about the ability of liberal democratic systems to produce
political leadership “adequate” to the needs of advanced industrial societies. Previously, American
scholars, nurtured on writings that put little emphasis on political leadership—such as the works of
John Locke and the Federalist Papers—tended to slight its analysis in favor of the exploration of
collective patterns of mass attitudes and behavior. But of late the realization that not just the
recruitment, but the actions of governmental decision-makers can be critical determinants of political
developments, has evidently led participant observers in the social sciences to concern themselves
more extensively with all the ramifications of leadership patterns.
3. In 1960 Eugene Jennings observed that “the attempt to study leadership scientifically has not
provided a widely accepted body of knowledge of what leadership is and does” (An Anatomy of
Leadership [New York, 1960], p. Iff). Twelve years later, the eminent editors of a popular comparative
text wrote: “When does leadership emerge? How, and under what conditions, can it manage to
restructure opinions? What techniques does it use to gain support without strongly antagonizing or
alienating powerful interests? Frankly we do not know.” Roy C. Macridis and Robert E. Ward,
“Introduction,” Comparative Political Systems (Englewood Cliffs, 1972), p. 24.
4. The very large number of studies on which this essay is based makes it inadvisable to include
extensive footnote citations. The interested reader is referred to the following bibliographies: Lewis J.
Edinger and Donald D. Searing, “Leadership: An Interdisciplinary Bibliography,” in Lewis J. Edinger,
ed. Political Leadership in Industrial Societies (New York, 1967), pp. 348-66; George T. Force and
Jack R. Van Der Silk, eds. Theory and Research in the Study of Political Leadership: An Annotated
Bibliography (Carbondale, 1969).
5. Fred Edward Fiedler, Leadership (New York, 1971), p. 1.
6. Cf. Edinger, Political Leadership, pp. 6-8.
7. Cf. Edinger, ed. Theory and Method in Comparative Elite Analysis (New York, 1969).
8. John Kenneth Galbraith writes, for instance, “On any form of organized activity—a church,
platoon, government bureau, congressional committee, house of casual pleasure—our first instinct is
to inquire who is in charge. Then the inquiry is to the qualifications or credentials which accord such
command. Organizations almost invariably invite two questions: Who is the head? How did he get
there?” (The New Industrial State [London, 1967J, p. 47). By way of contrast, consider the
proclamation of a short-lived, acephalous organization by students at Columbia University in 1968.
“Those who approach such organizations with the request, ‘Take me to your leader,’ are likely to be
met with the response ‘We’re all leader.’” (“What Leader?”, Student Voice [1 October 1967], p. 7.)
9. For some recent examples of this approach, see James MacGregor Burns, “Toward the
Conceptualization of Political Leadership” (Paper presented at 69th Annual Meeting, American
Political Science Association, New Orleans, September 1973); Léon Dion, “The Concept of Political
Leadership,” Canadian Journal of Political Science, I (March 1968), 1-17: Fiedler, Leadership (New
York, 1971); Cecil A. Gibb, “Leadership,” in Gardner Lindzey and Elliot Aronson, eds. Handbook of
Social Psychology, 2nd ed. (Cambridge [Mass.], 1969), vol. 4, pp. 206-81; Andrew S. McFarland,
Power and Leadership in Pluralist Systems (Stanford, 1969), pp. 153-76.
10. Leadership is usually associated with changes in followership behavior. But logically it can just as
well apply to successful efforts to prevent such changes in the presence of countervailing pressures.
11. Leadership strategy may be directed toward limiting the alternatives open to the objects of control
and toward making it appear that the benefits of compliance are greater than the costs. But when a
person is given no choice whatever, say over whether he wants to live or die, control over his actions
by another cannot be demonstrated. In this respect, variance and nonvariance in the behavior of the
targets of leadership control operate along a continuum between the poles of unlimited choice and
absolutely no choice.
12. Other factors may serve to explain how Hitler came to achieve control over the victims and their
immediate executioners. But when we ask whether these people would have died there and then had
he not ordered “The Final Solution,” the answer appears pretty conclusive in the light of the evidence.
13. Cf. McFarland, pp. 153-76, for the argument that in leadership analysis speculating about the “ifs”
of past events is the methodological equivalent of conducting controlled laboratory experiments on
leadership patterns in small groups. Note, however, Arend Lijphart’s persuasive caveat that “the
comparative method is not the equivalent of the experimental method but only a very imperfect
substitute,” in “Comparative Politics and the Comparative Method,” American Political Science
Review, LXV (September 1971), 685.
14. In this case leadership serves as a positive catalyst. Contrariwise, if it decelerates behavioral
changes induced by other factors, it can be said to have the effect of a negative catalyst or
preservative.
15. Confirming leadership may, for instance, take the form of instrumental actions that facilitate
established procedures, or it may legitimate and sanction old or new behavioral norms by symbolic
actions. The more closely it comes to approximate the performance of purely routine measures, or
what has been called “automatic leadership,” the more difficult it will be to differentiate confirming
leadership from actions that do not involve control over others. Here one might note the gradual
transformation of royal assent to legislative decisions from a means of control to a symbolic gesture of
legitimation, or the executive acts of the American president in what Richard E. Neustadt describes as
his role of a clerk. Cf. Neustadt, Presidential Power (New York, 1969), p. 19.
16. The latter approach characterizes most current investigations of leadership in formal organizations
and small groups. Thus, social-psychological studies of such collectivities usually evaluate and
compare the quality of leaders in terms of their ability to promote group harmony and effective task
performance. See, for example, Daniel Katz and Robert L. Kahn, The Social Psychology of
Organizations (New York, 1967).
17. Cf. Edinger, “Where are the Political Superstars?”, Political Science Quarterly, LXXXIX (June
1974), 249-68.
*This article owes a great deal to stimulating discussions with the author’s students at Michigan State,
Washington, and Columbia Universities. The author is indebted for financial assistance in support of
research to the John Simon Guggenheim Foundation, the National Science Foundation (Grant NSF-
GS-28086), and the Ford Foundation (through funds granted to the Institute on Western Europe at
Columbia University).
Leaders and Structures In Third
World’ Politics
Contrasting Approaches to Legitimacy
Paul R. Dettman
Introduction
The tendency of contemporary political scientists to neglect the significant
role which leadership plays in political life because of their preoccupation
with political structure has been deplored by Dankwart A. Rustow.1
Structurally oriented political analysis either ignores the leadership variable
or treats it as a variable that sets the limits within which political events can
be explained in terms of political structures. In political contexts in which
structural variables are overwhelmingly salient, the “error” introduced by
the leadership variable is certainly not large. There are, however, political
contexts in which structural analysis tells only part of the story and the
variable of leadership can be omitted only at the cost of a significant gap in
explanation. Such contexts are often those where political structures are
undergoing rapid and substantial change with particular leaders playing a
significant role in bringing this about. The politics of the “Third World” has
been characterized by this type of political context. Consequently,
explanations of “Third World” politics cannot be limited to structural
analyses but must give due consideration to the impact of the leadership
variable.
The student of Third World politics is faced, then, with the problem of
bringing the leadership variable into the picture. He can, of course, treat it
as a variable unrelated to structural variables—the method followed by W.
Howard Wriggins.2 The latter analyzes Third World political leadership in
terms of the strategies used to aggregate power and to build supporting
coalitions, e.g., projecting personality, building an organization, formulating
an ideology, providing patronage, suppressing opponents, fostering
economic development, encouraging or discouraging popular participation,
and pursuing an active foreign policy. In the end, however, Wriggins’
analysis tells nothing significant about the role of leadership in Third World
politics because the strategies he delineates are the same as those used by
politicians throughout the world. The reason for Wriggins’ failure to
discover anything peculiar in Third World political leadership is clearly that
he consciously deprecates the importance of the structural context in which
Third World political leaders function. ‘‘Politics,” he asserts, “is not a
matter of structure and process, neatly defined, but rather a drama of human
beings, responding to and attempting to affect the behavior of one
another/’3 Viewing politics as essentially a human drama abstracted from
the structural context in which it takes place, Wriggins inevitably arrives at
a universal picture of political strategies which fails to specify the
peculiarities of political leadership in the “Third” part of the world.
The meager theoretical results of Wriggins’ study make it clear that any
attempt to delineate the peculiar role that leadership plays in Third World
politics must give due weight to its structural dimensions. Such an attempt
cannot focus, as Wriggins does, on leadership variables alone; it must relate
these to structural variables. It is easy to say this but extremely difficult to
accomplish it. One way to begin would be to examine the relationship
between political leadership and socioeconomic and political structures in
several Third World countries in order to ascertain what, if any, cross-
cultural continuities emerge. The study which follows takes such a
comparative approach. This imposes certain limitations upon its scope,
however. In order to compensate for the difficulties involved in considering
several Third World countries within a single article, one must limit oneself
to a single aspect of leadership instead of taking up Wriggins’ complete
catalog. The single aspect that seems to be most important, legitimacy, has
been chosen on the understanding that, even if legitimacy stands
preeminent, Wriggins’ study has made clear that it is only one aspect of
leadership and that much more needs to be said on the subject before the
relationship between leadership and structures in Third World politics is
fully understood. In other words, what follows is only the beginning of a
long theoretical discussion.
The comparative approach calls for a common theoretical framework by
means of which the relationship between the legitimacy aspect of leadership
and the socioeconomic and political structures of several Third World
countries may be analyzed. So far as legitimacy is concerned, it would
appear that one can do no better than to follow Max Weber’s typological
distinctions between “charismatic,” “traditional,” and “legal-rational”
norms.4 The best analytical framework for the socioeconomic and political
structures seems to be Fred Riggs’ “model of the prismatic society,”5
although this model does not perfectly fit the structures of any of the Third
World countries which will be studied. Rather, his model will serve as a
yardstick to measure the degree to which these structures conform to or
deviate from those which he regards as characteristic of Third World
nations.
By combining the theoretical frameworks of Weber and Riggs, then, one
can attempt to relate comparatively the legitimacy aspect of leadership with
structural variables in several Third World political communities. These
communities can be chosen on the basis of variations in leadership
performance, including some in which civilian leadership has been
perpetuated and some in which civilian leadership has given way to military
rule. India, Tanzania, and Kenya are examples of the former, Indonesia,
Burma, and Ghana of the latter. Moreover, each of these political
communities began as an independent nation under a preeminent nationalist
leader: Nehru in India, Nyerere in Tanzania, Kenyatta in Kenya, Sukarno in
Indonesia, U Nu in Burma, and Nkrumah in Ghana. The interaction
between the approaches to legitimacy of these leaders and the structures
within which they led can be studied comparatively in order to ascertain
whether or not there is a relationship between this interaction and their
success or failure in maintaining authority.
Burma and U Nu
Burmese structures deviate from the prismatic model in only one respect,
but that deviation is in a “fused” rather than a diffracted direction.10 The
social structure is polycommunal, with a dominant Burman majority. The
non-Burmese minority is fragmented into a number of small tribal and
ethnic groups that are in conflict with each other as well as with the Burman
majority. The result of Burman domination is that society and community
correspond more closely than is typical of the prismatic case. It is important
to note, however, that this deviation is in the “fused” rather than the
diffracted direction because it results from the domination of a single
communal group rather than from the unification of a number of different
communal groups characteristic of the diffracted society.
U Nu exercised leadership within structures that are prismatic, except for
this one fused element. His legitimacy was based upon a mixture of
charismatic, traditional, and legal-rational norms, in that order. The
Burmese people looked upon him as the one man whom they could trust
and under whom they could unite. His personal integrity and selflessness
were conspicuous, as was his “other-worldliness.” He was a devout
Buddhist, who not only lived by the moral precepts of his religion but spent
long hours in prayer and meditation. His followers even considered that he
was on his way to becoming an “enlightened one,” a Buddha. As such he
exercised supernatural power in their eyes.
His charismatic appeal was based upon his qualities as a man of religion
more than upon his qualities as a politician. It was not he but Aung Saw
who had led Burma to independence; the latter was assassinated shortly
after independence was achieved. Nor had U Nu, like Aung Saw, suffered
so that the nation might be free. He was chosen to succeed Aung Saw as
head of the newly independent Burma mainly because his personal integrity
and selflessness were such that all the surviving nationalist leaders could
agree to serve under him. Nevertheless, U Nu’s legitimacy did derive
significant charismatic support from his role as a politician, particularly
because of his oratorical style and skill. He spoke to the people in their own
language and gave voice not only to their aspirations but also to their
grievances. He promised them a life of prosperity in which every Burman
would have a house, an automobile, and a monthly income of two hundred
dollars. He castigated the corruption of Burmese life and called for a return
to Buddhist purity. Finally, although U Nu had not worked a political
miracle in leading Burma to independence, he had achieved a miracle by
preserving Burmese unity when it appeared that independence would be
followed by the nation’s disintegration under the pressure of tribal and
Communist insurrections.
Traditional norms were next in importance as a legitimizing force for U
Nu’s authority. Because of the dominant position of the Burman majority,
the Buddhist religion and culture provided the lifeblood of Burmese society.
For this reason, Burmese nationalism had a religious as well as a political
dimension. Its political side provided the impetus for pushing toward the
conversion of Burma into a modern nation state. Its religious aspect,
however, pushed in the opposite or traditional direction—namely, recreating
the past glories of Buddhist society. U Nu’s personal predilections as a
devout Buddhist and his political position both led him to emphasize the
traditional religious rather than the modern political dimension. In the
hearts of the Burmese people, Aung Saw was the political “father of the
nation.” If U Nu aspired also to be a “father of the nation,” he must be a
religious father. Consequently, while he sought to achieve the political goal
of making Burma a prosperous modern nation, he tried to attain this goal
mainly by traditional religious means, especially by making Buddhism the
state religion in order to bring the nation back to the Buddhist ideals of
purity and selflessness.
Legal-rational norms were of only minor importance as a source of
legitimacy for U Nu’s authority. He did not play a major role in the
organization of the Anti-Fascist People’s Freedom League (A.F.P.F.L.), the
vanguard of Burma’s independence movement. And his insistence that the
disunity that plagued the League after independence was the result of
personal corruption among its members led him to undertake a party
“purge” which split the League into two factions.
Just as U Nu failed to capitalize on the legitimizing support of party
norms, so, too, he failed to capitalize on the legitimizing support of
bureaucracy. He politicized the civil bureaucracy by appointing only those
considered politically reliable and by making administrators subservient to
politicians. The result was to carry over into the civil bureaucracy the
factionalism and corruption which characterized the A.F.P.F.L. U Nu
attempted to cope with this development in a characteristic manner—by
purging the civil bureaucracy of corruption. Once again the result was
negative, for this anticorruption campaign only served to undermine further
the morale and efficiency of the civil bureaucracy and to diminish U Nu’s
legitimacy in the eyes of an important element of Burma’s political elite.
The Burmese leader’s dealings with the military bureaucracy were at best
ambivalent, at worst antagonistic. His personal predilection as a Buddhist
naturally placed him in opposition to the violence and authoritarianism
which provided the rationale of the military system. On the other hand,
Burma’s most serious problem was the threat to national unity posed by
tribal and Communist insurrections. An increase in the military bureaucracy
was absolutely essential if this threat were to be met. Consequently, during
the first years of his regime, U Nu was forced to support the military
bureaucracy. As a result of this joint political-military effort, the
government established its control over most of Burma, and the immediate
threat to the nation’s territorial unity was successfully countered. Following
the improvement in the internal security situation, however, U Nu’s policy
vis-à-vis the military became one of reluctant toleration coupled with
outright mistrust. Instead of relying upon military force to keep the
insurrectionists under control, he sought to win themover by conciliation
and compromise.
After two years of a military caretaker regime, U Nu returned to power in
1960 with a strong popular mandate, which gave him the courage to put the
army in a subservient place. He took over the defense portfolio, ousted
army officers from their acquired positions of political and economic
power, and removed the police from army jurisdiction. These actions
convinced the army leaders that U Nu had embarked upon a program which
threatened not only the interests of the military bureaucracy but of the
nation at large; consequently, they could no longer regard him as Burma’s
legitimate political leader. Their reaction was a military coup which not
only deposed U Nu, but placed him under detention so that he would be
unable to carry on further political activity.
Since he failed to capitalize upon the legal-rational norms that provided
the rationale of both party and bureaucracy, it is not surprising that U Nu’s
legitimacy was challenged by opposition from within the ranks of both
organizations. It is surprising, however, that this challenge met with such
easy success. Having recently won an overwhelming victory at the polls, U
Nu could theoretically have anticipated popular support against the military,
especially when the leaders of the coup went so far as to place him under
house arrest. This did not happen primarily because by 1960 U Nu’s
legitimacy had also lost the support of charismatic norms. Earlier he had
validated the faith of his followers by successfully countering the threat of
national disintegration. Unfortunately, as the decade of the 1950s drew to a
close his performance deteriorated; and, as the 1960s dawned, Burma stood
on the brink of economic and political chaos. Not only was the internal
security problem still serious, but the economy had reached a point of
almost complete stagnation. As a result, U Nu’s followers suffered both
political and economic deprivation from his leadership. They had looked
upon him as a leader who could bring about the economic and political
miracles he had promised them. When he did not fulfill his promises, his
followers lost faith in him, and his charismatic appeal evaporated.
On balance, U Nu’s legitimacy was based primarily on charismatic
norms, secondarily on traditional norms, and only slightly on legal-rational
norms. The relative degree of legitimacy derived from these three
categories of normative values was not congruent with the social, political,
economic, and normative structures within which he attempted to exercise
authority. To the extent that those structures contained traditional elements,
they sustained the legitimacy he derived from tradition. On the other hand,
to the extent that they contained modern elements, they undermined the
legitimacy based upon traditional norms. Nor was this undermining effect
offset by the slight degree to which U Nu’s legitimacy gained strength from
the legal-rational norms of Burmese society. In short, given the traditional-
modern mix of Burma’s prismatic society, it is quite clear that his
legitimacy was based too strongly on traditional, and not enough on legal-
rational norms.
To make matters worse for U Nu’s legitimacy, his primary reliance on
charismatic norms was not validated by the limited benefits conferred on
his followers. While he succeeded in averting national disintegration, he did
not achieve national unity. On the economic front, the results of his
leadership were negative. Having benefited to only a limited degree
politically and suffered economically, U Nu’s followers lost their initial
faith in him as a leader gifted with supernatural powers. Thus, his
legitimacy lost the support of charismatic norms. The army easily removed
him from office, and the Burmese people turned a deaf ear to his protests
that he was still Burma’s prime minister even though under house arrest.
Conclusions
A division of the six Third World political leaders discussed in this article
according to their abilities to maintain their regimes reveals that those who
fall into the successful category—Nehru, Kenyatta, and Nyerere—founded
their legitimacy upon a mixture of norms congruent with the structures
within which they led. None of these successful leaders placed primary
reliance upon charismatic norms. As a result, the limited benefits which
they were able to confer upon their followers measured up to the limited
extent to which their legitimacy depended upon their charismatic appeal.
On the other hand, the unsuccessful leaders—Nkrumah, U Nu, and Sukarno
—based their legitimacy upon a mixture of legitimizing norms incon-gruent
with their leadership structures. The legitimacy of each of these
unsuccessful leaders was based mainly on charismatic norms. When the
benefits they conferred upon their followers did not come up to the
charismatic promises they had made, the people’s faith in their
“miraculous” powers disappeared, and, along with it, the principal
foundations of their legitimacy.
It is not surpising that these opposite results should have obtained. Since
the end of the era in which kings ruled by “divine right,” the legitimacy of a
political leader has not resided in his person, nor even in his office.
Whereas a political office carries with it a corresponding authority, the
incumbent must be regarded by those over whom he is set as a legitimate
holder of that office, or he cannot effectively exercise that authority. The
paradox that the political leader is both the master and slave of his people
must not be ignored. He is their slave in the sense that gaining a position of
legitimacy among them is a prerequisite for his functioning as their master.
Whether or not the people of a Third World country confer legitimacy upon
their political leader depends in part upon their personal attitude toward
him; that is, upon whether or not he enjoys the support of charismatic
norms.12 This depends also upon the degree to which he conforms to the
norms of the structures that make up the society, for these structural norms
provide both guidance and standards of judgment for the conduct of its
members. If the political leader acts in a way which honors these structural
norms, he is legitimate in the eyes of his people. If he does not, he becomes
an illegitimate officeholder and rightly subject to deposition.
Max Weber was careful to emphasize the instability of the charismatic
norm that derives from a personal rather than a structural source.13 This
must be constantly renewed through personal contact and validated by the
benefits that the leader confers upon his followers. The accomplishment of
this two-fold task is particularly difficult for a Third World political leader.
During the struggle for independence from colonial rule he can tour the
country to rally his people to the cause; in the hectic days that follow
independence, he can make promises of a glorious future for the newly
independent nation. But when it becomes necessary at a later stage for him
to head a governmental apparatus and to spur that apparatus to implement
programs of social and economic development which will benefit his
people, he is forced to neglect the charismatic task of maintaining personal
contact with them. Moreover, the slowness which plagues progress toward
social and economic development in a Third World country—if indeed
there is progress at all—makes it virtually impossible for a leader to satisfy
the high expectations of those who followed him because he gave them a
grandiose vision of their nation’s future.
The six-nation comparative study described in this article provides no
settled conclusions on the theoretical question of the relationship between
legitimacy of political leadership and the societal structures of Third World
countries. It does, however, supply a measure of empirical evidence from
the Third World to support Rustow’s contentions (1) that politics in any of
the three worlds cannot be understood unless the relationship between
leadership and structures is taken into account and (2) that the past tendency
of theoreticians in the field of comparative politics to neglect this
relationship has hindered progress in arriving at such an understanding.
1. Dankwart A. Rustow, “The Study of Leadership,” in Rustow, ed. Philosophers and Kings (New
York, 1970).
2. W. Howard Wriggins, The Ruler’s Imperative: Strategies for Survival in Asia and Africa (New
York, 1970).
3. Ibid., p. 11.
4. Talcott Parsons, “Introduction’,’ in Max Weber, The Theory of Social and Economic Organization,
trans. Parsons and A. M. Henderson (New York, 1947).
5. Fred W. Riggs, Administration in Developing Countries: The Model of the Prismatic Society
(Boston, 1964). Riggs’ “model” is worked out in considerable detail and is replete with esoteric
vocabulary. It is founded, however, on the argument that the “prismatic” society is characterized by
several structures, each of which performs several functions. In this respect it is distinct from a
“fused” society in which one structure performs all functions and, at the opposite extreme, the
“diffracted” society in which there are many structures but each structure performs only one function.
He agrees with Lloyd I. and Suzanne H. Rudolph that the relationship between traditional and
modern in Third World countries is dialectical rather than dichotomous. See their, The Modernity of
Tradition (Chicago, 1967), pp. 3-14.
6. The most useful secondary sources on Nehru’s leadership in India are: Michael Brecher, Nehru: A
Political Biography (London, 1957); W. R. Crocker, Nehru: A Contemporary’s Estimate (New York,
1966); Donald E. Smith, India as a Secular State ( Princeton, 1965); and Hugh Tinker, India and
Pakistan (New York, 1967).
7. For Kenyatta’s leadership in Kenya, see: N. S. Carey Jones. The Anatomy of Uhuru (Manchester,
1966); George Bennett and Carl Rosberg, The Kenyatta Election (London, 1961); Cherry Gertzel,
The Politics of Independent Kenya (London, 1970); and Carl Rosberg and John Nottingham, The
Myth of Mau Mau: Nationalism in Kenya (New York, 1966).
8. Nyerere’s leadership in Tanzania is best evaluated in: Harvey Glickman “One Party System in
Tanganyika,” Annals of the American Academy of Political and Social Science, 358 (March 1965); J.
Clagett Taylor, The Political Development of Tanganyika (Stanford, 1963); Henry Bienen, Tanzania:
Party Transformation and Economic Development (Princeton, 1967); and William Tordoff,
Government and Politics in Tanzania (Nairobi, 1967).
9. David E. Apter has specialized in studies of Nkrumah’s leadership. See his Ghana in Transition
(New York, 1963), and “Nkrumah, Charisma and the Coup,” Daedalus, 97 (Summer 1968). Henry
Bretton, The Rise and Fall of Kwame Nkrumah (New York, 1966), constitutes a rebuttal of Apter at
several points. Valuable historical background is provided in Dennis Austin, Politics in Ghana: 1946-
1960 (London, 1964), and “Opposition in Ghana,” Government and Opposition, II (July-October
1967).
10. Richard Butwell has written an excellent biography, V Nu of Burma (Stanford, 1963). Valuable
additional interpretations are: Donald E. Smith, Religion and Politics in Burma (Princeton, 1965);
Joseph Silverstein, “Burma,” in George H. Kahim, ed. Governments and Politics in Southeast Asia
(Ithaca, 1964); and Hugh Tinker, The Union of Burma (London, 1967).
11. Sukarno’s leadership is best analyzed in Peter Leyon, War and Peace in Southeast Asia (London,
1969); Herbert Feith, The Decline of Constitutional Democracy in Indonesia (Ithaca, 1962); J. D.
Legge, Indonesia (Englewood Cliffs, 1964); and Bernard Dahm, Sukarno and the Struggle for
Indonesian Independence (Ithaca, 1969).
12. For a useful study of the particular importance of charismatic norms in legitimizing Third World
political leaders, see Ann Ruth and Dorothy Willner, “The Rise and Rule of Charismatic Leaders,”
Annals of the American Academy of Political and Social Sciences, 358 (March 1965).
13. Weber, “The Sociology of Charismatic Authority,” in H. H. Gerth and C. Wright Mills, eds. From
Max Weber: Essays in Sociology (New York, 1946).
Revolutionary and Managerial Elites
in Modernizing Regimes
John H. Kautsky*
I Introduction
Among the underdeveloped countries that have experienced a modernizing
revolution, the Soviet Union and Mexico are the most outstanding cases of
successful economic development. The political concomitants of that
development have differed widely in the two countries; but in each, the
revolutionary leadership that had overthrown the old regime and established
a new, modernizing one has in turn, as industrialization has proceeded, been
replaced by a new elite of bureaucrats and technocrats. It is not surprising,
therefore, that the hypothesis has been more or less explicitly advanced that
in underdeveloped countries generally there is some link between
industrialization, on the one hand, and, on the other hand, the replacement of
the leadership of revolutionary modernizers by that of managerial
modernizers.1 Depending on which of the two variables one regards as the
independent and which the dependent one, one may suggest either that if
industrialization is to proceed, the replacement of elites must take place or
that if and when industrialization does proceed, the replacement of elites will
take place.
In favor of the first suggestion one can argue that it is possible, after all,
that, apart from the obvious economic and cultural obstacles to successful,
rapid industrialization, the leadership of revolutionary modernizers,
committed though they are to the goal of industrialization, may itself stand
in the way of reaching that goal. Lacking the requisite skills and attitudes,
the revolutionary modernizers may have to be replaced if the goal is to be
attained. And since they resist being replaced, they adopt attitudes and
pursue policies which in fact serve the function of justifying their
continuance in power and which may slow or prevent industrialization. In
favor of the second suggestion, one can argue that an industrial economy not
only requires leadership by people with administrative, managerial, and
technical skills but also itself produces such leadership.
The hypothesis that the industrialization of underdeveloped countries and
the replacement of their revolutionary modernizing leadership by managerial
modernizers are linked seems to be in accord with common sense,2 but
sufficient evidence is not yet available to test this hypothesis, let alone to
confirm it. In only four major countries—Mexico, China, Russia, and
Turkey—did revolutionary modernizers come to power as much as a
generation ago. One can hardly expect that many of the other countries that
have experienced such a revolution much more recently would by now have
progressed very far on the road to industrialization and, hence, to managerial
and administrative rule. It is certainly striking that, in the only two of the
four countries mentioned that have in fact undergone substantial
industrialization, managerial modernizers have come to constitute a
powerful—perhaps even the most powerful—elite group. Brezhnev and
Kosygin in the Soviet Union and the successors to Mexico’s last
revolutionary leader, Cárdenas, particularly Avila Camacho, Alemán, and
Ruiz Cortines, may be regarded as representing managerial modernizers.
In China, too, industrialization has been progressing, and there, too,
managerial modernizers seem to be rising and competing for power. We may
also add that in Israel and in the Eastern European countries the transition
from revolutionary to managerial modernizers seems to have been in
progress for the past few years. In these countries, revolutionary modernizers
came to power only about two decades ago, but much of the population and
of the economy were, at the time of the revolution, no longer caught up in
the traditional order, as they are in most underdeveloped countries.
The general hypothesis that, concomitant with industrialization, an elite of
managerial modernizers rises to power does find support, then, in the
experience of the few countries that have made substantial progress toward
industrialization following a revolution of revolutionary modernizers. (This
process of political change is quite different from that which accompanied
the industrialization of Western countries and of Japan, with which we are
not concerned here). However, evidence derived from the experience of
these countries is too limited to be conclusive, and the hypothesis must
remain tentative for thé present. At best, it will be possible to establish its
validity only if and when, perhaps in a few decades, some of the many
underdeveloped countries that have in the past few years come under the
control of revolutionary modernizers make substantial progress toward
industrialization.
In the meantime, it is possible, of course, to look for bits and pieces of
relevant evidence to see to what extent processes that occurred in various
ways in the Soviet Union and Eastern Europe, in Mexico and Israel, may be
repeated in other underdeveloped countries. We furnish some such evidence
in the following pages, though our purpose here is not to present newly
discovered data but merely to organize some scattered well-known ones so
that they might provide us with some insights into, though not with
predictions of, the evolution of postrevolutionary societies in
underdeveloped countries. However, even such inconclusive attempts to test
the hypothesis of the replacement of revolutionary by managerial
modernizers are premature. For, clearly, before we can hope to test it, we
must be able to distinguish between the two types of elites. To try to do so is
the principal task of this article.
It is commonly and reasonably assumed that revolutionary modernizers
are distinguishable from managerial ones with respect to (1) their
background, experience, and training, (2) their attitudes, and (3) th^ir
policies. In order to test that assumption, we shall attempt to distinguish
between them in terms of these three categories. Under each, we shall try to
draw the picture of an ideal type of revolutionary modernizer and an ideal
type of managerial modernizer. Though we shall illustrate them with
selective evidence drawn from the real world, these ideal types as such
cannot be expected to be found in pure form in the real world. Rather, they
are meant merely to provide us with the analytical categories of
revolutionary and managerial modernizers that can help us organize the data
of the real world and determine to what extent real leaders fit into one
category or the other.
Without anticipating too much of what follows, it may be stated here that
the most significant result of our efforts lies not in their success but in
calling attention to the difficulties they encounter. Difficulties in “opera-
tionalizing” the distinction between the two types of modernizers, i.e., in
finding data which will permit us to place real leaders into one or the other
of the two categories, cast doubt on the validity of the distinction itself. And
this, in turn, renders questionable the possibility of testing—and hence of
ever confirming—a hypothesis which at the outset seems quite persuasive.
3 ( 9%) were trained in modern professions, but evidently did not practice
them, as follows:
III Conclusion
It would appear, then, that we cannot distinguish sharply between
revolutionary and managerial modernizers in terms of their experience and
training, their attitudes, or even their policies. This is true because there are
not, or not yet, enough regimes that could reasonably be regarded as being
controlled by managerial modernizers to provide us with sufficient data to
support some distinguishing criteria. It may also be the case, however, that
our two concepts, and particularly that of the managerial modernizer, are not
very useful. We have, of course, regarded them all along only as representing
ideal types, and we have assumed that, in the real world, modernizers are
ranged along some kind of continuum between the pure revolutionary and
pure managerial types. It may turn out, however, that, especially after
revolutionary modernizers have come to power in underdeveloped countries,
different kinds of leaders will evolve who are not just more or less
revolutionary and more or less managerial, but who simply cannot be
classified along these lines.
Thus, it appears that in some countries, such as India, where politicians
have to compete with each other for popular support, some of the emerging
leadership, particularly at the local level, is closer to the traditional culture
and hence perhaps less modernizing than the revolutionary leadership it
succeeds. This, of course, need not prevent the concurrent emergence of
managerial modernizers, particularly in those sections of government and
society that are directly concerned with the development of the economy, but
also in any others, such as health care and the military, that make extensive
use of modern techniques. Quite possibly, the expectation that managerial
modernizers will succeed the revolutionary ones in the course of economic
development will turn out to be too simple, and what will in fact occur will
be a diversification of elites. Managerial modernizers will be but one of the
successor elites to the revolutionary modernizers, and managerial
modernizers themselves will be subdivided from the beginning —and will
later become more so—among various types of administrative, scientific,
technical, and engineering personnel.
As we stated at the outset, we cannot test, much less validate, the
hypothesis that industrialization and the replacement of revolutionary by
managerial modernizers accompany each other if we cannot clearly
distinguish between the two types of modernizers. Since this is precisely our
position now, the hypothesis will have to remain no more than a hypothesis
for some time to come.
It is also possible, however, that a hypothesis relating industrialization to
the replacement of revolutionary by managerial modernizers cannot be
tested at all, because the two variables—industrialization and elite
replacement—are not clearly distinguishable. Whether there really are or are
not two distinct sets of attitudes that characterize revolutionary modernizers
and managerial modernizers is, as we have noted, difficult to determine and
by no means clear. It may or may not be true that it takes one type of man to
build a political movement and another to build a factory. But there is, in any
case, reason to believe that it is not so much attitudes and values that shape
policies (as was assumed above for the sake of argument) as it is the
prevailing system characteristics and, especially, the availability of wealth.
Whether revolutionary and managerial modernizers are, or are not,
different types of men we do not, and perhaps cannot, know. But we do
know that there are different types of societies, especially more or less
industrialized ones. Perhaps all we can put forth with certainty is the
commonplace that in industrial societies there are managers—since some
people must manage industry—and that in societies with little or no industry
modernizers cannot be managers. It may be merely the times and the types
of societies that create the distinctions between the two kinds of
modernizers, and it may therefore be hopeless to look for distinctions
between them apart from the times and societies in which they live.
Appendix
Biographical Information on Revolutionary Leaders
1. The present article grew out of an attempt to test this hypothesis. I am grateful to some of my
graduate students who, by trying to do the same thing with reference to a number of particular
countries, have not only provided me with a few of the data I utilize here but have given me a greater
awareness of the conceptual and practical difficulties involved in such an attempt.
2. James S. Coleman also distinguishes between, on the one hand, “the first wave of modern educated
elites,” or “the politically dominant leaders who carried out the revolution (nationalism or
Communist)” and, on the other hand, “the second post-revolutionary generation of technicians and
managers.” But he and some of the literature he cites cast doubt on the hypothesis that the latter will
replace the former. “Introduction to Part III,” James S. Coleman, ed. Education and Political
Development (Princeton, 1965), pp. 358-362.
3. I did this briefly in “Patterns of Elite Succession in the Process of Development,” The Journal of
Politics, XXXI (May 1969).
4. Pakistan News Digest (Karachi), July 1, 1965.
5. “We may conclude then that the government’s intensive concern wtih symbolic activity is a
reflection of intra-elite politics as well as of power maintenance. . . . All such activity which
underscores the doctrine of the unfinished Revolution justifies the retention of power by a larger group
of politician-administrators (including some prominent army officers), who have political
qualifications for the positions they hold but no technical ones.” Herbert Feith, “Indonesia’s Political
Symbols and Their Wielders,” World Politics, XVI (October 1963), 95. Ben Gurion concluded a
recent interview, “No, the state of Israel of which we dreamed has not yet come into being.” Le
Monde, Weekly Selection, April 23,1969.
6. Oginga A. Odinga, Not Yet Uhuru (London, 1967).
7. On the Great Proletarian Cultural Revolution, see A. Doak Barnett, China After Mao (Princeton,
1967); Roderick MacFarquhar, ed. China under Mao: Politics Takes Command (Cambridge, Mass.,
1966); and, for background, James R. Townsend, Political Participation in Communist China
(Berkeley and Los Angeles, 1967). On the conflict in Israel, see Dan Avni-Segre, “Israel, A Society in
Transition,” Worid Politics, XXI (April 1969), 343-365.
8. Feith, 81.
9. Sukarno introduced Mussolini’s “vivere pericoloso” as a national slogan in his Independence Day
address of August 17, 1964. Willard A. Hanna, “The Indonesia Crisis-Mid-1964 Phase,” American
Universities Field Staff Report Service, Southeast Asia Series, XII, No. 7 (Indonesia), 1.
10. Edgar Snow, “Interview with Mao,” The New Republic, February 27, 1965, p. 23.
11. Kwame Nkrumah, Challenge of the Congo (London, 1967), pp. ix, xi.
12. One observer of Indonesian politics describes some of the “themes recurrent in the rhetoric of
guided democracy” as follows: “The nation’s aims cannot be achieved by compromise and calculation,
but only by enthusiasm and faith; the goals themselves expand, become millenial: the revolution will
not be completed until imperialism has been crushed and the just and prosperous society established
over the entire world.” Ruth McVey, “Indonesia,” Survey, No. 54 (January 1965), 115.
13. Both programs are reprinted in Jan F. Triska, ed. Soviet Communism: Programs and Rules (San
Francisco, 1962), pp. 23-153.
14. V. I, Lenin, “Better Fewer, But Better,” Selected Works, IX (New York, 1937), 401.
15. Triska, p. 143.
16. Ibid., p. 130.
17. Ibid., pp. 143-146.
18. Ibid., pp. 71-89.
19. Ibid., p. 50.
20. On the conflict between the two, which seems to be at the heart of present-day politics in China,
see Franz Schurmann, Ideology and Organization in Communist China (Berkeley and Los Angeles,
1966).
21. Chu-yuan Cheng, “Power Struggle in Red China,” Asian Survey, VI (September 1966), 482. More
recently, it has been reported that, in line with Mao’s statement that “the lowly are the most intelligent,
the elite are the most ignorant,” hundreds of thousands of high school and university students are
being sent to do manual labor in factories and, especially, in the countryside and frontier regions, not
for a few years but presumably for life. Peggy Durdin, “The Bitter Tea of Mao’s Red Guards,” The
New York Times Magazine, January 19, 1969, pp. 28-35.
22. Quoted in (no author), Certain Aspects of the Inner Life of the Communist Party of China
(Moscow, n.d.), p. 15.
23. Ibid.
24. Translated from Red Flag (No. 8, 1960) in G. F. Hudson, Richard Lowenthal, and Roderick
MacFarquhar, The Sino-Soviet Dispute (New York, 1961), p. 92.
25. Ibid., p. 93.
26. A similar reliance on will power and manpower, as opposed to planning and expertise, and on the
military and guerrilla warfare as “schools” for economic development is to be found in Castro’s
thought: “The school of war taught us how men can do many things, how they can accomplish many
tasks when they apply themselves in a practical way. This was the school of war, where a small
nucleus of combatants developed into an army without bureaucracy. Without bureaucracy! It went to
war, waged war, and won the war without bureaucracy. . . . And war taught us what man can do when
he dedicates himself to working with enthusiasm, interest, and common sense.” Quoted from Castro’s
speech.of February 20, 1967, to farm-machine workers in Irving Louis Horowitz, “Cuban
Communism,” Trans-action, IV (October 1967), 9. See also Joseph A. Kahl, “The Moral Economy of
a Revolutionary Society,” Trans-action, VI (April 1969), 30-37.
27. Reported in The New York Times, September 14, I960, p. 3.
28. V. I. Lenin, “The State and Revolution,” Selected Works, VII, 48.
29. Ibid., 41-42.
30. In the early 1950’s, the ratio between the net incomes of the senior civil servants and those in the
lowest grades was 1.3 to 1 (as compared to 12 to 1 in the United States). Edwin Samuel, Problems of
Government in the State of Israel (Jerusalem, 1956), pp. 6364. I owe this reference and the one cited
in fn. 37 to Yael Ishai, “Transformation from Revolutionary to Managerial Leadership-Israel: A Case
Study” (unpublished, 1967).
31. Lenin, Selected Works, VII, 48.
32. Ibid., 47.
33. Ibid., 92-93.
34. Quoted from Jenmin Jih Pao in The New York Times, February 8, 1967, p. 5.
35. Leon Trotsky, Terrorism and Communism: A Reply to Karl Kautsky (Ann Arbor, 1961), p. 101.
36. V. I. Lenin, “Our Revolution,” Selected Works, VI, 510-511.
37. Among Zionists in the pre-State period, “there was a general belief that the strong determination
of pioneers could overcome all difficulties predicted by experts and could build a country despite the
warnings of scientific and professional knowledge.” Benjamin Akzin and Yehezkel Dror, Israel : High
Pressure Planning (Syracuse, Ν. Y., 1966), p. 12.
38. G. Glezerman, “Questions of Theory: Society, the Collective and the Individual,” Pravda, October
21, 1966, in Current Digest of the Soviet Press, XVIII (November 23, 1966), 15.
39. I dealt with this process at some length in “Myth, Self-Fulfilling Prophecy, and Symbolic
Reassurance in the East-West Conflict,” Journal of Conflict Resolution, IX (March 1965), 1-17.
40. For a listing of numerous other expensive prestige projects initiated by Sukarno, see Feith,
“Indonesia’s Political Symbols,” 83.
41. See my Communism and the Politics of Development: Persistent Myths and Changing Behavior
(New York, 1968).
42. Ishai, ‘Transformation.” Dayan illustrates the point that managerial leaders are not necessarily
more peaceable than revolutionary ones.
* This article is based on a paper presented to the Symposium on the Evolution of Established One-
Party Systems, sponsored by the Center for International Affairs of Harvard University and the
Institute of International Studies of the University of California at Berkeley and held at Timber Cove,
Jenner, California, April 5-7, 1968. I am indebted to the discussants at this symposium for valuable
suggestions and to my colleague Robert H. Salisbury for a critical reading of an earlier draft.
Political Mainsprings of Economic
Planning in the New Nations
The Modernization Imperative versus Social Mobilization
The quest for the appropriate model or strategy for economic development
in the new nations has created serious disagreements among economists.
The disputed alternatives have included: a gradualist or Big Push approach;
balanced or imbalanced growth; capital accumulation or institutional
change; industry or agriculture; centralized planning or laissez-faire; and
import substitution or export promotion. Notwithstanding the differences in
the debate over the preferred strategy, there is the common although often
unstated assumption that the objective of economic development programs
is the improvement of living standards for the masses. This assumption has
been at times explicitly stated not only as a description of reality, but also as
an exclusive normative goal for the new nations. In the course of a critical
examination of various growth models, economist Hans Singer states that
“this raising of the level of people’s life is both the objective of
development, and also its instrument,” and recommends that “the key
concept must be the improved quality of people’s life.”1
W. B. Reddaway, another economist, comments, “Long-range strategy
must have one overriding objective: to secure a progressive rise in the level
of consumption per head, by methods which will make possible an
indefinite continuance of the process and which also satisfy certain social
and political ideals (e.g., of liberty and reduced inequality). Other economic
objectives—such as increased capital-formation and industrialization or a
viable balance of payments—are essentially dealing with means and not
ends.”2 Similarly, John Kenneth Galbraith speaks favorably of the Popular
Consumption Criterion as the planning goal which “anchors economic
development to the consumption requirements, present and prospective, of
the typical citizen—to the consumption, statistically speaking, of the modal
consumer.” He further states that “the planning goal of the less developed
country ... is the improvement of the well-being of the modal person. This
goal is the counterpart of the poverty of the poor and it is not only central
but total.”3 The same concern for improved standards of living as the goal
for economic planning serves as the main theme of the three-volume survey
by Gunnar Myrdal, demonstrating the utter failure of Asian governments to
assure their populations a better life.4
Some economists realize that the reality does not always coincide with
this desired emphasis on mass welfare due to the existence of the forces of
“economic nationalism.” These forces lead to a stress on economic
autarchy, industrialization, economic planning, concern with the public
sector, and hostility to foreign enterprise. Such policies of “economic
nationalism” are negatively evaluated, however; they are considered
irrational, proceeding essentially from the psychic need of national elites to
imitate the West.5 Supposedly such policies do not serve the cause of mass
welfare; rather they promote the interests of specific classes. According to
these economists, mass welfare would be best served by following the law
of comparative advantage, entailing a concentration on the production of
primary products by underdeveloped countries.
For the most part, political scientists have not paid much attention to
questions of economic development. They have concentrated on political
structure and process, deterred by Max Weber’s advice not to examine state
behavior from the viewpoint of goals or purposes. Where attention has been
focused on economic development, political scientists have also operated
primarily from the perspective of mass welfare as well as from the
viewpoint of the desirability of a democratic regime and of political
stability. Accordingly, in discussions of the economic problems of the
underdeveloped countries, political scientists have emphasized
improvements in agriculture, expansion of consumer goods industry, and
encouragement of free enterprise. David Apter and Charles Andrain
underline Max Mil-likan’s belief that, for the advancement of mass welfare
and democracy,4 ’raising agricultural productivity should take precedence
over rapid industrialization.” They further state:
Those working on the Committee on Comparative Politics of the Social Science Research Council
take a similar position about the need to decentralize the economy and political institutions. For
example, according to Joseph LaPalombara, the role of government in furthering economic
development should be to create a framework for private enterprise, that is, provide law, order,
and security for private entrepreneurs and engage in limited economic planning. . . . Myron
Weiner also comes out for decentralized government and private entrepreneurship. To alleviate the
economic gap between the rich and poor nations, Lucían Pye supports population control and
concentration on agriculture, rather than industrialization.6
Conclusions
Studies in economic and political development have for the most part either
assumed or advocated mass welfare as the appropriate objective of
economic planning. The goal of national independence as a crucial factor in
plans for economic development has been treated with studied neglect vis-
à-vis today’s developing countries. Part of the explanation for this neglect
lies in the tendency among economists to concentrate exclusively on wealth
maximization, while students of comparative politics rigidly bifurcate
national and international affairs and study underdeveloped countries as
closed, self-contained systems, unrelated to an international structure of
power.
An examination of the historical experiences of Japan, the Soviet Union,
China, and India, as well as other major countries in Europe, demonstrates
clearly that the goal of national independence has served as a powerful
impulse—indeed, as an imperative force—in the modernization and
industrialization of these countries, and equally in the form such
modernization and industrialization have taken. It is the same goal that
often imparts a compulsive, frenetic, and at times desperate, character to the
struggle for modernity.
A more adequate understanding of the political development process
requires the incorporation of the international system and the activities of
foreign powers in the study of the concrete problems of the underdeveloped
countries rather than merely including it in abstract political system models.
For the present, the notion of the “modernization imperative” suggests
reconsideration of several aspects of political development theory.
In recent years specialists in political development theory have pointed
out the importance of such “crises of development” in the developing
nations as authority, legitimacy, and integration. These are largely seen as
emanating from internal forces, which in shorthand language are covered by
the term “social mobilization.”41 Insofar as modernization was introduced
under colonial rule, this is a correct interpretation. But it is only a partial
interpretation, for the same crises are precipitated in the developing
countries by the more serious problem of national security. Authority has to
be rationalized, society and polity have to be penetrated and integrated, and
legitimacy has to be engendered in order to cope with the defense challenge
presented by the industrial world. State activity in these directions is likely
to bring on crises in authority, integration, and participation. Thus, it is
misleading to look at “crises of development” as something generated
internally by self-contained political systems.
Similarly, the notion of the modernization imperative leads us to weigh
more critically the usual traditional-modern dichotomies in social science
scholarship. Such societal classifications—Henry Maine’s status versus
contract, Ferdinand Toennies’ Gemeinschaft versus Gesellschaft, Emile
Durkheim’s mechanical solidarity versus organic solidarity, Max Weber’s
traditional, charismatic, and legal-rational forms of domination, Talcott
Parsons’ pattern variables, or Fred Riggs’ fused, prismatic, and diffracted
types of society42—are obviously helpful in understanding some aspects of
social behavior; but these do not aid us in understanding the national goals
of the Third World nations. The only dichotomy that is meaningful for the
underdeveloped countries is the economic one that differentiates
agricultural from industrial societies, or one that distinguishes among
societies on the basis of animate or inanimate energy utilization in the
productive process.43 To modernize, societies consciously seek the
industrial wherewithal and not the characteristics associated with the
modern component of the other classifications.
It becomes obvious also that the only meaningful interpretation of the
much disputed term “political development” or “political modernity”44 is
effectiveness in the pursuit of industrialization, irrespective of the nature of
the political regime and processes involved and regardless of the use of
traditional structures and values to subserve that purpose. In this light, the
definition of political development as simply enhanced capacity or
differentiation only obfuscates matters, for it does not clarify what the
capacity or differentiation is for. Strictly speaking, there is no such thing as
“political development” that is concretely specifiable and isolatable in
political terms; there is only the “politics of development” within the
context of a variety of political regimes, and the crucial question becomes
how successful it is in achieving developmental goals.
Again, since political development here refers to the pursuit of what has
already been achieved elsewhere, it is valid to speak of the division between
“advanced” or “reference” societies and “follower” ones.45 The goals of the
follower societies—say, as included in Dankwart Rustow’s classification,
those of authority, identity, and equality but, equally important, those of
security and industry—have already been determined by history insofar as
modernization is an accomplished fact in the advanced societies. The
creative task for the leadership in the new states is, indeed, not the
determination of national goals—which are dictated by the international
system—but the forging of novel strategies that enable these states to
achieve these goals as rapidly as possible.
Emerging from the compulsion for national independence and political
autonomy, the modernization imperative also dictates that governments
cannot indefinitely abdicate from intervention in the process of economic
development. Oblivious of this, some authors recommend that economic
development should be allowed to take its own autonomous, evolutionary
course.46 Nor, for that matter, can the aim of political or institutional
stability make any special a priori claim for precedence, as others have
suggested.47 National leaders may at times be persuaded, under the
compulsions of the modernization imperative, to undertake measures which
may result in strain on, or erosion of, political institutions, thus creating
political decay.
It is not the intention of the author to suggest that national independence
replace mass welfare as the sole causal factor in explaining the nature of
economic development programs in the new nations. Rather, the purpose of
this article has been to provide a more balanced perspective by redressing
the neglect accorded to the goal of national power in studies of economic
and political development. This redress does not provide any definitive
answers concerning economic or political development, but raises further
questions for research. Among these questions are: the different ways
nations resolve the tension between the goals of mass welfare and national
power in their economic development programs; the conflict between class
interests of political elites and the aim of national independence; the
relationship of system types to success in the pursuit of economic
development and national power; the implications of the problem of
national integration for the goal of national power; the impact of failure in
economic development upon the national independence ideology; the
variations among nations in their perception of the goal of national
independence; the effect of population size and resource endowment on
such perception; the strategies that elites of small countries perceive as
available—regional integration, isolation, dependence; the impact of
political and economic penetration or domination by the industrial powers
on the pursuit and achievement of national goals in the realm of mass
welfare and national power. Hopefully, future studies of developing nations
will focus on these questions.
1. Hans W. Singer, “Social Development: Key Growth Sector,” International Development Review,
VII (March 1965), 3-5.
2. W. B. Reddaway, The Development of the Indian Economy (London, 1962), p. 48.
3. John Kenneth Galbraith, Economic Development (Cambridge [Mass.J, 1964), pp. 10-11, 62.
4. Gunnar Myrdal, Asian Drama: An Inquiry into the Poverty of Nations (New York, 1968).
5. Harry G. Johnson, ed. Economic Nationalism in Old and New States (Chicago, 1967), chaps. 1 and
8.
6. David E. Apter and Charles Andrain, “Comparative Government: Developing New Nations,” The
Journal of Politics, XXX (May 1968), 409-10.
7. Leonard Binder et al., Crises and Sequences in Political Development (Princeton, 1971), pp. 241-
73 (emphasis added).
8. Seymour Martin Lipset, The First New Nation: The United States in Historical and Comparative
Perspective (New York, 1963), p. 46 (emphasis added).
9. Karl W. Deutsch, “Social Mobilization and Political Development,” in Harry Eckstein and David
E. Apter, eds. Comparative Politics: A Reader (New York, 1963), pp. 582-603.
10. Samuel P. Huntington, “The Change to Change: Modernization, Development, and Politics,”
Comparative Politics, III (April 1971), 283-322.
11. Gabriel Almond, “A Developmental Approach to Political Systems,” World Politics, XVII
(January 1965), 183-214.
12. Binder, Crises.
13. John H. Kautsky, Political Change in Underdeveloped Countries: Nationalism and Communism
(New York, 1962), p. 98.
14. Fred W. Riggs, Administration in Developing Countries: The Theory of Prismatic Society
(Boston, 1964), pp. 42-49.
15. Kautsky, Communism and the Politics of Development: Persistent Myths and Changing Behavior
(New York, 1968), p. 174.
16. See, for example, the contributions in S. Tangri and H. Gray, eds. Capital Accumulation and
Economic Development (Boston, 1967), pp. 75-86, 86-94.
17. George Rosen, Democracy and Economic Change in India (Berkeley and Los Angeles, 1966), p.
250.
18. Alexander Eckstein, Walter Galenson, and Ta-Chung Liu, eds. Economic Trends in Communist
China (Chicago, 1968), p. 468.
19. Walt W. Rostow, Politics and the Stages of Growth (Cambridge [Mass.], 1971), pp. 65-69, 189,
195.
20. Lipset, p. 47.
21. E. Herbert Norman, Japan’s Emergence as a Modern State: Political and Economic Problems of
the Meiji Period (New York, 1940), pp. 125-26, 118.
22. See Eckstein, Economic Trends, pp. 50-51.
23. People’s Republic of China, First Five-Year Plan for Development of the National Economy of
the People’s Republic of China in 1953-57, p. 16.
24. Po I-po, cited in Dwight H. Perkins, “Industrial Planning and Management,” in Eckstein,
Galenson, and Liu, p. 598.
25. Perkins, p. 600.
26. Kautsky, Political Change, pp. 3-119.
27. Cited in Oleg Hoeffding, “State Planning and Forced Industrialization,’ Problems of
Communism, VIII (November-December 1959), 39.
28. Cited in Irving Louis Horowitz, Three Worlds of Development: The Theory and Practice of
International Stratification (New York, 1966), pp. 13-14.
29. Cited in Eckstein, Economic Trends, p. 460.
30. Robert C. Tucker, The Marxian Revolutionary Idea (New York, 1969), pp. 126-29.
31. Kautsky, Political Change, pp. 56-57.
32. New York Times, 17 February 1970.
33. Montreal Star, 17 February 1970.
34. Frank H. Golay et al., Underdevelopment and Economic Nationalism in Southeast Asia (Ithaca,
1969), p. 10.
35. Horowitz, p. 351.
36. Hla Myint, Economic Theory and the Underdeveloped Countries (New York, 1971), pp. 21-
23,32,178.
37. Data from United Nations, Statistical Yearbook 1971 (New York, 1972).
38. See Baldev Raj Nayar, The Modernization imperative and Indian Planning (Delhi, 1972).
39. On this, see Nayar, “U. S. Policy,” Seminar (New Delhi), No. 161 (January 1973), 46-55.
40. The Statesman Weekly (14 October 1972), pp. 7, 14.
41. Deutsch, “Social Mobilization”; Lucian W. Pye, Aspects of Political Development (Boston,
1966), pp. 62-67; Gabriel A. Almond and G. Bingham Powell, Jr., Comparative Politics: A
Developmental Approach (Boston, 1966), pp. 35-37; and Joseph LaPalombara and Myron Weiner,
eds. Political Parties and Political Development (Princeton, 1966), p. 14.
42. Pye, pp. 58-62; Riggs, pp. 19-31.
43. Gideon Sjoberg, The Preindustrial City: Past and Present (New York, 1960), pp. 7-18; and
Marion J. Levy, Jr., Modernization and the Structure of Societies: A Setting for International Affairs,
vol. 1 (Princeton, 1966), pp. 9-15.
44. LaPalombara ed., Bureaucracy and Political Devlopment (Princeton, 1963), pp. 35-39; and Pye,
pp. 31-48.
45. Reinhard Bendix, “What Is Modernization?” in Willard A. Beling and George O. Totten, eds.
Developing Nations: Quest for a Model (New York, 1970), pp. 3-20; and Fred W. Riggs,
“Modernization and Political Problems: Some Developmental Prerequisites,” ibid., pp. 60-82.
46. Robert T. Holt and John E. Turner, The Political Basis of Economic Development: An
Exploration in Comparative Political Analysis (Princeton, 1966), chap. 7.
47. See Samuel P. Huntington, Political Order in Changing Societies (New Haven, 1968), chap. 1.
*I would like to express my sincere appreciation to Professors Henry W. Ehrmann, Department of
Political Science, and Jagdish Handa, Department of Economics, both of McGill University, for their
comments on an earlier version of this article.
APPENDIX