INTERNATIONAL RESTRAINTS AND FOREIGN POLICY CHOICES:
BRITAIN FACES THE DICTATORS, 1931-1941
Stephen George Walker
A Dissertation Presented to the Graduate Council of
the University of Florida
in Partial Fulfillment of the Requirements for the
Degree of Doctor of Philosophy
UNIVERSITY OF FLORIDA
Stephen George Walker
To My Parents
Several individuals and institutions contributed materially to
this dissertation. My graduate committee, Professors John Spanier,
Al Clubok, Marvin Entner, Oscar Svarlien, and David Conradt, patiently
read and criticized the manuscript at various points in its development.
My fellow graduate students and faculty colleagues at the University of
Florida and Arizona State University provided an intellectual climate
that facilitated the research, as did my undergraduate pupils at both
schools. The University of Florida and Arizona State University were
indispensable sources of financial aid. Finally, my family contributed
encouragement and inspiration along the way, and made financial and
psychological sacrifices that were necessary for me to complete the
TABLE OF CONTENTS
ACKiNOULEDGMENTS . . . . . . .
LIST OF TABLES . . . . . . . . . .
LIST OF FIGURES . . . . . . . . .
ABSTRACT . . . . . . . . . . .
PART I. RESEARCH DESIGN
CHAPTER I. THEORIES OF BRITISH FOREIGN POLICY .
Historical Interpretations . . . . .
Balance of Power Theory . . . . .
Systems Theory . . . . . . . .
Decision-Making Theories . . . . . .
Summary . . . . . . . . . .
CHAPTER II. THE MEASUREMENT OF BRITISH FOREIGN POLI
Identifying British Decision Makers . . .
Scaling the Dependent Variables . . . .
Scaling the Independent Variables . . . .
Analytical Procedures . . . . . . .
PART II. DATA COLLECTION
CHAPTER III. BRITISH CONFLICT BEHAVIOR, 1931-1936
The Sino-Japanese Conflict . . . . .
The Italo-Abyssinian Conflict . . . . .
The German Occupation of the Rhineland . .
CHAPTER IV. BRITISH CONFLICT BEHAVIOR 1937-1941 .
Abyssinia and Spain . . . . . . .
Eden vs. Chamberlain . . . . . . .
Austria and East Europe . . . . . .
The Far East . . . . . . . . .
PA1TZ TTT. DATA ANALYSIS
CHAPTER V. INTERNATIONAL RESTRAINTS AND BRITISH FOREIGN POLICY . 135
Overview .................. ......... 135
British Behavior in Non-Military Situations . . . . .. 142
. . . . . iv
. . . . . vii
. . . . . xi
. . . . . xii
British Behavior in Military Situations . . . . . .. 155
Statistical Summary . . . . . . .. ... . . . 184
CHAPTER VI. BRITISH FOREIGN POLICY AND THE LEVELS OF ANALYSIS
PROBLEM . . . . . . . . . . . . 207
British Foreign Policy in Retrospect . . . . . . . 207
The Levels of Analysis Problem in International Relations . 215
APPENDIX I . . . ... .. . . . . . . . . . 228
APPENDIX II . . . . . . . . . . . . . . . 246
BIBLIOGRAPHY . . . .... . . . . . . . . . . 253
LIST OF TABLES
1. International Restraints and British Foreign Policy Toward
Japan During the Manchurian Conflict . . . . .. 54
2. International Restraints and British Foreign Policy Toward
Japan During the Shanghai Crisis. . . . . .. .60
3. International Restraints and British Foreign Policy Toward
Italy During the Italo-Abyssinian Conflict . . .. 72
4. International Restraints and British Foreign Policy Toward
Germany During the Rhineland Crisis . . . . . 78
5. International Restraints and British Foreign Policy Toward
Italy During the Spanish Civil War. . . . . . .86
6. International Restraints and British Foreign Policy Toward
Germany and Italy, 1937-1938. . . . . . . 94
7. International Restraints and British Foreign Policy Regarding
the Anschluss Question .. . . . . . . . .99
8. International Restraints and British Foreign Policy During
the Sudeten Crisis . . . . . . . . .. . 108
9. International Restraints and British Foreign Policy Toward
Germany and Italy During the Spring of 1939. . . ... 115
10. International Restraints and British Foreign Policy During
the Polish Crisis . . . . . . . . . . 119
11. International Restraints and British Foreign Policy During
the Sino-Japanese Conflict, 1937-1938 . . . . .. 126
12. International Restraints and British Foreign Policy During
the Sino-Japanese Conflict, 1938-1939 . . . . .. 131
13. International Restraints and British Foreign Policy Toward
Japan, 1940-1941 .. . . . . . . . . . 133
14. Decisions by Target and Intensity of Conflict Behavior,
1931-1941 . . . . . . . . . . . . 135
15. Decisions by Type of Policy Goal and Type of Policy Change,
1931-1941 . . . . . . . . . . . . 137
16. Decisions by Explanatory Models. .. . . . . . 138
17. Decisions by Type of Conflict Behavior and Type of Conflict
Situation . . . . . . . . . . . .. 140
18. Decisions in Military Situations by Type of Conflict Behavior
and Military Capabilities . . . . . . . .. 141
19. Decisions in Non-Military Situations by Type of Conflict
Behavior, Diplomatic Stakes, and Existence of Cross-
Pressures . . . . . . . . . . . 145
20. Decisions in Non-Military Situations by Type of Policy Change,
Diplomatic Stakes, and Existence of Cross-Pressures . . 146
21. Decisions in Non-Military Situations by Perception of Cross-
Pressures and Type of Diplomatic Stakes .. . . . . 147
22. Decisions in Non-Military Situations Over Time by Existence of
Cross-Pressures, Intensity of Conflict Behavior, and
Decision Maker . . . . . . . . . . . 152
23. Similarities and Differences in Decision-Making Patterns
Between Eden and Chamberlain in Non-Military Situations . 153
24. Decisions in Military Situations by Intensity of Conflict
Behavior, Existence of Cross-Pressures, and Relative Military
Capabilities . . . . . . . . . . . 156
25. Decisions by Intensity of Conflict Behavior and Diplomatic
Stakes for Military Situations with Cross-Pressures and Low
Capabilities . . . . . .. . . . . . 158
26. Decisions by Intensity of Conflict Behavior and Diplomatic
Stakes for Military Situations Without Cross-Pressures but
with Low Capabilities . . . ... . . . . . 160
27. Decisions by Intensity of Conflict Behavior and Diplomatic
Stakes for Military Situations with Cross-Pressures and
High Capabilities . . . . . . . . . . 162
28. Decisions by Intensity of Conflict Behavior and Diplomatic
Stakes for Military Situations with High Capabilities and
No Cross-Pressures . . . .. . . . . . 163
29. Decisions in Military Situations by Type of Policy Change,
Diplomatic Stakes, Cross-Pressures, and Relative Capa-
bilities . . .. . . . . . . . . 173
30. Decisions in Non-Military Situations by Type of Policy Change
and Complexity of the Situation . . . . .. . 175
31. Decisions in Military Situations by Perception of Cross-
Pressures and Type of Diplomatic Stakes . . ... .177
32. Decisions in Military Situations by Type of Conflict Behavior,
Cross-Pressures, Date, and Decision Maker. . . . 178
33. Decisions in Military Situations Between 1937 and 1941 by
Decision Maker, Cross-Pressures, and Target . . 179
34. British Conflict Behavior and Perceptions of Relative Capa-
bilities Under Cross-Pressures, 1937-1941 ..... .181
35. Coefficients of British Policy Change in Non-Military
Situations . . . . . . . . ... .. . .200
36. Zero Order and Partial Coefficients of British Behavior Toward
Italy and Japan, 1937-1941 . . . . . .... .202
37. British Conflict Behavior and the Relative Explanatory Power
of National Interests Theory and Systems Theory . . 203
38. Zero Order and Partial Coefficients of British Policy Change
Toward Italy and Japan, 1937-1941 . . . . .. 204
39. British Policy Change and the Relative Explanatory Power of
National Interests Theory and Systems Theory . . 205
40. Explanations of Conflict Behavior and Voting Behavior Based
Upon the Cumulative Effects Principle and Attitudinal Data 225
41. Foreign Policy Choice by Diplomatic Stakes . . .. 247
42. Foreign Policy Change by Diplomatic Stakes . . .. 247
43. Cross Pressures by Diplomatic Stakes ....... .247
44. Relative Capabilities by Diplomatic Stakes ..... .247
45. Foreign Policy Choice by Cross Pressures . . .. 248
46. Foreign Policy Change by Cross Pressures . . 248
47. Relative Capabilities by Cross Pressures . . .. 248
48. Foreign Policy Choice by Relative Capabilities . . 248
49. Foreign Policy Change by Relative Capabilities . . 249
50. Foreign Policy Choice by Diplomatic Stakes . . 249
51. Foreign Policy Change by Diplomatic Stakes . . 249
52. Cross Pressures by Diplomatic Stakes . . . . ... 249
53. Foreign Policy Choice by Cross Pressures . . . ... 250
54. Foreign Policy Change by Cross Pressures . . . ... 250
55. Foreign Policy Choice by Diplomatic Stakes . . . ... 250
56. Foreign Policy Change by Diplomatic Stakes . . . ... 250
57. Cross Pressures by Diplomatic Stakes . . . . ... 251
58. Relative Capabilities by Diplomatic Stakes . . . ... 251
59. Foreign Policy Choice by.Cross Pressures . . . ... 251
60. Foreign Policy Change by Cross Pressures . . . ... 251
61. Relative Capabilities by Cross Pressures . . . ... 252
62. Foreign Policy Choice by Relative Capabilities ...... 252
63. Foreign Policy Change by Relative Capabilities . . .. 252
LIST OF FIGURES
1. A Model of International Conflict Behavior Based Upon
Systems Theory . . . . . . . . . . .
2. A Model of Conflict Behavior Based Upon Decision-Making
Theories . . . . . . . . . . . .
3. Alternative Models of British Conflict Behavior
4. Different Cutpoihts for the Conflict Behavior Scale That
Measures British Involvement in Conflict Situations
5. Pattern of Observations for Hypothetical Foreign Policy
Decisions X and Y . . . . . . . . . .
6. Seven Alternative Definitions of the Situation and the
Theoretical Direction of British Conflict Behavior . .
7. Rules for Fitting British Policy Choices to the Different
Causal Models Suggested by Systems Theory and the Decision
Making Theories . . . . . . . . . . .
8. Alternative Models of British Conflict Behavior . . .
9. No Cause . . . . . . . . . . . .
10. Partial Cause . . . . . . . . . . .
11. Immediate Cause . . . . . . . . . . .
12. Remote Cause . . . . . . . . . . .
13. Single Cause . . . . . . . . . . .
14. Reciprocal Cause .........
15. Reverse Cause
16. A Logically-Based Hybrid of the National Interests and Systems
Theory Models . . . . . . . . . . . .
17. An Empirically-Based Hybrid of the National Interests and
Systems Theory Models for Military Situations .. . ....
18. An Additive Hybrid of the National Interests and Cross-
Pressures Models for Non-Military Situations ...
. . . . . . . . . . . 187
19. Alternative Causal Interpretations of the Intensity of
British Conflict Behavior in Military Situations,
20. An Additive Model for Intensity of Conflict Behavior in
Non-Military Situations, plus Supporting Evidence . .. 196
21. Alternative Causal Models of Foreign Policy Change in
Military Situations . . . . . . . .... . 197
22. An Additive Model for Foreign Policy Change in Non-Military
Situations . . .. . . . . . ... .199
23. Alternative Causal Interpretations of the Intensity of
British Conflict Behavior in Military Situations Toward
Italy and Japan, 1937-1941 . . . . . . ... .200
24. Alternative Causal Interpretations of British Foreign Policy
Change Toward Italy and Japan in Military Situations,
1937-1941 . . . . . . . . . . . . 204
25. Correlates of British Foreign Policy Behavior in Conflict
Situations . . . . . . . . . .. . 211
Abstract of Dissertation Presented to the Graduate Council
in Partial Fulfillment of the Requirements for the Degree of
Doctor of Philosophy
INTERNATIONAL RESTRAINTS AND FOREIGN POLICY CHOICES:
BRITAIN FACES THE DICTATORS, 1931-1941
Stephen George Walker
Chairman: John W. Spanier
Major Department: Political Science
This study attempts to identify and explain variations in British
conflict behavior toward Germany, Italy, and Japan between 1931 and
1941. The general hypothesis in the investigation is that British
conflict behavior varies according to the decision maker's definition
of the international situation. Specifically, changing perceptions of
cross-pressures, relative military capabilities, and diplomatic stakes
account for variations in the intensity of British conflict behavior
and its potential for rapid escalation.
In an examination of the principal diplomatic documents and mem-
oirs for the period, eighty-five British decisions and their deter-
minants were identified, coded, cross-tabulated, and subjected to
Pearsonian correlation analysis. The analysis confirmed the follow-
ing propositions. Regarding the intensity of British conflict be-
havior for the entire period, 1931-1941: (a) if cross-pressures ex-
isted, then the intensity of British conflict behavior was likely to
be low; (b) if British military capabilities were low, then the
intensity of British conflict behavior was likely to be low; (c) if
British diplomatic stakes were low, then the intensity of British con-
flict behavior was likely to be low; (d) if the number of these inter-
national restraints was high, then the intensity of British conflict
behavior was very likely to be low. Regarding the escalation potential
of British conflict behavior for the entire period, 1931-1941: (a) if
British diplomatic stakes were low, then their conflict behavior was
likely to escalate rapidly: (b) if the British were simultaneously in-
volved in several conflicts, they were likely to experience cross-
pressures; (c) if the British experienced cross-pressures, they were
likely to perceive their available military capabilities per conflict
as low; (d) if British perceptions of available military capabilities
were low, then their conflict behavior was unlikely to escalate rapidly.
These propositions fit two general theories, respectively: the
cumulative effects principle and the principles of international
systems theory. The findings regarding conflict intensity express the
cumulative effects principle, namely, that each of the independent
variables separately influences the intensity of British conflict be-
havior, and the cumulative impact of these variables is greater than
their individual effects. The findings regarding escalation potential
conform to the principles of international systems theory, which are
as follows: (a) no matter what the diplomatic stakes are, the greater
the number of members in the international system, the more likely it
is that each member becomes simultaneously involved in several con-
flicts; (b) cross-pressure from simultaneous involvement in several
conflicts lowers the members' perceptions of available military capa-
bilities per conflict; (c) perceptions of low military capabilities
restrain conflict behavior and thereby maintain the international
Two major conclusions emerge from the study. First, international
considerations offer a powerful, empirically-based, and theoretically-
acceptable explanation of British conflict behavior. Second, the re-
sults support the contention that international relations theory based
upon the perceptions of individual decision makers may be more pro-
ductive than theory based upon the characteristics of aggregate units,
such as groups or nations.
THEORIES OF BRITISH FOREIGN POLICY
Most scholars characterize the 1930's period as the era of ap-
peasement in British diplomatic history. Some criticize British policy
during this decade as-unrealistic, shortsighted, naive, or even con-
spiratorial.l Both historians and actual participants in these Brit-
ish decisions have written numerous books to condemn or defend British
policy toward the three fascist dictatorships, Germany, Italy, and
Japan.2 In view of the extensive and intensive treatment of this period
by several eminent observers and participants, the examination of
British policy by still another analyst calls for some justification.
Some books on the period with particularly suggestive titles in-
clude Margaret George, The Warped Vision (Pittsburgh, 1965); lan Colvin,
None So Blind (New York, 1965); John L. Snell (ed.), The Outbreak of the
Second World War: Design or Blunder (Boston, 1962); Donald N. Lammers,
Explaining Munich: The Search for Motive in British Policy (Stanford,
1966); John F. Kennedy, Why England Slept (New York, 1961); Martin
Gilbert and Richard Gott, The Appeasers (Boston, 1963); Martin Gilbert,
The Roots of Appeasement (New York, 1967); William Rock, Appeasement
on Trial (Hamden, 1966).
2Snell, ibid., surveys the representative historical works on the
period. A.J.P. Taylor's book, The Origins of the Second World War
(New York, 1962), is probably the most controversial interpretation. A
very critical review of Taylor's book by Hugh Trevor-Roper appears in
Encounter, XVII (July, 1961), 88-96; a more balanced critique is the
review by F.H. Hinsley in Power and the Pursuit of Peace (Cambridge, 1963),
Chapter 15. The most informative and least polemical memoirs by British
participants in my opinion are Viscount Templewood (Sir Samuel Hoare),
Nine Troubled Years (London, 1954), which defends the appeasement deci-
sions, and Anthony Eden (Lord Avon), Facing the Dictators (Houghton
Mifflin, 1962), which criticizes them. Both men served as Foreign
Secretaries in the 1930's.
A new look at a particular historical period is justified if new evi-
dence becomes available, or if the analyst brings with him a new analyt-
ical perspective, one that increases the understanding of already avail-
able information. This study does not offer new sources of information
as its justification. It does offer a new analytical perspective.
The purpose of this investigation is to re-examine British policy
during the 1930's and attempt to fit it into a theoretical framework.
Because of this focus the research is primarily an exercise in empirical
political theory rather than diplomatic history. The distinction between
empirical theory and history is an artificial one in that both the theo-
rist and the historian deal with the same events in their analysis.
Nevertheless, the distinction between a theoretical analysis and a his-
torical interpretation is a viable one. A review of the principal his-
torical interpretations for this period should help to illustrate this
point and provide an introduction to the theoretical framework that
guides this new look at an old topic.
The historical interpretations of British foreign policy during the
1930's are difficult to separate from those interpretations that explain
the occurrence of the Second World War. Historians often link them by
relating the British policy of appeasement to the outbreak of the war.
British policy becomes a "permissive" cause of World War II by allowing
Hitler to rearm Germany, consolidate German western frontiers, and
expand toward the east. The British failed to prevent Hitler from
J. Watkins, "Historical Explanations in the Social Sciences," in
Patrick Gardiner (ed.), Theories of History (Glencoe, Illinois, 1959),
pp. 503-514, especially pp. 513-514.
occupying the Rhineland in March, 1936, and merely protested the German
annexation of Austria two years later. Within six months after the
Anschluss Prime Minister Chamberlain accepted the cessation of the
Sudetenland to Germany at Munich. Finally, after Hitler conquered the
remainder of Czechoslovakia in the spring of 1939, the British guaran-
teed the territorial integrity of Poland. When Hitler attacked Poland
in September, 1939, the British government, with the French as their
allies, came to Poland's defense. The British decision to honor this
commitment belatedly but irrevocably reversed their earlier appease-
ment policy, which was to concede Hitler's territorial demands in an
attempt to reach a peaceful European settlement with Germany.
Most historians agree with this description of appeasement, but
they differ over its sources. Why did the British wait until 1939 to
oppose Hitler with force? Some historians suggest that British policy
did not take sufficient account of Hitler's personality. They believe
that Hitler was either a madman or a revolutionary, or both, but cer-
tainly not an orthodox political leader. British decision makers, on the
other hand, were reasonable men and expected Hitler to be reasonable and
satiable in his demands, an unrealistic expectation according to this
The following division of historians into schools is somewhat arbi-
trary. On the one hand, historical research is often very narrowly focus-
ed and does not pretend to offer more than a partial explanation of the
phenomena in question. On the other hand, the historical research that
synthesizes the monograph literature often either surveys it without
really synthesizing it or gives a "balanced" interpretation that is some-
what ambiguous. A few scholars, such as George, op.cit., Gilbert and
Gott, op.cit., and Taylor, op.cit., have attempted to definitively in-
terpret Anglo-German relations, but the number of variables that they in-
clude in their analyses still makes classification risky. Therefore, the
works cited as examples of the two schools below are approximations of
"ideal types" of interpretations: these examples may not exactly fit the
idealized interpretation, but do express their general themes.
interpretation. The British failure until after the Prague coup to see
that Hitler was unappeasable, therefore, accounts for the duration of
the appeasement policy.5 A second group of historians argues that the
British government realized Hitler's unreasonable designs but were in-
capable of stopping him. They contend that the principal restraints
upon the British elite were domestic political considerations and rela-
tive military capabilities. British public and parliamentary opinion
would not support the measures necessary to stop Hitler before 1939.
Painful memories of Wbrld War-I and similar visions of World War II made
it difficult for British leaders to justify policies of rapid rearmament,
alliances, and military intervention when a policy of negotiations existed
as an alternative. The futility of negotiations did not become completely
clear until after the Czech coup.
It is possible to synthesize these schools of thought. A theme that
appears in both interpretations is the British disposition to view some
of Hitler's publically-stated objectives as reasonable revisions of the
Versailles Treaty; the British primarily opposed Hitler's heavy-handed
methods of obtaining them. An integrated interpretation suggests that
British policy makers responded to the reasonableness of Hitler's initial
objectives and the climate of opinion in Britain by succumbing to
5William Newman, The Balance of Power in the Interwar Years, 1919-
1939.(New York, 1968), pp. 19-32. Martin Gilbert and Richard Gott, The
Appeasers (Boston, 1963), pp. 5-9. Arnold Wolfers, Britain and France
Between Two Wars: Conflicting Strategies of Peace From Versailles to
World War II (New York, 1966), pp. 209-211, 226-228, 251-253.
6John F. Kennedy, Why England Slept (New York, 1961), xxi, xxii
and passim. Margaret George, The Warped Vision (Pittsburgh, 1965),
pp. 39-55. A.L. Rowse, Appeasement (New York, 1961), pp. 14-15, 116-117.
Winston Churchill, The Gathering Storm (Boston, 1948), pp. 71-77, 119-123,
unrealistic expectations about resolving Hitler's later demands through
negotiations. Misperceptions of the international environment and
domestic considerations reinforced one another to jointly account for
the belated shift in British policy from appeasement to rearmament
and the use of force.
Balance of Power Theory
The different historical interpretations of British appeasement
provide the details for an empirical theoretical analysis. Theoretical
analyses attempt to explain historical events by explicitly subsuming
them under one or more general principles: historical analyses do not
usually explicitly attempt this task. Both the occurrence of World
War II and British appeasement policy toward Germany are susceptible to
This synthesis is most systematically expressed by George Lanyi,
"British Foreign Policy Revisited: The Domestic Sources of Appeasement,"
a paper prepared for delivery at the American Political Science Associa-
tion Meeting (Los Angeles, California, 1970).
This distinction follows Ernst Nagel, The Structure of Science
(New York, 1961), pp. 550-551...."There is an important asymmetry be-
tween theoretical (or "generalizing") science and history. A theo-
retical discipline like physics seeks to establish both general and
singular statements, and in order to do so physicists employ previously
assumed statements of both types. Historians, on the other hand, aim to
assert warranted singular statements about the occurrence and the inter-
relations of specific actions and other particular occurrences. However,
although this task can be achieved only by assuming and using general
laws, historians do not regard it as part of their aim to establish such
laws." Underlining Nagel's.
This classification is based upon types of explanations, and im-
plies that anyone who constructs theoretical explanations is a theorist
while anyone who does not is a historian. Actually, some people who
consider themselves historians do construct theoretical explanations and
vice versa. Two examples of theoretical explanations by diplomatic
historians are Edward Robert Burr, By Reason or Force: Chile and the
Balancing of Power in South America, 1830-1905 (Berkeley, 1965). James
Rosenau, "The Games IR Scholars Play," Journal of International Affairs
XXI (1967), 293-303, reviews the merits of these alternative styles of
analysis within the discipline of political science.
analysis by the principles of balance of power theory. Balance of power
theory is empirical in its treatment of international politics but
normative in its analysis of foreign policy. The theory's empirical
propositions describe international politics as a struggle for the pro-
tection or achievement of national interests by sovereign states. An
increase in one state's military capabilities by armament, conquest, or
alliances usually elicits corresponding increases in the capabilities
of other states. If all states behave this way, wars are infrequent
because no one state or group of states possesses the capabilities
necessary to insure victory in a military conflict. The theory's norma-
tive principles of foreign policy specify first that individual states
should not permit other states to change significantly the distribution
of military capabilities, and second that a state should not go to war
unless it possesses the military capabilities necessary to win.
William J. Newman's book, The Balance of Power in the Interwar
Years, 1919-1939, clearly reflects the normative foundations of balance
of power theory. Newman suggests four possible explanations for the
occurrence of World War II and the British failure to follow balance of
First, the principles of the balance of power may not have been
understood or properly applied....(Second)....The principles of
the balance may be inadequate to meet the really significant
problems of aggression and expansion...(Third)....the specific
historical circumstances of this period and especially of the
1930's made it impossible for the principles to be applied
effectively. Specifically,...Hitler was such a special phenomenon
The most comprehensive explication of balance of power theory ap-
pears in Inis Claude, Power and International Relations (New York, 1962);
a shorter treatment is Ernst Haas, "The Balance of Power: Policy, Pre-
scription or Propaganda," World Politics, V (July, 1953), 442-77.
10William Newman, The Balance of Power in the Interwar Years, 1919-
1939 (New York, 1968), pp. 15-19.
that none of the normal rules of international relations would
hold and that therefore none of the normal restraints of power
or control could have prevented him from starting a world war....
(Fourth)....the failure in the 1930's was a manifestation and
climax of a long-term historical trend of decline and decay of
the whole system of the balance.....World Wars I and II, from
this point of view become part of the same evolution and have
basically similar causes.
Newman accepts the first explanation as the most adequate one,
while simultaneously admitting that "...aspects of the system changed
during the 1930's in ways significant enough to warrant the use of the
term 'watershed' for the cumulative effects of these changes on the
balance of power."ll Although the British did not correctly apply the
balance of power strategy, there were extenuating circumstances. Hitler
was a high risktatker who could not be restrained by the normal appli-
cation of balance of power principles. The correct application of bal-
ance of power principles against an expanding state led by a risktaker
required changes in the mix of coercion and negotiation, changes in
the timing associated with the application of countervailing power
against the expanding state, and changes in the role of the holder of
the balance. As holders of the balance between France and Germany,
the British failed to realize that a clear preponderance of power against
Germany and the early use of coercion in response to minimal German
attempts to overthrow the balance were necessary to deter a risktakter
like Hitler and thereby limit German expansion.
Newman's theoretical analysis does not contradict the historical
interpretations elaborated earlier. His prescriptions simply state the
steps that the British elite should have taken to avoid World War II --
if they had not been hampered by domestic considerations or if they had
11Ibid., p. 38.
12Ibid., pp. 38-43.
correctly perceived Hitler as a risktaker. Together, the historical
analyses and balance of power theory give plausible answers to questions
that are significant for the 1930's period: Why did World War II occur:
What should have been done to prevent its occurrence? Newman has also
extrapolated lessons from the British experience with Hitler, which are
intended to have significance for contemporary policy makers. He notes,
"What might be called the Munich Syndrome...has become a model of what
should not be done and of a particular type of foreign policy situation
that is assumed to be true of international relations since Hitler."13
Evidence of the existence of the Munich Syndrome among contemporary
policy makers appears, for example, in the writings of Truman and
I recalled (at the outbreak of the Korean War) some earlier
instances: Manchuria, Ethiopia, Austria. I remembered how
each time that the democracies failed to act it had encouraged
the aggressors to go ahead. Communism was acting in Korea
just as Hitler, Mussolini, and the Japanese had acted ten,
fifteen, and twenty years earlier....If this was to go un-
challenged it would mean a third world war, just as similar
incidents had brought on the second World war.--
If we add the United States to Britain and France; if we change
the name of the potential aggressor; if we substitute the United
Nations Organization for the League of Nations, the Atlantic
Ocean for the English Channel, and the world for Europe, the
argument is not necessarily without its application today.--
The argument to which Churchill refers is the lesson that appease-
ment is not a viable foreign policy in a conflict situation between
democracies and revolutionary dictatorships--a belief shared by many
13Ibid., pp. 34-35
14Harry S. Truman, Memoirs, II (Garden City, 1956), p. 333.
15Winston Churchill, The Gathering Storm (Boston, 1948), p. 211.
contemporary academicians as well as policy makers.
The victims of the Munich Syndrome share several major premises.
First, they assume that Hitler's Germany was easily recognizable as the
principal foreign policy problem that faced the British, even if the
precise nature of the problem was initially unclear. Second, they
maintain that British decision makers could have and should have de-
voted most of their time to the German Question. Third, they attribute
the failure to solve the German riddle to domestic distractions, in-
experience in dealing with revolutionaries, or the limited nature of the
German challenge in its outward appearances. These premises are in-
complete, however, as still another analyst has noted:
Only close reading of several years of Cabinet papers can give
one an accurate impression of how...various foreign policy
problems crowded the policy makers and how they did not permit
them to concentrate on the German danger, certainly not before
the Spring of 1938. Only a few men, who were on the periphery
of the elite, foremost among them Winston Churchill, were able
to look at all foreign policies, certainly after 1935, from
the point of view of the one, chief and main danger: the
German challenge....This singleminded outlook was almost im-
possible to sustain for those whose daily business demanded a
highly fragmented attention to a variety of foreign policies.
That, one must conclude, is the inescapable destiny of any ex-
tensively committed, global power.17
If Lanyi's assertions are accurate, then historical and theoretical
analyses that concentrate primarily upon British domestic considerations
16See, for example, Henry Kissinger, Nuclear Weapons and Foreign
Policy (Garden City, 1957), Chapter IV, especially pp. 43-49, 58-64, and
John Spanier, World Politics in an Age of Revolution (New York, 1967),
Chapter III, especially pp. 69-76, 97-105. Two analysts who examine the
difficulties of drawing such lessons from history are Hans Morgenthau,
Politics Among Nations (New York, 1961), pp. 63-72, and John Herz, "The
Relevancy and Irrelevancy of Appeasement," Social Research, XXXI (1964),
Lanyi, op.cit., p. 20. (Underlining Lanyi's).
and Anglo-German relations are parochial oversimplifications and may
suggest misleading foreign policy prescriptions. A global examination
of British foreign policy, therefore, seems desirable.
International systems theory provides the conceptual framework
necessary to achieve a theoretical perspective that is global in scope.
Singer and Deutsch offer a systems theory that explains the intensity
of conflict behavior and its potential for escalation.8 Its principles
are more complex than'the ones that constitute balance of power theory,
although both theories focus upon features of the international situa-
tion and exclude domestic conditions. Figure 1 summarizes the systems
# of Independent Members: High
# of Interaction Opportunities: High
probability of Cross-Pressures: High
Relative Military Capabilities: Low
'-- Conflict Intensity: Low Escalation Potential: Low -'
Figure 1. A Model of International Conflict Behavior
Based Upon Systems Theory
The solid arrows (--) in Figure 1 indicate that if there are several
18Karl W. Deutsch and J. David Singer, "Multipolar Power Systems
and International Stability," World Politics, XVI (April, 1964), 390-407.
The version of Singer and Deutsch's model that follows does not conform
strictly to their original formulation, but its resemblance is close
enough for them to share the credit for its strengths; I am willing to
retain sole responsibility for its weaknesses.
members of the system and each of them interacts with the others, then
these structural characteristics are likely to generate simultaneous in-
volvement in several conflicts, creating cross-pressures and lowering
the level of military capabilities available for each conflict. These
conditions will tend to restrain conflict behavior, making the inter-
national system stable as the broken arrows (---4) in Figure 1 suggest.19
Stability, according to Singer and Deutsch, exists when the following
conditions are met:
From the broader, or systemic, point of view, we shall define
stability as the probability that the system retains all of its
essential characteristics; that no single nation becomes domi-
nant; that most of its members continue to survive; and that
large-scale war does not occur. And from the more limited
perspective of the individual nations, stability would refer
to the probability of their continued political independence
and territorial integrity without any significant probability
of becoming engaged in a "war for survival."
The two major criteria in this definition of stability are the inten-
sity of conflict and the continued existence of the members. The model
in Figure 1 states that these two characteristics are interdependent:
the conditions created by a high number of members restrain the level of
conflict; so long as international conflict remains limited the continued
existence of the members is likely. These hypothesized relationships,
in turn, correspond to what Kaplan calls, "A brief and nontechnical
An important underlying assumption of the model is that the inter-
acting independent members are roughly equal in military capabilities.
If this assumption does not hold, it is possible to imagine situations
in which a strong member simultaneously confronts another strong nation
and a weak member, or two weak members. Under these conditions cross-
pressures on the strong member might be weak or non-existent, and the
military capabilities may remain sufficiently high for military be-
havior toward both opponents.
Singer and Deutsch, op.cit., pp. 390-391.
description of the objectives of systems analysis...the study of a set
of interrelated variables...and of the ways in which this set is main-
tained....This definition emphasizes the articulation of the system and
of its components and the behaviors by means of which it maintains it-
self over time."21
The proposition that the over-all stability of the international
system is a function of its structure lies beyond the scope of inquiry
for this study. However, the hypothesis that the system's structure
influences the conflict behavior of its members does appear to be
relevant. The British were members of a large set of independent nations
and had a high number of interaction opportunities. They were simul-
taneously involved in Europe, the Mediterranean, and the Far East where
they interacted with several major powers, France, Germany, Italy, Russia,
Japan, and the United States, plus the less powerful nations in each area.
Among the Great Powers Germany, Italy, and Japan were the principal tar-
gets of British conflict behavior between 1931 and 1941. To what extent
did the structural features of the international system and Britain's
position in the system influence British conflict behavior toward these
three nations? If Britain's behavior corresponds to the systems theory
in Figure 1, then British policy makers (a) should very frequently have
experienced cross-pressures; (b) the existence of cross-pressures should
have co-occurred with low military capabilities; (c) these conditions
should have correlated with low levels of conflict behavior and incre-
mental changes in conflict behavior over time.
21Morton Kaplan, "Systems Theory," in James C. Charlesworth (ed.),
Contemporary Political Analysis (New York, 1967), p. 150. Underlining
The systems theory above is more elaborate than classical balance
of power theory, but it may still be too simple as a valid empirical
explanation of British foreign policy. Individual decision makers are
often explicitly goal-oriented in their behavior and selective in their
perceptions. To protect or achieve high-priority goals they may inten-
sify their conflict behavior and ignore the existence of cross-pressures
and low military capabilities. As Figure 2 illustrates, these latter
variables plus the diplomatic stakes may each independently affect
British conflict behavior.
Cross-Pressures: Yes Diplomatic Stakes: Low Relative Capabilities: Low
Conflict Behavior: Low -
Intensity and Small Change
Figure 2. A Model of Conflict Behavior Based
upon Decision-Making Theories
There is ample justification for advancing these bivariatc relation-
ships as mini-theories of conflict behavior. Each one reflects a promi-
nent theme in the literature on decision making. The cross-pressures hy-
pothesis appears as a theory in the literature on voting behavior, interest
groups, and international politics.22 Its most general formulation states
that conflicting influences upon a decision maker may be a function of the
differing positions that he takes on various issues or flow from the
different social strata and social groups to which he belongs. Whether
22Angus Campbell, Gerald Gurin, and Warren E. Miller, The Voter
Decides (Evanston, 1954), pp. 157-164; Seymour M. Lipset, Political
Man (Garden City, 1963), pp. 211-226; David B. Truman, The Governmental
Process (New York, 1951), pp. 157-167, 332-343; Frederick H. Hartmann,
The Relations of Nations (New York, 1967), pp. 5-10.
or not he is directly conscious of the conflict, "The essential element
is the pull from two sides, felt or latent." In all formulations of
the theory, the "conflicting pull" aspect is related causally to the
dependent variable, the decision. Research based upon the cross-
pressures model shows that the decision maker under cross-pressures
limits his behavior to low levels of activity until or unless a third
force resolves the restraints upon his behavior by either permitting
or compelling him to increase the intensity of his activity.
The diplomatic stakes hypothesis stems from the national interests
theory, which distinguishes between two types of foreign policy goals:
vital interests and secondary interests. Vital interests are those
goals that directly affect the nation's survival, while secondary inter-
ests are goals not so important to national survival. Critics of the
national interests theory argue that this distinction is very vague and
therefore of limited use for empirical theory-construction. The principal
justification for distinguishing types of interests is that a nation in a
conflict situation will presumably fight for vital interests but not for
secondary interests. National interests theorists often refer to the
This general formulation is from Bernard Berelson and Gary A.
Steiner, Human Behavior: An Inventory of Scientific Findings (New York,
1964); for an inventory of research findings from different disciplines
using this model see pp. 425-426, 434, 557-558, 575-584, 620 of their book.
Defenders of the national interest concept include Hans Morgenthau,
Politics Among Nations, Third Edition (New York, 1961), pp. 8-9, 562;
Hartmann, op.cit., Chapters 1, 4, 13; Morton Kaplan, System and Process
in International Politics (New York, 1957), Chapter 8. Its critics in-
clude Robert Tucker, "Professor Morgenthau's Theory of Political Realism,"
American Political Science Review, XLVI (March, 1952), 214-224; Tucker's
review of Hartmann's book, "The Study of International Politics," World
Politics, X (July, 1958), 699-747; Arnold Wolfers, Discord and Collabora-
tion (Baltimore, 1962), Chapter 10. A recent attempt to use the national
interests concept for systematic analysis is Thomas Robinson, "A National
Interests Analysis of Sino-Soviet Relations," International Studies
Quarterly, XI (June, 1967), 135-175.
territorial integrity and independence of the nation as vital interests.25
Beyond these examples the consensus regarding what constitutes vital
national interests weakens among foreign policy analysts. However, Arnold
Wolfers has developed a typology of foreign policy goals with sufficient
empirical content to distinguish between vital and secondary interests.
One can distinguish goals pertaining, respectively, to national
possessions and to the shape of the environment in which the
nation operates. I call the former "possession goals," the
latter "Milieu goals." In directing its foreign policy toward
the attainment of its possession goals, a nation is aiming at
the enhancement or the preservation of one or more things to
which it attaches value. The aim may apply to such values as
a stretch of territory, membership in the Security Council of
the United Nations, or tariff preferences. Here a nation finds
itself competing with others for a share in values of limited
supply; it is demanding that its share be left intact or in-
creased....Milieu goals are of a different character. Nations
pursuing them are not out to defend or increase possessions
they hold to the exclusion of others but aim instead at shaping
conditions beyond their boundaries.....Milieu goals often may
turn out to-be nothing but a means or a way station toward some
The equation of possession goals with vital interests and milieu goals
with secondary interests reduces the ambiguity of the national interests
theory and poses the hypothesis that the British are more likely to fight
for their own possessions than someone else's.
Foreign policy theorists continue to debate the importance of nation-
al power as an explanatory variable for international conflict. The
debates are multidimensional in that they deal with power as a goal of
policy, power as an instrument of policy, the distribution of power as a
25Hartmann, ibid., p. 76, Morgenthau, ibid., p. 562, and Wolfers,
ibid., pp. 154-155. The terms territorial integrity and independence
connote the preservation of the nation's cultural identity, or "way of
life," as well as the retention of its physical and political identity,
i.e., its boundaries and a government.
26Wolfers, ibid., pp. 73-74.
problem for policy makers, and the ambiguity of power--both as a term
and as a measurable phenomenon. Nuclear weapons strategists, for ex-
ample, focus upon one image of power, military capabilities. They con-
struct military strategies based upon the concept of deterrence, empha-
sizing the use of military capabilities to deter an opponent from attack-
ing in a conflict situation.28 This emphasis is not new; balance of
power theorists habitually note this feature in their analyses of inter-
national politics and foreign policy. This common theme from the classi-
cal balance of power literature and the contemporary works on nuclear
weapons suggests the power politics hypothesis that decision makers are
more likely to fight if their relative military capabilities are high
rather than low.
The cross-pressures, national interests, and power politics theories
imply a fourth theory, the cumulative effects principle. It attempts to
explain conflict behavior for international situations where the first
three theories either completely or partly reinforce one another as, for
example, when cross-pressures exist and diplomatic stakes and relative
capabilities are low, or when diplomatic stakes are high but cross-
pressures exist and military capabilities are low. The latter situation
2See Claude, op.cit., and Haas, op.cit., for the use of power in
the balance of power literature. Two interesting attempts to concep-
tualize and measure power, respectively, are Charles McClelland, Theory
and the International System (New York, 1966), Chapter 3, and David
Wilkinson, Comparative Foreign Relations: Framework and Methods (Belmont,
1969), Chapter 3.
2The literature on nuclear strategy is vast. The four books that
have probably had the most impact upon academic thinking are Herman Kahn,
On Thermonuclear War (Princeton, 1961); Henry Kissinger, Nuclear Weapons
and Foreign Policy (Garden City, 1958); Bernard Brodie, Strategy in the
Missile Age (Princeton, 1959); and Thomas Schelling, The Strategy of
Conflict (New York, 1963).
is more complex than the former, in which all three decision-making
theories hypothesize the same behavior. The cumulative effects theory
assumes that the relative potencies of its component principles are
equal and hypothesizes that the greater the number of international
restraints, the more likely is low conflict behavior. Consequently,
if two (or all three) decision-making theories predict low conflict
behavior, it will be more likely to occur than when two (or all three)
decision-making theories predict high conflict behavior.
Systems theory and the mini-theories of decision making offer
several explanations of British foreign policy decisions toward Germany,
Italy, and Japan between 1931 and 1941. A basic distinction between
systems theory and the mini-theories is that one of the latter, national
interests theory, incorporates goals as explanatory variables; systems
theory synthesizes the cross-pressures and power politics theories but
excludes goals as determinants of conflict behavior. Although these
theories include the same types of variables, they differ in the causal
relationships that they propose as links between the features of the
international situation and British conflict behavior. Systems theory
hypothesizes a developmental chain of bivariate relationships, while the
mini-theories postulate several bivariate relationships that show
different causal variables independently affecting a common dependent
variable. The actual relationships, if any, between these hypothetical
determinants and British policy may resemble either of these causal
patterns, or may approximate a hybrid version. Figure 3 illustrates the
hybrid possibility and reproduces the developmental and additive types
from Figures 1 and 2; the propositions that articulate the models also
appear in Figure 3.
The theories of conflict behavior in Figure 3 do not explicitly
identify the decision maker who is the subject of each theory. It is
possible to conceptualize the decision maker in each case as an indivi-
dual, a group, or a nation. This study focuses upon the individual as
the unit of analysis--a selection that implies a preferred solution to
the "level-of-analysis problem in international relations," raised by
J. David Singer:
In the vernacular of general systems theory, the observer is
always confronted with a system, its subsystems, and their
respective environments, and while he may choose as his system
(i.e., as his unit or level of analysis) any cluster of phe-
nomena from the most minute organism to the universe itself,
such choice cannot be merely a function of whim or caprice,
habit or familiarity....The responsible scholar must be pre-
pared to evaluate the relative utility--conceptual and method-
ological--of the various alternatives open to him, and to
appraise the manifold implications of the level of analysis
finally selected. So it is with international relations.
International conflict is a theoretical problem that is amenable
to analysis at several levels, but the results of conflict research based
upon aggregate units, especially the nation-state, have not been par-
ticularly productive.30 Conceptually, the individual appears to be a
logical and desirable alternative level of analysis, because the un-
satisfactory findings at other levels "can be interpreted as indicating
the potential relevance of psychological variables in the explanation
J. David Singer, "The Level-of-Analysis Problem in International
Relations," in Klaus Knorr and Sidney Verba (eds.), The International
System (Princeton, 1961), pp. 77-78.
3Richard A. Brody, "The Study of International Relations qua Science:
the Emphasis Upon Methods and Techniques," in Klaus Knorr and James Rosenau
(eds.), Contending Approaches to International Politics (Princeton, 1969),
S0 w wm i-w 0
_a 1-1 0 J OC ,-i
4-1 U co 0 a En i z
u) 0 W r= 4 0 COZ -
O 1 r- 4 0)
a, a) 41 Q L4
S03) 0 C
-- N E r1
U > ) 0 4 J Cn d r 0)
Pi Il >0 0 D C
e > o. ) E n E -4r. -t
C1) n 1-i CO >N a W W I w
0 0 r 0 w 0 0 4
-4 a. u c .
C* )O -
C I C/O0 C> l
$ Z a C
uI* a)0O u c
-1U 4 0.- u -ar) >1 o 0) 0
Z ca tli a)
0 4 C Ca) a) .
0)O *HU 0 *r.l C- 0 0.-i C !-i
c- Ur W:t C V 0 0
oa C: 4 4 a
\ 4- x 0 r 41 0 CO -4 n) 0 w
zI \ ( l >, -1 0) rc
n \a) a) -I E U 4 i- W u 0i)
w lM _- d C r4 C 0 O -4 (E-
4 01 C w E -4 co > i co 4-i
*l 4 0) t U ( 4 E C
-- I I0 m 11) C >, V C r-1 O r0 0 E
0n r4 a) no )i 0 4-Ci > 4-J (V *-i -4 1- c)o
(a) 40 o 1. .- -i O > .0 a >a E
14 C a I 1 0) W *- rC C 1 -
a Q. 4 C En 4J 4 4 4)r C W J 4 Co
ui C 0C u u O ca r u ) u '0 0
0 U 0H r O ) -4 --I r- 4J 3 E U C *
a U 4 > r Ca < o a) x 0 01
r -4 a -- 4-1 $4 t4- 4J U -H
PI > a U C 3C u a)
-i r a) C4 0 0 w E -- o 4 0 )
0 0 Q 4
co *o C
I 'l I IrCO 4J
Si U >n V)o 4J -4
co a) En O =U w
a) C 0 u E r U o C 3
a) I l r) > 0 0 4 U 4J r O
_3 r-1 0 0 -1 a) a L4J -4 Cal U
4 ) *- I 4 >- l C ., -I a C O U-4 .
n > Q n bz >z 0 >04 C) r
3 co 4.J 4 -4 H. 0 u 0 l U o IV O C
Scr aC aW o c u a u a)
AJ ) A a n ) a = 3 -r-l C _4 ll D
T> u w Q) E Cl. r-4 > r-4 (4 > G
E.1 >, = L Z a. 0 -4
4n 01 r r
1 0 4 C co
Ms, Q (1 c4 -) Id o
W ~ rr f^ J n r-1 tM
*-l 1 *
c W c 0
i r l
-H ca co
- 4-j 4
C nj S
of international conflict."31 Methodologically, the utility of the in-
dividual level of analysis depends upon the researcher's ability to
measure individual characteristics of theoretical interest. Because
there are several problems associated with this task, it is legitimate
to ask if it is worth the effort to solve them. By identifying such
problems and outlining solutions, the next chapter argues in favor of
the individual level of analysis for the British cases. Finally, in
accord with Singer's prescriptions for responsible scholarship, the
concluding chapter explicitly appraises the general utility of the in-
dividual level of analysis for the construction of international re-
THE MEASUREMENT OF BRITISH FOREIGN POLICY
In order to decide which propositions from systems theory and the
decision-making theories empirically explain British conflict behavior,
it is necessary to shift from the language of theory to the language of
research. The preceding discussion of systems theory and the decision-
making theories excluded criteria that indicated who were the British
decision makers, what constituted high and low levels of conflict be-
havior, and when British policy changed incrementally or radically. Nor
were the empirical referents for the values of the different independent
variables specified. Finally, the analytical procedures for fitting
these observations to the different models were omitted. All of these
problems need to be solved as part of the transition from the language
of theory to the language of research.
The sources of research data for this study are the perceptions of
British decision makers as reflected in government documents, personal
Hubert M. Blalock, Jr., Causal Inferences in Non-Experimental
Research (Chapel Hill, 1964), pp. 5-6. Blalock takes "the commonly ac-
cepted position that science contains two distinct languages or ways of
defining concepts...One thinks in terms of a theoretical language that
contains notions such as causes, forces, systems, and properties. But
one's tests are made in terms of covariations, operations, and pointer
readings.....There appears to be no purely logical way of bridging the gap
between these languages. Concepts in the one language are associated
with those in the other merely by convention or agreement among scien-
tists." Underlining Blalock's.
memoirs, and the historical works based upon them. The selection
of these data sources implies a choice between the individuals-as-
actors approach and the states-as-actors approach to foreign policy
analysis. As the principal advocates of the individuals-as-actors
approach point out:
It is...one of our basic choices to take as our prime analytical
objective the re-creation of the "world" of the decision makers
as they view it. The manner in which they define situations
becomes another way of saying how the state oriented to action
and why. This is a quite different approach from trying to
recreate the situation and interpretation of it objectively,
that is by the observer's judgment rather than that of the
The states-as-actors approach, on the other hand, assumes that a
decision maker's "definition of the situation" corresponds to its
objective structure. This assumption permits states-as-actors analysts
to focus upon the decision maker's environment as the data source and
avoid the problem of identifying the British decision maker for a
particular decision. Our selection of perceptions as a data source,
therefore, raises the problem of identifying the "real" British decision
makers in order to analyze the appropriate perceptions.
Identifying British Decision Makers
Decision making can be considered as a unified whole or separated
into its components and viewed as a process. The focus upon decision
making as a unified whole carries with it the risk of making certain
Richard Snyder, H.W. Bruck, and Burton Sapin (eds.), Foreign Policy
Decision-Making (New York, 1962), p. 65. Underlining Snyder's. Com-
parison of these two approaches is the subject of Wolfers, op.cit.,
3bid., pp. 90-91.
Whenever writers on international politics get down to dis-
cussing the behavior of decision-makers usually one5 of
five kinds of treatment results: (1) the same values and
perspectives are assigned to all officials; (2) motivation
is assumed to consist of a single drive; (3) the decision-
makers' actions are regarded as determined by "conditions"
and "resources"; (4) simple descriptions are made on a
very low level of generalization; and (5) diplomats are
often portrayed as isolated from any governmental organi-
Decision making viewed as a process, on the other hand, becomes a sequence
of activities in which many small decisions by several individuals result
in an organizational decision. From this perspective it is possible to
strip the decision of motives and view it as the output of a quasi-
mechanical organizational process or the outcome of a series of bargaining
moves by bureaucrats and politicians. In either case the decision is no
longer analyzed as a choice by an individual decision maker to protect or
achieve certain ends; it becomes instead a product of interactions by a
group, which may make it unintelligible from an ends/means perspective.6
The problems which these points raise for the investigation of
British foreign policy are important. If British decisions toward Germany,
Italy, and Japan are products rather than choices, then the measurement of
British decisions involves either collecting data on the perceptions of
several individuals in the British government7 or perhaps eliminating
4Ibid., p. 90,
5These treatments are not necessarily mutually exclusive.
Graham T. Allison, "Conceptual Models and the Cuban Missile Crisis,"
APSR, LXIII (September, 1969), 689-691.
7For a discussion of this option, see Ole R. Holsti, "Individual
Differences in 'Definition of the Situation'," Journal of Conflict
Resolution, XIV (September, 1970), 303-310.
perceptions as relevant data altogether. The communications patterns
within the British elite and the resources available to different
bureaucrats and politicians may better explain the pattern of British
decisions. Perhaps the "real" British decision makers are not govern-
ment officials at all, but an extra-governmental "power elite" who de-
cisively influence British foreign policy from their positions as
important British citizens.8
It is possible to resolve these problems by making two important
distinctions: distinctions among different phases of the decision-making
process and distinctions among different decision makers. Snyder suggests
that for any policy the "point of final decision" should be distinguished
from other phases of the policy process.
The point of final decision is that stage in the sequence
at which decision-makers having the authority choose a
specific course of action to be implemented and assume or
are assigned responsibility for it. At this point the
decision becomes official and thus binding on all decision-
makers whether they participated or not.9
The point of final decision usually follows "predecisional activities"
such as bargaining, "weeding out of information, condensation of memoranda...
The "power elite" thesis of C. Wright Mills is currently experiencing
a renassiance among students of American politics. See, for example, G.
William Domhoff, The Higher Circles (New York, 1970), Chapter 5, for an
analysis of "How the Power Elite Make (American) Foreign Policy." The
best discussion of the British foreign policy elite is D.C. Watt's essay
"The Nature of the Foreign Policy-Making Elite in Britain," Chapter 1 in
his book, Personalities and Politics (Notre Dame, 1965). During the early
1930's the "Cliveden Set," a group of upper-class British influentials,
did intermittently attempt to impose their views upon the British Govern-
ment through their spokesman, Thomas Jones, a close friend of Prime
Minister Stanley Baldwin. An account of their efforts to influence Brit-
ish police during the Rhineland crisis is in Thomas Jones, Diary With
Letters, 1931-1950 (New York, 1954), pp. 175-181.
9Snyder, op.cit., p. 91.
(that) all involve decisions which must be recognized as such by the
observer."10 Decisions involving more than a symbolic gesture of
assent or dissent also have an implementation phase that follows the
point of final decision and usually involves several decisions as well.
At each of these phases of the decision-making process the identity of
the decision makers may vary.
Between 1931 and 1941 the identity of the British decision maker
at the point of final decision depended upon the organizational processes
of the British elite and the relative authority of different participants
in the pre-decisional activities. During the inter-war period two com-
munications networks processed information and handled British foreign
policy decisions. The British Foreign Office processed most of the
information, and its officials made most of the decisions. Incoming
information to the Foreign Office fell into three categories:
(1) "Unimportant" messages for which a junior level member of a
regional department made the final decision, sometimes after
consultation with his immediate superiors or with officials
at his level in other ministries.
(2) "Important" messages which, accompanied by consultation with
other ministries, passed along (with memos) to higher levels
in the Foreign Office; these messages did not necessarily
11Lord Strang, The Foreign Office (London, 1955), pp. 154-156,
describes the formal communications network, as does Donald G. Bishop,
The Administration of British Foreign Relations (Binghamton, 1961),
pp. 108-109. Gordon A. Craig, "The British Foreign Office from Grey
to Austen Chamberlain," in Gordon A. Craig and Felix Gilbert (eds.), The
Diplomats, 1919-1939, I (New York, 1965), pp. 15-48, contains useful in-
formation about the origins of the informal network. Both networks are
elaborated in the following pages.
reach the Foreign Secretary before the point of final decision.
(3) "Very important" messages which either proceeded (with con-
sultations and memos) through channels to the Foreign Sec-
retary/Cabinet level or if "urgent" as well as "very impor-
tant," went directly to the Foreign Secretary's desk.
The great majority of British policy choices were responses to "un-
important" or "important" messages and reached the point of final
decision within the Foreign Office bureaucracy; the more multi-faceted
the decision in its implications, however, the more likely that horizontal
consultation occurred at various levels between Foreign Office officials
and other ministries during the pre-decisional activities.12
The other communications network for British foreign policy com-
bined the "very important" and "urgent" message channels of the Foreign
Office with semi-official correspondence, plus conference and summit
diplomacy. The face-to-face negotiations between chief executives at
the Versailles Peace Conference set a precedent for continued confer-
ence and summit diplomacy during the interwar period, supplemented by
semi-official correspondence between the chiefs and the appointment of
representatives to carry their messages back and forth. These procedures
often handled incoming messages from other capitals and transmitted
outgoing decisions by the British Prime Minister and the Foreign
12Strang, ibid., p. 156, estimates that 80 percent of the
Incoming messages are handled at the regional/functional department
level without reference to higher authority within the Foreign Office;
his estimate is not dated, but he served in the Foreign Office during
the 1930's and accompanied Chamberlain to Munich in 1938.
Traditionally, the authority to conduct foreign relations belonged
to the British monarch. With the evolution of parliamentary government
in Britain the responsibility passed to the Cabinet, which became
formally and collectively responsible for all Government policies.
Since this transfer of power, the Prime Minister as coordinator of all
Government policies and the Foreign Secretary as Chief minister of the
Foreign Office have exercised the most authority at the point of final
decision for very important foreign policy questions. Within the Cabinet
the relationships of authority and influence between the Prime Minister
and the Foreign Secretary have varied, ranging from a situation in which
one man held both offices simultaneously or where one's views dominated
the other's extensively, to situations where the two men either clashed
bitterly or worked harmoniously on roughly equal terms. In the final
analysis the Prime Minister can dominate the relationship if he chooses,
since he selects the Foreign Secretary and the other principal officials
in the Foreign Office and has access to all important Foreign Office
documents and dispatches.14
13One example of this communications network in operation is the Anglo-
American discussion between Chamberlain and Roosevelt in January, 1938, con-
cerning an American proposal for a conference in Washington among Germany,
France, Italy, Britain, and the United States to discuss current European
problems; another is the Anglo-Italian talks in February, 1938, concerning
British recognition of the Italian conquest of Abyssinia. In both cases
Prime Minister Chamberlain sent and received messages without telling his
Foreign Secretary, Anthony Eden. These tactics alienated Eden, who resign-
ed on February 20th. Full accounts of these incidents appear in John
Connell, The "Office" (New York, 1958), pp. 258-274, and Eden, Dictators,
op.cit., pp. 621-709. Chamberlain also used special emissaries to commu-
nicate with Hitler; his favorite choice was Sir Horace Wilson.
14Donald Bishop, The Administration of British Foreign Relations
(Binghampton, 1961), pp. 69ff. gives an overview of this evolution and
examples of different relationships between Prime Ministers and Foreign
Since the middle of the nineteenth century the career diplomat
appointed to the position of Permanent Under-Secretary in the Foreign
Office has customarily operated as the link between the Cabinet mem-
bers and the civil servants in the various sections of the Foreign
Office and Diplomatic Service.5 During and after World War I, how-
ever, the prestige and influence of the British diplomatic corps upon
very important foreign policy decisions declined for a variety of
reasons. This decline contrasted with an increase in the influence
of two other groups, the news media and the Prime Minister's "inner
circle" of personal friends inside and outside the government.1
The British patterns of communication and authority leading to the
point of final decision, therefore, vary from decision to decision, but
for very important and urgent decisions the perceptions of the Prime
Minister and the Foreign Secretary are the ones most likely to reflect
the international determinants of the policy choice. They occupy stra-
tegic positions at the top of the two communications networks and possess
the authority to make the final decision. Whether their perceptions are
the product of personal preconceptions, the influence of different ad-
visors, or a first-hand knowledge of the international situation is not
crucial. The decision maker's perception is an independent variable in
this research and not a dependent variable. The primary objectives of
15Zara Steiner, The Foreign Office and Foreign Policy, 1898-1914
(Cambridge, 1969), pp. 5-6.
16Discussion of these trends and their sources is in Gordon A.
Craig, "The British Foreign Office from Grey to Austen Chamberlain,"
Chapter 1 of Gordon A. Craig and Felix Gilbert (eds.), The Diplomats,
1919-1939, I (New York, 1965); Lord Vansittart, "The Decline of Diplo-
macy," Foreign Affairs, XXVIII (1950), 177-188; Harold Nicholson,
Diplomacy (New York, 1964), Chapters 3 and 4.
this analysis are to identify the decision maker at the point of final
decision and relate his policy choices to his perceptions. For very
important and urgent decisions the Prime Minister and the Foreign
Secretary are the key British decision makers.
Scaling the Dependent Variables
Since the theoretical orientation of this study is conflict be-
havior, very important and urgent British decisions in conflict situ-
ations are the ones that require systematic measurement. Conflicts
between Britain and one or more of the Axis nations, Germany, Italy,
and Japan, provide the British decisions for analysis. Their selection
reflects two methodological assumptions: that British behavior toward
this trio of nations generated a sufficient variety of decisions for
meaningful theoretical analysis; and that British involvement with
these countries in Europe, the Mediterranean, and the Far East repre-
sented the global scope of British foreign policy during the 1930's.
The list of conflict situations for this period includes the Manchurian
conflict (1931), the Italo-Abyssinian conflict (1934), the German re-
occupation of the Rhineland (1936), the Spanish Civil War (1936), the
Sino-Japanese War (1937), the Austrian Anschluss (1938), the Sudtenland
conflict (1938), the German intervention in Czechoslovakia (1939), the
Italian occupation of Albania (1939), the German invasion of Poland
(1939), and the Japanese encroachment in Indo-China (1940).
In these conflicts British decision makers could choose military
or non-military behavior to protect British diplomatic stakes; within
these two broad categories British policy could be essentially symbolic
or involve physical actions of one kind or another. These distinctions
combine to make a typology of four policy choices: symbolic non-military,
actual non-military, symbolic military, and actual military behavior.
Examples of behaviors that fall within each category include a diplo-
matic protest (symbolic non-military); withdrawing diplomatic or
economic relations (actual non-military); an ultimatum threatening the
use of force (symbolic military); and sustained use of military force
(actual military). Finally, it seems empirically justifiable to dis-
tinguish a fifth category, quasi-military behavior, such as a demon-
stration of military force, mobilization, or a border clash that is
not sustained; this type of behavior appears to differ significantly
from both the purely symbolic threat of force and its sustained, actual
use. In Figure 4 these five categories form a conflict behavior scale
to measure British involvement in a conflict situation.
SYM ACT SYM QUASI I ACT
Figure 4. Different Cutpoints for the Conflict Behavior Scale
That Measures British Involvement in Conflict Situations
This scale dichotomizes conflict behavior into high and low values.
There are several additional cutpoints in order to make any theoretical
generalizations from the scale as specific as possible. For example,
if British decision makers under cross-pressures do tend to select low
conflict policies, the additional cutpoints help determine more pre-
cisely in what sense the British select a low policy: a low (non-
military) policy, a low (symbolic non-military) policy, or a low
(actual non-military) policy. The conflict scale's additional cut-
points also measure the degree of change in British conflict behavior
over time. British policy either changes incrementally (low) or
radically (high), depending upon whether it shifts one position
(incremental change) or two positions (radical change) on the scale.
To operate the scale it is necessary to establish which obser-
vations of British behavior fall into the different positions. The
terminology in Figure 4 indicates the broad criteria for fitting
observations into scale categories. These criteria are crude in that
they do not permit systematic ranking of different British behaviors
within each category. For example, there appears to be a difference
of intensity between a carefully worded, polite expression of concern
and a bluntly worded grave expression of concern, but both diplomatic
notes fall into the symbolic non-military category. Furthermore, the
methodology of scaling can assume more complex forms; it is possible
to use scaling techniques that differentiate and rank such items.17
From this perspective the cutpoints question discussed above changes
from the problem of narrowing broad categories to the opposite one of
grouping data into broad categories.
The rationale for using broad categories rests upon both pragmatic
and theoretical considerations. First, as a practical matter a single
analyst can operate a broad scale without resorting to the validation
procedures necessary to operate a refined scale. The Q-sort and pair-
comparison scaling methods, for example, require panels of expert
judges to rank subtle types of behavior; the intersubjective consensus
of these panels determines the final ranking of the behaviors.18 Such
validation procedures are beyond the scope of the resources available
17Robert North, Content Analysis (Evanston, 1963), Chapters 4 and
5, discuss two such techniques, the Q-Sort and Pair Comparison methods.
for this study. Second, from a theoretical perspective there would
appear to be more doubt about the comparability of items from a re-
fined scale than from a crude one. The context in which an item
occurs as a historical act may vary and is not always identical to
the context that a panel of judges imagines in order to rank the item.
To the extent that this criticism is justified, the cruder, broader
scale is more likely to be valid for analyzing historical data.19
The broad criteria for scaling British conflict behavior, there-
fore, are as follows:*
s/n-m 1) Symbolic non-military behavior is a verbal
statement unaccompanied by higher activity,
proposed or actual, violent or non-violent.
a/n-m 2) Actual non-military behavior is an activity
that is more than verbal, either proposed
or actual, but which is non-violent.
s/m 3) Symbolic military behavior is a verbal state-
ment that proposes the use of military force.
q/m 4) Quasi-military behavior is the deployment of
Robert Jervis, "The Costs of the Quantitative Study of Inter-
national Relations," Chapter 10 in Klauss Knorr and James Rosenau (eds.),
Contending Approaches to International Politics (Princeton, 1969), dis-
cusses this problem with respect to North's research on the 1914 crisis;
North's rejoinder also appears in this volume. An extended discussion
of the theoretical and strategic implications of this question is the
concern of Jervis's own major work, The Logic of Images in International
Relations (Princeton, 1970), especially Chapters 2, 6 and 7.
20The military/non-military and symbolic/actual dichotomies below
are essentially identical to the violent/non-violent and bid/commission
dichotomies labeled by North as encompassing the "four broad categories
of action," which a decision maker may choose in pursuing goals and
which..."may also be viewed (sic)---and measured--by the investigator
as 'objective' behavior." The term objective refers to the high con-
sensus among foreign policy analysts and decision makers that these be-
haviors are easily distinguishable and have the same meaning to both
analysts and decision makers, i.e., they form a continuum of conflict
behavior according to the positions assigned to them in Figure 4.
North, Content Analysis, op.cit., Appendix A, pp. 152-154.
military forces without a complete commit-
ment to their use.
a/m 5) Actual military behavior is the sustained
use of military force.
Even the operation of such broad scale categories creates problems for
the analyst. Two that appear in this study are the problem of multi-
dimensionality and the problem of distinguishing symbolic from actual
The first problem, multidimensionality, arises because of the
variety of foreign policy options open to decision makers in conflict
situations. Decision makers can adopt different degrees of either
conciliatory or hostile behavior during a particular conflict, but the
conflict behavior scale above does not necessarily distinguish between
these two dimensions of conflict behavior. In the symbolic and actual
non-military categories the problem is most conspicuous. Here diplo-
matic protests (hostile behavior) and offers to negotiate (conciliatory
behavior) scale indiscriminately as symbolic non-military behavior,
while economic sanctions (hostile behavior) and efforts to mediate
(conciliatory behavior) both appear as actual non-military behavior.
It is permissible to lump such opposites only if the objectives
of the research make the distinction between them irrelevant or the
distinctions themselves appear to be intrinsically unimportant. The
latter contingency generally applies to the British cases, a situation
that also illustrates the scaling problem involving the comparability
of historical acts mentioned in an earlier paragraph. British offers
to negotiate with Germany, Italy, and Japan often occurred under duress,
or at least when the British thought themselves to be on the defensive;
their offers were not only conciliatory moves but also attempts to buy
time or minimize their losses. British decisions to mediate occurred
during conflicts between strong nations and weak nations, which meant
that British intervention defended the smaller nation from a settlement
dictated by the larger one--something that British decision makers
realized. Under these circumstances attempts to negotiate and mediate
are really subtle forms of opposition rather than conciliation. Under
other historical conditions, however, these same behaviors might be
The second problem differentiating between symbolic and actual
behavior, occurs at both ends of the scale. At the non-military end
the decision to mediate is not clearly different from a policy of
supporting negotiations that may be already taking place. In both
instances verbal forms of conflict behavior occur. However, in view
of the historical circumstances mentioned above in connection with
British decisions to mediate, such decisions appear to be similar to
actual non-military behavior in reflecting the level of British in-
volvement in a conflict. Consequently, British offers to mediate and
mediation itself both scale as actual non-military behavior in this
study. At the military end of the continuum it is questionable
2Treatment of these problems from a wide frame of reference, in
addition to the works by Jervis, op.cit., include Stanley Hoffmann,
"International Relations: The Long Road to Theory," World Politics,
XI (April, 1959), 345-377; Hayward Alker, Jr., "The Long Road to Inter-
national Relations Theory: Problems of Statistical Non-Additivity,"
World Politics, XVIII (July, 1966), 623-655; John Mueller (ed.),
Approaches to Measurement in International Relations (New York, 1969),
220n the other hand, British offers to negotiate and British
negotiations are not readily distinguishable from other symbolic be-
havior, since in this situation, Britain is directly involved in the
conflict itself rather than indirectly as a third party, as is the case
in mediation. Therefore, negotiations and offers to negotiate are both
scaled as symbolic non-military behaviors.
whether quasi-military acts are closer to symbolic or actual military
behavior. Do such acts manifest a real increase in the commitment to
fight or merely dramatize a commitment already made at the symbolic
level? The decision to deploy troops, planes, or ships appears to
alter significantly the conditions of the conflict situation and the
perceptions of the conflict by decision makers. The decision maker
who performs such an act generally looks upon it as a significant in-
crease in his level of involvement; analysts interpret it this way and
so do policy makers. Even if its intent is merely to dramatize a
previous verbal commitment, its impact seems to justify combining the
quasi-military and actual military categories for the statistical
analysis in later chapters.23
One other important methodological problem regarding the dependent
variable is how to decide when a new foreign policy decision occurs.
For example, does a new decision happen when a new decision maker re-
affirms an old policy? Does one occur when British policy continues
unchanged after another government changes its policy? The answers
to such questions are somewhat arbitrary and therefore should be made
explicit. A new foreign policy decision occurs regarding a given
diplomatic stake when one or more of the following events take place:
a) British conflict behavior changes from one scale type
b) Another government's policy changes scale types toward
23Support for this decision, based upon the folk axiom, "Actions
speak louder than words," appears in Thomas Schelling, The Strategy of
Conflict (New York, 1963), pp. 119-162; Alfred Vagts, Defense and
Diplomacy (New York, 1956), Chapters 7 and 10. Jervis, The Logic of
Images, op.cit., also offers qualified support for this position.
the British or toward other actors involved in the
conflict, and British behavior either changes or remains
c) The identity of the British key decision maker changes,
and British behavior either changes or remains the same.
d) The identity of a target's decision maker changes, and
British behavior either changes or remains the same.
e) If the British pursue a multiple policy toward the
same diplomatic stakes, such as employing economic
sanctions or mobilizing troops while simultaneously
negotiating, the policy choice that scales highest
is ordinarily considered to be the "real" British
Lastly, the relationship between British conflict behavior and
the British decision-making process needs elaboration. The decision to
adopt one of the five broad categories of conflict behavior or to change
from one type to another is almost certain to be a "very important"
decision, i.e., dne made by either the Prime Minister or the Foreign
Secretary. Some nuances in the formulation of the policy and its im-
plementation may not be the tasks of these officials, but the final
key decision to select, continue, or change one of these broad courses
of action falls within the realm of grand strategy and the establish-
ment of policy guidelines. In the British political system these
latter tasks are Cabinet-level functions, customarily performed by the
Prime Minister and the Foreign Secretary.25 Consequently, the explanation
24The most famous exception to this rule is the Hoare-Laval Plan,
which temporarily undermined the League of Nations policy of economic
sanctions toward Italy during the Italo-Abyssinian conflict.
25This generalization holds true for the 1930's. For a description
of Cabinet operations during this period, see Hans Daalder, Cabinet
Reform in Britain, 1914-1963 (Stanford, 1963), pp. 66-86; Sir Ivor
Jennings, Cabinet Government, Third Edition (Cambridge, 1959), pp.
for policy choices of this magnitude lies in an analysis of the per-
ceptions of these two officials.
Scaling the Independent Variables
The perceptions of the Prime Minister and the Foreign Secretary
contain the calculations that act as reasons for choosing a particular
foreign policy. Part II of this study presents the reasoning associated
with eighty-five British decisions in the 1930's and, for complexly
reasoned choices, ranks the reasons by their relative influence upon
policy selection. These reasons come from British documents and
private papers written at the time of the decision, and from memoirs
and secondary sources that have appeared since then. Where feasible,
extensive quotations that contain the evidence for judging the type of
reason and its rank appear in the text of Chapters III and IV. Whether
supported by quotes or not, the sources for such judgments are footnoted.
Finally, although the Prime Minister's and Foreign Secretary's calcu-
lations appear, the reasons offered by their supporters and opponents
inside and outside the government are ignored on the grounds that the
latter are not the "real" British decision makers.
Two steps are necessary to analyze the reasoning behind a par-
ticular foreign policy decision. First, the British diplomatic stakes
must be identified and coded as either milieu or possession goals.
Second, the primary reason or reasons associated with each decision
must be classified as one of the following types.
(cross-pressures) a) because of British involvement in
another conflict with a different
target, indicating cross-pressures
26International reasons only, and not domestic reasons, are ranked.
bilities ab) becuase of British military commit-
ments in another conflict with a
different target, which makes
British capabilities insufficient
(low) for exercising actual military
capabilities) b) because of British military capa-
bilities regarding this conflict or
another conflict with the same target;
the decision maker judges his military
capabilities as either sufficient (high)
or insufficient (low) for actual mili-
tary behavior without reference to con-
flicts with other nations.
A foreign policy decision's rationale contains cross-pressures if the
primary reasons for the policy include type a or if the primary reasons
include type ab; in the latter case cross-pressures and relative capa-
bilities combine to explain the decision. The relative capabilities
variable also influences a decision if a primary reason for the policy
is type b.
Figure 5 summarizes the observations necessary to scale a British
foreign policy decision, including the measurement of both dependent
and independent variables. The figure presents two hypothetical deci-
sions, X and Y. Decision X is an actual non-military behavior and an
incremental policy change (one scale type or less) from the preceding
decision regarding the milieu goals at stake. The rationale for the
decision reveals the presence of cross-pressures and indicates that
the decision maker's perception of.his relative capabilities is low.
Decision Y, on the other hand, is an actual military behavior and in-
volves a radical degree of policy change (two scale types or more) from
the preceding decision concerning the possession goals at stake. The
rationale for this decision indicates the absence of cross-pressures
and the existence of high relative military capabilities. Appendix I
records these observations for the eighty-five real British decisions
that constitute the cases for analysis. For some of these decisions
the policy maker did not calculate military capabilities as either
high or low. This absence suggests that two fundamental definitions
of the conflict situation exist among the cases, non-military and
military situations, distinguished by whether the decision maker con-
siders military capabilities as relevant or not.
It is possible to construct several other definitions of the situ-
ation from the independent variables in Figure 5, in addition to the
ones defined for hypothetical decisions X and Y. Decision X is an
example in which all the independent variables act as international
restraints, while Decision Y exemplifies a situation where none of them
restrain conflict behavior. Figure 6 contains the seven logically
possible definitions of the international situation, arranged according
to the type of situation and the number of international restraints.
The diagonal arrow represents the cumulative effects theory: as the
number of international restraints upon British conflict behavior
decreases in both non-military and military situations, the more likely
British conflict behavior will increase its intensity and degree
of change. Non-military situations are presumably less prone to high
conflict behavior than military situations. In the absence of military
considerations the maximum number of international restraints for non-
military situations is two (cross-pressures and milieu stakes); military
situations may have three (cross-pressures, milieu stakes, and low
Perception of the Situation
Non-Military Situation Military Situation
# of International
Restraints: Two Split None Three Majority Minority None
H s/m "
Figure 6. Seven Alternative Definitions of the Situation and the
Theoretical Direction of British Conflict Behavior
The cumulative effects hypothesis assumes that the potency of
each independent variable is equal; the number of international re-
straints rather than their identity influences British behavior. This
assumption is questionable. The causal relationships between conflict
behavior and cross-pressures, relative capabilities, and diplomatic
stakes, respectively, may vary in strength. Statistical tests of
assocaition can rank these relationships if they differ. Similarly,
the hypotheses from systems theory, which state a developmental causal
relationship for the independent variables, are susceptible to veri-
fication by statistical analysis. "Dummy variable" analysis, a form
of regression analysis applicable to dichotomous variables, can test
the direction and the strength of the causal hypotheses from systems
theory and the decision-making theories.27 Dummy-variable analysis
produces Pearson's r coefficients between pairs of variables, which
indicate the degree of association between them. The value of r
varies between -1 and +1. A high positive or a high negative value
indicates a strong relationship between the variables, while a low
positive or a low negative value indicates a weak relationship.
Two brief examples, based upon the cross-pressures and power
politics theories, illustrate how the Pearson's r statistic interprets
the possible relationships between pairs of variables. The power
politics hypothesis associates low conflict behavior with low capabili-
ties and high conflict behavior with high capabilities. A dummy vari-
able statistical analysis of this pair of dichotomous variables would
confirm the hypothesis if it produces a high positive Pearson's r. A
high negative Pearson's r, on the other hand, would suggest that the
counter-hypothesis is true, i.e., that low conflict behavior correlates
with high capabilities and high conflict behavior correlates with low
capabilities. Either a positive or a negative Pearson's r close to
27An exposition of the logic behind "dummy variable" analysis is
in Hayward Alker, Jr., Mathematics and Politics (New York, 1965), pp.
80-88; the technique of linear regression analysis is explained suc-
cinctly in William Buchanan, Understanding Political Variables (New
York, 1969), pp. 257-265, 275-277, 281-287, and extensively in Hubert
Blalock, Social Statistics (New York, 1960), pp. 273-301, 326-351.
28Dummy variable analysis generates Pearson's r's by assigning the
value of I to a case where an attribute is present and 0 to a case where
it is absent. Research examples that have used this technique for
analyzing dichotomous variables include Masakatsu Kato, "A Model of U.S.
Foreign Aid Allocation," in John E. Mueller (ed.), Approaches to
Measurement in International Relations (New York, 1969), pp. 198-216,
and Donald E. Stokes, "Some Dynamic Elements of the Contest for the
Presidency," APSLX (March, 1966), 19-28. According to Alker, ibid.,
p. 81, "Such a technique is as operational, reliable, and valid as the
observer making the prior qualitative assessments."
zero would indicate almost no relationship between conflict behavior
and military capabilities.
A comparison of the Pearson's r coefficient for the power politics
hypothesis with the r coefficient for the cross-pressures hypothesis
yields a preliminary evaluation of the relative potency of each
hypothesis. For instance, if the power politics r-value is +.71 and
the cross-pressures r-value is +.45, then there exists a stronger
relationship between relative capabilities and conflict behavior than
between cross-pressures and conflict behavior. Therefore, if one
assumes that each of the independent variables in the hypotheses from
decision-making theory independently affects British conflict behavior,
a comparison of Pearson's r's can indicate their relative potencies
in explaining British policy. But there are other possible relation-
ships among these variables. They may fit the systems theory hypotheses;
they may only appear to be causally related; or they may be related in
a fashion that is a hybrid of decision theory's additive causal model
and system theory's developmental causal model. Pearsonian correlation
analysis can also confirm or refute the existence of these more complex
possibilities, according to the rules of causal analysis summarized in
2These rules are adapted from Alker, "The Long Road to IR Theory,"
op.cit., pp. 645-653, and Alker, Mathematics and Politics, op.cit.,
pp. 119-126, and Blalock, Causal Inference, pp. 71-87. The use of
partial correlation coefficients is also sometimes necessary in causal
analysis; their use in Chapter V of this study is based upon Blalock,
Causal Inferences, op.cit., pp. 65-87, and Hubert Blalock, "Four-Vari-
able Causal Models and Partial Correlations," American Journal of
Sociology, LXVIII (September, 1962), 182-194.
Additive Model Developmental Model Spurious (No Cause) Model
If RC FP, then If CP RC FP, then: If CP i.e., RC FP, then:
CP rCPFP=(rCPRC)*(rRCFP). FP
rNICP=0 rNIFP= If RC CP FP, then: rRCFP(rCPFP) (rCPRC).
0 RCFP =0 rRCFP=(rRCCP)*(rCPFP). CP
NIRC RCFP If RC i.e., CP FP, then:
rCPRC=O rCPFP=0 F
rCPFP (rRCFP) (rRCCP)
Figure 7. Rules for Fitting British Policy Choices to the Different Causal
Models Suggested by Systems Theory and the Decision-Making Theories
*The Pearson's r-values for CPRC and RCCP are identical by
Figure 7 states that the relationships among r-values in a four-variable
situation vary predictably for different causal models. If the analyst
knows the Pearson's r coefficient for every pair of variables, he can
determine which causal model best fits the data by comparing the pre-
dieted values with the actual values.3
3For an example of this technique applied to international rela-
tions problems, see Alker, "The Long Road to IR theory," op.cit., p. 651-
653. Charles Cnudde and Donald J. McCrone "The Linkage Between Consti-
tuency Attitudes and Congressional Voting Behavior," APSR,LX (March, 1966),
66-72, also use this technique.
Two principal disadvantages accompany this type of analysis. The
first one is the necessity to collapse the categories of the dependent
variable (conflict behavior) into a dichotomy, which limits the gener-
alizability of the findings to distinctions between military and non-
military behavior. The other drawback is the difficulty in visualizing
the analysis itself, i.e., being able to see which British decisions
distribute where, in order to generate the Pearson's r-values that
summarize the analysis. To compensate for these limitations the chap-
ters in Part II and Part III adhere to the following format. Chapters
III and IV of Part II present a historical narrative of each decision
and the reasoning behind it, accompanied by tabular analysis as in
Figure 5, which should provide the reader with a sense of continuity
and an appreciation for the nuances surrounding each decision.
Part III analyzes the decisions collectively, but in a way that
permits the reader to identify individual decisions. The first section
of Chapter V gives an overview of the decisions. The second and third
sections separate the decisions for analysis by cross-tabulation of
2x2 and 2x4 tables, accompanied by commentary. These tables permit the
reader to inspect visually the distribution of individual decisions;
the commentary identifies the decisions by name and interprets the
tables. The final section of Chapter V condenses and refines these
findings by using dummy variable analysis and the rules of causal in-
terpretation. Chapter VI concludes the study by assessing its im-
plications for interpreting British foreign policy and building in-
ternational relations theory.
BRITISH CONFLICT BEHAVIOR, 1931-1936
Between 1931 and 1941 the British became involved in several
attempts to revive the status quo. From the beginning of this period
they had to focus their attention upon the demands of three revisionist
nations: Germany, Italy, and Japan. These next two chapters trace the
patterns of British involvement in the series of frontier disputes that
became the foci for revision. The analysis of British behavior in
these conflicts deals with each dispute separately, although there is
considerable overlap, chronologically, among them. This chapter ana-
lyzes British involvement between 1931 and 1936. The next chapter
covers the period 1937-1941.
Between 1931 and 1936 the principal British decision makers were
the Foreign Secretaries, Sir John Simon, Sir Samuel Hoare, and Sir
Anthony Eden. Of the three, Eden was the most independent in his
thinking about foreign affairs. Simon and Hoare appear to have relied
considerably upon the views of Robert Vansittart, the Permanent Under-
Secretary at the Foreign Office, an ironical situation considering
their later membership in Neville Chamberlain's Inner Cabinet of
Appeasers and Vansittart's anti-appeasement views. The two Prime
Ministers for these years, Ramsey MacDonald and Stanley Baldwin, did
The Marquess of Reading was Foreign Secretary from August to
November, 1931 and Arthur Henderson was his predecessor.
not initiate concrete policy and rarely vetoed the proposals of their
Foreign Secretaries. However, as the narrative in the next chapter
shows, the Prime Minister did initiate most of the key decisions after
Neville Chamberlain replaced Baldwin.
The Sino-Japanese Conflict
The first major challenge to the status quo came in the Far East.
Between September, 1931, and May, 1933, an "incident" between Chinese
and Japanese troops in the Chinese territory of Manchuria exploded into
a large-scale military conflict between China and Japan. It ended with
the establishment of an independent Manchuria (Manchukuo)--under Japa-
Britain's diplomatic stakes in the Manchurian dispute included the
following objectives. First, as a member of the League of Nations and
as a signatory of the Kellogg Pact and the Nine Power treaty, Britain
was committed in principle to collective security, the renunciation of
war as an instrument of national policy, and the territorial integrity
of China. Second, the British government became concerned with pro-
tecting British investments in Manchuria, especially the Peking-Mukden
Railway. Third, as a member of the International Settlement at Shanghai,
the British were involved in maintaining Shanghai's neutral status and
protecting the Settlement from damage by fighting Chinese and Japanese
troops. Together, these three areas of involvement outline the British
diplomatic stakes in the conflict between China and Japan. To protect
these stakes the British made several key decisions.
Baldwin's veto of the Hoare-Laval Pact in December, 1935, is
perhaps the outstanding example.
With respect to the diplomatic stakes of international peace and
China's territorial integrity, British policy initially operated at
the symbolic non-military level, changing to the actual non-military
level only in the Spring of 1933. From September, 1931, to January,
1932, the British government supported a series of League of Nations
resolutions which requested both China and Japan to adopt a cease fire
and make troop withdrawals to the status quo ante bellum. To comple-
ment the League's position the British government also made numerous
direct representations to the Japanese government to stop expanding
its military operations throughout Manchuria, and especially in the
A minute by Sir John Simon on December 25, 1931, regarding the
likely Japanese invasion of Chinchow,3 summarizes the international
restraints that conditioned British policy choices during this period.
I confess that I should have been glad if we had made a
representation about Chinchow rather earlier and more
firmly but that is not our Ambassador's fault as the in-
structions I had in mind were not sent. Ultimately, we
merely said ditto to the French. The important question
now is as to the outcome...I quite agree that good
relations with Japan are of the first order of requisite,
and must be safeguarded: but we must, consistently with
this, play our part as a member of the League, and use
such influence as we have.
The primary emphasis upon good relations with the Japanese reflects the
realization among British policy makers that Japan potentially threat-
ened the British position in Manchuria, Shanghai, and throughout the
Far East by virtue of her superior military capabilities and future
3The Japanese took Chinchow the following week.
4Her Majesty's Stationary Office, Documents On British Foreign
Policy (thereafter D.B.F.P.). Second Series, IX, #21, pp. 31, 33.
policy options. The secondary, although important, emphasis upon
British obligations to the League is best illustrated by two incidents.
The first one concerns the tone of the original British representa-
tions to the Japanese regarding the Manchurian conflict in September,
1931. The Marquess of Reading, who was the British Foreign Secretary
at the time, thought that the French government was going to make
"strong and unpalatable representations" to the Japanese government
and instructed the British Ambassador to Japan, F.O. Lindley, to
follow the French line. The British subsequently discovered that the
French had not -consented to jeopardize their relations with Japan by
such strong representations.
In the words of a British Foreign Office Memorandum which reviewed
His Majesty's Government, thus isolated, reaped the full
odium which the Japanese...would otherwise have distributed
among European nations generally, or at lease divided
between us and the French, and Sir F. Lindley's position
became painfully difficult.
When the Marquess of Reading discovered what had happened, he in-
structed Lindley to explain to the Japanese Minister of Foreign Affairs,
"the important fact that His Majesty's Government...acted...on the
understanding that other friendly Governments were acting similarly and
on the initiative of one of them."8 The rationale which lay behind
5Ibid. See also #238 and #239.
Sir John Simon replaced him on November 5, 1931.
D.B.F.P., Second Series, IX, #21. Cf. also Second Series, VIII,
#'s 520, 523, 538.
8D.B.F.P., Second Series, VIII, #538.
these maneuvers was that good relations with Japan could be maintained
while fulfilling League obligations, if the latter were fulfilled
jointly and therefore diffusely.
The second incident concerned British reactions between January
and March, 1932, to the Stimson doctrine of non-recognition of "any
situation, treaty, or agreement which may be brought about by means
contrary to the covenants and obligations of the Pact of Paris."
Stimson, the U.S. Secretary of State, attempted to persuade the Brit-
ish to support this non-recognition doctrine, which he intended as a
tacit warning to the Japanese. Sir John Simon responded by expressing
his desire to support the U.S., but only within the limits permitted
by British membership in the League. He wanted a British endorsement
of American policy to be supported by the other League members. The
League Assembly's Resolution of March 11, 1932, endorsing the principle
of non-recognition, met this requirement.
The League resolution did not name Japan as a violator and, when
the Japanese puppet state of Manchukuo asked for recognition the
following day, Simon expressed the carefully phrased British position
that it would be "premature" to recognize Manchukuo.0 The public
rationale behind these maneuvers concerning recognition was that they
were consistent with the League's policy of waiting for the Lytton
Commission's report in order not to prejudge the issues involved in
the Manchurian case. More importantly, British policy concerning
9Cited from official text in Dorothy Borg, The U.S. and the Far
Eastern Crisis of 1933-1938 (Cambridge, 1964), p. 9.
10Royal Institute of International Affairs, Survey of International
Affairs (hereafter cited as Survey), 1932, pp. 552-554.
recognition was compatible with the private rationale, which Simon had
expressed earlier, of not offending the Japanese.
The League Assembly finally met in late November, 1932, to con-
sider the Lytton Commission's report. After several months of debate
the League Assembly voted on March 24, 1933, to endorse the Lytton
Commission's conclusions that Japan was violating Chinese territorial
integrity and that the Chinese were not to blame. Britain voted with
the majority. On the same day the Japanese delegate stated that his
government disagreed with the League's decision and announced the
Japanese intention to withdraw from the League.11 The League's March
24th resolution did not impose sanctions upon the Japanese; it merely
instructed the Japanese government to take conciliatory measures to
resolve their dispute with the Chinese. As a symbolic non-military
policy it did not greatly risk a military response from Japan that
would threaten other British interests. The British government could
support it without abandoning their original strategy of minimizing
Japanese hostility toward Britain by diffusing it among the other
Three days later British policy shifted from the symbolic non-
military level to the actual non-military level, as the British
government imposed an arms embargo against both Japan and China. Simon
explained to the House of Commons that the policy depended upon the
support of other nations to be effective. The decision to include
China in the embargo would prevent potential collisions on the high
seas between British ships attempting to supply China and Japanese
11Ibid., 1933, pp. 504-509. The Lytton Commission was a League body
set up to investigate the Manchurian conflict and report to the League.
warships. The embargo was also subject to the maintenance of existing
contracts and the opportunity for international consultation about a
joint embargo.12 Approximately two weeks later the British abandoned
the arms embargo policy, due to lack of support by other nations during
the subsequent international consultations at Geneva.13 During its
short tenure the arms embargo policy was subject to the same inter-
national restraints as the previous British policy; these were Brit-
ish relative military capabilities and the future policy options of
the Japanese, plus the policy.choices of the other League members and
the United States. Simon's public rationale of requiring support by
other nations to make an effective embargo was consistent with the
private "diffusion strategy" regarding Japanese hostility. The in-
clusion of China in the embargo and the commitment to honor existing
contracts also minimized Japanese hostility.
Fighting between Japanese and Chinese troops continued until May
31, 1933, when the two disputants signed the Tangku truce. By this
time the Japanese controlled all of Manchuria, part of Inner Mongolia
(Jehol), and the area just inside the Great Wall above Peiping. During
this period the British government operated at the actual non-military
level with a policy of not recognizing Manchukuo. The Tangku truce
stabilized the frontiers between China and the Japanese sphere of
influence in Manchuria and northern China until 1937.14
121bid., pp. 512-514.
14Ibid., pp. 514 ff.
INTERNATIONAL RESTRAINTS AND BRITISH FOREIGN POLICY TOWARD JAPAN
DURING THE MANCHURIAN CONFLICT
Date Anal. Empir.
9/31 s/n-m Protest
3/32 s/n-m Premature
3/33 s/n-m League
3/33 a/n-m Arms
4/33 a/n-m Non-recogni-
Milieu China Terr.
Milieu China Terr.
Milieu China Terr.
Milieu China Terr.
SMilieu China Terr.
Cross-Press. Rel. Cap's.
The Si.no-Japanese dispute over Manchuria also potentially threat-
ened Britain's economic position in China. During the course of the
conflict the British government pursued a policy primarily designed to
protect this position. Soon after the fighting began to spread
throughout Manchuria the British government reminded the Japanese
government of British investments in that area and received assurances
from the Japanese that they would respect them. Throughout the con-
flict British policy regarding these diplomatic stakes in Manchuria
operated at the symbolic non-military level. The international
restraints which influenced this policy choice were primarily the
Japanese policy of respecting British interests and secondarily the
superior military capabilities of the Japanese and their future
Although the Sino-Japanese conflict over Manchuria never
seriously threatened British interests there, the position of the
International Settlement at Shanghai did become precarious. In re-
sponse to Japanese military activity in Manchuria the Chinese popula-
tion in Shanghai organized an economic boycott against Japanese goods
and services. By January, 1932, the resulting tension between Chinese
and Japanese in Shanghai had led to a series of violent clashes between
Chinese and Japanese civilians. After the deaths of several Japanese
Buddhist priests in one such incident the Japanese demanded that all
anti-Japanese movements in Shanghai be suppressed. On January 28th,
the Japanese government sent a note to the British government, which
stated that grave measures might be necessary to check anti-Japanese
movements in Shanghai.16
The British government responded on the same day at the symbolic
non-military level. Sir John Simon personally expressed his "grave
concern" to the Japanese Ambassador in London that such grave measures
might be necessary, and instructed the British Ambassador in Tokyo to
tell the Japanese government that Britain could not approve of the
International Settlement's use for other than defensive purposes. How-
ever, in the late afternoon of January 28th the local Japanese commander
15D.B.F.P., Second Series, VII and IX, passim.
16D.B.F.P., Second Series, IX, and Survey, 1932, pp. 498 ff.
demanded that the Chinese remove their troops from the Chapei section
of Shanghai. During the night of January 28-29th, Japanese marines
landed at Shanghai from warships sent earlier from Japan.17 By the
next morning Chinese and Japanese troops were fighting in Chapei.
The British response to these events was a two-level policy. At
the symbolic non-military level the British and American Ambassadors
in Tokyo made repeated demarches to the Japanese government concerning
the neutral status of the International Settlement. At Geneva the
British supported League resolutions asking for the creation of a
neutral zone and a ceasefire while in Shanghai the British Consul
worked to obtain a ceasefire. To protect British lives and property
in case the fighting spilled over into the International Settlement
from Chapei, British decision makers also adopted a policy at the
quasi-military level. Simon ordered several warships with artillery
and infantry aboard to Shanghai between February 1-5th, including H.M.S.
Kent, the flagship of Admiral Sir Howard Kelly.18
On the evening of January 29th, a cease fire began as the result
of British mediation, but the Japanese ended it on February 2nd.
Thereafter, British policy during the Shanghai fighting continued to
operate at the actual non-military level with respect to the Settlement's
neutral status, while British warships (and those of other Great Powers)
continued to stand by in case the fighting spread into the Settlement.
Continued attempts by British diplomats during February to mediate the
17Between January 21-26th, Japan sent one aircraft carrier, two
cruisers, and sixteen destroyers to Shanghai. Survey, 1932, pp. 472-480.
Simon':s instructions to his Tokyo Ambassador appear in D.B.F.P., Second
Series, IX, #195.
18D.B.F.P., Second Series, IX, #'s 114, 129, 130, 154, 156, 195, 200.
dispute failed, and a fresh Japanese military offensive on March 1st
pushed the Chinese troops out of the immediate vicinity of Shanghai.
After this successful Japanese move formal negotiations began on
March 9th between China and Japan with the British acting as mediators.
On May 5th the two disputants signed an armistice agreement based upon
a British compromise formula.19
During the Shanghai crisis the international restraints upon
British policy choices were once again Japanese military superiority,
future Japanese policy options, and the policy choices of other in-
terested nations, particularly the United States. The relative
potency of these restraints varied somewhat, although not critically,
according to the diplomatic stakes involved. With respect to main-
taining the International Settlement's neutral status, the British
wanted to stop the Japanese from using it as a base for launching
attacks against the Chinese and landing reinforcements for Japanese
units operating in the Shanghai vicinity. To accomplish these objectives
the British initially pursued a symbolic non-military policy in order
to minimize conflict with Japan. U.S. policy preferences were secondary.
This rationale governed Sir John Simon's thinking early in the crisis,
as the following quote from Simon's letter of January 29th to his P.M.,
Ramsey MacDonald, shows:21
19Survey, 1932, pp. 500-514.
201bid., and D.B.F.P., Second Series, IX, #153 and #238.
2D.B.F.P., Second Series, IX, #153. *British Ambassador to the
U.S. **Italics mine. ***Italics Simon's.
Stimson sent a message urging that we should at once join the
Americans in formal and categorical representations to Japan
that the International Settlement at Shanghai was sacrosanct,
and that we should take the gravest view of its being the
scene or source of violent conflict. I suggested to Lindsay*
a variant in Stimson's formula (the object being to avoid
offending the Japanese too much)**...we should jointly***
address both** China and Japan, urging China to comply fully
with Japanese reasonable demands, and to avoid action which
would lead to trouble, and at the same time urging Japan to
remember that the interests of foreign powers in the Inter-
national Settlement made it right for us to advise restraint
The Japanese attack on Chapei from their sector of the International
Settlement slightly altered the relationship among these influences.
The policy of the United States now assumed greater, although not
equal, consideration with Japanese policy and military capabilities.
Simon instructed the British Ambassador in Washington to ask the
Americans to protest to the Japanese about their attack on Chapei, as
the British were doing, and also instructed him to inform the U.S.
government that H.M.S. Kent was proceeding to Shanghai.23 The diplo-
matic stakes now included British lives and property as well as the
neutral status of the Settlement. A minute by Vansittart, addressed
to Simon on February 1st and initialed by him on February 3rd, states
the rationale behind these British moves to protect their diplomatic
2.B.F.P., Second Series, IX, #'s 154, 239.
2Stimson's note of January 25th, referred to by Simon in his letter
to MacDonald, had suggested sending reinforcements to Shanghai as well as
making a categoric statement to Japan that the International Settlement
was sacrosanct. Vansittart suggested including this message in the in-
structions to Lindsay in Washington. D.B.F.P., Second Series, IX, #151.
24Ibid., #238, footnote 2; see also #153, footnote 12.
(1) If Japan continues unchecked and increasingly, as she
indeed seems bent on doing, our position and vast interests
in the Far East will never recover. This may well spread
to the Middle East. The Japanese victory in 1904 was the
beginning of the trouble there.
(2) We are incapable of checking Japan in any way if she
really means business and has sized us up, as she certainly
(3) Therefore we must eventually be done for in the Far
(4) The United States are eventually prepared to use force.
(5) It is universally assumed here that the U.S. will
never use force.
(6) I do not agree that this is necessarily so. The same
was said of the U.S. in the Great War. Eventually she was
kicked in by the Germans. The Japanese may end by kicking
in the U.S. too, if they go on long enough kicking as they
(7) The Japanese are more afraid of the U.S. than of us,
and for obvious reasons. At present, however, they share
our low view of American fighting spirit.
(8) By ourselves we must eventually swallow any and every
humiliation in the Far East. If there is some limit to
American submissiveness, this is not necessarily so.
(9) We can therefore frame no policy and face no future till
we are sure on this all-important point. To assume that
there is no limit is a counsel of despair.
(10) We must let the provocation proceed further than at
present. At some point, however, we shall,....have to know
where we stand on this vital question. When that moment
comes, it will be impossible (sic) to make sure either by
telephone or telegram. The moment, however, has of course
not yet come.
(11) If and when this sounding has to be taken, there will
probably be a lull in Japanese aggression till we and they
know the answer. F.V., Feb. 1.
To Secretary of State
I think there is an universal tendency to go to great
lengths of (5) in my annexed minute. I suggest that you
should consider it in connection with Sir J. Pratt's memo-
randum, and put the logical sequence to your colleagues.
Till this sequence has been faced (see (9), in my minute)
we can have no longrange, or even shortrange, policy in
the Far East. We must live from hand to mouth--an humili-
ating process--unless we have made up, or cleared, our
minds upon the answer to (6). R.V., Feb. 1. J.S., Feb. 3.
Vansittart's minute reveals that British diplomatic protests were
based partly upon present Japanese policy, which did not yet directly
endanger British lives and property but did violate the Settlement's
TABLE 2.. INTERNATIONAL R
TOWARD JAPAN DI
Date Anal. Empir.
10/31 s/n-m Remind Japan
1/32 s/n-m Grave note
1/32 s/n-m Call for
1/32 a/n-m Offer to
2/32 q/m Send ships to
3/32 a/n-m Mediate for
neutral status. The British naval reinforcements anticipated Japan's
future policy options and reflected the British inability to do any-
thing else due to their own lack of military capabilities and no
present American support for a military policy.
The situation in Shanghai did not worsen sufficiently during
February to follow Vansittart's scenario and, after the Japanese
military offensive of March 1st removed the fighting from Shanghai,
RESTRAINTS AND BRITISH FOREIGN POLICY
JRING THE SHANGHAI CRISIS
Stakes Primary Rea
Anal. Empir. Cross-Press. Rel
Poss. Br. eco.
Poss. Br. Poss.
Poss. Br. poss.
British policy operated at the actual non-military level to negotiate
a truce. The policy choices of the Chinese and Japanese, expressed
as preferred terms for the ceasefire, acted as international restraints
upon the British mediator. The Japanese had occupied Chapei and would
not leave without a guarantee against resumption of the boycott and
other anti-Japanese movements by Chinese in Shanghai. The Chinese
maintained that they could not provide such a guarantee so long as
Japan occupied Chapei. The British mediator, Sir Miles Lampson, even-
tually formulated a troop-withdrawal scheme acceptable to both sides,
and the conflict ended.
The Italo-Abyssinian Conflict
The second dramatic challenge to British policy makers in the
1930's came from Italy. A border clash between Italian and Abyssinian
forces at Walwal in December, 1934, rapidly brought relations between
the two nations towards the threshold of war. The British diplomatic
stakes in this conflict indirectly consisted of local British imperial
interests. The upper waters of the Nile, upon which Egypt depended for
its existence, originated in Abyssinian territory at Lake Tsana, and
three British possessions, Egypt, Kenya, and British Somaliland, shared
frontiers with Abyssinia.26 Later, Abyssinia's territorial integrity
and the protection of British possessions in the Mediterranean from
Italian attack became the diplomatic stakes.
British policy, under the direction of Sir John Simon as Foreign
25Survey, 1932, pp. 510-513.
26Somali and Abyssinian tribesmen shared grazing privileges
across ill-defined frontiers in these areas.
Secretary, proceeded at the symbolic non-military level.27 In January,
1935, Simon instructed the British Ambassador at Rome, Sir Eric
Drummond, to try and persuade Mussolini of the advantages of negoti-
ating a peaceful solution without intervention by the League of Nations.2
Mussolini subsequently agreed to direct negotiations with the Abyssinians
at Addis Ababa. However, in February the Italian government mobilized
two divisions of the Italian army. This move plus the Italian refusal
to submit the dispute to arbitration led the Abyssinian government to
appeal to the League of Nations for investigation and consideration of
the conflict at the next ordinary session of the League Council in May.29
Between February and May, 1935, the British government continued
its attempts to persuade Mussolini to settle the dispute amicably. At
the suggestion of Sir Anthony Eden, the British representative at the
League, and with the approval of Simon, the British warned the Italians
that a dismemberment of Abyssinia might seriously disturb Anglo-
Italian harmony.0 On the eve of the Stresa Conference in April the
The fullest accounts of the early phases of this conflict are in
Survey, 1936, II, pp. 133-136; and in Anthony Eden's memoirs, Facing the
Dictators, op.cit., pp. 213-241. Sir John Simon's memoirs, Retrospect
(London, 1952) deal only with those phases of the dispute when he was
no longer Foreign Secretary. British Foreign Office documents cover-
ing this conflict are not yet available.
28The Abyssinian Emperor, Haile Selassie, was attempting to put
the dispute on the League Council's agenda.
29Cf. Survey, 1936, II, pp. 143-212. The Abyssinians attempted
to submit it in March at the Council's extraordinary session, but were
3Drummond, the British Ambassador, delivered this message in Rome;
at London Vansittart presented it forcefully to Grandi, the Italian
Ambassador during March, 1935. Eden, op.cit., pp. 222-224. The British
also suggested to the French that they send the same message, which they
Italian government announced that it would agree to make the necessary
arrangements for arbitration. But early in the following May, Mussolini
sent a message to the British through his ambassador appealing for a
friendly and helpful British attitude toward Italy's activities in
Abyssinia. Grandi, the Italian Ambassador, spoke to Simon of the
situation there "as a cancer which had to be cut out."
Simon interpreted Grandi's representations as conveying "in
veiled though unmistakable terms" that Mussolini was "contemplating
a forward policy of the most serious dimensions."3 Simon replied by
suggesting that prolonged Italian military operations in Africa might
weaken the front against Germany in Europe, which the Stresa Conference
had established. He also expressed concern about the effects of Italian
policy upon British parliamentary and public opinion. Grandi then
suggested that local British interests in Abyssinia would not necessar-
ily be endangered, but Simon answered that British concern over
Abyssinia was more fundamental.33
The meeting between Simon and Grandi in early May prompted two
British Cabinet meetings to consider British policy at the upcoming
session of the League Council, when the Abyssinians would present their
appeal. In a Foreign Office memo Simon stated his conviction that, "if
matters continue as they were, Italy would launch a large-scale offen-
sive when the rainy season was over in Abyssinia, at the end of September
or beginning of October." But he did not accompany this forecast with a
3Eden, ibid., p. 226.
3Eden, cites Simon in these terms, ibid., p. 226.
33Ibid., pp. 227-228.
policy recommendation.34 The outcome of these meetings was the
decision to remind Mussolini of how intensely the British felt about
the peaceful settlement of disputes through the League and ask that
Italian delegates at Geneva be instructed to discuss the best method
for securing a solution under the League's auspices. At Geneva the
British and French governments proposed that the League Council pass
two resolutions recommending arbitration of the dispute under the
Italo-Abyssinian Treaty of 1928 and stating that the Council itself
would consider the conflict if it were not resolved by August 25, 1935.36
Up to this point, at the end of May, 1935, British policy clearly
operated at the symbolic non-military level. Simon's conversations
with Grandi indicated the international restraints which conditioned
British policy. Simon's desire to retain Italian cooperation against
Germany and avoid a weakening of Italian military capabilities by
action in Africa reflected the influence of Italy's future policy
choices regarding European issues and British relative military capa-
abilities available for action on these issues. All of the British
diplomatic stakes in the Italo-Abyssinian conflict were not yet clearly
defined. Grandi's remarks indicated that local British interests were
not likely to be endangered.38 The "more fundamental" concern to which
3Ibid., p. 229.
35Ibid., p. 230.
Survey, 1936, II, pp. 153-154; Eden, ibid., pp. 230-240.
3Eden, ibid., pp. 226-227.
38An impression reinforced by the Maffey Commission's report in
June, 1935. Survey, 1935, II, pp. 42-44.
Simon had referred in his exchange with Grandi alluded to the expansion
of British diplomatic stakes to include Abyssinian territorial in-
tegrity, if it should become the League's business to settle the
conflict, and the accompanying deterioration of Anglo-Italian relations
Sir John Simon left the Foreign Office before the extent of
British involvement in the Italo-Abyssinian dispute increased. Sir
Samuel Hoare replaced him in June, 1935. He spent his first days in
office acquainting himself with the situation in Abyssinia and its
implications for British policy. In a series of meetings with
Vansittart and Eden he developed a rationale for future British policy
based upon the following "basic facts":39
First, Hitler's strength was becoming daily more formida-
ble, and his intentions more unabashed. Secondly,
Japanese aggression threatened us with war in the Far
East when we were not strong enough to resist Hitler in
Europe and at the same time fight in the Pacific. Thirdly,
it was essential to British security to have a friendly
Italy in the Mediterranean that would both guarantee our
lines of communication to the Far East and make it unneces-
sary for the Franch to keep an army on the Italian frontier.
Fourthly...Mussolini was at the time on very bad terms with
Hitler, his rival dictator.
From these "facts", Hoare concluded that, "the diversion of
Italian troops to a remote corner of East Africa, still worse,* the
breach of the Stresa front that the expedition involved, meant a
great and threatening accession of strength both to the Japanese in
the East and to Hitler in the West," and "decided to do what we could
to prevent its (Stresa) crumbling." For Hoare, therefore, the
^39Viscount Templewood (Sir Samuel Hoare) Nine Troubled Years, op.
cit., p. 153. Eden also refers to these discussions, op.cit., p. 246.
40bid *Italics mine.
Ibid. *Italics mine.
primary international restraints upon British policy were: (1) the
possibility of losing Mussolini's support against Hitler; (2) the
drop in British relative military capabilities even if Mussolini's
support continued, but his troops became tied down in Africa.
On the basis of this rationale Hoare and his Cabinet colleagues
decided to give Abyssinia a corridor to the sea from British Somaliland--
if Mussolini would accept Abyssinian territorial concessions to Italy
only along the disputed frontier between Abyssinia and Italian
Somaliland. However, Mussolini refused the British proposal and
replied that he had two alternative objectives in Abyssinia, depending
upon whether the conflict terminated peacefully or by force of arms.
Italy would not fight if Abyssinia agreed: (1) to cede to Italy those
territories conquered by Abyssinia and not inhabited by Abyssinia;
(2) to give control of Abyssinia proper to Italy while retaining the
formal sovereignty of the Abyssinian Emperor. If Abyssinia would not
agree to these terms, then Italy would invade and take over the whole
After Eden returned to London with Mussolini's reply, the Cabinet
met on July 3rd. At the meeting:
There was no dispute that the action the Duce contemplated
would involve a breach of the Treaty of 1906, of Article
10 of the Covenant and of the Kellogg Pact....The Govern-
ment concluded that everything depended upon the attitude
41This proposal came from Hoare, Vansittart, and Eden. Templewood,
op.cit., p. 155.
Eden, op.cit., pp. 247-256. Eden notes (p. 247) that his version
tallies with the official version in the records of the Italian Foreign
If British diplomatic stakes expanded to include these treaty obliga-
tions concerning Abyssinia's territorial integrity, then French support
would determine how effectively these obligations would be honored.
France was the only Great Power besides Britain (and Italy) that had
signed all three agreements. In this context Hoare and his colleagues
began to implement a two-level policy: (1) they warned Italy of
British obligations and encouraged negotiations at the symbolic non-
military level; (2) they took-steps to obtain an arms embargo against
both Italy and Ethiopia at the actual non-military level. Both
policies attempted to convince Mussolini of the dangers of Italian
military action in Abyssinia without damaging chances for a peaceful
settlement. The negotiations policy continued the earlier attempt
by Simon to operate within the restraints imposed by the desirability
of retaining Italian cooperation against Germany. The arms embargo
against both nations copied French policy and did not hinder Italian
military preparations. In his memoirs Hoare describes these maneuvers
as "The double policy that I was pursuing, of negotiations with Italy
and respect for our collective obligations under the Covenant, based
on Anglo-French cooperation."45
Between July and October, 1935, Mussolini maintained his previously
stated conditions for a peaceful settlement in informal talks at Paris
Eden, ibid., p. 266. The London Treaty of 1906 included the agree-
ment to respect Abyssinia's territorial integrity.
Eden, ibid., pp. 267, 323-325.
45Templewood, op.cit., pp. 160-161. See also Eden's comments and
efforts to lift the embargo against Abyssinia, pp. 323-325.
among the signatories of the London Treaty. Tension between Britain
and Italy increased. On September 10th, Hoare spoke at the League and
pledged British support of the obligations in the Covenant. British
intelligence reported the possibility of an Italian "mad dog" attack
on British territory and forces in the Mediterranean. The British
government responded by increasing the strength of the British navy in
the Mediterranean in order to insure these diplomatic stakes against
future Italian policy options. Undeterred, Mussolini's troops invaded
Abyssinia on October 3rd. During the next three months the British
government also asked for, and received, guarantees of assistance from
France, Greece, Turkey, Rumania, Czechoslovakia and Yugoslavia, if
Italy should attack Britain in the Mediterranean.47
46On July 5th, Hoare had warned Grandi that it "did not seem possible
to avert a great calamity" if Italy did not change her policy. Grandi
suggested a meeting of the 1906 Treaty's signitories. Mussolini even-
tually approved the informal talks in Paris.
4The rationale for taking these steps varied by individual, making
them difficult to scale. All of the participants at the Cabinet level
agreed that is was primarily necessary because of Italy's future Medi-
terranean policy options and present posture toward Abyssinia. Estimates
of relative British military capabilities in the area by British military
men varied from sufficient to inadequate, depending upon the contingency.
Against Italian ships the British navy was sufficient, but Italian dive-
bombers would make any battle with Italy very costly. This price would
be too high to pay, given other British commitments, unless total war
was contemplated. Hoare, on the other hand, felt that military action
against Italy was very unlikely and these moves were pro forma pre-
cautions. He communicated this interpretation to the Italian government.
Baldwin apparently viewed the moves as serious attempts to influence
Italy's Abyssinia policy toward a negotiated settlement. The decision
to scale the military considerations as secondary follows Hoare's
rationale and reflects the assumption that he was the "real" British
decision maker until his dismissal in December, 1935. Evidence to
support this position appears in the very recent, comprehensive bio-
graphy of Baldwin, based on access to Cabinet papers and Baldwin's pri-
vate papers. See Keith Middlemas and John Barnes, Baldwin (London, 1969),
pp. 850-858, 863, 876 ff., and also Templewood, op.cit., pp. 160 ff.
When Mussolini invaded Abyssinia, Hoare continued his "double
policy" of negotiations with Italy and cooperation with France and
the League until he resigned as Foreign Secretary in December. The
British supported the League's policy of economic sanctions, subject
to French endorsement of their scope. Simultaneously, British efforts
to formulate a territorial settlement that would end the conflict con-
tinued. The pursuit of the latter policy at the expense of the former
led to Hoare's downfall. In December, he and the French Prime Minister,
Laval, tentatively proposed a'settlement that would concede to'Italy
large segments of Abyssinian territory currently occupied by Italian
troops and make frontier adjustments between Abyssinia and Italian
Somaliland in favor of Italy. Hoare's Cabinet colleagues at first
supported this proposal then changed their minds and asked Hoare to
Hoare justified his proposal partly on the same grounds that had
provided a rationale for his negotiations policy before Italy had
invaded Abyssinia. He wished to keep Mussolini from aligning with
Hitler in Europe, a move that would decrease British support and rela-
tive military capabilities against Germany. In addition, however,
the implications of the example set by Mussolini's successful military
efforts in Abyssinia since October may have influenced Hoare. According
to his memoirs, he thought that unless negotiations ended the war,
48Vansittart was the other important sponsor of the Hoare-Laval
Pact; he had accompanied Hoare to Paris for the talks with Laval.
49Templewood, op.cit., pp. 191-192. See also Winston Churchill,
The Gathering Storm (Boston, 1948), pp. 181-182.
. eventually a total Italian military victory was certain, which would
disrupt the League and encourage German aggression.50 There is also
evidence that indicates Hoare was influenced by Laval's desire to try
and conciliate Italy before the League considered expanding economic
sanctions to include an oil embargo. There were two incentives for
Hoare to follow Laval's wishes: the potential lack of French support
in the League for oil sanctions and the French inability to come to
Britain's immediate aid in the Mediterranean if Mussolini should attack
there after the imposition of oil sanctions. These latter two con-
siderations appear to have been secondary in Hoare's mind.
From the beginning, Hoare and the British Cabinet regarded the
Hoare-Laval Pact as a tentative proposal, subject to consideration by
the League and the two disputants.53 Nevertheless, the domestic opposi-
tion in Parliament and among the British public induced the Cabinet
to ask for Hoare's resignation as a symbol of the Government's abandon-
ment of the proposal as British policy. Abyssinia also rejected the
Pact's terms. The opposition at home and abroad centered around the
belief that the territorial concessions to Italy in the Pact amounted
to "rewarding the aggressor."
Templewood, ibid., p. 183. This last reason may be an after-
thought to reinforce in retrospect his policy rationale.
51Ibid., pp. 178-179. The Hoare-Laval talks were on December 7th
and 8th; the League would consider oil sanctions on December 12th. See
also Survey, 1935 II, pp. 291-300.
52Compare Templewood, pp. 191-192 with Survey, 1936, II, pp. 291-294.
53Survey, ibid., pp. 300-301, Templewood, ibid., p. 182.
Middlemas and Barnes, op.cit., pp. 886-896.
Sir Anthony Eden succeeded Hoare as Foreign Secretary. He advo-
cated increasing the League's economic sanctions to include an oil
embargo against Italy, but eventually dropped this proposal when it
met with continuous French opposition.55 Eden also supported efforts
to conciliate the two disputants, so long as mediation proposals came
from League of Nations Committees rather than through channels outside
the League. The final Italian military conquest of Abyssinia in May,
1936, bankrupted both of these policies, and Eden then recommended
the abandonment of sanctions and the adoption of a non-recognition
policy by Britain. These proposals became British policy in July,
1936, when the League withdrew its sanctions policy. At that time the
British government also announced the imminent withdrawal of the
British warships sent to the Mediterranean from other stations during
Several international conditions influenced these changes in
British policy. Eden recommended the abandonment of sanctions and the
demobilization of the Mediterranean fleet because Italy had success-
fully defeated Abyssinia. He recognized that the French would support
his policy and that this change in policy might contribute to reaching
an agreement with Italy concerning Anglo-Italian relations in the
Mediterranean. Eden also realized that other members of the League
would not use force to restore the status quo in Abyssinia, nor were
5Eden had obtained Cabinet permission for oil sanctions if they
would be supported by other League members.
Eden, op.cit., p. 474.
57Eden, op.cit., pp. 430-432, 474, 669.
TABLE 3. INTERNATIONAL RESTRAINTS AND.BRITISH FOREIGN POLICY
DURING THE ITALO-ABYSSINIAN CONFLICT
Date Anal. Empir.
1/35 s/n-m Advocate
5/35 s/n-m Warn of
6/35 s/n-m Offer terr.
7/35 a/n-m Arms embargo
9/35 q/m Send ships
10/35 a/n-m Economic
11/35 s/m Obtain mil.
12/35 s/n-m Hoare-Laval
Milieu Abyss. Terr. Integ.
Br. Colonial poss.
Milieu Abyss. Terr. Integ.
Milieu Abyss. Terr. Integ.
Milieu Abyss. Terr. Integ.
Poss. British poss. in
Milieu Abyss. Terr. Integ.
Poss. Br. poss. in Medit.
Milieu Abyss. Terr. Integ. Yes
Milieu Abyss. Terr. Integ.
Milieu Abyss. Terr. Integ.
Poss. Br. Poss. in Medit.
they likely to maintain the League's existing sanctions policy. The
relative importance of these conditions in determining Eden's decisions
varied. In his memoirs Eden maintains that he "had recommended the
raising of sanctions, a particularly difficult decision for me, the
main purpose of which had been to improve Anglo-Italian relations."58
Eden hoped to coordinate the withdrawal of sanctions with other League
members, but beyond this point, their policies do not appear to be a
primary consideration in his decision to change British policy. Nor
did Eden emphasize the desirability of maintaining Italian cooperation
against Germany, as Hoare had done. Instead, from the start of his
tenure as Foreign Secretary he stressed "close Anglo-French under-
standing and coordinated action" in order to "find a way to survive the
tests which the next few months must bring from Hitler more importantly
The German Occupation of the Rhineland
The first test from Hitler came on March 7, 1936, three months
after Eden became Foreign Secretary, when German troops occupied the
demilitarized zone in the Rhineland. This move violated the Locarno
Agreements, in which Britain and Italy had agreed to come to the aid
of either France or Germany if one of these two nations should violate
the frontiers of the other or make any other "flagrant" violation of
58Ibid., p. 669. The same reasoning (p. 474) applied to de-
mobilizing the Mediterranean fleet. The French Government reported
that as of May, 1936, thirteen nations were violating the League
sanctions policy. Eden, p. 433.
59Eden, .cit., p. 355.
Eden, op.cit., p. 355.
the Locarno Agreements. The Locarno Agreements also provided for
procedures to deal with violations that were not "flagrant." For this
kind of act, the question was to be put before the League of Nations
Council. If the Council should judge that the violation had actually
occurred, then the signatories of the treaty were to come immediately
to the assistance of the injured Power. However, the injured Power
was to invoke the casus foederis.
For the case at hand France was the object of the violation.
British policy, therefore, depended partially upon whether France
should define the act as a flagrant or non-flagrant violation. The
French government on March 8th decided to classify it as a non-flagrant
violation and put it before the League Council.62 On March 9th, the
British government implemented a symbolic military policy by reaffirm-
ing the British guarantee in the Locarno Agreements to come to the aid
of France if French frontiers should be violated. At the symbolic
non-military level the British government protested against the German
movement of troops into the Rhineland. In announcing the occupation
6Rhineland Pact, Art. 4, paragraph 3, cited in Survey, 1936, p. 266.
The full text of this pact is in Newman, op.cit., pp. 207-211.
6Survey, ibid., p. 267 for citation from Locarno Treaty, Art. 4,
paragraphs 1 and 2.
Ibid., p. 268.
Cf. Eden's statement to the House of Commons on March 9, 1936, in
the R.I.I.A.'s Documents on International Affairs, 1936, pp. 52-56. The
guarantee no longer held for Germany, since Germany had repudiated
Locarno on March 7th, but it did apply to Belgium. Belgium's frontiers
had been covered under the Locarno treaty.
64bid., plus Eden's memoirs, .cit., pp. 380-382.
Ibid., plus Eden's memoirs, op.cit., pp. 380-382.
of the Rhineland on March 7th, the Germans had also expressed an in-
terest in negotiating a new agreement for organizing the security of
Europe, now that German sovereignty and equal rights had been fully
attained.65 The British responded to this overture at the symbolic
non-military level by promising to examine closely the German pro-
The diplomatic stakes for British policy makers during the Rhine-
land crisis, therefore, were the territorial integrity of France (and
Belgium), the demilitarization of the Rhineland, and a general settle-
ment with Germany in Europe. In conversations with French and Belgian
representatives between March 10th and 19th, Eden formulated a policy
which linked all of these objectives together by guaranteeing British
assistance to France under three contingencies, each one to be the
subject of military talks between the British and French General Staffs.
The first contingency encompassed the short-run period of emergency
characterized by German occupation of the Rhineland: Britain would
guarantee French security, attempt to negotiate a compromise regarding
the Rhineland, and establish a new security arrangement among inter-
ested powers to replace Locarno. The second and third contingencies
provided a British guarantee to France for the long run, either as
part of a new Locarno-type arrangement or even if negotiations aborted
without agreement on the Rhineland or a new security arrangement for
6The text of Germany's March 7th statement is in Documents on
International Affairs, 1936, op.cit., pp. 41-46.
6Eden's House of Commons speech, op.cit., footnote 63.
Survey, 1936, pp. 288-289 and Eden, Dictators, ..cit., p. 405.
Europe. The pattern of British policy in each contingency was the
same: a symbolic military policy on behalf of French territorial
integrity and a symbolic non-military policy regarding the Rhineland
and a new Locarno.
The international restraints that operated on Eden's formulation
of British policy were: (a) his desire for close Anglo-French relations,
which meant that France must support British policy; (b) Germany's
policy of simply occupying the Rhineland without threatening French
frontiers directly; (c) British naval involvement in the Italo-
Abyssinian conflict and an inadequate British army, which left the
British government unready to react militarily to Hitler's move; (d)
Germany's policy of offering to negotiate a new settlement.69 The
desire for close Anglo-French cooperation against Germany and Germany's
policy of simply occupying the Rhineland without threatening French
frontiers determined British symbolic military policy regarding French
territorial integrity. The lack of an immediate German threat to
French territory made Britain's unreadiness to react militarily above
the symbolic level a secondary determinant.
The determinants of the symbolic non-military policy regarding
the Rhineland and a general settlement were the British desire for
close Anglo-French cooperation and the inability to react militarily
to the occupation of the Rhineland. E.den believed that, "Britain's
armed forces were inadequate and unprepared and our support, except at
Eden, ibid., pp. 385-414.
sea, could only be token."70 The British Chiefs of Staff qualified
his assessment even further, saying that any risk of war with Germany
would require the withdrawal of British naval forces from the Medi-
terranean.7 Eden attempted to reconcile the French demand for
security with Britain's lack of military capabilities by adopting a
symbolic non-military policy of negotiations, backed by a British
guarantee of French territorial integrity irrespective of the outcome
of the negotiations.72 Germany's offer to negotiate a new general
settlement was a secondary consideration in Eden's calculations, since
he believed that Hitler was likely to "repudiate any treaty even if
freely negotiated (a) when it becomes inconvenient, and (b) when
Germany is sufficiently strong and the circumstances are otherwise
favourable for doing so." However, Eden did not advocate ignoring
Germany's offer to negotiate. In the memorandum to the Cabinet just
quoted, he concludes:
....owing to Germany's growing material strength and power
of mischief in Europe, it is in our interest to conclude
with her as far reaching and enduring a settlement as
possible whilst Herr Hitler is in the mood to do so. But
on entering upon this policy we must bear in mind that,
whatever time-limits may be laid down in such a settlement,
Herr Hitler's signature can only be considered as valid
under the conditions specified above.
701bid., p. 396.
Ibid., p. 400.
721bid., pp. 398-409
73Ibid., p. 387.
TABLE 4. INTERNATIONAL RESTRAINTS AND BRITISH FOREIGN POLICY
TOWARD GERMANY DURING THE RHINELAND CRISIS
Decision Stakes Primary Reasons
Date Anal. Empir. Anal. Empir. Cross-Press. Rel. Cap's.
3/36 s/m Will fight for Milieu Fr. Terr.
3/36 s/n-m Protest, but Milieu Eur. col Yes Low
will negotiate. security
for Eur. pact
Consequently, when the League Council declared on March 19th
that Germany had violated the Locarno Agreement, Britain pursued
a new general settlement throughout 1936, while at the same time pre-
paring to honor the guarantee to French territory. In April, conver-
sations between the British and French General Staffs began--accom-
panied by German protests. A series of diplomatic exchanges in the
summer and fall of 1936 between London and the other Locarno Powers
failed to arrange even a conference to discuss a new settlement. By
this time the attention of the Great Powers had turned to the Spanish
Civil War, which had begun in July, 1936. British policy makers now
had three foreign policy projects that would become intertwined during
the following year. They were the Italo-Abyssinian conflict, the
Spanish Civil War, and the search for a modus vivendi with Germany.
75Ibid., pp. 414-417.
BRITISH CONFLICT BEHAVIOR, 1937-1941
In 1937 Britain's foreign policy makers concerned themselves with
four major issue-areas. They included the status of Abyssinia, now
fully under Italian control; the Spanish Civil War; Germany's relations
with Austria and the problem of reaching a settlement with Germany in
central and east Europe; and the resumption of the Sino-Japanese con-
flict in northern China. The Abyssinian question and the Spanish Civil
War provided occasions for conflict between Britain and Italy, while
the latter two issues brought Britain into confrontation with Germany
Abyssinia and Spain
British diplomatic stakes in the Italo-Abyssinian conflict per se
were no longer important. The British policy of non-recognition was a
legal artifact. In withdrawing their legation from Addis Ababa, the
British government had bestowed de facto recognition upon the Italian
conquest. However, Eden hoped to use de jure recognition as a quid
pro quo in negotiations with the Italian government on other issues.
The principal issue that concerned Eden in this respect was the level
of Italian involvement in the Spanish Civil War.
When the conflict in Spain began in the summer of 1936, the
The other major source of friction was the anti-British propaganda
spread by Italy in the Middle East. Anthony Eden, Facing the Dictators,
op.cit., p. 485.
.British and'French governments attempted to gain an agreement among
the European Great Powers to pursue a policy of non-intervention.
They proposed that the governments of the Great Powers prohibit the
export of arms to Spain from their countries. The Italian government
accepted this proposal, but with some reservations. The Italian note
of acceptance distinguished between direct and indirect intervention
and based Italian reservations upon the latter, which would permit
"volunteers" from Italy to fight in Spain. The German and Soviet
governments replied favorably to the Anglo-French proposal, and the
French government then suggested that the Great Powers form a committee
to coordinate the details of their common non-intervention policy.
The Non-Intervention Committee met for the first time at London in
The British diplomatic stakes associated with the policy of non-
intervention included a government in Spain that was free from outside
domination, an objective related to other British diplomatic stakes,
including protection of British possessions in the Mediterranean, free
access to the Mediterranean, and the prevention of war in Europe among
the Great Powers. Eden believed that any policy other than non-inter-
vention in Spain might result in a Fascist government indebted to Italy
for its victory and operated under Italian influence. Such a government
might make territorial concessions to Italy or permit Italian military
2Eden, ibid., p. 453.
3bid., p. 452.
bases in the Balearic Islands. Gibralter would then be threatened
and, ultimately, British access to the Mediterranean would be en-
dangered. Finally, if Italy (and Germany) openly intervened, then
Russia and France might also intervene extensively, which could result
in a war among the Great Powers.
British policy toward Italy during the early stages of the Spanish
Civil War, therefore, operated at the symbolic non-military level. The
principal international condition which determined this policy was the
realization that intervention by Italy might precipitate future Italian
policy choices that would threaten the British position in Europe and
the Mediterranean. The French government's initiative in suggesting
the non-intervention policy was a secondary incentive to adopt the
policy, since Eden believed at the time that the policy was "the best
that could have been devised under the circumstances," and stated that
he "should have been glad to be able to say that non-intervention was
The qualified agreement by Italy to the non-intervention policy
accompanied feelers by the Italian government concerning the possibility
of improving Anglo-Italian relations in the Mediterranean. Between
September and November, 1936, the Italian Ambassador to London, Count
Grandi, visited Eden several times.6 The two men exchanged assurances
that Britain did not desire to threaten Italy's Mediterranean interests
and that Italy was not a danger to British Mediterranean interests.
41bid., p. 475.
5Ibid., p. 454.
6bid., pp. 475-481.
Grandi also asked Eden to recognize de jure the Italian conquest of
Abyssinia. Eden refused to take this step, and Grandi departed for
Rome on November 14th for instructions.
When Grandi returned to London on November 25th, he informed Eden
that Mussolini wished to formulate a Gentleman's Agreement, a joint
Anglo-Italian declaration that would refer to the complementary nature
of their interests in the Mediterranean. Grandi also told Eden that
the Italian government understood that Britain could not bargain over
Abyssinia, meaning that it would not be included in the negotiation
of the Gentlemen's Agreement. Eden accepted Italy's offer of dis-
cussions and instructed Sir Eric Drummond, the British Ambassador to
Rome, to begin negotiations, telling Drummond that the agreement must
not offend the French, must declare respect for the territorial in-
tegrity of Spain and the other Mediterranean countries, and give some
assurances that Italian propaganda against Britain in the Near East
would cease. Italy accepted these stipulations and the two nations
signed a joint statement in January, 1937.
The Gentleman's Agreement, however, failed to remove the two
major sources of tension between the British and the Italians.
Mussolini continued to send "volunteers" to General Franco's armies
Eden's reasons for refusing to recognize the Italian conquest of
Abyssinia were his desire for a quid pro quo and the fact that Britain's
military capabilities were sufficient to neutralize any threat by Italy
over this question. Eden, ibid., pp. 477-478. Britain's obligations
as a member of the League of Nations also would have made recognition
difficult, since Abyssinia still maintained a delegation there.
Ibid., p. 483.
Eden, ibid., pp. 484-486.
'in Spain, and the British government still did not recognize the Italian
conquest of Abyssinia. Eden felt that "Mussolini had used our negotia-
tions as a cover plan for his further intervention" and "that the in-
creased Fascist intervention in Spain was a violation of the spirit of
the Agreement."0 In a memorandum to the Cabinet Eden concluded:11
The Spanish civil war has ceased to be an internal Spanish
issue and has become an international battle-ground. The
character of the future government of Spain has now become
less important to the peace of Europe than that the dictators
should not be victorious in that country.....Unless we cry a
halt in Spain, we shall have trouble this year in one or
other of the danger points.....Memel, Danzig, or Czechoslo-
The principal mode of intervention in Spain was the sending of
"volunteers." The British proposed that the non-intervention policy
be expanded to include an embargo on foreign volunteers as well as
arms shipments. On February 20th, the Non-Intervention Committee
ratified the British proposal and set up a method of supervision de-
signed to monitor all Spanish frontiers and harbors for violations.
The Great Powers also agreed to withdraw foreign forces already fight-
ing in Spain.2
Italy and Germany agreed to these proposals, but did not honor
them. They continued to send "volunteers" to Spain throughout 1937
1Ibid., pp. 486-487.
Ibid., pp. 487-490.
12Ibid., pp. 493-495. Eden claims that the British proposal was
a weakened version of his ideas by the Cabinet.
and did not withdraw the forces already there.13 At the same time
tension between Italy and Britain increased in the Mediterranean, as
the Italians launched new propaganda broadcasts in the Near East, and
demands in the Italian press for de jure recognition in Abyssinia in-
creased. In July the British government received reports of an in-
crease in Italian troops in Libya and Italian naval maneuvers around
Sicily.14 Unidentified airplanes attacked a British tanker off the
coast of Spain in early August, and by the end of the month several
countries had reported submarine attacks on their merchant ships all
over the Mediterranean. The British Admirality believed that Italian
submarines had orders to attack any country's oil tankers sailing to
ports controlled by the Spanish government.5
British diplomatic stakes in the Spanish conflict thus increased
to include British lives and property endangered by "pirate" (Italian)
submarines. The British reacted by authorizing British ships to
attack and destroy any submarines that attacked them. At the suggest-
ion of the French government, the Mediterranean and Black Sea nations
met in September at Nyon, Switzerland to formulate a joint policy.
Italy decided not to attend the conference, stating that the acts of
piracy should be referred to the Non-Intervention Committee. Eden
13A disagreement between Russia and the Axis nations over including
financial aid in the embargo prompted Italy and Germany to declare that
they would delay troop withdrawals until agreement was reached.
14Eden,ibid., pp. 505-506.
15Ibid., pp. 515-517.
16Ibid., p. 523.
represented the British government at Nyon and proposed that the Med-
iterranean be patrolled by destroyers with orders to sink any sub-
marine attacking a non-Spanish ship. The plans for a destroyer patrol
divided the Mediterranean into different zones. The British and French
would initially conduct the patrol, but the plan made provisions for an
Italian zone if the Italian government should subsequently wish to
contribute destroyers. At the end of September Mussolini agreed to
join the patrol, and the "pirates" disappeared from the Mediterranean.
Eden's proposals' for this actual-military policy were based upon
the "overwhelming Anglo-French sea power in the Mediterranean," and
his belief that the British government must act decisively to maintain
Anglo-French prestige in Europe.8 He viewed the Nyon Conference as
an opportunity to deal effectively with one aspect of the Spanish
problem and demonstrate Anglo-French solidarity. The failure of the
Gentleman's Agreement and the success of the Nyon policy encouraged
Eden to deal firmly with Italy concerning the question of "volunteers"
in Spain and the de jure recognition of Abyssinia. He believed that
improved relations with Italy "could only be realized on a basis of
reciprocity." In the next few months Eden attempted to base British
de jure recognition of Abyssinia upon an Italian quid pro quo such
as withdrawal of Italian forces in Spain. This strategy, however,
brought him into conflict with Neville Chamberlain, the new British
17Ibid., p. 532.
18 ., pp 527-528.
Ibid., pp. 527-528.
19Ibid., pp. 535-538.