Edward Luttwak

Coup d'Etat
A Practical Handbook

Harvard University Press

Cambridge, Massachusetts
Contents

List of Figures 7
List of Tables 8
Foreword by Walter Laqueur 9
Priiace to the First Edition 12
Preface to the 1979 Edition 13

1. What is the Coup d' Btat? 19

2. When is a Coup d'Btat Possible? 28
3. The Strategy of the Coup d' Btat 57
4. The Planning ofthe Coup d' Btat 105
5. The Execution of the Coup d'Btat 146

Appendix A. The Economics of Repression 175

Appendix B. Tactical Aspects ofthe Coup d' Btat 182
Appendix C. Statistics 189

Table 1. Economic Development and theCoupd'Btat, 1945-78 190

Table II. Basic Lists of Coups and Attempted Coups, 1945-78 195
Table III. The Efficiency of the Coup d'Etat 1945-64 208
Table IV. The Frequency of the Coup d' Btat 209
Index 211
List of Figures

I. Alternative forms of government. Party govern-

ment in 'socialist' countries and 'People's Repub-
lics' (Communist countries). lIS
2. The formal government and the real one. 116
3. Telecommunication facilities available to govern-
ment (notional). 121
4. The physical targets of the coup. 123
S. Physical targets in coastal city. 126
6. Operational sequence and timing. 152
7. Intelligence 'noise' and analysis. 154
8. Lead-time required by teams to reach their target
by zero hour. 156
9. Simultaneous penetration of the defensive system
and team warning time. 158
10. Political survival limit on taxation. 179
II. The Duvalier formula. 180
12. 'Sophisticated' seizure of major defended targets. I8S
13. General structure of blocking position. 187
List of Tables

I. Formal structures and real chains of command. 66

2. Country X: potential forces of intervention. 70
3. Optimum infiltration strategy. . 74
4. The role of ethnic minorities in Syrian politics. 78
S. Class of 19-at Military Academy of Country X:
present career position. 80
6. The Aref brothers in Iraq, 1958-68: a study in
loyalty. 82
7. Battalion No. I: recruitment prospects. 83
8. Infiltration of the armed forces in Portugal
(notional). 88
9. Infiltration of the armed forces in West Germany
(notional). 89
10. The groups which would influence the formation of
American policy on Middle Eastern questions. 108
II. Alternative forms of government. I13
12. Mass communications in the Middle East and
North Mrica, mid-I967. II8
13. Police telecommunication facilities in Ghana. 120
14. The mechanics of intervention of the loyalist forces. 151
IS. The first communique: a choice of styles. 169
16. National accounts data, Country X. 177
17. National accounts data, Country X. (Mter propa-
ganda and police expenditure.) 178
18. National accounts data, Country X. (Funds avail-
able for development.) 178
19. Eve of the coup. Forces of the state fully subverted
(notional). 183
Foreword

Coup d'Etat, the brilliant and original book of a then very young
man, first published in 1967, attracted immediate attention and
appeared subsequently in the major languages. It is perhaps of
even greater interest today, simply because it has become dearer
during the last decade that far from being a fortunately rare
exception in an otherwise civilized world order, the coup d'etat is
now the normal mode of political change in most member states
of the United Nations. There are by now many more military
dictatorships in existence than parliamentary democracies, and
there are few cases on record in which such dictatorships have
been overthrown by 'popular revolts'. Far more often the military
men are replaced by one or more of their colleagues. Yet with all
this there has been a virtual taboo on the study of coup d'etats, and
some critics of the present book obviously did not know quite
what to make of it. It is in many ways easy to see why: the idea
that a coup d'etat can be carried out in many parts of the world
with equal ease by small groups of men of the left and the right
(and, for all one knows, also of the centre), provided they have
mastered some elementary lessons of modern politics, is, of
course, quite shocking. Marx and Engels wrote a great deal abollt
revolution but hardly ever about the technique of revolution; the
only nineteenth century left-wing leader who provided detailed
instruction in this respect was Blanqui, and he was not very
successful. There had been one other predecessor Gabriel Naude,
whose work was published in Paris in the late seventeenth
century; an English translation by Dr. William King appeared in
1711 (Political considerations upon Refined Politicks and the
Master Strokes of State). Some of this sounds very topical indeed:
'The thunderbolt falls before the noise of it is heard in
the skies, prayers are said before the bell is rung for
them; he receives the blow that thinks he himself is
giving it, he suffers who never expected it, and he dies
10 Foreword
that look'd upon himself to be the most secure; all is
done in the Night and Obscurity, amongst Storms and
Confusion. '
But Naude has been forgonen for a long time, and his concept of
the 'master stroke' was, in any case, much wider than that of coup
d'etat in its present meaning.
In our time whole libraries have been wrinen on the objective
conditions in which revolutions take place, about civil and
peasant wars, about revolutionary and internal war, about guerril-
la activities and terrorism, but almost nothing on coup d'etats, and
this despite the fact that there have been few, if any, revolutions
of late and that 'objective conditions' are always only one of
several factors involved in their genesis. Seen in this light coup
d'itats are annoying not only for practising politicians but also
from the point of view of the political scientist. For on the basis
of 'objective conditions' models and panerns can be built with-
out undue difficulty, whereas coups are quite unpredictable,
almost by definition they are mortal enemies of orderly hypo-
theses and concepts: how does one account scientifically for
the political ambitions of a few strategically well placed indiv-
iduals? .
All this is highly regrenable but it does not lessen the need for
more thorough and detailed study of coup d'etats. For according
to all indications this .seems to be the 'wave of the future' - much
more than other, far more often discussed, forms of political
violence. A study of guerrilla warfare led me to the conclusion
that the army in most third world countries is the strongest
contender for domestic power: during the last.fifteen years there
have been some hundred and twenty military coups, whereas only
five guerrilla movements have come to power - and three of these
followed the PortUguese coup in 1974. The funciion of the
guerrilla movement has reverted to what it originally was - that
of paving the way for and supporting the regular army: it holds
the stirrup so that others may get into the saddle, and the same
applies, a fortiori to terrorist groups. It is true that in some parts
of the world it has become more difficult to stage a military coup.
Once upon a time the commander of a tank brigade in a middle
Eastern country was at least a potential contender for political
power. This is no longer so, partly as the result of centralization
in military command, partly because the political police have
become more effective. But if in these pans coups have become
Foreword 11
less frequent they are still the only form of political change that
can be envisaged at the present time.
But even if coups are unpredictable, even if they defy known
methods of interpretation (let alone of prediction), they contain
certain ever recurring patterns - 'the same always different'-
from the time the conspiracy is first hatched to the actual seizure
of power. The present book is a major landmark in a field hitherto
almost uncharted.

Walter Laqueur
Washington - London
October 1978
Preface to the First Edition

This is a handbook. It is therefore not concerned with a theoreti-

cal analysis of the coup d'itat, but rather with the formulation of
the techniques which can be employed to seize power within a
state. It can be compared to a cookery book in the sense that it
aims at enabling any layman equipped with enthusiasm - and the
right ingredients - to carry out his own coup; only a knowledge
of the rules is required. Two words of caution: in the first place
in order to carry out a successful coup certain pre-conditions
must be present, just as in cooking bouillabaisse one needs the
right sorts of fish to start with. Secondly, readers should be
aware that the penalty of failure is far greater than having to eat
out of a tin. (The rewards, too, are greater.)
It may be objected that should such a handbook be inadequate
or misleading the readers will be subject to great dangers, while
if it is an efficient guide to the problems it may lead to upheavals
and disturbances. My defence is that coups are already common
and if, as a result of this book, a greater number of people learn
how to carry them out this is merely a step towards democrati-
zation of the coup - a fact that all persons of liberal sentiments
should applaud.
Finally, it should be noted that the techniques here discussed
are politically neutral, and are only concerned with the objective
of seizing control of the state, and not at all with subsequent
policies. Readers interested in pursuing this subject further may
wish to consult some of the standard academic works related to
the subject, among which are: The Role of the Military in Under-
Developed Countries, edited by J. J. Johnson, The Man on Horse-
back, by S. E. Finer, and the chapter on Armed Forces Organiza-
tions in Marion J. Levy Jr's Modernization and Structure of
Societies. E. L.
Preface to the 1979 Edition

During the years since the original publication of this book, I

have frequently been told that it has served as the planning guide
for this or that coup. But the only case of actual use for which
there is firm evidence would make a poor advenisement: the coup
in question was at first very successful but then failed, amidst
much killing. Its chief protagonist, a defence minister who
aspired to yet greater things, was caught and promptly executed.
When his house was searched, a heavily annotated copy of the
French edition was found in his study. I could take refuge in the
excuse that the book's prescriptions were not followed with
sufficient care, but in reality it was not my purpose to supply a
bonafide do-it-yourself manual. In writing this book, my true aim
was entirely different: to explore the meaning of political life in
those backward countries now officially known as 'less deve-
loped'.
When the ideas in this book were first conceived, the intellec-
tual classes of the Western world were passionately interested in
the affairs of the Third World. There was an atmosphere of
hopeful expectation about the new states of Africa and Asia, then
emerging on the world scene for the first time. Even for Latin
America there was a new interest, and new hopes - greatly
stimulated by Kennedy's 'Alliance for Progress' which, like all of
Kennedy's projects, enjoyed excellent publicity.
But it was undoubtedly Black Africa which stimulated the
greatest interest, often remarkably emotional. The dissolution of
the British and French empires was then still in progress, and the
new states of Africa were the newest of all. Their utter poveny
was not entirely concealed by the exotic scenery, and the almost
complete absence of an educated class was brutally obvious. Yet it
was only a few right-wing extremists, and the still smaller
number of Old African Hands, who argued that independence
was being granted too soon. This small minority was easily
dismissed as reactionary and racist. The enlightened knew better:
 14 Pr~face to the 1979 Edition

th.e new states would muster the fresh energies of the peoples
liberated from the lethargy of colonial rule; their youth w~uld
soon be educated to provide technicians, professionals and civil
servants; given some aid from the West, a great upsurge of
economic development was to be expected, and this would soon
remedy backwardness, and the. contrived poverty caused by
colonial exploitation. More than that, we were told to look for
moral leadership from the new states. The idealistic young
leaders who had struggled for independence would be a great
spiritual force on the world scene.
As a student at the London School of Economics the present
writer heard such things being said as if they were not merely
true but obvious. I had no desire to join the small band of
right-wingers who alone, opposed the devolution of the British
Empire. But I found the common view to be hopelessly removed
from reality; our best minds seemed to suffer a decomposition of
the critical faculties when the subject was the Third World. This
is not the place to speculate on the obscure emotional reasons that
alone could explain such a failure of the intellect. What is certain
is that a highly favorable vision of the future of the Third World
was given wide currency, even though all the factual evidence in
hand flatly contradicted the notions on which the prediction was
based.
It was not the poverty of the new states that made me dubious
of their future, and entirely pessimistic as to their contribution to
international life. Poverty does not necessarily inhibit cultural or
even social achievement, and in any case some of the least
promising of the new states already enjoyed vast unearned
I incomes from oil exports. As for the lack of adequate administra-
tive structures, this was certainly not a fatal deficiency; few things
grow as easily as state bureaucracies. Not even the ill-effects of
relative deprivation felt by the poor, confronted with luxuries by
way of the mass media, seemed to me to be all that serious. It
seems that the 'revolution of rising expectations' - yet another
slogan made by Western intellectuals to justify forthcoming
depredations - has remained unrealized.
But there was one deficiency that was, and is, fatal, which
would inevitably cause the new states to misgovern at home,
while degrading international standards abroad. There was one
thing that the new states lacked which they could neither make
for themselves nor obtain from abroad, and this was a genuine
Preface In the 1979 Edition 15
political community. It is difficult to give a formal definition, and
perhaps it is best to begin by evoking the familiar concept of the
nation, in contrast to the state. The new states carne into existence
because the colonial authorities handed over their powers to
political leaders who had agitated for independence; more
specifically, the new leaders were given control over the army,
police, tax collectors and administrators who had worked for the
colonial government. The old servants of the empire served their
new masters, ostensibly for new purposes. But their methods and
their operational ideology were those of the imperial power,
moulded by notions which reflected the values of its political
community. There was no organic nexus between the native
cultures and the instruments of state power, and neither could
such a link be formed. For one thing there were usually several,
quite different, native cultures, often incompatible. Moreover the
methods and operational ideologies that the native cultures would
organically sustain were usually utterly unsuited to the needs of
modern life, that is Western life. The problem was not that this
dissociation would make the state apparatus weak, but rather that
it would leave it entirely unconstrained and much too strong. The
consequences are now fully evident. The administrators of the
new states are vested with all the great power over the individual
that the entire machinery of files and records, vehicles, telecom-
munications and modern weapons give to the modern state. But
their conduct is not constrained by legality, or by the moral
standards which any genuine political community must enforce,
even if only in requiring hypocrisy on the part of violators. Above
all, their conduct is not restrained by political pressures, since the
oppressed are neither afforded the electoral opportunities availa-
ble in Western democracies, nor do they have social frameworks
within which political action may be concened. Hence the
universal misgovernment which has replaced colonial non-
government throughout the territories of the new states. Always
present, bribery is now a quite normal pan in any transaction
between citizen and state; a detailed and pervasive oppression has
replaced the distant authoritarianism of colonial days, since
neither bureaucrats nor policemen are restrained by the rules of
legality - or at least the procedures of legalism - which restrained
the colonial power. As a result, exactions may increase without
limit, and no citizen may assure his libeny, life or property by
complying with the law - since the law offers no protection
16 Preface to the 1979 Edition
against the violations of its custodians. If colonialism was a crime,
the greatest offence was in its undoing when fragile native
cultures, embryonic modern societies, and minority peoples
ill-provided to protect themselves, were everywhere abandoned
into the hands of political leaders equipped with the powerful
machine of the modern state. The brutalities of Idi Amin have
been sufficiently spectacular to attract the persiste~t attention of
Western media. But Idi Amin is fully justified in his complaint
that the Western media are unfair to him: from Algeria to
Zanzibar, the peoples of Africa are ruled by autocrats whose
unfettered control over the machinery of the state allows them to
indulge every vice and every excess of virtue: in one country the
ruler may be an alcoholic, in another he may forbid alcoholic
drink to all, because he deems it irreligious; in one country, the
ruler may openly claim for himself any woman or boy that meets
his fancy, in another he may have adulterers executed; in one
country the most useless of luxuries may be freely imported while
there is no foreign currency with which to buy essential
medicines, in another the ruler may proclaim that even books may
not be imported, although foreign currency is accumulating
uselessly in foreign bank deposits. Above all, there is the
systematic use of the instruments of defense and order for
internal oppression, and the appropriation of public wealth on a
fantastic scale. When the American Vice-President was forced to
resign because he had accepted bribes, or what were deemed to be
bribes, there was amazement at the size of the sums involved: in
the Third World not even a junior minister could be purchased
for so little. The logic whereby public power may easily generate
privane wealth is universal, and the enrichmen of the powerful is
a pervasive phenomenon found all over the world. But there is a
difference in the workings of this logic in the new states, and not
merely of scale: it is not an ancillary phenomenon but rather the
essential process of governance for those in control, and it is not
moderated by the need for discretion. The overt corruption
commonly found in the new states reveals the full consequences
of the absence of political community. Only from the latter can
effective norms arise, norms felt in the consciousness of each
citizen. Without political community there can be no effective
norms, and without the norms that arise quite naturally from the
values and beliefs of the community, the state is no more than a
machine. It is then that the coup d'etat becomes feasible since, as
Preface to the 1979 Edition 17
with any machine, one may gain control over the whole by seizing
hold of the critical levers. In writing of the coup d'etat then, I was
in fact writing of the true political life of the new states.
 I. What is the Coup d'Etat?
'I shall be sorry to commence the era of peace by
a coup d'etat such as that I had in contemplation.'
Duke of Wellington, ZBII

' ••• no other way of salvation remained except for

the army's intervention •••' Constantine Kollias,
21 April 1967, Athens

Though the term coup d'etat has been used for more than 300
years, the feasibility of the coup derives from a comparatively
recent development: the rise of the modem state with its pro-
fessional bureaucracy and standing armed forces. The power of
the modem state largely depends on this permanent machinery
which, with its archives, files, records and officials, can follow
intimately and, if it so desires, control the activities of lesser
organizations and individuals. 'Totalitarian' states merely use
more fully the detailed and comprehensive information which is
available to most states, however' democratic': the instrument is
largely the same though it is used differently.
The growth of the modem bureaucracy has two implications
which are crucial to the feasibility of the coup: the development
of a clear distinction between the permanent machinery of state
and the politica11eadership, and the fact that, like most large
organizations, the bureaucracy has a structured hierarchy with
definite chains of command. The distinction between the
bureaucrat as an employee of the state and as a personal servant
of the ruJer is a new one, and both the British and the American
systems show residual features of the earlier structure.*
*In Britain there is the constitutional fiction that civil servants - as their
name implies - are the servants of the Crown. In the U SA. while the days
when party hacks moved on to Washington after an election victory are long
past. some civil-service jobs are still given to political associates. rather
than to professionals.
20 COUp d'Etat
The importance of this development lies in the fact that if the
bureaucrats are linked to the leadership, an illegal seizure of
power must take the form of a 'Palace Revolution' and it essen-
tially concerns the manipulation of the person of the ruler. He
can be forced to accept new policies or advisers, he can be killed
or held captive, but whatever happens the Palace Revolution can
only be conducted from the 'inside', and by 'insiders'. The
'insiders' can be the palace guard as in ancient Rome, or the
Ethiopia of the 1960s, and in a dynastic system they would seek
to replace an unwanted ruler with a more malleable son.
The COup is a much more democratic affair. It can be con-
ducted from the 'outside' and it operates in that area outside
the government but within the state which is formed by the
permanent and professional civil service, the armed forces and
police. The aim is to detach the permanent employees of the
state from the political leadership, and this cannot usually take
place if the two are linked by political, ethnic or traditional
loyalties.
Both Imperial China and present-day African states use ethnic
bonds in the organization of their senior bureaucracy. The
Manchu dynasty was careful to follow native Chinese customs
and to employ' Han' Chinese in the civil service at all levels, but
the crucial posts in the magistracy and the army were filled by the
descendants of those who entered China with the founders of the
dynasty. Similarly, Mrican rulers usually appoint members of
their own tribes to the strategic posts in the security services.
When a party machjne controls civil-service appointments,
either as part of a general 'totalitarian' control, or because of a
very long period in office (as in post-war Italy), political asso-
ciates are appointed to the senior levels of the bureaucracy,
partly in order to protect the regime and partly to ensure the
sympathetic execution of policies. Thus 'party men' hold the
key jobs in the police and security services of France and Italy,
just as in the communist countries all senior jobs are held by
party 'apparatchiks '.
Saudi Arabia provides an instance of 'traditional bonds'.* In
*The bonds are religious in origin since the Saudi royal house is the
traditional promoter of the Wahabi interpretation of Islam.
 What is the Coup d'Etat? 2I

this case the lack of modern 'know-how' on the part of the

traditional tribal affiliates of the royal house has meant that what
could not be done individually has been done organizationally.
Alongside the modem army staffed by unreliable city-dwellers,
there is the 'White Army' of the Wahabi bedouin followers of
the Saudis.
Such possible bonds between political leadership and the
bureaucracy are not typical of the modern state. Apart from the
cases noted above, the civil service and the leadership of the
modern state may still be linked by class or ethnic affiliations,
but whatever the nature of the bonds, such groupings will
usually be large enough to be infiltrated by the planners of the
coup.
As a direct consequence of the sheer size of the bureaucracy,
and in order to achieve even a minimum of efficiency, the state
bureaucracy has to divide its work into clear-cut areas of
competence, which are assigned to different departments.
Within each department there must be an accepted chain of
command, and standard procedures have to be followed. Thus
a given piece of information, or a given order, is followed up in a
stereotyped manner, and if the order comes from the
appropriate source, at the appropriate level, it is carried
out.
In the more critical parts of the state apparatus, the armed
forces, the police, and the security services, all these charac-
teristics are intensified, with an even greater degree of discipline
and rigidity. The apparatus of the state is therefore to some
extent a 'machine' which will normally behave in a fairly
predictable and automatic manner.
A coup operates by taking advantage of this machine-like
behaviour: during the coup because it uses parts of the state
apparatus to seize the controlling levers; afterwards because the
value of the 'levers' depends on the fact that the state is a
machine.
We will see that some states are so well organized that the
'machine' is sufficiently sophisticated to exercise discretion,
according to a given conception of what is 'proper' and what is
not, in the orders that it executes. This is the case in some
C.E.-2
 22 COUp d'Etat
developed countries, and in such circumstances a coup is very
difficult to carry out.
In a few states the bureaucracy is so smaIl that the apparatus
is too simple and too intimately liDked with the leadership to be
suitable for a coup, as perhaps in the ex-British Protectorates in
southern Africa, Botswana, Lesotho and Swaziland. Fortunately,
most states are between these two extremes, the machine being
both large and unsophisticated and thus highly vulnerable to
those who can identify and seize the right clevers••
One of the most striking developments of the last century has
been the great decline in general political stability. Since the
French Revolution governments have been overthrown at an
increasing pace.* In the nineteenth century the French ex-
perienced two revolutions and two regimes collapsed following
military defeat. In 1958 the change of regime was a sophisticated
blend of both these elements. Peoples everywhere have followed
the French example, and the life-span of regimes has tended to
decrease while the life-span of their subjects has increased. This
contrasts sharply with the relative attachment to the system of
constitutional monarchy which was displayed in the nineteenth
century: when Greeks, Bulgars and Rumanians secured their
freedom from the Turkish colonial system they immediately
went over to Germany in order to shop around for a suitable
royal family.
Crowns, flags and decorations were designed and purchased
from reputable (English) suppliers; royal palaces were built;
and where possible, hunting lodges, royal mistresses, and a local
aristocracy were provided as fringe benefits. Twentieth-century
peoples have, on the other hand, shown a marked lack of in-
terest in monarchies and their paraphernalia; when the British
kindly provided them with a proper royal family the Iraqis made
numerous efforts to get rid of it, before finally succeeding in
1958. Military and other right-wing forces have meanwhile tried
to keep up with popular movements and have used their illegal
methods in order to seize power and overthrow regimes.
*Historica1ly speaking, the trend was initiated by the American Revolu-
tion; its impact on the world at large was, however, attenuated by America'
distance and exotic nature.
 What is the Coup d'Etat? 23
Why have regimes in the twentieth century proved so fragile?
It is, after all, paradoxical that this fragility has increased while
the established procedures for securing changes in government
have on the whole become more flexible. The political scientist
will reply that though the procedures have become more flexible,
pressures for change have also become stronger, and that the
rate of increase in flexibility has not kept up with the growth of
social and economic pressures.*
Violent methods are generally used when legal methods of
securing a governmental change are useless because they are
either too rigid - as in the case of ruling monarchies where the
ruler actually controls policy formation - or not rigid enough.
It was once remarked, for example, that the throne of Russia
was, until the seventeenth century, neither hereditary nor elective
but 'occupative'. The long series of abdications forced by the
great Boyar-landlords and by the 'Archers' of the Guard had
weakened the hereditary principle, so that whoever took
the throne· became Tzar, and precedence by birth counted for
little.
Some contemporary republics have ended in this position,
which comes about when a long series of illegal seizures of
power leads to a decay of the legal and political structures which
are needed to produce new governments. Thus post-war Syria
has had more than a dozen coups, and the provisions for open
general elections, written in the 'Hourani' constitution, could
not now be applied because the necessary supervisory machinery
no longer functions. Assuming, however, that there is an estab-
lished procedure for cl1anging the leadership, then all methods
other than this come within the range of illegality. What we call
themt depends on what side we are on but, skipping some of the
semantics, we have the following:
*Perhaps the ultimate source of destabilizing pressures has been the
spectacular progress of scientific discovery and the resultant technological
change. This is, however, a problem far beyond the scope of this book.
tThe equation 'Insurgency = Terrorism = War of National Liberation J
is particularly familiar.
 24 Coup d'Etat

Revolution
The action is conducted, initially at any rate, by uncoordinated
popular masses, and it aims* at changing the social and political
structures, as well as the actual personalities in the leadership.
The term has gained a certain popularity, and many coups are
graced with it, because of the implication that it was 'the people'
rather than a few plotters who did the whole thing. Thus the
obscure aims which Kassem had in mind when he overthrew the
Faisal-Nuri-es-Said regime in Iraq are locally known as the
'sacred principles of the July 14th Revolution'.

Civil War
Civil War is actual warfare between elements of the national
armed forces leading to the displacement of a government.
This term is unfashionable and if you are Spanish and pro-
Franco you call the events of 1936-9 '/a cruzada I - the crusade.
If you are not pro-Franco, but you may be overheard, you just
call them 'the events of •• .'

Pronunciamiento
This is an essentially Spanish and South American version of the
military coup d'etat, but many recent African coups have also
taken this particular form. In its original nineteenth-century
Spanish version this was a highly ritualized process: first came
the trabajos (literally the 'works ') in which the opinions of army
officers were sounded. The next step was the compromisos, in
which commitments were made and rewards promised; then
came the call for action and, finally, the appeal to the troops to
follow their officers in rebellion against the government.
The pronunciami"ento was often a liberal rather than a reac-
tionary phenomenon and the theoretical purpose of the take-
over was to ascertain the 'national will' - a typically liberal
*In the initial stages no aims are conceptualized but the scope of the action
may be clearly perceived.
 What is the Coup d'Etat? 2S
concept. Later, as the army became increasingly right-wing
while Spanish governments became less so, the theory shifted
from the neo-liberal 'national will' to the neo-conservative 'real
will' theory. This postulates the existence of a national essence,
a sort of permanent spiritual structure, which the wishes of the
majority may not always express. The army was entrusted with
the interpretation and preservation of this' essential Spain' and
to protect it against the government and, if need be, against the
people. *
The pronunciamiento was organized and led by a particular
army leader, but it was carried out in the name of the entire
officer corps; unlike the putsch, which is carried out by a faction
within the army, or the coup, which can be carried out by
civilians using some army units, the pronunciamiento leads to a
take-over by the army as a whole. Many African take-overs, in
which the army has participated as a whole, were therefore very
similar to the classic pronunciamiento.

Putsch
Essentially a wartime or immediately post-war phenomenon, a
putsch is attempted by a formal body within the armed forces
under its appointed leadership. The Kornilov putsch is a clear
example: Komilov, a general in charge of an army group in
north Russia, attempted to seize Leningrad, in order to establish
a 'fighting' regime which would prosecute the war. (Had he
succeeded, the city would perhaps bear his name instead of
Lenin's.)

Liberation
A state may be said (by supporters of the change) to be 'liberated
when its government is overthrown by foreign military or diplo-
matic intervention. A classic case of this was the installation of
the Communist leadership in Rumania in 1947. The USSR
forced the then King Michael to accept a new Cabinet by
*Various versions of this theory became popular in parts of the French
Armed Forces during the 19SoS and early 1960s.
26 Coup d'Etat
t1lreateDiDg to use the Soviet army in the country in the event
of a refusal.

War of Nadonal Liberadon, ~cy etc.

In this form of internal CODfiict the aim of the initiating party
is not to seize power within the state, but rather to set up a rival
state structure. This can be politically or ethnically based.
The Viet Cong in South VietDam aim at setting up a new
social structure and, incidentally, a new state. The Kurds in
Iraq, the Somalis in Kenya, the Karens in Burma and the Nagas
in India want to withdraw their areas from the state concerned.

The DefiDidoD of the Coup

A coup d'stat involves some elements of all these different
methods by which power can be seized but, UDlike most of them,
the coup is not necessarily assisted by either the intervention of
the masses, or, to any significant degree, by military-type force.
The assistance of these forms of direct force would no doubt
make it easier to seize power, but it would be umea1istic to think
that they would be available to the organizers of a coup. Because
we will not be in charge of the armed forces we c:aanot hope to
start the plannjng of a coup with sizeable military units already
under our control, nor will the pre-coup lJOvernment usually
allow us to carry out the propaganda and organization necessary
to make effective use of the 'broad masses of the people'.
A second distinguishing feature of a coup is that it does not
imply any particular political orientation. Revolutions are
usually 'leftist' while the putsch and the pronunciamitmto are
usually initiated by right-wing forces. A coup, however, is politi-
cally neutral, and there is no presumpdon that any particular
policies will be followed after the seizure of power. It is true that
many coups have been of a decidedly right-.character but
there is nothina inevitable about this.*
*The Greek: CIIUP has reinforced this image of the 'reactionary CIIUP' but
the Syrian coup of 1966, the Iraqi of 1958, the Yemeni of 1962 were aU
essentially leftist, if hardly llberal or progressive.
 What is the Coup d'Etat? 27
If a coup does not make use of the masses, or of warfare, what
instrument of power will enable it to seize control of the state?
The short answer is that the power will come from the state
itself. The long answer makes up the bulk of this book. The
following is our formal and functional definition:
A coup consists of the infiltration of a small but critical seg-
ment of the state apparatus, which is then used to displace the
government from its control of the remainder.
2. When is a Coup d'Etat Possible?

'The Bolsheviks have no right to wait for the

Congress of Soviets•••• They must take power
immediately•••• Victory is assured and there are
nine chances out of ten that it will be bloodless •
• • • To wait is a crime against the revolution.'
Vladimir Ilich Ulyano'{) Lenin, October 1917

Since 1945, the process of decolonization has more than doubled

the number of independent states, so that the opportunities open
to us have expanded in a most gratifying manner. We have to
recognize, however, that not all states make good targets for our
attentions. There is nothing to prevent us from carrying out a
coup in, say, Britain, but we would probably be unable to stay in
power for more than a short time. The public and the bureau-
cracy have a basic understanding of the nature and legal basis of
government and they would react in order to restore a legitimate
leadership.
This reaction renders any initial success of the coup meaning-
less, and it would arise even though the pre-coup government
may have been unpopular, and the 'new faces' might be
attractive. The reaction would arise from the fact that a signifi-
cant part of the population takes an active interest in political
life, and participates in it. This implies a recognition that the
power of the government derives from its legitimate origin, and
even those who have no reason to support the old guard have
many good reasons to support the principle of legitimacy.
We are all familiar with the periodic surveys which 'show'
that, say, twenty per cent of the sample failed to name correctly
the Prime Minister, and we know that a large part of the
population has only the vaguest contact with politics. Neverthe-
less in most developed countries those who do take an active
interest in politics form in absolute terms a very large group.
 When is a Coup d'Etat Possible ( 29
Controversial policy decisions stimulate and bring to the surface
this participation: pressure groups are formed, letters are sent to
the press and the politicians, petitions and demonstrations are
organized, and this adds up to a continuing dialogue between the
rulers and the ruled.
This dialogue does not depend necessarily on the existence of
a formally democratic political system. Even in one-party states,
where power is in the hands of a few self-appointed leaders, a
muted but nevertheless active dialogue can take place. The
higher organizations of the party can discuss policy decisions,
and in times of relative relaxation the discussions extend to the
larger numbers in the lower echelons, and to pUblications
reflecting different 'currents' - though only within the wider
framework of the accepted ideology and the broad policy deci-
sions of the leadership. The value of the dialogue that takes place
in non-democratic states varies very greatly. In Yugoslavia,* for
example, the Communist Patty has become a forum for increas-
ingly free and wide-ranging debates on major political issues; the
press and the parliament, though unable to assert their opinions
in the manner of their American or British counterparts, have
also participated in the 'opening' of the system. Policy decisions
are no longer merely applauded but are actually examined,
criticized, and sometimes even opposed.
It may well be that this particular one-party state is no longer
good coup territory. When people have learned to scrutinize and
question orders they will probably refuse merely to accept the
orders of the coup, any more than they accept the orders of the
government without ascertaining their legitimacy and desira-
bility.
In Egypt, on the other hand, the single patty, the Arab
Socialist Union (the ASU), is still a mere rubber stamp. Policy
decisions are made by Abd-el-Nasirf and his associates; they are
then carried out by the bureaucracy, and the ASU can only
cheer them along. When the question came up of whether the
ASU-dominated National Assembly would accept Nasir's with-
*And of late in ~zechoslovakia.
t Abd-el-Nasir = Servant of the Giver of Victory. Nasir or Nasser means
Victory, and this is rather inappropriate.
30 Coup d'Etat
drawal ofbis 'resignation', following the June 1967 debAcle, an
observer pointed out that the Assembly 'will jolly well do what
it is told'. Many one-party states are in this position.
The dialogue between the rulers and the ruled can only take
place if there is a large enough section of society which is
sufficiently literate, well fed and secure to 'talk back'. Even then
certain conditions can lead to a deterioration of the relationship,
and this can sometimes generate sufficient apathy and distrust of
the regime to make a coup possible.
The events of 1958 in France were in some respects similar to
a coup. Twenty years of warfare, which had included the defeat
of 1940, the Occupation and, from 1946, the long and losing
colonial wars in Indochina and Algeria, had undermined the
political structure. The continual changes of government had
lost for the regime the interest and respect of most Frenchmen
and left the bureaucracy leaderless, since the complex business
of the ministries could not be mastered by ministers who were
only in power for brief periods. The military were left to fight
the bitter Algerian war with little guidance from the Paris
authorities, because more often than not, the ministries were too
busy fighting for their survival in the Assembly to worry about
the other, bloodier, war.
The cost of the war, in both money and lives, antagonized the
general public from both the anny and the government, and
many Frenchmen felt a growing fear and distrust of this army,
whose sentiments and ideology was so alien to them - and to the
spirit of the times.
While the structures of political life under the IVth Republic
were falling apart, de Gaulle, in simulated retirement, gradually
emerged as the only alternative to the chaos that threatened.
When the army in Algeria appeared to be on the verge of really
drastic action - far beyond the earlier street theatricals of Algiers
- and yet another government was on the verge of collapse, de
Gaulle was recalled.
He was able to impose his own terms. When on 29 May 1958
Cety, the last President of the IVth Republic, called on him to
form a government (which was invested on 1 June), de Gaulle
was given powers to rule by decree for six months and to write
 When is a Coup d'Etat Possible? 31
a new constitution. Under the terms of this constitution,
presented for consultation in mid-August and approved by
referendum in September, elections were held in which de
Gaulle's newly formed UNR party won a majority. On 21
December de Gaulle became the first President of the Vth
Republic - an American-style President with wide executive
powers, but without an American-style Congress to restrain
them.
The France of 1958 had become politically inert and therefore
ripe for a coup. The political structures of most developed
countries, however, are too resilient to make them suitable
targets, unless certain 'temporary' factors weaken the system
and obscure its basic soundness. Of those temporary factors the
most common are:
(a) severe and prolonged economic crisis, with large-scale
unemployment or runaway inflation;
(b) a long and unsuccessful war or a major defeat, military or
diplomatic;
(c) chronic instability under a multi-party system.
Italy is an interesting example of an economically developed,
socially dynamic, but politically fragile country. Until the forma-
tion of the present relatively stable coalition between the
Democrazia Cristiana (DC) and the Socialists, the DC - though
the largest single party - could only rule by allying itself with
some of the many small parties of the centre, as it rarely polled
more than a simple majority.*
Coalition 'governments are always difficult to form, and in
Italy the problem became so complex that it resembled a parti-
cularly intricate puzzle. If you had been called to form a govern-
ment you would have been facing a set of mutually exclusive
propositions:
The DC is the single largest group in Parliament but, having
only thirty per cent of the votes, it cannot rule alone. If the
two smallleft-of-centre parties (the Social Democrats and the
Republicans) are brought in, the right-wing breaks away and
the government falls. If the right-of-centre parties (the
Monarchists and the Liberals) are brought in, the left-wing
*The stable coalition has since collapsed.
32 Coup d'Etat
breaks away and the government falls. If the large far-right-
wing party (the Nro-Fascists) is brought in, all the left-wing
groups will vote against the government. If •••
So the ruling party has stayed in power since 1948 without a
break, but it has not been able to do so at the same time reform-
ing the swollen and inefficient bureaucracy. Everybody wanted
reform, except that the right-wing resisted changes in the struc-
ture of local government and the police, while the left opposed
any checks on the fantastic activities of the State corporation.*
Meanwhile the parties out of power concentrated their efforts on
trying to break the coalition rather than on scrutinizing govern-
mental policy, so that abuses and inefficiency were unchecked.
Italians gradually lost interest in the goings-on in Rome, and
had there been an Italian de Gaulle the Republic would no
doubt have collapsed. (Even without a suitable man a coup almost
took place.)

The Pre-Conditions of the Coup

The France of 1958 and, to a lesser extent, the Italy of 1964 were
both countries where the dialogue between government and
people had broken down. But in many - indeed most - countries
of the world the dialogue cannot take place at all, regardless of
the temporary circumstances.
If we draw up a list of those countries which have recently
experienced coups, we shall see that though their ethnic and
historical backgrounds differ very considerably, they share
certain social and economic characteristics. If we isolate these
we can develop a set of indicators which when applied to the
basic socio-economic data of a country will show whether it will
make a good target for a coup.
*The activities of Enrico Mattei, the head of the National Hydrocarbon
Corporation (ENI) have been described in almost Bondian terms in a
number of books; he built up an industrial empire, battled the international
oil industry, interfered in Italian politics, forced foreign policy changes,
and died in unexplained circumstances.
 When is a Coup d'Etat Possible? 33

ECONOMIC BACKWARDNESS
In countries without a developed economy and the associated
prosperity, the general condition of the population is charac-
terized by disease, illiteracy, high birth- and death-rates, and
periodic hunger.
The average man in this state of deprivation is virtually cut off
from the wider society outside his village and clan. He has little
that he can sell. He has little with which to buy. He cannot read
the forms, signposts and newspapers through which society
speaks to him. He cannot write, nor can he afford to travel, so
that a cousin living as a city-dweller might as well be on the
moon. He has no way of knowing whether a particular tax is
legal, or merely the exaction of the village bureaucrat; no way of
knowing about the social and economic realities that condition
the policies that he is asked to applaud. His only source of con-
tact with the outside world is the governmental radio - and
he knows from past experience that it does not always tell the
truth.
The complexity of the outside world and the mistrust that it
inspires are such that the defenceless and insecure villager
retreats into the safe and well-known world of the clan, the tribe
or the family. He knows that the traditional chiefs of tribe and
clan prey on his very limited wealth, he often knows that their
mutual interests are diametrically opposed, but nevertheless they
represent a source of guidance and security that the state is too
remote and too mysterious, to be.
The city-dweller has escaped the crushing embrace of tradi-
tional society, but not the effects of ignorance and insecurity.
In these conditions the mass of the people is politically passive
and its relationship with the leadership is one-way only. The
leadership speaks to them, lectures them, rouses hopes or fears,
but never listens; the bureaucracy taxes them, bullies them,
takes their sons away for the army, their 1abour for the roads,
but gives very little in return. At best, in honest regimes, a dam
or a steel mill is being built somewhere, far away from their
village. These projects will not bring them any direct benefit, will
not lift them from their misery, but at least they are a consolation,
 34 Coup d'Etat
a hope of a better future for their sons. Elsewhere the poor are
even denied the consolation of hope: their taxes have been spent
on palaces, tanks, planes, and all the bizarre things that politi-
cians and their wives need. The urban poor living by expedients,
barely surviving in the day-to-day struggle for the necessities of
life, are treated to the spectacle of the cocktail parties, limousines
and grandiose villas of the ruling elite.*
The mass of the people is politically passive, but it is the
passivity of enforced silence, not inertia. All the time the terrible
anger caused by deprivation and injustice is there, and at times
it explodes. The mob may not have a clear political purpose, but
its actions do have political consequences.
The 1952 coup in Egypt, which led to the end of King Farouk's
'white telephone' monarchy, and to the rise of what eventually
became Abd-el-Nasir's regime, was preceded by one of these
sudden explosions. 26 January 1952 ('Black Saturday') was the
appointed date of an organized demonstration against the
presence and activities of the British forces in the Canal Zone.
The poor of the city streamed out from their hovels and joined
the procession, amongst them the agitators of the Muslim
Brotherhood,t who incited the crowd to arson and violence.
The agitators succeeded beyond their wildest dreams. The
poor seized the opportunity and destroyed the facilities of the
rich: hotels, department stores, the Turf Club, the liquor stores
and fashion shops in the centre of the city which was given the
appearance of a battlefield in one short day; only the wealthy
suffered, as these were places that had always been closed to the
poor. The middle-class organizers of the original demonstration
had no wish to destroy their own favourite gathering places; the
nationalists did not want to deprive Egypt of the 12,000 dwel;,.
lings and 500 businesses that were destroyed. They spoke of
anarchy, intrigue and madness. For the poor, however, it was a
general election: the voteless poor had voted with fire.
*Even then Bome of them retain their sense of humour; in some African
languages a new word has been coined from the Wa-prefix which indicates
a tribe. and where before there were only Wa-Kamba and Wa-Zungu now
a new tribe has appeared: the Wa-Benz and the Wa-Rolls-Royce.
t As in all Middle Eastern episodes. the affiliations of the agitators are
not certain. It has been said that the Palace organized the whole thing.
 When is a Coup d'Etat Possible ~ 35
Apart from the violent and inarticulate action of the mob in
response to some simple and dramatic issue, there is no arguing
with the power of the state; there is no interest in, and scrutiny
of, the day-to-day activities of government and bureaucracy.
Thus if the bureaucracy issues orders they are either obeyed or
evaded, but never challenged or examined.
All power, all participation, is in the hands of the small edu-
cated elite. This elite is literate, educated, well-fed and secure,
and therefore radically different from the vast majority of their
countrymen, practically a race apart. The masses recognize this
and they also accept the elite's monopoly of power, and unless
some unbearable exaction leads to desperate revolt they will
accept its policies. Equally they will accept a change in government,
whether legal or otherwise. Mer all, it is merely another lot of
'them' taking over.
Thus, after a coup, the village policeman comes to read out a
proclamation, the radio says that the old government was corrupt
and that the new one will provide food, health, schooling - and
sometimes even glory. The majority of the people will neither
believe nor disbelieve these promises or accusations, but merely
feel that it is all happening somewhere else, far away. This lack
of reaction* is all the coup needs on the part of the people in order
to stay in power.
The lower levels of the bureaucracy will react - or rather fail
to react - in a similar manner, and for similar reasons. Their
own lack of political sophistication will mean that the policies
and legitimacy of the old government were much less important
to them than their immediate superiors. The 'bosses' give the
orders, can promote or demote and, above all, are the source of
that power and prestige that make them village demi-gods. Mer
the coup, the man who sits at district headquarters will still be
obeyed - whether he is the man who was there before or not - so
long as he can pay the salaries and has links to the political
stratosphere in the capital city.
*The crowds that demonstrated in Cairo to dissuade Nasir from resigning
on 10 June 1967 gave impressive proof of his popularity. This was DOt.
however. an independent reaction: peasants from the Delta had been trLIcked
by the thousand to 'stimulate' the demonstration, which was widely publi-
cized by the state broadcasting service. .
 36 Coup d'Etat
For the senior bureaucrats, army and police officers, the coup
will be a mixture of dangers and opportunities. Some will be too
compromised with the old regime merely to ride out the crisis
and they will either flee, fight the coup, or step forward as sup-
porters of the new regime in order to gain the rewards of early
loyalty. The course of action followed by this group will depend
on their individual assessments of the balance of forces on the
two sides. But for the greater number of those who are not too
deeply committed, the coup will offer opportunities rather than
dangers. They can accept the coup and, being collectively indis-
pensable, can negotiate for even better salaries and positions;
they can create or join a focus of opposition; or, as in Nigeria in
1966, can take advantage of the temporary state of instability
and stage a counter-coup and seize power on their own account.
Much of the planning and execution of a coup will be directed
at influencing the decision of the elite in a favourable manner.
Nevertheless if, in an underdeveloped environment, they choose
to oppose the coup, they will have to do so in terms of political
rivalry. They would not be able to appeal to some general
principle of legality as in politically sophisticated countries,
because no such principle is generally accepted. So instead of
operating for the sake of legitimacy, they would be fighting the
planners of the coup as straight political opponents and therefore
on the same plane. This would have the effect of bringing over
to the coup their political or ethnic opponents. In any case,
fighting the coup would mean facing organized forces with
improvised ones, and in conditions of isolation from the masses
who, as we have seen, will almost always be neutral.*
As the coup will not usually represent a threat to most of the
elite, the choice is between the great dangers of opposition and
the safety of inaction. All that is required in order to support the
coup is, simply, to do nothing - and this is what will usually be
done.

*Many observers have commented on the lack of popular support for

fallen political idols. as in the case of N'krumah's overthrow when the
Accra citizenry seemed as happy to cheer his enemies as they had been to
cheer him shortly before the coup. This is not inanity. but highly rational
behaviour in the light of social and economic circumstances.
 When is a Coup d'Etat Possible? 37
Thus, at all levels, the most likely course of action following
a coup is acceptance: by the masses and the lower bureaucracy
because their interests are not tied with either side; by the upper
levels of the bureaucracy because of the great dangers of any
opposition conducted in isolation. This lack of reaction is the
key to the victory of the coup, and it contrasts with the spon-
taneous reaction that would take place in politically sophisticated
societies.
In totalitarian states the midnight arrests, the control over all
associations (however non-political), are all part of the general
tactic of insulating the individual who seeks to oppose the regime.
In underdeveloped areas the opposition is isolated from the
masses almost automatically by the effect of social conditions.
Our first pre-condition of a coup is therefore:
The social and economic conditions of the target country must
be such as to confine political participation to a small fraction of
the population.
By participation we do not mean an active and prominent role
in national politics, but merely that general understanding of the
basis of political life, commonly found amongst the masses in
economically developed societies. This pre-condition also im-
plies that, apart from the highest levels, the bureaucracy operates
in an unresponsive and mechanical manner because of its under-
educated staff. '
More generally the Ceconomic pre-condition' excludes the
possibility of a system of local government - or rather repre-
sentative local government. It is true that in underdeveloped
areas there is often. a system of local government, based on
traditional chiefs, but of their two possible roles, neither usually .
functions as a representative one.* They are either individually
powerful in their own right, which means in effect that the com-
moner is subjected to dual control, or else if their power has
collapsed they are little more than somewhat old-fashioned civil
servants. Neither of these two roles allows the commoner to
*There are some tribal groupings in Africa (notably the Kabyles of
Algeria) and elsewhere where an almost pure' Greek city' type of democracy
exists. But they are exceptions to the rule.
38 Coup d'Etat
participate in the small politics of the village or town in the
manner of his Western counterpart.
Thus, in an economically backward environment the diffusion
of power, which IS characteristic of sophisticated democracies,
cannot take place. There is either rigid centralized rule, or, as a
transitional phase, a degree of power for individual regions that
makes them de facto independent states (as was the case in
northern Nigeria before the coup). Everybody knows that it is
easier to take something concrete than something vague. Talking
loosely, power in the centralized state run by a narrow elite is
like a well-guarded treasure; power in a sophisticated demo-
cracy is like a free-floating atmosphere - and who can seize
that?
The pre-condition of economic backwardness can be tested
against the known facts of the degree of economic development
of countries that have had coups in the last ten years or so and
there is a clear connexion between the two.*
This does not necessarily mean that (a) all underdeveloped
countries are ipso facto vulnerable to a coup, nor (b) that the
developed areas are never good coup territory. It does mean
however, that only the intervention of special circumstances will
prevent a well-planned coup from succeeding in economically
backward countries, while only exceptional circumstances
(discussed in Chapter I) will allow it to succeed in the developed
areas.

POLITICAL INDEPENDENCE

It is impossible to seize power within a state if the major source

of political power is not there to be seized. The I956 Hungarian
Revolution, for example, was totally successful and the leaders
of the Revolution quickly found themseJves in control of all the
traditional instruments of power: the armed forces, the pollee,
the radio and the communication facilities. The one thing that
could not be seized in the streets of Budapest happened to be the
maior source of power to the previous regime: the presence of the
Soviet Army in and around Hungary.
*See Appendix C, Table I.
 When is a Coup d'Etat Possible? 39
These armed forces - vastly superior to the Hungarian Anny-
were a greater source of power to a Kremlin-backed government
than any element within the country. The control of the Red
Anny was in Moscow, thus the Revolution would only have
succeeded if it had been carried out in Moscow, not Budapest.*
This dependence of one government on another is a wide-
spread phenomenon; although there are very few colonial states
left there are many pseudo-independent ones. In some cases
these have less real autonomy than some of the old colonies.
In both East Germany and South Vietnam the armed forces
of the greater 'ally' are much more powerful than those of the
country itself. In both countries the international and the
domestic situations are so closely interrelated that the greater
'ally' must have e1iective control over vast areas of domestic
policy - even assuming that he had no wish to interfere for other
reasons. The rulers of East Germany have always been very
careful in ensuring the approval of the Soviet authorities for
even the smallest policy changes, and each·policy announcement
can be traced back to a r~cent trip to Moscow. Similarly, in
Saigon the US embassy is generally recognized as a greater
source of power than the Presidential Palace.
In these conditions a coup can only work with the approval
of the greater 'ally'. The first coup in Vietnam, which overthrew
the unpopular Diem, and the still less popular Ngo family, was
carried out by men who appreciated these realities of power.
When the events of Hue of 8 May 1963 resulted in a break
between Diem and the Buddhists, the long-suffering generals
decided to act: they 'sounded out' the opinion of the US
embassy in Saigon, and asked through an intermediary whether
the Americans would report to Diem 'possible consultations on
eventual changes in the prevailing political structures'. When,
after some considerable debate between the CIA, the embassy,
the Presidency and the Pentagon, the US authorities informed
the plotters that they would not be reported to Diem, the
following sequence of events took place:
*The other cause of the failure of the Revolution was, of course, the
fact that Moscow's intervention was not stopped by Washington but, again
the control of US policy cannot be seized in Budapest.
40 Coup d'Etat
.May 1963: Beginning of intensified conflict between
Buddhists and Diem.
May-September: Internal American debate on whether the
Buddhists are neutralists (to be opposed) or nationalists (to be
supported). The final conclusion reached was that 'Hinayana
Buddhists' were' bad' and 'Mahayana* Buddhists' were' good'.
October 1963: Standstill of all economic aid to Vietnam, i.e.
to Diem's regime.
22 October 1963: End of direct aid by CIA to Ngo Dinh
Nhu'st 'Special Forces'. These were the main source of direct
power to the regime, entirely financed and equipped by the
CIA.
I November 1963: Coup. 2 November: death of Diem ana
Ngo Dinh Nhu.
The Viet Cong accused the generals and their 'front-man',
Van Minh, of being stooges of the Americans, but in their
dealings with the US authorities they were merely being
realistic: they saw that whatever power there was to be seized
depended on the Americans. Seizing Saigon's fixtures and fittings
without US support would have been seizing an empty symbol.
East Germany and South Vietnam are two clear cases of
dependence, but there are many more countries which are in the
grey area between effective independence and total submission.
Ex-colonies provide many of these more subtle cases of depen-
dence, and the presence of the former mother-country is very
real, and very effective. Instead of large and expensive armies
there are military and economic 'advisers', there is economic
aid and, above all, the tight web of long-established dependence
in non-political areas. Thus schooling follows patterns origi-
nally established in colonial days, the organization of the pro-
fessions follows the 'metropolitan' system, and this is very
important where the ruling elite is composed largely of lawyers,
whose whole raison d'etre is based on the use of a particular
procedure and code of law. Trade is often largely tied to the
ex-colonial power because of the hold of inherited tastes, habits,
*See Buddhism by Christmas Humphries (Penguin Books, 1951). For a
fuller discussion of Buddhism in South Vietnamese politics see Chapter 4.
tDiem's brother-in-law and sometime 'strong man' of South Vietnam.
 When is a Coup d'Etat Possible? 41
and the fact that trade links are often based on established
relationships and communications. Thus, five years after in-
dependence, it was still necessary for adjacent ex-French and
ex-British colonies to route telephone calls through Paris and
London, though a direct line would have been 300 instead of
5,000 miles long.
The power of these links between ex-colonies and the former
colonial power varies from case to case. But quite unobtrusive
ties can sometimes give sufficient leverage to the latter to prevent
or oppose a coup. Thus in 1964 the French reversed an otherwise
successful ,,,up in Gabon - without needing to use more than
symbolic force. The power of the ex-colonial countries has
waned from its peak in most places, but it is still considerable
even in sheer military terms. Everybody remembers how a few
companies of commandos stopped the mutinies in East Mrica.
Although the French have opted for a policy of neutrality to new
coups in Gabon and elsewhere, should they want to intervene
they still (1968) retain six thousand troops in Africa, with air
transport and efficient equipment. * Six thousand does not sound
like a large force, but it is huge compared to most armies in the
ex-French colonies - the Central African Republic's army, for
example, numbers 500 and the average for the whole group is
about a battalion (1,000 men) per country.
A very specialized type of dependence is a by-product of
modern technology and is found oute.ide the ex-colonial sphere.
This is the heavy mortgage placed on political independence by
the acquisition of sophisticated weaponry, particularly aircraft.
The jet fighter is the crucial case: unlike ships and tanks, jet
fighters confer an absolute advantage. Better training and morale
can overcome the inferiority of equipment on the ground, but
not in the air. The best pilots in the world using old-generation
subsonic fighters can do very little about faster aircraft. There-
fore it is vitally important for any country to match its potential
rivals' aircraft. The political problem arises because (a) only a
few countries make jet fighters, (b) the planes need a continual
flow of spare parts, and (c) there is a long 'gestation' period
* In fact the post-coup military regimes have tended to be even more pro-
French than previous civilian regimes.
 42 Coup d~Etat
between the original order and the time when training and
ancillary equipment are sufficiently developed for operational use.
Thus, if a country wants to acquire jet fighters it has to be
reasonably friendly with one of five countries: Sweden, USA,
France, Britain or the US S R. * Once a deal is made, it will need
to stay friendly, otherwise the fiow of supplementary equipment
will stop. Thus the initial purchase is followed by years of
dependence. Jet fighters don't 'grow' in economically backward
countries where the whole industrial base is lacking, so that the
constant up-dating of electronics, air-to-air missiles, radar
equipment etc. have to be imported - and from the original
source to whose aircraft they are related.
Both sides of the bargain recognize this dependence and the
supply of sophisticated weaponry is usually allied with general
trade, ideological and political links. At what point is the degree
of dependence sufficient to affect the feasibility of the coup?
Consider the following time-table of relations between the Soviet
Union and Egypt from 1955 to 1967:

1955 'Czech' arms deal: this was the first arms supply contract
between the Soviet Uniont and any Arab state; it was of great
political importance for Egypt because it broke the Western
arms monopoly; and signified 'true' independence.
Effect: the commitment of (future) foreign currency earnings,
and the need to keep on friendly terms with the only possible
supplier of spare parts.
1956 Suez-Sinai War: the Egyptian defeat in the Sinai
resulted in the loss of much equipment; this was quickly
replaced by the Soviet Union.
Effect: the commitment to the US S R reinforced IUld financial
indebtedness increased.
*Some other countries manufacture or are in the process of developing
jet fighter-type aircraft, including Japan, Italy, Canada and India. But these
suppliers are limited either in terms of designs, specifications, resale freedom
or production capabilities.
tThe arms supply contract was Czech in name only. Kermit Roosevelt of
the CIA (State Department) was then Nasir's adviser and he suggested that
it should be called 'Czech' to pacify H. Trevelyan, the British Ambassador.
:j:Or rather, oligopoly.
 When is a Coup d'Etat Possible? 43
1962 Revolution and 'civil war'in theYemen: after the death of
Ahmad Ibn Yahya, ruler of the Yemen, the subsequent revolu-
tion and civil war involved Egyptian-supported republicans and
Saudi-supported royalists; Egyptian troops were sent to back the
Republicans.
Effect: Soviet help needed to keep 30-50,000 troops in the
Yemen. Moral and monetary debt increases.
1966 Final break with USA, end of us wheat shipments: the
shortfall in food supplies could not be covered by hard currency
purchases in the world market.
Effect: Soviet food aid started, making Egypt dependent on
the US S R for a significant part of its supplies.
1967 June, Six-Day War, Egyptian defeat in the Sinai. Israeli
sources estimated on 20 June that eighty per cent of Egyptian
military equipment of all kinds was destroyed or captured.
Effect: as a condition for the re-equipment of Egyptian forces,
the USSR required close supervision over army training, the
selection of senior military personnel and the organization of
intelligence services.

Thus, after twelve years a limited relationship, designed to

free Egypt of its dependence on the West for arms supplies, has
escalated to a much greater degree of dependence on the US S R:
Egypt is now dependent on Soviet goodwill for arms, wheat, and
general economic aid. The Soviet navy has shore facilities in
Alexandria and Port Said and there are several hundred Russian
advisers and instructors in the Egyptian armed forces. Is this
enough to allow the Soviet Union to oppose or reverse a
coup?
The Soviet embassy in Cairo could act as a focus of counter-
coup activity; it could inspire and deploy the forces of the many
Egyptians committed to the Soviet presence; it could certainly
regulate appropriately the flow of aid supplies. Even after a coup,
the USSR could bring about a collapse of the new regime
simply by cutting off all aid. Should the regime turn to the
USA for help, this will automatically mean that the opponents
of the coup will be joined by the anti-American element as well.
This would at least severely complicate the position of its
 44 Coup d'Etat
planners, and in present circumstances it would almost cer-
tainly lead to their defeat.
When countries are in such a position of direct and material
dependence, the coup must be integrated with some immediate
post-coup foreign policy planning; if the political inspiration of
the coup is opposed to the greater power, then the coup may well
fail unless this coloration can be concealed. The few remaining
colonies, largely Spanish and Portuguese, are of course formally
in a position of total dependence. There, too, as in South Viet-
nam and East Germany, no coup can succeed in situ but would
have to be carried out in Lisbon or Madrid.
The second pre-condition of the coup is therefore:
The target state must be substantially independent and the
influence of foreign powers in its internal politica1life must be
relatively limited.
It is a cliche that countries are interdependent rather than
independent; domestic political issues have international im-
plications, while foreign political developments have domestic
repercussions. The commercial, cultural and military ties that
link countries give each country a measure of influence in the
affairs of the other; even the most powerful nations can be so
influenced. Thus, in the period preceding the US intervention
in the Second World War, British-influenced and German-
influenced political groupings and pressure groups were
operating in American domestic politics, just as today the parties
in the Middle Eastern conflict exert pressure on the State
Department through their respective lobbies.
If even a super-power can be influenced by such weak powers
then any definition of independence must be as loose as reality.*
Nevertheless some more definite guidelines can be formulated:
(a) A coup is not worth attempting if a Great Power has
significant military forces in the country concerned. If these
forces are physically remote from the political centres, and/or
if the pre-coup regime was moving towards an unfriendly posi-
tion vis-a-vis the Great Power, the rule does not hold.
*In South Korea, for example, the civil disorders and subsequent coup of
1962 do not appear to have been influenced by the significant US military
presence in the country.
 When is a Coup d'Etat Possible? 45
(b) The coup must seek the endorsement of the Great Power
if large numbers of its nationals are serving as military or
civilian 'advisers'.
The application of these guidelines excludes many otherwise
potentially suitable targets: (a) the Spanish, Portuguese and
other colonies; (b) East Germany, Hungary, Outer Mongolia,
and perhaps Poland; (c) Thailand, Laos, South Vietnam, and
Cd) some of the Mrican territories still bound by defence agree-
ments with France or the UK subject to the qualifications made
above and limited to the smaller countries.

ORGANIC UNITY

In looking at the political consequences of economic backward-

ness we saw that the crucial element in these was the concen-
tration of all power in the hands of a small elite. Conversely, in
sophisticated political units power is diffuse and therefore diffi-
cult to seize in a coup.
We now face another possible obstacle to a coup; power may
be in the hands of sectional forces, who use the government as a
'front', or regional forces whose dependence on the supposed
political centre is only theoretical.
In both these cases the problem lies in the fact that the seizure
of the supposed political centre will not win the battle; the sources
of political power may be in other centres which may be too
difficult and/or numerous to seize. In both these instances the
realities of power are in conflict with the theoretical structure of
the state, just as in the cases where the political unit is not truly
independent. Here, 'power' is within the country - but not
where it is supposed to be; not in this instance because the unit
is unsophisticated enough to diffuse power through subsidiary
entities, but because the unit is not really organic.

Sectz'onal Interests
This is the age of the giant business enterprise. The same factors
which have led to the unprecedented prosperity of the modern
industrial economy have also systematically favoured the larger
46 Coup d'Etat
business organization; mass production and mass distribution
imply large business units. Where the advantages of large-scale
production are particularly great, as in the automobile and the
chemical industries, only the very large enterprise can survive.
Elsewhere, where there is no such economic imperative, the
giant corporation has developed because of the economies of
large-scale marketing, or simply because of the natural dynamics
of accumulation. In every industrially developed economy there
are such firms: leI in Britain, General Motors in the United
States, Philips in the Netherlands and Fiat in Italy are all firms
which have been able to grow sufficiently to emerge from the rest
of industry and to become one of its focal points. This position
gives them a great deal of economic power because their mana-
gerial decisions can affect the entire national economy.
In political terms, however, the power of the giant corporation
is just one more element within the business community, and
this in tum is just one of the forces competing in the political
life of the nation. The corporation may be a giant, but it is a
giant amongst many.
It is otherwise in economically underdeveloped countries. If
the availability of mineral deposits or of particular crops leads
to the development of industry, then because of the nature of
these sectors there will be one large firm rather than many small
ones. There is by definition little or no other industry; the tax
revenues will be small - except for the company's taxes - and
there will be very few jobs going, except for the company's jobs.
If there are roads and railways, they will have been built by the
company as 'company transport facilities'; most of the schools
and hospitals will be 'company welfare services'; 'company
housing' may dwarf the capital city, and company security
guards may be better equipped than the national police.
When the state is poor and fragile the rich and well-organized
mining or plantation company will be a great power in the land,*
whether it seeks or avoids this power. In fact, it will almost
*The material equipment available to the company will, in itself, con-
stitute a considerable source of direct power: its planes, trucks, and tele-
communication facilities may well compose a major part of the country's
infrastructure.
 When is a Coup d'Etat Possible? 47
always be forced to intervene in politics if only to preserve some
status quo. When the company acts it has a wide range of dif-
ferent weapons it can use, and it can use them at many different
levels. The company can slow the flow of tax income to the state
by transferring production to some other country in which it
operates;* it can boost a particular politician by giving real or
sinecure jobs to his supporters; it can buy or bribe the press, and
generally exercise the power it derives from being very rich
amongst the very poor.
What an industrial empire can do, when set in a backward
environment, was illustrated by the Katanga secession in the
early 1960s. When Tshombe launched his independent Katanga
Republic he had only the meagre resources of a Provincial
Governor of the Congolese RepUblic. Yet as the secession
proceeded Tshombe acquired an army with jets, heavy weaponry,
armoured cars as well as well-organized propaganda bureaux in
London and New York; he was able to recruit (and pay very
handsomely) mercenary soldiers and administrators. Katanga
has only one major source of wealth: the mining industry owned
by the Union Miniere, part of the interrelated mining groups
which operate in the Copper Belt and South Africa. It does not
need a Peking propaganda pamphlet to convince us that Tshombe
was financed by the Union Miniere - and largely acting as an
agent for the company.
But even the Union Miniere was operating in what was a
relatively unfavourable environment. The Congo is a large
country, and there are other mineral deposits worked by other
companies with different interests to protect. The typical large-
scale enterprise operates in countries where it is the only major
industry. Thus ARAMCO, the·oil company working in Saudi
Arabia, is the only major industrial organization in the country.
Its 'company town', built to house employees, dwarfs other
cities in the area; its taxes constitute almost ninety per cent of
government revenue; and it has been responsible for the build-
*The risks in the mining and plantation industries are very high and
therefore most fitms must be very large and they tend to operate in several
clliferent countries; they can therefore switch production from one to the
other, drastically affecting the counuies' finances.
 48 Coup d'Etat
ing of most of the educational, transport and medical facilities in
the country. The Saudi regime has always been efficie.nt at
retaining political control over what was, until recently, a loose
coalition of tribes; the old desert warrior and founder of the
kingdom, Abdul Aziz Ibn Saud, was a past-master at controlling
the tribes and he treated ARAMCO just as another tribe.
Nevertheless, it is clear that ARAMCO is a particularly power-
ful tribe.
A standard nationalist accusation against the large-scale
foreign enterprise is that it is 'a state within a state', and that it
exercises political power, either through its direct control of the
country's government, or by using the 'leverage' of its home
country on the 'host' country. The United Fruit Company has
often been accused of exercising power through corrupt local
cliques, while the oil companies in the Middle East have been
accused of using both methods.*
A much less plausible accusation against the foreign company
is that it engages in covert activities against the state, such as
sabotage and espionage. Just why it should undertake such
activities is not explained, but such accusations are widely
believed. When the new regime of Husni al-Za'im was set up in
Syria in 1949 one of its first actions was to limit the freedom of
action of the Iraq Petroleum Company. IPC was informed that
(a) its aircraft would have to obtain official permits for each
flight; (b) that the company's security guards would have to be
replaced by public security forces, and (c) that company person-
nel would need official permits to travel in the border zones.
However unfounded the allegations of complicity in espionage
(which were the supposed reasons for the rules), it should be
noted that such restrictions (except for the last one) are common-
place in most developed countries.
Even if the foreign company has no desire to interfere in the
political life of the 'host' country, it may be forced to do so
merely in order to protect its installations and personnel.
Typically, this is the case when the company is operating in
areas which are not under the effective control of the de jure
*Joseph Conrad's novel Nostromo is a brilliant and prophetic analysis of
the causes and consequences of 'neo-colonialism '.
 When is a Coup d'Etat Possible? 49
government, especially in remote areas inhabited by minority
groups or - which may amount to the same thing - controlled
by local insurgents. The French rubber plantation companies
in South Vietnam, for example, have often been accused of
financing the Viet Cong. There is no reason to impute sinister
motives to them, because since the official government - which
also collects taxes - is unable to guarantee law and order, the
French plantation companies are simply paying their taxes to
the de facto government.
The experience of the British oil company in Persia (originally
Anglo-Persian, later renamed Anglo-Iranian and finally British
Petroleum, before becoming part of the Iranian Oil Consortium),
illustrates the case of a business enterprise which was forced to
intervene in the domestic affairs of the 'host' country under the
pressure of local political realities. Anglo-Persian received its
concession from the Persian government in 1901, but it soon
discovered that the Tehran government had very little control
over the remote areas where the company was actually exploring
and later producing oil. The Sheikh of Mohammerah controlled
the area at the head of the Persian Gulf, and the neo-Mongolian,
Bakhtiari Khans, controlled the rest of Khuzistan; both the
Sheikh and the Khans were nominally subject to the Tehran
government, but in fact independent.
The company accepted local political realities and in order to
protect the safety of its installations entered into arrangements
with the local rulers. The British government, however, tried to
regularize the situation by supporting the autonomy of the
Sheikh against the central government, and the company, being
closely associated with the government,* identified itself with
the autonomy of the Sheikh. When Reza Pahlavi took power in
Persia and restored the authority of the central government the
company found itself pemUized for its support of the Sheikh.
The relationship between the company and the Bakhtiari
Khans was even more complicated. The company realized that
its wells and pipelines could only be 'protected' by coming to
an arrangement with the local de facto power. This time, how-
*The British government bought fifty per cent of the shares of what was
to become BP - certainly the best investment of tax-payers' money ever made.
so Coup d'Etat
ever, instead of one Sheikh there were many different Khans,
all involved in the mutual conflict of tribal politics, with only
loose coalitions whose instabilitY prejudiced the security which
the company had bought. The 'na~' solution was adopted:
the company, together with British consular authorities, entered
into tribal politics in order to promote a paramount chief who
would clarify and stabilize the situation. The feuds among the
Khans, however, were never concluded, and the tribal politics
of the company were only brought to an end when the central
govermnent of Reza Pahlavi disarmed the Khans and restored its
control over the area.
Thus the company, merely in order to protect its installations
and to avoid paying double taxation to two rival authorities, had
to enter politics at three different levels. It operated in tribal
politics to promote and maintain the power of the paramount
chief of the Bakhtiari; in national politics to preserve the auto-
nomy of the Sheikh of Mohammerah against the central govern-
ment; in international politics to 'detach' the Sheikhdom from
Persia acting in association with the British consular authorities
in the Gulf.
What action ml,lst be taken by the planners of the coup in the
event of the presence of such sub-states in the target country?
In a few extreme cases their consent may be necessary: they
tend to have their ears to the ground and will probably be aware
of the coup before the official intelligence outfits. This consent
can be obtained by a suitable mixture of threats and promises,
and in this case promises do not always have to be kept. Else-
where they will act as just one more factor with which the coup
has to deal, but increasingly - after the political education they
have received at the hands of nationalist forces everywhere -
foreign business interests have learnt that neutralitY is sweet.
Regional Entities
The essence of the coup is the seizure of power within the main
decision-making centre of the state and, through this, the
acquisition of control over the nation as a whole.
We have seen that in some cases the decision-making process
is too diffused through the entire state bureaucracy and the
 When is a Coup d'Etat Possible r 51
country at large; in other cases the supposed political centre is
controlled by another, foreign centre or by sectional forces which
are independent of the whole state machinery.
A similar problem arises where power is in the hands of re-
gional or ethnic blocs, who either use the supposed political
centre as an agency for their own policies, or ignore the claims
of the centre and regard themselves as independent. Practically
every Mro-Asian state has border areas, typically mountainous,
swampy or otherwise inaccessible, which are inhabited by
minority tribes, and where the control exercised by the govern-
ment is only theoretical. Where this sort of de facto autonomy
extends to major population centres, the problem of the lack of
organic disunity arises; it is, however, of no importance for the
coup if the organic unit is in itself large - the new regime can
deal with local autonomies when it has seized power. Sometimes,
however, the local units are so powerful that they control the
centre, or else the centre rules only the immediate suburbs of the
capital city.
This was often the case in the Congo in the period 1960-64
following Independence and the mutiny of the Force Publique.
Though the Congolese Republic was constitutionally a unitary,
and not a federal, state it quickly lost control of most of the
, provinces', which behaved as totally independent entities.
Within each province local factions were in conflict" but the
central government's faction tended to be one of the weakest:

Political situation in South Kasai I960-6I

The following groupings were contending for the control of
the province:
(a) The traditional chiefs. Forces available: tribal warriors.
(b) The South Kasai separatists led by eKing' Kal6njii.
Forces available: well-equipped, undisciplined troops led by
Belgian officers. (' Ex-Belgian' Belgians.)
(c) The central government. Forces avallable:young and
inexperienced administrators with loose control over the small
national army (ANC) contingent in the eastern part of the
province.
(d) The Mining company Forminiere. Resources available:
 S2 Coup d'Etat
financial support and air transport occasionally made available
to Kalon;iist and other groups.
The situation in Katanga was even more unfavourable to the
central government; the north-east and the Stanleyville area
were in the hands of the Gizenga forces; much of the rest of the
country could not be reached by government officials because of
the breakdown in law and order, and the disruption of trans-
portation facilities. Thus a successful coup in Leopoldville (now
Kinshasa) would only have won control of a very small fraction
of the great Congolese Republic; several different coups would
have been needed in the various de facto capitals, Stanleyville,
Elizabethville, Luluaburg, etc. in order to control the whole
country.
Federal states represent the overt and constitutional recog-
nition that regions have a local power base and are therefore
granted a corresponding measure of local autonomy. In some
cases the power of the centre comes from the voluntary union of
the regions, and until the central institution develops its own
sources of power and authority it is the regions that rule, only
using the centre as an agency for their common policies.
The United States was the product of a more or less volun-
tary union of states and, until the development of Presidential
authority in the course of the nineteenth century, the govern-
ment in Washington was little more than an agency for the
common problems of the states. Thus a coup staged in Washing-
ton in, say, 1800 would have seized an empty symbol, but by
1900 the development of federal authority was such that a coup
would have led to control over much of the country. The Soviet
Union, Canada, India and West Germany are all federal states,
but the degree of autonomy of each component state or province
varies from the near zero of the USSR to the increasingly
independent Canadian provinces. The fact that, constitutionally,
Soviet republics are supposed to be fully autonomous (and even
entitled to secede from the Federation) is another example of the
perpetual contrast between theoretical structures and political
realities.
In the event, the realities of power - and of its internal
dynamics - tend to lead to a decay of the federal system. The
 When is a Coup d'Etat Possible t 53
result is either growing centralization (e.g. the USA and
USSR) or else growing separation (e.g. India, Canada and pre-
coup Nigeria).
The idea that political power should be concentrated in one
controlling centre for the nation as a whole derives from the
presumption that the interests of each region are best served
by decision-making in a national framework. This presump-
tion, interestingly enough, is usually accepted only after the
destruction of the local power structures. Thus it is agreed by
most Englishmen and Frenchmen that major political decisions
ought to be made in London and Paris, rather than on a local
level. But this intellectual recognition followed, rather than
preceded, the crushing of the 'Barons' and of the independent
states of Burgundy, Provence, Aojou and Wales.*
In many underdeveloped areas the power of local 'Barons' is
still very real, and local movements, based on linguistic or
ethnic affiliations, are actively attempting to gain either greater
autonomy or else full de facto independence. As of January
I968 the central governments of India, Burma, Kenya, Somalia,
Ethiopia and Tibet were all experiencing armed conflicts with
separatist forces. Canada, India and (as admitted, belatedly, in a
grudging footnote) France and the United Kingdom are
experiencing political conflicts with separatist elements. In
Spain, Yugoslavia and Italy more or less violent separatist
groups are operating.
Amongst all these instances, where the local population does
not accept the superiority of centralized decision-making, we
have to differentiate between the various possible implications
for the coup:
. (a) The regions are the real centres of power: the coup must
either confine itself to one region, or extend to all of them; the
supposed centre must be just one more target area. This extends
and complicates the coup, while the weakness of the coup's
forces in each single capital may invite counter-coup activity.
(b) One or two regions dominate the whole country: this was the
*The rise of Celtic separatism in Wales, Scotland, France and Spain has
been so rapid and 'dramatic that perhaps only the absence of jungles is
preventing some of their members from starting a home-grown Viet Congo
54 Coup d'Etat
situation in pre-coup Nigeria. The Northern Region, ruled by
traditional Fulani and Hausa emirs, was the largest region by
far. Its ruler, the Sardauna of Sokoto (Ahmadu Belloh), was in
full control of its internal politics, whereas in the other regions
the situation was more fluid - and more democratic. Thus,
Ahmadu Belloh, in association with political forces in one other
region, dominated the whole Federation. The young Ibo officers,
who carried out the first coup, therefore had to allocate as much
of their efforts to Belloh and his 'capital' as to the federal capital
and the federal leadership. In the event, they killed both the
federal prime minister (Abubakar Tafawa Balewa) and Belloh.
But they were over-extended, so that Ironsi, the senior officer of
the army, acting with the police and bureaucracy staged a
counter-coup and seized power on his own account.
The existence of. these regional forces, strong enough to
control the supposed centre, may make a coup impossible. If
the regional or ethnic bloc is organized on tribal lines the
structures of its leadership will be too firm and intimate for a
coup to function from within. One of the few stable countries of
the Middle East, Lebanon, is based on such an arrangement:
the Christian, Muslim and Druze blocs are all mutually hostile,
but they recognize the fact that no single group can hope to
dominate all the others. Thus, the Beirut government functions
as a common clearing house for those policies which are accepted
by each ethnic bloc. If one carried out a coup in Beirut, it would
immediately lead to the collapse of the system, since each group,
backed up by their own armed forces, would seize power in their
own region. The coup would therefore only capture Beirut and
its suburbs; and it would probably be unable to retain its control
over even that small area.
Lebanon provides an extreme example of the role of ethnicl
regional forces. In each individual instance there will be a parti-
cular balance of power between the regions, and between the
regions and the centre. The efforts of the coup would have to be
allocated so as to deal with each ethnic or regional bloc on the
basis of an estimate of its role in the particular balance of forces.
In a few cases a coup may be impossible because the nature and
extent of regional power is such as to require resources beyond
 When is a Coup d'Etat Possible f 55
those likely to be available. Elsewhere it wUl be just one more
obstacle to overcome.
The third pre-condition of the coup is therefore:
The target state must have a political centre. If there are
several centres these must be identifiable and they must be
politically, rather than ethnically: structured. If the state is
controlled by a non-politically organized unit, the coup can only
be carried out with its consent or neutrality.
'Ethnically structured' is a rather awkward phrase. It is
intended to cover social groups whose leadership is evolved by
clear-cut and well-established (usually hereditary) procedures.
If a particular traditional leadership controls the state we cannot
seize power by carrying out a coup in the state's controlling
centre, nor can we. penetrate the traditional leadership because
we would be excluded automatically as usurpers and outsiders.
In Burundi, for example, the traditional Watutsi hierarchy
controlled the state and therefore to seize power in Burundi it
would have been necessary to penetrate the hierarchy, but this
would only be possible if (a) we were Watutsi, (b) we belonged
to the aristocracy, and (c) we were next in line for the succession.
In Rwanda, power was also controlled by traditional Watutsi
chiefs who had subjected the Bahutu majority. Then there was a
revolution, and now the leadership is Bahutu-political rather
than Watutsi-traditional. A coup is therefore now possible.
If a political entity is actually controlled by a group which is
not structured politically, then obviously political methods
cannot be used to seize power. This is the case of a country
dominated by a business unit. Imagine, for example, that
General Motors did control the USA, in the sense that the
Presidency and the Congress acted as its stooges. If that were
the case, power would have to be seized in Detroit, not Washing-
ton. In the unlikely event that one could raise enough cash to
buy fifty-one per cent of General Motors' shares, then Washing-
ton would be a fringe benefit added to all the other resources
controlled. But the coup is a political weapon, and its planners
have only political resources. Thus, G M-U S A would be out-
side the reach - and the scope - of a coup.
S6 Coup d'Etat
Returning to reality, Katanga in the early 1960s, and the
Central American' banana' republics of the 1950S were examples
of states whose real 'centres' were politically impenetrable -
unless you could raise 2-300 million dollars.
3. The Strategy of the Coup d'Etat

'Dean Acheson used to tell a story about Chief

Justice Taft relating a conversation he had just
had with an eminent man about the "machinery
of government". "And you know" - Taft said
with wonder in his voice - "he really did believe
that it is machinery.'" To Move a Nation,
Roger Hilsman

'Under totalitarian conditions knowledge of the

labyrinth of transmission belts [of the machinery
of government] equals supreme power.' The
Origins 0/ Totalitarianism, Hannah Arendt

Overthrowing governments is not easy. The government will

not only be protected by the professional defences of the state -
the armed forces, the police and the security agencies - but it
will also be supported by a whole range of political forces. In a
sophisticated and democratic society these will include political
parties, sectional interests, regional, ethnic and religious
groupings. Their interaction - and mutual opposition - results
in a particular balance of forces which the government in some
way represents.* In less sophisticated societies, there may be a
narrower range of such forces but there will almost always be
some political groups which support the status quo and therefore
the government.
If those who carry out the coup appear to shatter such a power-
ful structure merely by seizing a few buildings, arresting some
political figures and 'liberating' the radio station, it is because
their crucial achievement passes unnoticed. This is the dangerous
and elaborate process by which the armed forces and the other
*The language of celestial mechanics should not obscure the inevitable
distortions which affect the balance of forces.
58 Coup d'Etat
means of coercion are neutralized, and the political forces
temporarily forced into passivity. If we were revolutionaries,
wanting to change the structure of society, our aim would be to
destroy the power of some of the political forces, and the long
and often bloody process of revolutionary attrition can achieve
this. Our purpose is, however, qUite different: we want to seize
power within the present system, and we shall only stay in power
if we embody some new status quo supported by those very forces
which a revolution may seek to destroy. Should we want to
achieve fundamental social change we can do so after we have
become the government. This is perhaps a more efficient method
(and certainly a less painful one) than that of classic revolution.
Though we will try to avoid all conflict with the 'political'
forces, some of them will almost certainly oppose a coup. But
this opposition will largely subside when we have substituted
our new status quo for the old one, and can enforce it by our
control of the state bureaucracy and security forces. This period
of transition, which comes after we have emerged into the open
and before we are vested with the authority of the state, is the
most critical phase of the coup. We shall then be carrying out the
dual task of imposing our control on the machinery of state,
while at the same time using it to impose our control on the
country at large. Any resistance to the coup in the one will
stimulate further resistance in the other and if a chain reaction
develops the coup could be defeated.
Our strategy, therefore, must be guided by two principal
considerations: the need for maximum speed in the transitional
phase, and the need to neutralize fully the forces which could
oppose us both before and immediately after the coup. If, in the
operational phase of the coup, we are at any stage delayed, then
our essential weakness will emerge: we shall probably acquire
a definite political coloration, and this in turn will lead to a
concentration of those forces which oppose the tendency we
represent (or are thought to represent). As long as the execution
of the coup is rapid, and we are cloaked in anonymity, no
particular political faction will have either a motive, or an
opportunity, to oppose us. After all, we could be their potential
allies. In any case, a delay will lose us our principal advantage:
 The Strategy of the Coup d'Etat S9
the voluntary neutrality of the 'wait and see' elements, and the
involuntary neutrality of those forces which need time to
concentrate and deploy for action.
The need for maximum speed means that the many separate
operations of the coup must be carried out almost simultaneously;
this in turn requires a large number of people. Therefore,
assuming that we start the planning of the coup with only a small
group of political associates, most of the personnel we will need
must be recruited. Further, our recruits must have the training
and equipment that will enable them to take swift and determined
action. There will usually be only one source of such recruits:
the armed forces of the state itself.
It is true that in many countries there are ethnic minorities
which, being both traditionally anti-government and warlike,
may seem to be ideal recruits for a coup. These are typically
hill-peoples, like the Druzes of Syria, the Kurds of Iraq, the
Shans of Burma and the Pathans of West Pakistan and Mghani-
stan. Because of their poverty and traditions they would prob-
ably be easy to recruit, but to do so may lead to a nationalist
reaction on the part of the majority peoples, and since the
centres of government are usually located in the majority areas,
their opposition would be a further important obstacle to us.
Another possible substitute for the subversion of the forces of
the state is the organization of a party militia. When there is a
combination of political freedom with an ineffectual main-
tenance of law and order, such militias are sometimes formed in
order to 'protect' party activities. In Weimar Germany, for
example, apart from the Brownshirts, there were party militias
of the Social Democrats, Communists and the right-wing
nationalist parties. Similar organizations - Blackshirts, Green-
shirts, Redshirts and, in the Middle East, Silvershirts - spread
in many countries in the wake of Fascist and Nazi successes.
In spite of their military bearing, their uniforms, and often
extensive weaponry, in almost every instance of confrontation
between such militias and the forces of the state, the former
were defeated. * Thus, when the Nazis tried to use the embryonic
*The conflicts between the armed Bolivian tin-miners and army which led
to the 1952 revolution are among the very few exceptions.
 60 Coup d'Etat
Brownshirts in the 1923 Munich episode, they were easily
overpowered by the police and Hitler was himself arrested.
His subsequent rise to power was achieved by political means
and not by the efforts of the Brownshirts.
In any case, in order to organize and equip a party militia
two scarce resources are needed; money and the freedom to do
so. Recruiting forces from those maintained by the state requires
neither. Therefore, while we will have to neutralize a whole
range of forces, a distinctive approach will have to be used in the
case of the means of coercion of the state. In dealing with the
armed forces, the police and the security services, we will have
to subvert some forces while neutralizing the rest; in the case of
the political forces the objective will be limited to their neutraliz-
ation.
Because of their capability for direct intervention, the armed
forces and the other means of coercion of the state must be
fully neutralized before the actual coup starts; the 'political'
forces can usually be dealt with immediately after the coup. In
some situations, however, the political forces can also have an
immediate impact on the course of events and must, therefore,
be treated on the same basis as the means of coercion of the
state.
In Russia, during the period of instability which followed the
'bourgeois' February* revolution, the railwaymen's union
emerged as a major source of direct power. Vikzhel (the AlI-
Russian Committee of the Union of Railroad Employees)
played a decisive role in the defeat of General Kornilov's putsch
by simply refusing to work the railways carrying his soldiers on
their way to Petrograd. l.ater, when Kerensky fled the city
following Lenin's October coup, and took refuge with Krasnov's
army contingent, Vikzhel threatened to call a general strike
(i.e. to leave Krasnov's troops stranded) unless Kerensky
negotiated peacefully with the Bolsheviks. Since the Bolsheviks
had no serious intention of negotiating, this amounted to a
request for unconditional surrender.
In the peculiar conditions of Russia in 1917 the railways, and
those who controlled them, were of crucial importance from the
*Old Calendar. Otherwise March and November.
 The Strategy of the Coup d'Etat 61
military point of view. Elsewhere there will be other political
forces which can exert similar pressures: in poor countries,
where the majority of city-dwellers can only buy food on a day-
to-day basis, shopkeepers - if well organized - can bring great
pressure to bear on the government by refusing to open their
shops. Where there is a strong trade-union movement, strikes
can impede the vital process of establishing the authority of the
new government immediately after the coup. Religious and
ethnic leaders can use the structures of their communities to
organize mass demonstrations against the new regime. We must
therefore identify and evaluate these political forces and, if
necessary, their leading personalities and co-ordinating bodies
must be neutralized before the coup.* Other political forces
lacking such direct power will also have to be dealt with, but this
will be part of the process of conciliation and accommodation
which follows the coup.

Neutralizing the Defences of the State

One of the outstanding features of modern states is their
extensive and diversified security system. This is a consequence
of the genetal breakdown in external security and internal
stability which has been experienced in many areas of the
world in the last two or three generations. Every state main-
tains armed forces, a police force and some form of intelligence
organization. Many states find it necessary to have paramilitary
'gendarmeries', duplicate security services, and other variations
on the theme.
In the pre-I9I4 world, states were not noticeably less aggres-
sive than in present-day international society, but the lack of off-
rail transport and a residual attachment to diplomatic convention
resulted in a certain time-span between hostility and Chostilities'.
The modem pattern of military operations, the surprise attack
and undeclared war, has as a natural consequence the Cmilitary'
peace. Instead of small professional armies, acting as cadres for
wartime expansion, many states attempt to maintain permanent
armies capable of immediate defence - and therefore offence.
*See Chapter 4, in which the neutralization of political forces is discussed.
62 Coup d'Etat
The rise of ideology-based, revolutionary parties - on both
right and left - has led to a similar expansion of internal security
forces. 'Political' branches in police forces, para-military internal
security forces, under-cover secret police outfits, have all become
common features of many states, including 'democratic' ones.
In the 1930S the United States had fewer than 300,000 men in
its armed forces; the only significant intelligence operation was
a small (and supremely efficient) code-breaking outfit, while
internal security forces were limited to a few 'T' men of the
Presidential bodyguard and a relatively 'tame' and low budget
FBI.
Today, the US Marine Corps alone has more than 280,000
men in unifonn while the entire military establishment has a
'population' which at 3,400,000, outnumbers the civilian popu-
lation of many UN member-countries.
The Intelligence community has developed into a many-
headed bureaucratic monster, composed of the C I A, the equally
important N S A, the service intelligence outfits, and many
defence 'research' institutions. In the internal security field the
FBI has tolerated no competitors, except in specialized work,
but it has grown into a mini-CIA, with several thousand agents
working in the 'political', as opposed to the criminal, function.
No state has been able to emulate such a luxuriant growth,
and even the other super-power,* the Soviet Union, has found
it impossible to keep up with the USA - in spite of the fact that
it gets some of its facilities on the cheap - a la Kim Philby.
Without being able to keep up with the USA, most states have
done their very best. Even a medium-sized country like Italy,
with no hostile neighbours of military consequence, with no
serious insurgency and with a tame and 'parliamentary' Com-
munist party, has found it necessary to develop an extensive
system, comprising two national police forces, several security
agencies, a 3OO,ooo-man army, a 40,ooo-man navy and a large
air force.
Other more embattled states enroll practically the whole
*The astute editor of Le Monde contradicted de Gaulle's assertion that
there was only one super-power - the US - by pointing out that there were
still two: the United States and the CIA.
 The Strategy of the Coup d'Etat 63
population in various kinds of defence and security forces.
Israel, surrounded by avowed enemies, with weak natural
defences, unprotected by military alliances, is an extreme case:
though it only has the population of a medium-sized city it was
able to field more than 250,000 men and women in the June 1967
war.
From the point of view of the coup, the size and power of the
armed forces, police and security agencies is both a great
obstacle and a great help. On the one hand, as Trotsky pointed
out, the improvement in weapons, transport and communica-
tions has widened the gap between organized military forces and
civilians equipped with improvised means. Trotsky noted that,
while the French mobs of 1789 could 'rush' positions defended
by infantry soldiers, in 1917 a Russian mob - however large and
determined - would be cut down by 'modem' automatic
weapons. By 'modem' he meant the clumsy and slow-firing
Maxim machine-gun; today, every single soldier on mob con-
trol could be armed with a weapon giving a similar rate of
fire.
On the other hand, the increase in the size of uniformed
forces and the 'technological revolution' have improved the
characteristics of the state security systems as a recruitment
ground for the coup. The modem army or security force is
usually too large to be a coherent social unit bound by traditional
loyalties; the need for technically-minded personnel has broken
the barriers that often limited recruitment to particular social
groups within each country. Pathans and Bedouin may be
picturesque and politically very reliable, but they are inadequate
as pilots, tank crews, or even to staff a modem police force.
The fact that the personnel of the state security system is both
numerous and diverse means that we, the planners of the coup,
will be able to infiltrate the system. In doing so, we will have the
dual task of turning a few of its component units into active
participants of the coup, while neutralizing the others. This does
not mean that we have to fight them, but merely that we have to
prevent their possible intervention against us for the limited
time-span of the coup.
Whether the purpose of our infiltration and subversion of the
64 Coup d'Etat
defences of the state is to turn the unit concerned into an active
participant of the coup, or whether it is merely defensive, the
methods to be followed will depend on the character of each
particular organization. The raw material for our efforts is the
whole spectrum of the coercive forces of the state, and as these
vary substantially in their equipment, deployment and psycholo-
gical outlook, we shall examine them separately.

NEUTRALIZING THE ARMED FORCES

In June 1967 when the Israelis, having defeated the other Arab
armies, were turning to deal with Syria's, the head of Syria's
National Revolutionary Council, Salah Jadid, kept the two best
brigades of the Syrian army in their barracks at Horns and
Damascus.* The War Minister, Hafiz Assad, begged Jadid to
be allowed to send the 5th and 70th Brigades to the front, but
Jadid - after physically assaulting him - pointed out that, though
the brigades Inight save a few square Iniles ofterrltory, their use
at the front would jeopardize the survival of the regime. The
leftist Ba'ath government was not popular with any important
section of the populationt and the two brigades were the main
supports of the regime.
Though hardly patriotic, Jadid was at least realistic. When
he had taken power in February 1966, he had done so by means
of the two crucial brigades whose officers were politically and
ethnically allied to him, and which displaced the previous strong
man, Hafez, from power when his brigades happened to be away
from Damascus - or were infiltrated by Jadid's men.
Everywhere in the world, while the number of doctors,
teachers and engineers is only increasing slowly, the numerical
strength of arInies is expanding very rapidly. It is interesting to
note that while technical improvements in, say, agriculture, have
allowed a diminishing number of farmers to produce ever larger
amounts of food, armies have needed an ever larger 'labour
*R. Atallah, 'Six jours d'irresponsabilite' , Jeune A/rique No. 343, 6 August
1967, pp. 13-IS. Also Der Spiegel, 23 October 1967.
tExcept the small 'Chinese' faction of the Ba'ath party and the Alawite
Community.
 The Strategy of the Coup d'Etat 65
force' though their productivity - or rather destructivity - per
head has also increased very rapidly. A modem platoon of thirty
men has about three times the effective fire-power of its 1945
counterpart; it is doubtful whether farming techniques have
improved to the same extent.
The effectiveness of modem soldiers, with their rapid trans-
port, reliable communications and efficient weapons, means that
even one single formation loyal to the regime could intervene
and defeat the coup if, as is likely, our forces are small and the
mass of the people and the rest of the state's forces are neutral.
Our investigation of the armed forces of the proposed target
state must, therefore, be a complete one: we cannot leave out any
force capable of intervention - however small.
Though most states have naval and air forces, as well as armies,
we shall concentrate our attention on the latter because the
procedures to be followed are usually the same for all three
services, and because - with some exceptions - only land forces
will be important from the point of view of the coup. It is, of
course, possible to use fighter-bombers to 'take out' a presi-
dential palace instead of sending a team to arrest the occupant,
and this was done in the 1963 Iraqi coup, but it is a rather extreme
way of playing the game. Although the ratio of fire-power
achieved per man subverted is very high indeed, tactical bomb-
ing of one's future capital city - and prospective post-coup
residence - is not calculated to inspire confidence in the new
government.
In certain geographical settings, however, the transport
element of naval and air forces make them even more important
than the army, as for example in the case of Indonesia. With
its population centres scattered over several large islands and
thousands of small ones, and with the very limited road facilities
on the islands, a unit of naval marines - or paratroopers - will
be more effective than some much larger army unit located in the
wrong place. When the Communist-attempted coup-cum-
revolution took place in Indonesia the military commanders
were able to use their transport potential to great advantage:
though Communist-infiltrated army units were very powerful
they were in the wrong place; while they sat in the Borneo
 66 Coup d'Etat
jungles* the anti-Communist paratroopers and marines took
over Jakarta, and the country.
Armies are divided into certain traditional formations, which
vary from country to country, such as divisions, brigades,
regiments, battalions, companies, platoons. Beyond this theo-
retical structure, however, the focus of decision-making and the
real organizational framework is usually concentrated at one or
two particular levels. It is very important for us to identify
which level is the important one, and then concentrate our
efforts on it. Table I illustrates several possible alternatives
which we may face, though in order to achieve infiltration in
depth we may in fact have to operate on many levels below the
real centre; operating above it would be pointless.
TABLE I. Formal Structures and Real Chains of Command

(a) centralized
formal structure real chain of command

GHQ GHQ
army area HQ
division
brigade
battalion battalion
company company
platoon platoon

(b) de-centralized
formal structure real chain of command

GHQ GHQ
armyareaHQ armyareaHQ
division division
brigade brigade
battalion
company company
platoon platoon

* It is ironic that ex-President-for-life Sukamo sent them there in order to

oppose the Malaysian Federation in the 'confrontation' which the new
government eventually liquidated.
 The Strategy of the Coup d'Etat 67

(c) modern NATO-style structure

formal structure real chain of command

GHQ GHQ
army area HQ annyareaHQ
division
brigade brigade
battalion
company company
platoon platoon

In (a) in the table, the operational echelon is the battalion; if

there are persons holding the rank of divisional commanders
they will probably be officers who have been eliminated from
the real chain of command, and given gaudy uniforms and
exalted ranks as a sweetener. If, in this case, we were to subvert
a brigade or divisional commander, and he then issued orders
on our behalf to the battalion, the latter - used as it is to receiving
its orders direct from G H Q - will probably query or report the
order. Thus apart from mere ineffectiveness there could also be
a further risk in operating at the wrong echelon.
In (b) where almost every echelon is operational, we can
subvert the control mechanism at almost any level and orders
given on our behalf will be obeyed at each lower level. In (c)
again we can operate at all levels except those of division and
battalion.
Though it may seem that the location of the main focus of
control and communications is an arbitrary one, in reality it
depends on very firm psychological and technical factors.
Unless the standard of training and motivation is high enough,
men have to be welded into great uniform blocs under the firm
control of their superiors, because they have neither the disci-
pline nor the capability to fight as individuals. Even highly-
motivated soldiers cannot be allowed to operate far from the
concentration of troops unless they are linked by an efficient
system of communications which enables them to receive new
orders and to report on their situations. In general, the easier
68 Coup d'Etat
the terrain, the lower the degree of discipline and efficiency, the
larger will be the unit allowed to operate independently. Con-
versely, the more sophisticated the troops and equipment, and
the closer the terrain - as in jungles or swamps - the smaller
will be the unit operating on its own.
The two extremes came face to face in the Sinai in the 1967
Arab-Israel war when the Egyptian army was organized into
three large blocs under rigid H Q control and incapable of
independent action; the Israelis, on the other hand, operated in
many small brigade-sized groups which concentrated for mass,
and separated to infiltrate in a fluid and flexible manner.
When we have determined which is the true operational
echelon in the various formations of the country concerned, we
can go on to the next stage: identifying which formations have
the capability to intervene - for or against the coup. We shall
follow two main criteria: the nature of the unit concerned, and
the location of the particular unit. These are explored in a
case study of the Portuguese armed forces.

The Portuguese Armed Forces: Situation 1967

The present regime in Portugal can fairly be described as a
partnership between the land-owning classes, the newly-
emergent industrial and business interests, and the bureaucratic
middle class (this staffs the civil service and the officer-level of
the armed forces). As in Spain, the air force and navy are staffed
by elements which are traditionally less conservative than the
army officers; as in Spain, the two services have been kept' thin'
in numbers and resources.

ARMY: The total strength is about 120,000 men, distributed as

follows (excluding administrative personnel):
I infantry division, with some medium tanks, which is partially
used as a training formation and is at about one half of its
theoretical establishment. Of the total number of men in the
unit, only about 2,000 have any transport, apart from the small
number equipped with armour. At anyone time many of the
troops will be newly conscripted, with little training or discipline.
 The Strategy of the Coup d'Etat 69
Location: central Portugal.
I infantry divin'on: this formation is usually much below
strength, with perhaps 3,000 men with some degree of training.
Transport is sufficient for perhaps half this number.
Location: northern Portugal.
Rest of the army: the largest number of troops, around
100,000, with the highest degree of training and with the best
equipment, is spread over the Mrican territories: Angola,
Mozambique and Guinea.

NAVY: Though the Portuguese have a great naval tradition, and

though the overseas 'provinces' would justify a larger navy
(which the US military assistance programme could have
partially paid for), for the reasons suggested above it has been
kept relatively weak: one destroyer, fourteen smaller combat
ships, three submarines and thirty-six other vessels. Of greater
interest to us: twelve support ships, four landing-craft, and half a
battalion of marines. Because of the distance of the Mclean
provinces, even if the navy were particularly loyal to the regime,
it could not rapidly bring over many troops from there. The
marines tend to be in distant waters and, in any case, their
number is hardly significant.

AIR FORCE: About 14,000 men. It is equipped with a variety of

old American and Italian equipment. Its 3,000 paratroopers are
stationed in the African provinces, while the transport wing
would be able to carry back to Portugal only about a thousand
men every twenty-four hours.
In the case of Portugal, therefore, though the armed forces
number about 150,000 men, only a small fraction of this total
would be relevant in the event of a coup. Most of them would
be prevented from intervening physically in the Lisbon area
because of their location and their lack of suitable transport
equipment. Others would only be able to intervene ineffec-
tually, since their training and equipment would be unsuitable.
Thus, out of the entire armed forces, only three or four bat-
talions (perhaps 4,000 men) would have an effective intervention
capability. The small size of this force reduces the possibility of
70 Coup d'Etat
the coup being defeated, but it also limits our potential area of
recruitment.
If the Air Force or the Navy did bring back to Portugal some
of the troops stationed in Mrica, by the time of their arrival we
would be the government, and they would therefore be under our
orders. If we should fail to impose our authority by then, the
coup would have failed anyway and their arrival would not
change matters. Unless, that is, we had subverted the troops in
Mrica, which would be a rather tortuous way of going about
things.
This suggests the principal criteria by which we separate out
the forces relevant to the coup, whether military or not:
The forces relevant to a coup are those whose locations and/or
equipment enables them to intervene in its locale (usually the
capital city) within the I2-24-hour time-span which precedes
the establishment of its control over the machinery of govern-
ment.
Infiltrating the Armed Forces
Our initial survey of the armed forces of the target country will
have isolated two items of information which are crucial to the
planning of the coup. These are, the nature and composition of
the units which have an intervention capability, and the real
operational echelon within them. This data is illustrated in the
notional table below:

TABLE 2. Country X: Potential Forces of Intervention

(a) battalion-size/orce
1,000 men, organized in 10 companies, with mechanical transport
and anti-tank weapons.
Location: capital city. Operational echelon: battalion HQ.

(b) division-size/orce
1,500 men, organized in 20 companies, with armoured carriers,
25 tanks.
Location: 30 kilometres from capital city. Operational echelon:
brigade HQ; tanks under separate battalion HQ.
 The Strategy of the Coup d'Etat 71

(c) brigade-sizejorce
3,000 men, organized in 3 battalions.
Locatt"on: 300 kilometres from capital city; air transport available.
Operational echelons: brigade HQ and air force squadron HQ.

Hitherto we have been thinking in terms of formal military

units, but we must now carry our analysis further in order to
identify the 'key' individuals within each particular unit. If we
were dealing with a primitive military organization we could
readily isolate those who effectively lead the unit concerned. In
the tribal-war-band, for example, there will be a few obvious
'leader' types, distinguished by their appearance and less
obviously by their descent or personal repute; the other warriors
will only be functionally different from each other because of
their individual strength or dexterity. In modern military
organizations it is otherwise: the efficiency of the organization
depends on the use of many different types of weapons and
other facilities, handled by specialized personnel. In each situa-
tion there will be an appropriate 'mix' of these and the system
therefore depends on two kinds of 'key' individuals: the
'technicians' and those who co-ordinate them, the 'leaders'.*
Our next problem, therefore, is to determine who are the
'key' individuals within those units of the armed forces which
could intervene - for or against us - during the coup. As we have
already determined which is the 'operational' echelon within
each particular formation and thus implicitly identified the
, leaders', we now turn to the identification of the technicians.'
Who these are will depend on the nature of the organization and
the task to be carried out. If, for example, during the course of
the coup the government calls on the help of force (c) in our
notional Table 2, its arrival in the capital could be prevented with
the cooperation of just one of these groups:
(i) the staff operating the communication system between the
political leadership and force (c);
(ii) the pilots and/or ground-staff of the air transport squadron;
*The 'leaders' will usually be the operational officers of the unit con-
cerned, but this need not always be so. See footnote on page 73.
 72 Coup d'Etat
(iii) the guard force at the airport/s;
(iv) the control-tower personnel at either airport, especially
in difficult flying conditions.

In general, the more sophisticated the orgaruzation, the

greater is its efficiency - but also its vulnerability. Either force
(a) or force (b) in Table 2 could for example, operate successfully
even if quite a few of its personnel were not cooperating with the
leadership. For these forces, losing the cooperation of ten per
cent of their men would mean losing approximately ten per cent
of their effectiveness; in the case offorce (c), however, the loss of
perhaps one per cent of its men could lead to a total loss of
effectiveness for some particular tasks (such as intervening in the
capital city).
This indicates that insofar as we are trying to neutralize a
formation of the armed forces, we should do so through the
cooperation of technicians rather than leaders, because the former
are both more effective individually and easier (and safer) to
recruit. The second rule is that we should, other things being
equal, choose for neutralization those units which have the
most complex organization, while choosing the simplest ones for
incorporation. This will both reduce our vulnerability from a
sudden defection and minimize tlie total number of people that
must be ultimately recruited.
Before we go on to approach and persuade the 'key' indivi-
duals to join us (thus giving us effective control of their units)
we must have collected sufficient information on the armed forces
to know:
(a) which are the military units which could intervene at the
time and place of the coup;
(b) the real command structure within the relevant units, and
who are the leaders;
(c) the technical structure of the units and who are the
technicians.
To 'incorporate' a unit we will need the active cooperation
of a number of its leaders, and in the case of a technically simple
unit the defection of some technicians will not matter greatly. If,
. in otherwise well-infiltrated units, some of the leaders should
 The Strategy of the Coup d'Etat 73
remain loyal to the pre-coup regime, this should not prove to be a
major obstacle.*
Whether we concentrate on leaders or on technicians will
depend on the particular structure of the effective forces of
intervention and on the particular political climate. If there is a
sharp political division between the troops and their officers we
may be able to incorporate units without the cooperation of any
formal leaders at all. The problem of identifying the unofficial
leaders will, however, be a very difficult one, and in any case
there is no reason to believe that we are planning the coup at a
time when such a division has hardened. The technical structures
however, are more stable, and one of our principal considera-
tions will be to avoid being dependent on too many links of the
technical chain. Table 3 (p. 74) shows our optimum strategy
in infiltrating a typical set of potential intervention forces.
Of course, in countries prone to coups, those who order these
things are aware of their vulnerability to the defection of parts
of their armed forces. It is therefore quite likely that the 'easy'
battalion No. I has been carefully chosen for its reliability and
its commanders are trusted associates of the ruling group. If
this is the case, we may have to work on battalion NO.3. What
we must not do is to rely on battalion NO.2, because the defec-
tion from our cause of even a few of its 'technicians' would have
dramatic consequences.
Until we actually start to collect information about the
individuals and to make the first approaches, we may not know
which units are· politically , tied' to the regime, and more
generally we will not know what are our ultimate recruitment
prospects in each unit. Therefore, though we will have a rough
classification in mind, dividing the units into potential allies and
potential neutrals, we should keep the distinction flexible. As we
build up a picture of the recruitment potential in each unit, we
will concentrate our efforts on the units to be incorporated; the
reliability of a unit 'allied' to the coup will be increased if we
*Officers are amazingly expendable. In both France and Russia many
officers left their units following the respective revolutions, and yet the
armies seemed to experience a sudden increase in their efficiency. Certainly
the French military record after 1789 was a great improvement on the pre-
ceding thirty years, and so was the Russian after 1917.
 74 Coup d'Etat
TABLE 3. Optimum Infiltration Strategy

Unit: Battalion NO.1 Battalion NO.2 Battalion NO.3

Command 10 company commanders and S effective 'leaders' at the
H Q of each battalion. For infiluation in depth 30
platoon commanders may have to be subverted in each
battalion.

Key men: IS-4S 'leaders' IS-4s'leaders' IS-4S 'leaders'

Technical Very simple. Very complex. Medium.
Structure: Relies on To bring the Relies on land
ordinary force to the transport, but
communication scene of the radio links
and coup, airlift and needed to
transportation sophisticated operate
equipment. communications communications.
are required.

Key men: No 'technicians' 40 'technicians' S 'technicians'

Optimum Bring a Secure the If battalion
Strategy: proportion of passive No. I proves
the 'leaders' cooperation of difficult to
over to the some of the infiluate, this
coup (unit 'technicians ' one would be
incorporated). (unit the second
neutralized). choice.

infiltrate it in depth, but there is little point in over-infiltrating

a unit which will eventually be neutralized. Every approach to
an individual will involve an element of risk; every increase in
the number of those who know that something is up, will reduce
our overall security level. We must therefore avoid over-
recruitment.
If we go up to an army officer and ask him to join in a pro-
jected coup, he will be faced - unless he is a total loyalist - with
a set of options, offering both dangers and opportunities.
The proposition could be a 'plant' of the security authorities
to determine his loyalty to the regime. The proposition could be
genuine, but part of an insecure and inefficient plot, and, finally,
 The Strategy of the Coup d'Etat 75
the proposition could come from a team that has every chance of
success.
Should the proposal be a 'plant', accepting it could lose him
his job and much more, while reporting it would gain him the
rewards of loyalty. Should it be a genuine proposal he has the
uncertain prospect of benefiting after a coup, as against the
certain prospect of benefiting immediately from reporting it. The
natural thing for him to do, therefore, is to report it.
The whole technique of the approach is designed to defeat this
logic. Apart from the rewards of being part of a successful coup
(which can be portrayed as being significantly greater than the
rewards of loyalty) there is another factor which operates in our
favour. This is that the person to whom an approach is reported
may himself be a supporter of the coup. We must therefore
emphasize these two points as much as possible, while under-
playing the risk element. But, hopefully, our potential recruits
will be motivated by some considerations beyond greed and fear,
with other interests and affiliations entering their choice: links
of friendship with the planners of the coup, and a shared political
outlook will be important, but usually the crucial considerations
will be family, clan and ethnic links with those planning the coup.
In most eCQnomically backward countries the different ethnic
groups are only imperfectly fused into one entity and mass
education and mass communications have not by any means
broken down traditional rivalries and suspicions. In any case, the
first steps towards economic progress usually reinforce these
conflicts, and we may often find that ethnic links are far more
important than more recent political affiliations.
For example, when no steel mills were being built there could
be no regional conflicts on where to build them; when civil-
service jobs were all given to citizens of the imperial power
there could be little conflict between ethnic groups on the 'fair'
allocation of jobs. Conflicts over jobs or the location of steel
mills are necessarily more intense than the old conflicts over land:
while before, only the geographical fringes of the tribe were in
contact with the rival, now each tribe fights the other on the
national stage. While a conflict over land can reach a compro-
mise at some middle line, a steel mill has to be located in either
76 Coup d'Etat
area A or area B. (The alternative, of course, is to put it on the
border of the two provinces; although this is usually far from
roads and other facilities, it is sometimes done.)*
As old conflicts widen in scope and intensity, the instinctual
solidarity of the ethnic groups hardens. Mrican 'tribalism' is
merely an extreme case of a very general phenomenon "" for
example, sophisticated and utterly unreligious Jews will 'happen'
to marry other Jews, though they may regard themselves as
thoroughly assimilated. Despite Czech and Slovak protestations
of national unity, capital investment has had to be assigned
carefully to each area on an exact percentage basis, and conflict
over this was one of the factors which brought down the Novotny
(' the great survivalist ') government in 1968. In fact, all over
Eastern Europe the old conflicts are just below the surface, and
the new' socialist-national' policies are inevitably reviving them.t
In Rumania almost half a million Germans and a million and a
half Hungarians feel that they are not getting a fair d~; while
in Yugoslavia, Croats, Serbs, Dalmatians and Macedonians are
all involved in the ethnic balancing act, not to speak of the
smaller Albanian, Vlach and Slovene groups; In many areas,
ethnic divisions are complicated by a superimposed religious
conflict. The Ibo nation in Nigeria, for example, has been in
endemic conflict with the Muslim northerners for a very long
time, but the introduction of Christianity among them has
meant that the old Ibo/Hausa conflict has been intensified by a
new Muslim/Christian one.
We will therefore make the fullest use of the 'ethnic matrix'
without, however, aligning our coup with any particular ethnic
faction. In terms of petty tactics, we will match each potential
recruit with a recruiter who shares his affiliation and, if neces-
sary, the image of the coup will be presented in a similar vein.
*The problem is compounded by the fact that development programmes
are usually focused around one or two big projects, which attracts much of
the country's attention - and investment funds. The 'donor' countries
usually resist the fragmentation of industrial projects to appease local
feelings and this further complicates the political problem.
tThe 'Zionist plot' in Poland in early 1968 and the perpetual conflict
between Hungary and Rumania over the Transylvanian provinces are
particular examples of a general problem.
 The Strategy of the Coup d'Etat 77
But we must also take account of a special factor which is a
typical post-colonial phenomenon. Colonial regimes developed
the habit of recruiting army personnel among minority ethnic
groups, which were reputed to be more warlike and, more
important, could be trusted to join in the repression of the
majority group with enthusiasm. Mter independence, these
minorities naturally regressed in terms of political power and
social position, but they still staffed much of the armed forces.
This has led to the strange spectacle of minorities acting as the
official protectors of the regime which is putting the pressure on
them.
The Druzes and Alawites of Syria have been in this position
since the French departed in I945, and it is hardly surprising
that disaffected officers of the two groups have played a promi-
nent role in most of the many coups since independence.
In many parts of Africa the majority peoples are the reputedly
'soft' coastal tribes,* who have captured the political leadership
because of superior numbers and education, while much of the
army is made up of members of the smaller hill-tribes. This is
the result of the superficial ethnographic theory that the British
learnt in India and the French in Algeria~ but which, in African
conditions, was little less than absurd. As soon as the officers of
the colonial country landed in a new territory, they set about
finding the hills and, once there, tried to recreate their semi-
homosexualizing relationship with the 'wily Pathan' or 'Ie fier
Kabyle', by recruiting the supposedly 'tough' hill-men into the
army.
Without setting the stage for an inter-tribal civil war, there is
every incentive to make use of this factor but, to the extent that
there is an effective political life, the ideological outlook of the
potential recruit will also be important. As far as we are con-
cerned, combining all ranges of the political spectrum against a
right or left extreme will give the most suitable political 'cover'
to our coup. The Qassem regime in Iraq, which lasted for five
years as a pure balancing act, was finally brought down in I963
*Nigeria is the exception, where the coastal nations are much more
developed but fewer in numbers than the inland Hausas, and that is the
cause of the trouble.
T ABLB 4. The Role of Ethnic Minorities in Syrian Politics

TheDruzes
I949, April:
The first post-colonial regime of President Quwatli tries (and fails) to
destroy the power base of a major Druze clan. This was one of the
factors which led to the pioneering coup of Husni al-Za'im (the first
military dictator in the Arab world).
1949, August:
Husni al-Za'im overthrown by a group of officers, of whom many are
Druzes; this followed the attempt to intimidate the DruzeJabal area.
The crucial armoured-unit commanders were Druzes whose co-
operation had been enlisted by the planners of the coup.
I949J December:
The new regime starts its attempt to unite Syria with Iraq, and a new
coup is planned to overthrow it and stop the union. Druze officers of
the armoured unit carry out the coup, which leads to Shishakli's
military dictatorship.
I954, February:
Shishakli's regime overthrown. This was preceded by his military
occupation of the Jabal Druze area and his arrest of a Druze delega-
tion, which led to disturbances and reprisals. The group which carried
out the coup was composed of three factions, of which the Druze was
perhaps the most important.

The Alawite$
1966, February:
Coup by the leftist Ba'ath against the rightist Ba'ath regime of Hafiz
and the party founders, M. Afiak and S. Bitar. The coup was supposed-
ly based on an ideological rift within the Ba'ath movement. In fact,
the government of the leftist Ba'ath was a cover for a group of Alawite
officers headed by Salah Jadid, himself an Alawite.
I967, February:
The Chief of Staff', a Sunni Muslim, is replaced by an AIawite;
political power retained by the Alawite-controlled National Revolu-
tionary Council, with Sunni and Christian Arab ministers as figure-
heads.
 The Strategy of the Coup d'Etat 79
when the moderate nationalist, Aref (Abdel Salam), persuaded
all political factions, from left-wing Ba'ath to right-wing con-
servatives, to combine against the supposed Communist pene-
tration in the government.*
If there is no extreme faction available, however, we will have
to be content with the petty tactics of claiming political kinship
with potential recruits. But apart from the virtues of honesty
there is a need for consistency and a systematic presentation of
the coup in terms of divergent political 'lines' may eventually
lead to our undoing.
Finding out the ethnic group to which a particular officer
belongs is relatively easy; finding out what is his political out-
look is rather more difficult. But the hardest thing of all will be
to find out if he is personally alienated from the higher military
leadership. Only the family and the closest friends of an officer
will know whether he feels that his superiors are treating him
unfairly, or running things badly, to the extent that he would
welcome a radical change in the whole set-up. Unless we have a
direct line to the individual concerned, we will have to use out-
side information to determine his inner feelings.
A standard intelligence procedure is to follow the career
pattern of officers, in order to find out which ones have been
passed over for promotions, assuming - other things being equal
- that they will make good prospects for recrcitment. In many
countries, promotions within the armed forces are announced
in official gazettes, and starting from a particular class at the
military academy, one can follow the career of each officer from
their graduation to the present. In some countries, where pro-
motions are not published (for security reasons), one can carry
out the exercise by using back-copies of the telephone directory
where their names will be printed along with their changing
ranks. Where, as in the Soviet Union, neither telephone direc-
tories nor official gazettes are good sources of information, we
could use more desperate expedients: getting an old boy from
*One of the danger signs was the fact that Qassem started calling his
opponents 'fascist Hitlerites'. Adolf Hitler is a popular figure with most
shades of Arab opinion and only an unthinking transposition of Soviet habits
could have led to the use of this epithet.
80 Coup d'Etat
the relevant years to circulate proposals for a reunion, or build-
ing up mini-biographies from personal acquaintances; by what-
ever means, our aim would be to trace a reasonably accurate
career history for each graduating class from the military
academy. The competitive position of each officer will be
established vis-a-vis others of his year, rather than the other
officers of the form.ation in which he serves, and Table 5 pre-
sents the infonnation in the appropriate framework:

TABLE S. Class 0/19 - Military Academy o/Country X:

Present Career Position

lieutenant 7
captain SS
major 33
colonel. • IS
brigadier • 2
deceased or civilian • IS

The seven lieutenants will probably make eager recruits for

anything that will disturb - and rearrange - the order, but their
low rank may be a correct assessment of their abilities, in which
case their 'help' may be a liability. More generally, and more
usefully, we will know that the captains and majors in our table
may well be less enthusiastic about the set-up than the colonels,*
while the two brigadiers - if not actually appointed for their
political reliability - have probably become staunch supporters
of whoever gave them their exalted jobs.
Ethnic affiliation, poiitical outlook, and career patterns will all
serve as guides to the likely reaction of the potential recruit when
the approach is made. There are, however, two points that we
have to bear in mind, the first organizational, and the second
*Of course, colonels have always been prominent in military coups, but
these have been coups which they have initiated on their own behalf. Our
purpose is to use army officers, and captains are less likely to take the coup
out of our hands than more senior officers.
 The Strategy of the Coup d'Etat 81
deeply human. While alienated personnel will make good
recruits, we must remember that we need people who will not
only cooperate personally, as in the case of the technicians, but
also bring the units they command over to the coup. Thus, while
the leaders we recruit could (and should) be estranged from the
superior hierarchy, they must not be 'outsider' figures who are
not trusted by their fellow-officers and men. There will often be
a danger of attracting the inefficient, the unpopular, the corrupt,
as well as the disaffected. If we allow our coup to be assisted by
such men we will be endangering the security of the coup and
discouraging the recruitment of the better elements, and - most
important of all- we may find that our 'leader' recruits will fail
to bring their units with them.
The other.point to bear in mind is the basic unpredictability
of human behaviour. We have so far been trying to establish
which links could override the loyalty of army personnel to their
superiors and, of these affiliations, the strongest may be expected
to be a family link. We should not, however, place total reliance
on this factor. Although there is an Arab proverb: 'I and my
brother against my cousin; I and my cousin against the world',
we should remember the Aref family history in Iraq between
1958 and 1968 (Table 6).
The relationship between the brothers illustrates the diffi-
culty of predicting human behaviour. Between 1958 and 1962
one brother was in prison under a suspended death sentence,
while the other was in charge of a force that could probably have
moved on the capital at any time. The Ba'ath leaders, mindful
of this precedent, allowed Abd-el-Rabman to remain in charge of
the important armoured units near Baghdad, and this was their
undoing. There was a period, immediately after the first coup of
1963, when the position of the Presidential brother was weak and
the Ba'ath party militia, totally untrained but heavily armed,
could have been used to remove the military brother from his
command. The Ba'ath leaders, however, assumed that Abd-el-
Rahman would not collaborate with his brother and would be-
have as he did in 1958 and 1959"-62. This time he behaved
differently, in spite of the fact that he was helping a brother who
needed help much less badly than in 1958-62, when he was a
82 Coup d'Etat
TABLE 6. The Are! Brothers in Iraq, 1958-68: A Study in Loyalty

The recent ruler ofIraq, President Abd-el-Rahman Aref,* was chosen

as a compromise candidate by the army after the accidental death of
his brother, Abd-el-Salam, the previous dictator of Iraq in April 1966.
The career pattern of the two brothers shows that, while both were
prominent army leaders, one did not always cooperate with the other:

Abd-el-Salam Abd-el-Rahman
July 1958: Co-author of the Unaware of the plans
Coup overthrows coup with Qassem. and only intervenes at
the monarchy. the end - though
commander of an
important armoured
unit.
November 1958: Qassemarrests Promoted, and placed
Abd-el-Salam. in charge of a large
Accused of treason and army contingent.
given a (remitted)
death sentence.
1962: Placed in retirement.
February 1963: Released and made Placed in charge of
Ba'ath coup. President. the 5th armoured
Qassem deposed division, promoted to
and shot. brigadier-general.
November 1963: Planned by brothers together.
Anti-Ba'ath coup. Assumes full control. Promoted.
April 1966: Dies. Emerges as com-
promise Presidential
candidate of the army.

captive and under a death sentence (or perhaps because of this).

Despite such instances of human unpredictability, and bearing
in mind the individuality of our prospective recruits we can
nevertheless use the information we have collected to rank the
leaders in terms of their probable response.
Having established the career histories and ethnic and political
affiliations of possible recruits, we can proceed to weigh our
*Abd-el-Rahman Aref has since been overthrown by a coup d'etat.
 The Strategy of the Coup d'Etat 83
prospects as illustrated in Table 7. In evaluating the information
we must, of course, bear in mind that the importance to be
attached to each factor will differ from one environment to

TABLE 7. Battalion No. I: Recruitment Prospects, see Table 3 (p. 74)

Recruitment prospects of fifteen officers of higher rank. Experience,

opinions and affiliations from the point of view of the coup. (Repeat for
thirty platoon commanders.)
x = favourable n = unfavourable 0 = unknown
xx = very favourable nn = very unfavourable

Officer Political Ethnic Career Approach

No. Views affiliations pattern Yes No Doubtful
I 0 x xx V
2 n xx n V
3 xx XX xx V
4 nn x n V
5 nn nn nn V
6 X x x v'
7 0 XX nn V
8 0 n XX V
9 n XX n
10 o· nn XX V
II X n 0
12 X X X V
13 n n n
14 XX xx x v'
IS 0 n XX v'

another: in Latin America, for example, social background

would also have to be added, while in Western Europe and
North America political allegiance would be paramount -
ethnic affiliation would be of little importance, but social back-
ground would carry some weight.
Thus, out of fifteen potential recruits we see that NO.3 is the
only totally good prospect from the point of view of the factors
here taken into consideration; NO.5 is a totally bad one, and
probably dangerous to approach at all; the others, however, will
be somewhere in the middle.
84 Coup d'Etat
Once we have repeated the procedure followed in the case of
battalion No. I and we have covered all the other formations of
the armed forces (or at any rate those with an effective inter-
vention capability), we will know the overall recruitment
prospects of each unit, and within them, of each individual. We
will never be able to achieve a 100 per cent coverage; in some
cases, where the armed forces are very large in relation to our
resources, or frequently redeployed, our coverage may be very
incomplete.
This will not matter greatly if the 'unknown' units can be
neutralized technically. If, however, their intervention capability
does not depend on elaborate and vulnerable facilities, then the
coup may be jeopardized. We do not, however, depend on the
incorporation and the neutralization procedures alone, and we
will also be able to isolate physically those units which appear on
the scene unexpectedly and those which we have not been able
to infiltrate at all. Before looking at the problems involved in the
third, and least desirable, of our methods of dealing with armed
opposition, we must turn our attention to the subversion of
individuals in the units where we do have the requisite informa-
tion.
As soon as we emerge from the close security of the planning
and information stage, the danger factor in our activities will
increase very sharply. As we have pointed out earlier, every
single individual we approach will be a potential informer who
by telling the authorities about our efforts could lead to the
collapse of the coup. The most dangerous person to approach
will be the first in each particlllar formation, because until we
have his cooperation we will not have a really intimate source of
information about the unit and its members. Our first recruit
must, therefore, be a long-standing member of that particular
formation and, if at all possible, he must be a senior officer - or
even the commander. Once we have chosen our man, the first
step will be to arrange a meeting and to 'sound' him out jn
vague and generalized terms about the 'possibilities of achieving
political reform'. These soundings must be conducted by a man,
or men, who fulfils certain exacting qualifications: he or she must
be a trusted associate of high calibre without, however, being
 The Strategy of the Coup d'Etat 85
in the inner group planning the coup. In other words he must be
both valuable and expendable. This is an ideal which we can
only try to approximate, but it could be fatal to expose a member
of the inner group to the possibility of being betrayed to the
authorities. In the coup country· par excellence, Syria, political
leaders have in fact gone round the barracks 'canvassing' for
(armed) support, but the special conditions of Syrian political
life are not likely to be reproduced elsewhere.
Once the potential recruit has been brought to the state when
the possibility of a coup has been openly discussed, he should
be told three things about the coup: (a) the political aim; (b) that
we have already' recruited' other individuals and units and (c)
the nature of the task that he will be asked to perform. Every-
thing we say, or arrange to be said, to the potential recruit will
have to be studied carefully, and we will work on the assumption
that every recruit may be a 'double' working for the security
services.
We will not, of course, identify our coup with any particular
party (whose policies would be known) nor with any political
faction (whose leading personalities will be known). We will,
instead, state the aim of the coup in terms of a political attitude
rather than in terms of policies or personalities, because the
latter are necessarily more specific and therefore liable to specific
opposition. The attitude which we will project will have to be
calculated carefully: it should reflect the pre-occupations of the
target country, implying a solution to the problems which are
felt to exist, and inform it must reflect the general political beliefs
of the majority of its people. Thus, in Britain, we could speak of
'the need for more business-like government' - one can even
imply (whether truthfully or not) that the coup is linked with
prominent public figures such as newspaper proprietors, big-
business men or the chairman of a nationalized industry. In
Latin America the attitude presented may, for example, imply
that the 'sacred trust of the armed forces' requires intervention
to 'clear the mess made by the politicians' in order to achieve
'social/national progress, while respecting property rights!
individual rights'.
If the pre-coup government is itself the product of a seizure of
86 Coup d'Etat
power, then the aims of the coup can be presented purely in terms
of restoring 'normal political life', or, if we are outre leftists we
can speak about 'the need to restore Democracy'.
Making up slogans may seem to be an easy game, but in fact
our slogans will have to be calculated carefully to satisfy a
political optimum. We must, for example, avoid being specific,
but if the attitude we present is too general it will stimulate the
suspicions of the shrewder of our listeners, while failing to fire
the enthusiasm of the more idealistic ones. We must also
remember that the armed forces of many countries are often
politically, and psychologically, out of tune with civilian society,
and that they could have distinct - and perhaps antagonistic -
preoccupations and beliefs. As citizens, army officers may share
the belief that there ought to be economies in government
expenditure, but at the same time feel that the armed forces are
being starved of funds. Wh~re the social status of military
personnel has suffered a decline because of defeat in battle,
or just a long peace, we will always emphasize the need 'to
restore the defenders of society to their proper place within
it ...'
In presenting the aims of the coup to potential recruits we
should exercise a measure of flexibility in order to reach a 'good
fit' with what we know to be their beliefs: we cannot, however,
run the risk of being exposed as being grossly inconsistent.
Whether we hold the views that will make up our image does not
matter at all as long as the other conditions are satisfied. It is,
incidentally, polite to indicate that the coup is only being carried
out with extreme reluctance, and that we appreciate that this
reluctance is shared by our recruit.
Once the idea of the coup has gained a measure of acceptance
in the mind of our potential recruit, we should define the coup
in terms of his role within it. This will not imply that we will
reveal any of the operational detail, but we should make it quite
clear that:

(a) his role will be limited to a few specific actions;

(b) almost everybody in his unit is already with us, and
(c) therefore his role will be a safe one.
 The Strategy of the Coup d'Etat 87
When, and only when, the recruit becomes actual, rather than
potential, we can reveal to him the nature of his actual task.
This will be described in the greatest possible detail, but not so
as to enable the recruit to work out the implications of the task
he is asked to perform. If, for example, the recruit in question
is destined to use his unit to· provide 'muscle' for a roadblock
team, he will be told what equipment his men should have, how
many will be required and how he will receive the go-signal. He
will not be told the date of the coup, the place where the road-
block will be, or what the other teams will be doing.
Information is the greatest asset we have, and much of our
advantage in the planning stage will derive from the fact that,
while we know a great deal about the defences of the state, those
who control them know very little about us. We must therefore
make every effort to avoid giving any information beyond what
is actually required. In any case, while a recruit may feel that he
ought to know more about the coup before he agrees to partici-
pate in it, he will also feel more secure if we show concretely
that the operation is being run with great caution, and is there-
fore secure.
After the first few recruits in each unit have been made, the
others in it will be much easier to persuade; there will also be
more people to do the persuading, because this is the purpose
to which we will put our first recruits, in the interval between
their initial recruitment and the actual coup. Also, a 'snowball'
or, hopefully, an 'avalanche' effect will be generated by the
first recruits, who will gradually create a climate in which it will
be easy to recruit further.
After the approach and persuasion of the 'key' individuals has
begun to give its results, we will be able to identify the units
which will eventually be used as active participants in the coup.
These will be a small part of the armed forces as a whole but
hopefully, the only part that will be able to play an active role at
the time and place of the coup. We will concentrate our further
efforts on them because their infiltration in depth will be of
value to us, whereas the' over-neutralization' of the other forces
will merely involve further risk. Ideally, we will have neutralized
all those formations which we have not incorporated, but this is
88 Coup d'Etat
not likely to be the case. The methods that we will follow to
'isolate' those formations that we have not been able to penetrate
will be discussed in Chapter 4.
The degree of success required of our infiltration programme
before we can proceed to the operational phase will depend on
the military, political and geographical factors involved; the
same degree of penetration may ensure success in one country
while being inadequate in another. In our Portuguese example,
because of the extensive deployment of the active troops in the
remote African 'provinces', and the lack of training and
mechanization of the troops stationed in Portugal, we could go
ahead with minimal penetration (Table 8).

TABLE 8. Infiltration of the Armed Forces in Portugal (notional)

Total armed forces (Army, Navy and Air Force) 150 ,000
Incorporated as active participants: 3,000
Neutralized by the subversion of 'key' technicians: 12,000
Neutralized by unsuitable training and equipment: 45,000
Neutralized by their location: Angola 45,000
Mozambique 25,000
Portuguese Guinea 20,000

150,000

This is an extreme example of a small and poor country which

is trying to retain its African empire to the bitter end, and is
therefore only leaving a very small force in its own metropolitan
territory. The degree of incorporation achieved here is only
about two per cent, and yet the coup would not find any military
opposition in its way, unless it failed to impose its authority
within the time-span required to bring into Lisbon the troops
stationed in the African provinces. (The fact that the present
regime is far from universally popular would reinforce the
favourable 'military' factors.)
If, however, we take the case of a developed country with
good transport links and with no overseas commitments for its
troops, the same pelicentages of incorporation and active neutra-
 The Strategy of the Coup d'Etat 89
lization which in the Portuguese case would guarantee success
would lead to certain failure, as illustrated in Table 9.

TAB'LB 9. Infiltration a/the Armed Forces in West Germany (notional)

Total armed forces (Army, Navy and Air Force) 450,000

Incorporated as active participants: 9,000
Neutralized by the subversion of 'key' technicians: 40,000*
Neutralized by unsuitable equipment (mainly Air
Force and Navy): 180,000
Balance of forces under the control of the Government: 221,000

450,000

Since there is nothing that we can do to prevent the large

forces capable of intervention from doing so, we would almost
certainly fail- unless we were the higher leadership of the armed
forces.
Most situations will be between the two extremes, with a
small percentage of the armed forces incorporated, a larger
percentage neutralized by our efforts, and a very small percen-
tage to be 'isolated' by severing from the outside its communi-
cation and transport facilities. Apart from the military forces, the
government will also be defended by police forces and their
para-Inilitary extensions, and we now turn to the problem of
their neutralization.

NEUTRALIZING THE POLICE

The flags and uniforms of the military forces of different coun-

tries are very different but their structure and organization tend
to be very similar, since they reflect the universality of modern
technology. The tactical implications of weapons and ancillary
equipment impose a certain uniformity on military organization
*In a densely populated area, with extensive civilian telecommunication
facilities and a highly developed transport system, even this figure could
only be reached with very great efforts.
90 Coup d'Etat
and this has enabled us to study their infiltration in terms which
are generally applicable.
Police forces, however, are shaped by the social and political
conditions of the society they serve and are therefore very
diverse. Policemen can be armed very heavily or not at all;
they can be concentrated in mobile and hard-hitting units, or
dispersed in small groups; they can be controlled by the Ministry
of Defence, and thus have a military training and outlook, or by
the local community, and be extremely civilian-minded.
Though their structure is so diverse, police forces resemble
each other in the purposes they serve: the prevention and the
detection of crime* and the maintenance of public order. The
criminal side of police work involves the country-wide network
of police stations supported by special detective units in the
larger centres. The maintenance of public order, however, is
carried out by separate para-military forces or, where there are
no such forces, by concentrating and deploying ordinary police-
men taken from their other work. Police work also includes an
intelligence element. Information is gathered informally by the
entire police apparatus (and their informers), but there will
usually be a special section of the police whose only function is
in this area. The intelligence aspect of police work will
be effectively neutralized by our general defensive effort,
vis-a-vis the security services, which is discussed in the next
section.
Para-military forces do not exist in Britain, where there are
provisions for the army to act in support of the civil power, but
they are extensively employed in many other countries. In
France, for example, there are two police forces - the Surete
Natz"onale and the Prefecture de Police - but there is also a para-
military force which normally acts as the rural police, the
Gendarmerie.t
The Gendarmerie is controlled by the Ministry of Defence and

*By crime is meant an infraction of the laws of the land, and this means
different things in different countries - e.g. South African residential and
Soviet pUblication laws.
t In the spring of 1968 the Prefecture de Police was administratively merged
with the SUrete.
 The Strategy of the Coup d'Etat 91
its officers are integrated in the ranks of the armed forces, while
the men receive a light infantry training as well as their police
training. It numbers about 63,000 men, and is organized into
departmental forces which are scattered in small groups all over
the countryside and into 'mobile' groups which are concentrated
in large units (Legions). We can ignore the departmental forces
because they would probably be unable to intervene within the
short time-span of a coup: but the mobile units, each of which
consists of seven squadrons of trucked gendarmes and one
armoured car squadron, represent a very formidable force that
would have to be neutralized or isolated.
The mobile Gendarmerie live in military-type barracks and
are equipped with sub-machine-guns and heavier infantry
weapons; their armoured cars (13-ton wheeled vehicles with
40-mm armour) can only be stopped with standard anti-tank
weaponry. Officially, the Gendarmerie - unlike the other two
police forces - has no intelligence service; but during the
Algerian war a security section was set up and, as bureaucratic
organizations often do, has survived the demise of its original
function.
The Surete Nationale, which carries out police work in popu-
lation centres of more than 10,000 inhabitants (except for Paris
and its suburbs), is largely composed of C I D men and dispersed
policemen, but it also has a para-military force. This is the
Compagnie Republicaine de Securite (CRS). It numbers about
13,500 men, and they are trained and equipped in a manner like
that of the mobile units of the Gendarmerie - minus the armoured
cars. The eRS is staffed with personnel which has been care-
fully screened politically and it is headed by an assistant director
of the Ministry of the Interior. The Surete has an intelligence
service which largely concentrates on the more sophisticated
forms of crime, and a counter-intelligence service which also
carries out 'political' work and the surveillance of aliens. Both
intelligence organizations operate all over France, including
Paris,* unlike the rest of the Surete.
All police work in the Department de la Seine (the Paris area)
*But, in Paris, the jurisdiction of the Suret4's intelligence service is officially
limited to the railway stations.
92 Coup d'Etat
is the exclusive province of the Prefecture de Police, which has
been made internationally famous by one of its fictional inspec-
tors, Maigret. The Prefecture has influenced the organization of
police forces in many countries in southern Europe and the
Middle East, and we will study it in greater detail than the other
French police forces.

Anatomy of a Police Force: The Paris Prefecture

Hopefully, the police of the capital city which is the locale of
the coup will be less powerful than this force. It consists of about
24,000 men and is organized in several directorates, of which the
following concern us directly:
(a) The Police Municipale is the largest directorate and controls
the familiar uniformed flics, with their largely symbolic pistols
and their much-used truncheons. They are dispersed in twenty
district stations in .the city and twenty-six suburban ones; their
standards of training and discipline have been improved under
the Vth Republic, but their capacity for individual brutality does
not add up to effective intervention units. In the event of a major
disturbance they are deployed in columns of civilian-type buses
which could be stopped by suitable roadblocks; their training
and mentality will probably make them 'neutrals' if we can
prevent their concentrated deployment.
(b) The Police Judiciaire is the Paris CID and one of the
pioneers of scientific detection. Apart from the incidental in-
telligence aspect of their work, we can ignore this directorate.
(c) The Intelligence service, like its counterpart in the Surete, is
mainly concerned with sophisticated crime: drugs, vice and high-
class gambling. But it also has a political section which carries
out surveillance work. As in the case of other security agencies,
we will cover the appropriate defensive tactics in the next
section.
(d) Aliens'directorate is a small group, mainly concerned with
the bureaucratic routines of issuing and checking residence
permits. It exercises general surveillance over transient foreigners
(the fiches you fill in at the hotel are collected by this directorate),
and over foreign communities. Its work will only affect us if we
 The Strategy of the Coup d'Etat 93
have some connexion with foreign elements, and particularly
with those foreign communities which have a history of political
activity in its more violent forms.
(e) Safety of the President. This directorate is concerned with
the physical protection of the President, but it also carries out
a preventive intelligence function. Following the repeated
assassination attempts organized by the 0 A S and its affiliated
organizations, this section of the Prefecture was reinforced with
carefully screened personnel taken from the entire security
apparatus. Their security system at the Elysee Palace would be
a serious obstacle to its seizure during a coup.
(I) Garde Republicaine. Though controlled by the Prefecture,
this is part of the Gendarmerie and is equipped with light in-
fantry weapons and a variety of transport equipment. It provides
the horsed, helmeted and plumed Presidential guard on cere-
monial occasions, but its two regiments are hard-hitting mobile
forces whose neutralization would be an essential requirement in
the event of a coup.

The existence of separate police organizations is one of the

problems of neutralizing this part of the state security apparatus.
In Britain, the division is largely territorial and its purpose is to
give the local interest a measure of control over the police force,
but there are also specialized forces which reflect functional
divisions. Apart from the county-based police (now being
amalgamated into larger groups) there are the following inde-
pendent police forces:

Admiralty constabulary
Air Ministry constabulary
Atomic Energy Authority constabulary
Five independent harbour police forces
British Transport Commission police
Civil aviation constabulary
War Department constabulary

All these police forces are strictly confined in their operations

to the installations they protect, but similar organizations in
 94 Coup d'Etat
other countries, where bureaucratic propensities are subject to
weaker controls, have shown a remarkable ability to grow and
diversify.
Though the French police system is particularly extensive,
its basic features are shared by police forces in most of Africa,
Asia and the Middle East. The para-military element is usually
present in the form of a 'field force' attached to the regular
police, or else in the form of armoured car units. The riot-
control element is reproduced in the special squads of Middle
Eastern police forces which, like Beirut's Squad 16, can be very
effective in spite of their small size. Where, as in most parts of
Asia, a serious insurgency situation has been experienced, this
common pattern has been distorted by the proliferation of ad hoc
police forces which carry out combined internal security
and administrative functions. South Vietnam is the extreme
example of this trend, with no fewer than five different
formations.*
If the British police system can be said to be divided into
largely territorial units, and the French one into largely func-
tional ones, in the United States the division is largely con-
stitutional. Except for the specialized work of the police agencies
attached to various departments of the federal government, only
the FBI has a nation-wide jurisdiction and then only for certain
crimes legally defined as 'federal'. Most ordinary police work is
carried out by purely independent local forces maintained at the
municipal, county or state level. The fragmentationt of the
system means that the police as such would have a very limited
intervention potential, in spite of its extensive stock of weaponry
and communication equipment. There is, of course, the National
Guard, but this has not so far been organized in a manner that
would give it a real intervention capability, as Was shown by the
*Regional Forces (about 100,000 men), Popular Forces (about 30,000
men), Civilian Irregular Defence Groups (about 20,000 men), the regular
police, and the supposedly elite Police Field Force. .
tOf course the fragmentation of the police in the United States has
largely resulted from the deliberate intention of denying the federal govern-
ment a possible instrument of tyranny. Nowadays, however, the main
effect of the system is to impede efficient police work, viz. the case of the
'Boston Strangler'.
 The Strategy of the Coup d'Etat 95
events of the summer of 1967 when the Guard failed to perform
efficiently, even against untrained civilians.
The strategy of the coup, with respect to the police system of
our target country will, therefore, have to be as diversified as its
component parts:

The Para-Military Element

Para-military forces are usually able to perform a military as well
as a police function. This versatility has resulted in their rapid
growth, partly because they may be a genuinely economical
way of improving the security system in general, and partly
because funds are often easier to secure for them than for the
regular police. An opposition party, or public opinion, which
may resist an increase in the police budget can often be per-
suaded to allocate funds for the Ministry of Defence, and para-
military forces are usually under its administrative control. In
the newly-independent countries the para-military element of
the police can be a very serious obstacle to the coup because,
while the army is often a recent post-colonial development, the
police - and its para-military units - are usually old-established
organizations. This means that the police can be larger than the
army, and also sometimes superior in the quality of training and
equipment. If this is the case, it will not be possible to control
the para-military units by using that part of the army which we
have 'incorporated' against them.
Fortunately, the governments of newly-independent countries
are making every effort to increase the size of their armed forces
and thus this unfavourable (for us) balance of strength between
the army and the para-military police is usually reversed a few
years after independence. This is perhaps one of the explana-
tions for the sudden spate of coups in Mrica in the course of
1966-7, which came after a phase of very rapid expansion in the
armed forces. It is interesting to note that, while the 'ruthless
oppression' of the colonial powers was often carried out by
means of a village constabulary with few military pretensions,
the new era of freedom has often required the creation of heavily
armed para-military police forces.* In Ghana, for example, the
*But see Appendix A.
 96 Coup d'Etat
police system was extended after independence in 1957 and
armoured car units were added to the already existing mobile
police; the communication system of the police was made
independent of civilian services, and the 'Escort police', which
used to be a fezzed and barefooted force of amiable illiterates,
has been turned into an effective riot-breaking unit.
If the para-military police is large, as compared to those units
of the armed forces which we can incorporate, it will be necessary
to repeat the whole analysis and infiltration procedure on it. We
may, indeed, be able to concentrate on the para-military police
and content ourselves with neutralizing the army by technical
means. Normally, however, the balance of forces between the
means of coercion of the state will not require this and we will
be able to isolate the police for the duration of the coup by using
the army.
The first step in our neutralization of these forces is to establish
the size, deployment and organization of the para-military
police. This is usually easier than in the case of the army because,
unlike the latter, para-military forces are usually stationed in
permanent barracks. Next we will try to find out their degree of
attachment to the present regime. But this will not involve the
sort of study in depth we made of the army, and it will only be a
matter of finding out about their corporate, rather than individual
outlook. The mentality of the para-military police may be
'bureaucratic', i.e. concerned with jobs and careers - as in the
case of the Italian Pubblica Sicurezza and its Celere para-military
units; if this is the case, a minimal degree of intervention can be
expected. On the other hand, their mentality may be parallel to
that of the army, i.e. concerned with loyalty and honour* (as
well as jobs and careers) or reflect a political association, as in the
case of the KB G or Duvalier's Ton Ton Macoutes.
If the equipment, deployment and mentality of the para-
military police is such as to make them an effective intervention
force we will have to control them in the same manner as the
'hard core' loyalist forces of the army. (The ways and means of
this forcible isolation will be discussed in Chapter s.) Usually,
*The corporate mentality will of course be somewhat more complex than
is suggested here by way of illustration.
 The Strategy of the Coup d'Etat 97
however, we will find that the para-military police force is
essentially bureaucratic and, therefore, in spite of its impressive
military bearing and equipment, it will not intervene against the
armed support of a coup. I have been unable to find a single case
in the last twenty years of a para-military police which has
actually defended its political masters during a coup - though
there are several cases of their intervention on behalf of a
coup.

The Rural Police and Gendarmerie

In most underdeveloped countries, this element of the police
force is numerically the strongest; this is only to be expected
since most of the population of such countries lives in villages
and works in agriculture. In spite of its large size, this part of
the police will almost never have an intervention potential
against a coup. They are often retired NCOs, fully integrated
in the rural society they live in and, even where there are pro-
visions for their mobilization and concentrated use, they are
unlikely to be assembled, equipped and prepared in time to
intervene against us. Whether the rural policeman is a garde
champetre with an ancient pistol inscribed La Loi, or a Middle
Eastern Zaptie who plays the village boss, he will hardly want
to rush to a remote capital city to protect an equally remote
government.

The Urban and National Police

Though this part of the police system will be considerably less
dispersed than the village-based rural police, its main com-
ponents will be just as ineffective against a coup. The personnel
of the urban police will fall into three broad categories: (a) crime
detection and investigation; (b) normal surveillance, and (c)
traffic duties. The C I D element will be small, very bureaucratic-
minded and, apart from its incidental intelligence aspect,* it can
be ignored by us. The uniformed police which carries out all the
usual surveillance duties will be more numerous, but though
they may be useful as a riot-breaking force when suitably
* See next section, p. 99.
98 Coup d'Etat
concentrated they are unlikely to act against armed opponents in a
major political crisis. The municipal police, largely concerned
with traffic duties, will usually be staffed by middle-aged men of
retiring disposition, with small and rusty pistols. There are,
however, exceptions such as the Spanish Polida Armada y del
Trdfico whose personnel are politically screened and which is
equipped with adequate transport and telecommunications and
thus able (and probably willing) to intervene in a major political
disturbance. A detailed analysis of our target country's police
system will probably reveal a problem of composition: after
dividing the police force into 'hard' and 'soft' forces we may
find sizeable' hard' sub-divisions within the 'soft' elements.
Our brief survey has shown that only a small part of the police
force is likely to be able to intervene against us, and of this a yet
smaller part is likely to do so with any enthusiasm. The natural
inclination of the police will be to 'ride' the crisis out and, as
individuals, to avoid endangering their positions vis-a-vis their
possible future employers. The coup may well be planned as a
military operation, but it will not - unless partially or totally
unsuccessful- involve any actualjighting. Thus, the fact that the
police is not heavily armed does not fundamentally explain their
low intervention capability, as compared to the army. The real
difference between the two is in their degree of integration in the
civil society. While the army can develop a corporate ideology
and mentality which is divergent - or even opposed - to the
civilian one, the police is usually too intimately involved in
civilian life to do so.
This can be either an advantage or an obstacle from our point
of view. On the one hand the eccentricity of the army will mean
that a regime can retain its appeal in the closed world of the
military barracks after losing it in society at large. This might
interfere with our recruiting, but it could work the other way,
i.e. we may find that the army is fundamentally opposed to a
government which much civilian opinion accepts. Recruiting our
forces among the police will almost always be more difficult than
in the army. Firstly, the lower level of (automatic) discipline
will mean that recruiting an officer may not bring over 'his' men
as well. Further, the fact that policemen live among the public
 The Strategy of the Coup d'Etat 99
will mean that the internal dynamics which can be generated in
the closed world of a military unit would be dissipated in this
more open environment and the •snow-ball' effect which would
bring entire units over to us after a limited degree of infiltration
will not operate. All these factors point the same way: the low
degree of intervention capability - for, as well as against us -
and the difficulty of incorporation, both indicate that while the
army should be penetrated the police forces can be dealt with -
defensively - after the coup,

NEUTRALIZING THE SECURITY AGENCIES

The security agencies of our target country will be numerically

the smallest of the professional defences of the state, but often
also the most dangerous. Unlike the armed forces and the
police, the security agencies will be actively trying to identify
and defeat threats posed by groups such as ours; unlike the
armed forces and police, their organizations, deployment and
personnel cannot usually be studied from the outside, and even
their existence may not be known to us. Almost every state has
some sort of 'secret service'. Many have several such organiza-
tions which operate both within and outside the national terri-
tory, and which we have so far described with the blanket term
of •security agencies'. Our first task is to try to identify them
more precisely.
It is well known that the bureaucratic animal in its natural
state has certain characteristic patterns of behaviour: it grows
in size and extends its sphere of action until checked by some
outside force. This role is usually played by the financial
bureaucracy, which fulfils its instincts by opposing the growth of
all other bureaucratic organizations. Equally important as a
limiting factor, is the concerted pressure of individual bureau-
cracies, each of which is fighting to preserve and extend its
territory. The cumulative effect of these pressures is to limit to
some extent the growth of the bureaucracy as a whole, and
perhaps without them all the inhabitants of developed countries
would by now be employed by the state bureaucracy.
These pressures operate weakly or not at all in the case of the
100 COUp d'Etat
security services: their budgets are allocated by many different
departments and are usually secret so that they cannot easily be
known, let alone reduced; other bureaucratic organizations can-
not prevent them from poaching in their territories because their
activities cannot often be identified and thus declared to be off-
limits. Finally, the prestige which undercover men of all kinds
often enjoy allows them to break rules which other bureaucrats
must obey, and to operate in all areas of social activity. The
result of this freedom is predictable: in many countries security
agencies have grown in a more dynamic and more disorderly
fashion than the rest of the bureaucracy and tend to have over-
lapping spheres of activity.
Before a zoologist studies his animals he classifies them and
tries to relate them to the nearest known species. We will follow
this procedure both in functional terms (which are generally
applicable to all countries) and in organizational terms (which
are particular to each one).

The Pure Intelligence Function

This classification covers the collection and analysis of published
and unpublished information of all kinds and, because of the
high degree of specialized knowledge which is often required,
many different bodies can enter this field, which is the most
crowded of the whole sector. Tactical military intelligence (What
is the opposition doing?) may be collected by separate agencies
working for the three branches of the armed forces; in tradi-
tionally seafaring nations, naval intelligence is often the largest
and most developed service. Strategic information (What is the
opposition planning?) may be the province of separate and
competing agencies run by the General Staff, the Ministry of
Defence and the ministry in charge of foreign affairs. Scientific
information may be collected by the administrative entity in
charge of 'science' and also by specialized bodies in charge of
particular sectors, such as atomic energy, aeronautics and tele-
communications. Economic intelligence is one of the worst areas
of duplication, with demographic, energy and agriculture
authorities operating alongside the entity in charge of economic
affairs in general. Political intelligence may be handled by the
 The Strategy of the Coup d'Etat 101

foreign affairs ministry openly, through the diplomatic service

and also covertly by a separate agency.

The Counter-Intelligence Function

This covers the prevention of the activities listed above and may
be carried out by both generalized and specialized bodies. The
military may run their own agency, and the police of each
branch of the armed forces may do the same. The Ministry of
the Interior will almost always have a 'spy-catching' service (like
the Security Service of the Home Office) and particular bodies
will have a service to protect their installations (but these rarely
go beyond the ordinary police stage). From our point of view,
this sector will be the most important. We may - if we fail to
preserve our security position - come into contact with (a) the
police agency, such as Special Branch in Britain or the FBI in
the USA, (b) the separate ministerial body, or (c) the military
agencies. Much of our planning and infiltration work will be
indistinguishable from that which could be carried out by a
foreign intelligence service and it will therefore enter into the
territory of the counter-intelligence agencies.

The Counter-Espionage Function

This is the most subtle and sophisticated of all the functions. It
covers the deliberate contact with opposition intelligence services,
in order to feed them with 'disinformation' or to disrupt their
organization. It is unlikely that more than one agency carries
out this work because it requires an extremely precise control
over operations. The agency may be a subsection of any of
those mentioned above, but in order to function efficiently it must
be able to exercise some form of control over all competing
agencies - especially over counter-intelligence, which relates to
counter-espionage as a butcher does to a surgeon.

Internal (political) Security

This is another sensitive area from our point of view. Its specific
function is the 'prevention of exactly what we aim to do: the
overthrow of the government. In many countries there is a
102 COUp d'Etat
'political' police, with both uniformed and covert agents, and it
may be controlled by the bureaucracy of the Ministry of the
Interior or by the inner political leadership, either directly or,
in one-party states, through the party. Elsewhere, in more or
less democratic regimes the police has a political department, as
in France, Italy and West Germany, and its primary function is
the surveillance or extremist groups. In military dictatorships,
the territory of military intelligence often extends to this area;
in some countries the agency in charge of the physical protection
of the higher leadership may run an information service as well
as providing the bodyguards.

Internal Intelligence
This function is carried out by the information services which
are attached to the police and para-military forces of the state.
Thus, in Italy, apart from the police (Pubblica Sicurezza) which
has a ' political' squad, the para-military Carabinieri has an
information service (8 I FAR) which, though primarily concerned
with military counter-intelligence, also operates in the internal
political field.
Our behaviour in the midst of this bureaucratic jungle will be
purely defensive, unless we have a 'direct line' to one or other
of the security agencies. If that is the case, the security agency
concerned would provide an ideal 'cover' for all our activities.
Failing such a fortunate coincidence, we will not try to create a
'direct line' by infiltrating any security service, because if we
do so there will be the very great danger that they will use any
contact in order to infiltrate us. This is a standard procedure for
the security services to follow, and the elementary defensive
techniques used when infiltrating the armed forces (cut-outs,
one-way communication etc.) will probably fail to work in their
case.
In order to run a secure operation we will follow rules which
derive from the basic assumption that all information about our
activities is a source of danger as soon as it exists outside the
minds of our inner group. From this all the standard procedures
emerge: (a) no information to be communicated except verbally;
(b) no information to be communicated except on a 'need-to-
 The Strategy of the Coup d'Etat 103

know' basis; (c) all communication links from inner to affiliated

members to be on a one-way basis; (d) no activity to be carried
out by an inner member if an outer member can do the
job.
These rules are simple and well known; the problem is to keep
to them under the pressure of work and the emotions which it
generates. The most sensitive of our activities will be the
approach and persuasion of new affiliates to the coup, and the
nature of the security agencies can add an extra measure of
danger: in many countries some of the security services are
hidden within totally unexciting administrative bodies. Where,
as in the case of the US Treasury's Secret Service, this reflects
an administrative convenience, the fact is well known; elsewhere,
however, the department-within-a-department system is deli-
berate. We may therefore unwittingly try to infiltrate a 'safe'
department and discover that we are dealing with a security
agency. All we can do is to list some of the places where it is
'natural' for security services to 'grow': census and carto-
graphy services; central bank anti-counterfeiting agencies; post
office departments; press bureaUx; customs and immigration
departments and the taxation authorities.
It must not be thought that our entire operation will auto-
matically collapse if it is penetrated by a security agency. * If we
have followed the security procedures the chances are that only
a small part of our total effort will be identified, and therefore
its ultimate purpose may not be discovered. Even if it is dis-
covered that a coup is being planned, the security agency may
wait before taking any action, in order to capture all the plan-
ners - and this could be too late. As soon as our teams are on the
road, actually executing the coup, it will be too late for the
security services to oppose us on the 'information' side, while
*The Okhrana, the Tzarist secret police, was extremely efficient and it
had infiltrated the Bolshevik and other revolutionary parties without, how-
ever, impairing their activities. Roman Malinovsky, who was the leader of
the Bolshevik organization inside Russia until 19I4, was working for the
Okhrana, and they edited Pravda, whose chief editor was also one of their
agents. This, incidentally, later lent some slight plausibility to the accusations
made at the Moscow trials, that old Bolshevik leaders had been lifelong
stooges of foreign secret services (usually the British).
104 Coup d'Etat
their fighting power will usually be unimportant as compared to
the army units we have incorporated. Finally, political security
agencies are necessarily sensitive to political trends, and they may
decide to join the group planning a coup if they know that it is
well organized and ready to seize power.
 4. The Planning
of the Coup d'Etat

, • • • Even barricades, apparently a mechanical

element of the uprising, are of significance in
reality above all as a moral force ••• ' Lev
Davidovich Bronstein (Trotsky)

In the early morning of 23 April 1961, elements of the 1st

Foreign Legion Parachute Regiment seized the key points of the
city of Algiers in the name of Generals Challe, Zeller, Jouhaud
and Salan. The four generals, because of their personal prestige
and their position in the French hierarchy, quickly asserted their
control over the local military command and started to extend
their authority over all the armed forces in Algeria. At this time
de Gaulle's government was in the process of opening negotia-
ti~ns with the Algerian nationalists and the generals were
determined to replace him with a leader who would carry the
war to a victorious conclusion. The French armed forces in
Algeria were much more powerful than those stationed in France
and Germany and the four generals were hopeful that, once
their allegiance was assured, they would find it easy to take
effective control of the French government. Mter all, de Gaulle
himself had come to power after a similar episode in May 1958,
and there seemed to be no major obstacle to a successful second
edition of the famous treize maio
When the four generals made their declaration over Algiers
Radio, the 1st, 14th and 18th Colonial Parachute regiments
rallied to the coup. A few infantry units, some of the Marines
and much of the Air Force remained loyal to de Gaulle (as in
May 1958 they had remained loyal to the IVth Republic), but
most of the armed forces in Algeria were attentiste. Wait-and-
see is the attitude that usually favours a coup and when General
106 Coup d'Etat
Henri de Pouilly withdrew his headquarters from Oran to
Tlemcen, to avoid having to choose between fighting or joining
the coup, he was objectively favouring the coup.
The four generals seemed to be on the verge of victory. The
determined pieds noirs population of Algeria was 100 per cent
behind them. The powerful parachute units gave them a hard-
hitting force of intervention and the bulk of the armed forces
were either for them or neutral. Even the forces loyal to de
Gaulle's government did nothing actively to oppose the coup.
While the leaders of the coup started to gather support, the
French Defence Minister was on a visit to Morocco; Maurice
Papon, the head of the Paris police, was on vacation; Debre, the
Prime Minister and chief' :fire-fighter' of the regime, was ill; de
Gaulle himself was entertaining the visiting President of Senegal,
Senghor. Other ministers were on visits to Algiers itself, and
were promptly captured and held in confinement, together with
other representatives of the President. Everything pointed to an
early victory of the coup, and yet a few days later General Challe
was being flown to Paris for eventual trial and imprisonment,
Salan and the others were fleeing to the interior on their way to
exile or capture, and the 1st Foreign Legion Parachute Regiment
drove back to their barracks singing Edith Piaf's 'Je ne regrette
rien'" though their officers were under arrest and their unit was
to be disbanded.
Why did the coup fail? Perhaps the main reason was that the
four generals had utterly neglected the 'political forces' and had
allowed the immediate power of the armed forces to obscure the
somewhat less immediate, but ultimately decisive, role that they
could play. In the Gaullist coup of May 1958, the action of the
military and the population of Algiers had been supported by the
Gaullist infiltration of the civil service, and by the steady cor-
rosion of the will of other political groups to oppose the dissolu-
tion of the IVth Republic. This time the generals had simply
ignored the civilians.
De Gaulle went on television and called for popular support:
'Franfaises, Franfais" aidez-moi.' Debre, who followed him on
the screen, was more specific: 'Go ••• to the airports ••• con-
vince the soldiers who are misled •••' He also started to arm a
 The Planning of the Coup d'Etat 107

militia drawn from the Gaullist party. More important, the

trade-union organizations, the Communists (CGT), the
Christian Democrats (CFTC), and the Force Ouvriere, all
rallied round the government while most political parties did the
same; the left-wing Catholic movement started to organize sit-
down strikes among the national servicemen in Algeria; and, in
general, most organized forces of French society intervened and
refused to accept the authority of the coup.
The effect of this refusal was decisive; the larger part of the
'wait-and-see' element in the armed forces stopped waiting and
declared its support for de Gaulle, and this was the end of the
coup.
We will only be able to avoid a repetition of the crucial error
made by the generals if we can neutralize the political forces as
effectively as the military ones. .
Immediate political power is always concentrated in the
country's government, but in every country and under all
political systems there will be groups outside the government -
and even outside formal politics - which also have political
power. Their source of strength can be their ability to influence,
particular groups of voters (as in democratic societies) or their
control over certain organizations which are important in the
country's political life. Whether these groups, which we have
called 'the political forces', are pressure groups, political parties
or other associations does not greatly matter. What is of impor-
tance is their ability to participate in the formation of govern-
ments, and, later, to influence their decisions. The nature of the
forces which are important in the political life of a particular
country will reflect the structure of its society and economy, and
it will also depend on the particular context of decision-making.
If, for example, we were asked to list the most important
forces in British political life, we could produce the following
(rather conventional) list:

the two major political parties

the Trades Union Congress and certain major unions
the Confederation of British Industry
the senior civil-service-academic complex
108 Coup d'Etat
the City and its corporations
the press
But if we were asked to isolate the groups which would matter in
foreign policy decision about, say, the Middle East, we would
come up with a quite different list:
the two major British and part-British oil companies
the Foreign Office-academic' Arabist' group
the central organizations of British Jews

TABLE 10. The groups which would influence the formation of

American policy on Middle Eastern questions. Formal and
unofficial participants.

Official:
the President and the White House staff
the Department of State
the Pentagon
the CIA (as supplier of information)
the key Congressional Committees
Unofficial:
(a) Oil companies with interests in the Middle East. (These seek to
protect and extend their interests and therefore advocate sym-
pathetic understanding for Arab aspirations and policies.)
(b) Oil companies with exclusively domestic interests. (These
favour a continuation of the existing policy of excluding foreign
oil from the US market, and are therefore inimical to any general
US - Arab rapprochement.) .
(c) Other energy industries. (Which also oppose any relaxation of
present US import controls on - cheap - foreign oil.)
Cd) Politicians with significant Jewish populations in their con-
stituencies.· (These naturally follow a visible pro-Israel line on
Congressional voting and make appropriate speeches.)
(e) Pro-Zionist organizations of American Jewry.
(f) American Council for Judaism. (This body is anti-Zionist and
opposes the pro-Israel elements in this field.)
(g) Academic bodies with a special interest in Arab or Middle
Eastern studies. (These usually identify with Arab views and
seek a sympathetic hearing of Arab claims.)
 The Planning of the Coup d'Etat 109

In a sophisticated society, with its complex industrial and

social structure, there are hundreds of organizations which,
whatever their primary purpose, also act as pressure groups and
attempt to influence political decisions in a manner that serves
their members' interests. These organizations will reflect in their
divergent attitudes the diversity of a complex society. In econo·
mically backward countries, however, the structure of society is
simpler and any conflict of interests, though just as strong, is
played out in a much smaller arena and with fewer participants.
In Mrica south of the Sahara religious groups are generally
fragmented and apolitical, and where the local business com·
munity is still relatively small and weak, the major political
forces are limited to a few groupings:
Tribal and other ethnic groups
Trade unions
Students' and graduates' associations
Civil-service officials and officers of the armed forces
The activists of the ruling political party
In much of West Mrica one would have to add the local market
traders' association, and in immediate sub· Saharan areas the
traditional Muslim leadership structures. In Asia, religious
groups and their leaders would have to be added to the list, and
in some countries (such as Taiwan, Thailand, South Korea and
Hong Kong) the local business class will be of importance.
Missing from all the lists are the foreign business interests which
may play an important or even dominant role, but which repr~
sent a special problem already dealt with in Chapter 2. But
whatever groups dominate the political scene of our target
country in normal times, the special circumstances of the coup
will mean that only a few elements among them will be important
to us.
Political forces can intervene against the coup in two ways:
(a) they can rally and deploy the masses, or some part of them,
against the new government;
(b) they can manipulate technical facilities under their control
in order to oppose the consolidation of our power.
110 COUp d'Etat
The action of individual political, religious, ethnic or intellectual
leaders, who could use the framework of their party or com-
munity against us, is an example of the first kind of intervention;
a strike of the staff of the radio-television service is an example
of the second. A general strike would, in effect, combine both
kinds of intervention.

NeutraHzIng the Political Forces I: general

Politics, like economics, has its infrastructure. Just as industry
and commerce require a background of facilities such as roads,
ports and energy sources, direct political action requires certain
technical facilities. The mobilization of French public opinion
which took place during the attempted coup in Algiers - and
which was the principal cause of its failure - could not have taken
place without the use of a whole range of technical facilities. The
government appealed to public opinion by means of the mass
media, chiefly the radio and television services; the trade unions
and other organized bodies coordinated the agitation of their
members by means of their network of branches, connected to
the central headquarters by means of the public telecommuni-
cations facilities; finally the mass demonstrations could not have
taken place without the use of public and private transl?0rt.
Our general neutralization of the 'political' forces will be
conducted in terms of this infrastructure. We will seize and hold
such facilities as we require for our own purposes, while tem-
porarily putting out of action the others. If the means of com-
munication and the transport system are under our control, or
at any rate do not function, the potential threat posed by the
'political forces' will be largely neutralized: the leaders of the
pre-coup government will be arrested since they are part of the
infrastructure and they would probably be the major sources of
inspiration of any opposition to the coup.*
We will neutralize some political forces in particular by
identifying and isolating their leadership, and by disrupting
*In seizing the leaders of the government we wiD also contribute towards
the isolation of those segments of the army and police which we have been
unable to infiltrate, though more direct measures will be required as well.
 The Planning of the Coup d'Etat . III
their organizations; this will only be necessary for those forces
which are sufficiently resilient and sufficiently militant to inter-
vene against us even though the infrastructure has been
neutralized.
Both forms of neutralization will involve the selection of
certain objectives which will be seized or put out of action, by
teams* formed out of those forces of the state which we have
fully subverted or, in our terminology, incorporated.
Unless our target country is particularly small and its physical
and political structures particularly simple, its system of govern-
ment will be complex, its physical facilities will be extensive,
and its political forces will be many in number while their
intervention capabilities will be difficult to forecast.
We will, therefore, start by analysing the govemmentalleader-
s~ip in order to determine which personalities will have to be
isolated for the duration of the active phase of the coup and which
can be safely ignored. Next, we will study the physical facilities
and select those which are likely to be relevant during the coup,
in order to plan their seizure or neutralization. Finally, we will
investigate the likely nature of those political forces which could
still retain a degree of intervention capability after our' general'
measures, in order to prepare for their individual neutralization.

PERSONALITIES IN THE GOVERNMENT

However bloodless our coup, however progressive and liberal our
aims, we will still have to arrest certain individuals during and
immediately after, its execution. Of these the most important
group will be formed by the leading figures of the pre-coup
regime or, in other words, the leaders of the government and
their close associates, whether they are formally politicians or
not. The members of a Cabinet will form a fairly large group,
from ten to fifty people; adding their associates and intimate
advisers - who could organize opposition against us - we
could easily reaCh a figure four or five times this number. Apart
from being uncomfortably large, this will
also be an especially
*The nature and composition of the active teams of the coup are discussed
in Appendix B.
Ill. COUp d'Etat
determined and dangerous group. The personal repute, presence
and authority of its members could be such as to enable them to
rally against us the disorganized forces of the state, or the
unorganized masses: it could also enable them to impose their
will on the team sent to capture them, and tum their would-be
captors into their allies. General ChaIle, for example, was
regarded as the patron by the N C Os of the French Army in
Algeria, and even after the total failure of his attempted coup, the
Paris government could not entrust him to a military escort on
his way to France and arrest, and had to use the CR S,* who had
never experienced his personal authority. Mter all, if a young
soldier acting outside his familiar roles is facing a political
personality whose whole behaviour is calculated to make people
obey him, it is difficult to be absolutely certain that he will carry
out his orders, and not the counter-orders he may be given.
The large number of the separate targets, and the possible
'radiation' effects, indicate that the teams sent to arrest them
should be both large and particularly well chosen. Since our
resources will be limited, we will have to concentrate our efforts
on the most important figures within the group while leaving the
others to be picked up later when our means will have been
expanded by the allegiance of the 'wait-and-see' element. We
cannot arrest all those who may constitute an eventual danger,
but we must make sure that we do arrest the really dangerous
figures, that is, the key figures within the leadership, who may
or may not be the first in the formal order of precedence.
The formal structure of governments falls into two broad
categories (illustrated in Table 10): the 'presidential' type, where
the head of state is also the main decision-maker (as in the
USA, France and most of the Mro-Asian states), and the •prime-
ministerial' type, where the head of state has largely symbolic or
ceremonial duties and real decision-making is carried out at a
theoretically lower level (as in Britain, India and most of
Europe, as well as the US S R).
A third alternative form - which is not a structure at all, but
rather a denial of one - is the 'strong man' form of government.
*Compagnies RJpublicaines de SJcuriu, part of the police and outside the
military community.
 The Planning of the Coup d'Etat II3

TABLE II. Alternative Forms of Government

Presidential

~:~dent
(e.g. seventeenth-century England)
Real decision-} (e.g. twentieth-century America)
making level: Emperor (e.g. twentieth-century Ethiopia)
Ruler (e.g. twentieth-century Kuwait)
Prime (or Chief) Minister
Ministerial level
Junior ministers and civil service
Prime-ministerial
Ceremonial head} King (e.g. Belgium)
of State: President (e.g. Italy)
Real decision-} Prime Minister (e.g. United Kingdom)
making level: President of Council of Ministers (e.g. Italy)
Cabinet level ministers
Junior ministers
Higher civil servants

The 'strong man' may not be a top minister, and may hold no
official position at all, but actually rules by using the formal body
of politicians as a screen. This type ofregime is evolved when the
fabric of the state has been weakened to such an extent that only
the actual leader of some part of the armed forces or police can
control the situation and remain in power. If the person is him-
self even minimally acceptable as a political leader, he can take
over the formal posts as well and make himself the visible head
of the government. Abd-el-Nasir in Egypt, and Reza Shah (the
father of the present Shah of Persia) both accomplished this after
a short period of transition, but there can sometimes be racial or
religious reasons that bar the 'strong man' from an official
position. The man who controls the bayonets may be totally
unacceptable as a public figure, but he can still rule indirectly
by manipulating the official leaders which he keeps under control
by the ultimate sanction of force.
When in early 1966 the Syrian government of the moderate
wing of the Ba'ath party, headed by Michel Aflak, Salah Bitar
and the army leader Hafiz, was overthrown by an extreme left
114 Coup d'Etat
faction of the party, the new leadership found out that though
it controlled the army and the country it could not rule openly.
The army officers who led this latest coup were too young, too
unknown and, above all, they were Alawites. Salah Jadid, their
leader, is a dark, brooding figure who inspired fear and hatred
among that small part of the public that knew of him. And of all
the communities of Syria, the Alawites are amongst the least
prestigious. In colonial times the French had recruited most of
their forces of repression, the Troupes speciales du Levant, from
the minority communities, chiefly the Alawites, and they had
given the Alawite area in northern Syria a form of autonomy in
order - so the Nationalists claimed - to break up Syrian national
unity. Mter independence the Sunni majority community re-
garded the Alawites as renegades, and public opinion would
only have accepted an Alawite head of state with difficulty.
Salah Jadid overcame this problem by appointing a full set of
Cabinet Ministers, carefully chosen so as to balance the various
communities, while retaining the real decision-making power
within a separate body, 'the National Revolutionary Council',
headed by himself. Thus, though Syria has a President (Nurredin
Atassi), a Prime Minister (Youssof Zwayeen), and a Foreign
Minister (Ibrahim Makhous), all major political decisions are
made by J adid; the ministers go on state visits, make the public
speeches and appear in all ceremonial occasions, but power is
not in their hands.
The government of the Csocialist' countries is formally party
government, but it tends to break down into one of the two other
types. In its original form, real political power is concentrated
in the hands of the centr31 committee, or some other higher
party council, as illustrated in Figure I.
Once the purely ceremonial figures have been excluded, the
number of people still to be dealt with will be reduced, and by
applying our time-span criterion, we can reduce their numbers
still further. The Minister of Economic Planning may be a
crucial figure in the government, his position as a technocrat
may be unassailable, but he may be unable to rally public
opinion against us, or to assert his authority over the armed
forces. The dramatic nature of the coup will reduce political life
 The Planning of the Coup d'Etat liS

1. ALTERNATIVE FORMS OF GOVERNMENT

Party governmentin ~ociallst'countries and"'People's Republics'

Head of Party Machine

General Secretary of Party
Chairman of Pflitobureau etc.
J,
Head of State
Purely ceremonial
Prime tinister
Ministers
1
Regional party heads
President, etc.
1
Civil Service
1
Army
I Police
. "'s-ervlces
Secunty .

to its ultimate rationale, sheer force, and we will concentrate

on those figures in the government who could deploy it. The
obvious personalities will therefore be:
(a) the Minister of the Interior and his associates (who control
the police force);
(b) the Minister of Defence and his associates (who control
the armed forces);
(c) the party leaders (ifthere is a party militia);
(d) the Prime Minister or other central figure (who coordi-
nates all these).
We must remember that for various reasons figures in the
government may not always be what they appear to be. We
may discover that the apparently innocuous Minister of Educa-
tion controls an important students' militia, or the Minister of
Labour a powerful workers' militia. More important, the
'effective power may be held by an inner association of a particular
group of ministers who between them control the means of
coercion of the state. Thus, the government of Czechoslovakia
between the elections of May 1946 and the final Communist
 n6 Coup d'Etat

2. THE FORMAL GOVERNMENT AND THE REAL ONE

(Members of innerpower group shown In italics)

President/Head of State
Prime Minister
v---------------~
MiniSter 01 Minister of L-->. Ministerol Minister of
Defence finance .-... Interior Foreign Affairs
Minister ofJustice
Minister 01Labour ~~------I

Minister of Health Under Secretary of State for:

MinisterolEducationE-
~--+-4, The Army

Ministerof Development The Navy

Minister of Posts The Air Force
Minister ofTransport 4 The Police
Minister of Social Welfare Treasury

take-over in February 1948 was a coalition of all 'democratic'

parties, but the Communist ministers within it effectively
monopolized the real sources of power by means of the control
of the means of coercion. The existence of a group of associates,
whose alliance transcends the formal order of government, is
illustrated in Figure 2.
In this particular case, out of the eighteen or so members of
the government, the Prime Minister, the Ministers of Defence,
Labour, Education and the Under Secretaries of State for the
Army and Police, form a special inner council which actually
holds the reins of power.
The process of selection so far discussed should result in the
classification of the personalities of the pre-coup regime into three
categories :
 The Planning of the Coup d'Etat 117

The Ceremonial Figures

These will not be arrested. If the head of state is generally
popular he should be used as a symbol of continuity who will
help us to establish our legitimacy, provided he can be safely
manipulated and made to play this role. The other, lesser
ceremonial figures can simply be ignored.

The 'inner council' and the Controllers of the Afeans of Coercion

This small group must be sequestrated and held in isolation
until our authority is safely established. Apart from the service
ministers, etc., any government leader who is personally parti-
cularly popular should be included in this category.

The Other Ministers and Top Civil Servants

This larger group should be subdivided into priority groups and
put into the category above, as and when our resources expand,
or become available when other more urgent tasks have been
carried out.

PERSONALITIES OUTSIDE GOVERNMENT

The political weight of an individual in any large-scale political
community will usually only be important within the frame-
work of an organization which he heads or manipulates. It is
sometimes possible for an individual to achieve political im-
portance by becoming identified with an ideology or an attitude
in which some significant part of the public believes. Kossuth,
the leader of the Hungarian nationalist movement in the 1848-9
Revolution, was a poet by profession, and had no party machine
behind him, but he did have considerable power because the
masses (in the cities at any rate) identified his person with
Hungarian nationalism. Gandhi, who operated largely outside
the Congress party machine, also achieved personal power
because to many Indians he was the embodiment of nationalism.
The remoteness of the examples indicates that such figures are
very rare, and if we do have them in our target area they should
be treated as ceremonial figures.
lIS COUp d'Etat
PHYSICAL FACILITIES
Mass Media
Control over the flow of information emanating from the political
centre will be our most important weapon in establishing our
authority after the coup. The seizure of the main means of mass
communication will thus be a task of crucial importance. One,
though only one, of the causes of the failure of the Greek King's
counter-coup in late 1967 was this inability to communicate with
the masses, literally and otherwise. When Radio Larissa broad-
cast the King's messages it only reached a fraction of the
population: the transmitter was weak and the wave-length
unusual; instead of the booming voice of authority the declara-
tion took the form of a weak appeal for help. We must not make
a similar mistake.
Because of the short time-span of the coup, and because of the
likely social background of our target country, the press need not
be a primary target; we will establish our authority over it after
the coup, as with other aspects of the nation's life. Inevitably the
press can only play a marginal role in countries where illiteracy
is widespread; and, in any case, it is the radio-television service
which is mainly associated with the voice of the government.
The approximate comparative data for the Arab world in Table
12 illustrates the importance of the different media, in one part
of the •Third World'.
TABLE 12. Mass communications in the Middle East and North
Africa. mid-I967*

Estimated circulation of daily newspapers 1,500 ,000

Estimated number of television sets 1,000,000
Estimated number of radio sets 7,000,000

Even these figures understate the importance of radio and

television sets, because while the press figures refer to circula-
tion, i.e. estimated number of readers, rather than copies sold,
the radios and television sets reach a much wider public even
amongst the poorest groups, since every cafe has one.
*Data compiled from Orbis Yearbook and National Publications.
 The Planning of the Coup d'Etat II9
There are two problems associated with radio and television
facilities from our point of view: (a) there will often be many
different broadcasting services and associated facilities, and (b)
they are particularly difficult to seize. In some countries, where
the internal security position is· precarious, the governmental
radio is heavily guarded, but even where this is not the case, these
facilities are difficult to seize because their staff have a uniquely
extensive way of raising the alarm. As for the duplication of
broadcasting facilities, even Haiti, a very small and extremely
backward country, has eighteen different radio stations, con-
trolled by independent networks. Our objective is not merely to
control but also to monopolize the flow of information, and we
must therefore deal with every single facility. This would be
difficult (and would also lead to a dispersal of our forces) if we
tried to seize and hold every facility. Our strategy will therefore
be to seize and hold just one facility, the one most closely
associated with the voice of authority, while neutralizing the
others. This is best done with the cooperation of some technical
member of their staff who would be able to sabotage the facility
from the inside. A single cooperative technician will be able
temporarily to put out of action a radio station which would
otherwise require a full-scale assault team.
If we are unable to recruit an internal saboteur, the next best
alternative will be external sabotage. There is no need to cause
any extensive damage since it will usually be possible to remove
or destroy a small but essential part of the transmitter(s), thus
effectively neutralizing the facility. The one broadcasting facility
which we do have to seize and hold will present a special problem:
on the one hand, our need for the facility is absolute; on the
other, because it is such an obvious target, the governmental
forces will certainly try to recapture it. This means that the .
team assigned to this target will have to be adequately staffed
and equipped, and in order to obviate the need for the co-
operation of the facility's personnel, should also include a
skeleton technical staff. (Appendix B, on the military aspects of
the coup, deals inter alia with the composition of the various
teams.)
120 COUp d'Btat

Telecommunications
Technical progress has evolved in our favour, since all the
communication requirements between our own teams can be
carried out by the cheap and reliable two-way transistor radios
now universally available. We must, however, deny the opposi-
tion the use of their fixed communication systems, because by so
doing we will paralyse their reaction and prevent them from
deploying against us such forces as they still control. As Figure 3
shows, the neutralization of the telecommunication facilities will
be complicated by their multiplicity, and it will be essential to
achieve full coverage. The Left Socialist Revolutionary coup
against the Bolsheviks in July 1918 failed partly because it
failed to comprehend the need for a monopoly of all telecom-
munications. The Left Socialist Revolutionaries had infiltrated
a group of the Cheka, the main instrument of Bolshevik power,
and various army detachments; with these they arrested the
head of the Cheka, Dzerzhinsky, seized many public buildings
and the Moscow telegraph office. They failed, however, to seize
the Telephone Office as well, and while they were sending cables
all over Russia asking for generalized political support, Lenin
used the telephone service to mobilize his fighting forces, and
with these the coup was quickly crushed.
Internal security authorities are aware of the need for efficient
communications, and apart from the facilities illustrated in the
diagram on p. 121, there may also be independent networks for
the exclusive use of the security forces. The French gendarmerie
has a system of tegionallinks which by-passes the public tele-
phone and cable wires, and even in smaller countries, such as
Ghana, the police can have a fully independent system (Table
13).

TABLE I3. Police Telecommunication Facilities in Ghana

63 fixed wireless stations, both high frequency and VHF radio
telephones
6 dual-purpose mobile radio stations
Numerous man-portable radio sets
 3. TELECOMMUNICATION FACIUTIES AVA/LABLE TO GOVERNMENT

Telephones
Police
Telephones ) Exchange ) ISecurity Services Ordinary Facility to :
Armed Forces Government bureaucracy
In capital city area and subsidiary to :
Armed Forces

.
Telex/Telegrams
POS! Office
) Regional
Relay System
I
Relay Centre 'A'
) RelayCentre'B'
Relay Centre 'C' etc.
~
~
/
~
Police
Security Services
In regions'A"B"C' etc.

Police
Armed Forces in Capital City Security Services
Armed Forces
/ Capital Local Relay 'A'
Independent Military ) City Military ) Local Relay'B' ) Armed Forces
Radio Network Radio Station Local Relay 'C' etc. I in areas 'A"B"C' etc.

1,
Emergency Station
122 COUp d'Etat
In the USA there are no national police networks, but the
Department of Defense maintains a nation-wide and inter-
national system which is the largest single network in the world
and which connects every U S mili~ary installation with every
other throughout the world.
We cannot, of course, hope to seize every two-way set in the
hands of the police and the military authorities, but we should
neutralize, by external or internal sabotage,* those facilities
which can be identified and located. There is no need to seize
and hold any of these facilities, and it will therefore be a matter
of penetrating the central organization of each communication
system for the brief period required to sabotage its operation
though, again, internal sabotage will be easier and safer.

City Entry-Exit Road Links

During the active phase of the coup the unexpected arrival of
even a small contingent of loyalist or uninfiltrated forces could
seriously endanger our whole effort. When a government
discovers that troops of its own armed forces are taking part in
a~~~~~~~~~~~~~w~oo
troops stationed elsewhere, in the hope that the infiltration of
the armed forces is limited to those ~ the capiw city. As it is
not easy to infiltrate forces in the entire national territory, the
government's hope may not be unfounded. We will attack the
mechanism which could lead to the arrival of the loyalist troops
in the capital city at each separate level: we will arrest those who
would call them in, we will disrupt the telecommunications
needed to reach them and we will ~so try to isolate identified
loyalist forces by direct (though purely defensive) military
means. We must also prevent the intervention of these forces by
controlling the last level: the perimeter of the capital city and
scene of the coup.
If the loyalist forces are to intervene in time, they will have to
move rapidly and this will require the use of either the major
*The normal way of neutralizing an electrically-powered facility is to
detonate small plastic charges on the grouped wire links between the facility
and the public power supply (and independent generators, if any). These are
usually not difficult to reach from die outside.
 The Planning of the Coup d'Etat 123

4. THE PHYSICAL TARGETS OF THE COUP

- - - - - Perimeter within which active teams can operate in safety

0=

X Road block
0=

roads or alternatively air transport. If we can set up efficient

defensive road blocks at the appropriate places we should be able
to deny their entry into the capital city for the short period
required, that is, until we have established ourselves as the
government and received the allegiance of the bulk of the state
bureaucracy and military forces. Thus, by the time the forces of
intervention have reached the scene of the action, they will be
the isolated band of rebels. The most suitable places to block a
road with a small number of men and limited equipment, as well
124 Coup d'Etat
as the techniques and implication of such actions, are discussed
in Appendix B and also in Chapter 5 where we deal with the
direct neutralization of the identified loyalist forces. Figure 4
illustrates the locations which would be chosen in a particular
(synthetic) example. But our control of the physical access to the
capital city will also serve other purposes. It will be one of the
ways in which we will establish the physical presence of the
new regime, and it will also allow us to prevent the escape of
governmental leaders and other personalities which we have
been unable to arrest. One of the dangers which we will face will
be the revitalization of counter-coup opposition, which could
result if a major governmental figure escapes from the capital
city and joins loyalist elements outside it. Mter all the efforts we
have made to neutralize such forces by internal means and by
interference with their transport and communications, our whole
work could be endangered. The loyalist forces could fail to reach
the capital but the political leadership could reach them. The
means at our disposal will not be sufficient to seal hermetically
the entire capital city, though of course much will depend on its
location and spatial spread. Brasilia, though open on all sides,
would be easy to seal off simply by closing the airport, since the
surface links are inadequate to allow rapid movement to the rest
of the country. Helsinki, on the other hand, would be spatially
convenient because, though not remote from the rest of the
country, it is surrounded by sea and lake so that a small number
of road blocks would effectively seal it.

Focal Traffic Points

The sight of tanks in the main squares of the capital city has
become a symbol of the coup,* but is also an expression of a
very real practical requirement: the need to establish a physical
presence in the centre of political activity. Every capital city has
an area which is the local equivalent of Whitehall in the UK or
Capitol Hill in the USA, in or near which the main political-
administrative facilities are concentrated. We will select and
*Tanks in the main squares are a feature of the Middle Eastern and Latin
American, but not of the Mrican, military coup. Most African armies do not
in fact have tanks.
 The Planning of the Coup d'Etat I2S
defend certain positions around and within this area and by so
doing we will achieve a variety of purposes: (a) the positions will
form a ring round the main area within which our active teams
will operate so as to protect them from any hostile forces which
may have penetrated the capital city; (b) they will assist
in establishing our authority by giving visual evidence of our
power; (c) they will filter movement to and from the area, thus
enabling us to capture those whom we have been unable to arrest
directly.
In order to achieve these different objectives, our blocking
:positions must be individually strong, since otherwise they may
tempt any extant loyalist forces into a counter-attack. In any
case, unless adequately staffed they will be unable to act as
efficient filters to individual movements. We must therefore
resist the temptation to secure every important location by block-
ing positions which are individually weak. As only a few of the
possible locations will, in fact, be covered, it is essential to select
them with special care. Focal traffic points will be easier to
select in a coastal or riverine city, where a definite shape has been
imposed to the capital city, and to the traffic flows within it. This
is illustrated by Figure 5. In each particular case, the area which
is the centre of political and bureaucratic activity will he well
known to the local inhabitants, and it will therefore be a matter
of selecting a perimeter of straight and fairly broad streets at the
intersection of which we will establish our blocking positions.
(The avenues and boulevards of Paris are ideal from this point
of view.)

Airports and Other Transport Facilities

One of the classic moves in the period immediately following
the coup is the closure of airports and the cancellation of aU
flights. This is part of the general tactic which aims at 'freezing'
the situation by preventing the uncontrolled flow of people and
information. There will also be other, more specific objectives:
by closing the airport we will prevent the escape of those govem-
mental leaders whom we have been unable to arrest. We will also
prevent any inflow of loyalist forces into the area of the capital
city. Because of the short time-span in which the coup takes
 5. PHYSICAL TARGETS IN COASTAL CITY

x = Road blocks at focal traffic points and city entry points

P ...,Densely populated sea-front district
B '" Mixed residential areas
G ... Governmental administrative and other public buildings
R... Governmental leaders residences
- - - - '" Perimeter within which active teams can operate in safety
 The Planning of the Coup d'Etat 127

place, air transport will be of very great importance; either we

or the government could tip the balance of forces by flying in
quite small contingents of our respective supporters. The size of
the forces which can be moved by air may well be very small, but
in the context of the delicate balance of the active phase of the
coup, they could still playa decisive role.
Air transport is, however, very vulnerable insofar as it still
relies on long and uninterrupted landing strips; therefore, if at
all possible, we should avoid having to rely on it. To the extent
that we are independent of support arriving by air, we should
therefore prevent the use of all airfields in and around the area
of the capital city. Some of these airfields will be military ones,
but even if they are not they may still be heavily guarded. This
could be a serious obstacle if the government still controls
significant military forces outside the capital city and if transport
planes are available to bring them into it. Seizing a defended air-
field will certainly be difficult, but denying the use of one is very
easy. A few vehicles parked on the runway, either by covert
means or by a little cooperation from the inside, and 'covered'
by a small fire-team to prevent them from being moved, will
suffice to neutralize an entire airport. A few warning shots from
suitable positions could also prevent any landings taking
place.
Other organized forms of transport will only rarely be im-
portant in modern conditions. In many undeveloped countries
railways play a very marginal role in the transport structure.
Even where they are important economically, they will often be
removed from the main population centres, having been built
to connect mines and plantations with deep sea ports, as part of
the colonial export economy rather than as links between the
main population centres. In Europe and those parts of Latin
America where this is not the case, railways will still be unim-
portant from our point of view because of the time element. In
any case railways are extremely easy to neutralize. In the 1926
coup in Poland staged by Pilsudski, a great deal of the action
revolved around the railway system but rail-borne troops never
arrived in time to decide the issue, because both sides found it
easy to prevent the other's movements, though not to ensure
 128 Coup d'Etat
their own. Where, as in Ethiopia, the railways are important -
or rather the single Addis Ababa-Djibouti railway line is im-
portant* - technical neutralization should be used. Railways rely
on a technical chain system par excellence and if a single section
of rail or signals is sabotaged, the whole system will temporarily
stop. The gap between two sections of rail is easily crossed, but
probably there will be no rolling stock on the other side.
Public Buildings
The need to provide the bureaucracy and the masses with visual
evidence of the reality and power of the coup is one of the
continuing elements in our analysis. Otherwise, this will be the
least defined and coherent of our groups of targets. The build-
ings which we will have to seize, in every case, will be the resi-
dence of those government leaders whom we have selected for
arrest, and those buildings which house facilities that we require,
such as the radio-television building. In the first case it will be a
matter of a brief penetration to achieve capture or arrest; in the
second, however, we will have to seize and occupy the building,
and perhaps resist attempts made to recapture it. But there will
be other official buildings which we will also have to occupy, or
at any rate control the access to. Those can only be loosely
defined as those buildings whose possession is associated with
the possession of political power.
Most countries have some form of elected assembly, a
parliament or its local equivalent, but in many of them political
power emanates from the palace of the president or other ruler
(or the central committee of the party); we should not be
deceived by constitutional fictions, and after spending so much
effort distinguishing between effective political power and its
symbols, we will not make the mistake of using our scarce
resources on the latter.
Nevertheless, there will be certain symbolic buildings which
could play an important role in the crucial transitional phase of
the coup: their possession by one side or the other will act as a
signal to the masses and the rank and file of the bureaucracy in

*A significant part of the Ethiopian army which is not fighting the

Somali ski/tas is stationed in garrison towns along the single railway to the
sea.
 The Planning of the Coup d'Etat 129

the confused period when it is unclear which side is in control.

Our possession of those symbols will then give us the allegiance
of those who were waiting to choose one side or the other. Thus,
though useless in direct material terms, it may well be worth
while to seize those buildings which have a powerful symbolic
value. In the Ghana coup of I966 which brought down the
N'krumah regime the very efficient and practical-minded leaders
of the coup felt it necessary to fight their way into the presidential
residence, Flagstaff House, though it contained neither N'kru-
mah himself nor any important technical facilities. They realized
that though it was an empty symbol par excellence, its possession
was essential to secure the support of the Accra masses who
naturally associated the control of political power with that
particular building. Fortunately, by the very nature of such
symbols there will be one, or at most, two such symbolic build-
ings whose possession will be an essential requirement.
Apart from the purely symbolic buildings, there will be others
whose possession is highly desirable. These are the administra-
tive headquarters of the army, police and security services. Thus
in each case this group of targets will include the following:
(a) The seat of effective political power. This could be the Royal
or Presidential palace or the building of the elected assembly or
of the party Presidium or Central Committee;
(b) The main administrative buildings. The Ministry of Defence,
the Ministry of the Interior, police and military headquarters, if
separate;
(c) Symbolic buildings. Often the appropriate building will fall
into one or other of the classifications above; where, however,
there is a 'cultural' lag between the development of the country's
political life and the traditional attitudes, the masses will still
associate political power with an 'obsolete' building.
The coup will be practically over (in its 'active' phase) by the
time the citizenry wakes up and starts to investigate the posses-
sion of buildings symbolic or otherwise. We can therefore post-
pone the occupation of some of these targets to the later stages.
Since in direct practical terms other targets will be more
important, or at any rate more urgent, the best way of dealing
with the symbolic and administrative targets will be to use them
 130 Coup d'Etat
as assembly points for those teams which have already completed
their primary mission.

Neutralizing the PoHtical Forces D: particular groups

Which organized groups will be sufficiently strong to oppose us
even if the voice of the government is silent and the capital city
visibly in our hands? Not many, but we must remember that
even one well-organized demonstration, or a well-timed strike,
could pose a serious threat to the coup in the delicate transitional
phase. It is therefore essential to identify such groups and, once
identified, to neutralize them before the coup. Once it is known
that a coup has taken place, the leaders of the militant organiza-
tion(s) concerned will immediately prepare for action; they them-
selves will then be more difficult to arrest and their organizations
will be halfway underground.
In countries where political conflict is limited to the verbal
dimension, this kind of dramatic and rapid response to political
change will be unknown; but elsewhere, where political conflicts
can be violent and where all organized forces, whether primarily
political or not, can be drawn into them, this type of response is
more or less automatic. Political parties in the Middle East and
South America, trade-union movements in southern Europe,
religious movements in South Vietnam have little in common
except (a) their ability to respond in this way and (b) that even
without the weaponry that some of them have, they could be a
real threat to the coup. We will conduct our analysis in terms of
those three types of 'political force' because their features will
largely subsume those of other kinds of organized groups, which
may be relevant in particular countries. In the United States or
Britain, for example, where neither' trade unions nor religious
groups nor political parties are sufficiently militant to oppose a
coup after it has seized its initial targets, the groups which may
have this capability (such as para-military movements of the
paranoid right) will be organized in a manner which includes
features of all three.
One of the points which we must bear in mind is that not all
the organized groups which are important in normal political
 The Planning of the Coup d'Etat 131

life will also be important in the highly restricted and spasmodic

politics of the coup. Conversely, groups which in ordinary politi-
cal life are of very limited importance could emerge as real
threats. If, for example, we failed to neutralize the organization
of, say, the National Rifle Association in the USA, or the
National Union of Students in the UK, their reaction - however
ineffectual per se - could still endanger the coup by slowing down
the process of political stabilization in as much as they could
provoke conflicts which would re-open the whole issue. Other,
more prudent, groups would then re-examine the possibility of
challenging our position, while the use of violence to stop the
agitation of the groups we have overlooked could lead to further
opposition since the side-effects of violence would increase the
awareness of and hostility to the coup.
Finally, there are certain political forces which must not be
neutralized (apart from those groups which have agreed to
support us). These are those groups which are generally re-
garded as extremist but whose effective powers are limited. By
allowing them a certain freedom of action we will give them an
opportunity to oppose us, and their opposition will have two
favourable by-products: (a) we will be able to gain the support
of those political forces which fear them more than us; (b) we will
be able to step forward and fight other groups after having asso-
ciated them with the extremists in question. This can, however,
be a dangerous game to play; in the confused and dramatic
situation of the coup, the extremists could gain in power and
political support, and it is therefore possible that the time which
we have allowed them to discredit the opposition will work in
their favour.

RELIGIOUS ORGANIZATIONS
In many economically developed countries religious organiza-
tions no longer have much political power, though they may still
be an important social force. The leaders of religious groups can
be influential in social and, to a degree, political life, but the
allegiance of the believers is rarely expressed by direct and force-
fulactionin the political field. In economically backward countries,
 132 Coup d'Etat
and in those whose development is limited or very recent, it
is otherwise. Where the newer technology of man has only been
recently applied, or not at all, the older technology of God is still
of paramount importance. This can be a source of very consider-
able political power to the organizations which are identified with
the appropriate beliefs 'and which are able to channel the senti-
ments of the believers. Leaving aside local cults, which are too
fragmented to be important in terms of national politics, and
which in any case tend to be apolitical,* we see that even univer-
sal religions will differ in their degree of political involvement.
The role of the Catholic Church in Italy since the war illu-
strates the power which can be accumulated by a well-organized
religious group, even when operating in circumstances which are
unfavourable from the religious point of view. Though most male
Italians seldom or never go to church, Italian women are keen
and regular churchgoers. Italy being a democratic country
where women have the vote, it is obvious that if the organized
Church is willing to direct its followers to vote for a particular
party, that party will gain the bulk of the women's vote before it
even opens its electoral campaigns. The Church has generally
been willing to give such specific directions, and one particular
party has benefited: the Democrazia Cristiana (D C). Aided by
its assured majority of the female vote, the DC has ruled Italy,
alone or in various coalitions, since 1946, and it has done so
largely because of the support received from the Church. It is
hardly surprising therefore that the Church has been able to
dominate the DC and that, through the DC, it has influenced
every aspect of Italian national life.
This is no vague influence exercised on a plane of generalized
authority, but rather a constant supervision of political activity,
conducted at the provincial level by the bishops and at the
national level by the Pope and his associates. At each level of the
state bureaucracy the Church, directly or indirectly, exercises its
influence: on civil-service jobs and promotions; on the alloca-
tion of investment funds and of the various kinds of government
grants; ~n administrative decisions dealing with 'zoning' and
* Local cults may be important from the point ofview of the local adminis-
tration, but not in terms of national politics.
 The Planning of the Coup d'Etat 133
building regulations. This influence has brought its rewards.
While the facilities of the state bureaucracy have steadily
deteriorated compared with the dynamic private and semi-state
sector, the Church's educational and religious facilities have
steadily expanded; money to build and the permissions required
to do so have never been lacking.
If we failed to neutralize the organization of the Church in
Italy, it could inspire and co-ordinate opposition to us through
its capillary network of parish churches. Parishioners are used
to hearing political messages from the pulpit;* priests are used
to receiving detailed political briefs from their bishop and the
latter receive their instructions from the Vatican. Our neutrali-
zation of the telecommunications facilities will not prevent the
flow of instructions: the Vatican maintains its own radio station
and this could be used to contact directly the organization
throughout the country.
The Catholic Church plays a similar role in certain other
countries, where it has a 99'9 per cent nominal membership
and the status of the national religion, but the stronger state
structure of Spain and Portugal, let alone France, has denied it
the pre-eminent position it has in Italy. The intervention of the
Church would, however, bea powerful factor in much of the
Catholic world, including South America, especially if the motive
force behind the coup was identified as being anti-clerical.
Islam, which has the comprehensive nature of a religion, a
political system and a civilization, is still (though much decayed)
a major political force and its leaders play a recognized political
role. The 'doctors' of EI Azhar University in Cairo, one of the
main theological institutions of the Muslim world, are periodi-
cally prompted by the Nasir regime into openly political declara-
tions; no single leader in Islam has the authority of a Pope
because the system is less centralized, but in °each country the
local leaders are still very important. Even the spread of ' Arab
Socialism' has not impaired the position of Islam, and govern-
*In a recent and very successful Italian film, the parish priest was shown
explaining to his flock that he did not want to give them a pre-election brief,
but he merely asked them to vote for a parry which was Democratic and
which was Christian, ' ••• Democratico e Cristiano, Cristiano e Democratico••• •
134 Coup d'Etat
ments which follow an extreme left' line' in all foreign and some
domestic matters are still unwilling (or unable) to challenge the
status of Islam as the state religion. When such a course was
tentatively suggested by an obscure member of the present
Syrian government, the leadership (which follows an almost
Peking 'line' in all other fields) was forced to denounce him
officially. Whether this resilience means that the Islamic
leadership of particular countries could act as an active political
force is another matter. The structures of Islam as an organized
religion are fossilized; the fluid and dynamic aspect of the
movement in its early days has been replaced by a dogmatic and
extremely conservative set of beliefs, whose inflexibility is one
of the causes of the present travail of the Arab world.
The political sterility of Islam in recent times has meant" that,
though it has been used by governments to propagate their
political initiatives, Islam per se has only acted when a direct
attack has been made on religious orthodoxy.* Consequently,
unless our coup has a definite anti-Islamic colouring, religious
leaders in Muslim countries will not initiate any action against
us. We must therefore prevent our opponents imposing such a
colouring on our coup.
In the intermittent political warfare between' Arab-Socialists'
and the monarchies, while the latter are accused of being' tools
of the "Zionist-imperialist oil monopolies''', the former are
accused of wanting to eradicate Islam with their godless beliefs.
Actually even the soi-disant 'progressives' would not dream of
challenging Islam, which is after all the main factor - through
the language of the Koran - which binds the Arab countries,
separated as they are by both history and geography.
Thus, with the one qualification made above, we can ignore
Islam as an active political force. The same goes for Hinduism
which, though otherwise very different, shares the passive
political role of Islam. Though some political forces in India have
successively made use of Hindu sentiments, religious leaders as

*This and subsequent statements about Islam and the Arab world refer
to Sunni Islam: the heretical Shi's sects and their offshoots are a different
matter. Their political and religious leadership is often embodied in the
same person, and they are often politically very active.
 The Planning of the Coup d'Etat 135
such have never actually initiated any major political action.
(Even the periodic agitation against cow slaughter is instigated
by the extreme right parties.)
An extreme example of the potentialities of a dynamic
religious leadership is the C main line' Buddhist movement in
South Vietnam. The almost continual warfare of the last genera-
tion and the politically destructive effect of the Diem regime have
led to a collapse of the social and political structures of the
country, while its economy has been reduced to localized sub-
sistence agriculture, allied with urban dependence on US aid.
In this situation the newer economic, political and social forces
have become extremely weak and the groups based on the older
religious affiliations have emerged as the only valid civilian
forces in Vietnamese society. Apart from the main line Buddhist
movement, led by Thich Tri Quang and other regional leaders,
there is the following alignment of forces (early 1968):

Hoa Hao: a reformed Buddhist group with a large following

in the southern (Delta) part of the country. Their leadership is
politically orientated and, except for strictly local alliances, is
anti-Viet Congo They appear to have built up the rudiments of
an armed militia.
Cao Dai: an important Buddhist sect which has a history of
political participation.
Binh Xuyen: a small but very active part-sect and part-secret
society. Its main area of strength is in the Saigon region, and
before the Diem regime displaced them, the Binh Xuyen used
to 'own' the city's police force - and its underworld. The sect
has been influenced by the Chinese secret societies from across
the river in Cholon, and the effect of repression at the hands of
Diem has been to drive it underground rather than to destroy it.
Catholics: until Diem's fall, the large Catholic minority was
able to dominate the Buddhist majority. Many of the South
Vietnamese community are refugees from the North; under the
French many Catholics cooperated actively with the colonial
power and served in the French armed forces. Now that the
South appears to be going the way the North went in 1954, the
community has reached a desperate impasse. Their activity
 136 Coup d'Etat
against a pro-Viet Cong (or just pro-peace) coup would be
immediate and probably very effective.

All these religious groups could intervene against a coup: their

meeting places could be used to assemble and shelter our oppo-
nents; the priesthood could inspire and coordinate mass
agitation against us; finally, their direct influence on the army
and the bureaucratic rank-and-file could be used to resist the
imposition of our authority.
The religious groups which can be important in particular
countries will differ doctrinally, but will be organizationally
sufficiently similar to adopt a general method of neutralization.
If they operate private broadcasting facilities, such as the
Vatican Radio or the small radio stations of American missionary
sects in many parts of the world, we will put them temporarily
out of action. Religious meeting places should not be closed by
administrative methods, which are liable to foment rather than
stifle opposition, but access to them should be barred by
'incidental' roadblocks.
The leadership of religious organizations presents a special
problem in terms of neutralization: because of their particular
psychological role in the minds of public opinion it will usually
be extremely unwise to arrest the actual hierarchic leadership.
Fortunately, the actual decision-makers within religious organiza-
tions will often be younger men who are not in the public eye
but who are the key figures from our point of view. If the real
decision-makers are not also the hierarchical leaders, we will
arrest them; but if the two roles are embodied in the same person
or persons, we will not do so. In concrete terms, a Thich Tri
Quang, who is an effective decision-maker, but not formally in
the higher leadership, can (and should) be arrested; but a Pope,
who is both the representational and the effective leader, cannot
however be arrested without stimulating a great deal of opposi-
tion the impact of which will outweigh any advantage to be
gained from the arrest.
 The Planning of the Coup d'Etat 137
POLITICAL PARTIES
Unlike the other groups which constitute a potential source of
opposition to the coup, political parties are our direct competitors,
in the sense that their primary purpose - like our own - is the
accumulation of political power.* This will not necessarily make
them the main, or even a significant, potential threat to us, but
it will mean that their response to the coup will be particularly
prompt. Whether this response will be verbal and purely
declaratory, or whether it will be more direct and effective, will
depend on a variety of factors, including the nature of their
leadership, organization and membership. Because political
parties are as diverse as the countries within which they compete
for power, we will classify them in certain categories as a prelude
to examining the methods of their individual neutralization.

~ Machine' Parties
Where politics is a business like any other, parties take the form
of an association whose purpose is the procurement of votes in
exchange for ~pecific and material rewards. The local •boss'
secures votes for the party at election time in exchange for cash
and/or bureaucratic jobs for himself or his nominees. The
deputies in the Assembly then deliver their votes to the govern-
ment in exchange for definite favours, some of which are retained
and some of which are passed down to those who secured their
election. The 'machine' party can flourish in societies as dif-
ferent as early twentieth-century America, Egypt between the
wars and present-day South America. It needs two main in-
gredients: an elective parliamentary democracy and a socially
backward electorate. In the United States at the beginning of the
century the immigrant communities were largely composed of
eastern and southern Europeans, whose mother-countries were
economically, and often politically, unsophisticated. Thus the
newly-arrived immigrant lacked the political awareness required
to obtain direct concessions from the government, in the shape
of social welfare legislation or labour codes. He soon learnt,
however, to obtain indirect favours by promising his support to
*This is their purpose. Their junction, however, is to aggregate interests.
 138 Coup d'Etat
the local ward organization of the party - i.e. if the votes were
delivered on election day and the candidate elected, rewards
would eventually be received in return. Present-day 'machine'
parties, such as the Democrazia Cristiana in parts of Italy..
AccMn Democrdtica in Venezuela, and MNR in Bolivia, do not
distribute their rewards as widely as the old municipal machines
in the United States. This is because these parties participate
in the empleocracia, which dominates political societies in which
industry and commerce are undeveloped and agriculture the
monopoly of peasant subsistence farmers or aristocratic land-
lords. In these conditions politics and the associated jobs in the
state bureaucracy are the main avenues of middle-class enterprise,
and the party is the framework (together with legal training) for
the middle-class activity of office-hunting.
Machine parties have their rationale in the contrast between
the constitutional structures and the social ones in countries
which are both poor and 'democratic'. Their whole manner of
operation revolves around the exchange of votes for rewards at
every level; in other words it requires the functioning of the
parliamentary apparatus, with its periodic elections. In the event
of a coup this institutional framework would be frozen and the
machine made powerless. Even if the machine has a base of
mass support, its leadership, being a coalition of local power
structures without a national 'presence', will not be able to
mobilize it. We will therefore ignore the 'machine' parties and
will not need to take any particular action in their regard.

t Insurrectional' Parties*
Such parties mayor may not participate in open political life
(if it exists in our target country) but the primary purpose of
insurrectional parties is to destroy the system rather than ,to
work it. Like the Bolshevik party before 1917, the Muslim
Brotherhood in Egypt, Communist and extreme right-wing
parties in many parts of the world, these parties live a semi-
legal existence, with a cellular organization, an 'underground'
mentality and, frequently a para-military element. Such parties
*The alternative term, 'revolutionary parties' has left-wing connotations,
while insurrection covers both extremes of ~e spectrum.
 The Planning of the Coup d'Etat 139
are characterized by their adherence to a set of definite ideologi-
cal beliefs, a rigidly centralized organization and their preoccu-
pation with the use of direct methods to achieve political ends.
In the social and economic conditions of Western Europe
and North America, insurrectional parties are numerically weak
and their challenge to the system is usually conducted in an
atmosphere of unreality, though from time to time they can
gather a mass following among certain sectors of the population
which are outside the mainstream of national life. The 'Black
Power' movement in the United States, for example, has all the
traits of an insurrectional party, but it only operates amongst the
Negro communities in the ghettoes whose social and economic
conditions are those of an economically backward society. In the
'Third World', however, the constant pressure of economic
deprivation can generate a revolutionary mentality amongst wide
sections of the population, which insurrectional parties try to
channel and exploit. Their organization, however, is often
inadequate to the task and much of the endemic guerrilla activity
in Latin America and the insurgencies in South-East Asia are
outside their control.
Insurrectional parties can oppose us in three main ways: (a)
through the agitation of the masses, to the extent that they have
a mass following; (b) by direct means, such as assassination and
sabotage; (c) by syndicalist agitation. Insurrectional parties
usually have an authoritarian leadership structure and much of
their strength, in the confused circumstances which would
follow a coup, would derive from the coherence of a centralized
leadership. We should therefore make every effort to identify
and Cisolate' their key decision-makers. The emphasis on party
discipline and the habit of waiting for C directives' from the
higher leadership render many insurrectional parties powerless
once the leadership has ceased to function. Though the social
pressures which are the sources of strength of an insurrectional
party may lead to its revival, this will not take place in the short
period of time which concerns us. This vulnerability of in-
surrectional parties was strikingly demonstrated in the case of
the Muslim Brotherhood, Al-Ikhwan al-Muslimun. The Ikhwan
was a major force in Egyptian political life after the war, and its
140 Coup d'Etat
large mass following, its network of economic and educational
activities and its para-military youth groups gave it a great deal
of direct power. Its effectiveness, however, derived largely from
the coherent leadership of its founder, Shaykh Hasan al-Banna,
and the movement rapidly declined after his death (in unex-
plained circumstances) just after the failed coup of late 1948.
Where necessary, therefore, the committee or personal leadership
of the insurrectional party should be arrested and held in isola-
tion for the duration of the coup. Because of the emphasis on
party discipline, the beheaded movement will probably abstain
from action in the short but critical period following our seizure
of power.
Para-Bureaucratic Parties
In one-party states, such as those of Communist countries, much
of Africa, and Mexico, the party has lost its major role: the
competition for the allegiance of the masses. Being in a position
of monopoly, the party is also in danger of appearing super-
fluous. But, like any other bureaucratic organization, the party
can survive the loss of its primary function, either as a system of
spoliation or as an ancillary or supervisor of the administrative
bureaucracy of the state. Mrican parties, formed during the
political struggles which preceded independence, have often
legislated for their monopoly of power as soon as they have
attained it. Some, like T ANU (Tanzanian Mrican National
Union), have turned into constructive galvanizers of the com-
munal and state development programmes; others, like N'kru-
mah's old party in Ghana, became adjuncts to the personal
leadership and a system of outdoor relief for his Cactivist'
followers. The majority, however (until swept away by the
military dictatorships), have acted as the principal agent in the
main local industry: politics.
The para-bureaucratic party treats the state bureaucracy as
its subordinate. It investigates its activities, reports on its
behaviour to the higher leadership, and often demands special
privileges and concessions. These parties do not have a mass
following except within the framework of normal political life,
when they can be relied upon to produce demonstrations/or this
 The Planning of the Coup d'Etat 141

or that stand of the leadership. As soon as the hold of the

leadership is threatened, as soon as the police apparatus no
longer acts as its 'muscle', the para-bureaucratic party dissolves.
We can therefore ignore it in the active stage of the coup. How-
ever, its secondary function, intelligence and security, will be
important and will be dealt with as part of our general defensive
measures towards such organizations.

Parties in Developed Countries

Whether it is a two-party system, as in much of the Anglo-
Saxon world where parties are in effect coalitions of pressure
groups, or whether they are the class- or religion-based parties
of much of continental Europe, the major political parties in
developed and democratic countries will not present a direct
threat to the coup. Though such parties have mass support at
election time, neither they nor their followers are versed in the
techniques of mass agitation. The comparative stability of
political life has deprived them of ~he experience required to
employ direct methods, and the whole climate of their operation
revolves around the concept of periodic elections. Even where,
as in France and Italy, there are large and nominally revolu-
tionary parties, two or more decades of parliamentary life have
reduced their affinity with revolutionary methods.
The apparatus of the party, with its branches and local
organizers, can, however, allow them to perform a role of
information and co-ordination which could be potentially
dangerous. Even though their leadership may not take any
action, the apparatus can still serve as the framework for anti-
coup agitation. We will therefore close - administratively - the
network of branches, and this should be sufficient to neutralize
this particular threat.
The only serious threat from this direction will come from the
trade-union movements which are affiliated to the mass parties
of the Left. Their experience of industrial agitation has provided
a natural training for mass intervention against a coup, but this
will be dealt with separately below.
142 Coup d'Etat

TRADE UNIONS

Wherever there is a significant degree of industrial development,

and in many countries where there is not, trade unions are a
major political force. Because of their experience of industrial
agitation, which can be readily applied to political purposes, the
response of trade unions to the coup could constitute a serious
danger to us. The mass following of trade unions - unlike that
of political parties - is in continuous session: polling booths are
only open once every five years, but factories work all the year
round. The immediacy of the threat presented by trade unions
will depend on their size, cohesion and degree of militancy: the
fragmented syndicalism of Britain, with its purely electoral
politics, would not, for example, add up to the threat of, say, the
Italian movement with its centralization and long history of
political strikes.
The experience of Bolivia since the 1952 Revolution illus-
strates how a single trade union and its activities can dominate a
country's political life. Bolivia is one of the poorest countries of
South America, whose economy is characterized by subsistence
farming and the activities of the large tin industry. Before the
Revolution and the nationalization of the mines owned by the
Patino, Aramayo and Hochschild interests, the miners had
worked in physical and economic conditions of extreme harsh-
ness. Following their emancipation, they naturally wanted to
achieve immediate and substantial improvements in these condi-
tions and COM I B 0 L, the state tin-mining organization,
started immediate reforms.
It was soon discovered, however, that the geological alld
economic conditions of the industry required an increase in
productivity which could only be achieved by introducing much
new machinery and reducing the labour force. As the only source
of capital was the United States, the miners' leaders opposed the
reforms on the dual plank of no 'yanqui' capitalism and no
redundancies. Such problems are familiar from nearer home,
but the crucial difference was that the miners were also an army.
They had been armed by the middle-class leaders of the
revolutionary MNR (Movimento Nacionalista Revolucionario)
 The Planning of the Coup d'Etat 143
Party in order to act as a counter-weight to the old army domi-
nated by associates of the mine-owners. The Revolution dis-
banded the army so that the miners could not only exert political
and economic pressures but could also use more direct, military
methods. Until the MNR leadership found a counter-weight in
the unions organized amongst the peasant farmers, the Indian
campesinos, who were also armed, the miners had things pretty
much their way. Led by militants of the Catavi-Siglo Veinte
mines, the miners imposed their control on COM I B 0 L, and
therefore on the country which depends on it as the only major
source of foreign exchange. Certainly no coup could have held
on to power without the miners' consent, and even if the central
institutions in La Paz could have been seized, the real power base
in the mines would still have been under the control of the union
leaders.
Even without the special circumstances which operate in
Bolivia, trade unions will often be a major political force,
especially in terms of the situation immediately following a coup.
But much will depend on the particular organizational structure
of the trade unions, and crucially on the degree of effective
centralization and the nature of their political affiliations. In
Britain, the main focus of decision-making within the trade-
union movement as a whole is the executive of individual unions,
but in some of them it has shifted to the shop floor. Apart
from this fragmentation, which would at least impair the speed
of reaction to a coup, the largely mainstream politics of British
labour would not be a suitable framework for direct measures.
In France and Italy, the trade-union movement is not divided
on craft lines, as in Britain, or on industrial lines, as in the
USA and much of northern Europe, but on political lines.
Individual industrial unions are affiliated to central organiza-
tions which are associated with particular parties. In both
countries the largest organization is controlled by the local
Communist party, with smaller Social-Democrat and Catholic
ones. The Communist organizations, CGL in Italy and
CGT in France,* both have a long experience of political
*ConJederation Generale du Travail (CGT); ConJederazione Generale dl
La'/JOTo (C G L).
144 COUp d'Etat
activism expressed in 'political' and 'general' strikes, pheno-
mena of almost purely historical significance in Anglo-Saxon
countries.*
Unless our coup is directly linked to the Communist party,
the central organizations of French and Italian trade unionism
would react to it, and do so in predictable ways. Immediately
after the coup they would: (a) try to establish contact with other
'democratic forces' to form a popular front opposition; (b)
contact their national network of branches to coordinate a
general strike; (c) put into execution their contingency plans for
'underground' activity and illegal survival. The only tactic
which would present a threat to us is the general strike, which
would be organized with the deliberate intention of 'confront-
ing' the forces of the coup. Our general measures would affect
the overall performance of this emergency programme, but
specific action would be needed as well in order to avoid the
confrontation which the unions would probably seek. Both the
C G T and the C G L have memories of the wartime resistance
movements: both are aware of the destabilizing nature of open
repression and they would therefore try to provoke us into using
violence.
Though some form of confrontation may be inevitable, it is
essential to avoid bloodshed, because this may well have crucial
negative repercussions amongst the personnel of the armed
forces and the police. The avoidance of bloodshed· in tense
crowd situations is a matter of technique, and competent
handling of our incorporated armed and uniformed forces will be
essential.t
The incidents of Reggio Emilia in Italy in the summer of
1964 in which seven people died following a 'political' strike,

*Cf. the events of May I968 in France.

tThe renewal of political activism in recent years in both Britain and the
United States has intensified the search for e1fective means of controlling
mobs. The study of mass psychology and the development of assorted
gadgetry should not obscure the fundamental principles of mob controL
These are: (a) the need to keep the mob in open spaces so that claustro-
phobic and physical pressures are avoided, and (b) the need to break down the
anonymity of the individual in a mob by making selective arrests.
 The Planning of the Coup d'l1tat 145
illustrated how an incompetent police force can impair the
authority of the government it is trying to protect.*
If the trade unions of our target country approximate to
Franco-Italian levels of political effectiveness it will be necessary
- assuming that our coup is not politically linked with them - to
identify and arrest their leaders and close their headquarters in
order to impede the operation of their secondary leadership.
Elsewhere, it will be a matter of orientating our general measures
to deal with the particular threats which trade-union movement
could prevent.
*Cf. the destabilizing effects of the behaviour of the police in Paris on the
night of Friday 4 May 1968. which was the detonator of the crisis.
5. The Execution of the Coup d'Etat
C As soon as the moral power of national represen-

tation was destroyed, a legislative body, whatever

it might be, meant no more to the military than a
crowd of five hundred men, less vigorous and
disciplined than a battalion of the same number.'
Madame de StaeZ referring to Napoleon's coup
d'etat

C I came in on a tank, and only a tank will evict

me! Abu Zuhair Yahya, Iraqui Prime Minister,

1968

The active phase of a coup is like a military operation - only more

so. If the general principle of tactics is the application of force
at the right place, the coup achieves this with surgical precision
by striking at the organizational heart of the whole state; if
speed is often important in military operations, in the coup it is
an essential requirement. But the coup differs from most military
operations in one crucial respect: while in war it is often ad-
vantageous to retain some forces as reserves to be used in later
(and possibly more critical) phases of the conflict, in a coup the
principle of total commitment applies. The active stage takes
place in one short period of time and forces held back today will
be useless tomorrow: all our forces must therefore be used in the
one decisive engagement.
The fact that the coup has practically no time-dimension
means that we will rarely be able to correct errors made during
its execution; in war, tactics can be changed, weapons can be
replaced, plans re-shaped and men i-e-trained on the basis of
combat experience, but in the coup there will not be sufficient
time for any feed-back to operate. In this the coup is similar to
the most modem form of warfare, the strategic missile strike,
and the time factor places the entire burden of decision-making
 The Execution of the Coup d'Etat 147
in the planning stage. Every target must be studied in detail
before the coup. The team assigned to seize it must match it in
terms of size and composition; its every move must be planned
in advance and no tactical flexibility can be allowed.
With this degree of detailed planning, there will be no need
for any sort of headquarters structure in the active stage of the
coup; for if there is no scope for decision-making there is no
need for decision-makers and their apparatus. In fact, having a
headquarters would be a serious disadvantage: it would con-
stitute a concrete target for the opposition and one which would
be both vulnerable and easily identified. As soon as the coup
starts, the ruling group will know that something is happening,
but unless coups are very frequent in the country, they will not
know what that something is; it could be a mutiny, an insurrec-
tion, the opening of a guerrilla war, or even the beginning of an
invasion by a foreign power. All these forms of conflict represent
threats to the regime, but they are all different in terms of their
immediate significance and - more important - in terms of the
measures which are required to meet them. We should avoid
taking any action that will clarify the nature of the threat and
thus reduce the confusion that is left in the defensive apparatus
of the regime. Our teams will emerge from their bases and
proceed to seize their designated targets while operating as
independent units; their collective purpose and their coordina-
tions will thus remain unknown until it is too late for any effective
opposition. The leaders of the coup will be scattered among the
various teams, each joining the team whose ultimate target
requires his presence; thus the spokesman of the coup will be
with the team which will seize the radio-television station and the
prospective chief of police will be with the team whose target
is the police headquarters. As each team will be both small and
highly mobile, and as there is no headquarters throughout the
active phase of the coup, the opposition will not have any single
target on which it will be able to concentrate its forces. In this
way their numerical superiority will be dissipated and the
smaller forces of the coup will have local superiority in the area
of each particular target. This will be the key of the victory of
the coup.
On the Eve
In the last two chapters we have surveyed the planning of the
coup in terms of the neutralization of the 'professional' defences
of the state, and the selection of those targets which will assist
the neutralization of the 'political' forces. We analysed the
structure of the armed forces and the other means of coercion,
and we saw that much of the armed forces, a significant part of
the police system, and some of the security services, could not
intervene - either for or against us - in the event of a coup. This
was due to their remote location, dispersed deployment, or
because their training and equipment were inadequate - or over-
specialized. Then we infiltrated the relatively small part of the
apparatus which did have an intervention capability, so that
much of it was technically neutralized and some of it totally
subverted. This will ensure for us the neutrality of much of the
defences of the state and the active cooperation of some of its
parts. The infiltration of the army and police has given us an
instrument: the units which we have incorporated and which
form the forces of the coup. We have also prepared for the use
of this instrument by selecting the targets on which it will be
used: we have identified the physical targets which must be
seized and those which will have to be sabotaged or otherwise
interdicted; we have selected the leading personalities amongst
the potential opposition, both in and out of the government, and
prepared for their arrest.
But one major task, however, has not been covered in the
planning stage: the forcible isolation of the 'hard-core' loyalist
forces. Hopefully, the strength of those forces which we have
been unable to infiltrate and which also have an intervention
capability will not be very great. But even if they are weak in
absolute terms, we dare not ignore them. To do so, would be
to invalidate all the measures we have taken to insulate the
capital city - and ourselves - from the intervention of hostile
forces. The extreme instabHity of the balance of forces during
the active phase of the coup means that what in other circum-
stances would only be a minor threat could then have disastrous
 The Execution of the Coup d'Etat 149
consequences, and if the 'hard-core' loyalist forces are large in
relation to our own, we will indeed have to divert much of our
forces to their isolation.
Though we have been unable to penetrate these 'hard-core'
loyalist forces, two things will probably have been achieved:
(a) their number, quality and location will be known to us, and
(b) our general measures of neutralization will have reduced
their overall effectiveness. Their fighting capability will not have
been eroded but, as Table 14 illustrates, their intervention
against us will be delayed and disrupted.
Our purpose is not to destroy the loyalist forces militarily
(since we will be able to deal with their cadres administratively
after the coup) but merely to immobilize them for a few crucial
hours. The tactics which will be used must be exclusively
defensive: a ring of blocking positions around each concentra-
tion of loyalist forces or, if this is not possible, a similar ring
around the capital city. Thus, though we will be on the strategic
offensive (in the sense that we are the ones who want to change
the situation in general) we will also be on the tactical defensive,
and this will -give us important technical and psychological
advantages. By using defended roadblocks to isolate the loyalist
forces, we will put the onus of initiating any fighting on them:
our forces will be content to wait and it will be the loyalist forces
which will try to pass through. Should a column of loyalist
forces arrive at the roadblock, their leaders will be faced by
opposite numbers wearing the same uniform, and belonging to
the same corporate entity, perhaps even to the same regiment.
Both sides will state that they are 'obeying orders', but inter-
estingly enough, the 'orders' of the leaders of our forces will
probably appear more legitimate than those of the leaders of the
loyalist troops. Owing to our arrests and our interdiction of the
physical facilities, the 'legitimate' orders will probably have
taken an unusual form: the source of the orders to the loyalist
troops will probably be somebody other than the appropriate
superior in the hierarchy; the method used to convey them will
probably be an unusual emergency one; and the actual orders
indistinguishable 'in form from ones which might have been issued by
the planners of a coup.
e.E,-8
 ISO COUp d'Etat
Thus the officers of the loyalist forces may have received
orders stating, 'Move into the city centre, hold the Parliament
building and the radio station.' The leadership may have added
that they would be acting against the forces of a coup but, even
so, such orders would have insurrectional undertones. When
army officers find themselves doing unusual things, their natural
reaction is to try and fit them into familiar patterns; the most
familiar pattern of all will be to arrive at the conclusion that the
'politicians are guilty of yet another "mess" '. The most prob-
able course of action will be to request clarification from their
superior officers. It is to be hoped that these officers will have
either decided to remain neutral, or else have been arrested; in
either case the' clarification' will never arrive.
If, on the other hand, the loyalist units decide to force the
roadblock, we will benefit from the tactical advantages of the
defensive. These include the opportunity of choosing the place
(natural obstructions such as bridges and tunnels) and the
opportunity of deploying and camouflaging weapons and men.
In order to make the fullest use of both the psychological and
the tactical advantages, the blocking position should have a dual
structure: a (largely symbolic) first line composed of some suit-
able physical obstacle, such as cross-parked heavy, vehicles, with
a few men bearing' orders' which forbid all passage; beyond this,
there will be a second (military) line, much stronger numerically,
with weapons and men deployed to repel an eventual assault
(the operational detail involved is discussed in Appendix B). The
defenders of the blocking position should inform the incoming
loyalist forces that there is such a second line of defence: this
should be partly camouflaged and would then serve as a deter-
rent even if it is numerically weak compared to the opposition,
since its real strength will be difficult to ascertain.
The situation at the blocking position will require very delicate
hmdling and it will be necessary that the soldiers on our side
understand that their primary function is to avoid a conflict,
rather than to engage in it successfully. In concrete terms, their
mission will be a delaying operation rather than a decisive one
and this will have precise implications in terms of the weapons
and tactics to be employed.
 The Execution of the Coup d'Etat lSI

TABLE 14. The Mechanics of Intervention of the Loyalist Forces

Phase Effect of our

general measures

I. Police/security agency Telephone exchange has been

personnel raise initial alann" seized, telex cable links have
and seek to contact their been sabotaged, radio relays are
HQ. shut off. They must therefore
send a verbal message.
2. Police/security agency HQ As above for communications.
verify the reports and realize Some messengers fail to arrive
the seriousness of the threat. as focal traffic points are
H Q tries to communicate gradually occupied.
with political leadership.
3. Political leadership calls for As above for communications.
army and police intervention. Some units missing from their
barracks; others refuse to move;
others cannot move because of
technical neutralization.
4. Political leaders begin to As above for communications.
realize the extent of our Only military radio links
infiltration of the anned can be used to communicate
forces and police. Loyalist with loyalist forces.
troops respond.
5. Uninfiltrated forces assemble Many political leaders no
and prepare for intervention. longer available; some
They try to reach political arrested and some in hiding.
leadership for a confirmation
of their orders. Some defect
to us, others choose
neutrality, but some remain
under the control of the
government.
6. Loyalist forces move on to Airports closed and landing
capital city or if already strips interdicted. Railways
within its area, move in to interrupted and trains stopped.
the city centre. City entry points controlled by
our roadblocks.
Loyalist forces in capital city area are then isolated by direct means
 6. OPERATIONAL SEQUENCE AND TIMING
J - - - - - - - - - - - - - - - - - - - - z e r o hour
Area of surveillance of security agencies penetrated
!
A TEAMS
!
B TEAMS
~
t; TEAMS

1 ReSidenc! of target
reached and entered

."
Primary target
Targetprb'ated
(e.g. radio/TV station)
51
Facility neutralized ~
Q,
DefenSivJJI
sub-team (e.g. telephone exchange)
stays on
target " ~
Personality X brought III
to place of confinement ~
Team dispersed into ~
small groups assigned CI)
E

1
to secondary targets t=

1
Symb~lic
Target evacuated and
team joins pool of reserves
pubbg Team joins pool
building A • ofreserves
Symbolic
public /
building B

Focal traffic point Re$erves

Distributed in positions
at key traffic points
,
 The Execution of the Coup d'Etat 153

Timing, Sequence and Security

Ideally, the timing of the coup will be completely flexible so that
we can take advantage of any favourable circumstances which
may arise, such as the temporary absence of the leadership from
the capital city or the outbreak of some coincidental civil
disorders. This flexibility, which would be highly desirable, is
only rarely possible because the infiltration of the army and
police will be a dynamically unstable process: the circle of those
who have decided to join us will grow and continue growing as
a 'band-wagon' effect is generated; but unless the coup materia-
lizes, there will eventually be a movement into neutrality or
even opposition. Meanwhile, the danger of denunciation will
also increase as more and more people become aware that a
.coup is being planned, or at any rate that 'something is up'. The
timing of the coup will therefore be dictated by the progress of
our infiltration of the armed forces and police, and as soon as a
satisfactory degree of penetration is achieved the coup must be
executed. This implies that it will not be possible to designate a
date well in advance of the coup which can be communicated to
the various teams. This is just as well, since it means that the
date cannot be leaked to the security agencies. Actually, it is
quite likely that some information about us will have reached the
security agencies, but this should not affect the outcome. As the
preparations for the coup proceed, more and more information
about our actions will be in circulation but it will also be in-
creasingly obscured by 'noise'.*
Every move we make will generate information which could
eventually reach the security agencies, but the consequences and
misinterpretations of our actions will generate an equal or greater
amount of ' noise '. This will make it increasingly difficult for the
analysts of the security agencies to identify the nature of the
threat, since their capacity for processing information is not
ur.1imited. This process is illustrated by Figure 7, in which
O-Z is the normal level of 'noise' which is received at all times,
*An expression used in the Intelligence community to describe the false
or irrelevant information which is reported alongside 'hard' data.
 IS4 Coup d'Etat

7. INTELLIGENCE 'NOISE' AND ANALYSIS

I/)
w
U
zw
~
~
II:
:::l
(,)
w
VI
>
10
o
w
~
w
~ A Processing capacity of
~ analysts at security
~ agencies
:!!
II:
a
II.
:!:

z
LEVEL OF ACTIVITY OF PLANNERS OF THE COUP )
0<

O-A is the processing capacity of the analysts at the security

agencies, and X is the point beyond which the total flow of data
exceeds processing capacity so that .each item of real data is
accorded a diminishing amount of attention.*
Even if the security agencies could isolate the real data from
the 'noise', theywiU not usually take immediate action. Their
professional instincts will be to try to explore all the ramifica-
tions of the plot so as to be able to arrest all its participants. And,
*The specialized nature of the security agencies' work impedes a rapid
expansion of their facilities, and even if such an expansion could take place,
it would only be carried out if and when the real threat is identified. This is
precisely what the 'noise' problem prevents them from doing.
 The Execution of the Coup d'Etat 155
hopefully, the coup will actually take place while the security
agencies are still engaged on these explorations. But the latter
will be aware of this timing problem and are therefore likely to
respond to a possible threat by going ahead and arresting those
of the planners of the coup who have been identified. This
'nervousness' presents a special problem on the eve of the coup
because our final preparations will probably generate a sharp
increase in the total flow of information received by the security
agencies. Even without separating hard data from 'noise', the
mere ~ncrease in the total flow of information could be inter-
preted as a danger signal (as it certainly would be by competent
analysts) and this might trigger off the arrests.
In practice, it will rarely be possible to achieve total security
within the forces of the coup and we should assume as a working
hypothesis that they have in fact been infiltrated by the security
agency. This leads to the general defensive procedures discussed
in Chapter 3, but it will also have precise operational implica-
tions:
(a) Each 'team' will be told well in advance what equipment
and tactics will be required to seize its particular target, but not
the exact designation of the target.
(b) Each 'team' will only be told its designated target when it
actually receives the signal to proceed to its seizure.
(c) Each team will be alerted individually: with only as much
advance warning as it requires to prepare for its particular task,
instead of a general go-signal for all teams.
Since the teams will have different starting points and dif-
ferent targets to go to, the use of anyone general signal would
either give insufficient warning to some teams or an unneces-
sarily long one to others. The longer the time-span between the
announcement that the coup is 'on' and its actual execution, the
greater is the likelihood that information will reach the securitY
agencies in time to prevent the successful execution of the coup,
since this will be the moment at which their operative in our
ranks could send out a warning.
The problem of warning-time and lead-time is illustrated in
Figure 8.
If we give all our teams a ten-hour warning period, by sending
156 Coup d'Etat

8. LEAD-T.JME REQUIRED BY TEAMS TO REACH

THEIR TARGET BY ZERO HOUR

Team No.1
I
I
Team No.'2
I I
I r
I Team No.3
J I I
I
I
I
I Team No.4
r I I
I I I
I I : Team No.. 5_
I
I
: ,
I I :
Hour:-10-9 -8 -7 -6 -5 -4 -3 -2-1 0 ",1 +2 +3 +4
Intended start of th~ coup J

out a general call at hour - 10, then team No. I will just about
reach its target in time, but all the other teams will have received
C excess warning', or in other words, information will have been

distributed before it was essential to do so. If we give all the

teams a two-hour warning period, then 'excess warning' will be
zero but team NO.5 will reach its target several hours before
team No. I and those defending it will probably be on a full
alert. The solution appears to be a simple one: make warning-
time equal to lead-time so that each team is alerted just in time
to allow it to reach its target by the zero hour.
But in reality the problem is more complex. It is not a matter
of simultaneous arrival at the target but rather of the simul-
taneous penetration of the Cearly warning system' maintained
by the security agencies of the state. If, for example, team No.2
has to cross the entire capital city to reach its target, the security
agency will probably be alerted as soon as it enters the city at,
say, hour - 2. Thus by the time that team NO.4 reached its target,
the opposition would have had two hours to prepare for its
defence. We will probably have very little information on the
functioning of the security apparatus, but we can operate on the
assumption that a team (if it is large and/or equipped with
 The Execution of the Coup d'Etat IS7
armour) will be noticed and· reported as soon as it enters the
capital city. We must therefore secure: (a) the protection of our
security position against an internal threat, which is achieved
by minimizing 'excess warning time', and (b) the protection
of our security position against external observation, which
is achieved by simultaneous penetration of the capital city
area.
Both aims will be achieved by sending the teams into action
at a time corresponding to their 'lead-times' to the capital city
boundary (or other applicable perimeter). This is illustrated in
Figure 9.*

Into Acdon
The actual execution of the coup will require many different
qualities: skilful diplomacy at a blocking position confronted by
loyalist forces; instant personnel management at the radio-
television station in order to persuade its technical staff to co-
operate with us; considerable tactical abilities in the case of
targets which are heavily defended. Our resources will probably
be too limited to form fully-specialized teams out of the pool
of those units and individuals which we have incorporated, but
we should nevertheless match broad categories of targets with
appropriate teams. We can distinguish between three such
categories of targets and their corresponding teams:

A-TARGETS
These are the more heavily guarded facilities with strict pass
control, such as the Royal or Presidential palace, the central
police station and the army H Q. In times of crisis, of course,
such facilities may be provided with fully-fledged military
defences, and in many countries the crisis is permanent. Partly
in order to minimize bloodshed, which could have a destabilizing
effect on the situation and partly in order to reduce the total
*1n the diagram the 'early warning system' is shown as a clearly delin-
eated perimeter but, of course, in reality it will be a general area with
vaguely defined borders. We will adopt as a perimeter whichever approxi-
mation suits the circumstances.
1S8 Coup d'Etat

9. SIMULTANEOUS PENETRATION OF THE

DEFENSIVE SYSTEM AND TEAM
WARNING TIME

Perimeter

Team No. 1------~

I
,
I

Team No.2 -;-'------~

I ,
, I
, I
I I Team No.3 --~
I , I
I I I
I I I
Team No. 4---+----+
I I I
, I I I
, I I I
I I Team No.5 --07-
,I I I'
I I I I I
I I I I I
I I I I I
I I I I I
I I I I I

Time in hours 1 2 3 4 5 6 7 8 9 10 11 12

Team No. 1 2 3 4 5
Warninggivenathour 4 3 7t 5l 8t

manpower required, these targets will have to be seized by

'sophisticated' teams using a blend of infiltration, diversion and
assault.*
Though it will usually be necessary to prepare for a fairly
extensive military operation (and a complex one as well, unless
we have great numerical superiority in the area of the target)
this should not result in much actual combat: when those who
guard the target in question are confronted by our extensive
*The operational detail involved is discussed in Appendix B.
 The Execution of the Coup d'Etat 159
preparations, they are unlikely to put up much serious resistance.
The fact that our general measures of neutralization have cut.
off or impeded their contacts with the leadership, the fact that
the clear patriotic issues of international warfare will be missing
in an internal conflict, and the fact that we will make every
effort to allow them to give in gracefully, will all militate against
a prolonged defence.
If we are fortunate enough to have incorporated a very large
number of troops, and especially if they are equipped with
impressive weapons, such as armour, it will be still less likely
that actual combat will take place. These targets will neverthe-
less indirectly present us with a very serious problem, though
it is political rather than military: the formation of the large
teams required by these targets will raise the delicate issue of the
coup-within-the-coup. During the active phase of the coup the
situation will be confused and extremely unstable and while the
other teams will be too small to tempt their leaders into trying
to usurp our control, the opera:tionalleaders of the A-teams may
well succumb to temptation. The man who leads the tanks
which have just seized the Presidential Palace may easily
persuade himself that he can also seize power on his own behalf
and if the A-team is sufficiently powerful, he may do just that.
Our satisfaction at having carried out a coup successfully would
be an insufficient reward for all our efforts unless we also retain
power afterwards. It will therefore be necessary to adopt
measures to prevent the leaders of these large teams from chal-
lenging our position: this can sometimes be done by forming the
A-teams from many small sub-units under the overall command
of an inner member of our own group. Where this is not possible,
the A-teams will have to be dispersed into smaller groups
assigned to secondary targets as soon as they have fulfilled their
primary mission. Thus the possible threat presented by the
A-teams will be deflected by applying the energies of their
leaders to other tasks. The operational command.ers of the A-
teams will probably need a certain amount of time to readjust
to the fact that they are no longer isolated individuals engaged
on a dangerous endeavour and to start thinking in more ambitious
terms. Matters should be so arranged that they are deprived of
 160 Coup d'Etat
their large and unified teams before the transition is made in
their minds.

B-TARGETS
These are the technical facilities which will not usually be
heavily guarded, and which in any case we want to neutralize
rather than seize, such as the telephone exchange, the telegraph
office and secondary radio-television stations. Each of those
targets will be assigned to a small team whose personnel will
include a 'technician' whose presence should help to minimize
the amount of physical damage resulting from sabotage. If it is
possible to interdict these targets by minor and external sabotage,
the B-team may consist of just one or two technically-competent
operators. Even if the actual building has to be entered for a
short time, the B-team will still be a small one, but in this case it
should be overt and consist of uniformed soldiers or policemen.

C-TARGETS
These are the individuals we wish to hold in isola~ion for the
duration of the coup., In the case of the main leader(s) of the
government, the arrests will be subsumed in the seizure of the
Presidential Palace and similar A-targets, and therefore the
other C-targets should not present a penetration problem. But
they will present an evasion problem. A radio station or a Royal
palace can be very difficult targets to seize but at least they
cannot escape or conceal their identity. The personalities whom
we wish to arrest will try to do both. It will therefore be essential
to devote our early attentions to these targets so as to ensure that
they are seized before they can evade our teams. This will
usually imply that the C-teams will go into operation marginally
earlier than the other teams, and they can do so without breaking
the rule ofsimultaneous penetration of the 'early warning system',
because they should be sufficiently small and dispersed to act
covertly.
Because those targets are human they will be inherently more
problematic than some of our other objectives; the individuals
 The Execution of the Coup d'Etat 161

concerned, apart from escaping or concealing their identity,

could also try to subvert the very teams sent to arrest them. In
the case of particularly charismatic figures, our teams will have
to be formed from especially selected personnel; in some cases
it may even be necessary to include an inner member of our
group. These C-teams will be small, since their task will be a
matter of entering a private residence and overpowering one or
two guards. The exact size of each team will depend on the over-
all balance of resources and requirements under which we
operate, but will rarely exceed a dozen men or so.
Once the individuals which form this group of targets have
been arrested, we will have to ensure that they are kept under a
secure form of control. Our purpose in arresting them is to
prevent them from using their influence and/or charisma against
us and this can only be achieved if we can insulate them from
their public for the whole duration of the coup. Such individuals
are often the only casualties of otherwise bloodless coups,
because it is often easier to eliminate them, rather than to keep
them as captives; if we do keep them, the ad hoc prison used must
be both secret and internally secure. The liberation of a popular
public figure could be a powerful focus of counter-coup actions
on the part of the opposition and secrecy will be a more reliable
defence than any physical barrier.
While the teams are on their way to their respective targets
our other allies will also come into action: the individuals we
have subverted in various parts of the armed forces a.nd bureau-
cracy will carry out their limited missions of technical neutraliza-
tion. And the groups assigned to the blocking positions will be
moving to take up their planned locations designed to isolate the
loyalist forces. In the case of these dispersed individuals, whose
contribution will be extremely important though almost totally
invisible, there will be a signalling problem: since they are
scattered throughout the sensitive parts of the state apparatus,
it will be difficult to reach them individually. Further, they may
include informants of the security agencies because, unlike the
personnel of the various teams and blocking forces, they have
been recruited as isolated individuals and therefore the mutual
surveillance exercised on our behalf in the teams will not
 162 Coup d'Etat
operate. Thus it would be dangerous to give them advance
warning of the coup and their signal to go into action will have to
be our first broadcast on the radio-television station, except in
particular cases where the facility to be neutralized requires early
attention.
Our operational control over the various groups cooperating
with us will aim at achieving two objectives: (a) as always,
maximum speed in the execution of their tasks, and (b) the use
of the absolute minimum of force. This will be important not
only because of the psychological and political factors previously
mentioned, but also for a more direct, technical reason: the
external uniformity between the two sides of the conflict. Our
teams will, of course, be nationals of the country in which the
coup is being staged, and most of them will be soldiers and police-
men wearing the same uniforms as those of the opposition. This
uniformity will give us a measure of protection since the loyalist
forces will not readily know who is loyal and who is not. Usually
it would be a mistake to· prejudice this protective cover by
adopting distinctive armbands or other conventional labels,
since we will need all the protection we can get. Thus, as the
teams move around the capital city (probably at night) they will
probably not be fired upon, unless they open fire first; to do so
would be to facilitate the work of the opposition since this will
be their only way of distinguishing between their own forces and
ourselves. Further, since our teams have always been kept
separate, initially to prevent the penetration of the security
agencies, and now to protect our own position within the forces
of the coup, there will be a danger of conflict between our own
teams. The confusion we generate in the minds of the opposition
could therefore exact a price in confusion within our own ranks j
this may have serious consequences unless our forces respect the
rule of a minimal and purely defensive use of force.

The immediate Post-Coup Situation

Once our targets have been seized, the loyalist forces isolated and
the rest of the bureaucracy and armed forces neutralized, the
active (and more mechanical) phase of the coup will be over.
 The Execution of the Coup d'Etat 163
But everything will still be in the balance: the old regime will
have been deprived of its control over the critical parts of the
mechanism of the state, but we ourselves will not yet be in con-
trol of it, except in a purely physical sense and then only in the
area of the capital city. If we can retain our control over what we
have seized, those political forces whose primary requirement is
the preservation of law and order will probably give us their
allegiance. Our objective, therefore, is to freeze the situation so
that this process can take place. Thus, while until the actual
execution of the coup our aim was to destabilize the situation,
afterwards all our efforts should be' directed at stabilizing or
rather re-stabilizing, it.
We will be doing this at three different levels: (a) among our
own forces, where our aim is to. prevent our military or police
allies from usurping our leadership ; (b) within the state bureau-
cracy, whose allegiance and cooperation we wish to secure; and
(c) with the public at large, whose acceptance we want to gain.
In each case we will be using our leverage within one level in
order to control the next one, but each level will also require
separate and particular measures.

ST ABILIZING OUR OWN FORCES

During the planning stage, our recruits in the armed forces will
be fully conscious of the fact that the success of the coup - and
their own safety - depends on the work of coordination which
we perform. Immediately after the coup, however, the only
manifestation of all our efforts will be the direct force which
they themselves control. In these circumstances they may well
be tempted into trying a coup oftheir own, and they could do this
by establishing contact with the other military leaders we have
recruited, so as to secure their agreement to our exclusion from
the leadership. Apart from the dispersal counter-measures
discussed above, our only effective defence will be to retain full
control over all 'horizontal' communications, or in other words
to remain the only contact between each military leader we have
recruited and his colleagues. This can sometimes be done tech-
nically by keeping under our control the actual communication
 164 Coup d'Etat
equipment linking the various units, but this would only be
effective in unusually extensive capital cities and would, in
any case, break down after a relatively short period of time.
Usually, we will need somewhat less direct political and psycho~
logical methods designed to keep apart the various military
leaders we have recruited. This may involve promises of ac~
celerated promotion to selected younger officers who could not
otherwise expect very rapid advancement even within the limi~ed
context of those who have participated in the coup; it will also
be useful to remind our military and police allies that their
colleagues outside the conspiracy may try to displace them en
bloc unless they-and we-form a tight and mutually supporting
group. In general, we should ensure that all those who could
pose an internal threat are kept occupied on tasks which, whether
essential or not, will at least absorb their energies, and that there
are divisive factors operating between them. As soon as we begin
to receive the allegiance of military and bureaucratic leaders who
were previously outside the conspiracy our leverage with our
military and police recruits will increase very substantially. The
problem of retaining control against such internal threats will
therefore be largely short-term. As soon as our position has been
established our best policy may be to dispose of our dangerous
allies by using the usual polite methods available for the purpose:
diplomatic posting abroad, nominal and/or remote command
positions and 'promotions' to less vital parts of the state ap-
paratus. Since it is possible that an embryonic coup has existed
within our forces from the very beginning, the general security
measures we designed to protect ourselves against the penetra~
tion of the security agencies will also serve a useful supplemen~
tary function: they will prevent the lateral spread of the
conspiracy. If our internal security procedures are sufficiently
good to prevent all contact between the separate 'cells', so that
any infiltration by the security agencies is contained, they will
also prevent the co-ordination of this inner opposition.
It has been calculated* that in a defensive military situation
*The calculations are based on the performance of ex-Soviet Ukrainian
and Usbek troops used by the Germans in defensive positions during the
Normandy landings in the Second World War.
 The Execution of the Coup d'Etat 165
even if only twenty per cent of the troops of a unit are loyal, the
units concerned should operate successfully and perform their
assigned function. And though in aggregate terms our forces will
be operating offensively, vis-a-vis the uninfiltrated forces of the
state, their outlook will be defensive both psychologically and
tactically. Thus, even though it would be unusual to have the
complete loyalty of those who (since they joined our coup in the
first place) must be to some extent inherently disloyal, our forces
should still perform successfully.

STABILIZING THE BUREAUCRACY

Our attitude towards the second level, the armed forces and
bureaucracy which were not infiltrated before the coup, will
depend partly on the degree of control which we have over our
own 'incorporated' forces. Assuming that webave a reasonably
firm hold over them, we should not try to extract any early
commitment from the majority of soldiers and bureaucrats
whose first information of our existence will be the coup itself.
Not knowing the extent of the conspiracy, their principal pre-
occupation will be the possible danger to their positions in the
hierarchy: if most of the officers of the armed forces or the
officials of a ministry have joined the coup, those who have not
are hardly likely to be rewarded subsequently by rapid promo-
tion. If the soldiers and bureaucrats realized that the group
participating in the coup was in reality quite small, they would
also realize the strength of their own position: the fact that they
are collectively indispensable to any government, including the
one to be formed after the coup. In the period immediately after
the coup, however, they will probably see themselves as isolated
individuals whose careers, and even lives, could be in danger.
This feeling of insecurity may precipitate two alternative reac-
tions, both extreme: they will either step forward to assert their
loyalty to the leaders of the coup or else they will try to foment
or join in opposition against us. Both reactions are undesirable
from our point of view. Assertions of loyalty will usually be
worthless since they are made by men who have just abandoned
their previous, and possibly more legitimate, masters; opposition
 166 Coup d'Etat
will always be dangerous and sometimes disastrous. Our policy
towards the military and bureaucratic cadres will be to reduce
this sense of insecurity; we should establish direct communica-
tion with as many of the more senior officers and officials as
possible to convey one principal idea in a forceful and convincing
manner: that the coup will not threaten their positions in the
hierarchy and the aims of the coup do not include a reshaping of
the existing military or administrative structures.* This require-
ment will incidentally have technical implications in the plan-
ning stage, when the sabotage of the means of communication
must be carried out so as to be easily reversible.
The information campaign over the mass media will also
reach this narrow but important section of the population, but
it would be highly desirable to have more direct and confi-
dential means of communication with them. The general politi-
cal aims of the coup as expressed in our pronouncements on the
radio and television will help to package our tacit deal with the
bureaucrats and soldiers, but its real content will be the assurance
that their careers are not threatened. In dealing with particular
army or police officers who control especially important forces
or with important bureaucrats, we may well decide to go further,
in the sense that an actual exchange of promises of mutual
support may take place. We should, however, remember that
our main strength lies in the fact that only we have a precise
idea of the extent of our power. It would therefore be unwise to
enter into agreements which indicate that we need support
urgently; more generally, any information which reveals the
limits of our capabilities could threaten our position, which is
essentially based on the fact that our inherent weakness is con-
cealed. Again, as in the case of our own incorporated forces, we
should make every effort to prevent communication between the
cadres of the armed forces and bureaucracy outside our group.
Such communication would usually be indispensable to those

*Even when the coup is a vehicle for a political group which seeks to
achieve fundamental social change, the short-term objective is to stabilize
the bureaucracy and 'the anned forces. Later, when alternative sources of
direct force and political support have been established, the machinery of
state can be re-shaped into an instrument suitable for revolutionary change.
 The Execution of the Coup d'Etat 167
who may seek to stage a counter-coup; the ignorance of the extent
of the conspiracy will discourage such consultations: it is ob-
viously dangerous to ask somebody to participate in the opposi-
tion to a group of which he is himself a member. But we should
also interfere with such consultations directly, by using our con-
trol of the transport and communications infrastructure.

FROM POWER TO AUTHORITY:

STABILIZING THE MASSES

The masses have neither the weapons of the military nor the
administrative facilities of the bureaucracy, but their attitude to
the new government established after the coup will ultimately be
decisive. Our immediate aim will be to enforce public order, but
our long-term objective is to gain the acceptance of the masses so
that physical coercion will no longer be needed in order to secure
compliance with our orders. In both phases we shall use our
control over the infrastructure and the means of coercion, but as
the coup recedes in time, political means will become increasingly
important, physical ones less so.
Our first measures, to be taken immediately after the active
phase of the coup, will be designed to freeze the situation by
imposing physical immobility. A total curfew, the interruption
of all forms of public transport, the closing of all public build-
ings and facilities and the interruption of the telecommunication
services, will prevent, or at any rate impede active resistance to
us. Organized resistance will be very difficult since there will be
no way of inspiring and coordinating our potential opponents;
unorganized resistance on the part of a mob will, on the other
hand, be prevented because the people who might form such a
mob would have to violate the curfew while acting as individuals,
and not many will do this without the protective shelter of
anonymity which a crowd provides.
The impact of our physical measures will be reduced outside
the capital city but, to the extent that the capital city is the focus
of the national network of transport and communications, both
physical movement and the flow of information will be impeded.
The physical controls will be purely negative and defensive in
168 Coup d'Etat
character and our reliance on them could be minimal because
their concomitant effect is to enhance the importance of the
armed forces we have subverted.
Our second and far more flexible instrument will be our con-
trol over the means of mass communications; their importance
will be particularly great because the flow of all other informa-
tion will be affected by our physical controls. Moreover, the
confused and dramatic events of the coup will mean that the
radio and television services will have a particularly attentive
and receptive audience. In broadcasting over the radio and
television services our purpose is not to provide information
about the situation but rather to affect its development
by exploiting our monopoly of these media. We will have two
principal objectives in the information campaign that will
start immediately after the coup: (a) to discourage resistance
to us by emphasizing the strength of our position, and (b)
to dampen the fears which would otherwise give rise to such
resistance.
Our first objective will be achieved by conveying the reality
and strength of the coup instead of trying to justify it; this will be
done by listing the controls we have imposed, by emphasizing
that law and order have been fully restored, and by stating that
all resistance has ceased. One of the major obstacles to active
resistance will be the fact that we have fragmented the opposition
so that each individual opponent would have to operate in isola-
tion, cut off from friends and associates. In these circumstances
the news of any resistance against us would act as a powerful
stimulant to further resistance by breaking down this feeling of
isolation. We must therefore make every effort to withhold such
news. If there is in fact some resistance and if its intensity and
locale are such as to make it difficult to conceal from particular
segments of the public, we should admit its existence; but we
should strongly emphasize that it is isolated, the product of the
obstinacy of a few misguided or dishonest individuals who are
not affiliated to any party or group of significant membership.
The constant working of the motif of isolation, the repetition of
long and detailed lists of the administrative and physical controls
we have imposed and the emphasis on the fact that law and
 The Execution of the Coup d'Etat 169

TABLE IS. The First Communique: A Choice of Styles

The Romantic/Lyrical
'This is not a communique. but an avowal. an undertaking and an
appeal. It is an avowal of the situation in which the Army and the
People have been reduced by a handful of evil men ••• it is an under·
taking to wash clean the shame and disgrace suffered by the Army •••
it is finally a call to arms and to honour.....
Captain Mustafa Hamdun. Aleppo Radio. 6.30 a.m.
25 February I9S4
The Messianic
'The bourgeoisie is abolished ••• a new era of equality between all
citizens is inaugurated ••• all agreements with foreign countries will
be respected.•• .'
ColonelJean Bedel Bokassa, Central African Republic,
I5 January I966
The Unprepared
c ••• [This rebeIlion has been made for] a strong united and pros-
perous Nigeria free from corruption and internal strife.••• Looting.
arson, homosexuality [sic], rape, embezzlement, bribery, corruption,
sabotage and false alarm wiII be punishable by death. • •• '
Major Nzeogwu. Radio Kaduna. Nigeria. IS January I966
The Rational-Administrative
'The myth surrounding Kwame N'krumah has been broken ••• [he]
ruled the country as if it were his private property ••• [his] capricious
handling of the country's economic affairs ••• brought the country to
the point of economic collapse.••. We hope to announce measures for
curing the country's troubles within a few days ••• the future definitely
bright••• .'
Radio communique of Ghana's National Liberation Council
February I966

order have been re-established, should have the effect of making

resistance appear as dangerous and useless.
The second objective of our information campaign will be to
reassure the general public by dispelling fears that the coup is
170 Coup d'Etat
inspired by foreign and/or extremist elements, and to persuade
particular groups that the coup is not a threat to them. The first
aim will be achieved by manipulating national symbols and by
asserting our belief in the prevailing pieties: in the Arab world
the new regime will announce its belief in Arab unity and Islam,
or Arab unity and socialism as the case may be; where, as in
Egypt, the revolution has been institutionalized, it will be neces-
sary to assert our belief in Al- Thawra. In Mrica, the new regime
will announce its intention of fighting tribalism at home and
racialism abroad; in Latin America the need to secure social
justice (or to fight Communism and perhaps fidelismo) will be
invoked. Everywhere in the 'Third World', nationalist rhetoric
will be used and references made to the glorious people of X and
the glorious land of X which the last regime has degraded; above
all, repeated denunciations of neo-, and not so neo-colonialism
are de rigueur. Such denunciations will be particularly important
where there is a large foreign business enterprise operating in the
country in question; the inevitable suspicions that the coup is a
product of the machinations of the 'Company' can only be
dispelled by making violent attacks on it. These, being verbal and
not unexpected, will pacify the public without disturbing the
business interests, and the attacks should be all the more violent
if these suspicions are in fact justified.
While the religious attitude leads to the praise of the gods for
one's successes and self-blame for one's failures, the nationalist
attitude is to attribute successes to the nation and to blame
foreigners for its failures. Similarly, the chants in praise of the
gods have been replaced by ritualized curses variously addressed
to different groups of foreigners and their activities. Thus for the
phrase 'the imperialist-neo-colonialist power bloc' read the
English and the French if it is spoken by Africans of former
colonies of those countries; while the phrase' Zionist oil mono-
polist plotters' translates into Jews and Christians in the sub-
conscious of the Muslim. Arabs who make use of it.
There may be a purely ideological element in these denuncia-
tions, but even when the American extreme right speaks of 'the
international conspiracy of godless Communism' it is significant
that they stigmatize it as 'un-American' rather than as anti-
 The Execution of the Coup d'Etat 171

capitalist. We shall make use of a suitable selection of these un-

lovely phrases; though their meaning has been totally obscured
by constant and deliberate misuse,'* they will be useful as indi-
cators of our impeccable nationalism, and if that is not in reality
our position they will serve to obscure our true policy aims.
The flow of information emanating from all the sources under
our control should be coordinated with our other measures: the
impositions of physical controls will be announced and explained,
and the political moves, to which we now turn, will be suitably
presented. Physical coercion will deter or defeat direct opposi-
tion, while the information campaign will lay the basis of our
eventual acquisition of authority, but only political means will
secure for us a base of active support. Where the pre-coup regime
was exceptionally brutal, corrupt or retrograde, the leaders of
the coup will have little trouble in gaining a generalized form of
acceptance but, even then, the active support of specific groups
can only be gained by political accommodation, i.e. by sponsor-
ing policies which serve the interests of particular groups thus
giving them reasons for becoming committed to (or at least
interested in) our survival. In some Latin American countries, for
example, we could gain the support of the landless peasants by
announcing our intention of carrying out a programme of
agrarian reform. In West Africa we could announce our inten-
tion of increasing the prices paid to peasant producers by the
various commodity marketing boards; in Greece and Turkey,
where there is a heavy burden of peasant indebtedness, we could
announce a general cancellation of agrarian debts. Each of these
policy announcements will bind the interests of a large and
politically powerful group to our government unless we are
overtaken by other rival announcements, but it will also lead to
the hostility of other groups, whose interests are damaged by
our intended policies. In Latin America, where the peasants
would benefit, the landlords would lose; in Africa the urban
population would be the loser while in Greece the taxpayer would
bear the burden of peasant debt relief. Thus, the backing of one
*On 12 June 1967 the East German radio referred to the 'Nazi atrocities
committed by Jews against the Arabs of Gaza', who were described as
'victims of a Zionist-revanchist-Imperialist plot'.
172 COUp d'Etat
interest group will generally have as its concomitant the loss of
support of - or even actual hostility from - other groups. Clearly,
it will be necessary to estimate the net political support which a
given policy announcement will generate. This will mean taking
into account not only the political significance of each group but
also the immediacy of its political power. In the context of a
Latin American post-coup situation, for example, the goodwill of
remote and dispersed peasants will not help us much against the
immediate and powerful opposition of bureaucratic and military
cadres who could for the most part be the children of the land-
owning aristocracy. If, on the other hand, our short-term posi-
tion is strong but we are threatened by a longer-term usurpation
of power on the part of our military allies, our objective will be
to create a counter-weight capable eventually of becoming a
source of direct strength, such as a peasants' militia. Thus whether
we opt for a 'left' policy of land reform and longer-term
campesino support, or a 'right' policy of peasant repression and
immediate land-owner support, will depend on the balance
between the strength of our short- and our long-term positions.
The almost mechanical elements whic~ are important in the
special climate of the immediate post-coup period, will distort
the normal balance between the political forces of the country
concerned. If, therefore, our short-term position is not fragile,
we should repress the agitation of those forces which have a
disproportionate strength in the short term and concentrate
instead on cultivating the support of those groups whose longer-
term strength is the greater.
An element in our strategy after the coup is half-way between
the information and the political campaign; the problem of
'legitimizing' the coup. Clearly the coup is by definition illegal,
but whether this illegality matters, and whether it is possible to
counteract its effects, will depend on the total political environ-
ment of the country in question. We have seen in Chapter 2 that
in much of the 'Third World' the legitimacy or otherwise of the
government will not greatly matter; the government is treated as
part of nature, that is, something that one adjusts to rather than
questions. Elsewhere, however, the general attitude of the masses
could be more legalistic. One way of legitimizing the post-coup
 The Execution of the Coup d'Etat 173
government has already been mentioned in the discussion of the
selection of the personalities to be arrested - the retention of the
nominal head of state (where such a constitutional role exists)
as our own highly nominal head of state. In this way, the
appearance of continuity will be maintained and with it the
appearance oflegitimacy. Where the head of state is not nominal,
as in 'presidential' regimes, other tactics will have to be used:
the announcement of forthcoming elections or a referendum (as
a sort of ex-post facto legitimization) or, alternatively, the coup
can be openly admitted as an extra-constitutional intervention,
but one made against an un-constitutional regime. One illegality
will then be represented as being the cause of the other, but we
shall declare that whereas the illegality of the pre-coup regime
was voluntary and permanent, ours is necessary and temporary.
Such techniques will be of limited value in conducting the
political processes required to create a base of active support
and to secure our authority, since everything will depend on the
particular political environment in which we shall be operating;
one particular problem, however, requires further exploration:
recognition by foreign powers. This is almost always important,
but for many countries of the •Third World', whose pays riel lies
outside their own borders, it will be a crucial problem. When
much of the available disposable funds come from foreign
loans, investment or grants, and when foreign cadres carry out
vital administrative, technical and sometimes even military
functions, the maintenance of good relations with the particular
'donor' country or countries concerned may well be a deter-
mining factor in our political survival after the coup. Premature
recognition by a foreign power, i.e. recognition granted while the
old regime still retains some degree of control, is becoming
regarded as a form of aggression in international law. Beyond
this, however, recognition is usually granted to illegitimate
governments after a polite interval if there are convincing
assurances about their continuity in terms of foreign relations.
These assurances are conveyed simply and publicly by formal
announcements stating that membership in alliances and
groupings will' be maintained, that foreign agreements and
obligations will be respected and that legitimate foreign interests
 174 Coup d'Etat
in the country concerned will not be harmed. Thus the leaders
of Ghana's National Liberation Council, which was formed after
N'krumah's overthrow, announced that Ghana would retain her
membership of the Commonwealth, the Organization of African
Unity and of the UN, and would respect all obligations under-
taken by N'krumah's regime. Similarly, Arab post-coup regimes
announce that they will remain in the Arab League, and Latin
American regimes, in the Organization of American States. Far
more important than these declarations is the considerable
diplomacy activity which will take place after the coup (and
sometimes even before it). The purpose of these diplomatic
exchanges will be to clarify the political situation and, nowadays,
to indicate - or to dissemble - the ideological orientation of the
planners of the coup. Most countries of the world follow British
diplomatic doctrine in granting recognition to regimes on the
basis of the effective control of their territories. But this is a
doctrine as flexible as the definition of Ccontrol'; so that recogni-
tion can sometimes be withheld if the pre-coup regime retains
control over some part of the national territory, as in the case of
British non-recognition of the Yem.eni Republican regime.
Mter the necessary exchanges of information and assurances,
the new government will usually be recognized; ~his will be so
even if its illegality is an embarrassment, as in the case of the
United States and Latin American coups, or if its ideological
orientation is distasteful, as in the case of the Soviet Union and
the Ghanaian and Indonesian coups.
Diplomatic recognition is one of the elements in the general
process of establishing the authority of the new government;
until this is achieved, we will have to rely on the brittle instru-
ments of physical coercion and our position will be vulnerable to
many threats - including that of coup d'4tat.
Appendix A
The Economics of Repression

Once we have carried out our coup and established control over
the bureaucracy and the armed forces, our long-term political
survival will largely depend on our management of the problem
of economic development. Economic development is generally
regarded as a 'good thing' and almost everybody wants more of
it, but for us - the newly-established government of X-land - the
pursuit of economic development will be undesirable, since it
militates against our main goal: political stability.
An economy develops by extending and improving its stock of
human and physical capital and this requires investment, whether
to train people or to build factories. In order to invest, ,current
income has to be withdrawn from would-be consumers and
channelled away to create capital. Clearly, the higher the rate of
investment the faster will be the development of the economy,
but also the lower the present standard of living. The govern-
ments of economically backward countries - where the need for
development is manifest - are therefore faced with the alternative
of either slow economic development or further reduction of the
already desperately low standard of living. The more that can
be taxed from current incomes, the nearer will be the beautiful
dawn of prosperity - even if it is the prosperity of Spain or
Greece rather than that of Western Europe or North America.
But there are limits to the amount of saving that can be forced
out of a popUlation whose annual income per head is already
very low: there is an economic survival limit below which the
population - or a large part of it - would simply starve (or
retreat into the pure subsistence economy), but well before this
point is reached, there is a political survival limit below which
we, as the government would be overthrown. The economic
survival limit is more or less rigid: in any particular environment
with a given climate, pattern of nutrition, habits and traditions,
there will be a minimum annual income which an inhabitant of
average resourcefulness will need to satisfy his and his family's
176 Coup d'Etat
bodily needs. The 'political survival limit' is, however, very
flexible and it will depend on psychological, historical and social
factors, but also on the efficiency of the system of state security
and of the propaganda machine.
The problem is particularly acute in the newly-independent
states of the 'Third World'. The colonial regimes mayor may
not have tried to achieve economic development, but if they did
try it was without the urgency which the new post-colonial
regimes try to achieve. Immediately after independence, there-
fore, instead of the increase in the standard of living which the
native population had been led to expect, the opposite takes
place. The new 'independence' government has to increase
taxes and import duties in order to finance the great projects
with which economic development often starts: dams, road
systems, steel mills and harbours. Foreign aid, which many in
the 'donor' countries have been led to believe to. be very
substantial,* only contributes a fraction of the necessary funds.
Most of the money has to come out of current incomes so that,
instead of' having cars like the whites', the level of consumption
actually falls. This impoverishment of those who are already very
poor indeed is not easily tolerated - especially when the mecha-
nism of expectations has been built up.
Our basic problem, therefore, is to achieve economic develop-
ment - in order to satisfy the aspirations of the elite and would-
be elitet - without taxing the masses beyond the politically safe
limit, which could lead to their revolt. There are two main
instruments with which we can persuade the masses to accept
the sacrifice of present consumption for the sake of an increased
future income: propaganda and repression; or, more efficiently,
by a mixture of both. Imagine, therefore, that we have inherited
*Foreign aid has been !aUing as a percentage of GNP in the developed
countries in the last few years.
t For the elite, economic development subsumes the national goal of
modernization with the personal goal of expanded career opportunities. For
the new generation of educated citizens (the would-be elite) economic
development is a guarantee of employment - and the unemptoyed intelli-
gentsia is a major threat to many regimes in the 'Third World'.
:t:By propaganda is meant the whole range of activities whose content
is information or entertainment and whose function, in this case, is (a) to
 Appendix A. The Economics of Repression 177
a country with a backward economy whose vital statistics are
those shown in Table 16.
TABLE 16. National Accounts Data, Country X. (Assumed
egalitarian distribution)

Annual GNP
per head
£100 Present actual ievel of income in pounds per
inhabitant
Level of taxation accepted in the past (i.e. net
income left after tax)
£80
£70
£60
£5 0
£45 . . . Economic survival limit
£40
£30
£20
£10

Thus, in the past, in this poor (but not particularly poor)

country the gross national product per head was £100 per year,
and of this £10 was paid out in various taxes while £90 was
spent on current consumption, or saved. Now we know that only
£45 per inhabitant per year is needed for economic survival, and
the problem is to get hold of some of the difference in order to
distract attention from present hardship, (b) to justify it in terms of assured
future happiness. This mayor may not involve the presentation of the
outside world as even less well off, but it will almost certainly present the
past standard of living as much inferior. An equally important aim of
propaganda will be to persuade the masses that the present leadership is
the most efficient vehicle for modernization; this can be done in rational
terms by using statistical images, or by irrational ones which present the
leadership as super-human. By repression is meant the whole range of
political police activities which aim at: (a) suppressing individual political
activity by surveillaOce and imprisonment, (b) intimidating the masses by
displays of force, and (c) preventing the circulation of rival information by
controlling the media and itihibiting public discussion.
178 Coup d'Etat
finance development - and to do so without being overthrown.
If we simply increase taxes the chances are that part of the
population will refuse to pay them, and if administrative methods
are used in order to enforce payment a violent reaction may
ensue. We will therefore divert some of the modest tax pay-
ments received now (£ro in the diagram) from other uses, and
spend it on propaganda and the police. This may well result in
the situation shown in Table 17.

TABLE 17. National Accounts Data, Country X. (After

propaganda and police)

Annual GNP
per head
£100 Present actual level of income per inhabitant
£90
£80 Level of taxation accepted after, say, £1 per head
per year has been spent on propaganda and
police
£70
£60
£50
£45 •. . Economic survival limit
£40
£30
£20
£10

In the new situation, then, our sums work out as in Table 18.
TABLE 18. National Accounts Data, Country X. (Funds
available for development)
Before (Expenditure on propaganda and repression) After
£100 +- GNP per head per year -+ £100
£55 +-- Less £45 per year needed for' survival' - - £55
£10 +- Tax collected -----~ £20
£0 +-Less money spent on propaganda and
repression -+
+ - - - Net funds available for development ---+
 Appendix A. The Economics of Repression 179
Thus by spending £1 per man per year on propaganda and
an efficient police system, we have lowered the political survival
limit by £10, and after deducting the amount spent on the
system of repression and persuasion we still have £19 left. If we
spend another £1 per man per year the chances are that we will
be able to 'liberate' some more of the possible margin above the
survival limit, but as we spend more and more money on
repression we are likely to find that it will lower the safety limit
by less and less (see Figure 10). And, of course, as we spend
more and more on the police and propaganda we will find that
while the first extra £10 of taxes costs us £1 to obtain with
safety, the next £10 will cost, say, £2. Eventually the point is
reached where (as shown in Figure 10) further expenditure brings
us no increase in taxation at all. At that point we spend an extra

10. POLITICAL SURVIVAL LIMIT ON TAXATION

Theoretical maximum
~ Pounds per capita (i.e. economic.survivallimit)
40
30
/"
/
./
--
20 ./
i-'"
10

1 2 3 4 5 6 7 j
EXPENDrruRE ON PROPAGANDA AND REPRESSION Pounds per capita

£1 per year and get no increase at all in the taxes which can
safely be collected. Well before that point is reached, however,
there will be an earlier stage when we will spend, say, an extra
£1 on repression and persuasion and get exactly the same sum
in further taxes. Immediately before that point is the maximum
efficiency level of expenditure on the police and propaganda
machine.
180 Coup d'Etat

Maximum Safety and Zero Economic Development

This is the formula which Duvalier has applied in Haiti with
increasing thoroughness since his rise to power. Taxation, which
is heavy for a country with an annual income per head of about
£30, is spent almost entirely on the army, the 'police' (the Ton
Ton Macoutes) and on propaganda. The only major economic
development project is of doubtful value: the building of a new
capital, 'Duvalierville', which, in any case, has now been
suspended.
The Duvalier mix of efficient repression, massive propaganda
and no development investment has paid off; he has been in
power continuously since September 1957, and his regime still
appears to be more stable than that of most Latin American
countries. The Ton Ton Macoutes operate as a semi-public
presidential guard which carries out police and security functions,
and they supplement their (generous) salaries by private exac-

11. THE DUVALIER FORMULA

Political security -= maximum economic development = zero
$-
EXPEN"DITURl' ,I)I I
Ton Ton Macoutes

Army TAXATION

Propaganda r-Margin of income

I I~ above minimum
(Duvalierville
palaces etc.)
-subsistence
Minimum
subsistence >- DISPOSABLE INCOME
Health/Education level of income
per capita
Development
 Appendix A. The Economics of Repression 181

tions from what is left of the business sector. The propaganda

machine, which involves ceremonial parades, laudatory films
and the projection of 'Papa Doc' as a Voodoo expert, is almost
as expensive to run as the Ton Ton Macoutes, but equally use-
ful. The extreme poverty of the population means that their
level of political awareness and even vitality is extremely low;
the Ton Ton Macoutes terrorize the thin elite and the army
officers - who are constantly watched - and the Ton Ton them-
selves rely on Duvalier because their position depends on his
survival; the Voodoo mythology and the propaganda machine
have deified the man whom the Ton Ton defend, and if Duvalier
should disappear from the scene the army and/or the masses
would quickly liquidate the Ton Ton.
Kwame N'krumah of Ghana, and many other African leaders
now dead, in prison or in exile, have followed a policy of high
taxation IUld investment associated with clearly insufficient pro-
paganda and repression effort. N'krumah in spite of his eccen-
tricities, was largely defeated by his own success: the by-product
of the considerable economic development achieved by Ghana
was to stimulate and educate the masses and the new elite, their
attitude to N'krumah's regime becoming more and more critical
in the light of the education which the regime itself provided.
When this happens, more and more repression and propaganda
are needed to maintain political stability and, in spite of con-
siderable efforts, N'krumah was unable to build a sufficiently
ruthless police system. The cause of his downfall was not, there-
fore, the mismanagement of the economy - which was con-
siderable - but rather the success of much of the development
effort.
The middle way, efficient repression, extensive propaganda
and enough economic development to create new elites com-
mitted to the regime, has been followed successfully in both the
Soviet Union and China; the regimes of both countries have,
however, used different blends of repression and propaganda.
The two are interchangeable up to a point, but the nature of the
most efficient mixture will depend on the particular country, and
local conditions will dictate the most appropriate mixture.
Appendix B
Tactical Aspects of the Coup d'Etat
In the decisive (active) phase of the coup the forces which we
have acquired by infiltrating and subverting the system of state
security will be used to seize certain objectives or to neutralize
selected facilities. To do this force will be essential, but since
bloodshed could have unfortunate destabilizing effects we should
arrange matters so that the threat of force will suffice to achieve
our objectives, rather than its actual use. In this appendix we
will analyse two main problems, (a) the formation of the active
teams and their operational use, and (b) the deployment of the
blocking forces. In both cases our two major preoccupations will
be to avoid or minimize bloodshed and - more important - to
ensure that our position is not threatened after the coup by a
usurpation of our power on the part of the soldiers and police-
men we have recruited.

The Formation of the Active Teams

Our infiltration of the armed forces and police of the state may
have been either general and diffuse or concentrated within a
few large formations. In the first type of infiltration the forces
we have subverted will consist of many small sub-units, whose
commanders have decided to join us while their more senior
colleagues - who command the unit as a whole - have remained
outside the scope of our infiltration; in the second type of
infiltration a few large units have come over to us en bloc with all,
or mos~ of their entire establishment. The two alternatives are
illustrated in Table 19.
Both types of infiltration have their advantages and dis-
advantages: if we have subverted many small sub-units we will
have some additional protective cover, since the supporters of
the pre-coup regime will not readily be able to identify which
units have remained loyal and which have joined us; it will also
be useful to be able to confront loyalist forces with teams made
 Appendix B. Tactical Aspects of the Coup d'Etat 183

TAB LE 19. Eve of the coup: forces of the state fully subverted (notional)

Diffuse infiltration Concentrated

3 companies of brigade X 2 battalions of brigade X
6 companies of brigade Y I battalion of brigade Y
7 companies of brigade Z
I battalion of brigade V
4 companies of brigade U

Total forces = 3,000 men Total forces = 3,000 men

up from their own cadres. The subversion of a few large units,

on the other hand, will minimize the problems of coordination
and recognition and, more important, will increase the security
level before the coup, since within each large unit there would be
a measure of mutual surveillance which would deter defections
to the regime or leaks to the security agencies. Mter ~e active
phase of the coup, however, a composite force made up of many
small sub-units will be much more secure since it will reduce the
risk of a usurpation of our position on the part of our military
allies. This would be so for three main reasons: (a) the rank of
the officers concerned will obviously be lower if they are the
appointed commanders of small units rather than large ones, (b)
it will be easier to disperse our forces after the active phase of the
coup if their concentration is not organic but a construct of our
own, and (c) the larger the number of independent unit leaders
involved in the coup the less likely they are to combine in order
to exclude us from power.
Whatever the sources of the forces which we have incorporat-
ed, it will often be necessary to restructure them for the purposes
of the coup, since the many specialized tasks to be carried out will
require widely different teams; only if we have numerical
superiority or its equivalent over the loyalist forces will we be
able to use the formations we have subverted in their natural
state. We will need three types of team, as well as the blocking
forces, and these will correspond to the three types of targets
discussed in Chapter S, and we will therefore draw on the pool
184 Coup d'Etat
of subverted units and individuals to form the required number
of A-, B- and C-teams.
The A-teams will be needed in order to seize the major
defended objectives, such as the residence of the ruler, the main
radio-television station and the army and police headquarters.
These will be both larger in size and more sophisticated in
structure than the other two types of team. Each A-team will
consist of four elements whose relative size will vary with each
particular target:
(a) a 'civilian' penetration group. This will be very small and
will consist of a few men in civilian clothes carrying concealed
weapons or explosives. Their function will be to enter the
target as open and legitimate 'visitors' in order to assist in its
seizure from the inside. This assistance can be a direct internal
assault, or it can take the form of an internal diversion; in the
case of the broadcasting facility, however, their main function
will be to prevent the use of its installations to raise the alarm.
(b) a 'diversion' group. This group will be important in pro-
portion to the size of the forces deployed to protect the target.
Where, as in the case of the royal or presidential palace, there
could be an entire infantry formation assigned to protect the
target, a diversion, designed to attract part of the loyalist
forces, will be essential. The diversion group will carry out
its function by creating a disturbance, or by actually carrying
out an assault on a near-by secondary target. The diversion
should be timed to include the reaction time of the loyalist
forces and their route time to the scene of the disturbance,
after which the main assault on the primary target will take
place.
(c) a covering 'fire' group. This will be a small group but it
will include troops with heavier weapons, especially armoured
fighting vehicles. Its function will be to deter resistance on the
part of the loyalists, by giving demonstrations of fire-power,
and to prevent the intervention of loyalist forces from else-
where by covering approach routes.
(d) an assault group. This will be by far the largest, and its
 12. 'SOPHISTICATED' SEIZURE OF
MAJOR DEFENDED TARGETS

Gardens

SEQUENCE

1. Civilian panetration
2. Diversion designed to attract loyallsttroops away from palace
3. Interdicting fire to prevent their return and the passage
of the main loyallstforcea
4. AsS8ultgroup from atreet entaralnto action
A. Expected approach of main Ioyalistforcea
I86 Coup d'Etat
members should be chosen oli the usual criteria of combat
proficiency, though hopefully their skills will not be needed.

The integrated operation of the different groups of each A-

team is illustrated in Figure I2.
The B- and C-teams, whose function is respectively to arrest
political personalities and to sabotage selected facilities, will not
face significant tactical problems and their organization will be a
matter of forming small teams equipped with suitable transport,
of designating the target and of coordinating the timing. Each
team will consist of a couple of jeep-loads of troops or police
accompanied by a member of our inner group - in the case of the
major political personalities - or by a technician - in the case
where the sabotage requires a measure of expertise.

The Deployment of the Blocking Forces

Although it is to be hoped that the pre-coup regime will be
unaware of the timing of our particular coup, it will probably be
conscious of the danger from coups in general. Regimes in
politically unstable countries often go to great lengths to main-
tain a force ofpolitically reliable troops or armed police on which
they can rely against threats to internal security. Its officers and
men often share the ethnic and/or religious affiliation of the
ruling group and special safeguards are employed to ensure its
political reliability. The infiltration of such 'palace guards' is
very difficult, and we may well have deliberately decided to
exclude them from the scope of our infiltration. Elsewhere, even
where every major force has been fuUy subverted or internally
neutralized, we will still be vulnerable to unexpected defections
or co~cidental transfers of uninfiltrated troops. For all these
reasons, therefore, the blocking forces designed to insulate the
capital city from the intervention of loyalist forces will be
essential since, as has been repeatedly emphasized, the interven-
tion of detennined loyalist forces - however small in size -could
have effects disproportionate to their size.
The operation of a blocking force is the exact opposite of an
ambush: while the objective of an ambush is to inflict maximum
 Appendix B. Tactical Aspects of the Coup d'Etat 187

13. GENERAL STRUCTURE OF BLOCKING POSITION

Observation line to detect and prevent infiltration

',....../ Expectodpalhl"'IovaUstforces \ /....-

" , ~
~~
~ ~

Limit of area of '" .~ ~

constrained passage rc-.. -----

'Sympolic'
~.-...-.-.-.- roadblocks
Main defence forces
protected but not ._mlllEDmmz~
concealed

++++++ Anti-tank!
tank positions

To capital city and political-administrative centre

!
damage without COD,trol1ing passage, the objective of the block-
ing force is to prevent passage while inflicting minimum
damage. The general structure of the blocking position is shown
in Figure 13 but two essentials are missing: (a) correct intelli-
gence about the location and intentions of the loyalist forces, and
188 Coup d~Etat
(b) the efficient use of natural barriers (such as bridges, tunnels,
densely built-up areas3 etc.) and of subsidiary roadblocks to
channel any loyalist force into the blocking position.
The area of constrained passage on the diagram represents the
group of roads or streets which an intervention force must use in
order to enter the city from a particular direction; it is not
generally meant to represent a single road or street, though in
particular settings this may be the case.
The 'observation line' (or 'screen' in military terminology)
attempts to infiltrate round the blocking position which dis-
mounted loyalist troops may make. The 'symbolic' road-
blocks deployed across the set of roads or streets concerned will
dissuade the loyalist forces by appealing to 'orders' and com-
radeship; if dissuasion fails, they will try deterrence by pointing
out the main defensive forces and the anti-tank positions (or the
tanks if available). The operational leadership of the main
defensive forces, the 'teeth' of the blocking position, will have
to be chosen carefully to ensure a determined defence if force is
in fact used by the loyalist troops; they must also be made aware
of the damaging consequences which could ensue jf the blocking
position degenerates into an ambush.
APPENDIXC

Statistics

Table I Economic Development

and the Coup d'Etat, 1945-78
Table II Basic list of Coups
and Attempted Coups, 1945-7~
Table III The Efficiency of the Coup d'Etat, 1945-64
Table IV The Frequency of the Coup d'Etat
190 Coup d'Etat
TABLE IEconomic Development and the Coup d'Etat, 1945-78
Revised and updated by George Schon, August 8,1978

NOTE: All data for the individual countries are from the World
Bank Atlas: Population, Per Capita Product, and Growth Rates,
World Bank, 1977. Data refer to 1976 except for 12 ofthe smaller
countries for which the data refer to 1975.

Per Capita
Gross
National Dateo!
Product Last
Figures • Successful
Country Dollars Coup d'Etat? coup

AFRICA
Algeria 990 yes 1965
Angola 330 yes
Benin 130 yes 1972
Botswana 410 no
Burundi 120 yes 1976
Cameroon 290 no
Cape Verde 260 no
Central African Empire 230 yes 1966
Chad 120 yes 1975
Comoros ISO yes 1978
Djibouti 1,940 no
Egypt 280 yes 1952
Equatorial Guinea 330 yes ...
Ethiopia 100 yes 1974
Gabon 2,590 yes
Gambia 180 no
Ghana 580 yes 1972
Guinea 150 yes
Guinea-Bissau 140 no
 Appendix C. Statistics 191

Per Capita
Gross
National Dateo!
Product Last
Figures Successful
Country Dollars Coup d'Etat? coup

Ivory Coast 610 no

Kenya 240 no
Lesotho 170 yes 1970
Liberia 450 yes
Libya 6,310 yes 1969
Madgascar 200 yes 1975
Malawi 140 no
Mali 100 yes 1968
Mauritania 340 yes 1978
Mauritius 680 no
Morocco 540 yes
Mozambique 170 yes
Namibia 980 no
Niger 160 yes 1974
Nigeria 380 yes 1975
Reunion 1,920 no
Rhodesia 550 no
Rwanda 110 yes 1973
Senegal 390 no
Seychelles 580 yes 1977
Sao Tome and Principe 490 no
Sierra Leone 200 yes 1968
Somalia 110 yes 1969
South Africa 1,340 no
Sudan 290 yes 1969
Swaziland 470 yes 1973
Tanzania 180 yes 1964
Togo 260 yes 1967
Tunisia 840 no
Uganda 240 yes 1971
192 Coup d'Etat

Per Capita
Gross
National Date of
Product Last
Figures Successful
Country Dollars Coup d'Etat? coup

Upper Volta 110 yes 1966

Zaire 140 yes 1965
Zambia 440 no

ASIA
Afghanistan 160 yes 1978
Bangladesh 110 yes 1975
Bhutan 70 no
Bunna 120 yes 1962
Cambodia -* yes 1970
China, People's Rep. of 410 no
Hong Kong 2,110 no
India 150 no
Indonesia 240 yes
Iran 1,930 no
Japan 4,910 no
Korea, Dem. People's
Rep. 470 no
Korea,Rep. of 670 yes 1961
Laos 90 yes 1964
Macao 780 no
Maldives 110 no
Mongolia 860 no
Malaysia 860 no
Nepal 120 yes 1960
Pakistan 170 yes 1977
* In Kampuchean conditions no GNP estimate could have
meaning.
 Appendix C. Statistics 193

Per Capita
Gross
National Date of
Product Last
Figures Successful
Country Dollars Coup d'Etat? coup

Philippines 410 no
Singapore 2,700 no
Sri Lanka 200 yes
Taiwan 1,070 no
Thailand 380 yes 1977
Vietnam * yes 1965

LA TIN AMERICA
Argentina 1,550 yes 1976
Bahamas 3,310 no
Barbados 1,550 no
Bolivia 390 yes 1978
Brazil 1,140 yes 1964
Chile 1,050 yes 1973
Colombia 630 yes 1957
Costa Rica 1,040 yes
Cuba 860 yes 1952
Dominican Republic 780 yes 1965
Ecuador 640 yes 1976
El Salvador 490 yes 1961
Grenada 420 no
Guadeloupe 1,500 no
Guatemala 690 yes 1963
Guyana 540 no
Haiti 200 yes 1950
Honduras 390 yes 1975
* For Vietnam the per capita GNP is tentatively estimated at
$151.
194 Coup d'Etat

Per Capita
Gross
National Date of
Product Last
Figures , Successful
Country Dollars Coup d'Etat? coup

Jamaica 1,070 DO
Maninique 2,350 no
Mexico 1,090 no
Netherlands Antilles 1,680 no
Nicaragua 750 yes 1947
Panama 1,310 yes 1968
Paraguay 640 yes 1954
Peru 800 yes 1975
Pueno Rico 2,430 no
Surinam 1,370 no
Trinidad and Tobago 2,240 yes
Uruguay 1,390 no
Venezuela 2,570 yes 1952

MIDDLE EAST/SOUTHWEST ASIA

Bahrain 2,410 no
Cyprus 1,480 yes 1974
Iraq 1,390 yes 1968
Israel 3,920 no
Jordan 610 no
Kuwait 15,480 no
Lebanon -* yes
Oman 2,680 yes 1970
Qatar 11,400 yes 1972
Saudi Arabia 4,480 no
Syria 780 yes 1970
Turkey 990 yes 1971

* The 1977 GNP is tentatively estimated at $2.9 Billion.

 Appendix C. Statistics 195

Per Capita
Gross
National Date oj
Product" Last
Figures , Successful
Country Dollars Coup d'Etat? coup

United Arab Emirates 13,990 yes

Yemen 250 yes 1974
Yemen, Democratic 280 yes 1978

EUROPE, NORTH AMERICA AND OCEANIA

Czechoslovakia 3,840 yes 1948
Greece 2,590 yes 1973
Portugal 1,690 yes 1974

TABLE n Basic List ojCoups and Attempted Coups, 1945-78

Revised and updated by George Schott, August 8, 1978

Date Main Party Outcome

AFRICA
Algeria
June 19, 1965 elements from three services successful
Dec. 13,1967 army faction failed
Angola
May 27,1977 army and political faction failed
Benin (Dahomey)
Oct. 28, 1963 army faction successful
Dec. 22, 1965 army faction successful
Dec. 17,1967 army faction successful
196 Coup d'Etat

Date Main Party Outcome

July 12, 1969 army faction failed
Oct. 21, 1969 army faction failed
Dec. 10, 1969 army faction successful
Feb.28,1972 army faction failed
Oct. 26, 1972 army faction successful
Jan. 21,1975 army and political faction failed
Jan. 16, 1977 * [foreign-supported faction] failed
Burundi
Oct. 19, 1965 army and political faction failed
July 8, 1966 Prince plus army faction successful
Nov. 28, 1966 Prime Minister plus army faction successful
Nov. I, 1976 army and tribal faction successful
Central African Empire
Jan. 1, 1966 army faction successful
April 12, 1969 army faction failed
Chad
August 26, 1971 foreign-supported faction failed
April 13, 1975 army faction successful
March 31, 1977 army and tribal faction failed
Comoros
August 3, 1975 army and political faction successful
June 4, 1977 army and political faction failed
May 13, 1978 foreign mercenaries successful
Congo, People's Republic of
August 15, 1963 army and labour unions successful
June 28-29,1966 army and tribal faction failed
August 3-31,1968 army faction successful
Sept. 4, 1968 army faction successful
Nov. 8, 1969 army faction failed
March 23, 1970 army faction failed
Feb. 22, 1972 left-wing army faction failed
* It is widely believed that this coup was simulated by President
Mattneu Kerekou for internal political reasons.
 Appendix C. Statistics 197

Date Main Party Outcome

March 18, 1977 political faction failed
Egypt
July 23, 1952 army faction successful
Feb. 5, 1966 left-wing political faction failed
Sept. 24, 1966 political faction failed
August 27, 1967 army and political faction failed
June 1, 1975 political faction failed
Equatorial Guinea
March 21, 1969 army and political faction failed
Ethiopia
Dec. 13-17,1960 army faction failed
Sept. 12, 1974 elements from three services successful
Nov.22-24,1974 faction in ruling Military Council successful
Feb. 3, 1977 faction in ruling Military Council failed
Gabon
Feb. 17-18, 1964 * army faction failed
Ghana
Feb. 24, 1966 army faction successful
April 17, 1967 army faction failed
Jan. 13, 1972 army faction successful
Jan. 15, 1972 army faction failed
Guinea
Nov. 22, 1970 foreign-supported faction failed
May13,1976 army and political faction failed
Lesotho
Jan. 30, 1970 political faction successful
Liberia
Feb. 5,1963 political faction failed
* The government of Leon Mba was briefly overthrown by a
military coup, but was reinstated the next day when French
troops intervened under the terms of a 1961 defense agreement
[with France]. ,
198 Coup d'Etat

Date Main Party Outcome

Libya
Sept. I, 1969 elements from three services successful
Dec. 10, 1969 foreign-supported faction failed
August 14, 1975 army faction failed
Madagascar (Malagasy Republic)
April 1, 1971 left-wing army faction failed
MayI8,1972 political faction successful
Feb. 5-12, 1975 army faction successful
Mali
Nov. 19, 1968 army faction successful
April 7, 1971 army faction failed
Mauritania
July 10, 1978 army faction successful
Morocco
July 10, 1971 general's faction in army failed
August 17, 1972 air force faction failed
Mozambique
Dec. 17, 1975 army and police faction failed
Niger
Apri115, 1974 all forces successful
August 2, 1975 army and political faction failed
March 15,1976 army and political faction failed
Nigeria
Jan. 15, 1966 army faction successful
July 29, 1966 northern army faction successful
July 29, 1975 army faction successful
Feb. 13, 1976 army faction failed
Rwanda
July 5,1973 army and police faction successful
Seychelles
June 5,1977 police and political faction successful
 Appendix C. Statistics 199

Date Main Party Outcome

Sierra Leone
March 23,1967 army faction successful
April 18, 1968 army faction successful

Somalia
Oct. 21,1969 army and police faction successful
Apri121, 1970 army and political faction failed
May 25,1971 army and political faction failed
April 9, 1978 army faction failed
Sudan
August 18, 1955 army and tribal faction failed
Nov. 17,1958 army faction successful
Dec. 28, 1966 left-wing army faction failed
May 25,1969 left-wing army faction successsful
July 19-22, 1971 left-wing army faction failed
Sept. 5, 1975 army faction failed
Feb.2,1977 air force faction failed

Tanzania (Tanganyika and Zanzibar)

Jan. 12, 1964
(Zanzibar) political faction successful
Jan. 20, 1964
(Tanganyika) troop mutiny failed
Togo
Jan. 13, 1963 army and tribal faction successful
Nov. 21-22, 1966 political faction failed
Jan. 13, 1967 general's faction in army successful
August 8, 1970 army and political faction failed
Uganda
Jan. 23, 1964 troop mutiny failed
Jan. 25, 1971 army and police faction successful
March 23, 1974 army faction failed
Upper Volta
Jan 3,1966 army faction successful
200 Coup d'Etat

Date Main Party Outcome

Zaire
Nov. 25, 1965 army faction successful
May 30,1966 political faction failed

ASIA
Afghanistan
July 17, 1973 army and police successful
Nov. 30, 1976 retired army officers failed
April 27, 1978 army and air force successful
Bangladesh
August 15, 1975 army and political faction successful
Nov.7,1975 army mutiny failed
Oct.2,1977 army and air force faction failed
Burma
Sept. 26, 1958 army faction successful
March 2, 1962 elements from three services successful
July 24, 1974 left-wing political faction failed
Cambodia (Oem. Kampuchea)
.March 18, 1970 elements from three services successful
Indonesia
Dec.3,1950 navy faction failed
April 26, 1950 elements from two services failed
Oct. 1,1965 Communist Party failed
Nov. 16,.1965 elements from three services failed
Korea, Republic of
Oct. 20, 1948 army faction failed
May 16, 1961 elements from three services successful
Laos
August 9, 1960 neutralist army faction successful
April 19, 1964 right-wing army faction successful
Jan. 31, 1965 army and police faction . failed
Oct. 21, 1966 air force faction failed
August 20, 1973 air force faction failed
 Appendix C. Statistics 201

Date Main Party Outcome

Nepal
Dec. 15, 1960 King plus aony faction successful
Pakistan
Oct. 7-27, 1958 elements from three services successful
July 5, 1977 elements from three services successful
Sri Lanka
Jan. 29, 1962 political faction failed
Thailand
Nov.9,1947 army faction successful
June 29, 1951 navy faction failed
Nov. 29,1951 aony faction successful
Sept. 16, 1957 aony faction successful
Oct. 20, 1958 aony faction successful
Nov. 17,1971 Prime Minister successful
Feb. 24, 1976 army faction failed
Oct. 20, 1976 elements from three services successful
March 26, 1977 aony and political faction failed
Oct. 20, 1977 elements of three forces successful
Vietnam
Nov. 1-2, 1963 elements from three services successful
Jan. 30, 1964 elements from three services successful
Jan. 27, 1965 elements from three services successful
Feb. 21, 1965 elements from three services successful

LATIN AMERICA
Argentina
Sept. 28, 1951 elements from three services failed
June 16, 1955 navy faction failed
Sept. 16, 1955 elements from three services successful
Nov. 13, 1955 army faction successful
June 13, 1960 army faction failed
March 28, 1962 elements from three services successful
August 8,1962 troop mutiny failed
202 Coup d'Etat

Date Main Party Outcome

April 2, 1963 general's faaion in army failed
June 28, 1966 army and navy generals successful
June 8,1970 elements from three services successful
March 23, 1971 elements from three services successful
May 11, 1971 army and political faction failed
Oct. 8, 1971 army faction failed
March I, 1974 police failed
Dec. 18, 1975 right-wing air force faction failed
March 24,1976 elements from three services successful
Bolivia
May 16, 1951 army and air force faction successful
April 9, 1952 army and political faction successful
Nov. 3, 1964 army and air force faction successful
August 21, 1968 army faction failed
Sept. 26, 1969 army and air force faction successful
Oct. 6-7, 1970 army and air force faction successful
Jan. 10, 1971 right-wing army faction failed
Aug. 19-22, 1971 right-wing army faction successful
May 17, 1972 left-wing political faction failed
May 15, 1973 right-wing army faction failed
June 5,1974 army faction failed
Dec.9,1977 army faction failed
July 28, 1978 army and political faction successful
Brazil
Oct. 29, 1945 elements from three services successful
Nov. 11, 1955 army faction successful
August 26,1961 elements from three services successful
April 1, 1964 elements from three services successful

Chile
June 29, 1973 right-wing army and political
faction failed
Sept. 11, 1973 elements from three services successful

Colombia
June, 13, 1953 elements from three services successful
 Appendix C. Statistics 203

Date Main Party Outcome

May 10, 1957 elements from three services successful
Cuba
March 10, 1952 army faction successful
Dominican Republic
Jan. 13, 1962 army faction failed
Sept. 25, 1963 army faction successful
April 25, 1965 army and air force faction successful
Nov. 26, 1965 right-wing army and political
faction failed
Ecuador
March 14, 1947 elements from three forces failed
August 23,1947 elements from three services successful
Sept. 1-3,1947 elements from three services successful
Nov. 7-9, 1961 elements from three services successful
July 11, 1963 elements from three services successful
March 29, 1966 political faction successful
Feb. 15, 1972 elements from three services successful
Sept. 1, 1975 military and political faction failed
Jan. 11, 1976 elements from three services successful
El Salvador
Dec. 14, 1948 army faction successful
Oct. 26, 1960 military and political faction successful
Jan. 25, 1961 right-wing political successful
March 25, 1972 army and political faction failed
Guatemala
July 9, 1949 army faction failed
Jan. 20, 1955 political faction failed
Oct. 25, 1957 elements from three services successful
Nov. 13,1960 left-wing army and political
faction failed
March 30, 1963 army faction successful
Haiti
Jan. 11, 1946 army faction successful
May 10, 1950 army faction successful
204 Coup d'Etat

Date Main Party Outcome

August 5-7, 1963 armed political faction failed
Honduras
Oct. 21, 1956 army and air force successful
Oct.3,1963 army faction successful
Dec.3,1972 army and air force faction successful
April 22, 1975 army faction successful
Oct. 21, 1977 right-wing political faction failed
Nicaragua
May 26, 1947 army faction successful
Jan.22-23,1967 right-wing army and political
faction failed
Panama
Nov. 20, 1949 police successful
May 9,1951 army and political faction successful
Oct. 12, 1968 national guard successful
Dec. 16, 1969 national guard faction failed
Paraguay
March 7, 1947 army faction failed
June 3,1948 political party faction successful
Dec. 30, 1948 political party faction successful
Feb. 26, 1949 political party faction successful
May 5,1954 army faction successful
Peru
Oct.3,1948 navy faction failed
Oct. 27, 1948 right-wing army faction successful
Feb. 16, 1956 right-wing army faction failed
July 18, 1962 elements from three services successful
March 3, 1963 elements from three services successful
Oct. 3, 1968 elements from three services successful
August 29, 1975 army faction successful
July 9,1976 right-wing army faction failed
Trinidad- Tobago
April 21, 1970 army mutiny failed
Venezuela
Nov.23,1948 army and political faction successful
 Appendix C. Statistics 205

Date Main Party Outcome

Nov. 11, 1950 political faction failed
Dec.2,1952 elements from three services successful
June 24, 1960 foreign-supported faction failed
May 4,1962 right-wing navy faction failed
June 3, 1962 left-wing navy faction failed
Oct. 30, 1966 national guard faction failed

MIDDLE EAST/ SOUTHWEST ASIA

Cyprus
July 15, 1974 national guard successful
Iraq
July 15, 1958 army faction successful
March 8-9, 1959 left-wing army faction failed
Feb. 8, 1963 army and air force faction successful
Nov. 18, 1963 air force faction successful
Sept. 17,1965 army faction failed
June 30, 1966 Nasserist army faction failed
July 17,1968 right-wing Baathist army faction successful
Jan. 20, 1970 foreign-supported faction failed
June 30, 1973 police and political faction failed
Lebanon
Dec. 31,1961 army and tribal faction failed
Oman
July 23, 1970 Sultan's son plus palace guard successful

Qatar
Feb. 22, 1972 royal faction successful

Sharjah
Jan. 24, 1972 political faction failed
Syn'a
March 30, 1949 army faction successful
August 18, 1949 army faction successful
Dec. 17,1949 army faction successful
Nov.28,195l army faction successful
Feb. 25, 1954 army faction successful
 ,
206 Coup d'Etat

. Date Main Party Outcome

Sept. 28, 1961 army and political faction successful
March 28, 1962 army faction successful
April I, 1962 Nasserist army faction failed
March 8, 1963 left-wing army faction successful
Feb. 23, 1966 left-wing Baathist army faction successful
Sept. 8, 1966 army faction failed
Feb.2B,1969 army faction successful
Nov. 13, 1970 right-wing Baathist army faction successful
Turkey
May 21,1960 elements from three services successful
Feb. 27, 1962 army faction failed
May 20,1963 army and air force faction failed
March 12, 1971 * general's faction in army successful
March 2, 1975 army faction failed
Yemen
Nov.5,1967 political faction successful
March 2, 1968 left-wing political and tribal
faction failed
July 25, 1968 army faction failed
Jan. 26, 1969 army faction failed
June 13, 1974 army faction successful
Oct. 11, 1977 army and political faction failed
Yemen., Democratic
June 26, 1978 army faction successful

EUROPE
Czechoslovakia
Feb. 21, 1948 Communist Party successful

* Military leaders threatened to take over the government unless a

strong coalition government was established to replace that of
Prime Minister SUleyman Demirel. Prime Minister Nibat Erim
established a new coalition government two weeks later.
 Appendix C. Statistics 207

Date Main Party Outrome

Greece
April 21, 1967 right-wing army faction successful
Dec. 13, 1967 King failed
May 24,1973 naval mutiny failed
Nov. 25, 1973 army and naval faction successful
Portugal
Jan. 1, 1962 army mutiny failed
April 25, 1974 army and political faction successful
Sept. 24, 1974 army faction failed
March 11, 1975 right-wing air force faction failed
July 31, 1975 elements from three services failed
Nov. 25, 1975 left-wing paratroopers failed
208 Coup d'Etat
TABLE III The Efficiency o/the Coup d'Etat, 1945-64
Outcome as a Function of Conflict Type
Conflict Half
Con- Incon- and Don't
Type tinuing elusive Lost Half Won Know

civil disorder 4 25 47 26 18 3
coup d'etat 0 0 24 2 62
military revolt 0 3 17 0 3
military/insurrection 0 4 33 8 2
guerrilla war 5 2 6 4 8 1
civil war I 2 4 4 6
border conflict 8 14 3 3 0
limited war 0 1 3 1 1
covert invasion 2 5 1 I I
blockage, etc. 0 0 2 2 0
threat 1 1 0 0 1

21 57 146 51 102 4
 Apppendix C. Statistics 209

TABLE IV. The Frequency o/the Coup d'Etat

Time Distribution of Type of Conflict 1945-64 (based on starting
date of conflict)
Time Periods: A I January 1946-30 April 1952
B I May 1952-3 I August 1958
C I September 1958-31 December 1964

Type of Conflict A B C
Internal:
civil disorder 35 32 So
localized internal:
coup d'etat 2S 18 44
military revolt/mutiny 8 6 17
insurrection 15 I2 12
widespread internal:
guerrilla war II 10 5
civil war 4 7 6
International :
conventional:
border conflict 7 12 9
limited war 3 2 I
other:
covert invasion 6 3 7
blockade, etc. 2 I I
threat 0 2 I

TOTAL: n6 lOS 159

 Index

Mghanistan : Buddhists, political activities of.

ethnic minorities in, S9 39-40 , 135-6
Aflak, Michel, 78, II3 Bulgaria, 22
Airports, control of, 125-'1 Bureaucracy, 19-23
Alawites, 77, 78, II4 post-coup stabilization of, 165-7
Algiers, 30-31, 105-'1, 1I2 Burma, 26
Anglo-Persian Oil Company, see ethnic minorities, S9
British Petroleum Burundi, 55
Arab-Israel War (1967), 43, 68
Arab world, mass media in, u8 Canada, 52
ARAMCO, 47-8 Cao Dai, I3S
Aref brothers, 79, 81-2 Catholic Church, political influence
Armed forces, 62 of, 13z-3, 135-6
infiltration of, 70-89, 181 Celtic separatism, rise of, 53
neutralization of, 64-89 Challe, General, 105. 106, IIZ
Portuguese case study, 68-'10 China, Imperial:
Atassi, Nurredin, President of bureaucracy, 20
Syria, 1I4 Manchu dynasty, 20
Aziz Ibn Saud, Abdul, 48 Civil war, 24
COMIBOL,142-3
Balewa, Aboubakar Tafawa, S4 Communiques, post-coup, 169
Be1Joh,Ahmadu,54 Congo, 51
Binh Xuyen. 135 Conrad, Joseph, Nostromo, 48
Bitar, Salah, 78, II3 Counter-espionage function, 101
'Black Power' movement, 139 Counter-intelligence function, 101
Blocking forces, 150,186-8 Coup:
objective of, 187 basic list, 194-8
Bokassa, Colonel Jean Bedel, 169 definition, 26
Bolivia: efficiency, 198
trade union activities in, 142-3 frequency, 199
Britain: 'Czech' arms deal. 42
bureaucracy. 19,28 Czechoslovakia:
diplomatic doctrine, 174 Novotny govemment,
police, 94-5 displacement of (1968), 76
political life, forces in, 107-8
Security Service (Home Office). Debre, Prime Minister, 106
101 Defences of the State, neutralization
Special Branch, 101 of,61-104
trade unions, 143 De Gaulle, General, 30-32, 62.
British Petroleum, 49-50 105-'1
2I2 Index
De Pouilly, General Henri, 105-6 Government:
Diem, 39-40, 135 alternative forms of, 113, II4
Diplomatic recognition, 173-4 formal structure of, 108, 112-16
Druzes, 59, 77, 78 personalities, neutralization of,
Duvalier, 180-81 lOS-II, III-17
Dzerzhinsky, 120 Greece, 22, 26n., 1I8, 171

East Germany, 39,40 Hafez, 64, 78, 113

Economic backwardness, 33-8 Haiti, 170-72
Economic development, 190-94 Hamdun, Captain Mustafa, 169
Economics of repression, 175-81 Hasan AI-Banna, Shaykh, 140
Egypt: Hinduism, political influence of, 134
Arab-Socialist Union, 29 Hitler, Adolf, 60, 79n.
coup of 195 2 , 34 Hoa Hao, 135
demonstrations of 1967, 35n. Hue, 39
government, 1I3 Hungary:
Six Day War (1967), 43, 68 Revolution of 1956, 38
Soviet Union, relations with, Husni al-Za'im, 48, 78
42 -4
Suez-Sinai War (1956), 42 Ibn Yahya, Ahmad, 43
EI Azhar University, Cairo, 133 Ibo nation, 76
Ethnic minorities, 59 India, 26, 52
Indonesian coup-cum-revolution, 65
Farouk, King, 34 Internal intelligence, 102-4
France: Iraq, 26
Algiers, 30-31, 105-7, 112 Aref brothers in, 81-2
Gabon coup (1964), 41 coup of 1958, 26n.
Garde Repllblicaine, 93 coup of 1963, 65
Gendarmerie, 90-92, 120 ethnic minorities, 59
police, 20, 90-92, 14sn. July 14th revolution, 24
Prefecture de Police, 90-93 Quassen regime, 79
Security Service, 20 Iraq Petroleum Company, 48
South Vietnam rubber Islam, political influence of, 133-5
plantations, 49 Six Day War (1967), 43, 68
Surete Nationale,90-91 Suez-Sinai War (1956), 42
telecommunications in, 120 Syria and, 64
trade unions, 143 Italy, 31-2
Catholic Church in, 131-3
Gabon, 41 police, 20, 96-7, 102-3
Ghana, 140, 181 Reggio Emilia incidents (1964),
communique (1966), 169 144
coup of 1966, 129 security services, 20, 102
National Liberation Council, trade unions, 143
169, 174
police system, 96 Jadid, Salah, 64, 1I4
police telecommunications, 120 Jakarta, 66
 Index 213
Jet fighters, political importance of, police, 89-99
41-2 political forces:
Jouhaud, General, 105 general, 1I0-30
particular, 130-45
Kalonjii, 'King', 51 political parties:
Karens,26 in developed countries, 141
Katanga, 47, 52 'insurrectional', 138-40
Kenya, 26 'machine', 137-8
Kerensky,60 para-bureaucratic, 140-41
Kornilov, General, 25, 60 public buildings, 128-30
Kurds, 26, 59 religious organizations, 131-6
security agencies, 99-104
Latin America: telecommunications, 120-22
armed forces in, 85 trade unions, 142-5
Lebanon: traffic points, focal, 124-5
government of, 54 Ngo Dinh Nhu, 40
Leopoldville (Kinshasa), 52 Nigeria:
Liberation, 25 communique (1966), 169
wars of national, 26
counter-coup (1966), 36
Loyalist forces, isolation of
government, 53
hard-core, 148-52
Ibo nation and, 76
N'krumah, Kwame, 36n., 129, 140,
Makhous, Ibrahim, II4
174, lSI
Malinovsky, Roman, 103n. Nzeogwu, Major, 169
Mass media; neutralization of,
IIS-20
post-coup control of, 168-72 Okhrana, the, 103n.
Mattei, Enrico, 32n.
Michael, King of Rumania, 25 Pahlavi, Reza, 49
Militias, 59 Papon, Maurice, 106
Mob control, 1440. Para-military forces, 95-104
Muslim Brotherhood, 139-40 Pathans,59
Persia,49-50, II3
Nagas,26 Physical targets, 122-30
Nasir, Abd-el (Nasser), 29-30,34, Poland, 'Zionist plot' in, 76n.
35n., II3, 133 Police:
Nasser, see Nasir Abd-el British, 93
Nazis, 59 infiltration of, 182
Neutralization of: French, 90-94
airports and transport facilities, national, 97-9
125-8 neutralization of, 89-99
armed forces, 64-89 political department, 101
city entry-exit road links, 122-5 rural, 97
mass media, 118-19 urban, 97-9
personalities: Political independence, 38-47
inside government, III-I7 Political parties, neutralization of,
outside government, 117 137-41
 214 Index
Portugal, armed forces situation in, Security agencies, neutralization of.
68-'70 99-104
Post-coup situation, immediate: Security system, 61
communications, control of mass, Senghor, President, of Senegal, 106
168 Sequence,opera~,152, 153-7
cQmmuniqut!s, first, 169 Shans.59
curfew, 167 Shishakli, 78
diplomatic recognition, 173-4 Somalis, 26
'legitimization' or. 172- South Kasai:
public transport, interruption of, political situation (1960-61), 51
167 South Korea, 4411.
reassurance, 169 South Vietnam, 26, 39-41, 49
stabilization: Buddhists in. 135-6
of bureaucracy, 165-'1 Soviet Union:
of forces, 163-5 'boUtgeois' February revolution,
of masses, 167-'14 60
Pravda. 103n. East Germany and, 39. 40
Pre-conditions, 32 Egypt and. 42-4
Pronunciamiento, 24-5, 26 federal state. 52
Public buildings, occupation of, Hungary and, 38
128-30 Left Socialist Revolutionary coup
Pure intelligence function, 100 (1918), failure of, 120
Putsch, 25. 26 Spain:
police system. 98
Quang, Tri Thicb, 135 pronunci4miento, 24
Quassen, 79, 82 Syria. 23, 48
Quwatli, President, of Syria, 78 coup of 1966, 260., II3-14
ethnic minorities, 59, 77, 78
Regional entities, SO-56 government, 114
Religious organizations, political 'canvassing' in, 85
neutralization of, 131-6
Revolution, 24 Tactical aspects, 182-8
Reza Shah, (Reza Pahlavi). II3, Tanzanian African National Union,
49-50 140
Roosevelt, Kermit, 420. Tatgets classification, 157-62
Rumania. 22, 25, 76 Telecommunications:
Russia. Imperial: facilities available to government,
'occupative' throne of, 23 121
secret police (Okhrana). I03Q. neutralization of, 120-23
Russia, Soviet, see Soviet Union police, in Ghana. 120
Timing. in coup, 152, 153-7
Saigon, 39, 40 Ton Ton Macoutes, 180-81
Salan, General, lOS Trade unions, 142-5
Saudi Arabia: Traffic points. focal. 124-5
bureaucracy, zo-a3 Transport facilities, control of,
ARAMCO,47-8 125-8
 Index :us
Trevelyan, R., 420. Secret Service (Treasury), 103
Trotsky, 63 South Vietnam, 39-41
Tshombe, President, 47 telecommunications in, 120

Union Miniere, 47 Van Minh, 40

United Fruit Company, 48 Viet Cong, 26, 40, 49
United States of America:
armed forces, 62 Weimar Germany, 59
"Black Power' movement in, 139 West Germany, 52
bureaucracy, 19-20 infiltIation of armed forces in.
CIA,62 (notional), 89
Egypt, break with (1966), 43 West Pakistan:
FBI, 62, 94, 101 ethnic minorities, 59
federal government of, 52
General Motors and, 46, 55
Yemen:
intelligence system, 62
Middle East policy, influential coup of 1962, 26n., 43, 174
groups on, 108 Yugoslavia, 29. 76
NSA,62
National Guard, 94 Zeller, General, lOS
police, 93-5 Zwayeen, Youssof, II4