Origins of ww2 Reconsidered
Origins of ww2 Reconsidered
War Reconsidered
Second Edition
The Origins of the
Second World War
Reconsidered
Second Edition
Notes on Contributors ix
Preface xi
7 Appeasement
PAUL KENNEDY AND TALBOT IMLAY 116
viii Contents
10 More than meets the eye: the Ethiopian War and the origins
of the Second World War
BRIAN R. SULLIVAN 178
Sally Marks is the author of The Illusion of Peace: International Relations in Europe
1918–33; Innocent Abroad: Belgium at the Paris Peace Conference of 1919 and
the forthcoming The Ebbing of European Pre-eminence: An International History,
1914– 45.
Brian R. Sullivan has taught at Yale and the Naval War College, and done
strategic analysis at National Defense University He specializes in the history
of Fascist Italy and U.S. national security affairs and has co-authored Il Duce’s
Other Woman, a biography of Margherita Sarfatti. His current projects include
editing works by the naval theorist Romeo Bernotti and researching
Mussolini’s intelligence agencies.
Books have their own histories: they have births (usually painful), lives
(sometimes long, sometimes short), and deaths (often slow). When, in 1983, I
first had the idea for this one, it proved difficult to bring to fruition. Publishers
were wary of the project, suggesting that it did not seem “marketable.” Fifteen
years, thousands of copies and six printings have shown that the people who
commission books may be no better at forecasting the future than are those
who propose them. So I begin by paying homage once again to Jane Harris-
Matthews, then of Allen & Unwin, who was willing to take a chance on the
idea – and on its proponent who, at the time, was only beginning his editorial
career
During its life to date this book has passed from Allen & Unwin to Unwin
Hyman, then to HarperCollins, and now, perhaps finally, to Routledge. Its
editor at Routledge, Heather McCallum, believing that it has a future still,
asked me in 1996 if I would be prepared to prolong its life by producing a
new edition. Uncertain as to whether the book’s life deserved to be thus
extended and also of my own willingness to revitalize it – I consulted colleagues
who had been using the book in their teaching. Somewhat to my surprise,
they were unanimous in their opinion that a new edition would be helpful,
and they were forthcoming with suggestions about how it might be made more
useful. In producing this new edition, I have, in so far as it was possible, been
guided by their advice. Unfortunately, it has not been possible to include in
the second edition revised versions of all of the essays that appeared in the
first. The fact that the contributions of Norman Rich, Piotr Wandycz, Lloyd
Gardner, Akira Iriye, and Edward Ingram do not appear in revised form here
is certainly no reflection on the quality of their essays. Students and scholars
who wish to consult these authors may do so via the internet (at: http://
quarles.unbc.ca/history/hist.html).
The first edition was produced without the use of computers (or e-mail)
and, with almost everyone now using them, I thought it would be helpful to
contributors to work on a computerized version of their original essay This
work was willingly undertaken by my wife Valerie, who, in spite of her qualms
about the computer age (which, she argues, means mainly that I am now able
to work all of the time, anywhere), did the job with her customary dedication
xii Preface
and good humor. So the renascence of this book owes much to her The first
edition, decided upon during a splendid meal in San Francisco attended by
Valerie and hosted by Jane, would not have seen the light of day without the
support and the encouragement of these two wonderful women, and it is to
them that I dedicate this new edition.
1 The revisionist as moralist
A. J. P. Taylor and the lessons of
European history
Gordon Martel
Images of the 1930s continue to flash past us: Hitler’s moustache and
Chamberlain’s umbrella are still instantly recognizable; Nazi war criminals still
make the front pages; novels and films warning of a new menace emanating
from Brazil or Bavaria can be almost assured of popular success. The Second
World War, its symbols and personalities, continue to grip the modern
imagination. Thus the war – and its origins – functions today as a mental and
moral shorthand: anyone wishing to evoke an image of wickedness personified
need only mention “Hitler”; for stupidity, blundering or cowardice, substitute
“Chamberlain.” But political rhetoric extends the boundary beyond personality.
The systems we condemn are “totalitarian” or “dictatorships” (frequently both),
and we must never be guilty of “appeasement” in our relations with their leaders.
Politicians find these words useful because ordinary citizens agree that the Second
World War was caused by Hitler and his totalitarian dictatorship, and that it
might have been prevented had it not been for the policy of appeasement that
served only to whet his appetite.
Anyone who doubts that these simple assumptions are widely, almost
universally, subscribed to is invited to witness the effect of setting loose a class
of undergraduates on A. J. P Taylor’s The Origins of the Second World War. There
the effect is electric: they are stunned to read that Hitler neither planned nor
caused the war, that appeasement was not necessarily a bad thing, that new
ideologies such as fascism and communism were much less significant than the
aims and ambitions of statesmen, typical of all regimes, at all times. If the student
is converted to the Taylor view, war is almost certain to break out on the home
front; the young may be prepared to embrace new ideas, even if only as a
temporary fashion, but their parents are more likely to regard them as
treasonable. Two generations after its publication Origins has not lost its power
to provoke.
When the book first appeared in 1961 it created a storm. Professional
historians attacked Taylor for almost every imaginable sin: his evidence was
scanty and unreliable; he distorted documents by means of selective citation and
dismissed those he disliked by claiming they did not count; his logic was faulty;
he contradicted himself repeatedly and drew conclusions at variance with his
2 The Origins of the Second World War Reconsidered
own evidence. Nor was the storm confined to the citadels of academia – to
scholarly journals, college corridors, senior common-rooms and faculty clubs.
The debate was carried on in public – in newspapers, on television and radio.
Questions were asked in Parliament. Lifelong friendships were dissolved. Careers
were made and unmade. Taylor was soon the best-known historian in Britain:
his autobiography was a best-seller; an entire issue of The Journal of Modern History
was devoted to him; he has been honored with three Festschriften, and any book
with his name on it has been assured of popular success. One eminent historian,
when asked to contribute an essay to the first edition of this book, declined on
the ground that Taylor had no right to hold the first-mortgage on the subject of
the origins of the Second World War. He may not have the right, but hold the
mortgage he does. What other 38-year-old book on the war’s origins continues
not only to be available in paperback but can be seen to be stacked high in
university bookstores throughout the English-speaking world? Teachers wishing
to shake students out of their lethargy do well to introduce them to A. J. P. Taylor.
But the great man is now dead and most of the furor that emanated from his
book has gone with him. Nevertheless, interest in him remains strong, the
debate on the war’s origins continues and the book stimulates controversy still.
Book-length studies have now appeared in the form of Robert Cole’s A. J. P.
Taylor: The Traitor Within the Gates and Adam Sisman’s A. J. P. Taylor: A Biography,
the scholarship and insights of both of which will be surpassed by Kathy Burk’s
in her forthcoming biography.1 New surveys, especially Philip Bell’s Origins of
the Second World War in Europe and Akira Iriye’s The Origins of the Second World
War in Asia and the Pacific, but also Andrew Crozier’s The Causes of the Second
World War and Richard Overy’s The Origins of the Second World War, have certainly
replaced Taylor’s as books in which teachers can have confidence when
introducing students to the subject. But those very characteristics that make
these newer works more reliable make them less exciting, less challenging and
– ultimately – less enduring. Less careful, less balanced, more opinionated and
more provocative, Taylor’s book will remain in print long after his successors’
have ceased publication. It may be, as some argue, that he will continue to be
read largely as “an historical curiosity,” or mainly by graduate students
exploring the historiography of the subject, or by general readers looking for
someone who can be read for amusement and entertainment. But read Taylor
certainly will be.
The first question to be asked is why the book caused such a storm when it
appeared. The answer is that Taylor challenged an interpretation of the war’s
origins that had until 1961 satisfied almost everyone in the postwar world, and
because he conducted his challenge in flamboyant prose with such scathing wit.
Before Taylor launched his attack, the only point being debated was whether
the appeasers were foolish cowards who allowed themselves to be duped by
Hitler, or cunning capitalists who hoped to use Hitler to crush communism in
the Soviet Union. Blaming the war on Hitler certainly suited the Germans: with
the Nazis either dead or in hiding, they could claim to be blameless and to have
a claim to a respectable role in the new democratic alliance. This was equally
The Revisionist as Moralist 3
satisfactory in the west, where one might have expected an Orwellian unease to
emerge when the enemy was transformed into ally and the ally into enemy–but
the west now claimed to be united against “totalitarianism” rather than against
states or nations. The Second World War had been fought for a great and noble
principle, and this principle endured into the era of the Cold War. The enemy
had merely changed location: his ambitions and tactics remained the same.
Taylor would have none of this. The war had not been fought over great
principles, nor had Hitler planned its outbreak from the start. Taylor thereby
challenged two of the most confident assumptions of the 1950s. While others
saw in Hitler a demonic genius who was able to pull the strings of European
politics so masterfully because he had a carefully mapped out plan, Taylor saw
only an ordinary politician who responded to events as they occurred, who asked
only how he might benefit from them. Where others saw laid down in Mein
Kampf a blueprint, Taylor heard the confused babble of beer-hall chatter. Where
others saw a timetable for war in such documents as the “Hossbach
memorandum,” Taylor saw the petty intrigue and political machinations typical
of the Nazi system of government. If Taylor was right – if Hitler had not in fact
carefully plotted his route to world dominion well in advance and then followed
the route step-by-step this could only raise new, and possibly awkward, questions.
Some believed that Taylor was whitewashing Hitler, absolving him of guilt.
But Taylor did not stop with Hitler. He took a contrary view of almost every
significant figure of the interwar period: Chamberlain was neither a bungler nor
a coward, but a highly skilled politician who enjoyed the overwhelming support
of his party and his nation; Stresemann, the “good German” but for whose death
Germany might have followed a peaceful path, turns out to have shared Hitler’s
dreams of dominating eastern Europe; Roosevelt’s economic policies were
difficult to distinguish from Hitler’s; Stalin turns out to have been Europe’s most
conservative statesman, proposing to uphold the peace settlement of 1919 and
wishing the League of Nations to be an effective international institution, rather
than a monstrous ideologue plotting world revolution. If readers were not
offended by Taylor’s revisionist sketch of Hitler himself, they were almost certain
to find offense elsewhere in his book.
If readers discovered heroes and villains being turned upside down in Origins,
they also found states being turned inside-out. Anyone who believed in a
wicked Russia, a noble Poland, a beleaguered France, an efficient Italy or a
nationalistic Czechoslovakia would have their assumptions rudely challenged.
Russia never did more than ask to be accepted as a legitimate sovereign state;
Poland – corrupt and elitist as it was – was not a state such that one could be
proud of having fought to save it; France had consistently aimed to draw in
the new states of central and eastern Europe to fight on its behalf – while never
intending to assist them in any way; Italy was not the powerful representative
of a dynamic new political system, but the foolish plaything of a blustering
and blundering egomaniac; Czechoslovakia, even though democratic, “had a
canker at her heart,” its large German minority alienated from the Czech-
dominated centralized state.2
4 The Origins of the Second World War Reconsidered
It would be astonishing that someone who had reached maturity in the 1920s
should not have shown interest in war, nationalism, and revolution. The
consequences of 1914–18 were readily apparent: the physical destruction, the
disabled veterans, and the long lists of war dead inscribed on memorials; the
disappearance of the Habsburg, Ottoman, and Russian empires, and their
replacement by new “national” states and the Soviet Union. The greatest
historical controversy of the decade raged over the question of responsibility for
the outbreak of war in 1914; and Taylor later explained how he was struck by
the contrast between this stormy debate and the quiet complacency that
surrounded the origins of the Second World War. Even historians whose training
and work had been in other fields and earlier periods turned to recent diplomatic
history. One of them, the Austrian A. F. Pribram, had turned from Cromwell
and the Puritan revolution of the seventeenth century to recent Anglo-Austrian
relations. Thus, in one of the many “accidents” that transformed his “personal
history,” Taylor – who had gone to Vienna to work with Pribram on Cromwell
– was diverted from English domestic history to European diplomacy. His
personal and professional life exemplified the connection between the profound
forces of the age and the effect of chance and circumstance.
Taylor also explained that he followed the fashion of the 1920s in assuming
that Germany had been unfairly burdened with the guilt of having initiated
the First World War. But, as he set about investigating various aspects of
modern European history, he discovered that he and the “revisionists” were
wrong, that the peacemakers of Versailles were right: the responsibility for the
war lay with Germany. By the time he came to write Origins, he regarded
Germany as the dynamic element in European politics over the past century;
it was Germany that was growing, expanding, looking forward to a future when
it would be dominant in Europe and able to take up the position of a full-
fledged world power.
One of the distinctive features of Origins is the connections that are constantly
made between the interwar years and the nineteenth century; while most
historians concentrated on Hitlerism and ideology – usually treating these as
aberrations in statecraft – Taylor was keen to show the links between William
II and Hitler, the parallels between Chamberlain and Gladstone, the continuity
of Russian policies from the Romanovs to the Soviets. The great powers hoped
the Spanish Civil War would burn itself out, he insisted, “as Metternich had
hoped would happen with the Greek revolt in the 1820s” (p. 158). Schuschnigg
suffered from the perpetual illusion, peculiar to the Austrians, that exposing
nationalistic intrigue would stir the conscience of Europe into action – just as it
had “seemed to them axiomatic in 1859 that Cavour would be deserted by
Napoleon III” (p. 178). Stresemann shared Bismarck’s belief that peace was in
Germany’s interest, although he was “no more inclined to peace from moral
principle than Bismarck had been” (p. 79). These allusions to European politics
of the nineteenth century warn the reader not to trust those who would treat
the 1930s as if they had existed in a vacuum.
8 The Origins of the Second World War Reconsidered
Taylor admired few statesmen. His favourite book, The Troublemakers, extolled
the virtues of the dissenting tradition; his heroes were the critics and outsiders
of English history – the Cobbetts and the Cobdens, the Lloyd Georges and the
Beaverbrooks. And he showed how unreliable are the public utterances of men
in power. While Lamartine, foreign minister and romantic poet, was boldly
announcing a revolutionary French foreign policy in his manifesto of 1848, he
was also apologizing for it in private, pleading with the Duke of Wellington to
“understand its real sense” – that it was a gesture in public relations.3 When
Taylor treated Hitler’s speeches and Mein Kampf as meaningless or untrustworthy
pieces of self-advertisement, readers recoiled from the shock. But Taylor treated
all statesmen in this way “Great men in public life love power,” he explained;
“they fight to get it and they use it ruthlessly when it is in their hands.”4 They
make speeches, write books and strike poses in order to dupe an unsuspecting
public; it is the job of the historian not to be duped, to look for the reality behind
the facade.
The principal reality highlighted by Taylor is that statesmen seek to maintain
or extend the interests of their respective states. Those who take their own
rhetoric and ideals too seriously are likely to land themselves in trouble.
Palmerston’s greatness resided in his ability to recognize that, although he
trumpeted liberal Whig ideas, the interest of peace sometimes meant cooperating
with Austria in spite of the two countries’ ideological incompatibility. Palmerston
would do the right thing when opportunity and interest combined (as they did
during the unification of Italy), but he would do nothing for Poland or Hungary,
as “the one was beyond his reach, the liberation of the other he supposed would
have been against British interests.”5 Crispi, the prime minister who wished to
turn his country into a great colonial power, “lived in a world of illusions and
was leading Italy to disaster.”6 In all of Taylor’s work a motif is constructed in
which it is the dreamers, the speculators, and the ambitious who allow their grand
designs to overpower their appreciation of what is possible. Napoleon III, who
attempted to destroy the balance of power, “substitute his own hegemony,” and
replace the Holy Alliance with the “revolutionary association” of his dreams,
led the Second Empire to destruction.7
Dreamers who act out their dreams are not the only ones who are dangerous
in Taylor’s world – there are also those reactionaries whose nightmare visions
of the future lead them to oppose all change and often end with them flinging
themselves headlong into disaster in an act of national suicide. Such was
Metternich, who “dreaded action, sought always to postpone decisions and cared
only for repose”; when he fell from power he “brought down old Austria with
him.”8 Conservatives are especially likely to fall prey to the temptation of believing
that one dramatic act against liberals, radicals, nationalists, or revolutionaries
will destroy the enemy, restore the balance, and remove uncertainty. When
Austria–Hungary went to war in 1914, war was an end in itself: “the countless
problems which had dragged on so long could all be crossed off the agenda.”
Instead Austria–Hungary disappeared from the map.9 This is a complicated world
that Taylor describes, one in which it is as dangerous to try to stop the movement
The Revisionist as Moralist 9
of the world as it is to speed it up, and whether one chooses to act or stand still
the consequences can rarely be foreseen with any accuracy In this, the world of
Europe between the wars was little different from the Europe of the nineteenth
century. Men continued to dance “the perpetual quadrille of the Balance of
Power,” as much as some of them might wish the music to stop so “that they
could sit out a dance without maintaining the ceaseless watch on one another.”10
Although the faces were different after 1919, the problem confronting
European diplomacy remained the same: how to deal with the fact that Germany,
still the greatest of the European powers, was more convinced than ever that
the international system had been specially designed to thwart its designs. In
fact, ringing it with small states, and forcing the Soviet Union out of the European
equation, meant that when it recovered its cohesion and efficiency it would be
stronger than ever. Taylor was certainly convinced, long before writing The Origins
of the Second World War, that Germany had decided to exploit its potential for
establishing hegemony in Europe. This was signified most dramatically when
William II dismissed Bismarck and replaced him with advisers who favoured
“world policy,” who imagined that Germany was capable of realizing this
ambition “before she had secured the mastery of Europe.”11 The differences
between Bismarck, Bethmann-Hollweg, and Hitler were matters of temperament
and tactics: when the heirs of Bismarck rose up in 1944 it was “Hitler’s failure,
not his policy,” that drove them to resist.12 In the 1930s the Germans sought to
reduce Hungary to its true national size while incorporating the rest of the
Habsburg monarchy into the German Reich; but before 1914 they had been
restrained from this only by “dynastic scruples” and “twinges of Bismarck’s
caution.”13 The First World War sprang from Germany’s world policy, from its
decisions to challenge both Russia and Great Britain as world powers.14 By the
1950s Taylor was repeatedly referring to the war of 1939–45 as “the second
German war.”
It must have seemed bizarre to Taylor when Origins was criticized by some as
an apologia for Hitler, because the essence of his interpretation was the
continuous line of development in modern European history whereby Germany
sought to establish its domination of Europe by controlling the center and the
east. Although he dismissed the notion that Hitler was a careful and meticulous
planner, he clearly and unequivocally explained that Hitler “intended Germany
to become the dominant power in Europe.” But he also insisted that in this Hitler
was “like every other German statesman” (p. 171). These were the two arguments
that distinguished Taylor most clearly from other historians: Hitler ceased to be
the mad genius who pulled all the strings and had the whole play worked out
in advance; and he became just another German, struggling for mastery in
Europe. Even Hitler’s anti-Semitism had been “the Socialism of fools” for years:
“everything which Hitler did against the Jews followed logically from the racial
doctrines in which most Germans vaguely believed” (p. 100). If Taylor was right,
Hitler could not be expunged from the historical map because of his uniqueness:
he must be seen as a part of German, and even of European, history.
10 The Origins of the Second World War Reconsidered
countries as their satellites. Other Powers seek to defend their vital interests by
force of arms” (p. 27). What does it say about the way in which international
affairs are conducted that this personification of wickedness, when regarded from
the perspective of traditional European diplomacy, was simply an ordinary
statesman going about his business in a time-honored fashion? “In international
affairs there was nothing wrong with Hitler except that he was a German.”
So, in the final analysis, Taylor also chose to turn the interwar years into a
morality play. But unlike his predecessors, the play he has written bears the signs
of genius: few things are as they appear to the naked eye; honorable intentions
lead to tragic conclusions; wicked designs are facilitated by the well-intentioned.
His work will surely endure, if only because he rescued this vital part of the
human story from the vapid simplicities of good versus evil and returned it to
its proper place of complexity and paradox. “Human blunders...usually do more
to shape history than human wickedness” (pp. 265–6).
In the past forty years, much has been made of Taylor’s “philosophy of
history” – much too much. Taylor himself consistently denied having a
philosophy or even a systematic approach. These denials have the ring of truth.
In all his work he tried to tell stories, to explain how one thing led to another,
to answer the child’s question: “What happened next?” But good stories have
something to tell us, just as good jokes tell the truth. Such philosophy as may
be found in Origins can be summed up as a warning to mistrust historical truths
and parallels. We do learn from our mistakes, he says – we learn how to make
new ones. The attempt to extract simple policies from the lessons of the 1930s
was one of the enemies of clear thinking in the period that followed the Second
World War: Munich and the argument over appeasement supplied “superficial
parallels and superficial terms of abuse.”15 A conciliatory policy towards Russia
“would not be rejected so firmly now were it not for the recollection of the
appeasement towards Germany that failed a decade ago.”16 This warning against
the historical cliches offered up by historians and statesmen is as valuable a
message today as it was when first uttered.
These new essays on the origins of the Second World War are designed
neither to honor A. J. P Taylor nor to replace him. They undoubtedly testify to
his influence. If anyone is to hold the first-mortgage on the subject, we are
fortunate that that person should be someone able to write vigorous prose and
to stimulate debate – even among those not yet born when the book was written.
But Taylor has claimed that in writing Origins he wished to examine events in
detail, and the details of what happened behind the scenes are available to us
today in a way that was almost unimaginable when it was written. Not
surprisingly, the specialists contributing to this edition, having had the
opportunity to examine these events in great detail, have found much in his book
that requires revision or reconsideration. They have found that some of the
charges leveled at Taylor twenty-five years ago, especially those of contradiction
and overstatement, were justified. They have also found that he made mistakes
and overlooked material available to him, that he sometimes guessed wrong or
allowed prejudice to blind him.
12 The Origins of the Second World War Reconsidered
Nevertheless, one is left with the distinct impression that Taylor will continue
to be read and re-read with interest and profit for decades to come. These essays
have been collected with that premise in mind: to contribute to the ongoing
debate by clarifying vital aspects of Taylor’s interpretation, by synthesizing the
work that has been done over the past quarter-century, and by offering fresh
evaluations of major themes connected with the origins of the Second World
War. This book is intended not to signal the end of the debate but to show where
it stands today, and where new discoveries are being made.
Notes
1 I wish to express my gratitude to Kathy Burk for allowing me to read the early
chapters of her forthcoming biography. Although she had not yet reached the post-
1961 controversy in her writing, she assures me that nothing I say here will be
contradicted by her!
2 A. J. P. Taylor, The Origins of the Second World War, Harmondsworth, 1964, p. 190.
All references in this essay are to the Penguin paperback edition, which includes
the 1963 Foreword, “Second thoughts.”
3 A. J. P. Taylor, The Struggle for Mastery in Europe, Oxford, 1954, p. 5.
4 A. J. P. Taylor, Rumours of Wars, London, 1952, p. 25.
5 A. J. P. Taylor, The Italian Problem in European Diplomacy 1847–1849, Manchester 1934,
pp. 236–42.
6 Taylor, Struggle for Mastery, p. 324.
7 Taylor, Struggle for Mastery, p. 61.
8 A. J. P. Taylor, The Habsburg Monarchy, 1809–1918, London, 2nd edn, 1948, pp.
34, 56.
9 Taylor, Habsburg Monarchy, p. 232.
10 Taylor, Struggle for Mastery, p. xix.
11 Taylor, Struggle for Mastery, p. 294.
12 A. J. P. Taylor, Bismarck, London, 1955, p. 272.
13 Taylor, Habsburg Monarchy, p. 230.
14 Taylor, Bismarck, p. 268.
15 Taylor, Rumours of Wars, p. 76.
16 Taylor, Rumours of Wars, p. 80.
2 1918 and after
The postwar era
Sally Marks
The controversy over A. J. P Taylor’s The Origins of the Second World War centered
on the prewar years and the outbreak of war. Taylor’s dissent from the
assumption that Adolf Hitler was the primary cause and his views on
appeasement angered many, generating a debate on certain issues. Were Taylor’s
critics blinded by a conviction that Hitler caused the Second World War?
Regarding other leaders of the 1930s, did Taylor equate ineptitude or error with
evil? Above all, did Hitler “plan” the Second World War? On another level,
Taylor was charged with abuse and misuse of documentation, dismissal of
evidence, contradiction, overstatement, a narrow focus on diplomacy, and
ignoring fundamental causes.1
In the uproar, another question bearing on fundamental causes was ignored:
how sound are his early chapters addressing the pre-Hitler period? Taylor would
argue, and few historians now disagree, that the reasons for the Second World
War did not suddenly emerge in 1935 or 1936. After all, Hitler gained power
in the Weimar Republic because many Germans thought he could resolve their
discontents. His first success, in 1930, came during a recession but in an election
dominated by foreign policy, though many grievances about the Versailles treaty
had been partially or fully resolved. So why did his foreign policy have such
appeal? Taylor dismisses this problem by claiming Hitler’s policy was that of
his predecessors,2 but he also asserts that the entire interwar period forms a unit
and that events at the end of the first war were vital in provoking the second.
Certainly, what happened in 1918–19 affected what came after, especially in
forming the attitudes of most Germans, and Eugen Weber claims “the 1930s
begin in August 1914.”3 Thus, Taylor’s early chapters on the failure to solve
what he calls “the German problem” warrant examination.
Characteristically, Taylor starts with the balance of power. The unification
of Germany created a great nation in Europe’s center but one soon balanced by
a Russo-French alliance. Even so, Germany nearly won the First World War. In
early 1918, it was triumphant in the east and undefeated in the west, where an
impasse existed in France. Yet by November, Germany had lost the war in the
west. However, the armistice and peace treaty permitted it to remain a major
entity only somewhat diminished in a Europe providing fewer neighboring
checks upon it. This implied, after a period of recovery and of breaking treaty
restrictions, potential renewed German continental dominance. Some, including
14 The Origins of the Second World War Reconsidered
many Britons, soon found this acceptable if domination were economic, and not
military For them, German recovery was desirable if peaceful, and their concern
was German grievances, not fear of new aggression (pp. 24–30).
The Versailles treaty created Germany’s grievances and confirmed its
continuation as a great power. Taylor argues that, some territorial clauses aside,
the treaty focused on providing security against Germany. He says it “lacked
moral validity from the start,” meaning that Germans did not deem it fair, that
many others, especially in the English-speaking world, came to agree with them,
and that Germans were united in their hatred of the Versailles Diktat, in their
determination to break it and revert to continental domination. Without German
cooperation, enforcement became difficult and the German problem continued
unabated (pp. 30–3).
Taylor says that the Germans might have accepted the treaty had not one
portion of it remained unsettled. Disagreement among the victors deferred
decision about the amount Germany was to pay in reparation for economic
damage to the victors. Taylor says the question remained open throughout the
1920s, exacerbating all sore points, generating emotion, and becoming a chronic
grievance. In his view, the Germans blamed all woes on reparations, whether
or not they were so related, and soon transferred this grievance to the entire
treaty. Britain, despite initial avidity, had second thoughts, turned against the
reparations clauses, and projected these feelings onto the whole treaty, excepting
clauses to Britain’s benefit (pp. 46–52).
British attitudes were important in the other problem Taylor cites, the lack
of Allied unity. The narrow victory on the western front in 1918 derived from
Russia’s 1914 contribution and then a coalition of France, Britain, Italy, and
America. Postwar communist Russia was self-absorbed and widely distrusted.
American withdrawal was less than complete but soon substantial. Italy was not
really interested in the German problem. That left the British and the French,
who disagreed regularly about Germany. France feared future German aggression
despite alliances with Poland and Czechoslovakia to replace the Russian tie.
According to Taylor, Britain thought French fears foolish, was confident there
would be no war, and wished to revive Germany in a peaceful Europe. In one
pretense at compromise after another, the two powers canceled each other out.
Then Britain offered a meaningless paper guarantee to France, confident it would
never have to be honored if France gave up all else: great power status; treaty
enforcement; and in part its eastern allies. Though the French initially balked
at the price and struggled to enforce the treaty, Taylor says they learned the
futility of this. Thus the paper guarantee emerged as the 1925 Locarno treaty.
Taylor claims it ended the First World War and brought peace to Europe,
whereas its repudiation in 1936 “marked the prelude to the second war” (pp.
34–9, 44–58).
In examining Taylor’s analysis of the postwar period, one finds that most
questions prominent in the debate are irrelevant, for they focus on the Nazi era.
The charge of misuse of documentary evidence is extraneous, for Taylor cites
no documentation earlier than 1932. Of the usual complaints, there remain
1918 And After 15
and accepting Weimar’s view, which indeed sought equality. To Germans, this
meant removal of defeat and its consequences, and a reversion not to early 1918
but to early 1914.
Taylor’s analysis of the postwar power balance is penetrating but does not
differentiate between what it was and how it was then viewed. Granted, archival
evidence clarifies the picture. He does well in distinguishing – at least for
Germany – between the short-term power balance and what it was likely to be
later if the treaty were not enforced. He sees that Germany would then revert
to continental predominance, but again does not explore the distinction between
reality and the perception of it, notably in Britain and America. Despite attention
to reparations, he scants economic factors and emotional issues important in
democracies. Public opinion and propaganda mattered in the 1920s in setting
the outer limits of policy and fundamental assumptions, not least those of
historians.
Finally, the issue of causation warrants the analysis of Taylor’s opening
chapters, for here he addresses underlying causes. Naturally, reviewers focused
on later portions of the book, some praising the first section in passing.5 Taylor
asserts variously that the First World War, the armistice, or the treaty caused
the Second World War. This is overstatement, contradiction, and simplification,
and approaches preaching historical inevitability. Taylor begins his account in
1918–19 because he believes Allied failure to solve the German problem then
laid the basis for the Second World War. He is right that the German problem
was not solved (pp. 16, 23–8, 43–4, 71, 267), but to attribute the history of the
next twenty years exclusively to events in those months is too categorical and
omits other factors which require exploration, including the nature of the German
problem.
Taylor never defines this question crucial to the continent’s future, asking in
effect on what basis Europe would be reconstructed.6 He assumes a united
Germany would be dominant and disruptive. In the Preface to the 1983 edition
of Origins, he asks why Britain and France did not resist German reversion to
great-power status.7 He ignores how this was to be done, especially without
Russian aid, and Allied assumptions, which he indicates elsewhere, about what
had been accomplished by the 1918 military verdict; and he also forgets that
the question was not whether Germany would be a great power, but on what
terms. He disregards the nature of German nationalism, the self-delusions of the
citizenry and the reasons for each. Taylor is right that Weimar Germany was
resentful and revisionist, though his exploration of the reasons is inadequate.
He asserts that both Hitler and his predecessors wished to revert to the situation
of March 1918, which hardly explains why Hitler’s foreign policy was more
popular than Weimar’s. He assumes a united Germany would drive by force
toward an equality tantamount to continental predominance, that little or nothing
could have been done to deter it (thereby contradicting his own 1983 question),
and that Germany would revert to its place as “the greatest power in Europe
from her natural weight” (p. 70). His conclusion that a war over the settlement
between Germany and Britain and France “had been implicit since the moment
1918 And After 17
when the first war ended” (p. 267) ignores how little of the 1919 settlement
remained in September 1939 and how few of Hitler’s goals dealt with treaty
eradication; it ignores also the underlying reasons for German resentment,
revisionism, and later larger aims, and what the victors could have done in 1918–
19 and the 1920s to deflect what he considers the inevitable course of history
Taylor perceives many pieces of the puzzle but not their connections. For instance,
he looks at British, French, and German attitudes but does not examine the effect
of British policy assumptions upon Germany, France, the unraveling of the treaty,
the German drive to regain continental predominance, and thus the history of
the 1920s.
Taylor’s approach to fundamental causes deals in absolutes from the outset,
positing an either–or situation. In November 1918, he argues, the Allies could
either break Germany up or accept its continental predominance and revert to
the situation early in 1918 when Germany was triumphant in the east and
undefeated in the west. Because they did neither, the German problem continued.
In fact, both options were unrealistic. This was clear at the time, and neither
was seriously considered. Accepting German predominance after a long, bitter
war was politically impossible, amounting to surrender on the day of victory.
Breaking up Germany was no solution, given the relative modesty of Allied
territorial aims in Europe, Wilson’s opposition to dismemberment, and Anglo-
American eagerness to retire militarily except for token contingents. Without
Russian collaboration, enforcement of disunification would have required a
lengthy British and American military commitment politically impossible to both
powers. Taylor says that in the armistice, the Allies committed themselves “almost
without realizing it” to continued German unity. Elsewhere and more accurately,
he notes that the Allies had no desire to dismember Germany but rather wished
to demonstrate that aggression “could not succeed,” and this they thought they
had done (pp. 26–7).
Thus Taylor’s solutions to the German problem – supremacy or destruction
– were not viable in 1918. Were there other possibilities which neither the Allies
then nor Taylor later considered? As he mentions, in 1918–19 a gap existed
between reality and German popular perception of it. He says Germany was
thoroughly defeated and its leaders knew that in November; a few pages on, he
reports without comment that “no German accepted the treaty as a fair settlement
between equals ‘without victors or vanquished’” (p. 26). Here Taylor brushes a
key factor, the popular German perception that the First World War was a draw
and that Wilson’s “just peace” should mean the status quo ante bellum with
rectifications in Germany’s favor.8 However, Taylor fails to explain how this
situation arose and what could have prevented it.
Some answers are obvious. The war was fought on the soil of the victors;
they lay in ruins, not Germany. Clearly, German self-delusion was a factor, but
the Allies permitted that, perhaps their greatest mistake. In November 1918 the
victors could end the war or fight on. In retrospect, they probably should have
fought on into Germany, but they did not know how near the end was. Further,
as Taylor notes, beyond thinking their goal of German defeat accomplished,
18 The Origins of the Second World War Reconsidered
Britain and France were exhausted and feared American domination if the war
continued (p. 26). Still, the document ending hostilities should have been called
a surrender, not an armistice, a misleading term which fostered German illusions
about the war’s outcome. Taylor ignores this possibility and also another, not
considered by the victors – that the surrender terms might have provided for a
modest Allied occupation of Berlin and other key cities for a year. But the Allies
feared more bloodshed and the spread of Bolshevism to their troops;9 so, in the
absence of Allied armies, the Germans were free to enter what Ernst Troeltsch
called the “dreamland” of the armistice period.10
They clung to Wilson’s “peace without victory” speech of January 1917 when
the United States was neutral and he was seeking a negotiated peace. They
became convinced that Wilson’s 1917 statements about a peace “without victors
and vanquished” and a “peace between equals” were part of the armistice terms,
which they were not.11 Forgetting the harsh treaties of Brest–Litovsk and
Bucharest (not to mention plans, unknown to the public, to squeeze the last
pfennig out of the Allies if Germany won), they naively assumed that the Reich’s
past expansion could serenely continue, that Wilsonian self-determination meant
Germany would lose no territory, though the armistice terms indicated otherwise,
and that it could annex Austria. But the Allies had no intention of converting
victory into defeat by handing their foe the gateway to south-eastern Europe
and the middle east: “Germany’s enemies had not fought a war and made
sacrifices only to end up by turning Little Germany into Greater Germany.”12
Thus, in their “dreamland,” German expectations embarked on a journey away
from reality.13
Playing the “if” game is usually futile, but German perceptions were so
important that we may linger briefly. A durable peace must rest on what a
defeated major power will accept, but what it will accept is affected by its
perception of its circumstances. In 1814, Russia’s army and tsar were in Paris.
In 1871, the German empire was proclaimed at Versailles, and Prussian forces
trooped through the Arc de Triomphe.14 In 1918, the Allies occupied only the
Rhineland. Had they paraded soldiers in Berlin and Munich, reality would have
intruded. Chancellor Friedrich Ebert would not have hailed German troops in
December 1918 with “As you return unconquered from the field of battle, I salute
you,”15 were Allied armies likely to be marching in Berlin the next week. Taylor
attributes German refusal to accept the treaty simply to the continued existence
of a united Germany. Had symbols of defeat been visible in German cities,
perceptions of the nation’s circumstances would have been different and, hence,
perhaps also views of the treaty.
Ebert’s statement reinforced German self-delusion about the war’s outcome
and the German army’s Dolchstoss myth that it was undefeated in battle but had
been stabbed in the back by pacifists, Jews, and socialists. Ebert, who needed
army support, did not counter it, nor did the Allies. Between November and
May, this myth led many Germans to a belief that Germany had not lost the
war. The victors devising the peace in Paris proceeded on the premise of German
defeat, which Weimar rejected. Thus, any treaty they were likely to produce
1918 And After 19
1922 Russo-German treaty of Rapallo. This fact only heightened French fears
and German demands for treaty revision.
There is another factor which Taylor does not address directly. As he notes,
the kaiser was gone, Germany had a democratic republic, and the Allies expected
it to settle down and live in peace with its democratic neighbors (pp. 26–7).
Perhaps they were naive in assuming that democracies are peaceful, but they,
like Taylor, overlooked the tragedy of German democracy. The Weimar Republic
was born of defeat and as psychologically deformed as the ex-kaiser; further,
the Social Democrats, more democratic than other parties, were saddled with
defeat and the armistice, neither their doing, along with the treaty and what
followed. No wonder democracy was unpopular and deemed “unGerman.”
Domestic politics required Weimar coalitions to resist treaty enforcement to prove
their patriotism; this policy was so successful that it became axiomatic. The
fragility of the Republic and its concentration on foreign policy were factors in
Germany’s refusal to accept the treaty and its rapid fraying.
Taylor wonders why the Allies thought they could impose upon Germany a
treaty it would accept. They thought that Germany accepted defeat and assumed
that, since other defeated great powers had honored peace treaties, Germany
would do the same; France had done so after the 1871 treaty of Frankfurt, also
a diktat. Taylor notes that the Allies believed, at least at first, in the sanctity of
contracts (pp. 34, 51): Again, with eyes to 1871, they assumed Germany would
honor the treaty to liberate the Rhineland quickly But in 1871, France knew it
had lost. Moreover, the Allies ignored the infant Republic’s lack of political roots
and the adolescently assertive character of German nationalism, which was a
constant threat to governments in an unpopular democracy.
Taylor suggests also that the treaty was unenforceable because Germany was
responsible for implementing it. The Allies, assuming acceptance of defeat and
the treaty by a sturdier German Republic than existed, expected German
cooperation. For this reason, and because Britain and America rejected lengthy
continental involvement, the treaty lacked sufficient enforcement clauses,
especially automatic ones not requiring Allied negotiation. Taylor is fight that
the Allied assumption that Germany would carry out the settlement gave Weimar
a weapon against the treaty (pp. 28, 31, 33), but he oversimplifies. Any treaty
requires co-operation, but more important in the 1920s was an evaporation of
the will to enforce in Britain, America, and usually Italy.
Of course Allied unity collapsed soon after the last shot was fired. It generally
does. Also, American withdrawal from treaty implementation, erratic but rapid,
threw the peace structure and the power balance awry, exacerbating Germany’s
perception of injustice, as also did the reversal of British policy from March 1919
on.23 One must investigate the reasons for and effects of British policy, for it
was British support of German resistance which made it effective. This Taylor
does not do. Again, the archives and recent works clarify the picture, but the
basic outlines can be found in books which he lists. And while the extent and
speed of Anglo-American reversal were not evident in 1919 when key treaty
22 The Origins of the Second World War Reconsidered
decisions were made, France was slow thereafter to face facts, because it was no
more eager to accept solitude than Germany was to accept defeat.
Taylor comes close to Britain’s motive when he cites Ramsay MacDonald’s
1932 remark about Britain needing to support both sides (p. 56), but he sees
neither its implications nor that they predated 1924, to which he applies it. Britain
began reverting in 1919 to its traditional role as the fulcrum in the power
equation, seeking a strong Germany to balance France and block bolshevism. It
never abandoned France entirely, encouraging false hopes here,24 and never
supported Germany completely, creating bitterness there.25 But usually it sought
concessions to Germany at French expense, both to lure Weimar westward from
Russia and to weaken France. All British leaders pursued this largely instinctive
policy, though Austen Chamberlain did so less than most others. It derived from
isolationism, imperial crises, nostalgia for Britain’s prewar economic and political
predominance, reaction against the war’s cost in blood and money, tradition,
misreading the power equation, reluctance to enforce the treaty, and fear of
France. Here Taylor’s crystal-ball clouded. We now know that Britain fixed on
the superficial situation and misjudged the long-term power balance. The
momentary superiority of the French army and air force, along with France’s
submarine-building program, alarmed British leaders who seriously thought the
next war might be against France.26
Thus Britain swung to the temporarily weaker power, inadvertently facilitating
Germany’s return to continental predominance with profound implications for
the western entente and its treaty enforcement. Taylor says that Britain and
France blocked each other’s policies but not that Britain did so better, having
more allies and superior skills at propaganda and playing to the historical
galleries. He notes that Allied threats were progressively less effective (p. 32),
but not that Britain increasingly refused to threaten and gave Germany assurances
behind France’s back, which is why threats by continental victors lost force.
Taylor adds that America constituted world opinion; here he is wrong. America
could be useful, and each country wanted its support, but Britain was the key
power for the continental players, above all for Germany,27 which never defied
the entente when it was united, only when Britain stood aloof or backed Berlin.
As Britain moved from its allies toward the middle, and sometimes surreptitiously
to the German side,28 treaty enforcement dissolved.
Taylor sees that America was more involved in Europe than most realized,
but should add that the involvement was sporadic, financial, and often unofficial.
He overlooks the role played in the 1920s by New York bankers, giving
MacDonald unilateral credit for the 1924 reparations settlement, when in fact J.
P. Morgan and Co.’s agents were equally important in bringing France to heel
(pp. 55–6). America supported Britain out of distaste for European embroilment,
reluctance to bother, susceptibility to Anglo-German propaganda, and hope that
Europe would “settle down.”29 It was much easier not to enforce the treaty; like
Britons, Americans failed to see that unity rendered enforcement unnecessary.
Despite his denunciation of Italian fascism as illegal, “corrupting,” and
dishonest, and of Mussolini as “fraudulent” and “a vain blundering boaster
1918 And After 23
As Taylor sees, French policy stemmed from what Austen Chamberlain called
its “nightmare horror” of German renascence.38 France read the underlying power
balance correctly, and saw a disparity more severe than Taylor perceives. He
cites Germany’s demographic advantage, but its younger population and higher
birth-rate guaranteed a greater future advantage. While he notes that Britain
wished to reintegrate Germany in European trade, he does not see the effect on
France, for Lorraine’s iron industry depended on reparations coal and German
markets.39 If France did not gain a German trade treaty (which Britain hindered)
before the economic clauses of the Versailles Treaty expired in 1925, it would
be at Berlin’s mercy. And effective transfer of reconstruction costs from the losers
to the winners added to Weimar’s advantage, as did erasure of its war debts
during the 1922–23 inflation, which was caused primarily by internal factors
but which reduced the real value of Germany’s domestic debt to less than 1 US
cent. Weimar emerged debt-free while continental victors faced staggering
domestic and foreign war debts and reconstruction costs.40
Further, France’s predominance eroded as army equipment aged and military
service was reduced from three years to one; moreover, by 1925 the national
will was exhausted. France was no longer truly a great power. In 1921, a French
diplomat remarked that “in the foreseeable future the difficulty will be to slide
France reasonably smoothly into the ranks of the second-rate powers, to which
she belongs.”41 Understandably, French leaders did not accept this and struggled
to maintain France’s artificial treaty predominance against Anglo-German
revisionism. With Gallic illogicality, however, they wanted both deterrence and
conciliation, treaty enforcement and a British alliance, for France’s “English
problem” was linked to its “German problem.”42 But Britain read French
nightmares and efforts at partial treaty enforcement as vindictiveness or
imperialism, fearing French hegemony would replace that of Germany.43 Taylor
reflects this outlook when he remarks that in 1929 danger remained of aggression
by France, the continent’s only great power because it had the only great army
(p. 62). Though Taylor admits France had accepted defeat, his assertion ignores
the Russian army (easily as large by 1929, especially in Europe), France’s
weakness, and how it could attack Germany without losing the Locarno tie to
Britain.
Though Taylor rightly says France saw its eastern allies as assets against
Germany, not as potential liabilities, he errs in thinking that they dominated
French foreign policy (p. 42). Since a Russian treaty was not practical politics,
the primary goal was always a British alliance, for Paris assumed the next war
would resemble the last, when the British navy, empire, and ties to Wall Street
saved France. Every French premier wished to revive the wartime alliance, but
London’s price proved prohibitive, higher than Taylor notes. Yet a frightened
France continued to seek protection, whereas Britain, alarmed at the prospect
of being drawn into another continental bloodbath and wanting to turn away
from Europe, pressed France to concede more to Germany to pacify it.44
Germany wanted to break the treaty and regain the status quo ante bellum,
undoing the 1918 military verdict; thus the postwar became the continuation of
1918 And After 25
war by other means.45 Germany fought in many ways, using John Maynard
Keynes as an unofficial adviser and propagandist.46 The key battlefield at first
was reparations. Though the archives have proved him wrong on some points,
Taylor commendably faced this topic when most historians avoided it. His
assessment shows more comprehension of France’s dilemma than many early
accounts, though at times he approaches hinting that if France had only thrown
away its costly victory without protest, all would have been well.47 Saying that
Britain was not ready “to underwrite every French claim against Germany” (p.
38) omits that London increasingly opposed all French claims, however well-
grounded, particularly if Weimar was likely to refuse, creating a situation
requiring Allied action. Taylor rightly says that the Germans “deliberately kept
their economic affairs in confusion” to escape reparations; he deserves credit for
not blaming German fiscal and monetary chaos on French vindictiveness and
on reparations which mostly were not being paid.48
Because reparations schemes were excruciatingly technical, neither Taylor nor
historians after him penetrated to their core, which was political, not financial,
though the key documents were available early on. At the peace conference,
Allied reparations experts knew German capacity to pay was somewhere between
40 and 60 (French and American estimates) milliard gold marks.49 However,
this could not be revealed to Allied electorates, who were led to expect that
Germany would pay for reconstruction and “the costs of the war.”50 Thus the
problem was to bury a realistic sum in misleading devices implying a vast figure.
From the peace conference on, every reparations scheme used one or several
red herrings to inflate the ostensible total. At Paris, the German reparations offer
(contingent on massive territorial revision) did the same.51 The ostensibly large
figures fooled the public and the historians, partly because all governments
wished to continue the fiction of immense totals. Britain and Germany cited them
to press for treaty revision. Weimar’s leaders used them to encourage Germans
to blame all economic ills on reparations,52 whereas cabinets of continental victors
stressed them to assure electorates that Germany; not Allied taxpayers, would
pay. Given the actual claims, the real question was whether Germany would
contribute to Allied reconstruction costs or whether these would be added to
Allied domestic and foreign war debts. As that had major implications for the
power balance, reparations became the terrain on which the war continued.
The reparations struggle was complicated by its link to war-debts disputes
between American creditors and European debtors, a subject on which the
literature remains sparse.53 When Taylor says that reparations were perpetually
unsettled, he forgets they were settled in 1921 by the London Schedule of
Payments, then unsettled at British–German American insistence, as happened
after each subsequent settlement. The resultant uncertainty derived from
Germany’s resistance with Anglo-American support. German threats of an export
drive impressed the British, as did Keynes’s pronouncements, which influenced
opinion where it mattered. Taylor properly notes that German impoverishment
derived from the war, not from reparations, but should add that its scale was
modest compared to that of the continental victors, especially of France.
26 The Origins of the Second World War Reconsidered
The key reparations battle came with France’s “occupation” of the Ruhr in
January 1923. Premier Raymond Poincaré opposed military action and hoped
through 1922 for British cooperation in economic sanctions.54 But, by December,
he faced a British–American–German front against treaty application,55 and saw
coal for French steel production, funds for reconstruction of industrial areas, the
treaty itself and France’s wartime victory draining away. Caught between
Germany, Britain and the powerful French Right, which urged action,56 Poincaré
lacked maneuvering room and so reluctantly moved narrowly to avoid becoming
the loser of World War I. The Ruhr encirclement was profitable57 and caused
neither the German hyperinflation, which began in 1922 and ballooned because
of German responses to the Ruhr occupation, nor the franc’s 1924 collapse,58
which arose from French financial practices and the evaporation of reparations.
Poincaré won the battle, forcing Germany to acknowledge its treaty obligations,
but lost the war because he failed to capitalize on his victory while he had the
upper hand, and the Allies swung to the other side.
Poincaré lost the 1924 election more from the franc’s collapse and ensuing
taxation than from diplomatic isolation, so Edouard Herriot led France to the
first of two decisive defeats. Herriot made mistakes before and during the London
reparations conference but, given the British–American–German front against
France, whether another premier could have salvaged more is questionable.59
Taylor correctly gives MacDonald generous credit for the 1924 settlement. His
policy did not differ from his predecessors’ (and he gave France no more, except
blither) but his style and tone did, and that mattered. Moreover, France was
now at the mercy of American bankers who dictated many terms.60 Reparations
payments were sharply reduced, with gradual increases to a substantial amount
– at which point Germany would seek new revision. France lost preponderance
on the reparations commission, the right of sanctions against future German
default, and hope of a German trade treaty before 1925.61 Taylor notes that the
treaty provided artificial compensations to France for Germany’s inherently
greater power, but omits that many were temporary and that in 1924 France
lost several advantages vital to its economic future and to treaty enforcement,
losses which affected the power balance, accentuating French weakness.
Taylor says that France learned “the folly of coercion” (p. 54). In fact, it
learned the futility of seeking partial treaty enforcement without token British
co-operation and American consent. Taylor assumes that because treaty
enforcement largely collapsed in 1924–5 (while contradictorily assuming most
of the treaty remained intact in 1939), it was inevitable that it do so. Actually,
the treaty could have been enforced – with a different set of British and, perhaps,
American assumptions and policies. Non-military actions to compel fulfillment
were available – such as seizing customs receipts, taxing German exports to
victors,62 surprise disarmament inspections, requiring Germany to tax to the level
of the victors (as the treaty specified), or to transfer some railway profits – but
these required a degree of Allied unity which did not exist. Taylor never considers
non-military enforcement, but he knows Britain and the United States thought
the war ended in 1918 and were not alarmed at the prospect of German
1918 And After 27
as if it could lumber across the frontier – and assumed wrongly that, since the
Reischsvervand der deutschen Industrie representing the German iron–steel– coal
complex was so dominant, the same must have been true of the Comité des Forges
of French steel companies.
Next came American and British archives, reinforcing the German view. Of
the two, the American files are the less important, unless the papers of bankers
are added. The massive, minuted British records are invaluable – and misleading
since few British officials questioned their own assumptions or the policy they
conducted. As Taylor notes, “morality” mattered to the British (p. 183); others
call it self-righteousness. Whichever it was, it played well to historians. Though
one can infer from British files that Poincaré did not want to enter the Ruhr,
few researchers did. Again, scholars adopted the viewpoint of what they read,
and for some years assumptions about British far-sightedness and generosity were
reinforced.77
Finally. French, Italian, and Belgian archives (and that of the League of
Nations) offered new information and a corrective to one-sided evidence. The
massive Belgian records compensate on German questions for the paucity of
internal French memoranda and loss of some dossiers in the Second World War.
Surviving French files reveal disarray, disorganization, and confused planning,
not only among government departments but also in their dealings with elements
such as the Comité des Forges; the contrast with German methods and comparative
cohesiveness is startling. In administration and policy planning, France had not
reached the twentieth century, a factor which influenced the power equation.78
Poincaré-la-guerre emerges as a timid procrastinator, fearful of decisions and
ignorant of monetary matters.79 Italian files add to the evidence, and not only
about Italy. Numerous books have appeared based on these archives, some
providing a more balanced analysis.80 Others improve on the evidence or suppress
it to French benefit, as still happens with German records to enhance Weimar’s
reputation, for the craft imposes few controls on the practitioner.
Historians now have most of the evidence they can expect, except as to east-
European archives. Though some private papers and a few government files
remain closed, most German, west-European, and American records are available.
To the east, whence the surprises, if any, will probably come, the situation varies
from country to country and year to year, as material is opened or closed,
apparently on bureaucratic whim. But for what one author called Great Britain,
France and the German Problem,81 evidence is plentiful and historians need only to
master it. As D. C. Watt remarked,82 the quantity of material renders the task
daunting, but some scholars have the necessary fortitude.83 And do they confirm
or deny Taylor’s arguments? Despite the lack of consensus, they do a bit of both.
The records confirm some of Taylor’s judgments, and he was among the
leaders in certain insights. He asserted that relaxing the treaty would render
Germany “as strong, or almost as strong,” as before (p. 28); others since have
seen that comparatively it was stronger.84 Taylor’s approach to the 1919 Polish
settlement was calmer than others’. He pointed to German resurgence and
internal pressures for German rearmament, since confirmed by archival
30 The Origins of the Second World War Reconsidered
research;85 some still echo him in less-measured tones about Germany’s innate
right to rearm, failing to ask: rearmament to what end? Taylor’s focus on
reparations was sound, if not always accurate, and more judicious than some in
the early literature. He was right that Germans blamed all woes on reparations
and transferred their feelings to the entire treaty. He saw that Germans convinced
London that reparations were unjust (and harmful to Britain’s economy), leading
it to apply this view of some clauses to most (pp. 49–50), though he ignores the
parallel development and its significance. The absence of the vital connection is
an instance of Taylor’s failure to put the pieces together to discover why the
interwar era went awry.
Since key clashes in the Franco-German struggle came over reparations, with
implications for treaty enforcement and the power equation, historians have
tackled this abstruse topic.86 Many of their findings could have been ascertained
earlier, had their eiders consulted published documents instead of journalistic
accounts. Historians returned to the texts, and to studies of economic theory,
and reached a degree of consensus. Most now think the 1921 London Schedule
of Payments probably lay within Germany’s capacity to pay, had it been willing
to try, and amounted to 50 milliard gold marks in nominal value, the present
value being less. The rest of the ostensible 132 milliard gold marks (nominal
value) represented camouflage to bridge the gap between feasibility and popular
expectations and also, if possible, to provide worthless paper to exchange for
American war-debt cancellation.87 Most think the transfer problem, allegedly the
barrier to substantial payments, was exaggerated. Certainly, there is no dispute
about the importance of the issue.88
Historians agree with Taylor about the speed of Germany’s return to
predominance, but instead of ascribing this phenomenon to Germany’s natural
place and an inevitable process, they ask why it happened so fast. Instead of
declaring the treaty devoid of “moral validity,” they ask why Germans and others
rejected it and why it disintegrated so quickly, noting that, because of the former,
the answer to the latter was an anti-French coalition with powerful economic,
financial, and propaganda weapons. These historians deem the German problem
more complex than did Taylor, and they examine aspects in Taylor’s beloved
power balance which he ignored. They cast their net wider, focusing on German
determination and self-delusion, the assertiveness of German nationalism, and
British blindness in helping Berlin to reverse the 1918 military verdict. They
emphasize French frailties, Belgian ambivalence in balancing between Britain and
France, and pressure by American bankers toward “appeasement” of Germany.
Instead of attacking Taylor’s focus on the power equation, these historians
examine economic, financial, ethnic, and domestic political and emotional factors
which Taylor had scanted. In particular, they stress France’s dependence on
German coal, coke, and outlets, along with Britain’s obstruction of Franco-
German economic arrangements. Though they agree that Germany had greater
power potential than France in more senses than Taylor mentions, few view the
postwar era with Taylor’s determinism as an inevitable German drive “to restore
the natural order of things” (pp. 22, 61).
1918 And After 31
these facts, though Hitler bragged that he never accepted anything under the
treaty and took his first stride toward power in that election.
This brings us to Hitler in comparison to his Weimar predecessors. True, his
early diplomatic moves polished off items on Stresemann’s list. But Weimar
politicians were preoccupied with the Polish Corridor, whereas Hitler was not.
Though the war which Taylor addresses, the European war of 1939, started there,
Poland was not important in Hitler’s thinking, for he saw it as either an obedient
satellite or an area to cross en route to Russia.93 Some Weimar leaders hoped to
regain Alsace–Lorraine, but their wider aims did not extend far beyond the
boundaries of 1914, Anschluss aside. They shrank from confrontation with Britain
and would have swooned at the idea of attacking Russia. Granted, the opportunity
was not yet there, but many Weimar notables were horrified when it happened,
and Hitler had declared this as a primary goal long before it was possible.94 Without
becoming embroiled in continuity theory, one can say that Hitler’s aims were more
vast, his ideology different, and his methods more drastic than those of his
predecessors. Weimar politicians wanted treaty revision; even in the 1920s, Hitler
deemed that a mistake, and posited more sweeping goals.95 Germany did not fight
“specifically in the second war to reverse the verdict of the first and to destroy the
settlement which followed it.” Much of that was destroyed before Hitler took office,
and much more had become so by 1939. Despite Taylor’s assertion that Hitler’s
“foreign policy was that of his predecessors” (pp. 23, 70–1), his world view was
far more grandiose.96 Whether one looks back on the Weimar years from the
perspective of the late 1930s or forward from the 1920s toward the Nazi era, the
idea that the Second World War had been “implicit since the moment when the
first war ended” (p. 267) is simplistic, and the concept of Hitler as one more Weimar
politician struggling manfully against the shackles of Versailles contradicts the
evidence from both eras.
Notes
1 Most major essays are collected in: E. M. Robertson (ed.), The Origins of the Second
World War, London, 1971; W. R. Louis (ed.), The Origins of the Second World War: A.
J. P. Taylor and His Critics, New York, 1972; and Journal of Modern History, vol. 49,
1977. Also K. L. Nelson and S. C. Olin, Jr, Why War?, Berkeley, CA, 1979, and W.
H. Dray, “Concepts of causation in A. J. P. Taylor’s account of the origins of the
Second World War,” History and Theory, vol. 18, 1978. And see the biographies by
R. Cole, A. J. P. Taylor, New York, 1993, and A. Sisman, A. J. P. Taylor, London,
1994.
2 A. J. P. Taylor, The Origins of the Second World War, New York, 1961, p. 70. All citations
are of this edition unless otherwise noted.
3 E. Weber, The Hollow Years: France in the 1930s, New York, 1994, p. 11.
4 This was corrected in the 1983 edition. In both, he misdates Mein Kampf and inflates
into an orange the German lemon which Sir Eric Geddes wished to squeeze until
the pips squeaked.
5 G. A. Craig terms the early chapters “masterly” in his essay “Provocative, perverse
view of pre-1939”, in Louis, Origins, p. 110. They are praised in Robertson, Origins,
by T. W. Mason, pp. 105–6 and E. M. Robertson, p. 15.
1918 And After 33
6 R. Poidevin and J. Bariéty., Les relations franco-allemands, Paris, 1977, pp. 236–7.
7 Taylor, Origins, 1983 edn, p. xiv.
8 In March 1919, the American government privately reminded Germany that the
Allies had won and rejected Germany’s interpretation that the war lacked victors.
See Akten der Reichskanzlei, Weimar Republic [hereafter ARWR]: Das Kabinett Scheidemann,
Boppard am Rhein, 1971, p. 28.
9 J. F. V. Keiger, Raymond Poincaré, Oxford, 1997, p. 248; S. A. Schuker, “The Rhineland
question,” in M. F. Boemeke et al. (eds), The Treaty of Versailles, Washington, DC,
1998, pp. 275–312.
10 See K. Schwabe, “Germany’s peace aims and the domestic and international
constraints,” and F. Klein, “Between Compiegne and Versailles,” in Boemeke et al.,
Versailles, pp. 37–68 and 203–20.
11 M. F. Boemeke, “Woodrow Wilson’s image of Germany, the war-guilt question,
and the treaty of Versailles,” in Boemeke et al., Versailles, pp. 603–14. For a study
accepting German views, see A. Lentin, The Versailles Peace Settlement, London, 1991,
pp. 6, 14. Whether Taylor does so is unclear: see Origins, p. 32.
12 H. Heiber, The Weimar Republic, trans. W. E. Yuill, Oxford, 1993, p. 38. Also I. Geiss,
“The Weimar Republic between the Second and Third Reich,” in M. Laffan (ed.),
The Burden of German History, London, 1988, pp. 68–9.
13 M. Lee and W. Michalka, German Foreign Policy, Leamington Spa, 1987, p. 19.
14 N. Sauvée-Dauphin, “L’occupation prussienne à Versailles,” in P. Levillain and R.
Riemenschneider (eds), La guerre de 1870/71 et ses conséquences, Bonn, 1990, p. 231.
15 R. G. L. Waite, Vanguard of Nazism, New York, 1969 edn, p. 7.
16 K. Schwabe, Woodrow Wilson, Revolutionary Germany, and Peacemaking, trans. R. and R.
Kimber, Chapel Hill, NC, 1985, p. 444n, and see pp. 300–29; M. Kitchen, Europe
Between the Wars, London, 1988, pp. 1–4; Heiber, Weimar Republic, p. 36; Laffan,
Burden, pp. 81–3; Lee and Michalka, German Foreign Policy, p. 19.
17 S. Marks, “Smoke and mirrors,” in Boemeke et al., Versailles, pp. 337–70.
18 E. Glaser-Schmitt, “The Making of the Economic Peace,” in Boemeke et al., Versailles;
S. Marks, “German–American relations, 1918–1921,” in Mid-America, vol. 53, 1971:
5.
19 W. Murray, “The collapse of empire,” in Murray et al. (eds), The Making of Strategy,
New York, 1994, pp. 396–7; A. P Adamthwaite, Grandeur and Misery, London, 1995,
pp. 86–8; H. H. Herwig, “Clio deceived,” in K. Wilson (ed.), Forging the Collective
Memory, Providence, RI, 1996, pp. 87–127; F. Fischer, From Kaiserreich to Third Reich,
trans. R. Fletcher, London, 1986, p. 83.
20 Herwig, “Clio,” p. 87; Lentin, Versailles, pp. 14–16; Keiger, Poincaré, pp. 271, 291;
Fritz Fischer, Germany’s Aims in the First World War, New York, 1967. Fischer echoed
L. Albertini, The Origins of the War of 1914, London, 1952, 3 vols, but no German
historian had accepted this interpretation.
21 In August 1914 the chancellor told the Reichstag that the 1839 treaties committing
Prussia and, later, Germany to Belgium’s defense, which Germany had just violated,
were a “scrap of paper.”
22 In March 1919, several Weimar cabinet members acknowledged German culpability
in the war’s outbreak but declined responsibility for their predecessors’ actions. See
ARWR, Das Kabinett Scheidemann, pp. 78–88.
23 In March 1919, Lloyd George’s manufactured evidence about starving German
children and his Fontainebleau memorandum started a trend. Baron Riddell, Lord
Riddell’s Intimate Diary of the Peace Conference and After, London, 1933, p. 210; United
States, Department of State, Papers Relating to the Foreign Relations of the United States:
The Paris Peace Conference. 1919, Washington, DC, 1942–7, 13 vols [hereafter FRUS
PPC]: vol. 2, pp. 139–43; D. Lloyd George, Memoirs of the Peace Conference, 2 vols,
New Haven, CT, 1939; vol. 1, pp. 265–73. Britain had told Berlin that German
war guilt was “undeniably established,” ARWR, Das Kabinett Scheidemann, p. 85.
34 The Origins of the Second World War Reconsidered
John Maynard Keynes, London, vol. 1, 1983, pp. 356–7, 364; J. M. Keynes, Collected
Writings, 30 vols, New York, 1971–89; vol. 18, pp. 159–60.
47 Taylor claims (pp. 31, 48) that France retarded reparations settlements to remain
on the Rhine. Actually, after many coal-payment defaults had demonstrated German
bad faith, Paris argued the 5–10–15-year occupation terms would not begin. This
security-based argument was a casualty of the 1924 settlement.
48 Taylor confuses Germany’s economy, healthy compared to those of the continental
victors, with its fiscal and monetary shambles. Germany falsified its economic data
and postponed fiscal and monetary reform to gain reparations reductions. After mid-
1921, cash reparations were a trickle, financed as earlier by foreign loans repudiated
by Hitler. Beyond the export advantage, Germany profited (approximately totalling
all payments in cash and kind through 1923) from the inflation until mid-1922,
thanks to foreign speculation in the mark which, owing to its depreciation, equaled
a direct capital transfer. K.-L. Holtfrerich, “Internationale Verteilungsfolgen der
deutschen Inflation,” Kyklos, vol. 30, 1977/8, pp. 271–91; S. A. Schuker, American
“Reparations” to Germany, Princeton, NJ, 1988; Skidelsky, Keynes, vol. 2, p. 117; P.
Krüger, Deutschland und die Reparationen, Stuttgart, 1973, pp. 182, 210; Heiber, Weimar
Republic, p. 86.
49 P. M. Burnett, Reparation at the Paris Peace Conference, 2 vols, Washington, DC, 1940;
vol. 1, pp. 25, 54–8. Britons on the peace conference’s Commission on Reparation
produced astronomical figures, but Keynes estimated 40–60 milliard, favoring the
higher figure. Keynes, Collected Writings, vol. 16, p. 378. A milliard is an American
billion, but “billion” has another meaning in Britain, so “milliard” is more exact.
Whereas the paper mark depreciated rapidly, the gold mark held at 4 to the dollar
and 20 to the pound.
50 Peace conference estimates of civilian damage were 60–100 milliard gold marks.
The US Army Corps of Engineers’ final figure, excluding damage in Russia, Poland,
and Czechoslovakia, was 160 milliard gold marks. Burnett, Reparation, vol. 2, pp.
49, 46.
51 Marks, “Smoke and Mirrors”; Krüger, Deutschland, pp. 187, 198–9; Heiber, Weimar
Republic, p. 67; J. M. Keynes, The Economic Consequences of the Peace, New York, 1920,
pp. 222–3.
52 Krüger, Deutschland, pp. 210–13.
53 The sole major study remains D. Artaud, La Question des dettes interalliées, 2 vols, Lille,
1978.
54 When Poincaré decided to enter the Ruhr is unclear. Archival evidence suggests
he hoped for united action until January 4, 1923. See also Schuker, French
Predominance, pp. 21–5, Keiger, Poincaré, pp. 290–3; M. J. Carley, “The shoe on the
other foot,” Canadian Journal of History, vol. 26, 1991, pp. 581–7.
55 The American element consisted primarily of financiers.
56 Schuker, French Predominance, pp. 20–1; Carley, “Shoe”.
57 Profits were held on the reparations account for all victors, but for technical reasons
most went to Belgium. Profits (after Ruhr–Rhineland occupation costs) were nearly
900 million gold marks, FRUS PPC, vol. 13, pp. 487, 785. These figures should
not be compared to the defunct London Schedule but to the four-year cessation of
payments Britain demanded in January 1923.
58 Schuker, French Predominance, is definitive.
59 Schuker, French Predominance, pp. 386, 392.
60 J. Bariéty, Les Rélations franco-allemandes après la première guerre mondiale, Paris, 1977;
Schuker, French Predominance.
61 France’s defeat was evident to contemporary observers. Bariéty, Les Rélations, p.718;
S. Marks, “The myths of reparations,” in Keylor (ed.), Legacy, pp. 155–67.
62 Britain did this for several years, but when France tried to do likewise Germany
howled.
36 The Origins of the Second World War Reconsidered
63 E. Goldstein, “The evolution of British diplomatic strategy for the Locarno Pact,”
in M. Dockrill and B. McKercher, Diplomacy and World Power, New York, 1996, p.
115.
64 For Briand’s first effort toward a British treaty, see S. Marks, “Ménage à trois,”
International History Review, vol. 4, 1982, pp. 524–52.
65 A. Adamthwaite, The Lost Peace. New York, 1981, p. 65; Poidevin and Bariéty, Les
Rélations, p.268; DDB, vol. 2, p. 213.
66 G. Stresemann, Vermächtnis, 3 vols, Berlin, 1932; vol. 2, p. 233.
67 Texts: Great Britain, Parliamentary Command Paper [hereafter Cmd] 2525, 1925;
Stresemann’s interpretation: Vermächtnis, vol. 2, pp. 232–3, and M. J. Enssle,
Stresemann’s Territorial Revisionism, Wiesbaden, 1980, pp. 109–12.
68 DDB, vol. 2, p. 399.
69 Enssle, Stresemann’s Territorial Revisionism, pp. 85–6, 104, 113, 126, 128. Briand
challenged Stresemann’s view through diplomatic channels but not publicly lest he
jeopardize ratification.
70 Dutton, Austen Chamberlain, p. 262.
71 J. Jacobson, Locarno Diplomacy, Princeton, NJ, 1972, Part 4.
72 DBFP, Series IA, vol. 5, p. 335. Texts of the Young plan and Hague conference
documents: Cmd 3484, 1930, Cmd 3763, 1931, and Cmd 3766, 1931.
73 DBFP, Series 2, vol. 1, pp. 314–24.
74 Key recent works on the peace conference: A. Sharp, The Versailles Settlement, New
York, 1991, Boemeke et al. (eds), Versailles, and Keylor (ed.), Legacy. My thanks to
Prof. Keylor and Dr D. Mattern for pre-publication knowledge of the latter works.
English-language studies: A. J. Mayer, Politics and Diplomacy of Peacemaking, New York,
1967; I. Floto, Colonel House in Paris, Åarhus, Denmark, 1973; K. Lundgreen-Nielsen,
The Polish Problem at the Paris Peace Conference, Odense, Denmark, 1979; S. Marks,
Innocent Abroad, Chapel Hill, NC, 1981; Schwabe, Woodrow Wilson; A. Walworth,
Wilson and his Peacemakers, New York, 1986; L. S. Jaffe, The Decision to Disarm Germany,
Boston, MA, 1985.
75 R. J. Schmidt, Versailles and the Ruhr, The Hague, 1968; D. Felix, Walther Rathenau
and the Weimar Republic, Baltimore, MD, 1971, and, to a degree, H. J. Rupieper, The
Cuno Government and Reparations, The Hague, 1979.
76 H. W. Gatzke, Stresemann and the Rearmament of Germany, Baltimore, MD, 1954; A.
Thimme, “Stresemann and Locarno,” in Gatzke (ed.), European Diplomacy Between
Two Wars, Chicago, IL, 1972, pp. 73–93.
77 D. Marquand, Ramsay MacDonald, London, 1977.
78 The situation emerges in the archival material. See also Schuker, French Predominance,
pp. 365–7, 385; J. Gillingham, Coal, Steel and the Rebirth of Europe, Cambridge, 1991,
p. 14.
79 Keiger, Poincaré.
80 A. Cassels, Mussolini’s Early Diplomacy, Princeton, NJ, 1970; R. J. B. Bosworth, Italy
and the Wider World, London, 1996.
81 W. M. Jordan, Great Britain, France and the German Problem, London, 1943.
82 D. C. Watt, “Some aspects of A. J. P. Taylor’s work as diplomatic historian,”Journal
of Modern History, vol. 49, 1977: 30.
83 English-language surveys: Adamthwaite, Lost Peace; S. Marks, The Illusion of Peace,
London, 1976, and Kitchen, Europe Between the Wars. Monographs: P. S. Wandycz,
France and Her Eastern Allies, Minneapolis, MN, 1962, and The Twilight of French Eastern
Alliances, Princeton, NJ, 1988; Jacobson, Locarno Diplomacy; C. S. Maier, Recasting
Bourgeois Europe, Princeton, NJ, 1975; Schuker, French Predominance; Gatzke, Stresemann;
Anne Orde, Britain and International Security, London, 1978, and British Policy and
European Reconstruction after the First World War, Cambridge, 1990; D. H. Aldcroft, From
Versailles to Wall Street, Berkeley, CA, 1977; and K. L. Nelson, Victors Divided, Berkeley,
CA, 1975.
1918 And After 37
84 F. H. Hinsley, “The origins of the Second World War,” in Louis (ed.), Origins, p.
72; G. L. Weinberg, “The defeat of Germany in 1918 and the European balance
of power,” in Weinberg (ed.), Germany, Hitler and World War II, Cambridge, 1996.
85 E.W. Bennett, German Rearmament and the West, Princeton, NJ, 1979.
86 English-language works: Maier, Recasting Bourgeois Europe; Schuker, French Predominance
and American “Reparations”, M. Trachtenberg, Reparation in World Politics, New York,
1980; G. D. Feldman, Iron and Steel in the German Inflation, Princeton, NJ, 1977, and
The Great Disorder. Also the debate in Journal of Modern History, vol. 51, March 1979;
Marks, “Myths,” in Keylor (ed.), Legacy, and “Smoke and mirrors,” in Boemeke et
al. (eds), Versailles; A. C. Mierzejewski, “Payments and Profits,” German Studies Review,
vol. 18, 1995, pp. 65–86.
87 The 50 milliard included A Bonds (the 12 milliard unpaid balance on what Germany
owed by May 1, 1921) and B Bonds. C Bonds constituted the rest of the 132
milliard, subject to arithmetical adjustments. Germany paid about 22 milliard gold
marks, of which less than one-third was in cash, mostly borrowed. See FRUS PPC,
vol. 13, p. 409; Schuker, American “Reparations”, pp. 106–8.
88 Journal of Modern History, vol. 51, March 1979. Little else has been written about
the historiography of the 1920s except J. Jacobson’s review articles, “Strategies of
French foreign policy after World War I,” Journal of Modern History, vol. 55, 1983,
pp. 78–95, and “Is there a new international history of the 1920s?”, American Historical
Review, vol. 88, 1983, pp. 617–46; C. S. Maier, “Marking time: the historiography
of international relations,” in M.Kammen (ed.), The Past Before Us, Ithaca, NY, 1980,
pp. 355–87, stresses American diplomacy. For an entirely different approach see
Gordon Martel, “Reflections on the war-guilt question and the settlement: a
comment,” in Boemeke et al. (eds), Versailles, pp. 615–36,
89 R. M. Spaulding, “The political economy of German frontiers,” in C. Baechler and
C. Fink (eds), The Establishment of European Frontiers after the Two World Wars, Bern,
1996, pp. 231–3; Fischer, Kaiserreich, p. 83.
90 C. Fink, “The minorities’ question at the Paris peace conference,” in Boemeke et al.
(eds), Versailles, pp. 249–74 and “The protection of ethnic and religious minorities,”
in Keylor (ed.), Legacy, pp. 227–38.
91 DBFP, Series 2, vol. 1, pp. 487–8.
92 DBFP, Series 2, vol. 1, p. 502.
93 G. L. Weinberg, The Foreign Policy of Hitler’s Germany, 2 vols, Chicago, IL 1970, 1980:
vol. 1, pp. 13–14.
94 Hitler’s Secret Book, New York, 1961, pp. 139, 145.
95 Secret Book, pp. 144–5; G. L. Weinberg, “Friedenspropaganda und
Kriegsvorbereitung,” in W. Treue and J. Schmädeke (eds), Deutschland 1933, Berlin,
1984, pp. 121–2.
96 G. L. Weinberg, World in the Balance, Hanover, 1981, pp. 76, 81–4, 89–90, and
“Die Deutsche Politik gegenüber den Vereinigten Staaten im Jahr 1941,” in J.
Rohwer and E. Jäckel (eds), Kriegswende. Dezember 1941, Frankfurt-am-Main, 1984,
pp. 73–9.
3 The end of Versailles
Stephen A. Schuker
When did the Versailles system break down? A. J. P Taylor never quite makes
up his mind in The Origins of the Second World War. He devotes almost one-third of
his book to charting the successive stages in the disintegration of the peace
settlement over the course of fifteen years. But he attaches at most secondary
importance to these epiphenomena. He approaches the subject on the assumption
that the Versailles treaty “lacked moral validity from the start.”1
That does not necessarily reflect his own judgment, Taylor insists (not wholly
convincingly). It represents, rather, the view of the Germans and the many people
in the Allied countries who came to sympathize with them (p. xi). The essence
of the matter lay in the balance of demography and economic resources. Those
criteria indicated that Germany remained potentially by far the greatest power
on the European continent after 1919; indeed the dissolution of Austria–
Hungary, the withdrawal of Russia from international affairs, and the exhaustion
of France and Italy rather worsened the disparity compared with relative strengths
before the war. All Germans, Taylor notes, meant to shake off some part of the
peace treaty as soon as it proved convenient to do so. They differed at most
over timing (pp. 24, 28). The conclusion seems to follow with disarming
simplicity. The treaty was doomed from the moment that the Allies agreed to
make it with a united Germany, the great unified state created by Bismarck.
“Given a great Germany…the only question was whether the settlement would
be revised, and Germany become again the greatest Power in Europe, peacefully
or by war” (p. 51).
Was the course of events, then, foreordained once the Allies had failed to
march on Berlin in 1918 and to impose a dictated peace? Is Taylor not forgetting
the caveat of the distinguished medievalist F. W. Maitland – which he quotes
elsewhere to telling effect – that historians should avoid “after-mindedness” and
remember that “events now long in the past were once in the future?”2 Taylor,
however, is not so easily pinned down. In 1919 the danger lay “in a hypothetical
future; and who could tell what the future would hold?” Out-of-office Frenchmen
might peddle separatism by the back door. “High-flying historians” might lament
that the work of Bismarck remained intact. Yet after every great war alarmists
had feared that the defeated power would strike again. It did so rarely, or else
halfheartedly (pp. 23, 25).
The End of Versailles 39
he do but to mediate the impressions garnered from those works through his
own experience of the 1920s as a politically conscious young man?
By his own account, Taylor came from an unusual family of Nonconformist
Lancashire cotton merchants. With a competence of £100,000 sterling, his father
had retired from business to join the Gas Workers’ Union, deserted the Liberal
Party for Independent Labour, and for twenty years displayed his militancy on
the local Trades and Labour Council, hoping to transform himself into an
authentic working man. His mother, not to be outdone, progressed from anti-
conscription agitation in the First World War to a romantic communism that
did not preclude acting as a conduit for Soviet funds to the British Communist
Party. Taylor himself was a 17-year-old fifth-former of decidedly progressive views
attending Bootham, the Quaker-run public school, when Poincaré occupied the
Ruhr in 1923. He had become a Labour firebrand at Oxford who thrilled to
the “revolutionary enthusiasm of the common people” in Russia by the time
that the frock-coated representatives of the western powers negotiated the Locarno
agreements two years later. When Hitler took over the Reich chancellery in 1933,
Taylor had already won local notoriety as the brilliant university lecturer who
served as chief spokesman for the Manchester Anti-War Council.7
Of course, no one with Taylor’s temperamental independence and iconoclasm
could remain in an intellectual straitjacket of any sort for long. Still, hardly anyone
from this background in interwar Britain had a good word to say about the
Versailles treaty. None would have defended the sanctity of international
obligations or expostulated on the justice of reparations and war-debt payments.
Few would have grasped the importance of a strong military establishment for
Britain, let alone for France. Significantly, Taylor opposed rearmament by the
National Government until the mid-1930s because he feared that Britain might
take Hitler’s side in a war against Russia. When his anti-Nazi sentiments
overcame those scruples and he swung round on rearmament in 1936, he lapsed
into political inactivity. To favor armament expenditure, whatever the
circumstances, was an impossible position for a man of the Left!8 The wonder
is not that vestigial traces of interwar Labour views turn up in Taylor’s treatment
of the period. The marvel is that Taylor the mature historian could so far
transcend the sentiments of his youthful milieu as to approach objectivity at all.
Yet one must somehow explain a perceptible inconstancy in Taylor’s
interpretation of the 1920s. Can one speculate that this derives from the
unresolved tension between Taylor’s youthful recollections and his subsequent
reflections as a historian? How else can the reader account for his disconcerting
tendency to write on both sides of almost every issue? Let us examine, for
example, his successive statements on the question with which we started – when
and how the Versailles system broke down.
The German army had ceased to exist as a major fighting force after the war,
Taylor tells us; no one had to worry about armed conflict with the Reich for
years to come. Nevertheless, measures of coercion against Germany ultimately
could not work. The decision whether to comply with the treaty remained in
German hands (p. 42; see also p. 29). At each confrontation from the 1918
The End of Versailles 41
Armistice to the 1923 Ruhr occupation, it became more and more difficult for
the Allies to threaten an application of force against Germany. Yet, in fact, by
1921 much of the peace treaty was being enforced (pp. 28, 42). Resentment
against the treaty increased with every passing year, largely because of
reparations. On the other hand, appeasement began not with Neville
Chamberlain, but with Lloyd George, who carried it through successfully. “Even
reparations were constantly revised, and always downward, though no doubt
the revision dragged out tiresomely long” (pp. 42, 48). The Locarno treaties of
1925 ended the First World War and ushered in a period of “peace and hope.”
Indeed, the most popular cry in Germany, as late as 1929, was “No More War.”
All the same, appeasement was not achieved, and when the occupying forces
left the Rhineland in 1930 German resentment bulked larger than ever (pp. 55–
9).
The conflict between France and Germany seemed bound to continue, Taylor
asserts, so long as the illusion persisted that Europe remained the center of the
world. On the other hand, it appeared that “treaty revision would go on
gradually, almost imperceptibly, and that a new European system would emerge
without anyone noticing the exact moment when the watershed was crossed”
(pp. 58–9). The French had “never possessed” a mobile army capable, in case
of need, of an independent offensive against the Reich. Consequently, French
foreign policy stood in fundamental contradiction with French strategy. The
system of security against Germany, however, as devised in the Versailles treaty,
remained intact in 1929 (pp. 59–61).
More of this follows. No reason existed why the Depression should have
increased international tension. In depressions most nations turn away from
foreign affairs. The Brüning Government in Germany, however, had to seek
successes abroad in order to counterbalance the hardships imposed by deflation
at home, and the Japanese, too, had a good case for invading Manchuria, because
their trading interests had suffered devastation there. “Men who are well off forget
their grievances; in adversity they have nothing else to think about” (pp. 61–2).
No real negotiations were possible at the 1932–3 disarmament conference because
the German Government needed a “sensational success.” Yet prevailing opinion
“rightly feared” the collapse of Germany, not German strength, in the midst of
the Depression; and actually the Reich remained “virtually disarmed” even when
Hitler came to power. On the other hand, one could not assure security for France
and equality of status for Germany at the same time. By insisting upon that fact,
the French nevertheless “fired the starting pistol for the arms race,” although
characteristically they failed to run it (pp. 67, 74, 77). As for Hitler himself, he
merely took over the policies of his predecessors and indeed of almost all
Germans. In principle and doctrine, he seemed “no more wicked and
unscrupulous than many other contemporary statesmen.” If truth be told, he
did not initially concern himself much with foreign affairs, but spent the bulk
of his time at Berchtesgaden, “dreaming in his old feckless way.” Still, showing
little talent other than his gift for patience, he somehow within two years had
broken the Franco-Polish alliance, foiled an “eastern Locarno,” made a first move
42 The Origins of the Second World War Reconsidered
during the 1970s. To transfer such a sum would have required genuine sacrifices.
But it would not have constituted an insuperable burden for a nation resolved
to limit domestic consumption sufficiently to meet the levy. It would not have
reduced Germany, in the picturesque phrase that Taylor borrows from Keynes,
to “a state of Asiatic poverty” (p. 44). In fact, the London Schedule marked the
abandonment of the unrealistic proposals bandied about by some of the British
and French delegates at the Paris peace conference and an adjustment of the
bill to Germany’s capacity to pay.
Taylor does not deal specifically with the London Schedule. He outlines with
only the barest of brush strokes the many disputatious conferences of 1920–2
at which the French endeavored to compel payment, the Germans sought to
evade it, and the British twisted uneasily between the two, increasingly coming
to favor concessions with the hope of reviving their export trade. But he suggests
correctly what the archives have since confirmed: that Poincaré occupied the
Ruhr as a last resort in 1923 in order to oblige Germany to meet its financial
obligations, and not with a view to promoting the disintegration of the Reich
(p. 50).14 He shows how France snatched defeat from the jaws of victory.
Germany found itself obliged to abandon government-sponsored “passive
resistance” in the Ruhr. France, however, could muster neither the financial
strength nor the will to go it alone. Almost inevitably, fresh negotiations within
an interallied framework led to a new de facto reduction of the reparations bill.
The moral seemed to be that the Allies could reach a stable accommodation
with Germany only by engaging the latter’s voluntary cooperation, and not
through compulsion (pp. 44, 50).15
Taylor does not traffic in economic statistics, but he perceptively observes
that Germany ultimately emerged as the net gainer through the financial
transactions of the 1920s (p. 44). Subsequent calculations reinforce his analysis.
The Dawes plan of 1924 provided a four-year partial moratorium and reduced
the projected reparations levy afterward to 3.3 percent of national income. The
Young plan of 1929 marked a further limitation of the Allied claim on German
resources, effectively to 2.6 percent. Moreover, the Reich never paid those sums
in full. In 1932 the Lausanne conference canceled reparations altogether. For
the whole period 1919–31, Germany transferred to the Allies, in cash and kind
together, an average of only 2.0 percent of national income. At the same time,
Germany experienced a windfall profit resulting from the devaluation of foreign-
owned mark-denominated assets during the 1919–23 inflation. Then, after 1931,
it defaulted on most private foreign investment. These items combined yielded
a unilateral transfer equal to a startling 5.3 percent of German national income
for 1919–31. On balance, the United States and, to a lesser extent, the European
Allies subsidized Germany during the Weimar era, and not the other way
round.16
So far Taylor has gotten it right – as his hero Evelyn Waugh might have said:
“Up to a point, sir!”17 Characteristically, he then reverses field. The foreign
subsidy of the German economy, he asserts, provided “little consolation to the
German taxpayer, who was not at all the same person as the German borrower”
44 The Origins of the Second World War Reconsidered
(p. 44).18 Hence the average German held the “more or less rational belief” that
reparations pointed down the road to ruin.19
The United States “complicated the problem,” Taylor contends, by insisting
on repayment of Allied war debts. Thus Allied taxpayers obtained little relief
from reparations because they saw the proceeds immediately transferred across
the Atlantic. In addition, reparations fanned international suspicion and
resentment all around. This happened in part because the French “cheated”: some
of them did not wish to be paid, and instead attempted to exploit the situation
so as to keep troops in the Rhineland or to ruin Germany forever. Yet ultimately
reparations did almost as much damage to democracy in France, where they
led people to lose faith in the political leadership, as they did across the Rhine.
More than any other issue, in sum, reparations “cleared the way for the Second
World War.” Economically, however, they produced but one effect after the
various capital flows were balanced against each other: “to give employment to
a large number of bookkeepers” (pp. 27, 32, 44–5, 47).20 Here we have, in short
compass, a distillation of every cliché that in Britain between the wars nourished
self-righteous sentiment on the part of subscribers to the fashionable intellectual
weeklies and readers of the Manchester Guardian. Each of the substantive claims,
however, turns out upon further scrutiny to be seriously misleading.
The careful reader will observe here some gratuitous pulling of Uncle Sam’s
beard. Actually the American debt settlements of 1923–26 required no more
than token payments from the continental Allies during the first decade, and
the British could not reasonably complain about an annuity claiming a mere
0.8 percent of their foreign investment portfolio. An examination of the
magnitudes involved does not substantiate the common references to a circular
flow of funds.21
But Taylor’s misconceptions go beyond technical economics. He appears to
have lost from sight the point of reparations in the first place – to repair the
damage wrought in a war for which Germany bore primary responsibility. He
starts from the premise that both sides in the conflict “found it difficult to define
their war aims,” that both fought only for victory (pp. 19–20). No one who had
assimilated Hans Gatzke’s findings in Germany’s Drive to the West could have
sustained that point of view, even in 1961.22 Studies published since then have
rendered it wholly implausible. The discoveries of Fritz Fischer and his students,
and the revelations from the diary of chancellor Bethmann-Hollweg’s assistant,
Kurt Riezler, have underscored the almost megalomaniacal annexationism of the
German leadership, civilian as well as military, from the “September program”
of 1914 down to the end of the war.23 Scholars of Allied war aims have uncovered
nothing comparable.24 Taylor acknowledges this disparity in the “Second
Thoughts” prepared for his 1963 edition. Indeed, with not atypical overstatement
he now accuses Bethmann-Hollweg of having nurtured more extreme aims than
had Hitler, because the former sought Lebensraum in the west as well as the east
(p. xxv).25 Unfortunately, Taylor gives no outward sign of thinking through the
implications of the new evidence. If the nations of Europe in 1914 had blundered
into war by mistake, a peace of reconciliation involving a sharing of financial
The End of Versailles 45
burdens might have helped bind up the wounds. If, however, the elites of the
new Republican Germany inherited the territorial appetites of their Wilhelminian
predecessors, such largesse would prove neither expedient nor wise.
At no point does Taylor clearly articulate the central issue: that whether
Germany paid reparations and how much it paid would largely determine the
European balance of power in the 1920s. The costs of the most destructive
conflict in history had already been incurred. Someone would have to pay for
the treasure spent, the foreign investment lost, the land laid waste, the ships
destroyed, the men cut down on the battlefield at the start of their most
productive years. Someone would have to care for the maimed, the widowed,
and the orphaned. War also spurs economic growth, of course. But in this case
the greatest development had taken place in Asia and the Americas, and the
burgeoning manufacturing capacity outside Europe would dislocate trade patterns
and make recovery for the belligerents more rather than less difficult.26
At the peace conference European statesmen had maneuvered to induce the
United States to subsidize victors and vanquished alike by canceling war debts
and purchasing reparations bonds on commercial markets. American officials,
however, had declined to fall in with those schemes. Some Wall Street bankers
acknowledged that their country’s new-found creditor status imposed special
responsibilities. But economically the United States remained remarkably self-
sufficient (compared, for example, with the situation after the Second World
War). Political leaders in Washington, mindful that their constituents had already
spent $40 billion on the war effort, recognized from the spring of 1919 on that
public sentiment would not tolerate disbursing any more.27 Europeans had to
face the question squarely: would the taxpayers, bondholders, savers, and
consumers of Germany assume the main burden of reconstruction, or would
their counterparts in France, Britain, and other Allied nations have to foot the
bill?
Taylor, with habitual eloquence, explains how reparations took on symbolic
importance for Germans outraged by the punitive aspects of the peace (pp. 44–
5). If the German government won its test of wills with the Allies and undermined
the reparations clauses of the Versailles treaty, it would assume a stronger moral
position in its efforts to revise the territorial provisions of the European settlement.
Taylor neglects to add, however, that reparations also involved the transfer of
real resources, which assumed critical importance in the boom-and-bust cycle
and the prolonged period of trade disruption that followed the war.
If France, most importantly, could obtain sufficient coal, coke, and capital
through reparations, it might hope to rebuild its devastated districts and to
promote economic revival without overstraining the nation’s rickety tax and
financial system. In the generation before 1914 Germany had rapidly pulled
ahead of France and Britain in industrial production. It had quadrupled its
steelmaking capacity – the key to military power. Could France, having secured
Lorraine minette ore and Saar coal through other Versailles treaty provisions,
now deploy its reparations entitlement to make a comparable leap forward?28
46 The Origins of the Second World War Reconsidered
Despite the deficient mechanisms in the Paris bureaucracy for shaping long-
term economic policy, certain French officials perceived this challenge clearly.
That is why Jacques Seydoux of the Foreign Ministry, for example, pressed so
hard in the postwar years for reparations in kind. That is why, far from trying
to “ruin Germany,” he sought in vain to find a formula that would link the coal,
steel, potash, and chemical industries of the two nations to their mutual
advantage.29
Taylor evinces no demonstrable interest in such technical negotiations, possibly
because he believes that a united Germany would have come to dominate Europe
again.30 Admittedly, the prospects for any outcome other than this appeared slim
from the beginning. France would have had to create an industrial infrastructure
enabling it to sustain permanently the diplomatic position secured for the moment
by a paper treaty. For that purpose reparations appeared indispensable, yet
insufficient. France would also have had to maintain an unshakable alliance with
Britain. But the economic interests of the two erstwhile Allies diverged.31 If the
United States refused to act as deus ex machina by providing loans to Germany,
how could one reconcile those interests? The British suffered from their own
form of “devastated areas.” Two million unemployed queued disconsolately for
the dole in the Midlands, on Clydeside, and in Wales. Once war passions cooled,
Britain could neither tolerate the transfer of cash reparations large enough to
reduce the German standard of living nor encourage deliveries in kind coupled
with Franco-German industrial linkages: both prospects might further reduce
the Central European market for British goods. British difficulties in fact derived
as much from the wrong mix of industries, unprofessional management, and
backward labour relations as they did from the state of world trade. But few
discerned this at the time.32 The Franco-British conflict over reparations steadily
deepened. As it festered, all hope of forcing compliance with the treaty as a whole
slipped away.
If Taylor were writing today, presumably he would make more of the
connections between economics and political power. In the past generation,
practitioners of diplomatic history have changed the face of the discipline by
investigating questions of that sort. Today it is a commonplace that reparations
determined the outcome of the struggle for power in Europe between France
and Germany. The war-debt controversy similarly figured as a central element
in the duel between Britain and the United States for dominance in world finance.
But, for all his brilliance, Taylor approaches diplomacy the old-fashioned way.
He therefore mentions the 1924 London conference only in passing. He focuses
instead on the Locarno meeting fourteen months later as the turning-point of
the interwar era (pp. 53–4). Today, looking back, we see clearly that the London
conference provided the first unmistakable augury of the demise of the Versailles
system.33
At the London conference France agreed to evacuate the Ruhr without any
significant quid pro quo. Germany might for a while meet its reduced obligations
under the Dawes plan or successor schemes. It would never subsequently pay
reparations on a scale sufficient to change power relationships. And when it
The End of Versailles 47
defaulted, France pledged, in effect, not to take unilateral action again. Moreover,
the last chance slipped by to strike a deal on coal and steel before Germany
recovered its tariff sovereignty and took advantage of its natural strengths to
dominate European heavy industry. In addition, France conceded a heretofore
disputed legal point with an important bearing on security. The Allies confirmed
that, whether Germany had met its disarmament obligations or not, the
Rhineland occupation clock had begun to run. The occupation henceforth
constituted a wasting asset. French troops would have to withdraw by 1935 in
any case. Why squander scarce military resources and exacerbate ill-will by
remaining until the final hour?
Taylor would have us believe that, after Germany had freely accepted the
Locarno accords, the French could think of no rationale for preserving
reparations or perpetuating one-sided disarmament of the Reich (p. 58). Actually,
few Frenchmen suffered from a bad conscience – that malady principally afflicted
their friends across the English Channel. But, through financial weakness and
national lassitude, they had already made the decisive concessions. When they
fussed about arrangements for leaving the Cologne zone in 1926, fumed about
minor conditions for terminating Allied military control in 1927, and jockeyed
tiresomely for reparations advantages before withdrawing the last skeletal
divisions from the Rhineland in 1929–30, they knew that they were engaged in
desperate rearguard actions.34
In discussing security problems of the 1920s Taylor offers his usual mixture
of good sense and artful obfuscation. He makes a surprisingly favorable
evaluation of the Versailles territorial settlements. The new borders, even in
eastern Europe, rested on the “principles of natural justice, as then interpreted”
(p. 26). Germany lost only land, Taylor adds, to which it held no entitlement
on national grounds. With facile insouciance he dismisses even the most bitterly
voiced German grievances: the Polish Corridor had a predominantly Slavic
population; the arrangements for railway communications across that area to
east Prussia seemed adequate; and the lost colonies had proven a drain rather
than a source of profit (p. 47).
Other historians do not on the whole confirm the claim that the mass of
Germans expressed indifference to an Austrian Anschluss in 1919 or that they
cared little about the fate of German-speaking minorities in the Sudetenland and
elsewhere (though undoubtedly German plenipotentiaries at the peace conference
had felt obliged to rank their objectives and concentrate their fire).35 But
ultimately it makes little difference how much German resentment against the
new boundaries originated in 1919 and how much developed later. No treaty
enforces itself. Certainly the Versailles treaty could not do so. The Congress of
Vienna a century earlier had reached a modicum of consensus by appealing to
the universally acknowledged principles of legitimacy and compensation, suitably
diluted with the equally familiar values of convenience and hypocrisy. The world
of 1919 comprised too many ideological divisions to hope for a similar consensus.
Only the application of superior force had won the war. Only the threat of force
48 The Origins of the Second World War Reconsidered
France’s eastern alliances (p. 54). We can now show more explicitly than Taylor
did that the treaties were meant to open the way for territorial revision in the
east. If France went to the assistance of its Polish ally, and Germany then attacked
France, the British guarantee of the western frontiers would not apply. The British
Cabinet expected the French, in consequence, to adopt a defensive military
strategy and to limit their eastern commitments. The Poles and the Czechs would
draw closer to the German economic orbit. Forces would be set in motion that
would gradually lead to peaceful change and so stabilize the continent.40
It is never easy, however, to manage peaceful change. British Foreign Office
planners did not reckon with the increased urgency of German revisionism in
the Depression; nor did they expect the Poles to ignore the relationships of power
and to become ever more obdurate as time went on. All the same, the British
government did not issue a blank check, in 1925 or later. Taylor does not stand
on firm ground in drawing a direct comparison between the aims of Stresemann
and Hitler. He goes beyond the evidence when he argues that Hitler merely
sought satellites and not territorial gains, and also when he asserts in
contradictory fashion that Hitler aspired to restore the great eastern conquests
of Brest–Litovsk, but that many in the west accepted this as natural, even
desirable (pp. 70, 80).41
If Taylor had wished to strengthen his case for the continuity of Weimar and
Nazi foreign policy, he would have done better to focus more attention on the
years 1930–3. The Brüning, Papen, and Schleicher cabinets did seek more radical
concessions in foreign policy than had their predecessors, both to provide
distraction from the pain of deflation and to head off the growing Nazi electoral
menace. We now have more proof than Taylor did of the quantum leap in the
military preparations of the Reichswehr in 1931–2.42 One rubs one’s eyes in
disbelief at Taylor’s portrayal of Arthur Henderson as a latter-day Austen
Chamberlain, anxious to “reconcile disarmament and security” and to use
disarmament as a “lever for increasing British commitments to France.”43 The
Cabinet Secretary, Sir Maurice Hankey, was nearer the mark when he noted in
September 1931 that British pacifism and Britain’s disarmament policy drew
inspiration less from idealism than from realization of the country’s exhaustion
and economic weakness, and that this derived in turn from “our insistence on
maintaining a much higher standard of living than our economic circumstances
justify.”44
Whatever the mix of motives, Great Britain, like “equality of rights” for
Germany throughout the Geneva disarmament conference of 1932–3. They both
tried to persuade France to reduce its forces to a level that Germany was already
planning to exceed. American leaders considered the balance of power immoral;
British statesmen failed to perceive where the balance really lay. Taylor makes
much of the slow pace of German rearmament in 1933–5 and speaks slightingly
of the “false alarms” issued by Churchill and others (p. 75).45 One wishes he
had coupled this with an analysis of unreadiness in the British and French armed
forces over the same period and an explanation of how Depression-era budgetary
constraints determined the glacial pace of improvement.46 Most Englishmen, he
50 The Origins of the Second World War Reconsidered
notes, believed in the early 1930s that “great armaments were themselves a cause
of war” (p. 64).47 Once they acted on that belief, the European structure erected
at the Paris peace conference was bound to crumble. And crumble it did.
None of the objections made to The Origins of the Second World War – least of
all those to the early chapters appeared to bother Taylor a whit. In reply to earlier
criticism, he pleaded the simplest of motives. He had sought to produce “a
straightforward piece of hack diplomatic history.”48 He had tried to write without
thinking that he was English, or a radical or a Socialist. He took no delight,
“impish or otherwise,” in shocking readers; he just put down what seemed to
him right without worrying whether it would appear orthodox or not. If
inconsistencies turned up, they usually resulted from his having learned better
afterward.49 Doubtless those explanations represent much of the truth. The
French Socialists used to remark about Aristide Briand that he was “so guileful
one ought not to believe the opposite of what he says.”50 Yet elsewhere Taylor
observes, half in jest, that we have left behind the era of Ranke: nowadays we
read diplomatic history “for purposes of entertainment.”51 If the office of history
is to amuse, then Taylor stands without peer. Some four decades after publication
of The Origins of the Second World War, students still read the book avidly. No doubt
they will continue to do so far into the future, when the learned rebuttals of
critics lie moldering on library shelves. Such scholarly longevity speaks for itself.
André Gide, when asked whom he considered the greatest French poet, replied,
“Victor Hugo, hélas.” Taylor’s sternest critics will scarcely improve the line.
Notes
1 A. J. P. Taylor, The Origins of the Second World War, London, 1961, p.28. Citations in
this essay are of the latest American edition (New York, 1983), which uses arabic
pagination as in the original, but employs roman pagination for the 1963 Foreword,
“Second thoughts,” and the “Preface for the American reader” (dating from 1966).
2 See Taylor’s reference in Origins, p. 234; Maitland on after-mindedness in H. M.
Cam (ed.), Selected Historical Essays of F. W. Maitland, London, 1957, p. xix.
3 Despite his nonchalance about certain minutiae, Taylor conceived himself as writing
squarely within the empirical-history tradition. He proffered no theoretical challenge
to objectivity on the model later made popular by the “postmodernists.” Compare
A. Megill (ed.), Rethinking Objectivity, Durham, NC, 1994, with R. J. Evans, In Defence
of History, London, 1997.
4 E. M. Robertson (ed.), The Origins of the Second World War: Historical Interpretations,
London, 1971; W. R. Louis (ed.), The Origins of the Second World War: A. J. P. Taylor
and his Critics, New York, 1972.
5 A. J. P Taylor, A Personal History, New York, 1983, p. 233.
6 The abbreviated run of Documents diplomatiques belges, 1920–1940 reached completion
in 1966 but Her Majesty’s Stationery Office did not complete Series 1 and 1A of
the Documents on British Foreign Policy (covering 1919–29) until 1986. Series A and B
of Akten zur deutschen auswärtigen Politik (covering the years 1918–33) did not all appear
until 1995. The Quai d’Orsay began to publish Documents diplomatiques français for
the 1920s only in 1998, after a private German team under S. Martens took the
initiative with Documents diplomatiques français sur l’Allemagne, 1920, Bonn, 1992–93.
The original Russian series, Dokumenty vneshnei politiki SSSR, Moscow, 1959–77,
The End of Versailles 51
revealed remarkably little; the much franker volumes on 1939–41 appeared only
in 1995. The Italian, Swiss, and Austrian official series remain works in progress.
7 Taylor, Personal History, pp. 23–4.
8 Taylor, Personal History, pp. 125–8.
9 Taylor, Personal History, p. 60.
10 On Russian policy, see the standard synthesis by A. B. Ulam, Expansion and Coexistence
revised edn, New York, 1974; J. Braunthal, History of the International, vol. 2: 1914–
1943, New York, 1967; R. Pipes, The Bolshevik Regime, New York, 1994, and The
Unknown Lenin, New Haven, CT, 1996; R. K. Debo, Survival and Consolidation,
Montreal, 1992; also the meticulously compiled evidence on Soviet subversion in
Germany in W. T. Angress, Stillborn Revolution, Princeton, NJ, 1963. E. H. Carr,
Twilight of the Comintern, 1930–1935, New York, 1982, still portrays Soviet policy as
defensive, and some younger scholars who wrote before the opening of the Soviet
archives agree. See, for example, J. Haslam, Soviet Foreign Policy, 1930–1933, New
York, 1983. Helpful works on co-operation and competition in Anglo-American
relations include: C. Parrini, Heir to Empire, Pittsburgh, PA, 1969; M. J. Hogan,
Informal Entente, Columbia, MO, 1977; M. G. Fry, Illusions of Security, Toronto, 1972;
S. V. O. Clarke, Central Bank Cooperation, 1924–1931, New York, 1967; B. J. M.
McKercher, The Second Baldwin Government and the United States, 1924–1929,
Cambridge, 1984; and B. J. M. McKercher (ed.), Anglo-American Relations in the 1920s,
Edmonton, 1990. K. Burk, Britain, America and the Sinews of War, 1914–1918, London,
1985, provides background. R. N. Kottman, Reciprocity and the North Atlantic Triangle,
1932–1938, Ithaca, NY, 1968; C. A. MacDonald, The United States, Britain and
Appeasement, 1936–1939, New York, 1981; B. M. Rowland, Commercial Conflict and
Foreign Policy, New York, 1987; and D. Reynolds, The Creation of the Anglo-American
Alliance, 1937–1941, Chapel Hill, NC, 1981, carry the story onward. On the Far
East, the best works include A. Iriye, After Imperialism, Cambridge, MA, 1965; M.
D. Kennedy, The Estrangement of Great Britain and Japan, 1917–1935, Manchester,
1969; I. H. Nish, Alliance in Decline, London, 1972; R. Dingman, Power in the Pacific,
Chicago, IL, 1976; and J. Neidpath, The Singapore Naval Base and the Defence of Britain’s
Eastern Empire, 1919–1941, Oxford, 1981.
11 Quoted in E. Eyck, Bismarck and the German Empire, London, 1958, pp. 77–8.
12 Compare J. M. Keynes, The Economic Consequences of the Peace, London, 1919, and
the still less temperate A Revision of the Treaty, London, 1922 with E. Mantoux, The
Carthaginian Peace, London, 1946; reprinted Pittsburgh, PA, 1965. On Keynes’s
private motives for embracing the German cause, see S. A. Schuker, “The collected
writings of John Maynard Keynes,” Journal of Economic Literature, vol. 18, 1980,
pp.124–6; N. Ferguson, “Keynes and the German inflation,” English Historical Review,
vol. 110, 1995, pp. 368–91; and G. Martel, “The prehistory of appeasement:
Headlam-Morley, the peace settlement and revisionism,” Diplomacy and Statecraft, vol.
9, 1998, pp. 242–65.
13 On the London Schedule, Taylor could have consulted, but apparently did not, E.
Weill-Raynal, Les Réparations allemandes et la France, 3 vols, Paris, 1947, esp. vol. 1,
pp. 618–702. C. Holtfrerich, Die deutsche Inflation 1914–1923, Berlin, 1980, p. 221,
summarizes recent work reconstructing German national income for the period. S.
Marks provides the soundest orientation to scholarship on the reparations problem
in: “Reparations reconsidered,” Central European History, vol. 2, 1969, pp. 356–65;
“The myths of reparations,” Central European History, vol. 11, 1978, pp. 231–55; and
“In smoke-filled rooms and the Galerie des Glaces: reparation at the Paris peace
conference,” in M. Boemeke, G. Feldman, E. Glaser (eds), The Treaty of Versailles: A
Reassessment after 75 Years, Cambridge, 1998, pp. 337–70.
14 See S. A. Schuker, The End of French Predominance in Europe, Chapel Hill, NC, 1976,
pp. 20–6; J. Bariéty, Les Relations franco-allemandes aprés la première guerre mondiale, Paris,
1977, pp. 101–8; W. A. McDougall, France’s Rhineland Diplomacy, Princeton, NJ, 1978,
52 The Origins of the Second World War Reconsidered
pp. 214–49; also J. F. V. Keiger, Raymond Poincaré, Cambridge, 1997. For a summary
of the copious literature on passive resistance, see J.-C. Favez, Le Reich devant
l’occupation franco-belge de la Ruhr en 1923, Geneva, 1969.
15 On negotiation of the Dawes plan and French financial problems, see Schuker, French
Predominance, pp. 31–231; and Bariéty, Relations, pp. 289–320.
16 For economic analysis and statistics, see S. A. Schuker, American “Reparations” to
Germany, Princeton, NJ, 1988; also A. Klug, The German Buybacks, 1932–1939,
Princeton, NJ, 1993. Note that certain economists remain unconvinced: e.g. B.
Eichengreen, Golden Fetters, New York, 1992; a minority of historians, including B.
Kent, The Spoils of War, Oxford, 1989, continues to recycle older interpretations as
well.
17 Quoted in A. J. P. Taylor, Beaverbrook, London, 1972, p. 678n.
18 Precisely what Taylor means by this remains puzzling, since the German states and
municipalities – the greatest borrowers in 1925–28 – would otherwise have had to
draw upon tax revenue, directly or indirectly, for the services that they provided.
19 G. D. Feldman, The Great Disorder, Oxford, 1993, shows that, whatever the economic
balance, the German people paid a tremendous social price both contemporaneously
and later for adopting inflationary policies after the First World War. But it does
not follow that the actual reparations burden, as opposed to the subjective perception
of that burden, forced Berlin to make the fiscal and monetary choices that
perpetuated the inflation.
20 Keynes produced the original formulation: “The engravers’ dyes, the printers’ forms
are busier. But no one eats less, no one works more.” See “The progress of the
Dawes scheme [entry for September 11, 1926],” in E. Johnson (ed.), The Collected
Writings of John Maynard Keynes, London, 1978, vol. 18, pp. 277–82.
21 See B. D. Rhodes, “The United States and the war debt question, 1917–1934,” Ph.D.
thesis, University of Colorado, 1965, and “Reassessing uncle Shylock,” Journal of
American History, vol. 55, 1969, pp. 783–803; M. P. Leffler, “The origins of
Republican war debt policy, 1921–1923,” Journal of American History, vol. 59, 1972,
pp. 585–601. On linkage of the debts with other issues, note R. A. Dayer, “The
British war debts to the United States and the Anglo-Japanese alliance, 1920–1923,”
Pacific Historical Review, vol. 45, 1976, pp. 569–95; E. Glaser, “Von Versailles nach
Berlin,” in N. Finzsch et al. (eds), Liberalitas, Stuttgart, 1992, pp. 319–42; E. Wandel,
Die Bedeutung der Vereinigten Staaten von Amerika für das deutsche Reparationsproblem 1924–
1929, Tübingen, 1971; and especially the magisterial work by W. Link, Die
amerikanische Stabilisierungspolitik in Deutschland 1921–1932, Düsseldorf, 1970.
22 H. W. Gatzke, Germany’s Drive to the West, Baltimore, MD, 1950.
23 F. Fischer, Griff nach der Weltmacht, Düsseldorf, 1961, and Krieg der Illusionen,
Düsseldorf, 1969; trans. as Germany’s Aims in the First World War and War of Illusions,
New York, 1967 and 1975; I. Geiss, Der polnische Grenzstreifen, 1914–1918, Lübeck
and Hamburg, 1960; K. Riezler, Tagebücher, Aufsätze, Dokumente, ed. K. D. Erdmann,
Göttingen, 1972; H. Afflerbach, Falkenhayn, Munich, 1994. For attacks on the
Sonderweg thesis, however, see D. Calleo, The German Problem Reconsidered, Cambridge,
1978; also D. Blackbourn and G. Eley, The Peculiarities of German History, Oxford,
1984.
24 Standard works in this field include: V H. Rothwell, British War Aims and Peace
Diplomacy, 1914–1918, Oxford, 1971; D. Stevenson, French War Aims against Germany,
1914–1919, Oxford, 1982, and The First World War and International Politics, Oxford,
1988; M. Palo, “The diplomacy of Belgian war aims during the First World War,”
Ph.D. thesis, University of Illinois, 1977; S. Marks, Innocent Abroad, Chapel Hill,
NC, 1981, pp. 5–102; also the magisterial study of economic war aims by G. Soutou,
L’Or et le sang, Paris, 1989. On the earlier period, see D. C. B. Lieven, Russia and the
Origins of the First World War, London, 1983; and J. F. V. Keiger, France and the Origins
of the First World War, London, 1983.
The End of Versailles 53
Problem, 1924–1929, New York, 1987; and F. Knipping, Deutschland, Frankreich, and
das Ende der Locarno-Ära, 1928–1931, Munich, 1987. On military aspects, see M.
Salewski, Entwaffnung und Militärkontrolle in Deutschland, 1919–1927, Munich, 1966;
and J. M. Hughes, To the Maginot Line, Cambridge, MA, 1971. Several essays in G.
Schmidt (ed.), Konstellationen internationaler Politik, 1924–1932, Bochum, 1983, also
address these questions.
35 Given the confusion in Germany during 1918–19, and the illusions of German
statesmen about what was achievable at the peace conference, it is not surprising
that scholars have failed to reach a consensus on the importance of those objectives.
K. Schwabe provides the most reliable general overview in Deutsche Revolution und
Wilson-Frieden, Düsseldorf, 1971; revised version and ET: Woodrow Wilson, Revolutionary
Germany, and Peacemaking, 1918–1919, Chapel Hill, NC, 1985. For public opinion
during the subsequent period, see C. Höltje, Die Weimarer Republik und das
OstlocarnoProblem 1919–1934, Würzburg, 1958; and E. Hölzle (ed.), Die deutschen
Ostgebiete zur Zeit der Weimarer Republik, Cologne, 1966. C. M. Kimmich, The Free
City, New Haven, CT, 1968, covers Danzig; F. G. Campbell, Confrontation in Central
Europe, Chicago, IL, 1975, and R. Jaworski, Vorposten oder Minderheit?, Stuttgart, 1977,
deal with the German minority in Czechoslovakia from different points of view.
N. Krekeler, Revisionsanspruch und geheime Ostpolitik der Weimarer Republik, Stuttgart, 1973,
treats assistance to minorities in Poland; K.-H. Grundmann, Deutschtumpolitik zur Zeit
der Weimarer Republik, Hanover, 1977, addresses the Baltic. One cannot speak with
categorical assurance even about policy toward Austria: see S. Suval, The Anschluss
Question in the Weimar Era, Baltimore, MD, 1974, pp. 3–20; and A. D. Low, The
Anschluss Movement, 1918–1919 and the Paris Peace Conference, Philadelphia, PA, 1974.
36 On the Anglo-American guarantee, see L. Yates, The United States and French Security,
1917–1921, New York, 1957. The fundamental books on the eastern alliances, P.
Wandycz, France and Her Eastern Allies, 1919–1925, Minneapolis, MN, 1962, and
The Twilight of the French Eastern Alliances, 1926–1936, Princeton, NJ, 1988, emphasize
the importance of the Polish and Czech connections; N. Jordan, The Popular Front
and Central Europe, Cambridge, 1992, bathetically laments their abandonment. But
numerous recent articles in Relations internationales, the Revue d’histoire de la deuxième
guerre mondiale, and other journals have pointed to the frustrations experienced by
the French in training armies and trying to turn a profit on their investments in
eastern Europe. P.-E. Tournoux, Défense des frontières, Paris, 1960, examines the
successive strategic plans of the French army and confirms what Taylor suspected
– that they became steadily more defensive in the 1920s.
37 See J. M. King, Foch versus Clemenceau, Cambridge, MA, 1960; S. A. Schuker, “The
Rhineland question: west European security at the Paris peace conference,” in
Boemeke et al. (eds), The Treaty of Versailles, pp. 275–312; and McDougall, Rhineland
Diplomacy. There exists a large literature, much of it contentious in tone, evaluating
the seriousness with which the French at various points viewed the Rhineland
option. See K. D. Erdmann, Adenauer in der Rheinlandpolitik nach dem Ersten Weltkrieg,
Stuttgart, 1966, still the most solid account despite later revelations about Erdmann’s
career under the Third Reich; E. Bischof, Rheinischer Separatismus 1918–1924, Bern,
1969; Centre de Recherches Relations Internationales de l’Université de Metz,
Problèmes de la Rhénanie, 1919–1930/Die Rheinfrage nach dem Ersten Weltkrieg, Metz, 1975;
G. Steinmeyer, Die Grundlagen der französischen Rheinlandpolitik, 1917–1919, Stuttgart,
1979; H. Köhler, Novemberrevolution und Frankreich, Düsseldorf, 1980, and Adenauer
und die rheinische Republik, Opladen, 1986; H. E. Nadler, The Rhenish Separatist Movements
in the Early Weimar Republic, New York, 1987; S. Jeannesson, Poincaré, La France et la
Rühr, Strasbourg, 1998; and S.A. Schuker, “Bayern und der rheinische Separatismus,”
Jahrbuch des historischen Kollegs 1997, pp. 75–111.
38 Taylor, Beaverbrook, p. xiii.
The End of Versailles 55
39 For useful discussions of Stresemann’s aims, as set out in his September 7, 1925
letter to Crown Prince Wilhelm, see Hans Gatzke, Stresemann and the Rearmament of
Germany, Baltimore, MD, 1954; and Gaines Post, The Civil–Military Fabric of Weimar
Foreign Policy, Princeton, NJ, 1973. W. Weidenfeld, Die Englandpolitik Gustav Stresemanns,
Mainz, 1972, and C. Baechler, Gustave Stresemann, Strasbourg, 1996, mount the
barricades for the defense. W. Michalka and M. M. Lee (eds), Gustav Stresemann,
Darmstadt, 1982, group twenty essays by leading specialists who have elucidated
various aspects of Stresemann’s policies.
40 For the explicit formulations by Cabinet members and by the Foreign Office legal
adviser, see Schuker, French Predominance, pp. 389–90; also note the discussion in
Jacobson, Locarno Diplomacy, pp. 12–44; and S. E. Crowe, “Sir Eyre Crowe and the
Locarno pact,” English Historical Review, vol. 87, 1972, pp. 49–74.
41 On the growing rigidity of Polish policy, see R. Debicki, The Foreign Policy of Poland,
1919–1939, New York, 1962; J. Korbel, Poland Between East and West, Princeton, NJ,
1963; and L. Radice, Prelude to Appeasement: East Central European Diplomacy in the Early
1930s, Boulder, CO, 1981. On French frustration with the Poles, see the review
article by H. Rollet, “Deux mythes des relations franco-polonaises entre les deux
guerres,” Revue d’histoire diplomatique, no. 96, 1982, pp. 225–48. G. Weinberg, The
Foreign Policy of Hitler’s Germany: Diplomatic Revolution in Europe, 1933–36, Chicago, IL,
1970, esp. pp. 1– 24, puts to rest in elegant fashion the notion of Hitler as a
traditional diplomat. Neither J.-B. Duroselle, La Décadence 1932–1939, Paris, 1979,
nor any of the monographs on which he based his synthesis, offer support for
Taylor’s quixotic notion that “most” Frenchmen came to favor German dominance
of Russia on the model of the 1918 Brest–Litovsk treaty.
42 On foreign and military policy of the Reich in 1930–32, see particularly A. Rodder,
Stresemanns Erbe: Julius Curtius und die deutsche Aussenpolitik, 1929–1931, Paderborn,
1996; E. W. Bennett, Germany and the Diplomacy of the Financial Crisis, 1931, Cambridge,
MA, 1962 (1931), and German Rearmament and the West, 1932–1933, Princeton, NJ,
1979. The seriousness of German rearmament in those years remains disputed.
Much depends on whether one examines industrial mobilization or immediate
readiness, and short- or long-term planning. G. Meinck, Hitler und die deutsche
Aufrüstung, 1933–1937, Wiesbaden, 1959, minimizes early efforts under Schleicher;
M. Geyer, “Das Zweite Rüstungsprogramm, 1930–1934,” Militärgeschichtliche
Mitteilungen, vol. 16, 1975, pp. 125–72, and Aufrüstung oder Sicherheit: Die Reichswehr
in der Krise der Machtpolitik, 1924–1936, Wiesbaden, 1980, esp. pp. 237–362, stresses
elements of continuity. Note also E. W. Hansen, Reichswehr und Industrie, Boppard
am Rhein, 1981, which explores industrial rearmament; and T. Vogelsang, Reichswehr,
Staat und NSDAP, Stuttgart, 1962, which concentrates on political questions. G.
Schulz’s Introduction to I. Maurer and U. Wengst (eds), Politik und Wirtschaft in der
Krise: Quellen zur Ära Brüning, 2 vols, Düsseldorf, 1980; vol. 1, pp. ix–civ, reviews
the growing literature on Brüning’s foreign economic policy; K. Borchardt, Wachstum,
Krisen, Handelsspielräume der Wirtschaftspolitik, Göttingen, 1982, pp. 165–224, and W.L.
Patch, Heinrich Brüning and the Dissolution of the Weimar Republic, Cambridge 1998,
emphasize Brüning’s limited maneuvering room.
43 The balance of scholarly opinion sheds a less unkind light on Ramsay MacDonald
than it does on Henderson, although there is ample blame to go around. See D.
Carlton, MacDonald versus Henderson: The Foreign Policy of the Second Labour Government,
London, 1970; D. Marquand, Ramsay MacDonald, London, 1977; and H. Winkler,
Paths Not Taken: British Labour and International Policy in the 1920s, Chapel Hill, NC,
1994.
44 S. Roskill, Hankey: Man of Secrets, London, 1972, vol. 2: 1919–1931, pp. 544–5.
45 The debate continues over the pace of German rearmament before 1936. Still basic
is the semi-official history: W. Deist, M. Messerschmidt, H.-E. Volkmann, and W.
Wette, Das Deutsche Reich und der Zweite Weltkrieg, Stuttgart, 1979; vol. 1: Ursachen
56 The Origins of the Second World War Reconsidered
und Voraussetzungen der deutschen Kriegspolitik. K.-J. Müller, General Ludwig Beck, Boppard
am Rhein, 1980, shows that the Reichswehr pushed Hitler as much as the other
way round. W. K. Wark, The Ultimate Enemy, 1933–1939, Ithaca, NY, 1985, assesses
the perceptions of British intelligence; R. J. Overy, “German air strength, 1933 to
1939,” Historical Journal, vol. 27, 1984, pp. 465–71, calls attention to the difficulties
of using nominal figures to evaluate German offensive strength in the air.
46 For Britain, see N. H. Gibbs, Grand Strategy, London, 1976, vol. 1; B. Bond, British
Military Policy between the Two World Wars, Oxford, 1980, esp. pp. 161–214; G. C.
Peden, British Rearmament and the Treasury, 1932–1939, Edinburgh, 1979; R. P. Shay,
British Rearmament in the Thirties, Princeton, NJ, 1977; and R. Middleton, Towards the
Managed Economy, London, 1985. For France, see M. Vaïsse, Sécurité d’abord, Paris,
1981, esp. pp. 597–615; R. J. Young, In Command of France, Cambridge, MA, 1978,
pp. 13–75; R. Frankenstein, Le Prix du réarmement français, Paris, 1982; and M. S.
Alexander, The Republic in Danger, Cambridge, 1992.
47 For a notable example of the pro-disarmament mentality of the era, see P. Noel-
Baker, The First World Disarmament Conference, 1932–1933, and Why It Failed, New York,
1979. M. Ceadel, Pacifism in Britain, 1914–1945, Oxford, 1980, sums up the Zeitgeist
shrewdly. N. Ingram, The Politics of Dissent: Pacifism in France, 1919–1939, Oxford,
1991, demonstrates that Gallic contemporaries enjoyed no natural immunity to this
illusion.
48 Taylor, Personal History, p. 234.
49 Taylor to E. B. Segal, November 21, 1964, in Louis, Origins, pp. 26–7.
50 Quoted in D. B. Goldey, “The disintegration of the Cartel des Gauches and the politics
of French Government finance, 1924–1928,” DPhil. thesis, Oxford University, 1961,
p. 110.
51 Cited by E. B. Segal in Louis, Origins, p. 14.
4 Mussolini and the myth of
Rome
Alan Cassels
The Second World War left behind ample scope for anguished historiographical
debate in all the major powers involved in the conflict.1 This has proved
particularly true in Germany and Italy where historical disputes have followed
an interestingly parallel course, extending well beyond the fact that the two
countries were Axis partners in the war itself. Germany’s notoriously
“unmasterable past” raised the issue of Germany’s collective guilt for Nazism.2
To what extent did the origins of Nazism lie in German history and culture?
Applied to the field of international relations, the question bore on Nazi
expansionism after 1933. Was this simply a continuation of traditional German
foreign policy, or was it the working-out of a peculiar Nazi weltanschauung? In
answer to this foreign-policy question the Nuremberg Trials put the spotlight
on Hitler, his immediate circle and Nazi ideology. Not until the 1960s did the
scholarly search for the roots of the Third Reich’s international aggression really
switch to the pre-Nazi period. In large measure, this was due to the work of
two historians. Fritz Fischer, first implicitly and then avowedly, connected German
annexationist aims in the First World War to the Nazi search for eastern
Lebensraum a generation later.3 A. J. P. Taylor reached best-seller fame by
provocatively depicting the Führer as no more than a conventional if unusually
opportunistic German statesman.4 Since the 1960s, therefore, no account of Nazi
foreign policy can ignore the diplomatic legacy bequeathed by the Bismarckian,
Wilhelmine and Weimar regimes.
Historical literature on Italy’s role in the Second World War has undergone
a similar metamorphosis. After Italy’s entry into war in 1940, Winston Churchill
singled out Benito Mussolini as the “one man and one man only [who] was
resolved to plunge Italy…into the whirlpool of war.”5 Although at the time no
more than a propaganda ploy to divide the Fascist Duce from the Italian people,
it epitomized the view that Italy’s recent belligerence could be laid exclusively
at the door of the Fascist regime. After the war some Italians built on this idea
to develop the “parenthesis” argument, namely, that Fascism was a disjunctive
interlude in the “real history” of united Italy. Fascism, wrote Benedetto Croce,
58 The Origins of the Second World War Reconsidered
its wake he afforded a comfortable sanctuary for some of the Nazi fugitives, most
famously Hermann Goering.20 Despite the ignominious outcome of Hitler’s first
bid for power, Fascist Italy continued to pay close attention to the Nazi
movement,21 for on two counts at least the Nazis stood out and apart from run-
of-the-mill German nationalists. They described themselves as fascist, admittedly
a nebulous term, but the presumptive imitation flattered Mussolini’s thirsty ego.22
Of much greater importance, however, was Hitler’s unequivocal stand on the
Alto Adige. The Paris peace conference’s delimitation of the Austro-Italian border
at the Brenner Pass had consigned more than 200,000 German-speaking former
Habsburg subjects to the mercy of an intolerant Italian state. Hence German
nationalists, not excepting Hitler, referred to the region south of the Brenner by
its German name of the South Tyrol. But whereas all other pan-Germans put
the South Tyrol/Alto Adige high on the irredentist list for inclusion in a future
Reich, Hitler consistently repudiated any such objective. In Mein Kampf he
castigated imperial Germany’s diplomacy before 1914 for allowing Italy to slip
out of the Triple Alliance; in the future, Italian friendship would have to be
bought by abandoning the South Tyrol. This line of argument he reiterated
publicly in speeches and in the Völkischer Beobachter, as well as privately to
Mussolini.23
Readiness to traffic with Europe’s revisionist groups did not stop with
Germany. In the second half of the 1920s Mussolini turned to discontented
elements in the Danube valley in order to weaken the Little Entente and offset
French influence there, and he also began his long association with the Croatian
separatists in Yugoslavia.24 In 1928 he endorsed international revisionism before
the Italian senate: “I have sometimes had occasion to point out that the treaties
of peace are not eternal....Is there anyone who dares argue that the peace
treaties…are a work of perfection?”25 Yet despite this public avowal, most of the
Duce’s early revisionist activity was conducted sub rosa – even behind the back
of Italy’s diplomats. The Italian ambassador in Berlin, for one, was less than
fully aware of all the ramifications of Mussolini’s German policy, much of which
was handled by his man in Berlin, Major Renzetti.26 While the majority of Italy’s
foreign ministry officials were nationalistic enough to imagine they could use
the “wild” Mussolini to frighten the western states into concessions, they were
less ready to contemplate deserting the victorious First World War coalition.27
But their influence was circumscribed, and it declined with the passage of time.
The Duce, whether or not he held the foreign-ministry portfolio himself, always
kept the strings of Fascist foreign policy in his own hands. And, from the outset,
Mussolini was prepared to employ a diplomatic strategy as reckless as his goals
were extravagant; both methods and ambition placed him on the extremist wing
of Italian nationalism.
The pace at which Fascist Italy moved to realize its Roman empire was
inevitably contingent on the situation at home. For several years Mussolini’s first
concern was to secure his position in Rome, which he did in two ways. First, by
constructing, in the wake of the Matteotti affair of 1924, a one-party
dictatorship.28 Second, by agreeing a series of modi operandi with Italy’s old power
Mussolini and The Myth of Rome 61
structure – the latifondisti, the Confidustria, the armed forces, the monarchy and,
above all in 1929, the Catholic Church. These alliances resulted in the
transformation of Italian Fascism, in the phraseology of Renzo De Felice, from
a movement into a regime, denoting a shift from radical innovation to
bureaucratic stultification.29 Indeed, the PNF’s abandonment of any serious
program of social and economic change was glaringly revealed when the Great
Depression brought widespread distress to Italy.30 Having betrayed its pristine
revolutionary rhetoric, then, Fascism by the opening of the 1930s lacked any
raison d’être other than imperial expansion in conformity with the myth of Rome.
Accordingly, in 1932 Mussolini ordered that military plans be drawn up for the
conquest of Ethiopia, the east-African territory which had successfully resisted
Italy’s civilizing mission in the 1890s – the most humiliating reversal experienced
by any colonial power in the scramble for empire. The Duce’s directive thus
stemmed from a long Italian imperialist tradition and, it is worth noting, was
issued before Hitler reached office. Although the Ethiopian venture would end
by entrapping Mussolini in Hitler’s net, Fascist Italy was at first inhibited from
pursuing its planned African campaign by alarm at the resurgence of a German
nationalism increasingly located in the Nazi movement.
As one who had anticipated the renewal of German nationalist pretensions,
Mussolini was appropriately the first to advance a plan to meet the exigency.
His Four-Power Pact, floated in 1932, predicated embracing – and controlling –
Germany within a new concert of Europe (to the detriment of the League of
Nations, always a Fascist bête noire). As inducements, some revision of the postwar
settlement was proffered; parity in armaments was openly touted, and adjustment
of Germany’s borders hinted at. But this revisionist dimension offended France
and its eastern allies, and in the end killed Mussolini’s project.31
The failure of the Four Power Pact threw Mussolini back on his own devices
for coping with the German menace. The touchstone of Italo-German relations
was the double-edged problem of Austrian integrity and the sanctity of Italy’s
Brenner frontier. Whatever promise Hitler might make about the South Tyrol,
his program envisaged Anschluss, which would see German troops camped on
the Brenner. Italy’s palpable need to preserve an independent Austria as a buffer
had prompted Mussolini for some years to intervene surreptitiously in internal
Austrian politics in the hope of installing and patronizing a congenial government
in Vienna. By 1934 his preferred client was Engelbert Dollfuss, the Austrian
conservative and quasi-fascist chancellor. It was with the Duce’s encouragement
that Dollfuss attacked the Austrian Socialists in February, precipitating a state
of tension in his country which, four months on, led to the first Hitler–Mussolini
meeting in Venice. However, the encounter did nothing to dispel the
incompatibility of the two leaders’ aims in Austria, in part because Hitler went
on at great length in German without benefit of an interpreter.32 The innate
misunderstanding sprang into the open almost at once when, in July, the Viennese
Nazis murdered Dollfuss in a coup intended to bring about Anschluss. Mussolini
had the embarrassing task of informing Frau Dollfuss, who was his guest in
Rome, of her husband’s assassination. He relieved his feelings politically by
62 The Origins of the Second World War Reconsidered
War, widely if naively adjudged a contest between these two “-isms,” supplied a
perfect example. When Mussolini himself, on the very morrow of the Ethiopian
war, thrust his country into the Spanish conflict, geopolitical motives combined
nicely with propaganda against Soviet “popular frontism.” The same ideological
rationale came into play the following year when Italy joined the Anti-Comintern
Pact.45 In a more positive ideological context, Mussolini and Hitler had no trouble
recognizing each other as fellow-fascists. A. J. P. Taylor spotted the quintessence
of the new Italo-German relationship to which Mussolini gave the name “Axis”:
it was “ideological similarity.”46
Mussolini’s personality, social Darwinism and ideological thinking, then,
combined to create Fascist Italy’s pro-German orientation. In addition. it is worth
observing that history was here repeating itself. In 1882 Italy had joined
Bismarck’s Germany and Austria–Hungary in the Triple Alliance, and
subsequently, during the Crispi period, many of Italy’s elite out of admiration
for German power, wealth, and culture copied Teutonic fashions.47 Half a century
on, another political rapprochement between Rome and Berlin was followed by
Italian affectation of German manners and mores. In the latter instance,
unfortunately, one element of Nazi-German culture imported by Italy was
antisemitism. Not at the behest of Hitler but in manifest imitation, Mussolini
introduced race laws into Italy in 1938.48 The wheel had come full circle in Italy’s
search to fulfill the myth of Rome. Partnership with Germany had been
superseded by alliance with the western powers in the First World War; now it
was back to collaboration with Germany once more.
Mussolini was loath to admit the true extent of Italy’s absorption into the
Nazi orbit, and it is just possible that he sincerely believed his stance to have
been a neutral one. He was, after all, capable of immense self-deception to the
point of believing his own propaganda.49 At any rate, this circumstance has
opened the door to a good deal of historiographical discussion about the nature
and degree of Mussolini’s commitment to Hitler. Some have claimed that the
Duce actually sought to preserve “equidistance” and operate as a peso determinante
(decisive weight) between the western states and Nazi Germany up until a few
months before entering the Second World War in June 1940. This school of
thought takes its cue from the work of Renzo De Felice, the acknowledged dean
of Mussolinian studies, although he has been accused of immersing himself so
deeply in contemporary Fascist documentation as to have swallowed some of
its more self-serving commentary.50 Any thesis pleading the case for a Mussolinian
equidistant strategy of necessity lays stress on two Anglo-Italian agreements,
signed in January 1937 and April 1938, which bound the two powers to uphold
the status quo in the Mediterranean and the Middle East.51 Called – in the style
of the day – “Gentleman’s Agreements,” these accords, taken at face-value,
implied that Mussolini was appeasable. Predictably, there were those in the west
who, in light of the greater peril of aggression from Nazi Germany, were eager
enough to placate Mussolini.52 Thus, some historians have argued that the
western powers missed a chance to build on these agreements, appease Mussolini
further, and in the process break the Rome–Berlin Axis.53 A variation on this
66 The Origins of the Second World War Reconsidered
theme is the contention of one Italian scholar that the failure to attain a lasting
Anglo-Fascist détente should be attributed to the British government’s negotiating
in bad faith and denial of Italy’s legitimate interests.54
But the idea that Mussolini was open to western overtures, that in fact he
was striving to play an even-handed game between the democracies and Hitler,
constitutes a minority opinion.55 For the reality was that Fascist diplomacy after
Ethiopia was anything but impartial. Even the Gentleman’s Agreements were
not axiomatic pro-western gestures. It is significant that Rome consistently rejected
a triple pact which would include France as well as Britain.56 In this context,
the agreements can be construed as a Mussolinian diplomatic stratagem to split
the western Allies. This conjecture is given substance by the Duce’s obstinate
refusal to honor the spirit of his understandings with Britain. For instance,
implementation of the second Gentleman’s Agreement, also known as the Easter
accords of 1938, was delayed for six months by the ongoing presence of Italian
forces in Spain despite promises of withdrawal.
Furthermore, in the sequence of international crises leading up to the onset
of the Second World War, Fascist Italy’s record showed an unmistakable bias in
favor of Germany, and even more emphatically against Britain and France.
Mussolini’s acquiescence in Anschluss in March 1938 was a foregone conclusion;
he had forfeited any other option two years earlier. The crisis over the Czech
Sudetenland six months later was a different matter altogether. If Mussolini’s
foreign minister, Count Galeazzo Ciano, is to be believed – and there is no reason
to doubt him – his Duce was minded to fight alongside Nazi Germany, though
no Italian interests were directly engaged.57 And maybe it was Mussolini’s
professions of loyalty that persuaded Hitler to accede to his eleventh-hour request
for one last round of negotiations. Mussolini emerged from the ensuing Munich
conference as the savior of European peace, and Britain rewarded him by
activating the Easter accords and recognizing Italian rule over Ethiopia. But none
of this sat well with his image as warrior–leader, and en route to Munich he
had grumbled that war was averted, because “we could have liquidated France
and Great Britain for ever.”58 This reaction breeds the suspicion that Mussolini
was not content to use the Axis to blackmail the western powers into concessions,
but wanted war for its own sake – an experience to harden the Italian people
and an opportunity to displace Italy’s old-fashioned establishment.59 On the
international stage bellicosity was yet another bond with the other war-lover in
Berlin.
Mussolini’s estrangement from the western Allies, discernible from his conduct
at Munich, sprang into full view over the winter of 1938. On November 30 the
Italian parliament witnessed an orchestrated demonstration by Fascist deputies
who asserted Italy’s right to the French territories of Tunisia, Corsica, Nice and
Savoy. The key to achieving these ambitions in actual fact lay in London rather
than Paris. In the first place, Mussolini hoped to persuade the appeasement-
minded British administration to put pressure on the French to yield; this was
his purpose in inviting prime minister Chamberlain and foreign secretary Halifax
to Rome in the new year of 1939.60 If this gambit failed, as it did – not because
Mussolini and The Myth of Rome 67
Britain would not lean on France but because the French would not budge61 –
then the alternative was to turn to military force. And the foremost obstacle to
an Italian military campaign overseas was British sea power. This the Duce
spelled out in a secret session of the Fascist grand council in February 1939, the
only occasion he addressed this body at any length on international affairs. It
was a frank exposition of foreign-policy goals which bears comparison with
Hitler’s notorious disquisition, minuted by colonel Hossbach. Mussolini began
with the premise that a state’s freedom is “proportional to its maritime position,”
and went on to voice the familiar lament that Italy was a “prisoner in the
Mediterranean and the more populous and powerful Italy becomes, the more it
will suffer from its imprisonment. The bars of this prison are Corsica, Tunisia,
Malta, Cyprus: the sentinels of this prison are Gibraltar and Suez.” From this
situation he drew two conclusions:62
1 The task of Italian policy, which may not and does not entertain any
territorial objective on the European continent save Albania, is first to break
the prison bars.
2 Once the bars are broken, Italy’s policy can have only one watchword: to
march to the ocean. Which ocean? The Indian Ocean by linking Libya
with Ethiopia by way of the Sudan, or the Atlantic by way of French North
Africa.
On the walls of the basilica of Maxentius in Rome were affixed maps showing
side-by-side the conquests of classical Rome and Fascist Italy. The march-to-the-
ocean injunction suggested that Mussolini took this cartographical imagery
seriously. So far-ranging an empire necessitated that not just France but even
Britain should submit to Italy’s will. Indeed, Britain rather than France was
Mussolini’s long-range target, for it was Britain’s pre-eminent imperial status that
he really coveted. Egypt being the geopolitical crux of this aspiration, Radio Bari
directed a stream of anti-British propaganda to the Middle East.63 Anglo-Italian
relations went further downhill when Hitler’s extinction of Czechoslovakia
prompted Italy to invade Albania on Good Friday 1939. Britain’s response was
a series of unilateral guarantees to those small states deemed under threat by
the Axis powers. Whereupon Mussolini complained of British “encirclement”
(shades of Kaiser Wilhelm II whom the Duce so resembled), and he asked
London whether the Easter accords “possessed any further value.”64 A token
and disingenuous gesture to international equidistance.
In reality, notional equidistance had already given way to outright solidarity
with Nazi Germany by means of negotiations for a military-political pact. The
transformation of the informal Rome–Berlin Axis into a formal alliance had been
in the offing since the Anschluss. It was delayed while an effort was made to
inveigle Japan into a tripartite agreement. This proving fruitless, Mussolini had
determined as early as January 1939, the eve of Chamberlain’s Rome visit, on
a bilateral Nazi–Fascist accord. It was finally signed on May 22 and christened
the “Pact of Steel”. The manner of its negotiation reflected Mussolini’s growing
68 The Origins of the Second World War Reconsidered
subservience to his Axis partner: the Italians accepted verbatim a German draft
and agreed to a comprehensive engagement without the customary diplomatic
reservation of a defensive casus belli.65 Such cavalier behavior in so momentous
an undertaking is hard to credit, and is explicable solely on the grounds that
Mussolini and Ciano assumed Italy to be left with sufficient room to maneuver.
This fond hope arose from the fact that before signing the Pact of Steel the
Italians, in a rare burst of honesty, had admitted that their country could not be
ready for war until 1942 at the earliest, to which the Germans had replied that
they had no intention of launching war for several years – and then only in
consultation with Italy. A week after the conclusion of the pact the Duce
dispatched general Ugo Cavallero to Berlin with a memorandum restating this
verbal understanding.66 Here may be perceived Mussolini’s ultimate formula to
make good the myth of Rome. An Italian empire would be won on the battlefield
with the help of Nazi Germany, but Fascist Italy would fight at a time and in a
way of its own choosing. When Italy fought it would be in company with
Germany, yet it would not be fighting Germany’s war; it would fight what
Mussolini later came to call his “parallel war.” A parallel war, of course,
presupposed that Duce could manipulate the Führer; it proved Mussolini’s most
calamitous delusion of all.
The Italian miscalculation was soon exposed. Visiting Salzburg and
Berchtesgaden between August 11 and 13, Ciano learned that the Germans
planned to invade the Polish Corridor within a few weeks, regardless of the near-
certain war with the western democracies which the move would spark.67 In
Rome debate was immediately joined over whether to honor Italy’s commitment
under the Pact of Steel. Mussolini’s first inclination was to stand by Hitler, but
he was eventually dissuaded by reports of his nation’s unpreparedness for war.
This was a victory for those moderate Italian nationalists who, while themselves
not immune to the myth of Rome, did not share Mussolini’s extreme
interpretation of it and, most assuredly, disliked his compact with Nazism as
the agency for consummating it. Prototypical, and very influential in August
1939, was King Victor Emmanuel III. Some Fascist hierarchs also now vented
their reservations about the German alliance. One who, later in his memoirs,
would make much of his opposition to war was Dino Grandi, although how
much weight the opinion of this erstwhile Fascist foreign minister and ambassador
to the court of St James’s still carried is open to doubt.68 Of more significance,
Ciano, following his Salzburg conversations with Hitler and foreign minister
Ribbentrop, had conceived a profound distrust of the Nazis and counseled
neutrality in the coming war.69
Mussolini tried to save face by submitting an impossible list of war supplies
required from Germany before Italy entered the war. In Berlin the
communication was interpreted, as it was intended, as a declaration of neutrality.
Still anxious to cut a figure in the international drama, the Duce offered an
unsolicited mediation proposal. The French evinced some interest, but Germany
and Britain gave it short shrift.70 The rebuff seemed to illustrate Italy’s position
at the bottom of the pecking order of great powers and, by extension, of the
Mussolini and The Myth of Rome 69
marginal role that the country played in bringing on the Second World War. As
Taylor dismissively puts it: “The most Italy could do was hit the headlines, not
raise an alarm.”71 Yet, on receiving the news of Mussolini’s desertion, Hitler
evinced genuine shock; neither the Cavallero memorandum nor his generals’
contempt for Italy’s military capability had apparently registered with him.72 If
the Führer had known from the start of the Polish crisis that Italy would not
fight, would he have stepped back from war? Probably not, but clearly trust in
the Pact of Steel was one of several factors emboldening Hitler to force the issue.
The Italian impact on British policy can also lead to counterfactual speculation.
The driving force behind British appeasement was the triple threat, that is to
say, fear of having to fight Germany, Italy and Japan simultaneously.73 Had Fascist
Italy not adopted so warlike a posture in the Mediterranean, one can reasonably
presume that London would have been less inclined to appease and, however
unwittingly, encourage Hitler. For a lesser power which did not even join in the
hostilities in 1939, Italy’s contribution to the outbreak of the Second World War
was not negligible.
Staying out of war at the outset did not mean that Mussolini renounced his
dream of empire. He eschewed the word “neutrality” since it embarrassingly
recalled liberal Italy’s policy in 1914; instead, he employed nonbellicosity which
connoted, his entourage soon discovered, an intent to enter the war at the earliest
propitious moment.74 A parallel war against the democracies was still on the
agenda. Consequently, negotiations with Britain over the blockade of Germany
were difficult. In particular, an attempt to reach a commercial arrangement which
would mitigate the blockade’s worst effects on Italy was abruptly halted on
Mussolini’s orders in February 1940.75 Ominously, the Duce evaded US
endeavors to obtain a pledge of continued Italian neutrality (or nonbelligerency).76
The Nazi blitzkrieg in north-western Europe in May and the prospective collapse
of France brought on the moment of truth. Besides the patent chance of making
depredations on the western states’ colonial empires, what also weighed with
Mussolini was the scarcely veiled German hint that, without participation in the
war, Italy would be treated as a second-rate power in Hitler’s looming new world
order.77 A desperate French premier traveled to London with a proposal to buy
off Mussolini with an offer of colonial concessions, but he was blocked by
Churchill. In any case, when approached by the French alone, the Duce showed
no interest.78 On June 10, 1940, from the famous balcony overlooking the Piazza
Venezia he announced Italy’s entry into the Second World War in customary
social-Darwinian terminology: “An hour signaled by destiny is sounding....This
is a struggle between peoples fruitful and young against those sterile and dying.”79
In the event, Italy gained next-to-nothing from the fall of France. Then in
October came the real test of the parallel war: unannounced to Hitler beforehand,
Italian forces invaded Greece. But Fascist Italy’s dismal military performance in
the Balkans, coupled with other disasters in the Mediterranean and North Africa,
resulted in Germany assuming almost complete control over all Axis operations
in these theaters. By the end of 1940 the fantasy of a parallel war had been
exploded.80
70 The Origins of the Second World War Reconsidered
So, was June 10, 1940 “Mussolini’s war?” Or was it “the natural result of Italian
history?”81 The answer, as to all such simplistic queries, is “Both.” The myth of a
third Rome antedated Fascism. As one of those “assumptions’ or mentalités behind
foreign-policy making, its force might not have been quantifiable, but it was plainly
visible in the rhetoric of the liberal era before, during, and after the First World
War. In his pursuit of the myth of Rome Mussolini, then, was no radical innovator,
and he appeared to carry most Italians with him at first. The Corfu affair was a
transitory setback, and was offset by the acquisition of Fiume. The slow penetration
of Albania could be counted a success, and the conquest of Ethiopia was
undeniably popular. The first half of the 1930s saw at least a passive Italian
consensus in the Fascist regime’s favor.82 Mussolini’s slide in public esteem began
in 1936, and must be ascribed overwhelmingly to the Axis, an “iron cage” in which
he trapped himself.83 The rudiments of his idea of using Germany to forge a new
Roman empire went back to the early 1920s, and once it materialized the Axis
was put ahead of all other considerations. Italians did not so much resent the
German tie per se as the Duce’s deference to it – from the import of an antisemitism
alien to Italy to the subordination of Mussolini’s parallel war to Hitler’s geopolitical
priorities. To a certain extent, Italy’s miserable war effort of 1940–3 resulted from
the feeling of ordinary Italians that this was not their war. But it was among Italy’s
political class, those who had helped Mussolini into power and served him, that
the myth of Rome was most pervasive and its achievement most keenly anticipated.
Even in these quarters, and before June 10, 1940, many of the so-called Fascist
fiancheggiatori (flankers) had been expressing unease about the German connection.
When they followed the Duce into the maelstrom of war, it was not with any
confidence that he could now deliver a Roman empire, but rather out of nearly
twenty years of habit.
Like Hitler, Mussolini inherited a nationalist tradition and put his own stamp
on it. The Fascist leader contributed an ideological disposition, albeit less fanatical
than Hitler’s, and the same fatalistic social Darwinism which permitted him to
fix his gaze on the end without troubling about the means. These traits were
endemic in the hothouse atmosphere generated by the Rome myth, which asked
to be accepted as an article of faith and relied for fulfillment on Italy’s star being
in the ascendant. On the other hand, Mussolini’s elevation of social-Darwinian
ideology to be the deus ex machina of his diplomacy set him at least narrowly
apart from mainstream Italian nationalism. It was a distinction of degree more
than of kind, and it concerned methods and tactics, not aims. After 1936 Fascist
foreign policy amounted to an all-or-nothing gamble that placed all Italy’s eggs
in the German basket. This was a strategy based in the last analysis on a rash
belief in a favorable tide of history. As such, it tested the credence in destiny of
many old-guard Italian nationalists, and indeed stretched it to breaking-point.
In Germany Hitler’s analogous recklessness, born of blind confidence in his
historic mission, alienated numerous conservative and pragmatic nationalists. The
Italo-German parallelism held good to the end. On 25 July 1943 the Italian
political class belatedly mustered sufficient will to depose the Duce; the following
year their German counterparts tried to assassinate the Führer.
Mussolini and The Myth of Rome 71
Notes
1 See, for example, the exercise in comparative historiography by R. J. B. Bosworth,
Explaining Auschwitz and Hiroshima: History Writing and the Second World War, 1945–1990,
London, 1993.
2 C. S. Maier, The Unmasterable Past: History, Holocaust, and German National Identity,
Cambridge, MA, 1988.
3 F. Fischer, Griff nach der Weltmacht, Düsseldorf, 1961; From Kaiserreich to Third Reich:
Elements of Continuity in German History, 1871–1945, London, 1986.
4 A. J. P Taylor, The Origins of the Second World War, London, 1961.
5 Churchill speech on BBC, 23 December 1940, The Churchill War Papers, vol. 2: Never
Surrender, May–Dec. 1940, M. Gilbert (ed.), London, 1994, pp. 1284–8.
6 Quoted in C. Sprigge’s Introduction to B. Croce, Philosophy, Poetry, History: An Anthology
of Essays, London, 1966, p. lvii.
7 Italy, the Least of the Great Powers: Italian Foreign Policy before the First World War,
Cambridge, 1979, chs 1, 2; “Italian foreign policy and its historiography,” in Altro
Polo: Intellectuals and Their Ideas in Contemporary Italy, R. Bosworth and G. Rizzo (eds),
Sydney, 1983, pp. 65–85; Italy and A Wider World, 1860–1960, London, 1996, chs
1, 2. Also on the question of continuity in Italian foreign policy, see S. C. Azzi,
“The historiography of Fascist Italy,” Historical Journal, vol. 36, 1993, pp. 187–203.
8 F. Chabod, Storia della politica estera italiana dal 1870 al 1896: Le premesse, Bari, 1951.
The work has now been translated into English under the title Italian Foreign Policy:
The Statecraft of the Founders, Princeton, NJ, 1996.
9 Chabod, Statecraft, Part 2: “The idea of Rome.” With few exceptions, however, Italian
historians have been slow to follow Chabod either in seeking the roots of Italian
foreign policy in sociopolitical culture, or in locating twentieth-century Italy’s pursuit
of grandeur in post-unification mythology.
10 For an exemplary portrait of Italy’s old-guard diplomats as moderate and
responsible, see R. Guariglia, Ricordi, 1922–1946, Naples, 1950. Bosworth, Italy, Least
of the Great Powers, ch. 4, demonstrates that they were not so immune to delusions
of national grandeur as their memoirs would suggest.
11 C. J. Lowe and F. Marzari, Italian Foreign Policy, 1870–1940, London, 1975, pp. 9–
10, 54–68. See also the essays in Adua: Le ragioni della sconfitta, A. Del Boca (ed.),
Rome, 1997.
12C. Segrè, Fourth Shore: The Italian Colonization of Libya, Chicago, IL, 1974, pp. 22, 25, 39–40.
13 On nationalist sentiment in post-First World War Italy, see M. G. Melchionni, La
vittoria mutilata, Rome, 1981; H. J. Burgwyn, The Legend of the Mutilated Victory, Westport,
CT, 1993; M. Ledeen, The First Duce: D’Annunzio at Fiume, Baltimore, MD, 1977; R.
De Felice, D’Annunzio politico, 1918–1938, Bari, 1978, Parts 1, 2; A. De Grand, The
Italian Nationalist Association and the Rise of Fascism in Italy, Lincoln, NE, 1978.
14 “Romanità”, in P. V. Cannistraro (ed.), Historical Dictionary of Fascist Italy, Westport,
CT, 1982, p. 461.
15 Ministero degli Esteri, Italy, I documenti diplomatici italiani [henceforth DDI], series 7, vol.
4, no. 532. On Italy at Locarno, see S. Marks, “Mussolini and Locarno: Fascist foreign
policy in microcosm,” Journal of Contemporary History, vol. 14, 1979, pp. 423– 39.
72 The Origins of the Second World War Reconsidered
16 See the following articles by P. G. Edwards: “The foreign office and Fascism, 1924–
1929,” Journal of Contemporary History, vol. 5, no. 2, 1970, pp. 153–61; “The Austen
Chamberlain–Mussolini meetings,” Historical Journal, vol. 14, 1971, pp. 153–64;
“Britain, Fascist Italy and Ethiopia, 1925–1928,” European Studies Review, vol. 4, 1974,
pp. 359–74; “Britain, Mussolini and the ‘Locarno–Geneva system,’” European Studies
Review, vol. 10, 1980, pp. 1–16. For American opinion, see D. F. Schmitz, The United
States and Fascist Italy, 1922–1940, Chapel Hill, NC, 1988, chs 3, 4.
17 Mussolini’s imposture is the theme of my Mussolini’s Early Diplomacy, Princeton, NJ,
1970.
18 DDI, series 7, vol. 2, p. 238 n. 3; nos 360, 373.
19 Cassels, Early Diplomacy, pp. 160–6.
20 Cassels, Early Diplomacy, pp. 166–74; M. Palumbo, “Goering’s Italian exile,” Journal
of Modern History, vol. 50, 1978, no. 1, D1035.
21 Instances of ongoing Italian interest in the Nazis are to be found in DDI, series 7,
vol. 5, no. 680; vol. 6, no. 322; vol. 7, nos 413, 576; vol. 8, nos 367, 377, 384; vol.
9, nos 180, 254, 262, 289.
22 Use of the generic name “fascist” papered over some substantive ideological
differences between the Italian and German movements: witness the failure to
establish a Fascist International, M. Ledeen, Universal Fascism: Theory and Practice of
the Fascist International, 1928–1934, New York, 1972. The latest attempt to create a
typology of fascism is S. Payne, A History of Fascism, 1914–1945, Madison, WI, 1995,
esp. pp. 3–19, 487–95.
23 A. Hitler, Mein Kampf trans. R. Manheim, New York, 1943, pp. 128–30, 628–30,
655–66; K.-P. Hoepke, Die deutsche Rechte und der italienische Faschismus, Düsseldorf,
1968, pp. 159–65; D. J. Rusinow, Italy’s Austrian Heritage, 1919–1946, New York,
1969, pp. 215–18.
24 H. J. Burgwyn, Il revisionismo fascista: La sfida di Mussolini alle grande potenze nei Balcani
e sul Danubio, 1925–1933, Milan, 1979; S. Troebst, Mussolini, Macedonien und die Mächte,
1922–1930, Cologne, 1987; J. J. Sadkovich, Italian Support for Croatian Separatism,
1927– 1937, New York, 1987.
25 B. Mussolini, Opera omnia [henceforth, OO], E. and D. Susmel (eds), Florence–Rome,
1951–84, vol. 23, pp. 176–7.
26 Hoepke, Die deutsche Rechte, pp. 248, 306–8.
27 Guariglia, Ricordi, pp. 14–16; Legatus [R. Cantalupo], Vita diplomatica di Salvatore
Contarini, Rome, 1947, pp. 73–8.
28 A. Lyttleton, The Seizure of Power: Fascism in Italy, 1919–1929, 2nd edn, Princeton
NJ, 1988, ch. 10ff.
29 This theme is adumbrated in the early volumes of De Felice’s life-and-times’ biography,
Mussolini, Turin, 1965–98; 4 parts in 8 volumes. It is expressed more briefly in R. De
Felice, Fascism: An Informal Introduction to its Theory and Practice, New Brunswick, NJ, 1976,
pp. 43–55, and in M. Ledeen, “Renzo De Felice and the controversy over Italian
Fascism,” Journal of Contemporary History, vol. 11, 1976, pp. 269–82.
30 S. B. Clough, The Economic History of Modern Italy, New York, 1964, pp. 246–51; M.
Clark, Modern Italy, 2nd edn, London, 1996, ch. 13.
31 K. Jarausch, The Four Power Pact, Madison, WI, 1965; G. Giordano, Il Patto a Quattro
nella politica estera di Mussolini, Bologna, 1976.
32 Auswärtiges Amt, Germany, Documents on German Foreign Policy, 1918–1945
[henceforth, DGFP], series C, vol. 3, nos 5, 7, 10; G. L. Weinberg, The Foreign Policy
of Hitler’s Germany, vol. 1: Diplomatic Revolution in Europe, 1933–1937, Chicago, IL,
1970, pp. 100–1.
33 G.-K. Kindermann, Hitler’s Defeat in Austria, 1933–1934: Europe’s First Containment of
Nazi Expansionism, Boulder, CO, 1988.
34 W. I. Shorrock, From Ally to Enemy: The Enigma of Fascist Italy in French Diplomacy, 1920–
1940, Kent, OH,, 1988, ch. 7.
Mussolini and The Myth of Rome 73
35 The secret Laval–Mussolini accord is documented in DDI, series 7, vol. 16, nos
391, 399, 403; on the background, see F. Lefebvre D’Ovidio, L’intesa italo-francese
del 1935 nella politica estera di Mussolini, Rome, 1984. For what was both said and
implied at Stresa, see E. Serra, “La questione italo-etiopica alla conferenza di Stresa,”
Affari Esteri, vol. 9, 1977, pp. 313–39. Italian knowledge of the Maffey report was
first revealed by M. Toscano, Designs in Diplomacy, Baltimore, MD, 1970, pp. 412–
14. The overall appraisal of Italy’s international position in 1935 in E. M. Robertson,
Mussolini as Empire Builder: Europe and Africa, 1932–1936, London, 1977, explains why
it seemed to Mussolini a propitious moment for a colonial excursion.
36 G. Baer, Test Case: Italy, Ethiopia, and the League of Nations, Palo Alto, CA, 1976,
chs 2, 3.
37 R. Mori, Mussolini e la conqista dell’Etiopia, Florence, 1978, pp. 217–20, argues
persuasively that he would not.
38 DGFP, series C, vol. 4, nos 485, 506, 525; series D, vol. 1, nos 152, 155. Berlin’s
diplomatic recognition of Italian Ethiopia was announced on October 24, 1936.
39 M. Funke, Sanktionen und Kanonen: Hitler, Mussolini und der Abessinienkonflikt, 1934–1936,
Düsseldorf, 1970, pp. 48–81. For Italy’s place in Hitler’s global strategy, see his Mein
Kampf pp. 620, 664–5, and Hitler’s Secret Book, New York, 1961, pp. 173–4.
40 OO, vol. 15, p. 216.
41 Some of the Duce’s unorthodox sources of information are detailed in D. Mack
Smith, Mussolini’s Roman Empire, London, 1976, pp. 92–5.
42 Quoted in R. H. Whealey, “Mussolini’s ideological diplomacy,” Journal of Modern
History, vol. 39, 1967, p. 435.
43 B. R. Sullivan, “The Italian armed forces, 1918–1940,” in Military Effectiveness, vol. 2: The
Interwar Period, A. R. Millett and W. Murray (eds), Boston, MA, 1988, pp. 169–217.
44 The phenomenon of an ideologue without an ideology, or the distinction between
ideology and ideological thinking, is explored in my Ideology and International Relations
in the Modern World, London, 1997.
45 J. F. Coverdale, Italian Intervention in the Spanish Civil War, Princeton, NJ, 1976, pp.
12– 13, 78–82; G. Ciano, Diplomatic Papers, M. Muggeridge (ed.), London, 1948,
pp. 138– 41, 244.
46 Taylor, Origins, p. 111. Mussolini’s Axis speech is in OO, vol. 28, pp. 67–71.
Incidentally, the axis metaphor was not new in Mussolini’s vocabulary; he had once
described Germany alone as the axis of world affairs, OO, vol. 20, p. 31.
47 Clark, Modern Italy, pp. 97, 120; D. Mack Smith, Italy: A Modern History, revised edn,
Ann Arbor, MI, 1969, pp. 265–6.
48 M. Michaelis, Mussolini and the Jews: German–Italian Relations and the Jewish Question in
Italy, 1922–1945, Oxford, 1978, ch. 5.
49 D. Mack Smith, “Mussolini: artist in propaganda,” History Today, vol. 9, 1959, pp.
223–32.
50 R. De Felice, Mussolini il duce, vol. 2: Lo Stato totalitario, 1936–1940, Turin, 1981;
chs 5, 6.
51 C. Seton Watson, “The Anglo-Italian Gentleman’s Agreement of January 1937,” in
The Fascist Challenge and the Policy of Appeasement, W. J. Mommsen and L. Kettenacker
(eds), London, 1983, pp. 266–82; D. Bolech Cecchi, L’accordo di due imperi: L’accordo
italo-inglese del 16 aprile 1938, Milan, 1977.
52 See, for example, A. L. Goldman, “Sir Robert Vansittart’s search for Italian cooperation
against Hitler,” Journal of Contemporary History, vol. 9, no. 1974, pp. 93– 130.
53 For example, R. A. Lamb, Mussolini and the British, London, 1997.
54 R. Quartararo, Roma tra Londra e Berlino: Politica estera fascista dal 1930 al 1940, Rome,
1980.
55 A telling critique, of De Felice in particular, is M. Knox, “The Fascist regime, its
foreign policy and its wars: an anti-Fascist orthodoxy?”, Contemporary European History,
vol. 4, 1995, pp. 347–65. A recent survey, H. J. Burgwyn’s Italian Foreign Policy in the
74 The Origins of the Second World War Reconsidered
Interwar Period, 1918–1940, Westport, CT, 1997, attempts to find a “middle ground”
between De Felice and his critics.
56 P. Brundu Olla, L’equilibrio difficile: Gran Bretagna, Italia e Francia nel Mediterraneo, Milan,
1980.
57 G. Ciano, Diary, 1937–1938, A. Mayor (ed.), London, 1952, pp. 161–3.
58 Ciano, Diary, 1937–1938, p. 166.
59 M. Knox, “Conquest, foreign and domestic, in Fascist Italy and Nazi Germany,”
Journal of Modern History, vol. 56, 1984, pp. 1–57. The same postulate informs P.
Morgan’s Italian Fascism, 1919–1945, Basingstoke, 1995.
60 P. R. Stafford, “The Chamberlain–Halifax visit to Rome: a reappraisal,” English
Historical Review, vol. 98, 1983, pp. 61–100.
61 Shorrock, From Ally to Enemy, pp. 264–7.
62 OO, vol. 37, pp. 151–2.
63 C. A. MacDonald, “Radio Bari and Italian propaganda in the Middle East and
British counter measures, 1934–1938,” Middle Eastern Studies, vol. 13, 1977, pp. 195–
207. On the Anglo-Italian imperial rivalry, see also my own “Deux empires face à
face: La chimère d’un rapprochement anglo-italien, 1936–1940,” Guerre mondiales et
conflits contemporains, no. 161, Jan., vol. 41, 1991, pp. 67–96.
64 Foreign office, Great Britain, Documents on British Foreign Policy, 1919–1939, series 3,
vol. 5, no. 652. The British, hoping always to detach Mussolini from the Axis,
replied that they considered the Easter accords to be still valid, DBFP, no. 708.
65 M. Toscano, The Origins of the Pact of Steel, Baltimore, MD, 1967, ch. 4. The text of
the pact is printed in an appendix.
66 Ciano, Diplomatic Papers, pp. 283–4; DDI, series 8, vol. 12, no. 59.
67 DDI, series 8, vol. 13, nos 1, 4, 21.
68 D. Grandi, Il mio paese, Bologna, 1985, ch. 45; P. Nello, Dino Grandi: Un fedele
disubbidiente, Bologna, 1993, pp. 359–64.
69 G. Ciano, Diary, 1939–1943, M. Muggeridge (ed.), London, 1947, pp. 120–9.
70 P. R. Stafford, “The French government and the Danzig crisis: the Italian
dimension,” International History Review, vol. 6, 1984, pp. 48–87; R. A. C. Parker,
“The British government and the coming of war with Germany, 1939,” in War and
Society, M. R. D. Foot (ed.), London, 1973, pp. 3–15.
71 Taylor, Origins, p. 40.
72 F. Halder, Diaries, Boulder, CO, 1976, vol. 1, p. 28; P. Schmidt, Hitler’s Interpreter,
London, 1951, p. 146; Weinberg, Foreign Policy of Hitler’s Germany, vol. 2: Starting
World War II, 1937–1939, Chicago, IL, 1980, pp. 637–8.
73 L. R. Pratt, East of Malta, West of Suez: Britain’s Mediterranean Crisis, 1936–1939,
Cambridge, MA, 1975, pp. 4, 30–2.
74 M. Knox, Mussolini Unleashed, 1939–1941: Politics and Strategy in Fascist Italy’s Last War,
Cambridge, 1982, pp. 44–59.
75 W. N. Medlicott, The Economic Blockade, London, 1952, vol. 1, ch. 8.
76 S. Welles, The Time for Decision, New York, 1944, pp. 78–89, 135–47.
77 “If Italy was content with a second-rate position in the Mediterranean, then she
need do nothing more,” DGFP, series 9, vol. 3, no. 1. Interestingly, this explicit
Hitlerian threat is contained in the official German record of the Hitler–Mussolini
meeting on the Brenner in March 1940, but not the Italian. See Ciano, Diplomatic
Papers, pp. 361– 5, reprinted in DDI, series 9, vol. 3, no. 578.
78 On these last-minute maneuverings to keep Italy out of war, see Cassels, “Deux
empires face à face,” pp. 92–5.
79 OO, vol. 29, pp. 43–5.
80 Knox, Mussolini Unleashed, ch. 6.
81 Bosworth, “Italian foreign policy and historiography,” p. 79.
82 R. De Felice, Mussolini il duce, Gli anni del consenso, 1929–1936, Turin, 1974; vol. 1.
83 Ennio Di Nolfo’s phrase quoted in Burgwyn, Italian Foreign Policy, p. 228.
5 A. J. P. Taylor and the
problem with France
Robert J. Young
Rereading The Origins of the Second World War, more than three decades after its
first appearance, is like bumping into a familiar face at a school reunion. Years
ago we, the new boys, had been cautioned about this celebrity. Our mentors
distrusted his judgment, though they admired his “style’ – in part a temperament
distinguished by audacity, in part an inspired pen that too often made the
complex appear simple. It was clear to us then, even if the reasoning behind it
were less so, that Taylor’s Origins had passed some tests magnificently and failed
others rather miserably. Whatever the final assessment of his peers, we
undergraduates knew that Taylor had been responsible for an historical event
of its own kind. Almost forty years later that publishing event of 1961 remains,
as Taylor said of the war itself, “a matter of historical curiosity.”1
It is a mark of this book’s impact to be able to say that it remains central to a
debate that still bubbles away over the war’s origins. Of course, scholars may
persist in saying, as they have for nearly forty years, that Taylor got it wrong,
or at least much of it. They will continue to insist, as indeed they must, that his
evidence was often inadequate for his case, if not overtly incompatible with it.
An historical agent provocateur, he invited the blows of those who found his grasp
of economics rudimentary, his interest in ideology moribund, his predilection
for contrived aphorisms excessive. And yet he was never chased from the field,
never made to surrender. They may have worked around him; for a while they
may even have ignored him; but in the end they have had to return to his field.
Taylor opened the first edition of Origins with a chapter entitled “Forgotten
problem.” It was here that he acknowledged some of the archival obstacles that
he had encountered and which, by his own admission, may have contributed to
the “perhaps misleading impression that international relations between the wars
were an Anglo-German duologue” (p. 38). Misleading or not – and it was –
this is what Taylor concentrated on. Accordingly, it was here that one might
have expected him to vent his enthusiasm. He did not disappoint. Taylor tackled
Hitler’s Germany and Chamberlain’s England with the zeal of an iconoclast.
Hitler, contrary to all that had been said, had no great design, plan or blueprint,
76 The Origins of the Second World War Reconsidered
for the release of further archival material. In fairness, who would have wished
otherwise? Nevertheless, it is also fair to suggest that Taylor – intent as he was
on working with a broad continental canvas – did not come close to exhausting
the supply of very important French materials. By all appearances he did consult
Paul-Boncour, Bonnet, and Flandin, but he overlooked Baudouin, Cot, Fabry,
Laval, Monzie, Reynaud, Tardieu, and Zay. General Gamelin he read, but
evidently not Armengaud, Bourret, Gauché, Jacomet, Minart, Nollet, Prételat,
Réquin, or Weygand. More remarkable, the eleven volumes of reports and
testimonies produced by a French parliamentary commission just after the war
seemed to have escaped Taylor’s attention.6 Though such criticism may be
mistaken for cavil by some, the scholar will acknowledge the point. The fact is
that Taylor, for all his Anglo-German research, made no extraordinary or even
great effort to consult many of the French sources which were available to him
and which he would have found instructive.
Another part of the problem comes from what one is tempted to call a cultural
bias. In a word, it is this which makes it easy for many English to think ill of the
French, a state of mind which is known to have been reciprocated by the French
when given half a chance. The trick, of course, is to be able to sort out informed
from uninformed bias, a service sometimes afforded by one historian or another.
The “despairing cynicism” of the French in the 1930s, Taylor wrote, their “lack of
faith in their leaders and in themselves,” had a “long and complicated origin which
has often been dissected by historians” (p. 72). Reassured by the reasoned analyses
of these entirely unidentified scholars, and satisfied by the very predecessors whose
other judgments he was in the process of discarding, Taylor was ready to confront
the French. What ensued was a scattering of cryptic, unflinching, unsubstantiated,
judgmental remarks about the French. Indeed, Origins is peppered with the language
of some apprehensive day-tripper to Calais.
“Characteristically,” the French failed to see the implications of their decisions
on disarmament policy in April 1934. Faced with a renascent Germany, they
“looked plaintively to London.” Weak in spirit and devoid of resolution, they
were responsible for “dragging the British down with them” for, as everyone
knows, “weakness is infectious” (pp. 107, 197, 209). Daladier, when not being
“sullen,” either “gave way” or “wriggled and dodged.” He, the “most
representative of French politicians” led a people who regularly had “acquiesced
in” or “allowed” one German challenge to succeed after another, a people which
had “supposed all along that the advantages won in 1919 and
subsequently…were assets which they could supinely enjoy, not gains which they
must fiercely defend” (pp. 216, 219, 225, 322, 233). Indeed, for all of his
assurance that appeasement was a positive policy, generated by Chamberlain
without reference to military calculations, Taylor wrote of the prime minister’s
dilemma before the Commons in September 1938: “He could not stress the
unwillingness of the French to fight, which had been the really decisive weakness
on the western side” (p. 236). Despite gallant attempts to be the impartial
exponent, and not the judge, there is little doubt that Taylor was manifestly more
successful when it came to treating England’s enemy than he was her ally.
80 The Origins of the Second World War Reconsidered
Flandin’s apparent change of heart and the British decision of the same week to
commit Britain “for the first time in her history – to peacetime alliance with a
continental Great Power”; no connection between this milestone – designed only
for restraining France, for “ceaselessly holding her back” – and Flandin’s
triumphant return to Paris where he announced the dawning of a new day in
Anglo-French relations. If ever he suspected it, Taylor certainly never
acknowledged even the possibility that a British government had been finessed.
As for the retort that it was the French who had been finessed and unwittingly
harnessed, one has only to invoke Taylor’s murmured aside: “not that the French
needed much restraining” (p. 148). If not, then what explains his “alliance,” unless
Flandin’s bluff had worked and the British government had felt compelled to
defuse the situation?
Edouard Daladier receives similar treatment in the course of the Czech crisis
of 1938. After Chamberlain’s meeting with Hitler at Berchtesgaden, a series of
Anglo-French meetings took place in London in the last two weeks of September.
Daladier, by Taylor’s record, had proven obstreperous. However much he
“wriggled,” he seemed to have a nasty habit of “returning to the question of
principle.” For their part, the British “were urging the French not to take the
offensive.” They even declined the possibility of a last-minute meeting between
Chamberlain and Daladier, just on the eve of the Munich conference, for fear
that the French premier “would once more try ineffectually to coordinate
resistance” (pp. 225–6, 228). The trick was to keep the pressure on Daladier
until he was trapped, but the premier had demanded something in exchange,
one “essential condition.” His acquiescence was contingent upon a British
agreement to share in a formal guarantee to the forthcoming rump Czech state.
Chamberlain, the moralist, balked at such a commitment, but finally relented.
Daladier, Taylor’s weak cynic, though “convinced in his heart that Germany
was aiming at something far greater,” had stood farm. As a result, he managed
to commit Britain “to opposing Hitler’s advance in the east.” No mean
achievement this, had it been deliberate; but of course it was not, not by Taylor’s
light. Rather, Daladier “had built better than he knew” (pp. 219–20). In short,
he had succeeded by blunder and pure chance – a familiar Taylor motif –
unaware of what he had done, or why. How many times can one afford to lift
the glass to the blind eye?
There was something else that Daladier had demanded, and got, in London.
This was the inauguration of more formalized staff talks with the English. Indeed,
this was the one thing that Daladier was really after. All of his grudging
concessions to Chamberlain, over a period of several months, need to be
considered in the light of his determination to put some teeth into the “alliance”;
with Britain. This conclusion, to be sure, makes sense only if one rejects yet
another of Taylor’s contentions, namely that military considerations were not
central to British policy in 1938 and that neither Britain nor France in 1938
seriously envisaged the possibility of being defeated by Germany. This is not
the place, nor is there space, to argue this point. Sometimes, à la Taylor, it is
enough to rely on simple refutation. Recent research certainly raises many
82 The Origins of the Second World War Reconsidered
questions about his judgment of British policy – even to some extent where
Chamberlain himself is concerned – and who but Taylor would suggest that
the French were incapable of admitting the possibility of defeat?7 Clearly, they
were not. Accordingly, they were intensely interested in promoting joint Anglo-
French planning through the medium of extended staff talks. To secure agreement
here, the French government was prepared to adopt a policy in which it did not
believe, and which it actually considered to be inimical to the peace of Europe.
Taylor, distracted by seeing the French “dragged helplessly behind,” caught little
if any of this (pp. 202, 246).
But what made Britain so important to the French? Here is the second
illustration of the problem Taylor has with France. For him the answer lay in
the tirelessly repeated assertion that they were resolved to avoid war at any cost.
This, of course, was neither argued nor substantiated, only reiterated. The
combination of this technique, and the cultural bias previously discussed, meant
that the French emerged gutless from Taylor’s mold. In the absence of clear
historical exposition, there would appear to be no rhyme or reason to their fear
of finding themselves at war. But just as there is another construction that can
be placed on Flandin and Daladier, so there is another that may illuminate further
the condition of France between the wars. Since the argument has been advanced
before, and more thoroughly than is possible here, it should be sufficient to sketch
its outline – namely that French foreign and military policy may profitably be
appraised with some reference to the “guerre de longue durée.”
This notion of a long war of attrition, similar to that which ended in 1918,
was central to French official thinking between the wars.8 Little wonder, therefore,
that they found the prospect of some future war both awesome and repellent.
Nevertheless, for twenty years the conviction endured that the Germans were
bent on revenge, that France would be the principal adversary and the first victim
of German attack. In the absence of outside assistance, the struggle would be
uneven. Indeed, Germany was likely to win, thanks to its superior industrial
and demographic resources. Short of resigning themselves to such an outcome
– an alternative that even Taylor discarded when he said they did not fear defeat
– the French had to rebuild a coalition of allies, one that could be used to deter
Germany from an act of revenge, or to defeat it if deterrence should fail. To
that end, with very mixed results, the government of the Third Republic sought
to enlist the co-operation of Belgium, Poland, Czechoslovakia, Romania,
Yugoslavia, Italy and Russia. Each, for a time, was part of what is often called
the French alliance “system.” The term, though an exaggeration, is probably
better than having French policy confused with alleged attempts to establish a
new continental imperium. This was not what the French were after. Rather
they wished to have at the ready allies who would help to defend France. As
such, the contradictions between their reciprocal undertakings to each “ally,” and
their much remarked upon “defensive” military strategy, are perhaps more
apparent than real.
The keystone of this system, however, was supposed to be Britain. It was to
Britain that successive French governments turned for economic and financial
A. J. P. Taylor and The Problem with France 83
to use Donald Watt’s sobering metaphor, “the mouse remains a mouse and the
lion a lion.”10 It may be just as well, for the mice too have had their differences.
Those differences have a long history, indeed too long to be invoked in its
entirety here. Suffice to say that ever since the sudden collapse of France in 1940
there have been interpretive clashes among historians over the causes of that
defeat. And some of those interpretations have been consistent with Taylor’s
reading of interwar events in general and the behavior of France in particular.11
There is, for example, the work of the late Professor Jean-Baptiste Duroselle, a
legendary figure in postwar French scholarship and author of a weighty tome
called La Décadence.12 Published in 1979, this provocatively titled book had
overtones of Taylor’s work of twenty years earlier. Here we find a survey of the
principal flaws which had reduced the French republic to ignominy in the decade
that preceded its military collapse. To begin with, there was a fundamental
structural problem manifest in an overly powerful legislature and an emasculated
executive branch; this accounted for the sixteen different governments that
presided over France’s destiny between 1932 and 1940. From this chronic
upheaval came the associated problem of long-range planning for the purpose
of international affairs. Hence we have Duroselle’s root cause of the uncertainty
and indecisiveness which he detected in the country’s foreign policy; but there
was something more, something more serious. He was reluctant to suggest a
lethal blemish in the “national character,” and was only slightly more willing to
attribute this suspected decadence to the self-seeking, bourgeois, governing class.
More credible, or so he determined, was the prospect of linking whatever had
been wrong with France to the discredited Marshal Pétain, the defeated and
dismissed General Gamelin, the exiled and forever taciturn Alexis Léger. Here
was proof that the carnage of the First World War had denied the country really
first-class leadership when it came to facing her destiny in the interwar period.
Here, in the Third Republic, “men proved weaker than fate.”
His successor at the Sorbonne caught the same odor of decadence. “Decay
was in the air,” wrote René Girault. “They were preparing for defeat.” For him,
the lack of will-power was at the heart of the decadence, at once its symptom
and cause. Like Professor Duroselle, he was convinced that internal political
divisions had proven so fractious that there was no longer one France but two,
two nations divided over an ideological vision of where the greatest danger lay:
in the Left or in the Right. Hence, there was no unified, single, national will;
and in its absence, France gave to all observers the impression of being the new
sick man of Europe. Such imagery seemed none too harsh for another of Girault’s
Paris-based colleagues. If anything, Serge Berstein was even more convinced that
prewar rot pretty much explained the military disaster of 1940. The country
had dissolved in a sea of “national decadence,” its interwar military policy had
been “inane,” and its intellectual leadership woeful.13
Clearly, Taylor would be quite at home with all of this, entirely familiar with
the broad outline of this enduring view of interwar France. What could be more
encouraging than to have the support of three such eminent French historians?
Together, and expressly, they have ensured that the brief but sweeping judgments
A. J. P. Taylor and The Problem with France 85
of Origins will retain their relevance for years to come. Appraised in this light,
therefore, one must acknowledge that Taylor’s version of France remains very
much in the historical mainstream, despite the dissidence and doubts that have
been expressed earlier in these pages.
And if further proof were needed, one might turn to a few examples of
Anglophone scholarship in the 1980s and 90s, works which are roughly
consistent with Taylor’s view of France. Toward the end of his massive work
on the immediate origins of the war, Donald Watt observes, matter of factly,
that the French republic was in its “penultimate stage of decay” led by a buck-
passing foreign minister in a buck-passing “French system of government.”14
However reluctantly, the Yale-based scholar Piotr Wandycz settled upon
decadence as an explanation for France’s flaccid policy toward eastern Europe
in the 1930s, a conclusion shared with more enthusiasm by Nicole Jordan, who
is struck not only by France’s impotence but by her cynicism and dishonesty.15
The same dishonesty has caught the eye of the Ottawa-based historian, Michael
Carley, who has attributed much of it to capitalist machinations. Simply put,
Rightist influences in France’s public and private sectors, people blinded by their
animus toward the Soviet Union, severely underestimated the threat posed by
Nazi Germany.16 Whereas Carley puts a particular ideological twist on the
concept of Decadence, the eminent historian Eugen Weber is prepared to use it
more conventionally. As an entrée to a recent chapter entitled, simply enough,
“The decadence,” Weber invokes a saying about Frenchmen being “always
inferior to circumstances” and quickly subscribes to the impressions cast by
Professor Duroselle.17 For his part, Anthony Adamthwaite is more circumspect
about the language of decadence. Indeed, he finds it inappropriate. At the same
time, however, he concludes his most recent book on interwar France with the
remark that what was missing by the late 1930s was courage, a quality intended
to depict the national character, and a language not far removed from Taylor’s
own moralistic verdicts on France.18
It would take a debunker of Taylor’s own talent and stature to overturn all
such dismal verdicts on the failures of interwar France. In fact, no one has yet
tried, and no one seems likely to. The instances of brilliant policy brilliantly
conducted are rare in human affairs, and rarer still in the eyes of tough-minded
historians. This is neither more nor less true of France than elsewhere, despite
the disturbing proclivity of some observers – French included – to represent
France as perpetually exceptional. However, common sense, to invoke one of
Taylor’s favorite devices, would tell us that interwar France, like prewar or
postwar France, is likely to disclose a rather mixed legacy: on the one hand,
oversights, mistakes, missed opportunities; and, on the other hand, instances of
success and achievement, some delivered through good planning, others through
good fortune. Not surprisingly, this is indeed what we do find once we discard
the blurring vocabulary of French decadence.
Take, for instance, the matter of French public morale in 1939. While
refraining from calling it defeatist in character, Taylor detected a state of crippling
enervation. The French were, he judged, “at a loss what to do” and “therefore
86 The Origins of the Second World War Reconsidered
decided to let things happen” (p. 322). Yet Professor Duroselle himself led the
way in countering this particular over-simplification. In a public-opinion poll
taken shortly after the Munich settlement, some 70 percent of French respondents
believed that Britain and France had to resist any future demands from Berlin;
and a subsequent poll indicated that a substantial majority were actually in favor
of military action in the event that Germany should attempt to seize the Baltic
port of Danzig. The prospect of war did not please them, but they had had
enough of being bullied.19 Far from being supine, they had been conditioned to
think of resistance by patriotic school textbooks and state-sponsored cinema, by
a surfeit of popular literature on France’s imperial resources and by a wave of
media hype about the reappearance of the Anglo-French alliance.20 Saddened
though they certainly were by the prospect of renewed fighting, they were now
nonetheless resolved to stand up to Hitler. Indeed, pacifism, which had been in
the ascendant in 1938, had become the refuge of a minority in 1939, as the
public mood swung in favor of resistance. Though incapable of regenerating
the boisterous but naive spirit of belligerence of 1914, the French nation stoically
accepted the coming of war in August 1939.21
The growing mood of resistance was in turn associated with the considerable
upswing of the French economy through 1938 and 1939. Indeed, so substantial
was the acceleration of industrial production that Professor Duroselle understood
why some contemporaries had considered it miraculous. Without it, he said,
there certainly would not have been the “remarkable” progress made in aerial
rearmament.22 Two economic historians, Jean Bouvier and Robert Frank, added
their assent. What is more, they used this economic recovery to explain the
contrast between French international policy in 1938 and in 1939. In 1938 the
overall economic situation was so troubled that Daladier could not accept the
risk of war. A year later, when the economic mood was so much brighter, there
was an appreciable increase in the nation’s determination to check Hitler.23 Such
conclusions render two services. First, they credit the Daladier government with
an actual reason for having gone along with Chamberlain in 1938. Second, they
certainly suggest that the spirit of resistance was quickly rekindled once there
were some identifiable grounds for renewed economic optimism.
Similar conclusions may be drawn from the adjacent arena of military
rearmament. Far from shunning war as part of some adamant refusal to
countenance its return under any circumstance, the French government was
coldly pragmatic. It was concerned about the potentially disastrous disparity
between actual French and German military capacities.24 This appraisal, offered
by a team of scholars working for the army’s historical directorate, adds to the
list of legitimate reasons why Daladier acted as he did in 1938. By the same
token, the rearmament progress that was recorded through 1938 and 1939,
especially in the air, certainly contributed to the government’s decision to accept
war only a year after it had rejected it at Munich. By 1939 the country was
spending more on national defense than it had in 1914, and a higher percentage
of its national income.25 France still remained vulnerable, and doubtless
unprepared for undertaking a major offensive, but the combination of its domestic
A. J. P. Taylor and The Problem with France 87
arms’ revival, the recent loss of Czechoslovakia and the threatened loss of Poland,
made both its potential and its need for resistance that much more obvious. By
the spring of 1940 the rhythm of its arms’ production had become even more
impressive, yet another omen for those who would be captivated by images of
French despair, resignation, or apathy.
Then there is the related matter of military planning, a subject which can
embrace everything from strategy to doctrine, and from rearmament
administration to tactical theory. Once more, there are scholars who have
expressed doubts about Tayloresque characterizations of interwar France. If none
speaks of brilliance, a good number perceive foresight and well-honed
professionalism. Such was true of Jeffery Gunsburg. Neither he nor the
accomplished French scholar Henri Michel, whose Foreword appears in
Gunsburg’s book, was much drawn to the “moral decay” argument. In fact,
Gunsburg was impressed by the technical proficiency of the French army in 1940,
and explained its collapse within the context of a much broader Allied defeat.26
A similar chord has been struck by Martin S. Alexander, a British scholar who
regards as “old-fashioned” the decadence-inspired critiques of French policy
makers in the 1930s. In fact, Alexander writes, the often maligned general
Maurice Gamelin understood the German military peril more fully than most
of his contemporaries, which is why he became the “architect of a program of
unprecedented peace-time defensive preparation.”27 Colonel Robert Doughty is
equally chary about using decadence as an explanation. The French failure, he
asserts, “was not one of stupidity, decadence, disloyalty or defeatism.” Nor, thanks
to the intense rearmament effort of the late 1930s, was it due to any truly decisive
material imbalance between the French and German land forces. Rather, it
stemmed from a “vulnerable strategy” and “inadequate tactical doctrine.”28 More
recently, we have had the judgment of Eugenia C. Kiesling, who also rejects
sloth and stupidity as explanations for French doctrinal shortcomings. As for
decadence, she ignores the idea altogether, preferring to seek explanations in the
army’s organization and training.29
Students of foreign policy have also contributed to a more sympathetic, more
realistic, impression of interwar France. Whereas Taylor acknowledged the
growing mood of resistance in 1939, he was at a loss to explain it beyond panic
and desperation. Professor Duroselle subsequently enlarged our understanding
by exploring French foreign policy in the last year of peace. In the course of
that exploration he made it clear that neither Daladier nor Bonnet, unlike
Chamberlain, had any illusions about Hitler’s trustworthiness. That was why
they were not prepared to make significant economic concessions to Germany,
why they were not prepared to forsake French interests in eastern Europe, why
they were not happy with British foot-dragging over negotiations with Russia,
and why Daladier in particular was not going to trade French territory for the
sake of appeasing Mussolini.30 More recently, we have had the work of William
I. Shorrock and John E. Dreifort, neither of whom finds either complete
ineptitude or, for that matter, brilliance in French foreign policy. While sharply
critical of the Popular Front’s Italophobia, the former stresses the severe
88 The Origins of the Second World War Reconsidered
constraints imposed on France by her eastern allies and by the perverse caution
of the British government. Dreifort detects the same kind of constraints when it
comes to French policy toward the Far East, constraints imposed by the
Americans on the one hand and by the British on the other.31 And, more recently
still, we have had fresh insights from Peter Jackson who has examined the history
of the Anglo-French guarantee to Romania in the spring of 1939. The evidence,
Jackson argues, does not suggest an underlying French decadence, or drift, or,
in this case, even indecision. Rather, the French government, driven by its own
strategic interests, prevailed upon London to join in a guarantee of Romanian
security against German attack. In this context, too, one can speak of French
vigor and initiative.32
In other words, the evidence supports – as it usually does – a wide range of
interpretive impressions. Even without benefit of knowing what happened in
1940, one could say that this was a country plagued by economic distress,
ideological division, social fragmentation, and security anxiety. And one could
say with assurance that not all of these problems were addressed with aplomb
and dispatch – any more than problems of similar magnitude were deftly resolved
elsewhere. That said, I am inclined to agree with Michael Miller’s observation
that French “self-assuredness” between the wars has been underrated, and also
with William Irvine’s analysis of French domestic political opinion in the years
just prior to the collapse. France, he writes, “was morally and materially ready
to confront Nazi Germany” by the summer of 1939. That is why, “it was not
decadence that led to 1940.” Rather, it was 1940 “that has led us to view the
late Third Republic as decadent.”33 This was not an incompetent, morally
bankrupt, all-thumbs regime. Or if it were, we would be hard pressed to explain
the stunning economic revival in the last year before the war, the attendant burst
in rearmament, the stiffened public resolve to say “no” to Hitler once the material
conditions for resistance had perceptibly improved. Such, at any rate, seems to
me the most compelling interpretation, and partly because its claims are generally
more modest and measured than those tossed off in Origins.34
Whatever one thinks of the notion that there was something fundamentally
and peculiarly wrong with France in the 1930s, there is one final bone to pick
with the late Professor Taylor. In his “Second Thoughts” of 1963 he prescribed
that the historian’s “sole duty is to find out what was done and why.” A related
dictum, however, proscribes the historian from considering “what ought to have
been done” (p. 26). The argument, though he offers none, presumably is that
otherwise we would spend too much time fecklessly speculating on the historical
“what ifs.” Many will share this concern, but many may also worry that such a
proscription may also disguise irresponsibility within a cloak of good intentions.
For Taylor, history is deceptively simple. Tell it how it is and why, exposition
and not judgment. But if it is exposition left unmeasured against any realistic
yardstick of probabilities and “what ifs,” it may also be exposition for which the
historian cannot be held responsible. Unchecked from this quarter, and smitten
by one’s own impartiality, it becomes easy to fudge the difference between
explanation and judgment. And so Taylor is intent on “explaining,” sometimes
A. J. P. Taylor and The Problem with France 89
magnanimously, where the British and French went wrong, how they misread
Hitler and completely misunderstood him, thought he was planning to do them
in, when he was not. Hitler was only waiting; waiting for his chance to extract
more concessions. Their mistake, Taylor explains, beyond their basic misreading,
was to have resorted to concessions, whether from drift or deliberation.
Is that it? Should they have seen that Hitler was an arch opportunist, not a
planner, and then denied him his opportunities? That would have left the
initiative to him, would have put pressure on him, but with what likely result?
We do not know. Taylor does not know. All we know is that such a disobliging
approach had been characteristic of French policy over reparations and
disarmament, a policy that had drawn much criticism from London and one
which Taylor himself liked no better than the over-supple policy of concessions
in the later 1930s. Indeed, one has the nagging impression that the French can
never get things quite right. Their disarmament policy earlier in the decade was
criticized for being too rigid and unbending – which, for the most part, it
probably was.35 In 1938 their policy was similarly condemned, partly because
it appeared tough, and partly because it proved weak. Flexible or unyielding,
French policy just never seemed quite in time with the beat followed in Whitehall.
That led to contemporary British criticism and complaints, a perspective which
Taylor seemed to find eminently justified. Moreover, it is not clear from Taylor
how, when, or why the French – had they understood Hitler the way he did –
might have changed from hard to soft policy, or vice versa. Neither is it clear
how the British – endowed with Taylor’s vision – might have responded to a
more perspicacious French policy. It was enough for him to know where they
went wrong. It was not necessary to test or confirm it; no flicker of interest from
him over what would or could have happened had they known what he knew.
This, it must be reckoned, is the behavior of a judge, one who is forever
appraising negligence, if not culpability, and who recognizes no obligation to
justify any judgment. If it is not ostensibly a verdict, it is ostensibly more than
an explanation. This, however, did not much concern Taylor, a man not really
given to second thoughts. Over the years he surrendered little ground, made
few concessions, other than the late-in-life decision to minimize brushes with
academic colleagues. “My view is that they should get on with writing books in
their way and I will get on with writing them in mine.”36 Never before has that
advice seemed more sound for those whom he has left behind.
Notes
1 A. J. P. Taylor, The Origins of the Second World War, London, 1964, p. 336. All references
are to this Penguin edition, which contains his supplementary essay entitled “Second
Thoughts.”
2 See “Metternich,” in A. J. P. Taylor, Grandeur and Decline, London, 1967, p. 23.
3 There is some irony to this criticism. Taylor has been criticized for having treated
Hitler as some ordinary German, despite the enormity of the crimes committed by
his regime. By so doing, Taylor has been charged with having abdicated his moral
90 The Origins of the Second World War Reconsidered
responsibility as an historian. Here he is being criticized for making too many moral
judgments about the French, principally by means of implication and insinuation.
For the earlier criticism, see C. Robert Cole, “Critics of the Taylor view of history,”
in E. M. Robertson (ed.), The Origins of the Second World War, London, 1971, pp. 142–
57. See also the articles by Oswald Hauser and John W. Boyer in the Special Issue
of Journal of Modern History vol. 49, 1977, pp. 34–9, and 40–72.
,
4 Robert J. Young, In Command of France: French Foreign Policy and Military Planning, 1933–
1940, Cambridge, MA, 1978, p. 1.
5 A. J. P. Taylor, “Accident prone, or what happened next,” Journal of Modern History,
vol. 49, 1977, p. 18.
6 These and other materials are surveyed in Robert J. Young (ed.), French Foreign Policy,
1918–1945. A Guide to Research and Research Materials, Wilmington, DE, 1991.
7 In 1938 the British chiefs of staff admitted their uncertainty as to whether the
Germans were capable of delivering a knockout blow against England from the
air. For his part, the prime minister warned his cabinet in August that the country
could not face the prospect of war with much confidence. See R. A. C. Parker,
“Perceptions de la puissance par les décideurs britanniques, 1938–1939: le Cabinet,”
in René Girault and Robert Frank (eds), La Puissance en Europe, 1938–1940, Paris,
1984, p. 48. See also Parker’s more recent Chamberlain and Appeasement. British Policy
and the Coming of the Second World War, London and New York, 1993.
8 Conference paper by Philippe Masson, “La marine française et la stratégie alliée,
1938–39,” presented to the Colloque Franco-Allemand, Bonn, 1978. See also Eleanor
M. Gates, End of the Affair: The Collapse of the Anglo-French Alliance, 1939–1940,
Berkeley, CA, 1980, pp. 3–17; and Robert J. Young, “La Guerre de Longue Durée:
some reflections on French strategy and diplomacy in the 1930s,” in Adrian Preston
(ed.), General Staffs and Diplomacy Before the Second World War, London, 1974, pp. 41–
64.
9 See the two papers by René Girault and Robert Frankenstein entitled, respectively,
“The impact of the economic situation on the foreign policy of France, 1936–1939,”
and “The decline of France and French appeasement policies, 1936–1939,” in
Wolfgang J. Mommsen and Lothar Kettenacker (eds), The Fascist Challenge and the
Policy of Appeasement, London, 1983, pp. 209–26, and 236–45.
10 D. C. Watt, “Some aspects of A. J. P. Taylor’s work as diplomatic historian,” Journal
of Modern History, vol. 49, 1977, p. 33.
11 For a fuller discussion of the history of these interpretive debates, see the
historiographical chapter in my France and the Origins of the Second World War, London,
1996, pp. 37–59.
12 J.-B. Duroselle, La Décadence, 1932–1939, Paris, 1979.
13 René Girault, “Les décideurs français et la puissance française en 1938–1939,” in
La Puissance en Europe, p. 39; Serge Berstein, La France des armies 30, Paris, 1988, pp.
80, 169.
14 Donald Cameron Watt, How War Came. The Immediate Origins of the Second World War,
London, 1989, p. 617.
15 Piotr Wandycz, The Twilight of French Eastern Alliances, 1926–1936, Princeton, NJ,
1988; Nicole Jordan, The Popular Front and Central Europe: The Dilemmas of French
Impotence, 1918–1940, Cambridge, MA, 1992; and “Maurice Gamelin, Italy and the
eastern alliances,” Journal of Strategic Studies, vol. 14, 1991, pp. 428–41.
16 Michael J. Carley, “End of the ‘low, dishonest decade’: failure of the Anglo-French
Soviet alliance in 1939,” Europe–Asia Studies, vol. 14, 1993, pp. 303–41; and “Prelude
to defeat: Franco–Soviet relations, 1930–1939,” Historical Reflections/Réflexions
Historiques, vol. 22, 1996, pp. 159–188.
17 Eugen Weber, The Hollow Years: France in the 1930s, New York, 1994, pp. 111–46.
18 Anthony Adamthwaite, Grandeur and Misery. France’s Bid for Power in Europe, 1914–
1940, London, 1995, p. 231.
A. J. P. Taylor and The Problem with France 91
19 Duroselle, Décadence, pp. 355–6. See also his L’Abîme, 1939–1945, Paris, 1982, pp.
17– 18. This conclusion has been questioned by the remarkable, not to say
extraordinary, construction advanced by a psychohistorian who perceives a “suicidal
group fantasy” at work on the inexorable road to 1940. See Stephen Ryan,
“Reflections on the psychohistory of France, 1919–1940,” Journal of Psychohistory, vol.
2, 1983, pp. 225– 41.
20 See, for example, the following articles which appeared in a single issue of Relations
Internationales (no. 33, 1983), addressed to “Images de la France en 1938–1939”:
Christine Sellin, “Les manuels scolaires et la puissance française,” pp. 103–11; Rémy
Pithon, “Opinions publiques et représentations culturelles face aux problèmes de
la puissance. Le témoignage du cinéma français, 1938–1939,” pp. 91–101; René
Duval, “Radio–Paris,” in Olivier Barrot and Pascal Ory (eds), Entre deuxguerres. La
création entre 1919 et 1939, Paris, 1990, pp. 129–46.
21 See Maurice Vasse, “Le pacifisme français dans les années trente,” Relations
Internationales, no. 53, 1988, pp. 37–52; and Norman Ingram, The Politics of Dissent.
Pacifism in France, 1919–1939, Oxford, 1991.
22 Duroselle, Décadence, pp. 444, 457.
23 Jean Bouvier and Robert Frank, “Sur la perception de la ‘puissance’ économique
en France pendant les années 1930,” in La Puissance en Europe, pp. 182–3. For the
contribution of the Comité des Forges and the Conféderation du Patronat to this
new mood, see H. Coutau-Bégarie, “Comment les Français se sont préparés à la
guerre,” Revue d’Histoire Diplomatique, vol. 97, 1983, p. 347. For the range of industrial
responses to accelerated rearmament, see Richard Vinen, The Politics of French Business,
1936–1945, Cambridge, 1991.
24 Général J. Delmas, “La perception de la puissance militaire française,” in La Puissance
en Europe, pp. 127–40.
25 On the subject of French rearmament, see the work by Robert Frankenstein, Le
prix du réarmement français, 1935–1939, Paris, 1982; and Jean Crémieux-Brilhac, Les
Français de l’An 40, vol.2: Ouvrier et Soldats, Paris, 1990. On military strategy more
generally, see Gordon Martel, “Military planning and the origins of the Second
World War,” in Brian McKercher and Michael Hennessy (eds), Military Planning and
the Origins of the Second World War, Westport, CT, 1999.
26 Jeffery A. Gunsburg, Divided and Conquered: The French High Command and the Defeat
of the West, 1940, Westport, CT, 1979. For a dissenting, if familiar, view of the French
army, see the peculiarly self-possessed work by Williamson Murray, The Change in
the European Balance of Power, 1938–1939, Princeton, NJ, 1984.
27 Martin S. Alexander, The Republic in Danger. General Maurice Gamelin and the Politics of
French Defence, 1933–1940, Cambridge, 1992. For his remarks on the historiography,
see “Did the Deuxième Bureau work? The role of intellgence in French defence
policy and strategy, 1919–1939,” Intelligence and National Security, vol. 6, 1991, p. 302;
and “The fall of France, 1940,” Journal of Strategic Studies, vol. 13, 1990, pp. 12–21.
28 Robert A. Doughty, The Seeds of Disaster. The Development of French Army Doctrine, 1919–
1939, Hamden, CT, 1985, p. 188; The Breaking Point. Sedan and the Fall of France,
1940, New York, 1990, p. 245.
29 Eugenia C. Kiesling, Arming Against Hitler. France and the Limits of Military Planning,
Lawrence, KS, 1996, pp. 6, 144. See also Elizabeth Kier, Imagining War. French and
British Military Doctrine Between the Wars, Princeton, NJ, 1997.
30 Duroselle, Décadence, chs 12–15. See also Elisabeth du Réau, Edouard Daladier, 1884–
1970, Paris, 1993, pp. 310–54.
31 William I. Shorrock, From Ally to Enemy: The Enigma of Fascist Italy in French Diplomacy,
1920–1940, Kent, OH, 1988; John E. Dreifort, Myopic Grandeur. The Ambivalence of
French Foreign Policy toward the Far East, 1919–1945, Kent, OH, 1991.
32 Peter Jackson, “France and the guarantee to Romania, April 1939,” Intelligence and
National Security, vol. 10, 1995, pp. 242–72.
92 The Origins of the Second World War Reconsidered
33 Michael B. Miller, Shanghai on the Métro. Spies, Intrigue and the French Between the Wars,
Berkeley, CA, 1994, p. 5; William D. Irvine, “Domestic politics and the fall of France
in 1940,” Historical Reflections/Réflexions Historiques, vol. 22, 1996, pp. 86, 90.
34 Contrary to the suggestions of Douglas Porch, far from subscribing to the decadence
view, I have spent the last thirty years contesting it. See my France and the Origins of
the Second World War, London and New York, 1996, and his The French Secret Services.
From the Dreyfus Affair to the Gulf War, New York, 1995, p. 144.
35 See Maurice Vaïsse, Sécurité d’Abord. La politique française en matière de désarmament, 9
décembre 19 30–17 avril 1934, Paris, 1981. See also his “Against appeasement: French
advocates of firmness, 1933–1938,” in The Fascist Challenge and the Policy of Appeasement,
pp. 227–35.
36 Taylor, “Accident prone,” p. 17.
6 Misjudging Hitler
A. J. P. Taylor and the Third Reich
Richard Overy
rooted in the available scholarship and the published sources. Taylor sought to
make Hitler plausible: “Hitler had no clear-cut plan and instead was a supreme
opportunist, taking advantages as they came.”4 In Taylor’s hands Hitler was no
more a monster, but a vain power-seeker whose ideological rhetoric amounted
to mere incantations, “phrases to produce the popular roar.” Hitler was not
possessed of “genuine beliefs,” but “craved for power” alone.5
More than thirty years of scholarship on German foreign and military policy
in the 1930s, and on the role of Hitler himself, threaten to make Taylor’s view
of Germany nothing more than a historiographical curiosity. Yet it is important
to remember in the first instance what Taylor got right. German policy in the
1930s was rooted in the longer course of German history and did not represent
a sharp rupture with the past. It can be clearly demonstrated that the main
elements in Hitler’s view of foreign policy derived in almost a straight line from
the radical nationalism of the pre-1914 Reich. These elements were three: the
pan-German longing for the territorial unity and independence of all racial
Germans; the pursuit of Lebensraum (living-space) in order to achieve the proper
match between the territory and the economic needs of a people (with the strong
implication that space should rightfully be allocated to peoples with superior
cultures and forms of social organization); the pursuit of Woltpolitik, or global
policy, in which the united and enlarged state engaged in worldwide imperial
politics.6
Taylor, of course, did not express the continuities in this way. He
acknowledged that Hitler came from a Viennese pan-German background, but
he did not accept that Hitler’s ideological baggage mattered at all. In reality it
mattered a great deal. The infant Deutsche Arbeiter Partei (DAP) founded by Anton
Drexler in 1919, which was the immediate forerunner of the Nazi Party, was
directly linked with the pan-German Vaterlandspartei founded by Alfred Hugenberg
in September 1917 to rally patriotic support for the German war effort, and with
the radical nationalist Thule Society. Both organizations sought the moral and
spiritual regeneration of Germany, the unity of all Germans and the
predominance of the German race in Europe. Drexler found both organizations
too dominated by bourgeois intellectuals, and set up his own party. The DAP
admitted Adolf Hitler in September 1919, and the following year changed its
name to the Nationalsozialistische Deutsche Arbeiterpartei (NSDAP).7 The new Party’s
twenty-five-point program included demands for the unity of all Germans, and
the right to livingspace. These ambitions owed something to the Versailles
settlement and German revisionism, but in effect their pedigree considerably pre-
dated Versailles. Hitler’s program did not simply amount to treaty revision, as
Taylor claimed in 1981, but rested on a popular radical nationalist discourse
which held racial unity and violent economic imperialism at its core.
During the 1920s Hitler became an important spokesman for nationalist circles
in Germany that kept alive these prewar ambitions. German nationalism itself
was always a fractured movement. Many Germans, across the political spectrum
from right to left, sought revision of some kind. The terms of the Versailles
settlement were never accepted, and the well-known efforts of Gustav Stresemann,
Misjudging Hitler 95
foreign minister from 1924 to 1929, and Heinrich Brüning, chancellor from 1930
to 1932, to ameliorate or remove treaty restrictions scarcely merit repetition.8
Germany was reabsorbed into the states system when the Treaty of Locarno
was signed in 1925, respecting the western frontiers of Germany established in
1919, and when Germany was permitted to join the League of Nations a year
later. The Allied Control Commission in Germany, charged with monitoring
the disarmament clauses of the treaty, left in 1927. In 1930 the occupation forces
were finally withdrawn from the Rhineland. The most hated symbol of German
inferiority, the annual payment of reparations to the Allies for German “war
guilt,” were finally suspended at the Lausanne conference of 1932. It is almost
certain that if Hitler had not come to power in 1933, another German
government would have continued the revisionist thrust and might have achieved
through negotiation much of what Hitler ultimately achieved by a defiant
unilateralism.
The other elements of German nationalism in the 1920s were potentially more
dangerous, but were for much of the period confined to the radical nationalist
fringe. Hitler expressed them in Mein Kampf, which Taylor dismissed as fantasy,
and more elaborately in his so-called “Second book,” dictated in 1928 but not
published until 1961, when it appeared in German and in English. This latter
book came too late for Taylor’s first edition, but whether he would have treated
it any more seriously than he did Mein Kampf is open to question. Yet the second
book, even more than the first, shows Hitler’s rejection of what he called the
“patriotic-national bourgeois” circles in Germany, who had sold the Reich “to
an organization of pimps, pickpockets, deserters, black marketeers and hack
journalists.”9 Hitler was not concerned just with treaty revision, with its strong
implication of a “restoration” of the Germany of pre-1914, but saw in Germany’s
future the building of a solid racial core, the race-contest with international Jewry,
and the build-up of sufficient military power to allow Germany to seize an
economic empire in the spaces of the ill-defined “east.”10 This hardly constituted
a clear program, but it reflected a unique strand in German nationalism that
transcended treaty revision, or conventional balance-of-power politics, and
ultimately embraced just what the Nazi regime in fact embarked upon – ethnic
cleansing on a grotesque scale and a grandiose imperialism in the east. The
distinction between the cautious foreign and military policy of German
revisionists in the inter-war years, and the violent pursuit of a race-based New
Order is too fundamental to be dismissed lightly. Hitler had beliefs, borrowed
beliefs perhaps, but beliefs nonetheless. They were not, in Taylor’s trivial
formulation, “the conversation of any Austrian café or German beer-house,”11
but quite the reverse. Hitler’s ideological outlook was widely echoed in
universities and professional associations colonized by a nationalist intelligentsia
that also gazed beyond revisionism.12
Taylor was right, however, in another respect. Hitler did not act alone in the
conduct of foreign policy in the 1930s, neither did he dictate its course
exclusively. There were continuities of personnel and policy across the divide
of 1933. There was not one but several strands in the formation of German
96 The Origins of the Second World War Reconsidered
foreign policy in the 1930s, what Wolfgang Michalka has called “a plurality of
conceptions.”13 For much of the period between 1933 and 1939 foreign policy
was run by traditional career diplomats whose nationalism was based on a
cautious revisionism and the reassertion of Germany’s power-political position
that had been lost between 1914 and 1919. They were represented by the
German foreign minister Constantine von Neurath, a conservative nationalist
who was appointed before Hitler came to power, and kept his office until 1938.
Neurath, together with conservatives in the armed forces’ leadership and in
economic affairs, was concerned that the pace of revision should not prejudice
the establishment of domestic stability, nor invite the dangerous intervention of
other states. Hitler left much of the day-to-day conduct of diplomacy in Neurath’s
hands, just as he left economic policy to Schacht, and the practical achievement
of rearmament to the armed forces. Up until 1939 and the German demand for
the return of Danzig and renegotiation of the terms of the Polish Corridor (which
conservatives cared about more fervently than any other aspect of revision), little
that Hitler had achieved ran counter to the conservative view of revision (except
the Anschluss with Austria, with its implicit rejection of Bismarck’s conception of
a Prussia-centered Germany). Where conservatives parted company with Hitler
was over the methods used to achieve revision, methods which ran much greater
risks than they were prepared to accept.14
The conservatives also shared further preferences in foreign policy. They
hoped, as before 1914, to secure the friendship of Britain; they disliked Poland
but courted Russia; they favored the establishment of some kind of economic/
political bloc in central and eastern Europe (the prewar vision of Mitteleuropa);
they were hostile to collaboration with Italy (Albert Speer recalled after the war
that President von Hindenburg had once asked Hitler never again to enter
Germany into alliance with Italians15); finally, they wanted colonies and the
reintegration of Germany with the wider world economy. Some of these
preferences Hitler himself shared. Up until 1937 he too favored an alliance with
Britain, and had written so in the 1920s. Hitler was not a particular enthusiast
for Italy, despite the fact that it was governed by a fellow-radical nationalist. Hitler
was all for constructing a system of alliances and trade agreements in eastern
Europe, and did so right through to 1939 culminating with an agreement with
the USSR. There always existed sufficient congruence between Hitler and the
conservative nationalists to blur the differences, and it was in this sense that
Taylor could claim that Hitler’s foreign policy was “that of his predecessors, of
the professional diplomats at the foreign ministry, and indeed of virtually all
Germans.”16
There was more, of course, to German foreign policy in the 1930s than either
Hitler or the foreign ministry. The current view is that the Third Reich operated
in a polycratic way, with no single and consistent agenda and with a high level
of systemic tension generated by rivalry between the Party leaders. Alfred
Rosenberg, the Party’s official ideologue, harbored the idea of a German-
dominated Eurasia made up of ethnically defined states freed from Bolshevik
rule; Hermann Göring played a major part in securing better relations with
Misjudging Hitler 97
Poland, and with Italy, and tried to act as a brake on Hitler in 1938 and 1939
when he felt the risks his leader played were too great.17 Above all, Joachim von
Ribbentrop, one of the Party’s foreign affairs’ advisers, pursued a policy derived
more from pre-1914 Weltpolitik: the pursuit of colonies; re-entry into global politics
along with Italy and Japan; and a possible reconciliation with Russia to cover
potential conflict with Britain. Some historians have seen his appointment as
foreign minister in February 1938 to replace von Neurath as evidence that von
Ribbentrop now played a prominent part in formulating German foreign policy,18
particularly as he followed Hitler in 1937 in promoting a strongly anti-British
line, and apparently secured the German–Soviet Pact in August 1939. Though
his influence now seems to have been far short of decisive, von Ribbentrop did
clearly play some part in shaping Hitler’s attitude toward the balance of power
in the critical months leading to war, particularly in his efforts to persuade Hitler
to call Britain’s bluff over Poland.19 Yet none of the leading Nazis with foreign-
policy interests, Von Ribbentrop included, was able to impose them on Hitler
in any systematic way.
Finally, Taylor was surely right to view the outbreak of war in 1939 as part
of a broader crisis of the international order of which Hitler was able to take
particular advantage, as did Japan and Italy. Hitler clearly did perceive
opportunities to exploit as the League system broke up in the 1930s. The two
global powers, Britain and France, faced pressures on all fronts, domestic, foreign,
and imperial to which, in Hitler’s view, they manifestly failed to respond with
vigor. All the revisionist states profited from the partial decline of Britain and
France; Hitler hoped to exploit that decline when he sought war with Poland in
1939. Not all historians accept that Hitler wanted to isolate Poland and avoid a
general conflict, but the evidence weighs heavily in favor of this interpretation.
Take, for example, the notes of Hitler’s army adjutant, Gerhard Engel, published
in 1974. On August 22, 1939 he recorded Hitler’s views, expressed at a
conference with his generals: “He repeats again, he is convinced that Poland
remains isolated, England and France would only bluff, and he does not intend
to settle business with these in the foreseeable future.” On August 27 Engel heard
Hitler again argue that Britain would not intervene. Two days later Hitler was
determined to finish off Poland “but indeed wants no war at all with the others.”20
Hitler’s immediate entourage was less sanguine about the chances of avoiding a
general war, but as Göring concluded in one of his postwar interrogations:
“[Hitler’s] main idea was to try to keep the western powers out of the war....As
we saw it he held much too rigidly to this.”21
Hitler was not, of course, acting in wilful disregard of reality. He was supplied
with a diet of information, some of it culled from intercepted diplomatic traffic
between France, Britain and Poland, that seemed to confirm western hesitancy.
It is known that intelligence information which contradicted this interpretation
was deliberately withheld from him.22 For a man already predisposed to see in
the democracies only signs of decadence and double-dealing, the accumulating
weight of evidence suggested that Poland would be left in the lurch. Hitler’s
decision to invade Poland was a risk, but one which he judged to be worth taking,
98 The Origins of the Second World War Reconsidered
even more so after the German–Soviet Pact, which ended any prospect of a
revived strategy of “encirclement” which he, like any German statesman, was
anxious to avoid. In this sense he might well appear the unprincipled opportunist
of Taylor’s version of 1939, seizing a favorable moment to complete the program
of German revanchism for which millions had voted for him in 1932. Hitler
did not break down the European political order. It was already deep in crisis
when he delivered the fatal blow in 1939.23
The weaknesses of Taylor’s argument about Germany and Hitler lie not in
the realm of diplomacy but in that of domestic politics. His insistence that Hitler
had only a limited program of treaty revision, which was widely approved by
the conservative elites and by the German people, blinded him to the dynamic
nature of the dictatorship, and in particular to the sharp break in domestic politics
that occurred between 1936 and 1938. The structure and aims of German foreign
policy, and the means to achieve them, altered fundamentally in the mid-1930s
at the expense of the conservative nationalists who had applauded the early stages
of treaty revision. This was not perhaps immediately evident from the formal
foreign-office archive on which Taylor relied, which reflected what Klaus
Hildebrand has called “seemingly familiar historical phenomena,” but behind
the conventional diplomacy of the late 1930s lay, in Hildebrand’s words, “a
dogma which took over everything,” driven by the “destructive excess of [Hitler’s]
historical vision.”24
The break in the mid-1930s was rooted in one of the most important
documents of the whole period, the strategic memorandum that Hitler drew up
at Berchtesgaden in August 1936, which has come to be known as the “Four-
Year Plan Memorandum.” Its significance derives from the fact that Hitler hardly
ever put pen to paper throughout the entire dictatorship, but on this occasion
the substance of his thoughts was sufficiently important to the future development
of German policy for him to set them down himself. Hitler treated the document
with a solemn self-importance. When it was written he summoned Göring to
Berchtesgaden where he discussed its contents and presented him with one of
only four copies. Others were given to Fritz Todt, the engineer who was
responsible for Hitler’s grandiose construction plans, and to general Werner von
Blomberg, the war minister.25
Hitler’s central argument was simply that the world had reached a historic
climacteric: the French Revolution had worked its evils on European culture
for more than a century, and its natural progeny, Bolshevism, in alliance with
world Judaeism, was poised to do to Europe what the barbarian invasions had
done to ancient Rome. “No nation,” Hitler argued, “will be able to avoid or
abstain from this historical conflict.”26 The western powers had forfeited
leadership of the world struggle, corrupted as they were by democratic values
and infected by Marxism. Germany stood in the way of the worldwide triumph
of Judaeo-Bolshevism, aided by Italy and Japan. Hitler regarded conflict as
inevitable if Europe were to avoid “the most gruesome catastrophe” since “the
downfall of the states of antiquity.” All other aims paled into insignificance. Hitler
called for a program of massive militarization, and the unrestricted mobilization
Misjudging Hitler 99
of the nation’s economic resources to prepare for the apocalyptic struggle: “the
political movement among our people knows only one goal, the preservation of
our existence…”27
Taylor makes no mention of this memorandum, and that he would have taken
it at all seriously is questionable. It did not constitute a clear program of
objectives, and was couched in the same language of historical generality already
evident in Mein Kampf. Hitler was obsessed with the weight of the historical
moment he confronted and the struggle for the future which that moment made
necessary. The memorandum was not a detailed statement of policy; rather, it
had the quality of an oracular pronouncement, pointing the way towards the
harsh historical path that Germany must tread. What distinguishes the
memorandum from Hitler’s other writing is both its timing – for it was written
well into the dictatorship rather than in the years of political apprenticeship before
1933 and its practical consequences.
The 1936 memorandum cannot be regarded simply as some flight of fancy,
for it was developed as the basis for a complex and far-reaching transformation
of German foreign, economic, and military policy. It is well to remember that it
was written shortly after the outbreak of the Spanish Civil War and Hitler’s
decision to send military assistance to help the nationalist rebels. There were, of
course, practical reasons to explain German intervention. Valuable raw materials
– pyrites, wolfram – were imported from Spain and had to be safeguarded for
German rearmament.28 Göring was keen to find an opportunity to try out aircraft
capabilities and tactics. But intervention in Spain was also prompted by the fear
that Spain might undergo a communist revolution that would hem in the fascist
powers. The election of the Popular Front government in France in June, with
Communist support, posed a further threat. Taken together with the knowledge
of Soviet modernization and rearmament, Hitler’s belief in an imminent reckoning
of accounts with Bolshevism appears anything but fanciful.
It can scarcely be coincidence that the two states singled out in Hitler’s
memorandum as “standing firm in the face of the world peril,”29 Italy and Japan,
now became more closely aligned with Germany. Up until 1936 both the
German foreign ministry and the general staff had worked to create close links
with China rather than with Japan,30 whose value as an ally and as a market
was not regarded as high. Hitler, however, influenced by von Ribbentrop, took
a rosier view of Japan, whose anti-Soviet stance he shared. Four days after the
outbreak of the Spanish Civil War, Hitler spoke with the Japanese ambassador
at Bayreuth, where he agreed to pursue negotiations for a pact with Japan and
approved a secret protocol on benevolent neutrality. The discussions led to the
drafting of the Anti-Comintern Pact on October 23 1936, and its final signature
in Berlin in November, despite the continued hostility of the foreign ministry to
any policy that threatened Germany’s links with China.31
Intervention in the Spanish Civil War also drew Germany into alignment with
Fascist Italy. Again this was not the preference of the foreign ministry, nor was
it an alignment altogether welcomed by either Mussolini or Hitler. The Italian
leader had refused Franco’s request for aid until news came of German assistance.
100 The Origins of the Second World War Reconsidered
Isolated after the Ethiopian war, and uncertain of German ambitions in central
Europe, Mussolini was interested in closer links with Germany. When Count
Ciano, the newly appointed Italian foreign minister, visited Berlin in the autumn
of 1936 he was warmly received. In November Mussolini announced the Rome–
Berlin Axis, which was little more than a statement of mutual goodwill in defense
of the Fascist ideal. A year later Italy formally joined the Anti-Comintern Pact,
following further assiduous negotiation by von Ribbentrop.32
The re-alignment with Italy and Japan, expressed as a front against
international Marxism, made little strategic or military sense (nor did Germany,
in the end, derive much benefit from the connection). The German army thought
that Italy would be of more use to them as an enemy than as an ally; the generals
were keen to continue assistance to the nationalist Chinese against Japan. But
the alignment was made because it conformed with Hitler’s desire publicly to
sustain the struggle against the Bolshevik menace. The reorientation also
coincided with a more substantive shift in Hitler’s foreign-policy outlook away
from the idea of a British alliance, around which his pre-1933 diplomatic
conception had been based, to one in which Britain and France became obstacles
to be overcome or set aside. The development during 1937 of a policy ohne
England, without England, has been explained too often to be repeated here, but
it represented another important step away from conservative world policy and
the search for German colonies.33
The colonial question united a great many nationalists in Germany. It was
always assumed that at some point the return of colonies would be negotiable,
and in the context of declining world trade and raw material shortages German
leaders expected colonies to play an important part in sustaining Germany’s
world-economic position. When, in 1936, Hjalmar Schacht, minister of
economics, and a spokesman for conservative business circles in Germany, began
his own program of negotiations with western statesmen over the return of
colonies, he did so from traditional social-imperialist motives: colonies could be
used to divert the enthusiasm of radicals in the Nazi Party, and to soften economic
conditions at home in order to avert a revival of socialist agitation.34 Hitler paid
lip-service to this aspect of revisionism for as long as it gave him the prospect of
keeping diplomatic lines open to London, but there seems little doubt that he
was not essentially committed to the return of colonies at that juncture though
he would not have refused them had the opportunity arisen. He argued in the
second volume of Mein Kampf that German interests lay fundamentally in “the
strengthening of continental power by the winning of new soil and territory in
Europe,” a priority to which he remained consistently committed.35
Indeed, negotiations over colonies foundered on British insistence that
Germany should trade a colonial settlement for promises of good behavior in
eastern Europe, and were finally broken off in March 1938, after Schacht had
been forced to resign as economics minister. Göring told a British contact in
February 1937 that the regime wanted “a free hand in Eastern Europe” but was
happy to leave colonies to the British.36 The changing focus of German foreign
policy in 1936–37 was strongly driven by ideology rather than Realpolitik.
Misjudging Hitler 101
Intervention in Spain, closer ties with Japan and Italy, the gradual rejection of
any deal with Britain that did not grant a free hand to construct living-space in
the east, all stemmed from the desire to confront international Marxism and to
remodel the “east” in Germany’s favor. This reorientation was not achieved
without important political changes, since much of it was driven by the more
radical elements in the Party who regarded the regime’s conservative allies in
the armed forces, the economy, and the ministerial apparatus as a brake on
policies that were more assertively National Socialist. In this sense the changing
direction of German foreign policy was intimately related to the changing course
of German domestic politics. Between late 1936 and the spring of 1938
conservatives were slowly eased out of key areas of responsibility and were
replaced by Party appointments. Göring was given wide responsibilities for
economic policy and rearmament in October 1936, when the Second Four-Year
Plan was formally established, loosely based on Hitler’s memorandum of
August.37 Göring set out to develop a comprehensive program of autarky, or
self-sufficiency, which cut across Schacht’s aim to expand exports and living
standards. By November 1937 Schacht was sufficiently disillusioned with the
new direction in economic policy to resign, leaving the field clear for Göring
and another Party hack, Walther Funk, to dominate economic policy thereafter.
The armed forces were also unhappy with the implications of Göring’s
appointment, which relinquished, in their view, too much of a say in rearmament,
over which they had exercised close control since 1933. Conservatives in the
army feared, as did Schacht, that an irresponsible economic policy and excessive
levels of remilitarization would invite the danger of social unrest and hence undo
much of the valuable work done since the late 1920s in repairing Germany’s
military base. Nor were von Neurath and the foreign-office officials satisfied with
the increasing intervention of von Ribbentrop, and the tendency, explicit since
the 1936 negotiations with Japan and Italy but evident even before that, to
sidestep the foreign office altogether in the conduct of foreign affairs. They were
well aware that any congruence between Hitler’s policy and their conservative
nationalism was now more apparent than real, a distinction between what
Hildebrand has called “revisionist Great Power policy” and Hitler’s “expansionist
race policy.”38
The rumblings of discontent did not go unnoticed. When in November 1937
Hitler chose to reveal to von Neurath and senior military leaders his view of
how foreign and military policy would develop, the response was muted, even
hostile. In the spring of 1938 the political revolution was completed. The foreign
minister was replaced in February by von Ribbentrop, who personified the shift
in the power balance between the Party and the traditional elites. That same
month the war minister, von Blomberg, and the army commander-in-chief, von
Fritsch, were forced to resign on trumped-up charges of sexual scandal. Whether
or not Hitler was privy to the plots that ensnared Germany’s most senior soldiers
is still unclear, but he used the crisis as the opportunity to face the logic of his
growing personal power by abolishing the war ministry and establishing a
102 The Origins of the Second World War Reconsidered
Hitler did not, of course, do everything himself. He created what might be called
a “climate of permission” for the radicals in his entourage, and in the Party and
ministerial apparatus, to push on further and faster with plans for racial hygiene
and economic imperialism. However, they too could operate only to the extent
that their activities could be reconciled when necessary with Hitler’s current
priorities, ill-defined as they often were.
Under these circumstances it is essential to establish just what those intentions
were. Here the significance of the strategic memorandum of 1936 becomes clear.
In it Hitler developed lines of policy – some of them very specific – which
indicated the overall conceptual framework he was working with in the mid-
1930s. First, he expected some kind of major conflict, provoked by the threat
posed by Bolshevism and world Jewry, but by no means confined to them. There
are other fragments of evidence from the same period which show that he had
privately recognized that such a world-historical conflict was imminent and
inescapable. Second, he believed that Germany would survive that contest only
by developing into an economic and military superpower. The following passage
in the memorandum was put in italics: “The extent of the military development of our
resources cannot be too large, nor its pace too swift.” Such a program could be achieved,
Hitler continued, only by subordinating all other national and social tasks to
the one aim of strengthening Germany materially and psychologically. Finally,
Hitler remained committed to the idea of seizing Lebensraum at the appropriate
time, in order to secure the material foundation for global superpower status.43
These were broad goals and general expectations, certainly no blueprint for
aggression of the kind Taylor so resolutely criticized. The most obvious thing
that can be said about them is that between 1936 and 1945 Hitler’s regime gave
form and substance to them all. German society was heavily militarized and the
economy diverted to vast strategic projects; living-space was carved out of central
and eastern Europe; finally, Hitler decided in 1940 to launch the predicted
reckoning with Bolshevism, and in 1941 with the Jews. While it is obviously
tempting to argue that the thought was father to the deed, the actual course of
events after 1936 clearly depended on circumstances and opportunities, and was
neither pre-planned in any deliberate sense, nor remotely predictable. The
paradox can best be explained by recognizing that Hitler operated at two distinct
levels. The broad ideological and geopolitical aspirations acted as permanent
reference points or markers in the day-to-day conduct of affairs; on the other
hand Hitler acted like any politician in responding opportunistically to events
or an altered set of conditions.44 Improvisatory and reactive tactics in diplomacy
are no more inconsistent with a broad strategic vision than they are on the
battlefield.
Evidence of how the two levels interacted can be exemplified by a document
for which Taylor had little respect: the so-called “Hossbach memorandum.” The
memorandum, written by Colonel Friedrich Hossbach, an army adjutant, is the
record of a meeting at the Reich Chancellery on November 5, 1937. Taylor was
skeptical of its provenance and authenticity, and of the views it purported to
express, partly, no doubt, on grounds of scholarship, but partly because the
104 The Origins of the Second World War Reconsidered
document – taken at face-value – made it hard for him to argue that Hitler was
at heart a moderate revisionist with no discernible program. The authenticity
and accuracy of Hossbach’s account should now no longer be in doubt.45 Hitler
used the meeting with senior representatives of the armed forces and von Neurath
to give them a detailed insight into his views on foreign policy, and in particular
the acquisition of living-space for the German people in Europe in a future great
power conflict. This time, however, he gave a rough timetable, and an indication
of his immediate priorities. The final date for the conflict over living-space he
fixed at 1943–45, when the great military programs would be complete and
potential enemies not yet so heavily armed. The exact timing of the conflict and
against whom it would be fought depended on circumstances. Hitler never
confined himself to the promised conflict in the east. But he expected to be able
to incorporate Austria and to conquer Czechoslovakia, his initial aims, when
the opportunity arose, and that, he told his audience, might be sooner rather
than later.46
Both the projected date for conflict and the plans for Lebensraum are
independently confirmed – indeed Hitler’s colleagues made little secret of German
aims in central Europe. In February 1937 Göring was reported to have told a
British acquaintance: “Austria will come into our Reich of its own free will, but
if the Czechs remain unyielding, we shall have to take Böhmen und Mähren
[Bohemia and Moravia]. We don’t want the province of Slovakia…some day
the Corridor and Danzig must come back into the Reich.”47 It was Göring too
who told another British visitor in December 1937: “First we shall overrun
Czechoslovakia, and then Danzig, and then we shall fight the Russians.”48 Hitler
had told Goebbels earlier in 1937 that the “great world conflict” would come in
five or six years’ time, and perhaps be over by 1950, the approximate date by
which Albert Speer had been told to complete the “victory” buildings in Berlin.49
It is no doubt possible, even if the Hossbach memorandum is taken at face
value, to argue that it amounts to little more than treaty revision in its final stages,
as Taylor suggested. But such a perspective is convincing only if the issue of
Lebensraum is set aside, for the conquest of territory outlined in November 1937
took Germany beyond treaty revision, just as the hints of world conflict suggest
a vision of the 1940s that transcended revisionism entirely. There is not enough
evidence to demonstrate a clear blueprint, and it would be surprising if any
statesman, however committed to long-term aspirations, would draw one, but
the balance between long-term aims and short-term goals suggest that
Hildebrand’s idea of a Stufenplan, moving on step-by-step from immediate
revisionist goals to the seizure of living-space, and ultimately a contest for world
power, may more accurately reflect Hitler’s outlook.
Even the “step-by-step” formulation is, perhaps, too programmatic to be
explained by what Hitler actually said before 1939. At the height of the struggle
with the Soviet Union, in November 1942, he told an audience that three years
beforehand he “could not even have suspected this outcome”50 – who indeed
could have done so? Hitler did not plan to destroy Poland and wage war on the
Soviet Union in the sense that the Schlieffen Plan before 1914 predicated for
Misjudging Hitler 105
years a war with Russia and France, but neither course was inconsistent with
his more general views about the prospect of war and its proximate cause. On
the basis of the prewar evidence there is little that can be known with certainty
about Hitler’s detailed plans for the future, beyond the stated intention to destroy
Poland and dominate eastern Europe, as Taylor maintained. But there are
elements of a broader conception that cannot be disregarded if sense is to be
made of what followed on from the war with the Poles. These were not
necessarily shared by all of Hitler’s colleagues, and certainly not by the German
public as a whole, but because by the late 1930s Hitler had come to play such a
central role in the establishment and conduct of foreign policy it is his views
that bear the greatest historical weight.
First, Hitler was obsessed with the idea of waging war at some time. War
was for him a necessary condition of the international system, as the struggle
for survival was natural to the evolutionary process. War was the instrument
for altering the conditions for a people’s existence. War was the means to keep
a people aware of its historic racial mission and to build a population committed
to self-sacrifice and physical regeneration. When he looked at the Czech issue
in 1938 he chose war above any of the other options for resolving German–
Czech disputes. In 1939 Danzig and a revised Corridor might have been acquired
by negotiation, but in April 1939 he made it clear that war against Poland was
his preferred option. In the same speech in November 1942 he reflected on the
dangers that might have overtaken Germany if Britain had decided to give him
Danzig: “I have felt a shiver run down my spine when I read these proposals
again, and I can only thank Providence for dictating a different course…”. “War,”
as he wrote in 1936, was part of Germany’s “destiny.”51
Second, Hitler saw the solution to this destiny in turning Germany into an
economic and military superpower capable of absorbing any and every crisis
by striking the opponent with massive force. “If you want to lead a war,” he
told Goebbels in November 1937, “then this must be its name: destruction of
the enemy with every means.”52 The key to becoming a superpower was to seize
living-space in central and eastern Europe by force, so that Germany could get
access to the resources that the post-1919 German state lacked. At the least it
required the establishment of the Greater Economic Area (Grossraumwirtschaft)
that began to take shape in the late 1930s as a complement to the strategy of
domestic autarky. Food, materials, and labor were essential to sustain a large
war effort whoever the enemy was and however long the war lasted. Hitler’s
concern was to avoid the risk-taking of 1914 by ensuring that Germany possessed
sufficient armed might to be able to emerge from the imminent re-ordering of
the world system as the victor.
Third, Hitler remained a consistent enemy of Bolshevism and Judaeism to
the extent that ideology colored some of the critical choices made in German
foreign policy. That is not to say that Hitler had definite plans in the late 1930s
to annihilate Europe’s Jews and destroy the Soviet Union. But he believed that
Jewish world opinion was mobilized to stir up anti-German hatred in Britain
and America, and thus sow the seeds of conflict between them – a belief that
106 The Origins of the Second World War Reconsidered
fits ill with the idea that in foreign policy the Hitler regime was governed by
balance-of-power rationality. He also regarded Soviet Marxism as the most serious
threat posed to Europe’s future, and although he could make a pact with the
devil in August 1939, it is impossible not to see the decision taken in 1940 to
move eastwards as a product of the ideologically inspired battle with Bolshevism
as much as it stemmed from calculations of military expediency or Realpolitik.
Hitler’s worldview was shaped by these preconceptions and governed the choices
he made when he was face-to-face with the issues of ethnic cleansing and living-
space in 1941.53
Some of this outlook was shared by those around Hitler, even by wider circles
in the radical nationalist constituency. For German conservative nationalists,
however, the sum of Hitler’s views on foreign and military policy ran directly
counter to the conservative agenda. They did not want to risk waging war,
though they were prepared to make Germany defensible if war should break
out. They did not share the more fantastic elements of Hitler’s vision of
superpower status, though they hoped to constitute some form of central
European power bloc dominated by Germany through political pressure and
economic collaboration, which might restore Germany as an equal of the other
great powers. They disliked the risks for social peace that the costs of becoming
a military superpower entailed, and were fearful that if the German people were
forced to accept sacrifices through heavy arms’ spending there might be a return
to the revolutionary crisis of 1918–19. Finally, they did not share Hitler’s savage
xenophobia and the “barbarous utopia” it presaged in the 1940s, though many
of them became instruments in the attempt to achieve it. The problem faced by
those who might have opposed the radical aspects of Hitler’s strategy lay in the
shifting political balance, which favored a Hitler-centered system and promoted
the more radical elements of the Party to the forefront of German politics. The
possibilities for resistance or dissent did not disappear entirely, particularly in
the more routine day-to-day conduct of foreign or military policy, but they were
greatly restricted by the nature of the political structures that emerged after 1936.
Taylor had one more string to his bow. Using the work of the American
economist Burton Klein, Taylor argued in his 1963 supplement “Second
thoughts” that the economic evidence confirmed his argument that Germany
had not armed heavily in the 1930s, and had indeed a very narrow military
base, designed to achieve the kind of quick Bismarckian victories that a limited
revision required. Taylor felt that he, like everyone else in the 1930s, had been
tricked by Hitler and that there had been “no overwhelming advance in
armaments” after all. The arms’ gap was “pure myth.”54 Hitler, Taylor concluded,
was “pretending to prepare for a great war” but in fact “put butter before guns.”55
Klein’s work on the German economy, first published in 1959, was based
upon his experience as a young economist on the United States Strategic
Bombing Survey team in 1945, which was sent to Germany to assess the
economic and moral impact of bombing on the beaten enemy. The investigation
arrived at the conclusion, after a few weeks’ study, that Germany had armed
only lightly, in width, during the 1930s and had then maintained a “peace-like”
Misjudging Hitler 107
war economy in the first two years of war until forced to adopt fuller mobilization
in 1942. The motive for limited rearmament was believed to be political – Hitler
was thought to be anxious to improve living standards to avoid a social crisis.
Although there was little detailed evidence to suggest that these were actual policy
choices, Klein based his argument on the fact that German weapons’ production
was relatively modest in 1939–42, while the output of “consumer goods” was
maintained at almost a constant level in the first years of war.56 This was the
kind of opportunistic rearmament that matched the brief opportunistic wars Hitler
was believed to favor.
Once again the strategic memorandum of 1936 is the starting-point for a very
different interpretation of German rearmament in the 1930s. Until then military
procurement was left to the armed forces, who built upon the Second
Rearmament Plan drawn up in 1932 and introduced in 1933.57 Military spending
was relatively low during this early period, constituting only 1.3 percent of GNP
in the period 1932–35. Priority was given to the slow build-up of the military
infrastructure of the state which had been dismantled in 1919–20, and to the
reestablishment of a specialized armaments industry. The military capability of
the armed forces was extremely limited, and aroused fears that Germany’s
neighbors might intervene to prevent the restoration of German armed power.
In 1935 conscription was reintroduced, leaving the armed forces with the
substantial and expensive administrative task of rebuilding mass armed forces
after almost two decades of enforced disarmament. In 1936 the armed forces
suggested a new armaments program, published in August, which was to expand
the peacetime army to 800,000 men and complete the final phase of the
remilitarization programme. This, too, was an expensive short-term project, but
it was widely assumed that once the necessary level of military reconstruction
had been achieved the yearly costs of maintaining the system would reach a
plateau, or might even decline.58
This period of cautious military reconstruction was brought to an end in 1936,
when Hitler’s memorandum made it clear that he wanted “the premier army in
the world” and the mobilization of resources for war preparation on the largest
scale. The economy, Hitler wrote, has no other task than the “self-assertion of
the nation.” He continued: “Parallel with the military and political rearmament
and mobilization of our nation must go its economic rearmament and
mobilization, and this must be effected in the same tempo, with the same
determination, and if need by with the same ruthlessness as well.” The
memorandum concluded with the injunction that “the German economy must
be fit for war in four years.”59
There were practical explanations for the decision to increase the scale and
tempo of rearmament in 1936. The economy was stronger and the state more
stable than it had been in 1933, when a great rush for arms might have
compromised the economic revival and resurrected social tensions. The
international situation deteriorated sharply in 1936 with the Ethiopian crisis,
during which the western states showed their willingness to resort to economic
pressure on Italy, and with the Civil War in Spain. Hitler sensed that the
108 The Origins of the Second World War Reconsidered
unraveling of the existing order was now underway. In February 1937 Goebbels
recorded in his diary a conversation with Hitler: “He expected a great world
conflict in five or six years. In 15 years he would have liquidated the Westphalian
peace. He developed grandiose visions of the future. Germany will be victorious
in the coming conflict or will live no more.”60 For Hitler the most dangerous
development was the modernization and rearming of the Soviet Union, and he
referred to the “menacing extent of this development” in the memorandum.
When he discussed rearmament with Goebbels in January 1937 he measured
German achievements against “Russia’s strength”; in a speech that same year
Hitler told his audience that on the issue of rearmament “the USSR as the leader
of the tempo should show the way.”61
Above all, rearmament was governed by Hitler’s conception of the nature of
modern war. He accepted the widely held view that future conflicts, like that of
1914–18, were likely to be on a large-scale and costly, drawing on the full material
and moral resources of the nation. Here he was at one with much of the military
leadership in Germany, which had since the mid-1920s been urging the case
for a “defense-based economy” (Wehrwirtschaft). The concept was widely discussed
and understood in 1930s’ Germany. At root it represented a strategy designed
to avoid the mistakes of the First World War: the provision of adequate foodstuffs
to prevent internal unrest; the supply of essential war materials from domestic
resources to avoid the effects of blockade; the preparation of detailed mobilization
plans for industry and labor to ensure their rapid and effective conversion to
war production; the reorientation of domestic priorities to ensure that
developments in key areas such as transport, energy, and communications
conformed with potential military requirements.62
The object was to ensure that Germany would be capable of fighting a “total
war” if called upon to do so (though it obviously did not exclude fighting wars
of lower intensity). Hitler did not have a particular war in mind. His aim seems
to have been to create the sinews of a military and economic superpower which
would then be in a position to emerge victorious from any possible conflict that
might develop in the 1940s, a vision not very different from the one that had
driven the Soviet build-up in the 1930s. The record that we have of Hitler’s
thoughts on the nature of war preparation all point in this direction, most
famously in his address to the generals on May 23, 1939:
Everybody’s Armed Forces and Government must strive for a short war.
But the Government must, however, also prepare for a war of from ten to
15 years’ duration. History shows that wars were always expected to be
short. In 1914 it was still believed that long wars could not be
financed....However, every state will hold out as long as it can....The idea
of getting out cheaply is dangerous, there is no such possibility.63
In 1939 Colonel Thomas, head of the OKW war economy staff, recorded
Hitler’s view “that any mobilization must be a total one,” and in September 1939,
when Britain and France declared war, that is what he ordered.64
Misjudging Hitler 109
Throughout the period 1936–39 the language that Hitler used to describe
mobilization and war preparation was entirely consistent with the views expressed
in the memorandum – the nature of military preparation “cannot be too large,
nor its pace too swift.” The view that he deliberately ran a calculated risk by
preferring butter to guns is a speculation rooted in nothing more than the
impression that Germany was less-heavily armed in 1939 than the generation
that had lived through the 1930s, Taylor included, were led to expect. From
1936 the regime made no attempt to hide the fact that sacrifices would have to
be made to satisfy the broad ideas of the “defense-based economy,”65 The clearest
way to demonstrate the military priority of the regime lies in the macroeconomic
picture. By 1938 Germany’s GNP was 39 percent greater than in 1928, the pre-
Depression peak, but aggregate consumer expenditures were only 9 percent
higher than a decade before, and in per capita terms had increased by only 4
percent, an increase largely accounted for by the higher proportion of adults in
the population in the late 1930s.66 As a proportion of national income, private
consumption fell from 71 percent in 1928 to 59 percent in 1938, a fall of
exceptional magnitude in an economy the size of Germany’s. The additional
spurt of growth in the economy in the late 1930s was almost all accounted for
by war-related projects and high state spending.
The consequences of the increased tempo in 1936 were many. The military
budget expanded rapidly, taking up 17 percent of GNP in 1938–39. In the last
peacetime year 52 pfennigs out of every mark the German government spent
went on defense. These were not remotely moderate proportions. In 1913, at
the height of the pre-1914 arms race, the German government spent an estimated
3 percent of GNP, and devoted 24 percent of a much smaller state budget to
defense purposes. In the 1960s western states spent between 3 and 7 percent of
GNP on the military. There can be no question but that the level of military
spending in the late 1930s in Germany was exceptionally high by any peacetime
measurement.67
Why then did Klein, and afterwards Taylor, argue that German preparations
were limited? Part of the explanation lies in a misreading of the regime’s economic
priorities in the 1930s. In 1936 Hitler used the August memorandum as the
opportunity to bring the economy into relation with his military thinking. The
instrument for achieving this was a new Four-Year Plan organization, set up in
October 1936 under Göring who saw his new office as the centerpiece of a broad
strategy for creating a defense-based economy. The Four-Year Plan was ostensibly
concerned with the achievement of self-sufficiency, or autarky, in a range of sectors
deemed to be vital for a blockade-free war economy. These included chemicals,
iron-ore production, aluminum, synthetic rubber, synthetic textiles, synthetic fuel
oil and foodstuffs. There was never any intention to achieve full self-sufficiency,
and ultimately the regime intended to enlarge the resource base by taking over or
controlling the material resources of central and eastern Europe. Nevertheless the
programs that were initiated were very large in scale, absorbing almost two-thirds
of all industrial investment in Germany between 1936 and 1939. Investment in
the new iron-ore and iron producer set up in 1937 – the Reichswerke “Hermann
110 The Origins of the Second World War Reconsidered
Göring” – topped 800 million RM before the outbreak of war; state investment in
synthetic rubber production was 280 million RM, out of a total investment of
940 million. These sums were vastly greater than the amounts spent on weapons
production, since they were large, complex, and capital-intensive sectors. They
were the fruit of what Hitler had called “economic rearmament”, a concept now
generally described by the term “indirect rearmament” to distinguish it from the
regular military budget which devoted considerably less to investment purposes
and spent a great deal on wages and administration.68
Indirect rearmament took many other forms. Agriculture was given subsidies
and technical aid to try to raise yields. Labor was retrained in skills more
appropriate for a war economy, 1.2 million in total. The Four-Year Plan also
became involved in expanding the machine-tool industry, and the research and
development of a range of new substitute products. The broad framework for
war preparation set up by Hitler in 1936 meant that much of the economy was
engaged by 1939 on war-related activity, either direct or indirect. By May 1939
it was estimated by Reichsqruppe Industrie (Reich Group Industry) that 21 percent
of the workforce was engaged on direct orders for the armed forces (28 percent
in manufacturing).69 If precise details on the labor force of the other strategic
sectors built up under the Four-Year Plan were available, the final claim on the
civilian workforce for war-related activity would be considerably higher, perhaps
as high as one-third. To have more than one-quarter of the manufacturing
workforce engaged on war contracts suggests not a regime committed to a
strategy of “butter before guns” but an exceptional diversion of resources away
from civilian consumption and exports.
The second explanation for Klein’s argument lies in the comparatively low
level of finished weapons produced in 1939. Taylor had scant regard for British
and French rearmament, but he was surprised to discover that Germany was
producing at levels that were no higher. The reason for the apparent discrepancy
between national commitment to defense and the output of finished weapons
lies partly in the large diversion of resources in Germany to establish the
economic foundation for future war-making (raw materials and factory capacity),
but largely from the fact that German forces were not yet prepared for a general
war in 1939, and had not expected to fight one. The major arms programs were
set up only in 1938 and 1939, and were far from ready when war broke out. In
the summer of 1938 an explosives plan was drawn up which dwarfed the output
figures of the First World War; in October Göring was ordered by Hitler to
treble the general level of arms output and to expand the air force five-fold; in
January 1939 Hitler authorized the naval Z-Plan (Z = Ziel, or “goal”) for a large
battle-fleet by the late 1940s, a decision that has encouraged some historians to
see Hitler moving towards a global strategy, Weltpolitik, in the late 1930s.70 All of
these many programs were to be realized by 1942 at the earliest, the time when
Hitler had suggested in the “Hossbach” meeting that German armaments would
reach their peak in relation to those of other great powers.
These armaments plans were to provide Germany with its superpower
status. They are entirely incompatible with a policy of short local wars and a
Misjudging Hitler 111
Notes
1 A. J. P. Taylor, “1939 revisited,” German Historical Insitute, London, Annual Lecture
1981, 1982, p. 9.
2 See A. Sisman, A. J. P Taylor: A Biography, London, 1994, pp. 294–301; R. Cole, A.
J. P. Taylor: The Traitor Within the Gates, London and New York, 1993, pp. 190–203.
3 Taylor, “1939 revisited,” pp. 14–15.
4 A. J. P. Taylor, A Personal History, London, 1983, p. 299.
5 A. J. P. Taylor, “The supermen: Hitler and Mussolini,” in Europe: Grandeur and Decline,
London, 1967, p. 221.
6 See particularly W. D. Smith, The Ideological Origins of Nazi Imperialism, Oxford, 1986,
esp. pp. 244–52.
112 The Origins of the Second World War Reconsidered
43 Noakes and Pridham, Nazism, vol. 2, pp. 282–4. On Hitler’s world aims, see the
discussion in M. Michaelis, “World power status or world dominion?” Historical
Journal, vol. 15, 1972, pp. 345–59.
44 This was a conclusion arrived at by the prosecutors at the Nuremberg Trials. See
Michaelis, “World power,” p. 336.
45 See J. Wright and P. Stafford, “Hitler, Britain and the Hossbach memorandum,”
Militärgeschichtliche Mitteilungen, vol. 42, 1987, pp. 78–84, for a full discussion of its
authenticity. See, too, B.-J. Wendt, Grossdeutschland: Aussenpolitik und Kriegsvorbereitung
des Hitler-Regimes, Munich, 1987, pp. 11–37.
46 Minutes of the conference in the Reich Chancellery, November 5, 1937, Documents
on German Foreign Policy, series D, vol. 1, pp. 29–39.
47 Report of meeting with Göring, July 28, 1937, in the Christie Papers, 18/1 5.
48 Lord Gladwyn, The Memoirs of Lord Gladwyn, London, 1972, p. 66.
49 E. Fröhlich (ed.), Die Tagebücher von Joseph Goebbels: Sämtliche Fragmente, 4 vols, Munich,
1987, vol. 3, p. 55; A. Speer, Inside the Third Reich, London, 1970, p. 154.
50 W. Maser (ed.), Hitler’s Letters and Notes, New York, 1974, p. 189–90.
51 Noakes and Pridham, Nazism, vol. 2, p. 282; Maser, Hitler’s Letters, p. 190. On Hitler’s
commitment to war, see J. Fest, “Hitlers Krieg,” in N. Frei and H. Kling (eds), Der
nalionalsozialistische Krieg, Frankfurt am Main, 1990, pp. 103–21.
52 Goebbels, Tagebücher, vol. 3, p. 348.
53 See particularly Jacobsen, Aussenpolitik, pp. 446–8, 452–60; K. Pätzold,
“Antikommunismus und Antibolschewismus als Instrumente der Kriegsvorbereitung
und Kriegspolitik,” and Y. Bauer, “Antisemitismus und Krieg,” both in Frei and Kling
(eds), nationalsozialistische Krieg, pp. 122–36, 146–61. Bauer concludes: “Hitler’s war
was from first to last a war against the Jews.”
54 Taylor, “1939 revisited,” pp. 6–7.
55 Taylor, “Second thoughts,” in Origins (1963 edn), pp. 17–18.
56 See B. H. Klein, Germany’s Economic Preparations for War, Cambridge, MA, 1959, and
an earlier article which Taylor did not see, “Germany’s preparations for war: a
reexamination,” American Economic Review, vol. 38, 1948. The genesis of the thesis of
limited war is discussed in J. K. Galbraith, A Life in Our Times, London, 1981, pp.
212–25.
57 H. J. Rautenberg, “Drei Dokumente zur Planung eines 300,000-Mann Friedenheeres
aus dem Dezember 1933,” Militärgeschichtliche Mitteilungen, vol. 22, 1977, pp. 103–39.
58 Details in W. Deist, The Wehrmacht and German Rearmament, London, 1981, pp. 36–
53. Costs for the army were to rise sharply during 1937–39 and then, for the 1940s,
fall back to an annual level little higher than that of 1936.
59 Noakes and Pridham, Nazism, vol. 2, p. 283. In his “Secret book” (p. 96) Hitler had
written: “[I]n the future the enlargement of people’s living space…will require staking
the whole strength of the people.”
60 Goebbels, Tagebuch, vol. 3, p. 55.
61 Goebbels, Tagebuch, vol. 3, p. 33; Notes of a 1937 speech by Hitler, Christie Papers,
180/1 5.
62 On the theory of Wehrwirtschaft, see W. M. Stern, “Wehrwirtschaft: a German
contribution to economics,” Economic History Review, series 2, vol. 13, 1960–1, pp.
270–81. For a useful summary of the concept, see E. Hesse, “Wehrwirtschaft auf
lange oder kurze Sicht?’ Der deutsche Volkswirt, vol. 10, 1936, pp. 1,384–5. The
background can be found in H.-E. Volkmann, “Aspekte der nationalsozialistischen
‘Wehrwirtschaft,’ 1933 bis 1936,” Francia, vol. 5, 1977, pp. 513–38.
63 Documents on German Foreign Policy, series D, vol. 6, p. 577.
64 IWM, EDS files, Mi 14/377 (file 2), Thomas memorandum, March 28, 1939, p. 2;
Mi 14/328 (d), OKW conference, September 3, 1939, pp. 1–2.
65 In December 1936, for example, Germany’s most important economic journal, Der
Deutsche Volkswirt (vol. 11, December 24, 1936, p. 625) included the following in a
Misjudging Hitler 115
leading editorial on government economic policy: “It requires the sacrifice of living-
standards, at least temporarily…in the end it is better to limit the demand for
luxuries in order to build up ourselves what is necessary for food, clothing, housing
and [military] security.”
66 Klein, Economic Preparations, pp. 251–3; C. W. Guillebaud, The Economic Recovery of
Germany, London, 1939, pp. 204–6.
67 R. J. Overy, War and Economy in the Third Reich, Oxford, 1994, p. 20.
68 On the Reichswerke, see Overy, War and Economy, pp. 190–1; on chemicals, A.
BagelBohlan, Hitlers industrielle Kriegsvorbereitung, 1936 bis 1939, Koblenz, 1975, pp.
117–21; on synthetic rubber, G. Plumpe, “Industrie, technischer Fortschritt und
Staat. Die Kautschuksynthese in Deutschland, 1906–1944/5,” Geschichte und
Gesellschaft, vol. 9, 1983, p. 594.
69 Statistical material on the German manpower position during the war period, July
31, 1945, table 7, in IWM, Speer Collection, FD 3056/49.
70 On naval rearmament, see M. Salewski, Die deutsche Seekriegsleitung, 1939–1945, 2
vols, Frankfurt am Main, 1970, vol. 1, pp. 58–65; see, for example, G. Weinberg,
“Hitler and England, 1933–1945: pretense and reality,” in Germany, Hitler and World
War II, Cambridge, 1995, who argues that changed armaments priorities indicate a
clear intention to turn westward in 1939. This is a view that sits ill with the reality
of German naval and air planning. Neither service was ready to face Britain in 1939,
and their operational studies carried out that year made this situation clear to senior
German commanders.
71 For Hitler’s views on massive military build-up, see the interesting remarks of Albert
Speer in Speer interrogation, July 13, 1945, in IWM, Box 8368:
[Hitler] anticipated an intensification of the war…he had repeatedly
drawn attention to the dangers of the second front or of additional theatres
of war....He knew the supply figures of the last war in detail and could
reproach us with the fact that output in 1917/18 was higher than we could
show for 1942. I only knew that these were the requirements which had
been fixed in his mind for a long time. They were in nearly every case
three to six times the armament production of 1941.
72 Noakes and Pridham, Nazism, vol. 2, p. 282.
7 Appeasement
It is all too easy to comment upon the deficiencies contained in a book published
thirty-eight years ago on the basis of the then available evidence, and to list the
changes that would be needed to bring that volume up to date with more recent
scholarship.1 Both the questions asked by historians, and the materials open to
them (especially in respect of twentieth-century sources) change significantly from
one decade to the next. If historians are, in E. H. Carr’s phrase, part of a vast
caravan winding through time, it is hardly surprising that perspectives about
“appeasement” have altered between 1961 and 1999 – a much more considerable
period of time than that between the end of the Second World War and the
publication of A. J. P Taylor’s book. Since the past three decades have also seen
the opening up of the vast trove of British official records2 on the interwar years,
it is inconceivable that The Origins of the Second World War would not be “dated”
in many respects – as its author later acknowledged.3 What may perhaps be
more surprising is the extent to which many of Taylor’s judgments and (for want
of a better word) “hunches” have stood the test of time. This was true when
the first edition of this collection of essays appeared in 1986; and because there
have been no dramatic advances in the scholarship on appeasement, it remains
true today.
A greater difficulty in an essay such as this is to deal with a single strand –
that of British appeasement policy – in isolation. To do so is difficult not merely
because British attitudes and actions were, in Taylor’s book, integrated into the
overall story of why the Second World War occurred but because our own
judgments of how well-founded, say were Whitehall’s worries about the size of
the Luftwaffe will be affected by new researches on German aerial rearmament.
Similarly, our assessments of British policy towards Poland, Russia and the USA
in the 1930s can be placed in a different light by newly released archival materials
from those countries, as well as from France, Japan and other actors. Above all,
the issue of how well, or how poorly, the British understood Hitler’s real
intentions can be fully analyzed only by reference to scholarship on German
policy, which is outside the bounds of this particular essay.4 Students wishing to
comprehend British appeasement will always need to understand other, non-
British, perspectives as well.
Appeasement 117
While this policy of revision worked successfully in the two crises of 1938 –
that is, concerning the incorporation of Austria and the Sudeten Germans into
the Reich – it broke down over the Polish issue in the year following. By that
time, and especially after the German acquisition of the rump state of
Czechoslovakia in March 1939, British public opinion wanted Chamberlain and
his cabinet colleagues to take measures to “stop Hitler.” In a tragicomedy of good
intentions going astray, the British government – with France in its wake – found
itself tied into binding military commitments to a stubborn and reckless Polish
regime under Beck. Since Chamberlain was still determined to settle things
peacefully and was sending messages to that effect to Berlin, Hitler felt that he
could proceed to solve the Danzig dispute by hints of action but without serious
risk of war with the west; indeed, with the very strong chance that Chamberlain
would arrange things in just the same way as he had done at Munich. It was
only because the Poles declined to be as conciliatory as the Czechs that Hitler’s
– and Chamberlain’s – expectations went awry. Although none of them planned
to be at war with each other, by September 3 that very state of affairs existed.
Far from being a maniac, Hitler had acted in a rational (if calculated) manner.
But the appeasers, having willingly undermined the European status quo on
numerous occasions in the 1930s, had now bungled things, and had gone to
war “for that part of the [1919] peace settlement which they had long regarded
as least defensible” (p. 335).
The cries of outrage which greeted the publication of such views thirty-eight
years ago are understandable, and perhaps even more so now in the light of
recent scholarship on the Holocaust.7 On the one hand, there was Taylor’s refusal
to make moral judgments, or to give much weight to the significance of Nazi
ideology, domestic politics, and racial doctrines. Then there were the critics who
were alarmed at the possible implications of Taylor’s suggestion that German
hegemony in Europe was “natural,” and ought not to have been resisted; if that
applied to Hitler’s Germany in the 1930s, might it not also apply to Khrushchev’s
Russia in 1961 – the year of the Berlin crisis as well as of the publication of
Taylor’s book? Above all, there were those infuriated by his flippantly throwaway
style and sweeping remarks: that Hitler had no “preconceived plan” (p. 98); or
that the Hoare – Laval scheme was “perfectly sensible” (p. 128); or that Munich
was “a triumph for all that was best and most enlightened in British life” (p.
235); or that “it seems from the record that [Hitler] became involved in war
through launching on 29 August a diplomatic manoeuver which he ought to
have launched on 28 August’ (p. 336). All this was strong stuff
The greatest indignation was, of course, reserved for Taylor’s implicit (and
sometimes explicit) “de-demonization” of Hitler, an interpretation that many
critics thought untenable on both moral and factual grounds.8 By contrast,
Taylor’s view of the British appeasers appeared less controversial, if only because
the prevailing image of Neville Chamberlain was already a negative one. The
Origins of the Second World War may have portrayed Britain’s appeasement policy
in a more dynamic and purposeful way than was hitherto imagined, but by
showing how eager London was to comply with, or even anticipate, the führer’s
Appeasement 119
wishes, it still seemed unflattering. Consequently, the notion that policy was both
unwise and immoral was scarcely shaken, as could be seen in Gilbert and Gott’s
swingeing indictment, The Appeasers, published two years later.9 Taylor himself
might not wish to draw moral judgments about Chamberlain, or Samuel Hoare,
or Horace Wilson, but many other historians were very willing to do so.
Apart from his general argument that British appeasement policy unwittingly
contributed to the outbreak of the Second World War, Taylor offered detailed
remarks on the leading British personalities in this story, the arguments they
deployed, and the phases they went through. All of this was based upon published
British and German diplomatic documents for the interwar years, and upon older
memoirs and biographies. Within another decade, however, virtually the entire
official records for the 1930s (not to mention more and more private collections)
were opened to historians – giving them the unexpected opportunity to measure
Taylor’s book, and other works, against the government’s own documents. By
an Act of 1967 the Labour government reduced from fifty years to thirty years
the period of time in which public records were kept confidential, though there
were exceptions. Had that not happened, historians today would only now be
digesting the cabinet files on, for example, “colonial appeasement” in 1937, and
much of what follows in this essay would have been impossible to write. For
the past thirty years, therefore, there has been a vast swathe of scholarly books
and articles dealing with British appeasement policy, although it would be fair
to say that, during the past decade, there have been distinct signs of a tapering-
off in the originality of viewpoints and arguments.
Many of Taylor’s observations, it ought to be said at once, have stood the
test of time rather well. His somewhat cynical view of statesmen, strengthened
no doubt by his years of studying Bismarckian diplomacy, put him in good stead
in describing the role of people such as Simon, MacDonald and Hoare. Recent
biographies – of Hoare, Simon, even Eden – have fleshed out their personalities
but have not fundamentally altered Taylor’s portrait.10 His coverage of Sir Nevile
Henderson’s debilitating functions as British ambassador in Berlin – constantly
toning down the firmness of the foreign office’s messages, and making
deprecating noises to his German listeners about the Jews or the Czechs or the
Poles – have required no amendments now that the files are open.11 Above all,
Taylor’s observations on Halifax, although brief, ring very true: the foreign
secretary’s aloofness, his sense of conscience (occasionally fostered by his foreign-
office staff) his sensitivity to what the Conservative Party and the country at
large would think fair, made the “Holy Fox” one of the few people – perhaps
the only one – who could influence the prime minister during the Munich and
Prague crises (p. 188–9 and ff.).12
By contrast, the portrait of Chamberlain in these pages seems one-dimensional.
Taylor captures the prime minister’s personal decisiveness and sense of purpose,
the businessman-turned-politician who knew how to run an organization on
efficiently utilitarian lines; and many a later book, benefiting from the cabinet
papers, has shown how dirigiste was the administration that Chamberlain
controlled.13 On the other hand, historians who have gained access to the prime
120 The Origins of the Second World War Reconsidered
minister’s private letters, especially those to his sisters, have shown Chamberlain
to be increasingly uncertain about Hitler as 1938 turned into 1939. One week’s
expressions of confidence that all was going well, and that the likelihood of war
was fading, mingle with much more gloomy assessments in the week following.14
No doubt these shifts of mood can be explained in part by the fluctuating
reports of happenings on the continent, but Chamberlain’s letters suggest a more
complex figure than Taylor presents: sometimes briskly efficient, sometimes proud
and privately boastful of his successes, sometimes worried and even bewildered
at the turn of events. Perhaps this is why recent studies present such contradictory
interpretations, with Aster returning to the “guilty men” view, Charmley
portraying the prime minister as a far-sighted realpolitiker, and Parker seeking to
draw a balance. We still await the second volume of Professor Dilks’s biography
of Chamberlain15 before we have the full picture, but the image which is emerging
has already shown how difficult it is to assess the prime minister’s character in
a few swift sentences. As it is, this continued focus on the “high politics” of the
Chamberlain cabinet distracts scholarship from broader issues.
On the internal politics of Great Britain in the interwar years, and their effect
on foreign policy, Taylor has not much to say, although the remarks he makes
are usually accurate enough. For example, he shrewdly notes that Baldwin’s pro-
League statements in 1935 were intended to outwit the Labour Party just as a
general election was pending. Similarly, Taylor’s discussion of the spring 1939
considerations of an Anglo-Russian alliance to assist Poland nicely captures the
dilemma in which Chamberlain found himself; if London negotiated with
Moscow (which the prime minister and his colleagues greatly disliked) and was
successful, it would be seen as vindicating the arguments of such varied critics
of the government as Churchill, Lloyd George, and the Labour Party; if London
refused to negotiate, or did so and failed to reach a settlement with Stalin, it
would be blamed – by the British public, by a suspicious Hitler, and by posterity
(as indeed it was). While Taylor does not provide Maurice Cowling’s full picture
of the internal political dynamics of appeasement diplomacy – that is, of a
Chamberlain needing a successful “deal” with Germany not only to preserve
peace but to secure his own political position and confound his critics to the left
and the right16 – he does hint at this domestic dimension.
Researches into the internal political aspects of British policy have therefore
tended to supplement Taylor’s version rather than replace it. Cowling, for
example, has gone even further in his argument, suggesting that Chamberlain’s
deeper concern was that another great war (with its total mobilization of national
resources) would lead to significant advances by Labour and the Trades Union
movement – just as the First World War had done. The preservation of peace
was, therefore, intricately linked with the fate of the Conservative Party, a fact
which (Chamberlain felt) the more reckless or “irregular” Tories like Churchill
did not comprehend.17 Just as such an account does not contradict Taylor, so
also do the writings about the “anti-appeasers” scarcely affect his picture: for
the message of such works has generally been that Chamberlain’s opponents,
too, were uncertain of how to respond to the unprecedented circumstances of
Appeasement 121
the late 1930s. There were all sorts of divisions among the ranks of the
Conservative critics, although this still remains inadequately researched (there
is no study of the Tory Party and foreign policy, for example, and little use of
constituency records). What we do know is that some Conservatives disliked
the appeasement of Germany, but strongly urged the appeasement of Italy; most
of them – even Churchill – softened their attacks when the prospect of being
invited to join the government seemed closer; the “Eden group” tried to keep
its distance from the “Churchill camp,” and so on. In the same way, the Labour
Party was neither as forthright nor as consistent in its criticism of appeasement
as it later liked to think. Attlee and his colleagues (who deserve fuller study)
were very wary of being portrayed as warmongers, and warier still of co-
operation with the old imperialist war-horses on the right of the Conservative
Party. The revelation of such uncertainties gives us a better idea nowadays of
how Chamberlain was able to preserve his commanding position in British
politics for so long.18
British public opinion – the press, varied pressure-groups and the legendary
“man in the street” – is not a key feature in The Origins of the Second World War.
To be sure, Taylor refers to that general mood of pacifism, non-interventionism,
and dislike of “foreign politics” which conditioned the entire interwar period
and made every administration, from Lloyd George’s coalition onward, reluctant
to accept commitments in Europe and eager to see an amicable settlement of all
international disputes. Over the past three decades, the study of British public
opinion – especially the ideas and movements associated with pacifism, the Peace
Ballot and the League of Nations’ Union, but also strands of opinion on the
Right19 – has become a major growth industry. The press’s views of Germany,
the Left Book Club, the public’s attitude towards the Abyssinian crisis, or the
Spanish Civil War, have all found their historians.20 Here again, we are talking
about additions to Taylor’s version of events, not challenges to it.
Public opinion’s two most significant disruptions of the official policy of
appeasing the dictators occurred, first, in late 1935, when the news of the Hoare–
Laval pact provoked an explosion of discontent against this undermining of
League of Nations’ principles; and, second (and more importantly), in the spring
of 1939, when large segments of British public opinion, including many former
supporters of Chamberlain’s appeasement policies, decided that Hitler had to
be stopped and urged all manner of embarrassing proposals upon the
government: guarantees to east-European states, an alliance with Russia, further
rearmament, closer ties with the French, and so on. Taylor gives a good account
of how Chamberlain became increasingly trapped between two uncontrollable
forces – the exogenous force of Hitler, moving to further actions or threats of
action, and the endogenous force of a resentful British public – but much more
might have been said about the vital change in mood and circumstances. There
is little or nothing in his account, for example, of the anger and disgust produced
in Britain by news of the Kristallnacht, (November 9, 1938), nor of the rabid
speeches of late 1938 in which Hitler denounced Chamberlain’s interfering
122 The Origins of the Second World War Reconsidered
diplomacy and proclaimed the Munich settlement a victory for brute force –
exactly the opposite of what the prime minister was saying.21
Where Taylor seems less correct is in his assumption (which he repeated on
many occasions in the 1960s and 1970s) that the Munich agreement was
overwhelmingly supported by the British press, with only Reynold’s News in
opposition. In fact, both left-of-center papers like the Manchester Guardian, the Daily
Herald and the News Chronicle, and the distinctly right-wing Daily Telegraph, wanted
a firmer line taken toward Nazi Germany, and were joined in this by many
individuals.22 What is even more significant, and until recently less wellknown,
were the persistent and very determined efforts made by Chamberlain and his
colleagues to control the media – by influencing the press lords and editors, by
getting critical talks suppressed on BBC Radio, by censoring the contents of the
newsreels shown at the enormously popular cinemas so – as to give to the world
the impression that the nation was behind the prime minister and his policies.23
In view of this recent evidence, the older idea of a general consensus in British
public opinion in favor of appeasement which broke only with the news of the
German entry into Prague in March 1939 now looks distinctly wrong. The much
more likely position was that opinion was already divided during the Czech
drama, although this was obscured by the combination of the government’s
censorship efforts, the reluctance of Chamberlain’s critics to appear as
warmongers, and the cautiousness produced by natural apprehension at the
prospect of a major war. As soon as the shudders of relief at the avoidance of
hostilities were over, however, the sense of unease returned, reinforced by one
of shame at the fate of the Czechs, and anger at Hitler’s speeches and programs.
Seen in this light, the uproar over the Prague crisis was but one step (even if the
most important one) in the dramatic switch of British public opinion against
appeasement.24 This metamorphosis of British opinion, and the government’s
efforts to steer it, still requires more work.
Appeasement – in the older sense of an attempt to settle differences by
negotiation and concession – was not a new feature in British diplomacy: as
historians have pointed out, many elements of appeasement went back to
Gladstone’s time, or even further.25 What was quite new, and altogether more
difficult for the British government to handle, was the unprecedented state of
the international system after 1919. By that time, the USA was by far the most
powerful financial and industrial (and, potentially, military) state in the world –
yet it rapidly abandoned most of its diplomatic responsibilities, even while the
ups and downs of its enormous economy continued to affect trade, investment,
and prosperity across the globe. The other great continent-wide power, Russia,
had been shattered by the First World War and was now ruled by the mysterious
and threatening Bolsheviks. The Austro-Hungarian empire had dissolved into
a cluster of intensely jealous rivals. By contrast, Germany’s territories (despite
reductions in size, especially in the east) remained basically intact; and its power
potential, as measured in terms of population, industrial capacity, and national
efficiency, was great – greater than France’s in the long run. If and when the
Germans organized themselves to assert their claims for a revision of the 1919
Appeasement 123
treaty, they would be inherently in a very strong position. Neither the “successor-
states” of eastern Europe nor a nervous, politically fragmented, and economically
weaker France would be able to resist for more than a relatively short period –
unless aided by another Great Power. Yet, with the USA excluding itself, and
the USSR in partly enforced, partly self-chosen isolation, only Britain remained;
and it found it less easy to escape into isolation, much as it wanted to.
This fundamental change in the international balances as compared with the
pre-1914 era Taylor captured very well, illuminating basic trends to which he
had already drawn attention in his important earlier work, The Struggle for Mastery
in Europe, 1848–1918.26 The First World War, then, had not “solved” the German
question: if anything, it had made it “ultimately more acute.” “If events followed
their course in the old ‘free’ way,” Taylor suggested, “nothing could prevent the
Germans from overshadowing Europe, even if they did not plan to do so” (p.
48). To be sure, Britain could have carried out her traditional balance-of-power
policy, but many things conspired to make that seem less useful than ever before.
In the first place, for the entire 1920s it was Germany’s weaknesses and France’s
(and even Poland’s) strengths which caught the eye. Second, as noted above,
the British public in the post-Versailles era did not want any further commitments
in Europe; and, like most British ministers, soon came to feel that the 1919
boundaries ought to be revised – by peaceful means, of course, and under the
aegis of the newly created instrument of the League of Nations. The fact that
Japan now appeared as a potential threat in the far east, where Britain had much
more substantial interests than any other European country, also made it easy
“to understand why the British felt distinct from the Powers of Europe and why
they often wanted to withdraw from European politics” (p. 68). After the
Abyssinian crisis and the Spanish Civil War had revealed not only a new potential
enemy in the form of Mussolini’s Italy but how ineffectual was the League of
Nations, the international reasons for settling German grievances seemed more
pressing than ever – or so it appeared to the firm-minded Chamberlain when
he took over in 1937.
In discussing the general external structure in which the British government
now had to carry out its diplomacy, Taylor’s book is very lucid. It is also
essentially correct in its portrayal of British policy toward potential allies among
the other Great Powers. Thus, London’s dismay at the difficulties of persuading
the United States’ government to do anything substantial either in Europe or in
the far east, and Neville Chamberlain’s personal suspicion that the Americans
were “all words, and no actions,” have been amply confirmed in the excellent
studies by David Reynolds and Callum MacDonald.27 In the same way, Taylor’s
picture of the far greater dislike shown by Chamberlain and his colleagues
towards the Soviet Union – as a general threat to the western order of things
and, more specifically, in the context of a possible Anglo-Russian alliance to
support Poland in 1939 – has not been shaken by the newer literature. And,
since Taylor distinguishes between this general mistrust of Russia, on the one
hand, and a (non-existent) policy of trying to provoke a German–Russian war,
on the other, his portrayal is much more balanced than those strained pro-
124 The Origins of the Second World War Reconsidered
to which the appeasers merely “lacked guts.” Watt felt that a future investigation
of the files might reveal, inter alia,
Some of these aspects are indeed referred to by Taylor, but only briefly, such as
defense weaknesses, or the influence of the treasury; and the role of the
dominions, or the empire as a whole, is not mentioned at all. Yet if one feature
of the historiography of appeasement since the opening of the official records
stands out, it is the massive attention which has been paid to the evidence of
Britain’s frightening economic and strategical–global weaknesses in the 1930s.
Taylor is brief but reasonably good in referring to “economic appeasement,”
presumably since the published German and British documents detailed at least
some of the efforts made by Whitehall to soften German resentments by offers
of trade credits, access to raw materials, exchange arrangements, and outright
loans; but his remarks upon the baneful influence of the treasury on British
rearmament now look very dated. As a flood of works has shown, it was simply
not true that a nice burst of Keynesian “pump-priming” by means of higher
armaments’ spending would have solved Britain’s problems, reducing
unemployment and strengthening the armed services. It is of course likely that
in the early 1930s some extra expenditure on the forces would have had beneficial
effects in strategic, industrial, and employment terms; but the amount of cash
that was needed to rebuild a two-ocean navy, to provide the Royal Air Force
with both its fighter defenses and its long-range bombers, and to equip the army
for a European field role – all of which the chiefs of staff desired – was well
beyond the industrial and financial capacity of the country. The long economic
decline, exacerbated by the world slump after 1929, had eroded the British
industrial base to an alarming extent. There were incredibly few skilled workers,
especially in the vital engineering trades. There were insufficient machine-tools.
There were few modern factories, and no modern shipyards. What was more,
simply throwing money at these problems could never produce easy and fast
solutions; it might, indeed, weaken the British economy still further by provoking
inflation, hurting the balance of payments, and producing bottlenecks. For such
an ailing patient, only a gentle stimulus seemed proper.36
By the late 1930s, the treasury’s arguments were proven to be correct, even
when – or, rather, especially when – it had lost its battle to keep defense spending
126 The Origins of the Second World War Reconsidered
[W]e cannot foresee the time when our defence forces will be strong
enough to safeguard our trade, territory and vital interests against
Germany, Italy and Japan at the same time....[We cannot] exaggerate the
importance from the point of view of Imperial Defence of any political or
international action which could be taken to reduce the number of our
potential enemies and to gain the support of potential allies.42
Appeasement 127
Here was an argument for appeasement which at first sight was utterly
compelling. Yet apart from a brief mention of Britain’s supposed vulnerability
to aerial attack – which Chamberlain used in order to cow cabinet critics during
the Munich crisis – the reader gains little sense from Taylor’s account of the
significant role of defense weaknesses in appeasement policy. This is not a charge
of negligence on his part: the mass of evidence was simply not available to
scholars in the late 1950s.
As if this catalogue of gloom were not enough, the global international crisis
of the 1930s threatened to split the British empire apart. Ever since 1919
Afrikaners and French-Canadians – not to mention, after 1921, the fiercely
independent Irish Free-Staters – had bitterly opposed any idea of “imperial
defense” and expressed even more hostility to the notion of being dragged into
a war in consequence of European quarrels. And while Australia and New
Zealand were more willing to cooperate with Britain, they too were worried that
European issues would divert resources from the more immediate danger of
Japanese aggression in the Pacific. In addition, the dependent empire was much
less tractable than in the days of Disraeli or Salisbury. A widespread Indian
nationalist movement, Egyptian discontents, a potential civil-war situation in
Palestine by the late 1930s, were all pinning down British troops and resources
and, last but not least, reinforcing Whitehall’s arguments for not being committed
to Europe.43 Moreover, these fissiparous movements within the empire could not
be completely separated from the external threats to the Mediterranean route and
especially to the British possessions in the far east – a region whose significance,
were one to measure it in terms of books published on that aspect of British
policy alone – overshadowed everything else!44
Not only did the chiefs of staff and the treasury have contradictory ideas of
the British government’s existing priorities – with one side pressing for more
defense spending, and the other pressing for financial stability – but their
prognostications for a future conflict, should one actually come, were actually
at odds. From the armed services’ viewpoint, Britain had a chance of successfully
fighting Germany and Italy only if the war were a long one, during which the
population and material resources of the empire could steadily be mobilized. In
the treasury’s opinion, Britain could afford to fight only a short war, since it
would very swiftly run out of gold and dollar holdings.45 Impaled on the horns
of this dilemma, was it surprising that cabinet ministers should endeavor to avoid
a conflict of any kind?
Because the official archives have revealed this catalogue of industrial,
financial, strategic and imperial weaknesses with which successive British
governments grappled in vain during the interwar years, the tendency of recent
writings has been much more emphatic (and even sympathetic) toward the
appeasers. In consequence, the “guilty men” interpretation of the 1940s and
1950s looks unbalanced and unfair. Far from finding Chamberlain’s policy in
the late 1930s inexplicable, it now seems quite understandable to many historians.
As one of them has put it: “If one begins to tot up all the plausible motivations
for appeasement…one sees that these are far more than enough to explain it. It
128 The Origins of the Second World War Reconsidered
was massively over-determined; any other policy in 1938 would have been an
astounding, almost inexplicable divergence from the norm.”46
All this newer evidence of British weakness affects Taylor’s arguments only
indirectly. To the extent that many of these writings have suggested that
appeasement was unavoidable and predetermined, they do place Chamberlain
and his colleagues in a more favorable light than that in which they appeared in
Origins. But such materials would probably not have affected his central thesis,
that the coming of war in September 1939 was an accident, and one caused
more by the erratic moves of the appeasers and the stubbornness of the Poles
than by Hitler’s own calculations. Nor, one suspects, would they have altered
his own skeptical view that the politicians rarely consulted “their military experts
in a detached way before deciding on policy. They decided policy first; and then
asked the experts for technical arguments with which this policy could be
justified” (p. 155). It was because of that habit, Taylor writes, that even when
British leaders used such “practical arguments” as aerial weaknesses during the
Czech crisis it was to reinforce their own conviction that appeasement was
morally right (p. 254).
Given the weight of this newer evidence, few historians today will be as cynical
and cavalier as Taylor was then about the role of military (or treasury) advice
on British policy. None the less, his remarks may be useful in reminding us that
strategic memoranda are not the “be-all and end-all” of historical causation, and
that we still have the task of properly integrating the newer evidence into our
larger understanding of what appeasement meant.
As noted above, the weakness of the older “guilty men” argument appeared
to be that it denounced Chamberlain and his colleagues for a failure both of
morality and of willpower without much appreciation (or knowledge) of the
difficulties under which British governments of the 1920s and 1930s labored.
By contrast, most of the later works have focused upon the seemingly compelling
strategic, economic, and political motives behind British policy at that time, but
without much concern for the moral and ideological aspects of it. That is to
say, the mass of cool treasury memoranda and the well-honed strategic
assessments of the chiefs of staff, available for everyone to see in the Public
Record Office, now occupy such a prominent position in the story that they are
in danger of overshadowing those very important personal feelings behind
appeasement: the contempt and indifference felt by many leading Englishmen
towards east-central Europe, the half-fear–half-admiration with which Nazi
Germany and Fascist Italy were viewed, the detestation of communism, the
apprehensions about future war.
Of course the warnings of the treasury and the chiefs of staff about Britain’s
impending financial and strategical bankruptcy were important; but the facts are
that such statements were not infallible, and that they were sometimes used by
Chamberlain to justify policies he already wanted to pursue. For example, as
Correlli Barnett and Williamson Murray have pointed out, both the chiefs of
staff and the cabinet were making some excessively gloomy predictions during
the Czech crisis. Germany’s own weaknesses were not considered. The value
Appeasement 129
of the Czech army was ignored. Britain’s vulnerability to aerial attack was
repeatedly stressed, but without consideration of whether the Luftwaffe would
or could throw itself against London while Germany was engaged in a central
European war. Furthermore, the cabinet minutes reveal that when some ministers
(Duff Cooper, Stanley) actually wanted to take a stronger stand against Hitler,
despite the risks to Britain and its empire, they were swiftly overwhelmed by
counter-arguments from Chamberlain and his friends.47 Objectors within the
cabinet had to be silenced, just as the press and the BBC had to be controlled.
Even when, by early-to-mid 1939, British public opinion was moving strongly
against appeasement, when Britain’s aerial defenses were much improved, and
when the dominions were more supportive of a firm line, Chamberlain and his
fellow-appeasers were still seeking, in secret rather than in the open, to buy off
Hitler. After Prague, making concessions to Germany was neither as logical nor
as “natural” as it might have been in 1926 and 1936; on the contrary, it seemed
to many a policy lacking in both practical wisdom and moral idealism. Yet it
was still being attempted by Downing Street, which suggests that the convictions
of individuals – in this case, Chamberlain’s – must play a central part in our
explanation of British policy, which cannot be fully understood simply in terms
of “objective” strategical and economic realities.
Appeasement, then, is not a simple phenomenon which can be defined in a
few sharp words. Older histories tend to see it as a shameful and bankrupt policy
of surrender to the dictator-states. Taylor has portrayed it as a series of well-
meaning bungles which eventually embroiled both Hitler and the west in a war
neither of them desired. Some scholars have seen it as a natural and rational
strategy in the light of Britain’s weaknesses in the world by the 1930s. Others
have pointed out that it was, albeit in a more intensified form, a normal
continuation of the British diplomatic tradition of attempting to settle disputes
peacefully.
Appeasement was, in fact, all of the above, and needs to be understood as
such. It also needs to be investigated at different levels of causality, so that
distinctions can be made between the nebulous, sometimes confused mentality
of the appeasers on the one hand, and the cluster of military or economic or
imperial or domestic–political motives which justified, or seemed to justify,
concessions to the dictators on the other. Only when it is approached in such
a way will the historians rise above simplistic one dimensional descriptions,
and deal with appeasement as the complex, variegated, shifting phenomenon
which it really was. This essay, then, closes with a call for further work on
appeasement. Taylor’s book, together with the opening of official and private
archives, provoked and inspired a flood of scholarship on British policy in the
1930s. Although the tide has ebbed, it should be clear that many subjects
remain inadequately explored. One can hope that the recognition of
appeasement’s complex nature will spur scholars to examine all aspects of
British policy before the Second World War.
130 The Origins of the Second World War Reconsidered
Notes
1 This essay is concerned with the text of the original (1961) edition of Taylor’s The
Origins of the Second World War, and not with the “Second thoughts” Foreword of the
1963 edition, nor with either “War origins again,” Past & Present, no. 30, 1965, or
“1939 revisited,” the 1981 Annual Lecture of the German Historical Institute,
London.
2 Intelligence records remain a notable exception, and thus far only official historians
have gained access to them. See F. H. Hinsley et al., British Intelligence in the Second
World War: Its Influence on Strategy and Operations, vol. 1, London, 1979. For efforts to
trace intelligence records by roundabout means, see D. Dilks, “Appeasement and
intelligence,” in D. Dilks (ed.), Retreat from Power, 2 vols, London, 1981, vol. 1, pp.
139–69; and Wesley K. Wark, The Ultimate Enemy: British Intelligence and Nazi Germany,
1933– 1939, Ithaca, NY, 1985.
3 Taylor, “1939 revisited”.
4 See Chapter 6, “Misjudging Hitler,” in this collection of essays; G. L. Weinberg,
The Foreign Policy of Hitler’s Germany, 2 vols, Chicago, IL, 1970 and 1980; N. Rich,
Hitler’s War Aims, vol. 1, New York, 1973; W. Carr, Arms, Autarky and Aggression,
London, 1972; E. Jäckel, Hitler’s World View, Cambridge, MA, 1981; K. Hildebrand,
The Foreign Policy of the Third Reich, London, 1973; Wilhem Deist et al. (eds), Germany
and the Second World War, vol. 1: The Build-Up of German Aggression, Oxford, 1990, are
all helpful here.
5 The general survey literature is now so large as to be almost beyond control; but
historiographical pieces to note are D. C. Watt, “The historiography of
appeasement,” in A. Sked and C. Cook (eds), Crisis and Controversy, London, 1976;
D. Dilks, “Appeasement revisited,” University of Leeds Review, 1972, pp. 38–49; P.
Kennedy, “Reading history: appeasement,” History Today, October 1982, pp. 51–3.
There are also very important analyses by German scholars such as B. I. Wendt,
G. Schmidt, G. Niedhart, W. Gruner, R. Meyers, and others – some flavor of which
can be gleaned from the important collection, edited by W. J. Mommsen and L.
Kettenacker, The Fascist Challenge and the Policy of Appeasement, London, 1983, and
summarized in part in P. Kennedy, “The logic of appeasement,” Times Literary
Supplement, May 28, 1982.
6 The works by Wheeler Bennett, Namier, and Churchill are very much in this tone,
as are pro-Moscow books and articles. For a recent restatement of this view, see
Sidney Aster, “‘Guilty Men’: The case of Neville Chamberlain,’” in Robert Boyce
and Esmonde M. Robertson (eds), Paths to War: New Essays on the origins of the Second
World War, London, 1989, pp. 233–68.
7 The literature on the Holocaust is massive, but see Raul Hilberg, The Destruction of
the European Jews, New York, 1985 edn; Michael Burleigh and Wolfgang
Wippermann, The Racial State: Germany, 1933–1945, Cambridge, 1991; Christopher
Browning, Ordinary Men: Reserve Battalion 101 and the Final Solution in Poland, New
York, 1992; Saul Friedlander, Nazi Germany and the Jews, New York, 1997. A useful
historiographical summary is Michael Marrus, The Holocaust in History, Hanover, NH,
1987.
8 Some sense of this outrage can be gleaned from contributions in E. M. Robertson
(ed.), The Origins of the Second World War, London, 1971, and W. R. Louis (ed.), The
Origins of the Second World War: A. J. P. Taylor and His Critics, New York, 1972. The
most sustained repudiation of the Taylor line is in volume 2 of Weinberg’s Foreign
Policy of Hitler’s Germany, which is meaningfully subtitled Starting World War II, 1937–
1939. D. C. Watt’s panoramic study also identifies Hitler as chiefly responsible for
the war: see How War Came: The Immediate Origins of the Second World War, 1938–1939,
London, 1989, p. 610. See also Deist et al., Germany and the Second World War, vol. 1.
9 M. Gilbert and R. Gott, The Appeasers, London, 1963.
Appeasement 131
10 David Dutton, Simon: A Political Biography of Sir John Simon, London, 1992; Robert
Rhodes James, Anthony Eden, London, 1987; David Dutton, Anthony Eden: A Life and
Reputation, New York, 1997.
11 Compare Taylor’s remarks on Henderson with A. L. Goldman, “Two views of
Germany: Nevile Henderson vs. Vansittart and the foreign office, 1937–39,” British
Journal of International Studies, vol. 6, 1980, pp. 247–77. A good recent treatment of
Henderson is D. C. Watt, “Chamberlain’s ambassadors,” in Michael Dockrill and
Brian McKercher (eds), Diplomacy and World Power: Studies in British Foreign Policy, 1890–
1950, Cambridge, 1996, pp. 145–54.
12 Compare M. Cowling, The Impact of Hitler, Cambridge, 1975, pp. 271ff. R. A. C.
Parker also has some astute comments on Halifax in Chamberlain and Appeasement:
British Policy and the Coming of the Second World War, London, 1993, pp. 122–3; as does
D. C. Watt, How War Came, pp. 79–80. Also see Andrew Roberts, The “Holy Fox”:
A Biography of Lord Halifax, London, 1991.
13 Cowling, Impact of Hitler; Colvin, The Chamberlain Cabinet, London, 1971; K.
Middlemas, Diplomacy of Illusion, London, 1972; C. Barnett, The Collapse of British Power,
London and New York, 1972. The best recent study of Chamberlain is Parker,
Chamberlain and Appeasement.
14 Some of these ups-and-downs are covered in L. W. Fuchser, Neville Chamberlain and
Appeasement, New York, 1982. See also D. C. Watt’s perceptive essay, “Misfortune,
misconception, mistrust: episodes in British policy and the approach of war, 1938–
1939,” in M. Bentley and J. Stevenson (eds), High and Low Politics in Modern Britain,
Oxford, 1983, pp. 214–54.
15 Aster, Guilty Men; John Charmley, Chamberlain and the Lost Peace, London, 1989;
Parker, Chamberlain and Appeasement. D. Dilks, Neville Chamberlain, vol. 1, Cambridge,
1984, only goes to the year 1929.
16 Cowling, Impact of Hitler.
17 Cowling, Impact of Hitler. Charmley, in Chamberlain and the Lost Peace, argues this point
with particular gusto, suggesting even that resistance to German expansion to the
point of war was not in Britain’s long-term strategic interests. Charmley, however,
downplays the potential costs to Britain – both material and moral – of a German
conquest of Europe.
18 N. Thompson, The Anti-Appeasers, Oxford, 1971, confirms Cowling’s descriptions
of these divisions, as does D. Carlton in his critical biography Anthony Eden, London,
1981. For a study of an earlier period which makes use of Conservative constituency
records, see Stuart Ball, Baldwin and the Conservative Party: The Crisis of 1929–1931,
New Haven, CT, 1988. The best study of Labour’s foreign policy remains J. F.
Naylor, Labour’s International Policy, London, 1969, pp. 252ff. But also see Ben Pimlott,
Labour and the Left in the 1930s, Cambridge, 1977, and Hugh Dalton, London, 1985.
19 The literature is now too extensive to be listed in its entirety but readers can consult
M. Ceadel, Pacifism in Britain 1914–45, Oxford, 1980; D. S. Birn, The League of Nations’
Union, 1918–1945, London, 1981; D. Lukowitz, “British pacifists and appeasement,”
Journal of Contemporary History, vol. 9, 1974, pp.115–28. For the right (apart from the
works on fascism in Britain), see R. Griffiths, Fellow-Travellers of the Right, London,
1980.
20 F. R. Gannon, The British Press and Germany, 1936–39, Oxford, 1971; J. Lewis, The
Left Book Club, London, 1970; D. Waley, British Public Opinion and the Abyssinian War,
1935– 36, London, 1975; K. W. Watkins, Britain Divided: The Effect of the Spanish
Civil War on British Public Opinion, London, 1963.
21 On which, see Weinberg, Foreign Policy of Hitler’s Germany, vol. 1, pp. 516ff.; T. Taylor,
Munich: The Price of Peace, London, 1979, pp. 937ff. See also the excellent analysis of
the erosion of pro-appeasement feelings during late 1938 and early 1939 in L.
Kettenacker, “Die Diplomatie der Ohnmacht,” in W. Benz and H. Graml (eds),
132 The Origins of the Second World War Reconsidered
Sommer 1939. Die Grossmächte und der Europäische Krieg, Stuttgart, 1979, esp. pp. 239,
247ff. Watt, in How War Came, pp. 99–108, treats the shift in official opinion.
22 This is not to say that the Telegraph or the Manchester Guardian did not share, to some
extent, the relief that war had been avoided in 1938, but their line was altogether
much firmer than the government’s. See Gannon, British Press and Germany;
Thompson, The Anti-Appeasers, pp. 165ff.; P. Kennedy, “Idealists and realists: British
views of Germany, 1864–1939,” Transactions of the Royal Historical Society, series 5, vol.
25, 1975, pp. 154ff.
23 A. Adamthwaite, “The British government and the media,” Journal of Contemporary
History, vol. 18, 1983, pp. 281–97. Also see Richard Cockett, Twilight of Truth:
Chamberlain, Appeasement, and the Manipulation of the Press, London, 1989.
24 Kettenacker, “Die Diplomatie der Ohnmacht.”
25 P. Kennedy, “The tradition of appeasement in British foreign policy, 1865–1939,”
British Journal of International Studies, vol. 2, 1976, pp. 195–215; I. W. D. Gruner,
“The British political, social and economic system and the decision for peace and
war: reflections on Anglo-German relations, 1800–1939,” British Journal of International
Studies, vol. 6, 1980, pp. 189–218; M. Gilbert, The Roots of Appeasement, London,
1966.
26 A. J. P. Taylor, The Struggle for Mastery in Europe, 1848–1918, Oxford, 1954. Also
see P. Kennedy, The Rise and Fall of the Great Powers: Economic Change and Military Conflict
from 1500 to 2000, London, 1988, pp. 275–343.
27 D. Reynolds, The Creation of the Anglo-American Alliance, 1937–41, London, 1981; C.
A. MacDonald, The United States, Britain and Appeasement, London, 1981.
28 Compare G. Niedhart, Grossbritannien und die Sowjetunion, 1934–1939, Munich, 1972,
as well as the two very good articles by Niedhart and Hewdon, in Mommsen and
Kettenacker (eds), Fascist Challenge. Also see Parker, Chamberlain and Appeasement, pp.
216–45, which echoes Taylor on Chamberlain’s anti-Soviet prejudices. Also useful
is Anita Prazmowska, Britain, Poland and the Eastern Front, 1939, Cambridge, 1987.
The pro-Moscow versions are briskly (perhaps too briskly?) dealt with in D. N.
Lammers, Explaining Munich: The Search for Motive in British Policy, Stanford, CA, 1966.
29 For the military aspects, see especially M. Howard, The Continental Commitment,
London, 1972, chs 4–6. For the general political and cultural side see A. Wolfers,
Britain and France Between Two Wars, New York, 1966 (first published 1940); J. C.
Cairns, “A nation of shopkeepers in search of a suitable France, 1919–1940,” American
Historical Review, vol. 79, 1974, pp. 710–43, and the brief but pertinent comments
in E. M. Gates, End of the Affair, London, 1981, pp. 895ff. More recent treatments
are Martin S. Alexander, The Republic in Danger. General Maurice Gamelin and the Politics
of French Defence, 1933–1940, Cambridge, 1992, pp. 236–78; P. M. H. Bell, France
and Britain, 1900–1940, London, 1996, pp. 167–231.
30 See F. M. Hardie, The Abyssinian Crisis, London, 1974, whose definition of
appeasement (p. 4) is “not mere failure to resist an act of aggression but connivance
at it.” For Abyssinia, see also Gaines Post, Jr, Dilemmas of Appeasement: British Deterrence
and Defense, 1934–1937, Ithaca, NY, 1993, pp. 81–115. Locarno and its results are
covered in J. Jacobson, Locarno Diplomacy: Germany and the West, 1925–1929, Princeton,
NJ, 1972; A. Orde, Britain and International Security, 1920–1926, London, 1978. The
1932–3 Disarmament conference is covered in the excellent book by E. W. Bennett,
German Rearmament and the West 1932–1933, Princeton, NJ, 1979.
31 A. J. Marder, “The Royal Navy and the Ethiopian crisis of 1935–36,” American
Historical Review, vol. 75, 1970, pp. 1,327–56.
32 On which see the important study by C. Thorne, The Limits Of Foreign Policy: The
West, the League and the Far Eastern Crisis of 1931–1933, London, 1972.
33 E. H. Haraszti, Treaty-Breakers or Realpolitiker? The Anglo-German Naval Agreement of June
1935, Boppard, 1973; D. C. Watt, “The Anglo-German naval agreement of 1935,”
Journal of Modern History, vol. 28, 1956, pp. 155–75; J. Dülffer, “Des deutsch–englische
Appeasement 133
and the Imperial Japanese Navy, Oxford, 1981; P. Haggie, Britannia at Bay: The Defence
of the British Empire Against Japan, Oxford, 1981; P. Lowe, Great Britain and the Origins
of the Pacific War, 1937–1941, Oxford, 1977; A. Shai, Origins of the War in the East:
Britain, China, and Japan, 1937–41, London, 1976.
45 See, again, Kennedy, “Strategy versus diplomacy.”
46 P. W. Schroeder, “Munich and the British tradition,” Historical Journal, vol. 19, 1976,
p. 242. Also see Charmley, Chamberlain and the Lost Peace, p. 210–12.
47 Barnett, Collapse of British Power, pp. 505–20. For more assessments of the balance,
see W. Murray, The Change in the European Balance of Power, 1938–1939, Princeton,
NJ, 1984, esp. ch. 7.
8 Debating the role of
Russia in the origins of the
Second World War
Teddy J. Uldricks
The central characters in A. J. P Taylor’s The Origins of the Second World War are
Hitler and a succession of British and French statesmen. Soviet Russia is ascribed
a lesser role, as earlier critics have pointed out.1 For Taylor, the really important
action takes place in Berlin, London and Paris, not Moscow. This is a serious
weakness because, by “tilting” his analysis to the west, Taylor limits himself to
an incomplete and interpretively distorted account of Europe’s descent into war.
Ironically, this western orientation leads Taylor seriously to underestimate the
influence of anti-Communism and Russophobia within the British leadership.
The Soviet state is largely absent from the early chapters of the book. In
Taylor’s view Russia had fallen from the ranks of the Great Powers by 1918.
He even goes so far as to refer in several places to the “disappearance” of Russia.2
It is certainly true that the Bolshevik regime did not at first enjoy the same
material base of military strength relative to the other European powers that its
imperial predecessor had possessed, and that this shift in the balance of power
benefited Germany. Yet, contrary to Taylor, the USSR had not disappeared. In
fact, the western nations – mesmerized by the threat of revolution – may well
have accorded more attention to communist Russia than they ever had paid to
the Romanov empire. This oversight, fostered by Taylor’s disinclination to give
weight to ideological factors in international relations, causes him to misinterpret
western policy in a number of important ways. For example, he criticizes the
Allies for leaving Germany united and, thus, still potentially the strongest country
on the continent. But this judgment misses the critical point that the victors, for
all their hatred for the defeated enemy, felt that they needed Germany intact as
an anti-Communist rampart in central Europe.3
Similarly, Taylor’s aversion to the ideological factor causes him to write of
the Bolsheviks’ “sense of security” during the decade after the First World War
(pp. 40–1). He assumes that the Soviet leaders must not have felt threatened,
because he knows in hindsight that no other power attacked or even seriously
planned to strike the USSR after the end of foreign intervention in the Russian
Civil War. Bolshevik perceptions were, in fact, quite different. The belief that
the imperialist powers intended to renew their bloody assault on the homeland
136 The Origins of the Second World War Reconsidered
of socialism was an article of faith in the Kremlin. The Dawes plan, the Locarno
accords, German membership in the League of Nations and the Young plan were
each interpreted in Moscow as parts of a concerted imperialist strategy to undo
the October Revolution. These fears may have been unrealistic, but they were
real.4 Having dismissed these fears as either rhetorical or manipulative devices,
Taylor underrates the tremendous importance the Rapallo treaty had for the
Russians (pp. 52–3).5 Both emotionally and ideologically, it was the centerpiece
of the Soviet diplomatic system. It gave the Bolsheviks some minimal reassurance
that a united phalanx of capitalist powers was not yet poised to crush the Soviet
experiment.
Taylor’s treatment of Soviet foreign policy, and also of the USSR as a factor
in the diplomacy of other powers, is somewhat better for the 1930s than for the
previous decade: his self-proclaimed historical “intuition” seems to serve him
better in this period.6 There is for the 1930s a somewhat more extensive coverage
of Moscow’s significance in European affairs, though the amount of attention
accorded Soviet Russia is still not proportional to its actual role in the origins of
the Second World War. The Soviet Union still did not fully re-emerge, in his
view, as a European Great Power until the signing of the Nazi–Soviet Pact in
August 1939 (p. 216). Taylor is also willing to give Soviet anxieties more credence
in the Hitler era. He sees that the Kremlin perceived every diplomatic
combination which excluded them (Mussolini’s projected Four-Power Pact, the
Munich conference, etc.) as a conspiracy against the USSR. He recognizes that
the defensive pacts which the USSR signed with France and Czechoslovakia
did not allay those fears and suspicions. Because the French refused to give it
substantive military content, the Franco-Soviet agreement remained an “empty”
gesture (p. 85).7 That, in turn, vitiated the alliance with Prague, which was
dependent on relations between Paris and Moscow.
The subject of British views of and policy toward the USSR is, on the whole,
perceptively handled in The Origins of the Second World War. Taylor demonstrates
that the British government constantly rebuffed the attempts of the Kremlin to
draw it and France into a system of collective security against the menace of
Nazi aggression. Far from aiming at alliance, it was the exclusion of the Soviet
Union from European affairs that was a consistent British objective. Chamberlain
did not trust the Kremlin, nor did he respect its strength. Moreover, he assumed
that an alliance with Russia would provoke Hitler, thus making war more, rather
than less, likely. The only really significant gap in Taylor’s description of British
policy toward the U SSR is, once again, the omission of its ideological
underpinnings. He seriously underestimates the strength of anti-Communism
as a motive force in British foreign policy, not only toward Soviet Russia, but in
regard to Germany as well.8 Thus, Taylor notes, the great purges (especially the
destruction of the Soviet officer corps) reinforced the Prime Minister’s belief that
the USSR was scarcely worth having as an ally, but he misses the more important
point that Chamberlain was doctrinally opposed to any real alliance with the
Communist state – even had Stalin been a benevolent ruler instead of a bloody
tyrant.9
The Role of Russia in the Second World War 137
French foreign policy and diplomacy is treated in much less detail. Taylor
follows the traditional interpretation of France as a self-imposed captive of British
policy. The French, as he notes, grew more anxious for a binding military
convention with the USSR as the German threat continually mounted in the
late 1930s. A powerful Franco-Soviet alliance never emerged, however, due to
British opposition. Instead of recreating the Triple Entente, Paris had to choose
between two alternative dual alliances – London or Moscow.
The steady progression of German advances from the announcement of
rearmament to the absorption of Czechoslovakia finally produced a partial
reassessment of British appeasement. After the German seizure of Prague,
London responded more positively to the Soviet collective security campaign.
Yet even at this juncture, Taylor suggests, “the British government, in fact, were
not interested in solid military co-operation with Soviet Russia; they merely
wanted to chalk a Red bogey on the wall, in the hope that this would keep Hitler
quiet” (p. 246). Opening talks with Moscow was also a strategy for keeping their
French ally in line. The British reluctantly took the lead in negotiations with
the Soviets during 1939, Taylor suspects, because they feared that if they did
not do so, France might make binding military commitments with the USSR
which could subsequently drag Britain into war as well (p. 226).
The negotiations between the western powers and the Soviet Union in the
spring and summer of 1939 never had much chance of success. The two sides
came to the talks with divergent goals and expectations. The British still hoped
that war could be avoided. If it could not be prevented, they thought that western
forces, together with those of Poland, would bear the brunt of the fighting. Given
the purge-ravaged condition of the Red Army, they looked to Russia as only a
supply base, not as the main striking force of the prospective alliance. In addition,
western strategy anticipated a long war of attrition, much like the First World
War, in which Germany would be worn down by a prolonged blockade. The
Soviet role in such a conflict would be limited to supplying and supporting a
static eastern front in Poland and Romania. The Soviets saw things differently.
They had little faith that war could be avoided, and even less confidence in the
Polish army. They sought an alliance with Britain and France that included iron-
clad military commitments, in which the Soviet army would play an aggressive
and leading role from the onset of hostilities. They were convinced that only a
great Franco-Soviet offensive on two fronts could prevent German victory.
Differing forecasts of war or peace as well as divergent expectations of the nature
of a possible conflict, thus erected further barriers to co-operation between the
western democracies and the USSR (p. 247).10 Chamberlain doubted the value
of a defense pact with the Soviet Union, but he was sure that, if such a treaty
were signed, the presence of the Red Army in east-central Europe would be
neither necessary nor desirable. Therefore, he saw no need to force the Poles or
the Romanians to grant transit rights to large Soviet military formations. The
Polish regime’s refusal to permit the passage of Soviet troops was allowed to
stand as the apparent barrier to an Anglo-Russian alliance which the British
government did not want in the first place (pp. 205–6, 247).
138 The Origins of the Second World War Reconsidered
however, he admits the possibility that Stalin may have had territorial designs
on his neighbors but he still criticizes western statesmen for rationalizing the
rejection of an alliance with the USSR on such hypothetical grounds. He is
willing to accept the Soviet contention that a major Russian offensive through
Poland and Romania was necessary to defeat Germany (pp. 246–8). He also
believes that the Kremlin was reluctant to abandon its collective security
campaign as late as August 1939, and he even suggests that Molotov may have
deliberately stalled negotiations with Germany in order to give the democracies
one more chance (p. 250). When the Soviets finally chose to deal with Hitler,
Taylor contends, “the Russians, in fact, did only what western statesmen had
hoped to do” (p. 252). Finally, he argues that the Nazi–Soviet Pact should not
be considered an aggressive alliance – a conclusion which is difficult to support
in view of Soviet territorial gains and the degree to which the Soviet economy
functioned as the primary supplier of several otherwise unobtainable resources
for the German war machine from September 1939 to June 1941.
The Nazi–Soviet Pact and, more broadly, the nature and aims of Soviet foreign
policy throughout the 1930s have been subjects of intense debate. The
controversy encompasses extreme polarities and a wide spectrum of opinion
between them. At one extreme is the official interpretation sponsored by the
former Soviet government, according to which the USSR pursued a clear,
unambiguous, and even noble policy of building a European-wide shield of
collective security against Nazi aggression. This diplomatic course was based
not on cynical self-interest or the desire for aggrandizement but on high principle,
since the Soviet Union was said to represent the forces of historical progress
and was, therefore, bound to take the lead in opposing the barbarous and
retrogressive schemes of Nazi Germany. If collective security ultimately failed,
it was not for lack of unstinting and sincere Soviet effort, but rather because of
the treacherous failure of the western democracies to oppose Hitler’s murderous
plans. In the words of the authoritative History of Soviet Foreign Policy, co-edited by
former foreign minister Andrei Gromyko:
When the nazis seized power in Germany, the threat of another world war
became very real in Europe. However, at the time it was still possible to
avert fascist aggression through the concerted efforts of countries desiring
peace. Had the Soviet proposals for collective security been put into effect
it would have been possible to erect a powerful barrier to any
aggressor....But this project was wrecked by the joint efforts of the fascist
states and Poland with British encouragement....In this atmosphere the
Soviet Union never for a moment relinquished its efforts to create a system
of collective security.12
could not be considered in any sense an alliance with Nazi Germany, while they
ignored, and sometimes even denied the existence of, the secret protocol that
divided all of eastern Europe between Hitler and Stalin.13
At the opposite extreme is the allegation that collective security against
aggression was never the Kremlin’s real objective, but only a front behind which
Stalin sought throughout the decade to woo a reluctant Hitler into an aggressive
alliance.14 Robert Tucker has advanced an especially radical version of this
interpretation. He contends that, as far back as 1925, Stalin determined to divide
the capitalist states against each other and maneuver them into a mutually
destructive inter-imperialist war, from which the USSR would emerge unscathed
and in a strong position to expand territorially all along its borders. To bring
about this war Stalin allegedly aided Hitler’s rise by deliberately steering the
policy of the Comintern and the German Communist Party on a suicidal course.
The Nazi–Soviet Pact was, in this view, always implicit in Stalin’s plans, while
the collective security line was never anything more than a mask for his designs
and a bait to attract Hitler.15 Arguing along similar lines, R. C. Raack contends
that Stalin wanted the outbreak of war in Europe as much as Hitler did. Stalin
is said to have anticipated a prolonged war of attrition which would weaken
both Germany and the western democracies, breed revolution throughout
Europe, and thereby create an opportunity for the Red Army to aid proletarian
insurrections in the west.16
There is very little evidence to support these theories. Lacking any direct
evidence, Tucker must rely on a painstaking (and often strained) exegesis of
certain portions of Stalin’s published writings. Yet the Soviet dictator’s statements
were sometimes ambiguous and often ran counter to the thesis Tucker is trying
to sustain.17 The ambitious and aggressive policy of collusion with Germany,
which Tucker ascribes to Stalin, is not a continuation of the Rapallo orientation,
as he claims. The Rapallo policy, though certainly designed to split the imperialist
camp by courting Weimar Germany, was a defensive strategy aimed at preventing
the outbreak of a war in central Europe that could easily draw in the USSR.
The course of action which Tucker describes is a reckless high-risk strategy. Hitler
was that sort of desperate gambler; Stalin was not. Tucker also has suggested
that the purges were necessary to clear away the opposition within the Bolshevik
elite to an opportunistic deal with Hitler. Thus the great purges provide evidence,
he argues, for his theory that collective security was always a ruse.18 The problem
with this argument is that Stalin had the wrong people killed. Among Soviet
diplomats, for example, numerous senior officials with a strongly pro-Rapallo
orientation, such as Nikolai Krestinskii and Lev Karakhan, were purged, while
many of their pro-western colleagues, like Litvinov, Ivan Maiskii and Aleksandra
Kollontai, were spared.19
A less extreme version of the theory that Stalin always preferred cooperation
with Germany (whether Weimar or Nazi) to a defensive alliance with the western
powers is advanced by Gerhard Weinberg. Although his major two-volume study
concentrates on German foreign policy, Moscow’s policy toward Berlin is an
important sub-theme which is explored in some detail. He is especially interested
The Role of Russia in the Second World War 141
in contacts between David Kandelaki, head of the Soviet trade mission in Berlin,
and Hjalmar Schacht.20 Although their tentative discussions did not lead to a
Russo-German rapprochement, Weinberg is convinced that Stalin tolerated the
collective security line only as a poor second choice while repeated Nazi rejections
of Soviet feelers kept his preferred alternative – an agreement with Hitler – out
of reach. Weinberg’s version of Stalin’s alleged preference for Germany over the
democracies is certainly more temperately argued and more thoroughly
researched than Tucker’s or Raack’s, but it, too, lacks sufficient evidence to be
convincing. It makes 98 percent of all Soviet diplomatic activity in the 1930s a
brittle cover for the remaining covert 2 percent. Lacking conclusive evidence to
the contrary, it is more reasonable to assume that Soviet representatives spent
the majority of their time and effort trying to accomplish their real objectives.
Similarly, enormous energy was expended to propagate the Popular Front line
in the Comintern and all foreign communist parties. These efforts were
counterproductive to achieving a rapprochement with Nazi Germany. Moreover,
the eventual abandonment of collective security and signing of the Nazi–Soviet
Pact did incalculable harm to the communist movement around the world.21 This
suggests that the policy shift in August 1939 was not the culmination of a long
and covert campaign, but a painful reorientation forced on Moscow by the failure
of its collective security strategy.
When Taylor wrote Origins there was very little Soviet documentary and
memoir material available on which to build an interpretation of Soviet conduct.
His bibliography lists “Soviet Russia: Nothing.” Since 1961 useful sources have
appeared. The Dokumenty vneshnei politiki SSSR series began publication in 1957
and continued with a new volume each year until 1977. The series stopped
suddenly, without any explanation, after the appearance of volume XXI,
containing documents for 1938. The editors and, more importantly, the
leadership of the Communist Party of the Soviet Union could not bring
themselves to release sensitive, and potentially embarrassing, materials about the
shift from collective security to the pact with Hitler in 1939. Not surprisingly,
these published documents tend to show the Soviet Union pursuing a consistent
and principled policy of anti-Fascism throughout the decade. Nonetheless, this
official collection contains a great deal of valuable material, although it was
tendentiously edited to support the party line.22 Unfortunately, the editors omitted
all documents relating to secret attempts by Kandelaki, Karl Radek, and Sergei
Bessonov to seek a rapprochement with Berlin. Other governments, most notably
the French and the Italian, have since 1961 issued substantial documentary
collections which contain important evidence of Soviet diplomatic activities.
An interesting memoir, Evgenii Gnedin’s Katastrofa i vtoroe rozhdenie, provides
some glimpses behind the scenes of Soviet diplomacy.23 Gnedin was first secretary
at the Soviet embassy in Berlin and then press spokesman at the foreign
Commissariat in Moscow during the latter half of the 1930s. In 1962 he was
asked to prepare a report on Russo-German relations in the 1930s, in connection
with a Central Committee investigation into the past “miscalculations and errors”
of Molotov. In other words, Gnedin was requested to help with the compilation
142 The Origins of the Second World War Reconsidered
the war about to burst over Europe by avoiding commitments to any of the
imperialist powers. Stalin, he believes, did not take a resolute position on either
side of the argument between Litvinov and the isolationists. Instead, the Soviet
dictator followed the alternative that appeared most likely to protect his state
from a disastrous war – which meant the collective security line for most of the
1930s, until he despaired of its success at the end of the decade. Given the
fragmentary evidence available, delineating Kremlin factions is a perilous
enterprise for any scholar, but Haslam’s conclusions are carefully reasoned and
suitably tentative.29
This approach contrasts strongly with many earlier accounts of Soviet foreign
policy in the 1930s which were influenced by the Cold War and by the then-
dominant totalitarian model of Soviet behavior. That model depicts the Soviet
state as a totalitarian monolith in which policy debates and political process, as
normally understood, simply did not exist. Instead, this view portrays Stalin as
the sole political actor in the Kremlin. He appears as a virtually omniscient,
though diabolical, political chess master who plots his nefarious course many
moves in advance. The model also suggests that totalitarian states require foreign
enemies, war, and territorial conquests in order to maintain their domestic
stability. Haslam’s abandonment of the totalitarian perspective and his emphasis
on genuine political differences in the Kremlin is supported by the work of Paul
Raymond, who characterizes the Politburo debate over foreign policy options
in the 1930s as “controlled pluralism.” He sees the Soviet elite divided over a
complex of crucial domestic and foreign policy issues, with Litvinov, Marshal
Tukhashevskii, and others advocating collective security, while Molotov,
Kaganovich, and others argued for a re-establishment of the old Berlin–Moscow
nexus.30 R. Craig Nation agrees that an “important body of opinion within the
party hierarchy, including leaders closely linked to Stalin such as Viacheslav
Molotov, Andrei Zhdanov, and Lazar Kaganovich, never accepted the legitimacy
of Litvinov’s program…”; but he cautions that “the most basic reality working
to undermine support for collective security in Moscow was the unwillingness
of the western powers and their eastern European allies to engage themselves
unambignously.”31 That conclusion is reinforced strongly by Michael Jabara
Carley, who demonstrates the consistent and tenacious opposition of London
and Paris to any collective security arrangement with Moscow.32
Haslam’s book, along with the work of Geoffrey Roberts (discussed below),
are the most realistic treatments now available of the evolution of Soviet foreign
policy from collective security into the Nazi–Soviet Pact. Not only does Haslam
attempt to demonstrate the diversity of opinion within the Kremlin leadership,
he also gives full weight to the plethora of Soviet fears – fear of German
aggression, fear of an anti-Soviet alliance between Germany and the western
powers, fear that the democracies would “buy off” the fascist regimes by
encouraging them to move eastward, fear of a Japanese attack, etc.33 He avoids
the trap of seeing Soviet policy as either sublimely “principled” or diabolically
unprincipled. Instead, along with Taylor, Haslam assumes that the Kremlin elite
was searching for whatever course would best protect the interests of the Soviet
The Role of Russia in the Second World War 145
Union. Thus, Stalin’s “preferred” policy is neither a defensive alliance with Britain
and France nor a pact with Germany, but whatever policy would provide the
greatest degree of security. Given the almost unrelenting hostility of the Nazi
regime, the only alternative (before August 1939) was the collective security line.34
Haslam’s interpretation of two crucial periods, the Czech crisis of 1938 and
the dual set of negotiations in the summer of 1939, are particularly well argued
and based on thorough research. He demonstrates that Stalin followed Litvinov’s
collective security line throughout 1938, as opposed to the “isolationist”
alternative, and that the Soviets would have aided the Czechs against a German
attack (at least if France did so, too). The Kremlin was still convinced that a
show of resolution by Britain, France, and Russia together would have forced
Hitler to back down, but the western powers were determined to appease
Germany, so they undermined Soviet efforts for joint action.35 Haslam places
the blame for the breakdown of the collective security drive and for the signing
of the Nazi– Soviet Pact squarely on Britain. He believes that Stalin did not make
the final decision for a deal with Hitler until August. Up to that point, Soviet
diplomacy still sought an alliance with the democracies, but their mutual
suspicions and vastly different reading of the international situation prevented
such an agreement.36 “Confronted with the evident unwillingness of the Entente
to provide immediate, concrete and water-tight guarantees for Soviet security in
Europe…the Russians were left with little alternative,” Haslam contends, “but
an agreement with Germany creating a condominium in Eastern Europe.”37
While debate over the nature and objectives of Stalin’s foreign policy raged
in the west, the writing of international history in the USSR remained caught
in the icy grip of party orthodoxy well into the 1980s. Then, suddenly, that
grip was broken in the glasnost era. Echoing poet Evgenii Evtushenko, Mikhail
Gorbachev called for filling in the “blank spots” of Soviet history. However,
Gorbachev initially had no wish to disseminate information which might
embarrass the Soviet Union or undermine the legitimacy of Communist Party
rule. The pace of change was slow at first. Well into the mid-1980s traditionalist
senior historians still published essentially orthodox accounts of Soviet diplomacy
before the Second World War, and they continued to deny the existence of the
secret protocols to the Nazi–Soviet Pact. Inevitably, however, history soon became
a weapon in the wider struggle between conservatives and reformers. Among
the latter, Iurii Afanas’ev admitted the existence of the secret protocols, while
Mikhail Semeriaga argued that it had been a mistake to accept a pact with Hitler
rather than intensifying the pursuit of an alliance with Britain and France.38
Seeking independence for Latvia, Lithuania, and Estonia, Baltic nationalists
published the previously secret protocols, and Baltic historians denounced the
Nazi–Soviet Pact and all that flowed from it as an illegitimate imperialistic
conspiracy between Hitler and Stalin.39
The campaign to replace the pieties of party orthodoxy with a candid look
at the Soviet past had become unstoppable. Colonel-general Dmitri Volkogonov,
with access to state, party and military archives, published a massive, bombshell
biography of Stalin. In the chapters on the origins of the Second World War,
146 The Origins of the Second World War Reconsidered
gave the Soviet leadership no choice but to seek an anti-German alliance with
the western powers. Collective security was pursued with determination and
patience by Moscow from December 1933 to August 1939. Of course, the
Kremlin would have been happy to negotiate a rapprochement with Germany
at any time during the decade, but only on terms which would have required a
complete reversal of Hitler’s aggressive foreign policy. Roberts argues powerfully
that the Soviet Union sincerely pursued co-operation with Britain and France
up to the time of the Munich conference in order to deter Nazi aggression.
Munich convinced Stalin that war was now unavoidable, so thereafter the Soviet
Union sought a full-scale war-fighting alliance with the western democracies.
Only the collapse of these alliance negotiations in mid-August 1939, combined
with tempting offers from Hitler, brought about the reversal of that policy and
the signing of the Nazi–Soviet Pact.
The availability of recently published Soviet documents enables Roberts to
analyze Russo-German relations in a much more sophisticated way. A revealing
example concerns Aleksei Merekalov’s memorandum of his meeting with Ernst
Weizsäcker on April 17, 1939. This apparently pivotal meeting, known previously
only through the German account, had led many western scholars to identify
this as the point at which the Soviet government took the initiative in pursuing
an agreement with Germany. However, Merekalov’s report does not show any
Soviet initiative for an improvement of political relations between Moscow and
Berlin, and it indicates that it was the Germans who took the role of the pursuer
in this courtship.46 More broadly, after a number of contacts, Soviet diplomats
reported to Moscow that their German counterparts took the lead in suggesting
improved relations. The Soviets gained the impression from these gambits that
there were important elements within the German diplomatic, industrial, and
military elites who did not share the sharp antagonism of the Nazis toward the
USSR. The Nazi regime, however, did not alter its anti-Soviet policies, so these
hints of a possible rapprochement came to nothing.47
Roberts also departs from the common assumption that discipline, consistency,
and overarching purposiveness typified Kremlin policy making. He characterizes
Soviet foreign policy in the 1930s as often reactive, sometimes haphazard and
accidental, and occasionally adrift. Whereas the formerly dominant totalitarian
model portrayed the Soviet dictator as a bloodthirsty chess master, plotting his
diabolical course many moves in advance, Roberts sees Stalin responding with
increasing desperation to the uncontrollable rush of events. Thus, in his view,
the Nazi–Soviet Pact did not inevitably lead to the invasion of eastern Poland,
the Winter War with Finland, or the absorption of the Baltic republics. Instead,
an initial policy of genuine neutrality gave way to one of expansion and quasi-
alliance with Germany only under the pressure of events.48
The publication of new documents and the partial opening of Soviet archives
have forced scholars to reexamine the question of whether the USSR would
have come to the aid of Czechoslovakia against German aggression at the time
of the Munich crisis, at least had France done so as well. Hochman and Igor
Lukes have argued strongly that Stalin had no intention of honoring his treaty
148 The Origins of the Second World War Reconsidered
obligation to assist the Czechs.49 Lukes has even gone so far as to suggest that
Stalin was the only “winner” in the outcome of the Czech crisis since, he argues,
it made war between Germany and the western powers almost inevitable.50
Opposing the views of Hochman and Lukes, Hugh Ragsdale and Geoff Jukes
demonstrate that there is evidence suggesting that Stalin would have fought for
Czechoslovakia. Jukes uses the memoirs of Marshal M. V. Zakharov to show
that, at the height of the Munich crisis, the Kiev Special Military District was
ordered to mobilize and deploy a large proportion of its forces along its western
boundary. The Belorussian Special Military District received similar instructions.
In total the Red army may have deployed as many as 330,000 men along the
central portion of its east-European frontier.51 Critics have disputed the veracity
of Zakharov’s testimony, but Ragsdale has confirmed the accuracy of the
marshal’s account (and of documents subsequently published by the Soviet
authorities) through his research in the records of Politburo meetings contained
in the archives of the Communist Party. Ragsdale has found the original telegram
transmitting the order for mobilization to the Kiev Special Military District.52 It
is, of course, impossible to prove definitively a hypothetical statement about what
Stalin would have done had France honored its pledges to Czechoslovakia and
had the whole crisis not been resolved so dishonorably at Munich. However,
the evidence uncovered by Ragsdale suggests strongly that the USSR would
have honored its treaty obligations had conditions for doing so materialized.53
The most recent major contribution to the study of Russia’s role in the origins
of the Second World War was written by the late Aleksandr Nekrich, who
emigrated from the USSR in 1976 after his path-breaking 1965 work June 22,
1941 proved to be too revealing in the repressive environment after Khrushchev’s
fall. Nekrich argues that the Kremlin followed a dual policy throughout the
1930s, both trying to restore a cooperative Rapallo-style relationship with
Germany and, failing that, pursuing a collective security arrangement with Britain
and France against the menace of Nazi aggression. This dual policy was
maintained well into 1939. He emphasizes contacts between Red Army
commanders and Reichswehr officers, as well as Kandelaki’s approaches to
Hjalmar Schacht as Soviet attempts to court non-Nazi German elites (the officer
corps and big business). At least in the mid-1930s Stalin mistakenly thought these
forces might be able to redirect German foreign policy away from its anti-Soviet
course.54 This line of analysis supports the work of Ingeborg Fleischhauer,
discussed above.
In contrast to Haslam, Raymond, and Kulish, however, Nekrich assumes that
Stalin dominated all foreign policy decision-making. In his account there are no
debates or factions. Nekrich also argues that the dismissal of Litvinov as foreign
commissar marked a new stage in Stalin’s foreign policy. From this point on, he
argues, territorial expansion became a significant goal.55 This conclusion contrasts
sharply with the work of Roberts, who sees the territorial gains of 1939–40 as
defensive reactions to a rapidly deteriorating international situation.56
The last decade also has seen renewed interest in the life and work of the
main apostle of collective security, Maksim Litvinov. Zinovy Sheinis, a Russian
The Role of Russia in the Second World War 149
Notes
1 See T. M. Mason, “Some origins of the Second World War,” Past & Present, no. 29,
1964, pp. 67–87; and D. C. Watt, “Some aspects of A. J. P. Taylor’s work as
diplomatic historian,” Journal of Modern History, vol. 49, 1977, pp. 19–33.
2 A. J. P. Taylor, The Origins of the Second World War, Greenwich, CT, 1965, p. 25 and
throughout. The reference here and elsewhere in this chapter is to the American
150 The Origins of the Second World War Reconsidered
second edition which includes a new Preface for American readers as well as a
postscript in which the author defends himself against some of his early critics.
3 On the importance of anti-communism and counter-revolution in the foreign policy
of the western states, see A. J. Mayer, Politics and Diplomacy of Peacemaking: Containment
and Counterrevolution at Versailles, New York, 1967; J. M. Thompson, Russia, Bolshevism,
and the Versailles Peace, Princeton, NJ, 1966; and two works by R. K. Debo, Revolution
and Survival: The Foreign Policy of Soviet Russia, 1917–18, Toronto, 1979, and Survival
and Consolidation: The Foreign Policy of Soviet Russia, 1918–1921, Montreal, 1992.
4 For evidence of the Soviet regime’s fundamental insecurity in the international arena,
see T. J. Uldricks, “Russia and Europe: diplomacy, revolution, and economic
development in the 1920s,” International History Review, vol. 1, 1979, pp. 55–83; J. P
Sontag, “The Soviet war scare of 1926–27,” Russian Review, vol. 34, 1975, pp. 66–
77; J. Haslam, Soviet Foreign Policy, 1930–33: The Impact of the Depression, New York,
1983; and J. Jacobson, When the Soviet Union Entered World Politics, Berkeley, CA, 1994,
esp. pp. 273–280.
5 Compare with C. Fink, The Genoa Conference: European Diplomacy, 1921–1922, Chapel
Hill, NC, 1984, and, “The NEP in foreign policy: the Genoa conference and the
Treaty of Rapallo,” in G. Gorodetsky (ed.), Soviet Foreign Policy, 1917–1991: A
Retrospective, London, 1994, pp. 11–20.
6 A. J. P. Taylor, “Accident prone, or what happened next,” Journal of Modern History,
vol. 49, 1977, p. 7. Borrowing a gardening term from Sir Lewis Namier, Taylor refers
to his historical “intuition” as “green fingers”: “Some may say that I have relied on
my green fingers too much. I think that I have relied on them too little.”
7 Compare with J. E. Dreifort, “The French Popular Front and the Franco-Soviet Pact,
1936–37: a dilemma in foreign policy,” Journal of Contemporary History, vol. 11, 1976,
pp. 217–36.
8 For a sophisticated discussion of the bases of British appeasement policy which takes
the factor of anti-communism into account (though, perhaps, still not sufficiently),
see M. Gilbert, Roots of Appeasement. Also useful are M. Thomas, Britain, France and
Appeasement, Oxford, 1996, and N. Jordan, The Popular Front and Central Europe,
Cambridge, 1992. Compare with G. Niedhart, “British attitudes and policies towards
the Soviet Union and international Communism, 1933–9,” in W. J. Mommsen and
L. Kettenacker (eds), The Fascist Challenge and the Policy of Appeasement, London, 1983,
pp. 286–96.
9 For western perceptions of Soviet military strength (or weaknesses), see R. R. Rader,
“Anglo-French estimates of the Red army, 1936–1937,” Soviet Armed Forces Annual,
vol. 3, 1979, pp. 265 80; and J. H. Herndon, “British perceptions of Soviet military
capability, 1935–9,” in Mommsen and Kettenacker, Fascist Challenge, pp. 297–319.
10 Since Taylor developed this argument, the matter of western expectations of the
nature of future wars has been studied in much greater detail. See A. Preston (ed.),
General Staffs and Diplomacy before the Second World War, London, 1978; R. J. Young, In
Command of France: French Foreign Policy and Military Planning, 1933–1940, Cambridge,
1978; and E. R. May (ed.), Knowing One’s Enemies: Intelligence and Assessment Before the
Two World Wars, Princeton, NJ, 1984. These more detailed studies tend to support
Taylor’s suggestion that differing strategic conceptions added to the gulf between
London and Moscow.
11 For an alternative interpretation stressing the logical continuity of theory and practice
in Nazi foreign policy see G. L. Weinberg, The Foreign Policy of Hitler’s Germany, vol.
1, Diplomatic Revolution in Europe, 1933–36, Chicago, IL, 1970; and vol. 2, Starting
World War II, 1937–1939, Chicago, IL, 1980.
12 B. Ponomaryov, A. Gromyko, and V Khvostov (eds), History of Soviet Foreign Policy,
1917–1945, Moscow, 1969, pp. 337–8. For the orthodox Soviet view also see
I. K. Koblyakov, USSR: For Peace, Against Aggression, 1933–1941, Moscow, 1976;
I. F. Maksimychev, Diplomatiia mira protiv diplomatii voiny, Moscow, 1981; and A.
The Role of Russia in the Second World War 151
Louise Young
On September 18, 1931, Japan’s Kwantung Army initiated the military conquest
of northeast China now known as the Manchurian Incident. With the occupation
of Manchuria, Japanese imperialism entered a new and critical period. During
this phase of empire building, Japan moved aggressively to expand its overseas
territories, occupying first China and then south-east Asia, and initiating a
succession of military conflicts against the nationalist and communist forces in
China, the Soviet Union, the United States, and the British empire. During the
rapid military expansion of the 1930s and 1940s, what Japanese officials called
“autonomous diplomacy” signified two departures from past practice. First, it
meant liberating imperial interests in Asia from a consideration of relations with
the west. In the past, fearing diplomatic isolation, Japanese policy makers took
careful stock of how a potential move in Asia was likely to be received in the
west, and interventions were preceded by cautious multilateral negotiations. After
1931, however, the “Manchurian problem,” the “China question,” and the
“advance south” were each decided unilaterally and in the face of western
opposition. Second, autonomy signaled a new independence for Japan’s colonial
armies. In the wake of the Manchurian Incident, military fait accomplis followed
one upon another, as aggressive field officers took their lead from the success of
the Manchurian occupation. Since the 1890s, imperial expansion had begun with
military conquest. But by the 1930s, the imperial garrisons had multiplied and
the institutional complexity of the armed services had opened new possibilities
for subimperialists. The trigger-happy proclivity of the garrison armies turned
the boundaries of the empire into a rolling frontier.
In the conventional chronologies of modern Japanese history, the Manchurian
Incident ushered in a new era in Japanese politics and foreign policy. It marked
the beginning of a period of militarism, fascism, and war, characterized by
repression and mobilization at home and aggressive expansion in Asia and the
Pacific. The chain of events set off by the Manchurian Incident is thus viewed
as the onset of a regional crisis and the first step on the road to the Second World
War in Asia. Although most historians would agree with a chronology that
156 The Origins of the Second World War Reconsidered
censorship. Mitchell’s analysis takes the reader through the dense web of
censorship laws, providing a draconian portrait of the Japanese state and its
control over freedom of expression. Kasza’s description of the relationship
between the state and the mass media allows for more autonomy on the part of
journalists and editors, suggesting that the state censorship apparatus, though it
grew in the 1930s, was less than comprehensive.
One of the shortcomings of the literature on thought control is the tendency
to minimize the genuinely popular support that existed for many of the
government initiatives of the 1930s. The problem emerges in part from the
concept of “thought control,” which renders the Japanese public as passive, robot-
like, consumers of an ideology of war produced by government propaganda and
indoctrination. In my own work I have tried to call attention to the ways that
consumers and producers of mass culture, in and out of government, collectively
produced a new vision of empire that ennobled the military aggression of these
years. This was particularly true during the Manchurian Incident’s war fever
of the early 1930s, when the Japanese public became eager consumers of war
mongering and China bashing in the mass media. After the story broke of the
military clash between Chinese and Japanese troops on September 18, 1931,
the news of the latest action on the China continent commanded the headlines
for months. War songs set the fashion in popular music and battlefield dramas
filled the stage and screen. Such media sensationalism flooded popular
consciousness with images of war and empire. Marketing militarism, the mass
media became, in effect, unofficial propagandists, helping to mobilize popular
support for the army’s policy of military aggression against China.6
Leading the crusade were Japan’s major dailies, the Asahi and the Mainichi
newspapers. Both papers responded to the outbreak of hostilities by deploying
recently purchased fleets of airplanes and cars, and mobilizing the latest printing
and photo-telegraphic machinery in a drive to use the Manchurian Incident to
expand their hold on the national news’ market. The newspaper companies used
airplanes to shuttle teams of correspondents and equipment back and forth
between Japan and Manchuria, keeping the home front abreast of the daily
progress of the occupation through morning and evening editions saturated with
news from the front, plus additional photograph-filled Manchurian Incident
“extras” hawked by bell-ringing newsboys throughout the day.
The press supplemented more conventional news presentation with reporting
in the form of a public spectacle: newspaper-sponsored newsreel screenings,
traveling lecture series, and exhibits of military paraphernalia. All of these were
enormously popular, public enthusiasm driving the newspaper companies to new
heights in their sensationalizing of the Manchurian Incident. Asahi and Mainichi
newsreels that tracked the occupation of Manchuria, stage by victorious stage,
filled public halls and packed city parks. Although both newspaper and film
companies had made sporadic attempts at producing regular film news during
the 1920s, newspaper company footage of the Manchurian Incident brought
newsreels into widespread use for the first time. As fast as new film canisters
could be flown in from Manchuria, the Asahi and Mainichi screened the newsreels
158 The Origins of the Second World War Reconsidered
in city parks in Osaka, Kobe, Kyoto, and Tokyo, and circulated the films for
additional showings in department stores, elementary schools, and elsewhere
throughout the country In Osaka, for example, the first installment, “The military
clash between the Japanese and Chinese armies,” opened on September 21 –
just three days after the clash began – and required several showings a night to
accommodate the crowds. An account of the onset of the campaign for northern
Manchuria, “The Nen River battle-front,” proved to be the city’s favorite for
November, playing for 20,000 spectators on a single night. Five thousand stood
outdoors on a chill January evening to watch marching columns of Japanese
soldiers “entering Jinzhou.” Since the free newsreels were a marketing tool, they
were shown widely outside the urban areas, particularly in rural districts where
the large dailies hoped to expand circulation.
An equally enthusiastic reception for traveling lecture series and exhibitions
of military paraphernalia rewarded the big dailies with popular acclaim. On
November 25, the Osaka Asahi touched off a lecture boom with a three-day
lecture series on “Reports from the battlefield,” with special correspondents
lecturing to full houses in Osaka, Kobe, Kyoto, and Nagoya on their impressions
of conditions at the front. A December 3 session drew a crowd of 6,000 in Osaka,
and a report in Tokyo on the invasion of Jinzhou in January inspired standing
ovations and three banzais for the Asahi from the enthusiastic crowd.
The large department stores offered space to both the Mainichi and Asahi
newspaper companies for exhibitions, held in November and December, of
military paraphernalia commemorating the Manchurian Incident. After opening
in Tokyo, an exhibition of “Souvenirs of the fierce campaign to take the Fengtian
Beitaying,” sponsored by the Tokyo Asahi, went on to tour seventy locations to
the north and the west. In Tokyo, exhibition-goers numbered 11,000 daily, while
nation-wide a total of 600,000 saw the exhibition. The fusillade of Manchurian
Incident “extras,” the newsreel screenings, the lectures and exhibitions glorified
military action, lionized the Kwantung army, and extolled the founding of
Manchukuo. Such messages dovetailed beautifully with what the army wanted
its public to hear about the Manchurian Incident. But the culture industries
needed no arm-twisting to advertise the army’s cause: they became unofficial
propagandists because militarism was all the rage. Audiences flocked to watch
the dramas of death in battle, and readers bought up the magazines
commemorating the glories of the empire. From this perspective it was not the
system of “thought control” that created popular support for the Manchurian
Incident, but rather the operation of a commercial market for mass culture.
Another dimension of wartime ideology that has commanded the attention
of historians is the problem of tenkô, the ideological conversions that devastated
the ranks of communists, socialists, and liberals. Responding to new social and
government pressures, scores of prominent intellectuals abandoned their public
support for oppositional politics, effectively gutting the left-wing social movement
in the 1930s. The literature on tenkô, while stressing the growth of state repression,
also focuses on the agency of private individuals and organizations in the
ideological shift of the 1930s, examining different motivations for erstwhile critics
Japan At War 159
of the Japanese state to become its new supporters.7 Intimately tied to the problem
of tenkô are the issue of war responsibility and the question: Why were Japan’s
progressive intellectuals not a stronger voice against the militarism and
ultranationalism of the 1930s? Tatsuo Arima, Victor Koschmann, and
Matsumoto Sannosuke argue that, over the course of the modern period,
Japanese intellectuals failed to carve out a position of individual autonomy from
which they could develop a voice of genuine protest. This “failure of freedom,”
in Arima’s words, fatally weakened the ability of intellectuals to stand against
the state as it mobilized for war.8 Others portray intellectuals in more forceful
terms, quite capable of independent intervention to direct the course of culture
and politics. The research of Miles Fletcher, for example, illustrates the active
support given by many left-wing intellectuals to state goals in the 1930s. Drawn
to fascist ideas of using the state as an agent of renovation, intellectuals helped
shape the agenda for the wartime reconstruction of the economy and the New
Order in Asia.9
Evidence tends to support the latter interpretation, particularly when one looks
at the role played by progressive intellectuals in the expansion in China. Over
the course of the 1930s, scholars and China experts flocked to Manchuria to
take up research and planning posts in the rapidly expanding colonial state. The
context for this mobilization of intellectual talent by Manchukuo was a
transformation of the academic climate in Japan. Responding to a new assault
on academic freedom in the academy by conservative bureaucrats and right-wing
scholars, a series of university incidents brought scientific analysis of Japan’s
history, polity, and society increasingly under attack and purged from the
academy liberal intellectuals like Takigawa Yukitoki and Minobe Tatsukichi. The
increasing limitations on freedom of expression profoundly affected Japanese
Sinologists, forcing them to maintain a show of public support for official
government aims on the continent. At the same time, the expansion of the
imperial project in Manchuria brought new pressures on Sinologists to cooperate,
just as it opened up new opportunities for them. Depending on political
convictions, scholars responded to these changes in diverse ways. In the last
analysis, however, they had two options: mobilization or repression.
Numerous right-wing intellectuals, like Kyoto University professor Yano
Jin’ichi, who had long advocated a larger Japanese presence in Manchuria,
enthusiastically greeted the new situation. Immediately after the establishment
of Manchukuo, Yano packed his bags for China, where he became a propagandist
for the puppet state. For many left and liberal critics of Japanese expansion,
however, to remain true to their beliefs meant the loss of a job or the threat of
more serious punishment. Yanaihara Tadao, the liberal head of colonial studies
at Tokyo University, was forced along with countless others to sit out the war
years under intellectual house-arrest.
Although Yano Jin’ichi and Yanaihara Tadao took divergent paths in response
to the suppression of the 1930s, both remained faithful to political beliefs that
they had held in the 1920s. A whole segment of Japanese intellectuals, however,
abandoned their former faiths, suddenly lending their support to a policy of
160 The Origins of the Second World War Reconsidered
flatly refused to participate in a party cabinet. Civilian statesmen were told that
young officers were “fundamentally in agreement with the principles held by
[the assassins]. Should the cabinet again be handed over to a political party,
second and third incidents would recur.”14
In these ways army authorities made use of right-wing terrorism to exclude
political parties from control of the cabinet and to assume much broader powers
for themselves. The story of military radicalism, ably told by Wilson, Shillony,
and others, looks at the growth of militarism at the level of high politics. An
alternative approach to the problem of militarism in the 1930s is to focus on popular
or grassroots militarism, as Richard Smethurst, Thomas Havens, and others have
done. Both Smethurst and Havens frame their discussions in terms of wartime
mobilization. Smethurst argues that the support of rural Japanese for the war effort
and the turn to the right in the 1930s was mobilized through the army reservist
associations and their affiliates: youth groups and the Women’s Self-Defense
Association. In this sense he sees the rise of militarism in the countryside as the
product of an elaborate institution-building effort on the part of the army ministry
and of the organizational genius of army bureaucrats like Tanaka Giichi.15
In addition to the historians cited earlier, Ben-Ami Shillony and Thomas
Havens have also written on the subject of war and popular culture. Both frame
their narratives in terms of mass control and mass mobilization, dealing with
the variety of government campaigns that took place in the late 1930s and early
1940s: the campaign for spiritual mobilization through the mass media and state
propaganda; the campaign against “extravagance” in household finance; social
mobilization through the formation of an increasingly dense network of voluntary
organizations that culminated in the creation of block associations. The story
that Shillony and Havens tell, in other words, is about the ways in which the
state sought to organize people into units for war.16 Such an approach tends to
depict “the people” in passive terms; the state becomes the agent of history and
the people are merely its pawns. However, other writings on wartime mobilization
challenge this picture of an all-powerful state. Gordon Berger’s work on political
parties in the 1930s suggests that the old-style political parties were never
effectively absorbed into the totalitarian formation of the Imperial Rule Assistance
Association. Richard Rice also stresses the imperfection of wartime controls,
arguing that business interests managed to resist many provisions of the economic
mobilization law.17
Whether they depict the wartime state in stronger or weaker terms, these
writings highlight one of the important effects of the process of mobilization:
the growth of a socially-interventionist state in the 1930s and a new, more
intimate, relationship between state and society. A point that is sometimes lost
in these discussions is that the forces reshaping the relationship between
community and state did not emerge solely from the state itself. While
government initiatives exerted new kinds of pressure on local communities, at
the grassroots level the bureaucracy’s operation of mobilization relied on the
mediation of local elites. Ultimately, the extent to which the state was permitted
into the local community was determined by these elites – the gatekeepers of
Japan At War 163
local autonomy. In their desire to invite state intervention in what had been
private affairs, elites were driven by different motivations – anxiety about an
economy that appeared deadlocked by global depression, inability to handle the
social devastation the Depression had wreaked on their communities, fears that
social conflict over shrinking resources was spinning out of control. Whatever
their motivations, actions by local elites to mobilize the state were as much
responsible for the extension of state control over their lives as were the initiatives
of bureaucrats in the central government.
In these ways, the literature on the war at home has taken up the question of
“what happened” in the 1930s from diverse angles. Historians have given the
transformations of these years a long list of labels, describing wartime Japan
variously as ultranationalist, Japanist, militarist, corporatist, totalitarian, and
fascist. Each label comes with its own interpretive framework, defining the
character of state and society in the 1930s, and postulating the causes of Japan’s
turn to rapid military expansionism. Of all these labels, that of fascism has been
the most frequently suggested and the most ardently contested. The “fascism
debate,” as it is known, began almost immediately after the end of the war and
has been joined by historians on both sides of the Pacific.18
The opening round in the fascism debate was delivered by Japanese Marxist
historians, who argued that the intensified military aggression abroad and political
repression at home of the 1930s were the products of accumulated contradictions
in the structures of Japan’s political economy. For these historians, the roots of
the 1930s’ crisis went back to the Meiji reforms of the 1870s and 1880s, when
the architects of Japanese modernization made “emperor-system absolutism” into
the framework for the Japanese nation–state. As Herbert Bix has summarized
it, the “emperor-system” represented a formula for rule that encompassed both
an “ideology of absolutism, prompted by a regime in which the sovereign had
virtually unlimited powers and was regarded as an object of popular veneration”
and “a nationalism generated by periodic national crises.” The emperor-system
entered its fascist stage in the 1930s, when the absolutist political structures
acquired fascist functions. When they proposed the concept of “emperor-system
fascism,” Japanese Marxists were underscoring the differences between fascism
in Japan and fascism in Europe. While fascism in Japan and Europe emerged
out of the same crisis in the capitalist system when it reached the stage of
monopoly capitalism, the nature of imperial rule and the peculiar composition
of the ruling class set Japan apart from its European counterparts. Thus their
analysis stressed the pre-eminent position of the bureaucracy as the instrument
of imperial absolutism, supported by a ruling class comprised of the military,
landlords, and monopoly capitalists. In the 1930s, then, the bureaucracy became
the prime agent of emperor-system fascism, “fascizing” the structures of the state
in order to salvage Japanese monopoly capitalism.19
This interpretation was challenged by Maruyama Masao, in his enormously
influential series of essays on Japanese politics, published in English under the
title Thought and Behavior in Modern Japanese Politics.20 Maruyama agreed with the
Marxists in labeling Japan fascist, but took issue with their exclusive focus on
164 The Origins of the Second World War Reconsidered
had previously been moving along so successfully. In the end they glossed over
the problem by minimizing the importance of the 1930s as an aberration.21
The writings of the modernization school stimulated a new debate over the
utility of the category of “fascism” to Japan in the 1970s. Most Japanese historians,
while critiquing the earlier formulation of emperor-system fascism for its
monolithic conceptualization of the Japanese ruling class, and “fascism from
above” for neglecting the role of mass movements and support for fascism on
the part of ordinary Japanese people, still upheld the propriety of “fascism” as a
description of the 1930s. They attempted to salvage the concept by looking at
fascism as the outgrowth of modernizing trends in the development of Japan’s
political and social system. For example, the newer literature on fascism focused
on the pre-eminent role of the military in politics and postulated that for Japan
fascism represented the product of a new political system designed to cope with
total war.22 Others, notably American historians such as Peter Duus, Daniel
Okamoto, Ben-Ami Shillony, George Wilson, and Gordon Berger, followed the
lead of the modernization scholars and questioned the use of “fascism” in the
Japanese case. Approaching the problem with more analytical focus, these
American scholars identified a number of key differences between Japanese and
European fascism: Japan had no charismatic leader like Hitler or Mussolini, and
unlike Italy and Germany the Japanese government was not taken over by a
mass movement but remained in the hands of the old guard, the political elite.
As Duus and Okamoto put it, though there may have been fascists in Japan
there was nothing that could reasonably be defined as a fascist political system.23
In the place of the rejected paradigm of “fascism” its critics have offered a
host of alternative rubrics: corporatism, totalitarianism, and militarism, among
others. What is at stake in these labels? At the heart of the fascism debate is the
question: Who are the villains of the debacle of the 1930s? What led Japan to
war? Whether blame is placed on a government-manipulated system of thought
control, or on an institutional structure peculiarly susceptible to military takeover,
or on a form of late-capitalism which is driven by contradictions to repress at
home and aggress abroad, historians in Japan and the United States have looked
to domestic causes for the domestic crisis. Indeed, one of the striking aspects of
the literature on wartime Japan is its tendency to segregate the domestic from
the foreign context. In the works discussed thus far, problems in Japan’s empire
and in its relationship with the western powers are not considered as potential
causes of the domestic turmoil of the 1930s, but only as its effects. The literature
on the foreign crisis, to which I now turn, is characterized by a similar tendency
to treat Japanese actions abroad in isolation from the profound transformations
taking place in the domestic arena.
If the literature on the home front is prodigious, the literature on the foreign
crisis is equally vast. Increasing the confusion is the fact that the Second World
War in Asia was fought against many enemies and on many fronts: the war in
China against the nationalists and the communists, the war along the Siberian
border of Manchuria against the Soviet Union, the war against the European
colonial powers in south-east Asia, and finally, the war in the Pacific against the
166 The Origins of the Second World War Reconsidered
United States. Even though these separate conflicts were ultimately connected in
a single great war, they remained in many ways distinct from one another – distinct
in their origins and chronology, in the nature of the conflict and the war aims of
each side, and in the impact of their denouement on the postwar order. These
distinctions are reflected in the literature and add to the complexity of the picture
of the Second World War in Asia. Still, the question remains: How did these
separate conflicts become connected? What were the steps that led the Japanese
from a small war in Manchuria to the hell-fires of Hiroshima and Nagasaki?
The war in China began in Japan’s sphere of influence in the northeast. The
literature on the Manchurian Incident of 1931 focuses on two dimensions of
the conflict. One set of studies looks at the diplomatic fall-out in the League of
Nations and the escalating tensions between Japan and the western powers,
including the American attempt to contain Japan through “moral diplomacy,”
the French and British attempts at appeasement, and the appointment of the
Lytton Commission, whose critical report culminated in Japan’s withdrawal from
the League. Nish, Thorne, and others have debated the significance of Japan’s
withdrawal in terms of what it meant for international peacekeeping and whether
it signified the evisceration of the League as an instrument for maintaining
international order. A point they do not touch on is its significance in terms of
Japanese diplomacy, for Japan’s withdrawal signaled a profound departure from
past practice. Since the forced opening of Japan in the nineteenth century, the
overriding goal of foreign policy had been to join the great powers on terms of
international equality. Choosing to stand against the western powers in the
League in 1932–33, Japanese diplomats abandoned this goal in favor of
“autonomous diplomacy.” Of their own volition, Japanese statesmen withdrew
from the great power club to which they had labored so long to gain entry. In
the process, they ushered in an era of international isolation and entered on a
collision course with rival imperialists.
The second set of studies on the Manchurian Incident focuses on the gathering
conflict between Japan and its problematic ally in Manchuria, the warlord Zhang
Xueliang. Documenting the series of events that led the army to act – the increasing
intractability of Zhang Xueliang, the specter of the Chinese nationalist movement
spreading to Manchuria, dissatisfaction with the civilian authorities’ handling of
the Manchurian question – these studies highlight the role of the Kwantung army,
Japan’s garrison force in Manchuria, and narrate the Manchurian Incident as
military history.24 What made the Kwantung army such a key player in the
Manchurian Incident and in the military origins of the Second World War in Asia
was its peculiar character as a rogue army. In this sense the fact that the Manchurian
Incident was initiated by a Kwantung army conspiracy is of more than passing
significance. On the evening of September 18, 1931, several Kwantung army
officers secretly exploded a section of Japanese railway, blamed the explosion on
Chinese agitators, and used this as the pretext to attack the forces of Zhang
Xueliang. Such a modus operandi was nothing new; the Kwantung army had been
a hotbed of intrigue for years and had been responsible for the assassination in
1928 of Zhang Zuolin, Xueliang’s father and Japan’s previous collaborator in the
Japan At War 167
region. The conspirators of 1928 were not punished; nor were those of 1931.
Instead, upon finding themselves in possession of a vastly expanded territory in
northeast China, government authorities in Tokyo decided to endorse the
Kwantung army’s action as a fait accompli and ignore the insubordination of the
officer corps of the most powerful of Japan’s colonial garrisons.
While the Sino-Japanese conflict was temporarily resolved by the Tanggu truce
of 1933, which recognized Japan’s new position in Manchuria, tensions continued
to simmer until they erupted again in 1937. Although the China Incident (as
the undeclared China War was officially known) was a direct outgrowth of the
Manchurian Incident, the Anglo-American literature treats it as a separate conflict.
Books by John Boyle and Lincoln Li on the Japanese occupation look at the
difficulties of maintaining effective control over local collaborators, but draw no
comparisons with the Manchurian case.25 Other works on the conflict focus on
the military strategies employed by both sides, analyzing their respective successes
and failures. On the Japanese side the most pressing question is: How did Japan
become bogged down in what proved an unwinnable war? The answers are
found in a series of tactical errors, minute missteps, and misjudgments. By
beginning their narratives after 1933, these works suggest that the “success” of
the Manchurian campaigns bore no relation to the failures of the China
Incident.26 And yet, the facility with which Japan occupied Manchuria nurtured
the hubris of Japanese ambitions to crush the Nationalists in a quick and decisive
strike, and fed mounting ambitions to absorb all of China’s vast territories.
Moreover, both incidents arose from the contradiction between Japanese
imperialism and Chinese nationalism, and what Japanese military planners were
blind to in both 1931 and 1937 was that the Nationalist movement could not
be vanquished by military means. The occupation of Manchuria was an
enormous stimulant to Chinese nationalism and steeled Chinese determination
to resist Japan at all costs. As much as it reflected Japanese overconfidence, this
determination to resist was the product also of the Manchurian Incident, and
both attitudes helped precipitate the outbreak of fighting on the Marco Polo
Bridge in 1937 and its escalation into all-out war.
Much like the analysis of the Manchurian and China Incidents, the histories
of the other fronts in the Second World War in Asia are contained in separate
literatures, each highlighting its distinctive origins and character.27 The conflict
with the Soviet Union that flared into the series of Manchukuo–Siberian border
wars in 1937–39 was justified by a combination of Russophobia and anti-
communism. Though fears of the “red peril” emerged powerfully in the wake
of the Russian Revolution and the failure of Japan’s military efforts to contain
communism in the Siberian Expedition of 1918–22, Russophobia had deeper
roots, going back to the competition over Korea and Manchuria that culminated
in the Russo-Japanese War of 1904–5. When the Japanese occupied the Russian
sphere of influence in north Manchuria in the early 1930s, they destroyed the
geopolitical balance of power that had been laid down in 1905 and had lasted
for twenty-five years. Both sides began preparations for war.
168 The Origins of the Second World War Reconsidered
On another front, the Japanese sweep through south-east Asia was launched
in the name of Asian liberation and to drive the “white peril” out of Asia. In
the face of the Japanese assault, the European colonial edifice fell like a house
of cards, and the Japanese took credit for exposing the hollow core of European
racial arrogance. Although the ideology of white peril had long-term rhetorical
roots, the decision to strike south was the product of an increasingly acute need
for war resources (particularly the oil fields of the Dutch East Indies) and the
desire to cut off China’s supply routes through south-east Asia. These came to
a head in the late 1930s, just when the outbreak of war in Europe diverted
attention from Asia and weakened the defenses of the colonial powers.
In the meantime, along yet another front a naval war had broken out with
the United States. Like the conflicts with China, the Soviet Union, and the
European colonial empires, Japan’s tensions with the United States had deep
roots. The two had begun to jockey for strategic domination of the Pacific before
the turn of the century, when a rivalry over Hawaii precipitated a war scare
and ended in the American decision to annex. The two powers fought over who
would inherit the German-held Pacific islands after the First World War; they
prepared war plans and built up their navies against one another. Tensions rose
in the 1930s with Japanese aggression in China and a new naval arms’ race,
but America did not take concrete action to stop Japan until it felt its position in
the Philippines threatened by Japan’s southward advance.
Clearly, the military conflicts between Japan and its many enemies in the
Second World War were in fundamental ways quite distinct from one another.
Yet the question remains: What, if anything, tied them together into a single
great war? The analytic isolation of these different conflicts in the Anglo-
American literature tends to minimize the connections. Conservative historians
in Japan echo this position, arguing in particular that a line be drawn between
the Manchurian and China incidents. The implication is that had the army
stopped in Manchuria, Japan might well have avoided being drawn into the
Second World War. Progressive Japanese historians have objected strongly to
this argument, and many use the rubric “fifteen-year war” to stress the origins
of the Second World War in the Manchurian Incident. As the historian who
coined the term explained: “The term ‘Fifteen-year War’ is used in order to imply
that the various military incidents which occurred between 1931 and 1945 were
in fact a continuous undeclared war.”28 The point is that the Manchurian Incident
set in motion a process of military expansion that became difficult to contain.
There were many reasons for this, not the least of which concerned the “shoot
first, ask questions later” approach to empire-building inaugurated by the army
in Manchuria. Again, one cannot underscore too heavily the ways the new
military imperialism linked the Manchurian Incident to the later wars of the
decade. Unlike subsequent conflicts, the Manchurian Incident was a success, in
the sense that Japan’s war aims were achieved. Yet “autonomous diplomacy”
worked, in this instance, because Japan’s diplomatic partners were too
preoccupied to mount an effective resistance.29 China and the Soviet Union, for
example, responded initially to Japanese bellicosity with concessions. In the early
Japan At War 169
1930s, the nationalists were too busy fighting the communists to resist the
takeover of Manchuria. Stalin, preoccupied with agricultural collectivization, the
five-year plans and purging the party, decided to sell off the Chinese Eastern
Railway in 1935 and retreat before Japan’s advance into north Manchuria. But
after the formation of a united Chinese communist–nationalist front in 1936 and
the Soviet fortifications of the Manchurian–Soviet border, both China and the
Soviet Union began to stand their ground. War broke out with China in 1937,
and with the Soviet Union in 1938 and 1939.
Similarly, American and European interests in Asia were initially consumed
with domestic economic problems and the dissolution of the international financial
system. The day before the Manchurian Incident, Great Britain went off the gold
standard, so there was little attention to spare for the far east. Although after 1937
the United States opposed Japan indirectly by supplying Jiang Jieshi with war
materials, only in 1940, after the outbreak of war in Europe and the Japanese
advance into Indochina, did the United States begin embargoes on strategic
materials to Japan. The tightening of the economic screws led the Japanese to the
decision, once again, to attack, and from December 1941 Japan was fighting a
war against Britain and the United States, and the boundaries of the empire became
an endless war front. In the process, the empire and the war grew indistinguishable.
From a small war in Manchuria to the titanic naval battles of the Pacific, the
hallmark of the new imperialism of the 1930s was a state of perpetual war.
Following the model pioneered in Manchukuo, the autonomous phase of
empire also denoted a new kind of colonial rule. This was signaled initially in
official rhetoric which sought to depict the Japanese colonial state as the ally of
anti-colonial nationalism. First, Manchuria was “liberated” from China by a
movement for independence; later, the Japanese set up an administration in south-
east Asia under the slogan “Asia for the Asiatics.” As vacuous and self-serving
as these declarations seem in retrospect, at the time they were effective in initially
mobilizing support both among Japanese at home and among the Asians who
helped Japan create the new colonial institutions.
The organizational structure of the puppet state which was developed in
Manchuria subsequently became the prototype for the creation of a string of
collaborationist regimes in occupied China. In south-east Asia, the picture was
more complex. With the support of local nationalist movements, Japan drove
out western colonial rulers, establishing two types of administration. In Thailand
(the sole independent country at the time of Japanese occupation) and, after
January 1943, in Burma and the Philippines, alliances gave Japan the power of
indirect rule. In Indonesia and Malaya, the occupying forces governed through
a military administration. With the exception of French Indochina, where Japan
ruled in collaboration with the French authorities and was opposed by Ho Chi
Minh’s newly organized Vietminh, south-east Asian nationalists cooperated with
Japanese colonial rule, especially in its initial phase.
Strategies of mobilization were part of the Manchurian formula. Military,
political, economic, and cultural institutions were created or reshaped to organize
new communities of support for Japanese rule. Ambitious young Chinese found
170 The Origins of the Second World War Reconsidered
the Manchukuo army and military academy a route of advancement, as did their
counterparts throughout the empire. Military institutions formed in the late
colonial period in Burma, Korea, and elsewhere became the training grounds
for postcolonial elites. Similarly, the Japanese established mass parties such as
the Putera in Indonesia and the Kalibapi in the Philippines, patterned on
Manchukuo’s Kyôwakai. Throughout the empire, the Japanese created joint
ventures with local capital. Sometimes this was a mask for Japanese control,
sometimes a cover for appropriation of native capital, and sometimes, as in Korea,
a means of cultivating a collaborative elite and splitting the nationalist movement.
Assimilationist cultural policies were widely applied over the course of the 1930s
and 1940s, in an attempt to create an elite cadre of youth loyal to Japanese rule.
These policies went furthest in Taiwan and Korea, where the kôminka,
(imperialization) movement sought to erase native cultural traditions, replacing
them with the Japanese religious practices of shrine Shinto, the use of the Japanese
language, and the Japanization of given names.
It was not only colonial state institutions, but also the experiment with
economic autarky in Manchukuo that became the guiding spirit of the wartime
Japanese empire. The integrated industrial and trading unit formed with the
Japan–Manchuria bloc economy was extended first to include north China, then
the rest of China, and finally south-east Asia in a self-sufficient yen bloc. In Korea,
Taiwan, and north China this involved industrialization and heavy investment,
as it did in Manchukuo. The lessons of economic management learned in
Manchukuo, including currency reunification, production targets, semi-public
development companies, and other tools of state control, were applied also in
these new economies.
In all these ways the experiment in Manchukuo marked the beginning of a
new imperialism, made necessary by the upsurge of revolutionary nationalist
movements throughout the empires of Asia. European powers responded to the
rise of Asian nationalism with a policy of appeasement, attempting to shore up
the crumbling colonial edifice through political concessions in the middle east
and India. The Japanese dealt with the same challenge by claiming a unity with
Asian nationalism. They tried to co-opt the anti-colonial movement by declaring
the Japanese colonial state to be the agent of nationalist liberation.
While this question of continuity between the various military incidents of
the 1930s has constituted one of the interpretive divides in the literature on the
foreign crisis, another division of opinion emerged over the issue of Japan’s war
responsibility. Stripped to its bare bones, the basic point of contention in the
latter debate is whether the foreign crisis was the result of belligerent, extremist,
and ultimately irrational Japanese policies, or whether those same policies
represented a reasonable and realistic response to external pressures and
provocations. In other words: Did Japan create the crisis in Asia, or was it merely
drawn into a crisis of another’s making?30
The former position is articulated in the argument that in the 1930s a cabal of
reckless ultra-nationalist extremists in the army seized the reigns of government
from civilian moderates and, reversing the course of foreign policy, led Japan down
Japan At War 171
the path to war. This interpretation owes its origins to the “Japan crowd” in the
US State Department, and especially to Joseph Grew, ambassador to Japan for
much of the 1930s. Grew subscribed to a “pendulum theory” of Japanese history,
arguing that the nation alternated between periods of intense nationalism and anti-
foreign sentiment, followed by periods of internationalism and cooperation. Grew’s
view was that the Japanese government was divided into two hostile factions, the
extremist nationalists and the moderate internationalists, and that American policy
needed to co-operate with and encourage the latter at all costs.31
In postwar history books, this interpretation re-emerged in a “civilian– military
split” thesis: the idea that liberal civilian moderates in government were thrust
aside in the 1930s by military extremists. This was the position taken by the
prosecution in the Tokyo War Crimes’ Trial, officially known as the International
Military Tribunal for the Far East. The IMTFE indicted Japanese militarism,
embodied in the persons of twenty-eight defendants, for the crime of conspiracy
to wage aggressive war. The hearings of the Tokyo trial involved a long excursion
into Japanese history, and traced to its origins a putative Japanese plan to conquer
the world, put into action through the series of military incidents that began in
Manchuria.32 Robert Butow’s 1961 biography of wartime leader Tojo Hideki
drew heavily on trial records and enshrined the judgments of the tribunal as
academic orthodoxy. Echoing the Tokyo trial’s verdict that a “militaristic clique”
took hold of the government, Butow dated the military seizure of power and
the first step on the road to Pearl Harbor from the Kwantung army’s adventurism
in the “neighbor’s garden” of Manchuria.33
This viewpoint is given a slightly different inflection in the seven-volume series
The Road to the Pacific War (Taiheiyô sensô e no michi: kaisen gaikôshi), published by
the Japan Association of International Relations (Nihon kokusai seiji gakkai) in 1962–
63 and translated more recently into English. This work aimed to criticize the
structural arguments of Japanese Marxists. Eschewing any notion of a ruling
class unified behind emperor-system fascism, the volumes in this series focused
narrowly on the military as an organization. Unlike the Marxist historians, these
researchers were given privileged access to the archives of the self-defense forces.
Later they published many of the materials they discovered in a documentary
supplement to the series.
The thrust of this scholarship is twofold. First, it takes issue with the “fifteen-
year war” approach, choosing rather to treat the conflicts of the Second World
War as isolated military incidents. Second, it looks for the causes of the various
military conflicts in the minute day-to-day decisions of Japanese officers on-the-
spot. Consequently, the Japanese military, and in particular its officer corps, bears
the main burden of war responsibility: given its single-minded pursuit of military
security to the exclusion of all other considerations, escalation of military conflict
was inevitable, as was the eventual clash with the United States. Diplomats are
secondary figures in this narrative: they appear after the fact, attempting
ineffectually to repair the damage done by precipitate and irresponsible military
officers. In this analysis the explanation for war becomes, again, a military that
has run amok. Yet while Butow’s argument stresses the military seizure of power
172 The Origins of the Second World War Reconsidered
at the center, The Road to the Pacific War asserts the importance of military
autonomy at the periphery of the Japanese empire.34
The military-conspiracy thesis is echoed in many ways in the work of
progressive Japanese historians like Ienaga Saburô. Ienaga’s book on the Pacific
War, written in the mid-1960s, was part of a broader attempt to educate the
Japanese population and to “stimulate reflection and self-criticism about the war.”
The book is concerned with the issue of responsibility, and examines the choices
made during the war by ordinary citizens like Ienaga himself. In some ways his
book follows from the line of interpretation of Japanese Marxist historians who
located responsibility for war more broadly in the power structure of prewar
Japan, that is to say the entire ruling class of landlords, monopoly capitalists,
and state bureaucrats, of which the military was one part. Yet, though Ienaga
acknowledges the importance of support given by other elements of the ruling
class, the principal blame for the war falls on the army. The peculiar autonomy
granted the military under the prewar constitution permitted this “authoritarian
and irrational” institution the leeway to mobilize people for war, inculcating in
them racist and hateful attitudes toward other Asians, promoting mass conformity
and blind obedience to authority, and driving them into a frenzy of self-
destruction for an unwinnable war.35
In contrast to such depictions of Japanese foreign policy in the 1930s driven
by an irrational and out-of-control military, other historians have analyzed these
same policies as the rational and realistic response to a crisis in the international
system. In this literature, the key decision-makers appear not as ultra-nationalist
extremists or trigger-happy colonial officers determined to overturn the world
order but as sober and experienced bureaucrats laboring as responsible
government officials to secure Japan’s legitimate economic and political interests
in the face of a increasingly unstable world situation.
This was the thrust of writing by American New-Left historians and critics
of American imperialism such as Noam Chomsky. Placing the blame for the
Pacific War squarely on the shoulders of the United States, Chomsky argued
that Japanese foreign policy in the 1930s was a defensive response to American
imperialism in Asia. Echoing the case made by Japanese statesmen in the 1930s,
Chomsky wrote that American claims to have protected China from Japanese
aggression were hypocritical and self-interested. Using economic strangulation
to block Japanese access to China, America was guilty of a gross double-standard
– denying Japan a guaranteed market in Asia like that possessed by America in
Latin America and by Britain in its colonial territories throughout the world.
American Asian policy was motivated not by paternalism toward China but
rather by the interest of American capitalists in the China market. And since
China was necessary for Japan’s survival, US pressure to end its occupation
propelled Japan to fight in self-defense. What Chomsky was suggesting, in other
words, was that while American activities in Asia were aggressive and
imperialistic, Japanese activities were driven by a legitimate desire for economic
survival.36 Guilty of falling prey to his own double-standard, Chomsky’s
vindication of Japanese aggression in Asia ironically echoed the views of
Japan At War 173
Notes
1 In the Meiji Restoration of 1868 the feudal state was overthrown and a series of
modernizing reforms inaugurated, culminating in the creation of a constitutional
monarchy.
2 In addition to the items discussed here, see also works listed in E. R May and J. C.
Thomson, Jr, American–East Asian Relations: A Survey, Cambridge, MA, 1972. See the
essays in W. I. Cohen (ed.), New Frontiers in American–East Asian Relations: Essays
Presented to Dorothy Borg, New York, 1983; A. Iriye, “The Asian factor,” in G. Martel
(ed.), The Origins of the Second World War Reconsidered: The A. J. P. Taylor Debate After
Twenty-five Years, Boston, MA, 1986, pp. 227–43; J. W. Dower with T. S. George,
Japan At War 175
Japanese History and Culture From Ancient to Modern Times: Seven Basic Bibliographies, 2nd
edn, Princeton, NJ, 1995; and S. Asada (ed.), Japan and the World, 1853–1952: A
Bibliographic Guide to Japanese Scholarship in Foreign Relations, New York, 1989. The
Dower bibliography will be especially useful for students seeking to do more
specialized reading on a particular topic. For students who read Japanese, the Asada
volume provides an excellent overview of the Japanese academic debates together
with an extensive bibliography of primary and secondary materials. Within the text
of this essay Japanese names are written in Japanese order: family name first,
personal name second. In citations of English-language translations of Japanese
authors, names are written as published.
3 The Meiji, (1868–1912), Taishô, (1912–1926), and Shôwa, (1926–1989) eras demarcate
imperial reigns.
4 J. W. Dower (Empire and Aftermath: Yoshida Shigeru and the Japanese Experience, 1878–
1954, Cambridge, MA, 1979, p. 85) uses this phrase to stress the continuity of
Japanese expansionist policies from the 1920s to the 1930s:
For further discussion of the sea-change of the early 1930s, see L. Young, Japan’s
Total Empire: Manchuria and the Culture of Wartime Imperialism, Berkeley, CA, 1998, pp.
55–182.
5 R. H. Mitchell, Thought Control in Prewar Japan, Ithaca, NY, 1966, and Censorship in
Imperial Japan, Princeton, NJ, 1984; Kazuko Tsurumi, “The army: the emperor
system in microcosm,” and “Socialization for death: moral education at school and
in the army,” both in her Social Change and the Individual: Japan before and after Defeat
in World War II, Princeton, NJ, 1966, G. J. Kasza, The State and the Mass Media in
Japan, 1918– 1945, Berkeley, CA, 1988.
6 The discussion which follows on the mass media and the Manchurian Incident is
drawn from Young, Japan’s Total Empire, pp. 62–7.
7 Shunsuke Tsurumi, An Intellectual History of Wartime Japan, 1931–1945, London,
1986; P. G. Steinhoff, “Tenkô and thought control,” in G. L. Bernstein and
Harahiro Fukui, (eds), Japan and the World: Essays on Japanese History and Politics in
Honour of Ishida Takeshi, London, 1988, pp. 78–94; P. G. Steinhoff, Tenkô: Ideology
and Social Integration in Prewar Japan, New York, 1991; Kazuko Tsurumi, “Six types
of change in personality: case studies of ideological conversion in the 1930s,” in
her Social Change and the Individual, Princeton, NJ, 1966; I. Neary, “Tenkô of an
organization: the Suiheisha in the late 1930s,” Proceedings of the British Association
for Japanese Studies, vol. 2, 1977, pp. 64–76.
8 Tatsuo Arima, The Failure of Freedom: A Portrait of Modern Japanese Intellectuals,
Cambridge, MA, 1969; J. V. Koschmann, “Introduction: Soft rule and expressive
protest,” and Matsumoto Sannosuke, “Special introductory essay: The roots of
political disillusionment: ‘public’ and ‘private’ in Japan,” both in J. V. Koschmann
(ed.), Authority and the Individual in Japan, Tokyo, 1978, pp. 1–53.
9 W. M. Fletcher, The Search for a New Order: Intellectuals and Fascism in Prewar Japan,
Chapel Hill, NC, 1982; and “Intellectuals and fascism in early Shôwa Japan,” Journal
of Asian Studies, vol. 39, 1979, pp. 39–63. Other discussions of intellectuals and the
New Order include: B.-A. Shillony, “Japanese intellectuals during the Pacific War,”
Proceedings of the British Association for Japanese Studies, vol. 2, 1977, pp. 90–9; James B.
Crowley, “A New Asian Order: some notes on prewar Japanese nationalism,” in B.
176 The Origins of the Second World War Reconsidered
S. and H. D. Harootunian (eds), Japan in Crisis, Princeton NJ, 1974, pp. 270–98; J.
B. Crowley, “Intellectuals as Visionaries of the New Asian Order,” in J. Morley (ed.),
Dilemmas of Growth in Prewar Japan, Princeton, NJ, 1971, pp. 319–73; Donald Keene,
“Tenkô literature: the writings of ex-communists,” in his Dawn to the West: Japanese
Literature in the Modern Era, New York, 1984, pp.846–905.
10 For a more elaborated version of this discussion, see Young, Japan’s Total Empire,
pp. 268–82.
11 See the chapters “The totalitarian solution’ and “Tenancy and aggression’ in R. P.
Dore, Land Reform in Japan, New York, 1985 (1959), pp. 86–114, 115–25; T. R. H.
Havens, Farm and Nation in Modern Japan; Agrarian Nationalism, 1870–1940, Princeton,
NJ, 1974.
12 B.-A. Shillony, Revolt in Japan: The Young Officers and the February 26, 1936 Incident,
Princeton, NJ, 1972; G.M. Wilson, Radical Nationalist in Japan: Kita Ikki, 1883–1937,
Cambridge, MA, 1969; R. Storry, The Double Patriots: A Study of Japanese Nationalism,
NY, 1957.
13 H. Byas, Government by Assassination, New York, 1942.
14 This discussion is drawn from Young, Japan’s Total Empire, pp. 127–128.
15 R. Smethurst, A Social Basis for Prewar Japanese Militarism: The Army and the Rural
Community, Berkeley, CA, 1974.
16 T. R. H. Havens, Valley of Darkness; The Japanese People and World War Two, Lanham,
MD, 1986; B.-A. Shillony, Politics and Culture in Wartime Japan, Oxford, 1981.
17 G. Berger, “Politics and mobilization in Japan, 1931–1945,” in P. Duus (ed.),
Cambridge History of Japan, vol. 6: The Twentieth Century, Cambridge, MA, 1988, pp.
97–153; G. M. Berger, Parties Out of Power in Japan, 1931–1941, Princeton, NJ, 1977;
R. Rice, “Economic mobilization in wartime Japan: business, bureaucracy, and
military in conflict,” Journal of Asian Studies, vol. 38, 1979, pp. 689–706.
18 For a sampling of the debate, see G. Kasza, “Fascism from below? A comparative
perspective on the Japanese Right, 1931–1936,” Journal of Contemporary History, vol.
19, 1984, pp. 607–30; I. Morris (ed.), Japan 1931–1945: Militarism, Fascism, Japanism?,
Boston, MA, 1963; Fletcher, “Intellectuals and fascism,” pp. 39–63; G. McCormack,
“Nineteen-thirties’ Japan: fascism?’, Bulletin of Concerned Asian Scholars, vol. 14, 1982,
pp. 20–33; P. Duus and D. I. Okimoto, “Fascism and the history of pre-war Japan:
the failure of a concept,” Journal of Asian Studies, vol. 39, 1979, 65–76, H. P. Bix,
“Rethinking ‘emperorsystem fascism’: ruptures and continuities in modern Japanese
history,” Bulletin of Concerned Asian Scholars, vol. 14, 1982, 2–19; G. M. Wilson, “A
new look at the problem of ‘Japanese fascism,’ ’ Comparative Studies in Society and History,
vol. 10, 1968, 401–12; R. Dore and Tsutomu Ôuchi, “Rural origins of Japanese
fascism,” in Morley (ed.), Dilemmas of Growth, pp. 181–209.
19 The “emperor-system fascism” thesis is restated in Bix, “Ruptures and continuities,”
pp. 3–5, the quoted statement from p. 4. For a discussion of the Marxist position,
see G. A. Hosten, Marxism and the Crisis of Development in Prewar Japan, Princeton,
NJ, 1986, pp. 256–63, and McCormack, “Nineteen-thirties’Japan,” pp. 30–1.
20 I. Morris (ed.), Thought and Behavior in Modern Japanese Politics, Oxford, 1963.
21 The volumes in this series include: M. Jansen (ed.), Changing Japanese Attitudes toward
Modernization, 1965; W. E. Lockwood (ed.), The State and Economic Enterprise in Japan,
1965; R. P. Dore (ed.), Aspects of Social Change in Modern Japan, 1971; R. E. Ward
(ed.), Political Development in Modern Japan, 1971; D. Shively (ed.), Tradition and
Modernization in Japanese Culture, 1971; and Morley (ed.), Dilemmas of Growth. For critical
evaluation of this series, see J. W. Dower, “E. H. Norman, Japan and the uses of
history,” in Dower (ed.), Origins of the Modern Japanese State. Selected Writings of E. H.
Norman, New York, 1975, pp. 31–65.
22 For a discussion of this literature, see Hatano Sumio, “Japan’s foreign policy, 1931–
1945: historiography,” in Sadao Asada, Japan and the World, 1853–1952: A Bibliographic
Guide to Japanese Scholarship in Foreign Relations, New York, 1989, pp. 218–33.
Japan At War 177
Brian R. Sullivan
“No concrete ‘interest’ was at stake in Abyssinia – not even for Italy: Mussolini
was concerned to show off Italy’s strength, not to acquire the practical gains (if
any such exist) of Empire.... But Italy still stood condemned as an aggressor; and
the two Western Powers could not bring themselves to recognize the King of Italy
as Emperor of Abyssinia. The Stresa front was gone beyond recall, Mussolini
forced on to the German side. This outcome was unwelcome to him. In attacking
Abyssinia, Mussolini had intended to exploit the international tension on th
eRhine, not to opt for Germany. Instead he had lost his freedom of choice.”
A.J.P Taylor, Origins, pp. 96, 106–7.
The influence on the origins of the Second World War of the crisis provoked
by the Ethiopian War appears clear from a glance at the evidence. Prior to the
Italian invasion of Ethiopia, Mussolini had made military agreements with the
French and formed a coalition with the British and the French to prevent German
aggression in Europe. Following the Ethiopian War, Fascist Italy and Nazi
Germany moved from providing joint military aid to the Nationalists in the
Spanish Civil War, to formal alliance in May 1939, to waging war together on
France and Britain in June 1940.
The argument explicating this sequence of events goes as follows. British
attempts to prevent the Italian conquest of Ethiopia created deep bitterness within
the Fascist regime and among the great majority of the Italian people. Hoping
to save the anti-German coalition, the French had tried to mediate the British–
Italian quarrel. After that attempt failed, the French reluctantly selected the British
as the superior partner. This choice struck Fascist leaders as betrayal. Considering
that the British were thus revealed as hostile and the French as treacherous,
Mussolini began a détente with the Germans in January 1936, to escape
diplomatic isolation.1
There remained many sources of friction between Italy and Germany,
particularly Italian protection of Austria and German penetration of the Balkans.
However, common ideology and compatible goals, as well as Hitler’s skillful
wooing of Mussolini, drew the two regimes closer over the next three years.
This relationship helped Hitler to remilitarize the Rhineland, annex Austria,
extort the Sudetenland, and seize Bohemia–Moravia and Memel. But these events
More Than Meets The Eye: The Ethiopian War 179
Britain might shut the Suez Canal, the conquest of Ethiopia proceeded. In early
May 1936 Italian forces entered Addis Ababa, while emperor Haile Selassie fled
into exile. From his balcony overlooking the Piazza Venezia, the Duce announced
the establishment of a Fascist Roman Impero in East Africa. Still, most of Ethiopia
remained unconquered.9
Yet another possibility of war flared in early June 1936. Mussolini worried
about even harsher League sanctions on Italy as retaliation for the erosion of
Ethiopian independence. In that case, he was prepared to invade Yugoslavia in
concert with the Hungarians, to seize the rich natural resources of Croatia and
Serbia. Over the following weeks, however, Mussolini grew ever more certain
that the League would end the economic blockade of Italy. Under strong British
and French pressure the League revoked sanctions against Italy in July 1936.
For a few weeks Europe returned to tranquillity.10
The eighteen months prior to the Italian attack on Ethiopia provide a distinct
contrast with the eighteen months following Mussolini’s proclamation in May
1936 of Vittorio Emanuele III as emperor of Ethiopia. In early June, Mussolini
advised Kurt von Schuschnigg, Dollfuss’ successor, to settle his differences with
Hitler. A few days later, Mussolini dismissed Suvich and Aloisi, and made his
pro-German son-in-law Galeazzo Ciano foreign minister. In late July, Italian and
German air crews began cooperating to transport General Francisco Franco’s
rebel Army of Africa from Morocco to Spain. Over the next few weeks, Mussolini
and Hitler agreed to coordinate their aid to the Nationalists in the Spanish Civil
War. That October, Ciano visited Germany, concluding his trip with friendly
conversations with Hitler. Speaking in Milan on November 1, 1936, Mussolini
described the new ties between Rome and Berlin as “an axis around which may
cooperate all European states…”.11 The following month the Italians and
Germans agreed to pursue coordinated diplomatic and economic policies in the
Danubian region.
Hermann Goering visited Rome in January 1937. He and Mussolini agreed
that Italy should predominate in Axis assistance to the Spanish Nationalists.
While Mussolini responded coldly to the suggestion that an Anschluss was
inevitable, he stressed his enthusiasm for co-operation with Hitler. Afterwards,
Ciano hinted to Goering that Italy might eventually acquiesce to German
annexation of Austria. In May, Mussolini accepted Hitler’s invitation to visit
Germany, and his September tour proved a triumph of Nazi propaganda.
Mussolini returned deeply impressed with German power. In November 1937
Italy adhered to the Anti-Comintern Pact between Germany and Japan. As Ciano
noted, the agreement was “anti-Communist in theory but in fact [was]
unmistakably anti-British.”12 The previous day, Hitler had said that he expected
Mussolini to attack Britain and France, possibly that coming summer, and that
Germany would supply Italy with raw materials but would seize the opportunity
to annex Austria and Czechoslovakia. The Führer expected the Duce to accept
this with good grace.13
Judged by these events, the A. J. P. Taylor thesis that French and British
inflexibility in 1935–36 drove an unwilling Mussolini into an alliance of necessity
182 The Origins of the Second World War Reconsidered
with Hitler appears unexceptionable.14 But his contention that British opposition
to and French inconsistency over the Italian conquest of Ethiopia proved
instrumental in creating the Axis reflects post hoc ergo propter hoc reasoning: logic
undermines Taylor’s argument, as does evidence available to him at the time he
was writing Origins.
Mussolini’s statements and actions from May 1936 to June 1940 make it easy
to distinguish the degrees of his hostility toward Germany, Britain, and France.
Toward Germany, Mussolini moved from proclaiming the Rome–Berlin Axis,
in November 1936, to paying a state visit to Hitler in September 1937, to
acceptance of the Anschluss in March 1938, to receiving Hitler in Italy in May
1938, to willingness to join Germany in war in September 1938, to a formal
Italian– German alliance in May 1939, to a summit conference with Hitler in
March 1940 where he pledged intervention on Germany’s side, and, finally, to
partnership in a war against France and Britain in June 1940.
With Britain, Mussolini proved willing to make diplomatic settlements: the
“Gendemen’s Agreement” (January 1937), the Nyon patrol arrangements
(October 1937) and the Easter Accords (April 1938). In September 1938
Mussolini and Chamberlain hurriedly worked out a proposal for the Munich
conference; they conducted private discussions during the meetings, and
Mussolini received the British head of government in Rome in January 1939.
Finally, throughout the second half of August 1939, Mussolini allowed Ciano
to attempt an Italian–British mediation to prevent the German invasion of Poland.
Only in the period January–March 1940 did the Italian–British diplomatic
relationship crumble, due to Mussolini’s refusal to sell Britain arms and the
British blockade of sea-borne deliveries of German coal to Italy.15
Toward France Mussolini directed rage and contempt from 1936 to 1940.
Neither the Duce nor Léon Blum hid their disdain for one another after the
latter became premier of France in June 1936. Two months later, they became
engaged in a bitter proxy war in Spain, where they were almost drawn into
fighting one another, particularly when the Italian-led Nationalist offensive
approached the Pyrenees. Another potential conflict arose from French fears that
the Italians would establish permanent bases on Majorca. Because of their refusal
to recognize the King of Italy as Emperor of Ethiopia, the French had no
ambassador in Rome between October 1936 and November 1938. In October
1937 Mussolini withdrew his ambassador from Paris. In March 1937, with the
Italian– Yugoslav accord, Mussolini effectively detached Yugoslavia from its
alliance with France. A year later, Mussolini rejected French attempts to join the
Easter Accords.
In May 1938, following Hitler’s visit to Italy, Mussolini insisted that his
country and France were enemies, fighting “on opposite sides of the barricade”
dividing Spain, and that the spirit of the Stresa conference was “dead and buried.”
After the French government finally appointed a duly-accredited ambassador to
Rome in November 1938, the Fascist regime replied with a stinging diplomatic
insult to Paris. Eleven days following the new French envoy’s arrival, a pre-
arranged demonstration took place in the Chamber of Fasces and Corporations.
More Than Meets The Eye: The Ethiopian War 183
the Fascist militia for the fiscal years 1935/36–1938/39, as well armed forces’
spending hidden in colonial budgets, amounted to some $5 billion. From January
1935 to December 1938 the Wehrmacht received the equivalent of $18.4 billion.34
Huge military expenditures made Nazi Germany much stronger, while they
made Fascist Italy comparatively weak. This arose from differences in their
expenditures: Germany’s were concentrated in the Reich; Italy’s were dispersed
abroad. As his military power grew at home, Hitler used it as a powerful adjunct
to diplomacy to annex Austria, the Sudetenland, Bohemia–Moravia, and Memel;
in 1933 he ruled 66 million Germans; by 1939 he directly controlled 87 million
people. The number of males under German domination aged 20–39 had
increased from 11.2 million to 14.3 million. Furthermore, Nazi Germany gained
riches from its bloodless conquests. Not only did it acquire the armaments and
resources of two central European countries but it captured their gold and foreign
exchange reserves. Confident in the economic potential of the expanded Reich,
Hitler’s new subjects retrieved monies they had stashed abroad. All in all, the
Nazi regime laid its hands on bullion and hard currency worth the equivalent
of 13.4 billion Italian lire in the seventeen months following the Anschluss. By
the spring of 1939, the industrial production of the Greater German Reich
equaled that of the United States. Its economy was annually producing goods
and services four times higher than the value of Germany’s gross national product
when Hitler had taken over the country six years before.35
Italian demographics and economics developed much less favorably. With the
conquest of Ethiopia and the seizure of Albania, Mussolini added 10–11 million
Africans and 1.1 million Europeans to his empire. Yet in the invasion and
pacification of Ethiopia Italian forces slaughtered hundreds of thousands of
Ethiopians, perhaps as many as 7 percent of the population. In the process, the
Italian colonial forces also suffered: at least 125,000 Eritreans and Somalis were
killed or disabled. As a result, Mussolini enjoyed only minor demographic
advantages from his conquests. In 1931, Italy had a population of 41.2 million,
with a further 2.3 million in its colonial empire. By 1940, the Italian population
amounted to 44.5 million and that of the colonies and Albania to 15 million. But,
of the latter, all but the populations of Albania, Libya and the Dodecanese Islands
– about 2.1 million – were cut off from Italy after it entered the Second World
War. Mussolini had hoped to form a “Black Army” of 1 million men in Ethiopia,
to coerce the population into subjugation for slave labor and then to undertake a
march of conquest across the Sudan and Egypt to the Suez Canal. Instead, by
1940, his East African colonial troops numbered only 200,000, and could do little
more than hold the roads and towns of an Impero ablaze in rebellion.
Furthermore, Italian East Africa had drained Italy of the 250,000 soldiers and
the mostly male settlers there by June 1940. Throughout the Mediterranean at
this time Mussolini commanded only 6.7 million male Italians, and about
180,000 Albanians and Libyans, in the 20–39 age bracket, less than half the
German and Czech manpower available to Hitler. The Fascist regime’s pronatalist
efforts had been heavily offset by its colonial policy As a result, it had gained
only 500,000 more men of military age in Europe and North Africa than it had
More Than Meets The Eye: The Ethiopian War 189
commanded in 1933. During the same period, Nazi Germany had acquired 3.1
million more men fit for war service or heavy labor.36
Albania contained significant unexploited mineral wealth, and Ethiopia
possessed huge agricultural potential. Mussolini believed his East African empire
also contained great deposits of iron, copper, coal, gold, platinum and steel alloys.
He anticipated establishing a vast network of mines, foundries and armaments’
manufacturing plants there, certainly a practical gain of empire to his way of
thinking. But the geological riches the Italians hoped to find never materialized,
and the ongoing insurgency prevented introduction of large-scale mechanized
farming. The Italians did increase Albanian oil, copper, chrome, and bitumen
extraction in 1939–40. But petroleum production remained less than 7 percent
of Italian imports in 1939. The Italians needed more time and investments to
realize significant output from their Balkan colony Both Albania and East Africa
probably would have yielded major profits eventually to Italy But between 1936
and 1940, they only absorbed Italian capital – vast amounts in the case of the
Impero. Between 1936 and 1938, Italy’s exports to East Africa were valued at
5.3 billion lire, her imports at only 477 million.37
Italy’s economy did expand from 1933 to 1939: industrial production rose
by about 40 percent in those years, mostly due to armaments’ output. But this
was much less than German production and without the rich additions the Reich
derived from the factories, farms, and mines of Austria and Bohemia–Moravia.
In early 1933, Italy’s gross national product was about half of Germany’s; by
mid-1939 it had been reduced to 18 percent. Per capita income for middle-class
Italians rose perhaps 5–6 percent between 1930 and 1940; for workers and farm
laborers it actually dropped by about 10 percent. For all levels of German society,
average income doubled during the same time. Small wonder that the Italians
had lost out to the Germans in their economic competition in Spain and south-
east Europe.38
The costs of the Ethiopian War, the East African pacification campaigns, and
involvement in Spain combined to gravely weaken Italy’s economy and military.
Of the $5 billion the Italians spent on their armed forces between mid-1935 and
mid-1939, $2.9 billion went into the conquest and pacification of Ethiopia. With
another $700 million expended on intervention in Spain, this meant that only
28 percent of Italian military outlays between 1935 and 1939 – some $1.4 billion
– was used to strengthen the armed forces in Italy, Libya, and Albania. The
allocation of 34.8 billion lire ($1.8 billion) to the armed forces in 1939–40 – of
which 2.1 billion lire ($110 million) was spent on the military in East Africa –
could not suddenly make up for the previous four years of profligacy Meanwhile,
the Germans were rearming at an all-out pace, followed belatedly by the French
and the British.
The result was a rapid decline in Italian strength from a high point in 1934–
35, when Mussolini could have prevented the Anschluss and defied Britain in the
Mediterranean, to a position of near-impotence in 1938–39. Mussolini could not
halt the spread of Nazi power throughout central Europe and the Balkans, nor
restrain Hitler’s attack on Poland, nor extract territorial concessions from the
190 The Origins of the Second World War Reconsidered
French, nor even risk joining the Germans in war against the western Allies.
Yet over the winter of 1939–40 he decided that Italy had become so vulnerable
that it could not remain neutral. Whichever belligerent won the ongoing conflict
would punish the Italians for perfidy – however defined – at the war’s end.
Furthermore, an unprecedented historical opportunity had arisen. The territorial
aims of the Fascist regime indicated intervention on the German side. It was as
easy for Hitler to promise Mussolini the Mediterranean, Middle Eastern, and
African possessions of the French and British empires, as it was difficult – even
impossible – for Paris and London to concede them. However, the Nazis had
treated Fascist Italy so poorly in the 1933–39 period, when Hitler had needed
Mussolini’s support, that an obvious question arose: how would a victorious
Third Reich, the master of a prostrate Europe, deal with its far weaker Italian
ally after they had together defeated the western powers? The fluid European
relationships of 1933–36 had solidified considerably by early 1940. Nonetheless,
Mussolini still faced alternatives: he was not compelled to enter Hitler’s war
against the western Allies.39
Although the full consequences of the Ethiopian War were not yet clear to
Mussolini in mid-1936, he knew that he had miscalculated. He understood
that the East African conflict had sapped Italy’s strength far more than
anticipated. But this does not explain his decision to tie Fascist Italy so closely
to Nazi Germany later that year. Mussolini had agreed to something like an
Italian– German condominium in Austria when, the previous January, he
had desperately needed Hitler’s support. But there were no practical reasons
why he had to honor this agreement: he lied habitually when he considered
it advantageous to do so. Furthermore, Hitler had broken several promises
to Mussolini, notably in July 1934 and again over his shipments of arms to
Ethiopia. Well into 1937, Hitler himself suspected that Mussolini was
deceiving him by avowing a new Austrian policy. The Germans enjoyed no
greater influence in Austria immediately after the Ethiopian War than before.
Thus, they could not bring any greater pressure on Rome or Vienna. But,
from June 1936, Mussolini urged Schuschnigg to reach a modus vivendi with
Hitler. Mussolini sometimes refused to admit the consequences to others –
occasionally even to himself – but he knew the July 1936 agreement marked
a major step toward an Anschluss.40
That Mussolini chose to ally with Hitler, rather than being forced, is
indicated by the approaches made to Rome by the French and the British
throughout 1936. For their part, the French repeatedly made clear their wish
to reinvigorate the Stresa Front and maintain the Mussolini–Laval pact of 1935.
France’s leaders displayed an almost humiliating determination to retain Italy
as an ally. This is remarkable. The remilitarization of the Rhineland had ended
Mussolini’s obligation to aid France against such a German action. Yet the
French were offering unilateral assistance to Italy to prevent an Anschluss. The
French also guaranteed that they would help end League sanctions against Italy,
which they did.41
More Than Meets The Eye: The Ethiopian War 191
But Mussolini despised the leftist Popular Front government that came to
power in Paris in early June 1936, considering it another sign of French
decadence. This undermined his willingness to maintain an alliance with France.
He ignored Premier Blum’s appeals – seconded by both the French and Italian
high commands – to set aside their differences to prevent German hegemony
over Europe. But the failure of the previous French government to respond with
force to the German move into the Rhineland made him doubt that he could
count on the French to help stop a similar move into Austria. He considered
the new Popular Front government even weaker. Certainly, Blum’s decision to
aid the Spanish Republicans and Mussolini’s to aid the Nationalists in late July
made further cooperation between their two governments very unlikely. By late
August 1936, the Fascists and Nazis were coordinating their military efforts in
Spain, and the possibility of reviving the Italian – French anti-German partnership
had evaporated. However, Mussolini made his decision to intervene in Spain
slowly and deliberately. He was not forced into a proxy war alongside Germany
against France: he had acted freely.42
The British made similar appeals to Mussolini immediately before, during,
and after the Ethiopian War. In September 1935, the Baldwin government had
confidentially informed Mussolini that Royal Navy movements in the
Mediterranean were meant only to bolster public opinion. Britain would not
take military action against Italy, nor would it close the Suez Canal. Even before
the Italians invaded Ethiopia, British representatives had assured Mussolini that
their government’s supposed pro-League position was only a sham to satisfy the
British public. British diplomats insisted that London wanted to maintain the
closest possible relations with Rome.43
By February 1936, after Italian victories in Ethiopia had made Mussolini’s
adventure appear successful, members of the governing Conservative Party began
assuring Grandi, the Italian ambassador, that they considered war with Italy out
of the question, that they opposed sanctions, and that they considered the League
a troublesome nuisance. Behind many of these pleas to restore Italian–British
amity lay exhortations from the Admiralty which was concerned about the
vulnerability of imperial communications through the Mediterranean. The
strongest proponents of regaining the cooperation of Mussolini were Neville
Chamberlain and, most of all, Winston Churchill, who had long admired
Mussolini and who now praised him to Grandi as the greatest statesman in
Europe.44 With an Italian triumph in East Africa a fait accompli, Churchill and
Chamberlain hoped to regain Mussolini as an ally – against Hitler. Baldwin,
they insisted, had been misguided in his pro-League policies, encouraged by that
unbalanced Italian-hater and fool, Anthony Eden.
Through Grandi, Mussolini encouraged his English friends. Together with the
French, the British had removed League sanctions in July 1936. And, like the
French, their longing for Italian friendship and their obvious fear of Germany had
convinced Mussolini that they too were scared and weak. Italian aggression had
been rewarded not only with a great empire in East Africa but with demonstrations
of enhanced respect from Paris and London. The Stresa Front was hardly gone
192 The Origins of the Second World War Reconsidered
beyond recall. In the spring and summer of 1936 London and Paris begged
Mussolini to revive it. They did not condemn Italy as an aggressor, except in the
most formal sense. The French and British were hailing Mussolini as the potential
savior of Europe from Hitler. Unfortunately for them, such importuning raised a
fatal question in Mussolini’s mind: What might Italy achieve if it pressed its
advantage, especially in concert with Hitler’s new Germany?45
British appeasement, which had steadily intensified by mid-1936, was applied
more toward Fascist Italy than Nazi Germany. This emphasis increased after
Mussolini intervened in the Spanish Civil War. Britain and France must remain
neutral in that conflict, Churchill insisted to the Italian chargé d’affaires Leonardo
Vitetti in August. But Italy, he argued, had to join in the fighting to prevent the
spread of communism in western Europe – and many British officials privately
agreed. Churchill himself worried that the Popular Front government presaged a
communist France. Mussolini must not let that happen in Spain as well, he warned.
A few days earlier, Laval had said almost the same to the Italian ambassador in
Paris about the Duce’s intervention in Spain. Two months later Robert Vansittart,
the British permanent under-secretary of state for foreign affairs, delivered a similar
message to Grandi, although in more guarded terms. Mussolini ordered intelligence
reports showing the wide extent of French and Soviet military aid to the Spanish
Republicans to be circulated in London. This produced dismay among British
leaders. Only a few, notably Vansittart and Churchill, eventually realized that the
Fascists were playing them for dupes. But, in Churchill’s case, illusions about
Mussolini lasted far longer than he later would admit.
The British Conservatives continued to believe that stopping the spread of
communism in Spain and reaching a lasting understanding with Britain were
Mussolini’s principal goals. Among those taken in by Mussolini’s lies, Chamberlain
stood out. Seven months after he became prime minister in May 1937, Chamberlain
shunted aside Vansittart in favor of the more optimistic Alexander Cadogan. A
few months later Eden, in protest over Chamberlain’s relentless appeasement of
Italy, was forced to resign. What conclusions could Mussolini draw from such
short-sightedness, other than that the French and British were divided, each
hopelessly in moral decline? In early 1938 it certainly appeared that Fascist Italy’s
best course was to align itself with dynamic Nazi Germany.46
Yet even after the Anschluss, when German expansion to Italy’s northern
frontier seemed to have locked Mussolini permanently into the Axis, Hitler
himself remained unsure of the future course of Italian foreign policy. In April
1938 Hitler remarked to his adjutant that Mussolini might be satiated, having
conquered Ethiopia and gained British recognition of its annexation through the
Easter Accords. In that case, he mused, further German territorial
aggrandizement would occur only in the “distant future.”47
Despite Hitler’s worry over Mussolini’s support, the Nazi treatment of Italy
most likely to alienate the Duce took place from March 1938 to August 1939.
Any one of the incidents previously mentioned presented Mussolini with an
excellent reason to break with Hitler. Even after the signing of the Pact of Steel,
Hitler’s broken promises to avoid a major war for three or four years and to
More Than Meets The Eye: The Ethiopian War 193
consult prior to any major foreign policy initiative, as well as the shock of the
Nazi–Soviet Pact, offered sufficient grounds to renounce the Axis. As late as
December 1939, this struck many well-informed observers as a distinct possibility.
Certainly it was Ciano’s preferred option. Mussolini recommitted himself to
Hitler in March 1940, well before German victories in Scandinavia and the west.
The explanation lies in Mussolini’s foreign policy and ideology, both set years
before the Nazis came to power.
Mussolini had proclaimed himself an imperialist as early as 1920. In 1922
he declared that Italy must dominate the Mediterranean, recreate the Roman
empire and destroy that of Britain. Mussolini also sought to end French power
in the Mediterranean. In February 1923, he ordered his war ministry to instigate
an insurrection on Corsica. After the army leadership refused, Mussolini assigned
the Fascist Party the task. Six months later he prepared to fight Britain for Corfu,
backing down only when his admirals convinced him that to do so would prove
suicidal. Shortly after he became dictator in January 1925, he adopted a long-
term program of expansion. He planned to conquer Ethiopia, but as a step toward
joining “the two shores of the Mediterranean and of the Indian Ocean into a
single Italian territory.”48 This required the acquisition of territory from Tunisia
to Kenya. In late 1926, Mussolini explained his plans to the army general staff:
“A nation which does not have free access to the sea cannot be considered a
free country. A nation which does not have free access to the oceans cannot be
considered a great power. Italy must become a great power.”49
Escaping the “Mediteranean prison” became the focus of Mussolini’s foreign
policy. Secure passage to the oceans, however, required war with France or
Britain, or their surrender. Conquering Ethiopia strengthened Italian control of
Somalia’s coast and facilitated any offensive to link Libya to the Impero. But these
were only steps toward the goal. Thereafter, Mussolini’s actions – especially
intervention in Spain – and his words openly indicated his geographic aims.
When declaring war on June 10, 1940 Mussolini was explicit. The Italian people,
he insisted, wanted “to break the bars which suffocate them in mare nostrum,
because a people…is not truly free if they do not have free access to the ocean.”50
However, the Corfu crisis and an appreciation of the balance of power taught
Mussolini harsh truths. Italy was too weak to threaten the French or British
empires. After studying the demographics, he concluded that prospects would
improve after 1934, when the impact of losses in the First World War would
begin to reduce British and, especially French, military manpower. This largely
explains Fascist pro-natalist policies. The Duce wanted large armies for war with
the West. However, even with growing manpower advantages, Mussolini decided
he needed allies for a successful struggle. Countries like Hungary and Bulgaria
could assist in conflicts against lesser states, such as Yugoslavia. But to defeat
the imperial powers, Italy needed major partners. In Europe, only Germany and
the Soviet Union met his specifications. He flirted with Moscow in 1933–34, in
case playing the French and Germans off against each other went wrong. But
given his ardent anti-Bolshevism, from 1924–25 Mussolini’s preferred co-
conspirator to overthrow the Versailles order had been Germany.51
194 The Origins of the Second World War Reconsidered
In August 1923, two weeks before the Corfu incident, another SIS report
reached the foreign office. The Indians had met Mussolini again. He had promised
them financial support. However, the nationalists had grown suspicious and began
to wonder if Mussolini’s intentions were to expel the British only to impose Italian
rule on India. Once again, however, British official reactions were dismissive: “It
is all very vague and intangible....I take all such reports with a considerable amount
of reserve”; and “‘Cum grano salis’ applies to all these reports.” Among those who
had annotated the SIS documents were Lord Curzon, Eyre Crowe, Harold
Nicolson, Miles Lampson, Robert Vansittart, and Alexander Cadogan.55
French intelligence on Mussolini’s intentions to attack France in the 1920s is
less certain. But, like the British, the French had broken some Italian codes. Italian
planning for conflict with France arose from plans to destroy Yugoslavia. In
November 1927 the French and Yugoslav governments signed a pact of amity
and understanding. Sixteen months later, Mussolini ordered his high command
to plan an attack on France simultaneously with an offensive against Yugoslavia.
Studies continued until at least late 1931. Evidence strongly indicates that the
Deuxième Bureau gained knowledge of such plans. In July 1932, Mussolini
contacted Kurt von Schleicher, the German chancellor, about Italy and Germany
waging war together against France; Schleicher replied favorably, but added that
Germany needed time to rearm. Later that month Mussolini ordered that plans
be drawn up for an invasion of Corsica. He also authorized Ustaše landings in
Dalmatia to spark a Croat revolt, although the uprising was crushed that fall.
In January 1933 he proposed to invade Yugoslavia that summer, and to instigate
another Croat uprising. Next, he considered using his army’s August maneuvers
to cover a surprise attack on France. He believed that France, politically unstable
since mid-1932, was nearing collapse.56
However, the Deuxième Bureau discovered Mussolini’s intentions. General
Weygand urged a preventive attack on both Italy and Germany, guaranteeing
Italian defeat in a few months. His confidence indicates that the French general
staff had acquired a thorough picture of Italian military dispositions and planning.
Soon after, Italian military intelligence learned that the French had uncovered
Italian plans and were making their own war preparations. Mussolini postponed
invasion of Ethiopia and delayed his proposed attack on France. He preferred
to wait, to join the Germans in crushing the French in 1934. Meanwhile, the
Austrian question had begun to hinder Nazi–Fascist cooperation. Furthermore,
Hitler stated that Germany would remain unready for a major war for several
years. Mussolini’s advisers persuaded him that these circumstances made conflict
with France a folly, and he reluctantly agreed. His thoughts returned to Ethiopia
and the need of an agreement with France to render practical his East African
conquest under the prevailing European conditions. Nonetheless, Italian military
planning since 1927 indicates that Mussolini’s post-Ethiopian War hostility to
France was of long duration, not a reaction to the crisis of 1935–36.57
In May 1934 Mussolini and the Italian high command marked Britain as
Italy’s main European enemy. In discussing the impending attack on Ethiopia,
Marshal Pietro Badoglio, chief of the supreme general staff, advised deploying
196 The Origins of the Second World War Reconsidered
Twelve years later Mussolini explained his reasons for invading Ethiopia to the
Hungarian military attaché in similar terms:
For the past thirteen years I have been asking, begging, threatening so that
the Italian people get their own place within lawful boundaries. I want the
Italians to be able to earn their own bread and to be liberated from having to
More Than Meets The Eye: The Ethiopian War 197
work for starvation wages at the arbitrary wishes of foreign powers. I want
these poor masses to have enough in order to eat sufficiently at least once a
day, because at this moment they eat no more than once a day. But even so,
they are eating very poorly. What I could achieve on Italian soil by improving
this earth, I have already done. This cannot be forced any further…
Even if we sacrificed billions on our existing colonies, the first crops from olive
trees would be ripe no sooner than in fifteen to twenty years. We need territories,
otherwise we shall explode. Thus, when I gained a free hand from Laval, I myself
have “exploded.” There is no stopping here....We shall do everything in our power,
we shall sacrifice everything we must, but we shall not surrender our aims!60
Mussolini believed Fascism gave Italy the power to solve its domestic
problems through war. But he also calculated that military victory would
strengthen his regime’s grip on Italian society. Furthermore, both Duce and
Führer believed their systems provided revolutionary solutions to the crisis
afflicting Europe. Their alliance would smash the “pluto-democracies” of the
west, then bend all mankind to their will. As Hitler remarked to an Italian
diplomat in early 1933: “Fascism is a force which must be imposed on the
world. This common ideal will make us constantly stronger and more
united.”61 Before the Nazi takeover and during his regime’s first year, Hitler
flattered Mussolini, calling himself the Duce’s disciple. As he gained
confidence, the Führer preferred to refer to the Duce as his equal. In any case,
both agreed that they had nothing in common with the “capons” and “worms”
who governed Britain and France.62
Although Hitler’s words were meant to manipulate Mussolini, he was sincere.
He viewed Fascist institutions as models he intended to emulate. Equally important,
while Nazi Germany remained weak Hitler needed Fascist Italy for support. The
Führer’s personal daring, unshakable beliefs and the nature of Nazism led him to
the failed Anschluss of July 1934. Although the coup infuriated Mussolini, Hitler did
not reciprocate that hostility. He was certain fate and faith linked Nazism and Fascism
indissolubly. The Duce’s reckless courage in defying Britain and the League during
the Ethiopian War deeply impressed Hitler. He also recognized the danger to
Mussolini’s survival. That is why Hitler ended German arms’ shipments to Ethiopia
in late 1935 and also explains his welcome of Mussolini’s January 1936
rapprochement. True, the Führer used Italian difficulties to advance his plans for
Austria and, more immediately, to move forces across the Rhine. But Hitler hardly
wanted the Duce to succumb to economic sanctions or military defeat.63
Mussolini also concluded that the survival of the two regimes was linked.
As father of radical-right totalitarianism, both pride and lust for power motivated
him to seek the expansion of Fascism and Nazism. On the tenth anniversary of
the March on Rome, Mussolini had proclaimed: “the 20th century will be the
century of Fascism, will be the century of Italian power, will be the century during
which Italy will return for the third time as the director of human
civilization....Within a decade all Europe will be Fascist or fascistized.”64
198 The Origins of the Second World War Reconsidered
By late 1937, Mussolini had come to the painful recognition that the goal lay
beyond his power. He disliked Hitler but needed Nazi power to fulfill his historic
mission. In September 1938, when Mussolini feared general war due to Hitler’s
demands on Czechoslovakia, he pledged solidarity with Germany. He knew the
parlous state of Italian armaments and the grave dangers in joining a conflict
with the West; he believed, however, that Hitler could not win such a war alone
and, if Nazi Germany fell, that Fascist Italy would become the target of British
and French revenge.65
A year later, Mussolini still had not entered the war started by Hitler, though
he longed to do so. Under a barrage of arguments from his advisers, he agreed
that his armed forces were too weak – the result of his profligate use of resources
since 1935. But he estimated that Germany had grown so powerful, and the
Western leaders so timid, that the Nazi regime could survive fighting Britain
and France on its own. As soon as possible, however, Mussolini intended to
intervene on Hitler’s side. Neutrality negated the meaning of Fascism. It also
meant missing the opportunity to gain all that he had been seeking since 1922.
When the opportunity presented itself again in the spring of 1940, Mussolini
seized it in order to achieve his dream of re-establishing the Roman empire.66
But Mussolini’s decision was carefully and consciously made; British and
French policies had not forced him on to Hitler’s side. Instead, in a manner no
one then realized, the Ethiopian crisis had exposed vulnerabilities and created
opportunities that he seized to realize his imperial vision. He had long believed
Italy could gain world-power status through conquest of an empire stretching
from the Strait of Gibraltar to the Strait of Hormuz. In June 1940 he acted on
that faith and in the hope that he had nurtured for two decades. Over the next
five years, he and all Italians would pay dearly for that fatal mistake.
Notes
1 The better studies – including the works they cite – of European diplomacy during
the Ethiopian crisis are: P. M. H. Bell, The Origins of the Second World War in Europe,
London, 1986; Hans-Jürgen Döscher, Das Auswärtige Amt im Dritten Reich: Diplomatie
im Schatten der Endlösung, Berlin, 1986; A. R. Peters, Anthony Eden at the Foreign Office,
1931– 1938, New York, 1986, pp. 114–206; Harold G. Marcus, Haile Selassie I. The
Formative Years, 1892–1936, Berkeley, CA, 1987, pp. 124–80; Arianna Arisi Rosa,
La diplomazia del ventennio. Storia di una politica estera, Milan, 1990, pp. 73–130; Alan
Cassels (ed.), Italian Foreign Policy, 1918–1945, revised edn, Wilmington, DE, 1991;
Patrick Finney (ed.), The Origins of the Second World War, London, 1997.
2 In addition to the books previously mentioned, among the better works dealing
with European international relations for 1936–40 are: Wolfgang Michalka, Ribbentrop
und die deutsche Weltpolitik, 1933–1940, Munich, 1980; Brunello Vigezzi, L’Italia unita e
le sfide della politica estera. Dal Risorgimento alla Repubblica, Milan, 1997, pp. 177–259.
3 Mussolini’s policies, August 1939–June 1940, are brilliantly analyzed in MacGregor
Knox, Mussolini Unleashed, 1939–1941. Politics and Strategy in Fascist Italy’s Last War,
New York, 1982. For a view of Mussolini’s diplomacy incorporating the Taylor
thesis, with documentation unavailable in 1961, see Richard Lamb, Mussolini and
the British, London, 1997.
More Than Meets The Eye: The Ethiopian War 199
16 DDF, 2e série: 1936–1939, tome 2, nos 275, 482; tome 3, nos 46, 104, 106, 341,
502, 509; tome 9, nos 335, 339, 355, 360, 372; tome 12, nos 288, 433; tome 13,
nos 1, 9, 15, 17, 26, 53, 166, 193–4, 270, 313, 354, 377, 426, 448, 459; tome 14,
nos 6, 82, 121, 138, 176, 237; tome 15, nos 212, 286, 510; DGFP, series D, vol. 1,
nos 736–41, 746, 751, 766, 799; vol. 4, no. 412; OO, vol. 29, pp. 99–102, Mussolini
quotations, pp. 100, 101; Ciano, Diario, pp. 123, 212, 217–28, 234–7; Ciano’s
Diplomatic Papers, pp. 247–53; William Shorrock, From Ally to Enemy: The Enigma of
Fascist Italy in French Diplomacy, 1920–1940, Kent, OH, 1988, pp. 181–251.
17 Ciano, Diario, pp. 238–40, 244, 246–7, 254, 275, 289, 291, 311, 329, 396, 421–2,
434–5; Ciano’s Diplomatic Papers, pp. 257–66; DBFP, series 3, vol. 3, nos 500–2; DDF,
2e série, tome 14, nos 11, 26, 46, 112, 131, 281; DGFP, series D, vol. 4, nos 447,
452; Shorrock, From Ally, pp. 252–86.
18 Contemporary knowledge of events, 1933–40, can be obtained from the relevant
volumes of Survey of International Affairs, London, 1934–58.
19 Brian R. Sullivan, “From little brother to senior partner: Fascist Italian perceptions
of the Nazis and of Hitler’s regime, 1930–1936,” Intelligence and National Security, vol.
13, 1998, pp. 90–6.
20 DDI; series 8: 1935–1939, vol. 3, nos 124, 261, 275, 322, 396, 433, 466; DGFP, series
C, vol. 4, nos 212, 485, 575, 579, 592, 598; Manfred Funke, Sanktionen und Kanonen:
Hitler, Mussolini and der internationale Abessinienkonflikt, 1934–1936, Düsseldorf, 1970, pp.
43–5, 59 n. 70, 68–9; James Thomas Emmerson, The Rhineland Crisis, 7 March 1936. A
Study in Multilateral Diplomacy, London, 1977, pp. 75–103, 152–4, 202–3.
21 DDI, series 8, vol. 4, nos 715, 787, 800, 819; vol 5, nos 46, 95, 140, 305, 316, 397,
425, 429, 520, 531, 546, 598, 605, 614, 627, 630, 644, 648, 649, 680, 691; DGFP,
series D, vol. 3, nos 30, 47, 84; De Felice, Mussolini il duce, vol. 2, pp. 368–84.
22 For examples of such statements, see DGFP, series C, vol. 1,no 14 [Feb. 7, 1933];
no. 166 [June 24, 1935]; DDI, series 8,vol. 5, no. 277, p. 317 [Oct. 24, 1936]; vol.
6,no.60 [Jan.15, 1937]; DGFP, series D,vol. 1, no. 769 [May 20, 1938]; Ciano, Diario,
p. 299 [May 23, 1939].
23 DGFP, series C, vol. 3, nos 13, 18, 169, 175, 207, 250, 316; vol. 4, nos 22, 110,
209, 307; Hjalmar Schacht, My First Seventy-Six Years, London, 1955, pp. 327–34.
24 DGFP, series C, vol. 4, nos 283, 310, 312, 369, 434, 446, 459, 481, 539, 557; vol.
6, nos 86, 368; series D, vol. 1, nos 84, 745; vol. 4, nos 399, 414, 427–9, 438,
445–6, 448, 457; DDI, series 8, vol. 5, no. 567.
25 DGFP, series C, vol. 4, nos 303, 330, 445, 450; John F. Coverdale, Italian Intervention
in the Spanish Civil War, Princeton, NJ, 1975; Alberto Rovighi and Filippo Stefani,
La partecipazione italiana alla guerra civile spagnola, 1936–1939, 2 vols, Rome, 1992–
93; vol. 2: Dall autunno 1937 all estate del 1939. Testo, pp. 505–6.
26 DGFP, series D, vol. 1, no. 795; vol. 2, nos 198, 248, 284, 296, 390, 395, 402,
541, 554–5, 557, 577; vol. 4, nos 163, 165, 167, 179–83, 198–9, 214, 217, 237, 243;
Maria Ormos, “L’opinione del conte Stefano Bethlen sui rapporti italo – ungheresi,
1927–1931,” Storia contemporanea, vol. 2, 1971, pp. 283–314; Alfredo Breccia, “La
politica estera italiana e l Ungheria, 1922–1933,” Rivista di studi politici internazionali,
vol. 47, 1980, pp. 93–112; C. A. Macartney, October Fifteenth. A History of Modern
Hungary, 1929–1945, Edinburgh, 1956, part 1, pp. 330–63; Thomas Sakmyster,
Hungary, the Great Powers, and the Danubian Crisis, 1936–1939, Athens, GA, 1980, pp.
223–7, and Hungary’s Admiral on Horseback. Miklós Horthy, 1918–1944, New York, 1994,
p. 234; George Punka, Hungarian Air Force, Carrollton, TX, 1994; Jörg K. Hoensch,
A History of Modern Hungary, 1867–1994, New York, 1996, pp. 132, 143–5.
27 Dov B. Lungu, Romania and the Great Powers, 1933–1940, Durham, NC, 1989, pp.
140, 153–237; Keith Hitchins, Romania, 1866–1947, Oxford, 1994, pp. 441–50;
Mark Axworthy, Cornel Scafe, and Cristian Craciunoiu, Third Axis, Fourth Ally.
Romanian Armed Forces in the European War, London, 1995, pp. 17–30, 266–76; Angela
Raspin, The Italian War Economy 1940–43, London, 1986, p. 410.
More Than Meets The Eye: The Ethiopian War 201
28 DDI, series 8, vol. 4, nos 53, 99, 248, 293, 371, 555, 558, 782; vol. 5, nos 329,
458, 459; vol. 6, nos 118, 211, 340; DGFP, series C, vol. 6, nos 20, 27, 254, 565;
series D, vol. 4, no. 411; vol. 6, nos 15, 45, 55; Ciano’s Diplomatic Papers, pp. 276–
81; J. B. Hoptner, Yugoslavia in Crisis, 1934–1941, New York, 1962, pp. 31–161.
29 DDI, series 8, vol. 5, no. 251; vol. 6, nos 29, 348; vol. 12, nos 10, 100, 214, 300,
419, 467, 510, 559, 792; vol. 13, nos 165, 240, 413; Ciano, Diario, pp. 206–7, 243,
245, 262–88; DGFP, series D, vol. 4, nos 449, 450; Marshall Lee Miller, Bulgaria
During the Second World War, Stanford, CA, 1975, pp. 5–23; R. J. Crampton, A Concise
History of Bulgaria, Cambridge, 1997, pp. 168–70.
30 Ciano, Diario, pp. 108–14, 183–6, 214–23, 244–9, 262–71, 299–300, 325–42, 374–5;
Ciano’s Diplomatic Papers, pp. 282–8, 296–304; DGFP, series D, vol. 1, nos 152, 155, 160–
1, 207–8, 348–50, 352, 361, 399; vol. 2, nos 610, 627, 661, 662; vol. 4, nos 43, 57–55,
68, 74–80, 84–98, 105, 118–34, 139, 365–76, 382, 384, 408, 412, 447, 463; vol. 6, nos
86, 87, 140; DDI, series 8, vol. 13; DBFP, series 3, vol. 7, no. 679; DDF, 2e série, vol.
13, nos 87, 93, 202, 348, 351, 428, 456; Gerhard L. Weinberg, The Foreign Policy of Hitler’s
Germany, 2 vols, Chicago, 1970 and 1980; vol. 2: Starting World War Two, 1937–1939;
Watt, How War Came; De Felice, Mussolini il duce, vol. 2, pp. 467–674.
31 Alberto Sbacchi, “The price of empire: toward an enumeration of Italian casualties in
Ethiopia, 1935–40,” in Alberto Sbacchi (ed.), Legacy of Bitterness. Ethiopia and Fascist Italy, 1935–
1941, Lawrenceville, NJ, 1997; Brian R. Sullivan, “The Italian–Ethiopian War, October
1935–November 1941: causes, conduct, and consequences,” in A. Hamish Ion and E. J.
Errington (eds), Great Powers and Little Wars. The Limits of Power, Westport, CT, 1993, pp. 185–94.
32 Brian R. Sullivan, “Fascist Italian involvement in the Spanish Civil War,” The Journal
of Military History, vol. 59, 1995, p. 713; Virgilio Ilari, Storia del servizio militare in Italia,
4 vols, Rome, 1989–91; vol. 4: Soldati e partigiani, 1943–1945, pp. 23, 28; B. R.
Mitchell, International Historical Statistics. Europe, 1750–1988, London, 1992, p. 6.
33 Ministero del Tesoro, Ragioneria Generale dello Stato, Il Bilancio dello Stato negli esercizi
finanziari dal 1930–31 al 1941–42, Rome, 1951, pp. 203, 205, 257, 373, 403, 407;
Francesco A. Répaci, “Le spese delle guerre condotte dall Italia nel ultimo
quarantacinquennio, 1913–14 – 1957–58,” Rivista di politica economica, vol. 50, 1960, pp.
695–713; Sullivan, “Italian–Ethiopian War,” p. 185; Maurice F. Neufeld, Italy: School
for Awakening Countries, Ithaca, NY, 1961, pp. 401, 411; Vera Zamagni, “Un analisi
macro-economica,” in Zamagni, ed., Come perdere la guerra e vincere la pace,
Bologna, 1997, p. 46; Alberto Sbacchi, “Italian mandate or protectorate over Ethiopia,
1935 36,” Rivista di studi politici internazionali, vol. 42, 1975, p. 588; Raspin, Italian War
Economy, pp. 434–6; Virgilio Ilari and Antonio Sema, Marte in Orbace. Guerra, esercito e
milizia nella concezione fascista della nazione, Ancona, 1988, p. 261; Sullivan, “Fascist Italian
involvement,” pp. 711, 723; Knox, Mussolini Unleashed, pp. 292–6; Mitchell, Statistics,
pp. 800, 895; The World Almanac & Book of Facts for 1942, New York, 1942, p. 515.
34 DGFP, series D, vol. 4, nos 170, 201; Williamson Murray, The Change in the European
Balance of Power 1938–1939. The Path to Ruin, Princeton, NJ, 1984, pp. 149–53, 290–
2; Mitchell, Statistics, pp. 3, 4, 13, 16, 22, 892, 894.
35 Giorgio Rochat, Il colonialismo italiano, Turin, 1973, pp. 28, 103, 146, 187; Mitchell, Statistics,
pp. 3, 6, 12, 28; Sullivan, “Italian–Ethiopian War,” pp. 191, 193; Aloisi, Journal, p. 382;
Ferdinando Quaranta, Ethiopia: An Empire in the Making, London, 1939, pp. 95–6; SP,
1936, file 11: “Memorandum about the talks with the Duce on June 12, 1936,” p. 2.
36 Aloisi, Journal, p. 382; Sbacchi, “Italian colonization in Ethiopia: promises, plans and
projects, 1936–1940,” Legacy, pp. 103–22; James C. McCann, People of the Plow. An
Agricultural History of Ethiopia, 1800–1990, Madison, WI, 1990; Renato Mori, “Delle
cause dell impresa etiopica mussoliniana,” Storia e politica, vol. 17, 1978, pp. 689–706;
Quaranta, Ethiopia, pp. vii, 19–110; Derek Hall, Albania and the Albanians, London,
1994, pp. 22–4, 102–5; Ciano, Diario, pp. 330–1, Raspin, Italian War Economy, p. 407.
37 Mitchell, Statistics, pp. 894–5; The World Almanac 1942, p. 515; Raspin, Italian War
Economy, pp. 11–51, 62–104; Neufeld, Italy, pp. 530, 538, 540.
202 The Origins of the Second World War Reconsidered
38 Vigezzi, L’Italia unita, pp. 199–204, 243–59; Salerno, “Multilateral strategy,” pp. 66–
70; Brian R. Sullivan, “The Impatient Cat: Assessments of Military Power in Fascist
Italy, 1936–1940,” in Williamson Murray and Allan R. Millett (eds), Calculations.
Net Assessment and the Coming of World War II, New York, 1992, pp. 112–35.
39 DGFP, series C, vol. 4, nos 486, 506, 525, 545, 569; series D, vol. 1, nos 152,
155, 207–8; DDI, series 8,vol. 4, nos 192, 448, 489, 496, 499; vol. 6, no. 330; De
Felice, Mussolini il duce, vol. 1, pp. 755–6.
40 Anthony Adamthwaite, France and the Coming of the Second World War, 1936–1939,
London, 1977, pp. 32–6; Robert J. Young, In Command of France: French Foreign Policy
and Military Planning, 1933–1940, Cambridge, MA, 1978, pp. 110–14; Jean-Baptiste
Duroselle, La Décadence, 1932–1939, Paris, 1979, pp. 147–57.
41 DDF, 2e série, vol. 2, nos 248, 275, 374, 454, 482; vol. 3, no. 114; DDI, series 8,
vol. 4, nos 190, 223, 227, 294, 326, 383, 490, 589, 598, 601, 677, 684, 685, 738,
819; Jordan, “Maurice Gamelin,” pp. 434–5.
42 DBFP, series 2, vol. 14, nos 591, 603–4, 611, 620, 630, 662, 663; DDI, series 8,
vol. 2, nos 166, 181, 189, 209, 236; Baer, Coming, pp. 323–66.
43 DBFP, series 2, vol. 15, nos 98, 154, 175, 198, 317, 427.
44 “Mr Churchill on Fascism,” The Times, Jan. 21, 1927; Grandi, Il mio paese, p. 234; Yvon De
Begnac, Taccuini mussoliniani, Bologna, 1990, p. 352; De Felice, Mussolini il duce, vol. 1, p. 553.
45 DDI, series 8, vol. 3, nos 251, 491, 808, 815, 846; vol. 4, nos 94, 236, 242, 251, 262,
283, 286, 338; Alberto Pirelli, Taccuini 1922/1943, Bologna, 1984, pp. 170–7; Peters,
Eden, pp. 196–206; Salerno, “Multilateral strategy,” pp. 65–6; Daniel Waley, British Public
Opinion and the Abyssinian War, 1935–6, London, 1975, pp. 77–88, 117–35.
46 DDI, series 8, vol. 4, nos 678, 686, 701, 708; vol. 5, nos 182, 368; vol. 6, nos 144, 425; Winston
Churchill, The Gathering Storm, New York, 1948, pp. 220–2, 240–67; The Diaries of Sir Alexander
Cadogan, 1938–1945, David Dilks (ed.), New York, 1972, pp. 27–51; Peters, Eden, pp. 321–51;
M. L. Roi, “From the Stresa Front to the Triple Entente: Sir Robert Vansittart, the Abyssinian
crisis and the containment of Germany,” Diplomacy & Statecraft, vol. 6, 1995, pp. 61–90.
47 DGFP, series D, vol. 2, no. 132.
48 OO, vol. 12, p. 323; vol. 18, pp. 160–1, 439, 535–6; G. A. Chiurco, Storia della
rivoluzione fascista, 2 vols, Florence, 1929, vol. 1, p. 270; DDI, series 7, vol. 1, nos
427, 548, 705, 742; vol. 3, no. 604; vol. 4, nos 420, 460; Ezio Ferrante, “Un rischio
calcolato? Mussolini e gli ammiragli nella gestione della crisi di Corfù,” Storia delle
relazioni internazionali, vol. 5, 1989, pp. 231–39; DBFP, series 1, vol. 24, nos 664,
704; Pirelli, Taccuini, pp. 54, 123–4; Gustavo Pesenti, Fronte Kenya, Borgo San
Dalmazzo, 1953, p. 12, quotation; MacGregor Knox, “The Fascist regime, its foreign
policy and its wars: an ‘anti-anti-Fascist’ orthodoxy?’, in Finney (ed.), Origins, p.162;
Archivio Centrale dello Stato, Carte Badoglio, busta 4, fascicolo 6, nos 1, 2; Philip
Morgan, Italian Fascism, 1919–1945, London, 1995, pp. 131–40.
49 Emilio Canevari, La guerra italiana, 2 vols, Rome, 1948, vol. 1, pp. 211–12.
50 National Archives [hereafter NA], microfilm series T586, reel 405, frames 39–46;
DDI, series 9, vol. 3, no. 669; OO, vol. 29, p. 404, quotation; MacGregor Knox,
“Il fascismo e la politica estera italiana,” in R. J. B. Bosworth and Sergio Romano
(eds), La politica estera italiana (1860–1985), Bologna, 1991, pp. 296–302.
51 Papers of General Pietro Gazzera, [henceforth GP], supplied to author by Renzo De
Felice. Meetings with Mussolini, Aug. 7, 1929, Jan. 27, 1931; OO, vol. 22, p. 386; vol.
23, p. 177; Enrico Caviglia, Diario, Rome, 1952, p. 60; Augnsto Turati, Fuori dell’ombra
della mia vita. Dieci anni nel solco del fascismo, Brescia, 1973, p. 21; J. Calvitt Clarke, Russia
and Italy Against Hitler. The Bolshevik–Fascist Rapprochement of the 1930s, New York, 1991;
De Felice, Mussolini l’alleato, vol. 1, p. 59; Knox, “The Fascist regime,” pp. 160–1.
52 Jens Petersen, Hitler e Mussolini. La difficile alleanza, Bari, 1975, pp. 11–173; Knox, “The
Fascist regime,” pp. 160–1: GP, meeting with Mussolini, Jan. 27, 1931; NA, record
group 331, reel 476A, extract no. 11: “Address by Mussolini to the Fascist Grand
Council – 7 April 1932,” pp. 11–13, 18–19; Grandi, Il mio paese, 315–16, 341– 2, 352–
More Than Meets The Eye: The Ethiopian War 203
3, 359–60; DDI, series 7, vol. 12, nos 38, 393, 449, 460, 534; vol. 13, nos 305, 406;
Gregory Alegi, “Balbo e il riarmo clandestino tedesco. Un episodio segreto della
collaborazione italo-tedesco,” Storia contemporanea, vol. 23, 1992, pp. 305–17.
53 Alan Cassels, “Deux empires face à face: la chimère d’un rapprochement anglo-
italien (1936–1940),” Revue d’histoire de la deuxiéme guerres mondiales et conflits contemporains,
vol. 41, 1991, pp. 67–96; Romain Rainero, La rivendicazione fascista sulla Tunisia, Milan,
1978; Juliette Bessis, La mediterranée fasciste. L’Italie mussolinienne et la Tunisie, Paris, 1980;
Shorrock, From Ally, pp. 49–84, 240–51; René Albrecht-Carrié, Italy at the Paris Peace
Conference, New York, 1938; Brian R. Sullivan, “The strategy of the decisive weight:
Italy, 1882–1922,” in Williamson Murray, MacGregor Knox, and Alvin Bernstein
(eds), The making of strategy, rulers, states, and war, New York, 1994, pp. 343–51; R.A.
Butler, The Art of the Possible, London, 1971, p. 79, Chamberlain quotation.
54 PRO, FO 371/8889 C9802, June 5, 1923; PRO HW 12/40.
55 PRO, FO 371/8889 C13963 August 14, 1923.
56 Antonello Biagini and Alessandro Gionfrida (eds), Lo Stato Maggiore Generale tra le
due guerre. (Verbali delle riunioni presiedute da Badoglio dal 1925 al 1937), Rome, 1997, pp. 93–
292; GP, meetings with Mussolini, Mar. 16, Nov. 18, Dec. 27, 1929; May 30, June 30,
July 14, Nov. 6, Dec. 23, 1930; May 5, July 22, 1931, July 22, Aug. 11, 1932; Jan. 4, Jan.
8, 1933; DDI, series 7, vol. 10, no. 174; vol. 11, nos 147, 155, 169; DDF, le série, tome 2,
nos 140, 154, 182, 201, 236, 242, 365, 419; DGFP, series C, vol. 1, nos 83, 122; Sergio
Pelagalli, “Il generale Pietro Gàzzera al ministero della guerra (1928–1933),” Storia
contemporanea, vol. 20, 1989, pp. 1,038–44; Hoptner, Yugoslavia, pp. 12–14; Douglas Porch,
The French Secret Services. From the Dreyfus Affair to the Gulf War, New York, 1995, p. 141;
Salvatore Loi, Le operazioni delle unità italiane in Jugoslavia (1941–1943), Rome, 1978, pp.
15–17; Aloisi, Journal, pp. 39–40, 45–50, 60–1; Alexander, Republic, pp. 211–15; Fortunato
Minniti, “L’ipotesi più sfavorevole. Una pianificazione operativa italiana tra strategia militare
e politica estera (1927–1933),” Nuova Rivista Storica, vol. 79, 1995, pp. 613–50; Knox, “Il
fascismo,” pp. 302–8, 314– 17; William A. Shirer, The Collapse of the Third Republic. An Inquiry
into the Fall of France in 1940, New York, 1969, pp. 170, 194–6.
57 GP, meetings with Mussolini, Jan. 8, Feb. 4, Feb. 14, March 14, April 21, May 5,
1933; meeting with Vittorio Emanuele III, Jan. 9, 1933; meeting of council of
ministers, May 27, 1933; Emilio Faldella, L’Italia nella seconda guerra mondiale. Revisione
di giudizi, Rocca San Casciano, 1959, pp. 16–17; “Le guerre segrete di Mussolini,”
La Stampa, Jan. 9, 1982; Pelagalli, “Il generale Pietro Gàzzera,” pp. 1,045–9; Knox,
“Il fascismo,” pp. 317–19, and “The Fascist regime,” pp. 160–1.
58 Lo Stato Maggiore Generale tra le due guerre, pp. 293–5, 311–27, 373–82, 420, quotations,
pp. 295, 373; Fortunato Minniti, “‘Il nemico vero.’ Gli obiettivi dei piani di
operazione contro la Gran Bretagna nel contesto etiopico (maggio 1935–maggio
1936),” Storia contemporanea, vol. 26, 1995, pp. 580–1.
59 PRO FO 371/8889 C9802, p. 126.
60 SP,1935, Szabótoarmychief of staff, Oct. 11, 1935, “AudiencewiththeDuce,” Oct.3, 1935.
61 DDI, series 7, vol. 13, no. 69.
62 MacGregor Knox, “Conquest, foreign and domestic, in Fascist Italy and Nazi Germany,”
Journal of Modern History, vol. 56, 1984, pp. 1–57; De Felice, Mussolini il duce, vol. 1, pp.
32–53, 177–322, 544–87, and Mussolini il duce, vol. 2, pp. 265–330; Sullivan, “From little
brother,” pp. 95–6; The Empire at Bay. The Leo Amery Diaries, 1929–1945, London, 1988,
p. 406; Ciano, Diario, pp. 130, 238; DGFP, series D, vol. 7, no. 171.
63 Petersen, Hitler e Mussolini, pp. 383–428; Funke, Sanktionen, pp. 43–5, 59, 68–9.
64 OO, vol. 25, pp. 147–50.
65 OO, vol. 29, pp. 144–65; DGFP, series D, vol. 2, nos 494, 565; Ciano, Diario, pp. 179
85; De Felice, Mussolini il duce,vol. 2, pp. 515–18; Jens Petersen, “L’Italia fascista tra
impegno e neutralismo. I rapporti italo-tedeschi, 1938–1940, Qualestoria, vol. 13, 1990.
66 De Felice, Mussolini il duce,vol. 2, pp. 677–793; Knox, Mussolini Unleashed, pp. 42–
112, and “The Fascist regime,” p. 164; Sullivan, “The Impatient Cat,” pp. 114–35.
11 The Spanish Civil
War and the origins of the
Second World War
Mary Habeck
In his treatment of the Spanish Civil War, as with so much else in his seminal
work on the origins of the Second World War, A. J. P. Taylor managed to
delineate the causes, course, and consequences of a complex international crisis
in a most succinct and skillful way. Using less than fifteen pages, Taylor showed
the profound impact that the nationalist rebellion had on great power politics.
He argued effectively that this war, perhaps more than any other crisis before
the general conflict began, prevented the unification of Great Britain and France
against the dictators, alienated the Soviet Union from the western democracies,
and distracted attention from far greater perils. Scholars since Taylor have added
greatly to our knowledge of the facts surrounding the Civil War, but have not
bettered his interpretation of its significance.
Yet the conflict continues to exert a strong fascination. One of the reasons
for this interest is the enduring perception that Taylor was right: this war
was important both for how it affected Spain and in terms of broader events
taking place concurrently in other parts of Europe. Although the war
stemmed from fissures within Spanish society itself, the struggle between
Nationalists and Republicans soon became caught up in the wider ideological
and great power conflicts of the time. Within days of the uprising both sides
had appealed for aid from those nations that they knew shared their views
of government and society. Weeks later military assistance had begun to
arrive, and would continue to pour in for the next two-and-a-half years.
Volunteers, some there truly of their own free will and others “volunteered”
by their governments, swiftly came to help their ideological brothers in the
struggle. Less than two weeks after the rebellion began, policy makers in
Britain, France, Germany, and the Soviet Union recognized the possibility
that the war might spread to other nations and start a general European
conflict. This realization, and other considerations, led every major European
country but one to seek to limit interference in Spain – in effect to quarantine
the Civil War until it died of its own accord. True believers at either end of
the ideological spectrum refused to accept this decision and worked either
directly, by volunteering, or indirectly, through publicizing and glorifying the
conflict, to aid their chosen comrades.
The Spanish Civil War and the Second World War 205
The result was that some observers of the war shifted their focus from its
Spanish roots, perceiving it to be an extension of the larger struggles occurring
elsewhere at the time, a sort of preview of the much greater conflict that would
shortly play itself out on the battlefields of the world. This was, they argued, the
first clash between despotism and democracy, between Fascism and liberalism, that
later erupted into the more intense flames of the Second World War. It was no
wonder, then, that Claude Bowers, the American ambassador to Spain and a
fervent supporter of the Republicans, would sub-title the reminiscences of his time
spent in that country “Watching the rehearsal for World War II.”1 In a more radical
version of this same opinion, the Spanish Civil War became part of a great “master
plan” designed by Hitler and Mussolini to dominate first Europe and then the
world. No one stated this viewpoint better than José Alvarez del Vayo, the former
president of the Spanish Republic, who argued in 1940 that the Spanish Civil War
was preceded and followed by a series of aggressions – in Manchuria, the
Rhineland, Abyssinia, Austria, Czechoslovakia, Albania – but it was in Spain that
the battle against totalitarian barbarism had been fought with the greatest intensity;
in Spain that there had been a chance to stop the aggressor powers; and in Spain
that these powers had, instead, been helped to victory and given encouragement
for further advances which have since become history.2
As with other supposed examples of a Hitlerian master plan, Taylor decisively
rejected this view of the Civil War. He argued that, while the conflict played an
important role in shaping great power politics and the origins of the Second
World War, it had begun for purely internal reasons. With the exception of a
few scholars, such as Pierre Broué and Emile Témime, who continued to see
the war as little more than another aspect of the struggle between the great powers
of Europe, historians over the last thirty years have generally agreed with Taylor’s
view of the conflict.3 Scholars today emphasize the deep indigenous roots of the
war, discussing the ways that great wealth clashed with dire poverty, conservative
traditions with anarchism and other radical views, even deeply-felt Catholicism
with liberal, (or socialist) anti-clericalism, to produce smoldering resentments that
needed only the slightest provocation to burst into flame. The problems of land
wealth were immense. In 1931, just 2 percent of the people held 65 percent of
the land while about 75 percent owned less than 5 percent.4 This meant that
incomes were also wildly skewed, with 1 percent of the landlords averaging
30,000 pesetas a year and the “bottom” 95 percent averaging about 200 pesetas.
Even worse, the majority of this latter group averaged only twenty-four pesetas
a year.5 A radical anarchism found fertile soil in the discontent created by these
inequities, while Spanish liberalism was more extreme than in other European
nations. On the other side was a conservative backlash that sought to counter
this extremism by emphasizing native Spanish traditions and Nationalism. As
Burnett Bolloten has written, the social tensions of the time meant that “no
foreign intervention was necessary to ignite the tinder of civil strife.”6
Yet intervene foreign countries did, and soon the war had become entangled
in great-power politics. There is thus truth to the assertion that this conflict was
more than just civil war. In fact it could be argued that the conflict in Spain,
206 The Origins of the Second World War Reconsidered
The new administration immediately set about trying to right some of the social
injustices that they perceived within Spanish society. Thus encouraged, peasants,
workers, and anarchists began to demand more extensive changes, holding
strikes, staging demonstrations, seizing land, and attacking churches when they
thought that the government was moving too slowly. The problem was that,
due to the way that voting was weighed in the elections, the Right had actually
received several hundred thousand more votes than had the left-of-center parties.7
Conservatives and other critics of the Left thought that the elections had been
“stolen” or that, at the very least, the Popular Front regime had no mandate to
impose on them its vision of a just society. As the summer began, the unrest
initiated by the Left, and answered by the Right, escalated, until there was an
increasing number of violent exchanges and assassinations. On July 18, 1936
Spanish military forces, led by generals Franco and Sanjurjo, tried to carry out
a coup d’etat against what they perceived as a communist conspiracy to set up
a “Soviet dictatorship.”8 Their failure to immediately topple the Popular Front
government set the stage for the next two-and-a-half years of war.
Almost immediately both the Republicans and the Nationalists turned to the
great powers for military support, the former appealing to France and the Soviet
Union and the latter to Italy and Germany. On July 19, José Giral, the Spanish
Prime Minister, sent a telegram to Léon Blum, the leader of the French Popular
Front government, requesting aircraft to put down the uprising. Blum met with
several members of his cabinet two days later, and all agreed that they should
aid the rightful government of Spain in its attempts to suppress the rebellion.
The primary reason that Blum gave for helping Spain was that a Franco victory
would mean German and Italian bases in the Canaries and Balearics.9 Besides
this strategically sound reason for agreeing to the request, the French had a 1935
commercial treaty with Spain that allowed for the purchase of matériel, and there
was some feeling that the two Popular Front governments ought to support each
other whenever possible. Yet within four days the Blum cabinet reversed its
decision and announced that France would not send the aircraft to its supposedly
ideological brothers in Spain. Shortly afterward the French would propose and
then strongly promote a policy of non-interference in Spanish affairs. By early
August Blum’s cabinet had developed a plan to quarantine the war that included
a non-intervention agreement and the creation of a non-intervention committee
(NIC) to oversee the agreement.
The abrupt nature of this about-face has provoked intense speculation over
Blum’s motives. Many scholars, most prominently Jill Edwards, Antony Beevor,
and John Dreifort, have argued that Blum was persuaded to back away from
supporting the Republicans by British opposition to his plans; France in any case
wished to avoid upsetting its closest partner.10 On July 22, Blum had traveled to
London for meetings with English officials to discuss an unrelated matter. It was
just after he had returned from Britain that the French government announced
the decision not to ship the aircraft to the Republicans, leading many people to
the obvious conclusion that the British had discouraged French involvement in
the developing war.
208 The Origins of the Second World War Reconsidered
As will be outlined below, the British government did indeed have motives
for keeping its ally out of the conflict.11 Nevertheless, it was apparent even at
the time that there was more to the French decision than simply doing whatever
the British suggested. Almost as soon as Blum’s administration had determined
to support the Republicans, its intentions were leaked to the public by a pro-
Franco supporter at the Spanish embassy in Paris.12 This created a huge furor
both in the press and among conservatives in the senate.13 When the cabinet
met to discuss the problem on July 25, the rightist campaign against supporting
the Republicans and opposition from conservative senators, as well as British
attitudes, were high on its list of concerns.14 Members of the cabinet were worried
also about the conflict spreading to other countries, or even turning into another
general European war.15 It was this combination of factors rather than British
pressure alone that dampened the original enthusiasm of Blum and his cabinet
for aiding the Republicans.16 The French now consulted the other great powers
on a non-intervention pact, an initiative that received general approval.17 Here
was a policy that could replace individual interference in the war, please the
British, and avoid the risks of a large-scale conflict. After initially sending some
aircraft – while publicly denying that he was doing so – Blum repudiated any
military support for the Republicans, and worked hard over the next several
months to keep the war from spreading beyond the Spanish borders.18
The change of heart by the Popular Front government showed that France
would not chance a great war without the full support of its people and of Britain.
Although the Blum administration felt a profound ideological closeness to the
Republicans, and knew that there was a strong possibility that the conflict would
result in a regime favorable to enemies directly on France’s borders, neither
strategic nor ideological arguments were strong enough to convince the French.
Throughout the rest of 1936, Blum and Delbos would continue to express their
fears that the war might spill over into a larger conflict. In December Blum would
claim that it was “a miracle” that the war had so far been confined to Spain
alone.19 He would also tell Alvarez del Vayo that “the crux of the matter was in
London,” and that the British had to be convinced to abandon non-intervention
if any help was to come to Spain.20 These were not the words of a man who felt
that France could afford to risk entering a war on its own, regardless of the
incentives to do so.
At about the same time, José Giral wrote the Soviet ambassador in Paris and
asked the Soviet Union to supply the Republicans with arms for their struggle
against the rebels.21 Despite numerous reports from agents in Spain that the war
was going badly for the government, Stalin would not decide to send military
aid to the Republicans until early September.22 The causes for the delay are still
not clear, although there has been much speculation about what he had in mind.23
It is clear, however, that the Soviet Union, like France, had as much to lose as
to gain from intervening in the Civil War.24 The only good reason for helping
Spain was ideological: as the world’s first socialist nation, the Soviet Union was
seen as the natural protector of the Republicans. Stalin was soon faced with
demands from European socialists and communists that he save Spain’s embattled
The Spanish Civil War and the Second World War 209
Nationalists, and Stalin told the few hundred Soviet advisors to “stay out of range
of the artillery fire” when he sent them off to the war.34 After the summer of
1937 the amount of arms and number of men sent to Spain tapered off sharply,
as Stalin and the Politburo decided that it would be more advantageous to have
the war drag on as long as possible in order to keep Hitler tied up in a low-
intensity conflict.35
There was, in fact, little desire in Germany to escalate the conflict. Although
Hitler invested more in Spain in terms of men, matériel and money than did
Stalin, he also wanted to avoid a general war. At the same time, he had more
reason than either France or the Soviet Union to become involved in the Civil
War, and he was willing to take more chances. Hitler, like Stalin, saw himself as
reacting to provocation rather than initiating any interference in Spanish affairs.
As early as July 23, the Germans knew that the French were contemplating
sending matériel to the Republicans. Johannes Welczeck, the German ambassador
to France, informed his superiors that supplying the amount and sorts of
weaponry that he had heard of would cause Franco’s situation to deteriorate
decisively.36 Two Germans in Spain, both Nazis and one an agent of the
Auslandsorganisation (AO), a kind of Nazi Party foreign office, may have had
similar fears. The rebellion had stalled in southern Spain for lack of
reinforcements from Franco’s troops, which were stranded in Africa, and needed
only this sort of aid to the Republic to fail completely. When Franco asked the
two Germans to act as his emissaries, they left for Berlin almost immediately
with a letter for Hitler requesting military aid.37
In their meeting with top Nazi officials, it became clear that Hitler was alone
in supporting some sort of intervention in the war. Werner von Blomberg, the
head of the Wehrmacht, was, as usual, extremely cautious, as was Göring.38 Even
the foreign office was not asked to participate in the decision-making process,
but later statements from the heads of that ministry would show their fears that
the war could expand into a more general conflict.39 Despite this lack of support
from his advisors, Hitler decided to provide the aircraft necessary to airlift
Franco’s forces from Africa to Spain. Over the next two months, though Hitler
would sign the non-intervention agreement, he gradually expanded his
involvement in the war, providing 5,000 German troops and hundreds of millions
of marks in military equipment.40
At the various stages of his involvement in the Civil War, Hitler proved willing
to face the risks attending intervention. However, each time he decided to make a
greater commitment to Franco’s cause, he did so in a way which guaranteed that
the war would not spread beyond Spain’s borders. The supplies he first sent to
the Nationalists were limited in scope and were almost directly proportional to
those that the French were about to send to the Republicans. The initial German
involvement thus did not pose much danger, since it was simply a response to the
French initiative, and could have very great returns, in the form of a regime friendly
to Germany installed on France’s southern border.
Increasing the aid to Franco would have seemed a different matter, since the
chances for provoking retaliation were higher, but once again Hitler had good
The Spanish Civil War and the Second World War 211
peril in view of the fact that she [was] faced on the East by Slav peoples racially
affiliated with the Russians and who might in time fall victim to Communism
and thus close Italy in on two sides.”56 Fascist concerns about communism were
not limited to fears of foreign countries falling to its abhorrent ideology: as a
German observer pointed out, Italian officials also saw that a Republican victory
might encourage the spread of communism to Italy itself.57
Prompted by these apprehensions, the Italians went further than any of the
great powers in their attempts to control the outcome of the Civil War. While
other countries were careful to restrict their aid sufficiently to prevent the war
spreading abroad, Italy took more risks. The reasons for this were twofold: the
perception of Nationalist victory as tied to Italian prestige; and the desire to make
up for the economic costs of the war. Once Italy had invested so many men in
the war, recognized Franco’s regime, and made themselves publicly into the
firmest supporters of the Nationalists, there was little chance that the Fascist
regime could see Franco defeated without enormous domestic and international
repercussions to itself.58 On a more practical note, Italy put so much money into
the war that it could ill-afford to withdraw before making up for its investment.
By October 1937 Mussolini declared that the war in Spain had to be won at all
costs, and that he would persevere until it was successfully concluded. He had
already spent 3 billion lire by this point, and he intended to get his money back,
which he believed to be possible because of Spain’s richness in raw materials.59
Mussolini’s determination to win in Spain served to encourage both the Germans
and the Soviets to invest more there than they might otherwise have done; it
served also to heighten tensions throughout Europe, and so push the world to
the brink of war.
In the end, however, Mussolini’s adventures did not widen the conflict into
the European conflagration that so many feared, primarily because Britain was
willing to compromise almost everything in order to avoid a general war. In
comparison to French agonizing, the British apparently had no difficulty in
deciding to avoid a conflict by not involving themselves in the Civil War. The
main reason for the difference was quite simple: the British saw no reason to
wish success for either side in the conflict. The Conservatives thought that the
Republicans were nothing more than communists in disguise, but they had no
desire to see the Nationalists win either.60 In his diary entry for August 8, 1936,
Harold Nicolson expressed perfectly the attitude of many British officials:
With this slight bias toward the Nationalists, it should come as no surprise that
early on in the crisis Stanley Baldwin, then prime minister, gave Eden explicit
instructions that “on no account, French or other, must he bring us into the fight
on the side of the Russians.”64
Given the absence of a real desire to see either side win, the succeeding British
administrations were determined to support any policy that would keep the war
confined to the Spanish borders. After giving their French allies no
encouragement in July to implement Blum’s plans for intervention, British
officials were pleased to be approached about a non-interference pact the next
month. Over the course of the war, the British would be the most enthusiastic
supporters of the NIC and its agreement, even when it was obvious that three
of the great powers were blatantly violating its terms. Despite its problems, non-
intervention suited all of the British thinking about the Civil War,65 and even
those members of parliament who were most critical of the government’s other
foreign policies found good reason to support this imperfect pact rather than
taking sides and risking a European war.66
Non-intervention also fitted well into Neville Chamberlain’s policy of
appeasement by allowing him to contain the war and ignore most of its
The Spanish Civil War and the Second World War 215
German economic interests at times forced Mussolini to change his plans for
the benefit of Germany.88 Even more important was the rise of Hitler at the
expense of Mussolini, especially after the defeats suffered by Italian forces at
Guadalajara in early 1937. Before the war, Hitler saw Mussolini as a role model
and a great man, whose wishes had to be respected. The more established
Mussolini had greater international prestige than Hitler and acted as the leading
statesman in the cause of radical nationalism. This relationship changed during
the Civil War, as Hitler was gradually able to assert his will to achieve the ends
that he wanted, sometimes at the expense of Italian interests. By the time the
conflict ended, Hitler had completely eclipsed Mussolini in the Axis alliance.
Although there were other reasons for this transformation, the Italian
commitment to the Nationalists’ victory played a vital role in the gradual decline
of Mussolini’s power and influence. Because he had less at stake in Spain Hitler
could further his own interests at the expense of a Mussolini absorbed in his
Spanish troubles. Hitler recognized this, commenting during Mussolini’s state
visit in September 1937: “Italy was sinking deeper in the mire of the civil war
in Spain, leaving Germany free to move forward in the east.”89 As Mussolini
saw his influence fading, he became obsessed with having Franco win as a way
to shore up both his personal prestige and that of Italy, thus fulfilling Hitler’s
wish to prolong the war.90
Mussolini’s new lowered – status – in the Axis in turn affected two significant
milestones on the way to the Second World War: the Anschluss and the Anglo-
Italian accord. There were, of course, other causes for these events, but Italy’s
involvement in the Civil War was an important factor in both. Before the war
Mussolini had expressed his sharp disapproval of Hitler’s plans for Austria. As
Italy became more deeply drawn into the conflict, his concern over a German–
Austrian union, and his ability to prevent such a union, lessened. Schuschnigg
was told early on in the war that Italy could no longer supply Austria with arms
because of her commitments in Spain.91 By 1938 Hitler’s dominance in the Axis,
fed by Mussolini’s distraction over Spain, was so well-established that it was clear
that he could not stop Germany from taking Austria even had he desired to do
so. At almost the same moment as he saw himself becoming the junior partner
in the alliance Mussolini decided to re-open negotiations with Britain, resulting
in the accord signed that April.92 The agreement did little to halt the precipitous
decline of Italy’s influence in the alliance with Germany. By the time the
Nationalists had defeated the Republicans in early 1939, Germany and Italy were
bound more tightly than ever, but Hitler was now firmly in control of the Axis.
If events in Spain added greatly to Hitler’s power and prestige, it also made
a difference in German economic strength. As mentioned above, Germany hoped
to use the war to secure the raw materials necessary to equip the army with
modern armaments. In November 1936 Hitler was already ordering general
Wilhelm von Faupel, the new chargé to the Nationalist government, to concern
himself with improving commercial relations between the two countries. The
opportunity afforded by the war would keep Britain from taking away the market
at a later stage and secure for Germany a permanent supply of iron for Hitler’s
The Spanish Civil War and the Second World War 219
Four-Year Plan.93 Negotiations with Franco to cement these economic ties took
up a great deal of time, but ended with agreements that Hitler prized more than
any friendship treaty with Franco.94 By 1937, Germany had managed to obtain
1.6 million tons of iron ore and 956,000 tons of pyrites without spending any
foreign currency.95 As in other matters, Hitler was not averse to using under-
handed means to secure the iron that Germany desperately needed. When
Nationalist forces ran into difficulties in the summer of 1938 and Franco was
compelled to beg for help, Hitler agreed to send additional troops, but at a price:
the Spanish leader was told that his partners had a “justifiable interest” in a
cooperative attitude on Spain’s part to guarantee Germany’s ore supply.96 Franco
must have acquiesced, for it was shortly afterward that the most important of
the accords over economic ties was signed between the two powers. While
Germany was able to obtain iron from other countries, such as Sweden, Spain
provided a cheaper source of this and other ores essential to German war
industries.
As the example of economic benefits shows, the Spanish Civil War was a
vital stepping-stone on the way to world war, but not in the way that people at
the time understood it. Although Fascists and Nazis would battle communists
and liberals, the war, as Taylor argued, was less a preview of the next conflict
than it was another in a series of crises that exposed the underlying weaknesses
in the international system. As with other events in the Rhineland, Abyssinia,
Austria, and Munich, the war in Spain showed that Britain, France, and the
United States could not risk war in order to deter the dictators; that Hitler and
Mussolini were willing to take calculated risks to achieve their ideological,
strategic, and economic interests; and that the Soviet Union, while trying to avoid
alienating the western democracies, would follow its own course. The war also
confirmed that the alliance between Britain and France, dominated by a Britain
that was firmly committed to appeasement, could not keep the two revisionist
powers from doing almost all that they wished. Meanwhile, Hitler used the
conflict in Spain, as he would other crises, to gain the upper hand in the Axis
and to move toward his ultimate goal of living-space in the east, taking as he
did so the raw materials that he needed for any further steps. Perhaps even more
significant for later events, successes in Spain encouraged Hitler, and Mussolini
to believe that they could employ their armies on the European continent without
fear of retaliation, as long as they did so step-by-step, while the non-intervention
agreement was one of the first that Hitler signed and yet broke with impunity,
laying the groundwork for his later unkept promises.
Notes
1 See also his description of what was at stake in the war in Claude G. Bowers, My
Mission to Spain. Watching the Rehearsal for World War II, New York, 1954, p. 272.
2 J. Alvarez del Vayo, Freedom’s Battle, New York, 1971 [1940], p. xix. See also Ivan
Maisky, Spanish Notebooks, London, 1966, p. 9; Bowers to Roosevelt, August 11, 1937,
Foreign Relations of the United States [henceforth FRUS], 1937, vol. 1:General,
220 The Origins of the Second World War Reconsidered
Washington, 1954, p. 372; and Richard Griffiths, Fellow Travellers of the Right. British
Enthusiasts for Nazi Germany,1933–9, London, 1980, p. 261.
3 Pierre Broué and Emile Témime, The Revolution and the Civil War in Spain, Cambridge,
MA, 1970.
4 Dante A. Puzzo, Spain and the Great Powers, 1936–1941, Freeport, CT, 1962, p. 5.
5 Alvarez del Vayo, Freedom’s Battle, p. 142.
6 Burnett Bolloten, The Spanish Civil War. Revolution and Counterrevolution, Chapel Hill,
NC, 1991, p. 21. The best treatment of the numerous underlying causes for conflict
in Spain during the early part of this century can be found in Gerald Brenan, The
Spanish Labyrinth. An Account of the Social and Political Background of the Civil War,
Cambridge, 1967.
7 See Anthony Rhodes, The Vatican in the Age of the Dictators, 1922–1945, London, 1973,
p. 122, for this little-noted fact.
8 See Franco’s declaration to the Germans that the uprising was necessary “to
anticipate a Soviet dictatorship, which was already prepared.” Wegener [Tetuán
consulate] to foreign ministry, July 24, 1936, Documents on German Foreign Policy, 1918–
1945 [henceforth DGFP], series D, 1937–45, vol. 3, p. 8.
9 Michael Alpert, A New International History of the Spanish Civil War, London, 1994,
p. 14.
10 Among those scholars who have emphasized the role that the British had in
discouraging the French from supporting the Republicans are Jill Edwards, The British
Government and the Spanish Civil War, 1936–1939, London, 1979, pp. 25–30; Antony
Beevor, The Spanish Civil War, London, 1982, pp. 109–10; John E. Dreifort, Yvon
Delbos at the Quai D’Orsay. French Foreign Policy during the Popular Front, 1936–1938,
Lawrence, KS, 1973, pp. 50, 54; Robert H. Whealey, Hitler and Spain. The Nazi Role
in the Spanish Civil War, 1936–1939, Lexington, KY, 1989, p. 15; William E. Watters,
An International Affair. Non-Intervention in the Spanish Civil War, 1936–1939, New York,
1971, p. 38; Broué and Témime, Revolution and the Civil War, pp. 328–9; Puzzo, Spain
and the Great Powers, 82–8; Pierre Cot, Triumph of Treason, Chicago, IL, 1944, pp. 94,
344–6.
11 And at least one British official, the ambassador to France, warned the French against
helping out the Republicans, although he said that he was speaking “personally”
and not officially. See Dreifort, Delbos, pp. 46–7.
12 Cot, Triumph of Treason, p. 338.
13 Cot, Triumph of Treason, pp. 344–6; Alpert, New International History, pp. 14, 21–2;
Dreifort, Delbos, p. 35.
14 Dreifort, Delbos, p. 40.
15 Anthony Adamthwaite, France and the Coming of the Second World War, 1936–1939,
London, 1977, p. 42. Throughout the rest of 1936 Blum and Delbos would continue
to talk about their fears that the war would widen into a greater conflict, showing
the strength that this held in their decision to support non-intervention. Telegram
from Bullitt to acting-secretary of state, November 28, 1936, FRUS, 1936, vol. 2:
Europe, pp. 578–80; Welczeck to the foreign ministry, December 15, 1936, DGFP,
series D, vol. 3, 167–8.
16 Most recent scholars have recognized that it was all of these ingredients working
together that militated against French involvement in the war. Even those who
believe that it was primarily British warnings that convinced the French to back
down are careful to list the other reasons for Blum’s change of mind. See Puzzo,
Spain and the Great Powers, pp. 82–8; Broué and Témime, Revolution and the Civil War,
pp. 328–9; Dreifort, Delbos, p. 53.
17 Anthony Eden, Facing the Dictators. The Memoirs of Anthony Eden, Earl of Avon, Boston,
MA, 1962, p. 455. Delbos was among those most enthusiastic about the policy of
non-intervention. Cot, Triumph of Treason, p. 339; Dreifort, Delbos, pp. 39–40, 43.
18 Cot, Triumph of Treason, pp. 339–42; Alpert, New International History, pp. 22–3.
The Spanish Civil War and the Second World War 221
19 Welczeck to the foreign ministry, December 15, 1936, DGFP, series D, vol. 3, pp.
167–8.
20 Julio Alvarez del Vayo, Give Me Combat, Boston, MA, 1973, pp. 159–60.
21 Letter from Giral to Soviet ambassador in Paris, July 25, 1936. Russian State Military
Archive [hereafter RGVA], f. 33987, op. 3, d. 991, pp. 56–9. The timing of the letter
is important because it refutes those historians who have argued that the
abandonment of the Republic by the western democracies threw it into the arms of
the Soviet Union and thus gave substance to the association of the Republican cause
with communism. See e.g. Paul Preston, “The creation of the Popular Front in
Spain,” in Helen Graham and Paul Preston (eds), The Popular Front in Europe, London,
1987, p. 84.
22 See e.g. “Information on military situation in Spain,” September 8, 1936, and
September 19, 1936, RGVA, f. 33987, op. 3, d. 845, vol. 2, pp. 20–3, 46–8.
23 See Bolloten, Spanish Civil War, p. 110.
24 David T. Cattell, Communism and the Spanish Civil War, Berkeley, CA, 1955, pp.
162–3.
25 Broué and Témime, Revolution and the Civil War, pp. 367–9.
26 For Baldwin’s attitude toward the Republicans, see Edwards, British Government,
p. 18.
27 Carr believes that the Soviets agreed to non-intervention because they wanted to
keep in step with Britain and France and because they lacked the capacity to send
military supplies to Spain on any scale matching those of Germany and Italy. E.
H. Carr, The Comintern and the Spanish Civil War, New York, 1984, pp. 17–18.
28 List of the cost of all ABT material sent to Spain from the ABTU to Langov, dated
November 1, 1936, RGVA, f. 33987, op. 3, d. 832, pp. 272–5.
29 Ivan Maisky, Spanish Notebooks, London, 1966, p. 49.
30 Memorandum from Manuil’skii to Stalin, November 1936, RGVA, f. 33987, op. 3,
d. 832, p. 309.
31 W. G. Krivitsky, In Stalin’s Secret Service, New York, 1939, p. 80.
32 See R. H. Haigh, D. S. Morris, and A. R. Peters, Soviet Foreign Policy, the League of
Nations and Europe, 1917–1939, Totowa, NJ, 1986, pp. 62–3.
33 Davies (ambassador in the Soviet Union) to Hull, March 26, 1937, FRUS, 1937,
pp. 263–5.
34 Krivitsky, Secret Service, p. 77.
35 Alexander Orlov, The Secret History of Stalin’s Crimes, New York, 1953, p. 238.
36 Telegram from Welczeck to the foreign ministry, July 23, 1936, DGFP, series D,
vol. 3, p. 4.
37 Manfred Merkes, Die deutsche Politik gegenüber dem spanischen Bürgerkrieg, 1936–1939,
Bonn, 1961, p. 19; Alpert, New International History, p. 24; William Carr, Arms, Autarky,
and Aggression. A Study in German Foreign Policy, 1933–1939, New York, 1972, p. 69.
38 For the attitudes expressed at this meeting see André Brissaud, Canaris, London,
1973, p. 37; Alpert, New International History, pp. 27, 30.
39 See Memorandum by Neurath, August 4, 1936, DGFP, series D, vol. 3, p. 29;
Memorandum by Weizsacker, director of the political department, July 4, 1937,
DGFP, series D, vol. 3, p. 391.
40 Hans-Henning Abendroth, Hitler in der spanischen Arena, Paderborn, 1973, pp. 63–4;
Whealey, Hitler and Spain, p. 8; Raymond L. Proctor, Hitler’s Luftwaffe in the Spanish
Civil War, Westport, CT, 1983, p. 253.
41 See e.g. Abendroth, spanischen Arena, pp. 30, 34–8; Merkes, deutsche Politik, p. 25;
Carr, Arms, p. 69; Gerhard L. Weinberg, The Foreign Policy of Hitler’s Germany. Diplomatic
Revolution in Europe, 1933–36, Chicago, IL, 288–9.
42 Glenn T. Harper, German Economic Policy in Spain during the Spanish Civil War, 1936–
1939, The Hague, 1967, pp. 16–17.
222 The Origins of the Second World War Reconsidered
43 See e.g. Memorandum from the foreign ministry to the war ministry, July 24, 1936,
DGFP, series D, vol. 3, p. 7; Memorandum by Dieckhoff, director of the political
department, July 25, 1936, DGFP, series D, vol. 3, pp. 10–11 and Ernst von
Weizsäcker, Erinnerungen, München, 1950, p. 129. For secondary source discussions
of this, see Harper, German Economic Policy, pp. 13–14, 40; Merkes, deutsche Politik,
pp. 20–1.
44 Almost every student of Germany’s part in the war has mentioned the German
desire for raw materials as a primary reason for further involvement in the war.
See e.g. Broué and Témime, Revolution and the Civil War, pp. 359–60, 364; Carr, Arms,
p. 69; Harper, German Economic Policy, pp. 16–17, 40–1; Weinberg, Diplomatic Revolution
in Europe, pp. 288–9.
45 Quoted in Harper, German Economic Policy, p. 65.
46 For one statement of this argument see Harper, German Economic Policy, pp. 16–17.
47 Carr, Arms, pp. 75–6; Memorandum [hereafter The Hossbach Memorandum],
November 10, 1937, DGFP, series D, vol. 1, pp. 36–7. For a fuller discussion of the
effects that Spain had on the relationship between Italy and Germany see below.
48 One historian who doubts this interpretation is Abendroth; see spanischen Arena, pp.
30, 34–8.
49 Neurath to Blomberg, December 15, 1936, DGFP, series D, vol. 3, p. 168; Telegram
from Hassell (ambassador in Italy) to foreign ministry, December 17, 1936, DGFP,
series D, vol. 3, p. 169; Memorandum by Neurath, January 13, 1937, DGFP, series
D, vol. 3, p. 222; The Hossbach Memorandum, p. 37. See also Weinberg, Diplomatic
Revolution in Europe, pp. 297–8.
50 See Alpert, New International History, pp. 35–8 for a complete discussion of Mussolini’s
decision.
51 Whealey, Hitler and Spain, p. 12.
52 Ciano’s Diary, 1937–1938, London, 1952, pp. 50–1.
53 Ciano’s Diary, 1937–1938, p. 32. See also p. 26 for Ciano’s attitude toward this issue.
54 For an excellent discussion of this factor in determining Italian policy, see John F.
Coverdale, Italian Intervention in the Spanish Civil War, Princeton, NJ, 1975, pp. 75–6,
127ff.
55 Merkes, deutsche Politik, p. 31.
56 Memorandum by Murray (chief of the division of near-eastern affairs), November
2, 1936, FRUS, 1936, p. 549.
57 Plessen (chargé d’affaires in Rome) to foreign ministry, August 14, 1936, DGFP,
series D, vol. 3, pp. 38–9.
58 Gerhard L. Weinberg, The Foreign Policy of Hitler’s Germany. Starting World War II, 1937–
1939, Chicago IL, 1980, pp. 144, 149
59 Memorandum from von Bülow-Schwantz (protocol department, German foreign
ministry) to Neurath, October 2, 1937, DGFP, series D, vol. 1, pp. 4–5.
60 Edwards, British Government, pp. 35–7.
61 Stanley Olson (ed.), Harold Nicolson. Diaries and Letters, 1930–1964, London, 1980,
p. 101.
62 Telegram from Bingham to Hull, July 6, 1937, FRUS, 1937, pp. 353–4.
63 Letter from Johnson (chargé in London) to Hull, August 23, 1937, FRUS, 1937,
pp. 375–6.
64 Quoted in Edwards, British Government, p. 18.
65 Edwards gives reasons other than the fit non-intervention with British policies,
including the fact that it would obey a chief of staff dictum to remain on good terms
with whichever side won, prevent France from going “red,” maintain a veneer of
impartiality to quieten the British Left, and impede the growing alienation of Italy
and Germany from the rest of Europe. See Edwards, British Government, p. 61.
66 For a sophisticated discussion of how Conservative critics in particular were
convinced to support Chamberlain’s policies, see Neville Thompson, The Anti-
The Spanish Civil War and the Second World War 223
Appeasers. Conservative Opposition to Appeasement in the 1930s, Oxford, 1971, pp. 116–
19. For Eden’s attitude see Eden, Facing the Dictators, p. 451.
67 Eden, Facing the Dictators, p. 454; Larry William Fuchser, Neville Chamberlain and
Appeasement. A Study in the Politics of History, New York, 1982, pp. 61–2.
68 Keith Middlemas, The Strategy of Appeasement. The British Government and Germany,
1937– 39, Chicago, IL, 1972, p. 42; Ian Colvin, The Chamberlain Cabinet, London,
1971, p. 45.
69 Eden and other cabinet members thought this a good first step, but wanted to go
even further, making the war inseparable from events in Czechoslovakia. See e.g.
Middlemas, Strategy of Appeasement, pp. 183–4.
70 Thompson, Anti-Appeasers, pp. 41, 143.
71 Hull to Bingham (ambassador in London), March 27, 1937, FRUS, 1937, p. 268.
72 Bingham to Hull, March 31, 1937, FRUS, 1937, pp. 270–2.
73 See e.g. Memorandum by Welles (undersecretary of state), October 18, 1937, FRUS,
1937, pp. 425–7.
74 Telegram, Bullitt to acting-secretary of state, November 28, 1936, FRUS, 1936, pp.
578–80.
75 For complete discussions of US policy during the war, see Richard P. Traina, American
Diplomacy and the Spanish Civil War, Bloomington, IN, 1968; Alpert, New International
History, pp. 109–11; C. A. MacDonald, The United States, Britain and Appeasement,
1936–1939, London, 1981, p. 1. Little argues that the US and Britain were
motivated by fears of a communist state in Spain to adopt a “malevolent neutrality”
in the war. See Douglas Little, Malevolent Neutrality. The United States, Great Britain,
and the Origins of the Spanish Civil War, Ithaca, NY, 1985.
76 Cordell Hull, The Memoirs of Cordell Hull, vol 1, New York, 1948, p. 477.
77 Letter from Bowers to Roosevelt, August 11, 1937, FRUS, 1937, p. 372; Bowers,
Mission to Spain; Harold L. Ickes, The Secret Diary of Harold L. Ickes, vol 2: The Inside
Struggle, 1936–1939, New York, 1954, pp. 93, 424–5.
78 At which point he seems to have regretted his choice of policy. See Ickes, Secret Diary,
vol. 2, p. 569.
79 John Harvey (ed.), The Diplomatic Diaries of Oliver Harvey, 1937–1940, London, 1970,
p. 118.
80 Telegram from Bullitt to Hull, May 9, 1938, FRUS, 1938, p. 192.
81 See Oliver Harvey’s note in his diaries: “The French are getting increasingly restive
as a result of having closed their frontier as a result of our urging,” in Harvey (ed.),
Diplomatic Diaries, p. 157.
82 Hugh Thomas, The Spanish Civil War, New York, 1986, pp. 740–1 has this
interpretation of the British decision to support the French.
83 Dreifort, Delbos, pp. 76–7.
84 Report from Hassell to the foreign ministry, December 18, 1936, DGFP, series D,
vol. 3, pp. 170–2.
85 Coverdale, Italian Intervention, pp. 110–13.
86 See e.g. Telegram from Neurath to the embassy in Italy, January 12, 1937, DGFP,
series D, vol. 3, p. 220; Telegram from Hassell to the foreign ministry, January 13,
1937, DGFP, series D, vol. 3, p. 221.
87 Whealey, Hitler and Spain, pp. 12–13.
88 Weinberg, Starting World War II, pp. 144, 149; Memorandum from von
BülowSchwantz to Neurath, October 2, 1937, DGFP, series D, vol. 1, pp. 4–5; The
Hossbach Memorandum, p. 37.
89 Carr, Arms, pp. 75–6.
90 C. J. Lowe and F. Marzari, Italian Foreign Policy, 1870–1940, London, 1975, p. 298.
91 Lowe and Marzari, Italian Foreign Policy, p. 298.
92 Coverdale, Italian Intervention, pp. 352–3.
224 The Origins of the Second World War Reconsidered
93 Minute by Sabath (economic policy department), November 27, 1936, DGFP, series
D, vol. 3, p. 142.
94 Von Ribbentrop actually drew up a “treaty of friendship” between Spain and
Germany, but Hitler thought that a commercial treaty was better than something
like this, which had “little value.” See Memorandum from Ribbentrop to Hitler
(1938), DGFP, series D, vol. 3, pp. 631–4; Memorandum from Spitzy (foreign
minister’s secretariat) to Ribbentrop, DGFP, series D, vol. 3, pp. 634–5.
95 Carr, Arms, p. 69.
96 Stohrer (ambassador in Madrid) to foreign ministry, July 6, 1938, DGFP, series D,
vol. 3, p. 716.
12 The phantom crisis
Danzig, 1939
Sean Greenwood
discredited in Paris and London, and Beck himself suspected of collusion with
Hitler. He was more than unpopular in Moscow. And now, confronted with Nazi
demands, Poland was alone.”4
Nevertheless, if Danzig, which was regarded in Warsaw as the barometer of
German–Polish relations, seemed set to stormy at the beginning of 1939, the
Poles were not unduly alarmed. Beck and his two close associates Józef Lipski,
the Polish ambassador in Berlin, and Jan Szembek, the Polish deputy minister
for foreign affairs, were convinced that the demand for the return of Danzig
was merely a temporary phase in German foreign policy and that Hitler’s long-
term aspirations in the east meant that Poland would be a necessary ally to
Germany.5 Though Beck decided to stand farm in the face of what was judged
to be German bluster, it is not quite the case, as Taylor would have it, that he
kept the issue in the way or that the Poles “were determined not to yield an
inch” (p. 242). Over Polish rights in Danzig itself, Warsaw was adamant.
However, Beck was not averse to some form of compromise over
communications across the Corridor which might appease Hitler without turning
Poland into a German satellite. His confidence that Hitler’s demands were a bluff,
and that Polish military strength was sufficient to deal with any aggression which
might arise, survived until war broke out, supporting Taylor’s image of Poland
as a state with a “false pride as a Great Power” (p. 241).6
The vital issue as the crisis enfolded, however, was not Danzig. Nor was it
simply the future of Poland. It was the scope of Hitler’s ambitions. This, of
course, is the central problem with Taylor’s account. Why the Hitler of Taylor’s
construction – who supposedly spent his time after his “dazzling success” at
Munich “drawing dream-plans for the rebuilding of Linz, the Austrian town
where he went to school” (p. 238), and who had no desire for “theatrical displays
of glory” (p. 239) but sought a steady extension of German power at which he
would arrive through his usual process of letting others do his work for him
should now have bothered to put his sketches aside in order to raise the future
of Danzig presents a problem which Taylor attempts to resolve by resorting to
equivocation. He presents the issue as having materialized spontaneously: it was
principally a matter of selling a partnership with Poland to a German public
which bitterly resented the loss of the Polish Corridor in 1919 but would be
appeased by the restoration of Danzig to Germany. The objective of this
German–Polish association would be to allow the two to “act together in the
Ukraine” (p. 242). But Taylor doubts that Hitler was interested in territorial gains
in the east. If, on the other hand, he “really aspired to reach the Ukraine he
must go through Poland” (p. 240).
It may be, though Taylor does not mention this, that Hitler was indeed
stirred into action by the Poles themselves, when they raised the Danzig
question in the midst of the Czechoslovak crisis in the hope of fixing the status
quo.7 This would provide some support for Taylor’s general thesis that Hitler’s
foreign policy was fueled by the initiatives of others, and the Polish approach
may have drawn attention to Poland’s vulnerability. Hitler could hardly have
failed to recognize that Polish hostility toward Prague during and immediately
228 The Origins of the Second World War Reconsidered
after the Czechoslovak crisis had, by tilting Warsaw towards Berlin, further
compromised their ideal of preserving a balance between Germany and Russia.
Also, as von Ribbentrop noted, Poland’s position had been weakened by “the
evident chill in Poland’s relations with the western powers.”8 Perhaps there was
sufficient here to stimulate the imagination of a musing dictator. Maybe it was
not outsiders at all but members of Hitler’s own entourage who manipulated
events so as to prompt him to bring Poland further into the German embrace.9
But the Hitler of The Origins of the Second World War is, as is now generally
accepted, Taylor’s particular fabrication, and the irrational, tortured, vindictive
individual, consumed by a neuralgic impulse to act before his own early death
– the Hitler who so many sources assure us existed – is absent. Yet, despite all
his efforts to banish him, Hitler the man of ambition and purpose has survived
the daydreaming Chaplinesque creation of Taylor. Weinberg, for example, is
convinced that after Munich Hitler continued to pursue his long-term goal of
securing living-space in the east. More specifically, he “wanted to be able to
concentrate all his forces in the west without having to worry about trouble
with Poland on his eastern border, and he therefore tried by persuasion and
pressure to convert that country into a satellite.”10 Donald Watt provides us
with a similar picture: a Hitler who, even before the crisis over Czechoslovakia
had reached its climax, had begun to prepare for war against Britain and
needed Poland to be negotiated onto his side as one of the preliminaries for
this.11 His is an angry, brooding, subtly semi-Taylorian Hitler, one lacking the
psychological capacity to follow through long-held goals, letting events take
their course, and trusting to instinct. His decision, around 1937, to take a more
pro-active course merely complicated matters and destroyed his earlier sureness
of touch. This is not a passive, not even a barely patient, dictator. After Munich,
for perfectly sound reasons, Poland was expected to be compliant. Tension
began to boil over when it was discovered in Berlin that, against the odds, the
Poles refused to be submissive.
By the time von Ribbentrop raised Danzig with the Poles for the fourth time
on March 21, post-Munich Czecho-Slovakia had already disappeared and the
process of coercing Lithuania, which would lead to the annexation of Memel
on the March 23, had started. This deteriorating climate had its impact on
German–Polish relations, but had much more momentous repercussions in
London. If, as Taylor surmises, the German pounce on Memel – a question
which otherwise, apparently, “exploded of itself” (p. 258) – held an implicit
warning to the Poles to be co-operative, it had the opposite effect, making opinion
in Warsaw “indignant, alarmed and at the same time defiant.”12 On March 28
Beck informed the German ambassador to Poland that any attempt by Germany
to alter unilaterally the status of Danzig would be viewed as an act of war. “Until
this moment,” Taylor says, “everything had gone on in secret, with no public
hint of German–Polish estrangement. Now it blazed into the open” (p. 258).
Here certainly was the kindling for a crisis, but as both Berlin and Warsaw
continued to be publicly evasive over the significance of their discussions on
Danzig the issue continued to flicker rather than to flare. Ignition came because
The Phantom Crisis: Danzig 1939 229
cope with it, that the British were embarking upon an essentially haphazard
attempt to regain the initiative lost to them over the last half dozen years is much
more open to question.
British policy now acquired a harder edge, and was rather more systematic
and purposeful than Taylor allows. The two weeks between Tilea’s démarche and
the guarantee to Poland was a period of intense activity which saw the British
consciously reviving and developing the concept of a barrier to Hitler that they
had begun to stumble towards back in January. Taylor’s abbreviated account
provides only a half-glimpse at what was taking place. In Origins the start of this
process was a proposal for a declaration of support for any state threatened by
Germany in which the French, Soviet, and Polish governments were also asked
to participate. The initial draft was Chamberlain’s own, though it had to be
firmed-up by the French to include action as well as consultation. Its intention
was, as Taylor says, “geared to the supposed threat to Rumania” (p. 255). But
the British, Taylor implies, were diverted, even gulled, from this purpose by
Beck’s refusal to associate Poland with Russia and his insistence upon a simple
Anglo-Polish declaration.
In fact, the Four-Power Declaration was not Chamberlain’s invention but a
proposal made on the morning of March 18 by the chiefs of staff.
Understandably, given the Tilea episode and their ignorance of the details of
German– Polish discussions on Danzig, their focus of concern was Rumania. It
continued for a while to be so. Ten days after the Tilea scare Halifax still thought
Rumania “may be the State primarily menaced by Germany’s plans for Eastern
expansion.”20 Rumania was crucial to the British because of her oil and grain:
Britain’s ability to wage economic warfare against Germany would be seriously
diminished if Hitler controlled these.21 The chiefs of staff advised that an
immediate diplomatic initiative to forge an alliance with Russia and Poland might
deter Germany from action against Rumania. An earlier and wider request for
reactions to the Rumanian scare, which Taylor ignores, had already been made
in Paris, Warsaw, and Moscow as well as in Bucharest, Athens, Ankara, and
Belgrade. The inspiration for this was probably Tilea himself.22 Chamberlain
has been criticized for abandoning this larger scheme for the more modest
proposal of the chiefs of staff.23 But this was forced upon him because the
responses from the eastern-European states approached were an unsatisfactory
mix of suspicions of British firmness of purpose and resistance to an alignment
with the Soviet Union, such that there was little substantial on which to work.
Chamberlain dismissed the foreign office as “pretty barren of suggestions” and,
encouraged by the chiefs of staff, came up with his own solution.24 This moved
British policy forward from a vague intention of simply supporting what others
might do in the event of further German aggression toward the more precise
construction of an east-European bloc to hem Hitler in – though the intention
remained that this should be achieved by warnings rather than by specific new
British obligations.
The British needed to associate Poland with the Four-Power Declaration not
out of their worries for Polish security but because of their fears for Rumania.
232 The Origins of the Second World War Reconsidered
on the most effective partner for an eastern front, their responses also favored
the Poles. They advised that Russia would be a more useful ally against Germany
than would Poland – but only marginally so. Should there be a simultaneous
strike from Germany, Italy, and Japan, even an Anglo-Russian alliance, they
believed, would be of limited value. Added to this, there was evidence to suggest
that a British arrangement with the Russians would alienate a range of potential
support and might even tip the balance in pushing Italy and Japan toward closer
partnership with Germany. If a choice had to be made, then choosing Poland
seemed less complicated. What was at stake, in any case, was not a decision
between peace and war but a policy of restraint through the construction of an
eastern front. At the same time, account had to be taken of the possibility that
deterrence might not work, and, if Britain had to fight, the estimation was that
she would do better with Poland than the Soviet Union on her side.
As Taylor puts it, “the British hardly reflected that, by choosing Poland, they
might lose Russia” (p. 256). Whilst bowing to the reservations of both the
Rumanians and the Poles to any association with Russia, the British worked on
the assumption that once Poland was embraced by “the organization which we
were trying to build up for the defence of Roumania” the Poles would then accept
“indirect Soviet assistance.”29 Here is Taylor’s image of Russian assistance being
“turned on and off at will like a tap” (p. 278). The Russian complication was,
however, instrumental in maneuvering the British away from the Four-Power
Declaration and toward a more restricted and committed agreement with the
Poles. The spur continued to be to protect Rumania. On March 22 Halifax
pondered that, “in order to persuade Poland to commit herself to support
Roumania, Great Britain and France would have to give Poland a private
undertaking that, if Poland came in, they would both come in also.”30 An
approach from Beck on March 24 accelerated the idea of a possible bilateral
understanding. But developments did not go entirely Beck’s way. Confident of
being able to deal with Germany alone and “terrified” of anything more overt
for fear of antagonizing Hitler, he pressed for a secret arrangement with Britain.31
In this way, between March 17 and 24 the number of states to be associated
with Britain and France in an eastern bloc was on the point of being reduced
from three to one – yet the intention remained to construct the larger edifice.
The outline of the British guarantee to Poland is thus already discernible –
though, as it turned out, Beck was to get more publicity than he had hoped for.
The process was to be completed when a new war scare switched the spotlight
from Rumania to Poland.
By the end of March the rumor factory in Whitehall had resumed full
production with warnings of threatening moves by Germany toward France and
Belgium, and of possible air raids on London. Poland, viewed initially as not
much more than an element in the construction of an eastern security bloc, at
this point entered the eye of the storm with the warning from Ian Colvin, the
Berlin correspondent of the News Chronicle, that “an attack on the Polish Republic”
was imminent.32 Colvin’s warnings of a general threat to Poland were presented
to Chamberlain shortly after the cabinet had agreed to opt for reciprocal
234 The Origins of the Second World War Reconsidered
guarantees with Poland and Rumania. Once again, this information was false,
having been planted by the German opposition to Hitler. Whether, as Taylor
argues, “after the occupation of Prague and the supposed threat to Rumania,
the British were ready to believe anything” (p. 260), or whether, as some have
more recently suggested, Halifax simply manipulated the incident to press for
action, everything now changed. Reciprocity was ignored, Rumania was
temporarily forgotten and, on March 31, a unilateral guarantee was offered to
Poland.
Few would dispute Taylor’s view that the guarantee was “a revolutionary
event in international affairs” (p. 264). Unintended by and unknown to the
British, it also marked the beginning of the process which would lead to war.
Incensed at the “ingratitude of the Poles” and at what he regarded as a process
of encirclement by the British, Hitler reversed his policy by withdrawing his
recent proposals to Poland, revoking the non-aggression pact of 1934 and also
the Anglo-German naval agreement.33 Prior to this, on April 3 he ordered
preparations for a possible attack on Poland to mature by September 1. That
Britain now “plunged into alliance with a country far in Eastern Europe, and
one which, until almost the day before, had been not worth the bones of a British
grenadier” (p. 264), is Taylorian hyperbole. The undoubted scramble
surrounding the issuing of the guarantee – not yet an alliance – is deceptive. As
we have seen, British thinking was already shifting towards the idea of some
direct assurance to Poland, and “Colvin’s warning precipitated the public
announcement of a policy already being formulated.”34 To Taylor, the British
were now “committed to resistance” (p. 263), though he also has them
simultaneously “faced with the choice between resistance and conciliation.”
Anyhow, they “preferred the second course” (p. 264), and appeasement was to
be continued. But in their haste to provide “some vague and generous gesture
to moderate the speed of Germany’s advance,” Taylor argues, the British found
that they had given all the aces to Beck. “The assurance was unconditional. The
Poles alone were to judge whether it should be called upon. The British could
no longer press for concessions over Danzig; equally they could no longer urge
Poland to cooperate with Soviet Russia” (p. 260).
Taylor has significant support here. Geoffrey Roberts accepts that the
guarantee constricted British and French freedom of maneuver, “tying them into
an inherently dangerous situation.”35 More specific is Watt’s assertion that “[i]f
the Poles took up arms then Britain fought too. The decision, war or peace,
had been voluntarily surrendered by Chamberlain and his cabinet into the
nervous hands of Colonel Beck and his junta comrades-in-arms.”36 Anna
Cienciala makes a rather different point, arguing that the guarantee served the
double purpose of persuading Hitler to abandon aggression while allowing Britain
and France to press the Poles into giving Hitler what he apparently wanted. To
her, “A. J. P. Taylor was closest to the mark in viewing the guarantee as a
continuation of appeasement because it envisaged further territorial revision in
Eastern Europe.”37 But Chamberlain did not believe he had overplayed his hand
with the Poles and, whereas adjustments to preserve the peace remained on the
The Phantom Crisis: Danzig 1939 235
cards – over Danzig, for instance – the guarantee had drawn a line beyond which
Hitler must not go. With Strang’s view that the guarantee meant that “Britain
had committed itself to fight for Polish independence, and had allowed the
decision on war or peace largely to pass out of its hands” we get closer to the
nub of the issue.38 The key word here is independence. The guarantee was not
envisaged as a “blank cheque” to the Poles. It was certainly not regarded by the
British as a provocation, a “deliberate challenge,” and a stimulant to Polish
intransigence which increased the probability of war.39 Beck admitted at the time
that the attitude of the Poles was not altered by the guarantee.40 They had already
made up their minds to resist Hitler if necessary. An arrangement over Danzig
was still a possibility so far as the British were concerned. There was no written
commitment on their part to underwrite the existing situation. Nor did
Chamberlain intend one. As Taylor admits, Danzig did not seem to be an urgent
issue at the end of March. Poland’s integrity did. Fears were for a weekend fall
on Poland, not on the Free City. Press comment within hours of the declaration,
probably government-inspired, sought to remove any implication of a
commitment to Danzig.41 Chamberlain still hoped, if it remained a possiblity, to
remove grievances which might be the cause of war, and these might well include
Danzig. As he privately stated, “what we are concerned with is not the boundaries
of states but attacks upon their independence. And it is we who will judge
whether this independence is threatened or not.”42 Halifax also made it clear to
Warsaw that continuance of the guarantee was dependent upon the creation of
reciprocal guarantees between Britain, France, Poland and Rumania.43 Only in
this particular sense does Taylor’s view that “the British were no sooner
committed than they realized the flaws in what they had done” (p. 261) and
were determined to remedy them have pertinence.
The principal purpose of the guarantee was to defer war by restraining Hitler.
It was premature to argue that the guarantee marked “a retreat from the initial
comprehensive British proposal towards central and southeastern Europe which
had been considered in the wake of German action in Prague” or that with it
“the possibility of an eastern front against Germany was lost.”44 This is to echo
Taylor’s opinion that once negotiations with Beck were under way the “peace
front” and collective security vanished from the scene.” This may have been the
outcome. It was not the intention. The objective of the Four-Power Declaration
to provide what Halifax called a “rallying point” to other Balkan states survived.45
When apprised of the need for a guarantee to Poland, the cabinet and the House
of Commons were told of its “interim” nature to meet the possibility of a sudden
move by Germany. It was envisaged as part of an interconnecting east-European
web. As Halifax pointed out to the cabinet, “if Poland was over-run, there would
be no hope of saving Roumania.”46 And if, as Taylor says, it was the case that
the British could now no longer urge the Poles to include the Soviet Union in
this future arrangement, their inclination to do so, as we have seen, had always
been wafer-thin.
Taylor, along with others, has pointed out that there was no practical way
for Britain to fulfill the guarantee.47 But, as he also indicates, “the British
236 The Origins of the Second World War Reconsidered
government were striving to preserve the peace of Europe, not to win a war”
(p. 278). Arrangements were to proceed to build a wider security front to contain
Germany. Although these were focused on eastern Europe, the possibility of an
attack in the west could not be ruled out. Should the dam in the east fail to
deter, the calculation seems to have been, it would nevertheless provide an
additional capability for Britain to conduct a war with some expectation of
success. The fact that the guarantee was not followed up with military or financial
support to Poland – that, as Taylor remarks, “no credit had passed by the time
war broke out; no British bomb or rifle went to Poland” (p. 271) – is not an
indication of an empty policy of bluff or panic. Rather, it was the strategically
rational, if cynical, extension of a perceived need to harbor all resources for a
long war. It might not have been possible to assist Poland were an attack to be
made. Yet the very process of her defeat would stretch German resources, and
victory over Poland would require significant manpower to defend the new
German–Soviet frontier.48 This point was made emphatically in cabinet by Lord
Chatfield, the minister for co-ordination of defence, on the day before the Polish
guarantee was issued.49 This may cast some doubt on Taylor’s view that British
policy “was determined by morality, not by strategical calculations” (p. 278).
The crucial failure of the British lay less in their neglect to provide material
support for the Poles than in the collapse of their attempts to build a solid “eastern
front.” By mid-May guarantees had been extended to Rumania, Greece and
Turkey – a somewhat disparate and erratically constructed “rallying point,”
admittedly, but close to the original design of mid-March. The real prize, the
Soviet Union, eluded them. Soviet intentions during the summer of 1939 are
examined elsewhere in this volume. The vain hope of Russian assistance without
strings for any neighbors of the USSR faced with aggression – turning the
Russian tap on and off at will – was dashed by Moscow’s refusal to play along.
The British reluctance to enter into anything more formal produced fatal
hesitation. Taylor seems uncertain whether to blame this dilatoriness on
incompetence or on a calculated attempt to avoid being dragged by the French
into an unwelcome arrangement with Russia. Chamberlain’s profound reluctance
and its ideological overtones are undeniable. Yet condemning British scruples to
avoid over-riding the attitude of the lesser states as “the narrow moralism of the
reformed drunkard” (p. 277) is to miss the point. The proposed front might
collapse if Poland and Rumania were lost at the expense of the Soviets. Moreover,
Soviet insistence upon a full alliance implied a military, rather than a diplomatic,
slant to the scheme which perhaps would further excite German fears of
encirclement. These were not mere rationalizations but real concerns which
Halifax, most of the cabinet, and the foreign office initially shared with
Chamberlain.50 By the end of May, pressure from the French, from public and
parliamentary opinion, plus a change of mind by the chiefs of staff and the
recurrent worry that the Russians might do a deal with the Germans, had isolated
Chamberlain from the majority of his colleagues on this issue. Taylor’s charge
that the British were not interested in an alliance with the Soviets is overstated,
though without access to cabinet records it was an easy assumption to make.
The Phantom Crisis: Danzig 1939 237
The Soviets, of course, eventually drew the same conclusion. The signs are that,
once it had dawned that the Russians would settle for nothing less than a full-
blown alliance, the British did reluctantly move forward, but did so with such
evident distaste as to drive the Russians into the arms of the Nazis, thus burying
the possibility of an eastern deterrent of any substance.
Chamberlain’s culpability for the failure to reach agreement with the Soviets
was considerable, but the burden should not be uniquely his. Though he was
its keenest advocate, Chamberlain was not the sole vessel of appeasement. Taylor,
however, has very little to say about Chamberlain’s colleagues. Indeed, his
account of the events of 1939 is largely a two-man show, a kind of disembodied
dialogue between Hitler and Chamberlain. The same might be said of his image
of Nazi Germany, with hardly a nod being made to the so-called “polycratic”
nature of the Nazi system. In Taylor’s drama Halifax, the foreign secretary, is a
relatively minor character, “coached by the foreign office,” or helping shake
Chamberlain’s confidence after the Prague coup on hearing “the call of
conscience in the watches of the night” (pp. 243, 253). Halifax’s stock has risen
since the publication of Origins, and his impact in redefining the direction of British
policy and in urging and manipulating an increasingly isolated Prime Minister
into taking it has been emphasized.51 Though the evidence for this remains
inconclusive, it does provide some of the depth that is lacking in Taylor’s rather
flat canvas.
It is possible to see Halifax, ever since Hitler’s upping of his demands against
the Czechs at Godesberg, nursing a developing, if not unswerving, pessimism
about German objectives. This increasingly put him out of step with the Prime
Minister – for example, over the need to widen the base of the government after
Munich, the need to set up a ministry of supply and over the appropriate response
to the extinction of Czechoslovakia. Vindicated by the seizure of Prague, Halifax
now possessed the confidence to press his opinions on Chamberlain. This line
of argument is not without its problems. The government was not broadened
until after war broke out. The ministry of supply, half-heartedly set up in April,
was as much the product of pressure from others as from Halifax. It has been
suggested that without Halifax’s presence “Chamberlain would have been able
to view Prague as a setback instead of being forced to see it as a major reversal.”52
This may be so, though it is difficult to believe that Chamberlain’s obstinacy
would have been sufficiently dented had he not himself also been shaken by
the starkness of what had now happened and at the end of an extended period
of nerve-wracking rumors. The ease with which the tough-minded Chamberlain
was purportedly persuaded or finessed is a hurdle which the supporters of Halifax
do not always effectively clear. Maybe Halifax “inserted a passage into the
Birmingham speech” envisaging war as preferable to dishonor, but those who
say so are not explicit about their sources.53 We might, just as reasonably, take
Chamberlain’s own explanation for the “very restrained and cautious exposition”
in the Commons on March 15, which he put down to having had “no time to
digest” the news of the Prague coup, “much less to form a considered opinion
of it.” His intention two days later at Birmingham was “to correct that mistake.”54
238 The Origins of the Second World War Reconsidered
The Danzig crisis had, as Taylor indicates, a feel very different from that over
Czechoslovakia the year before. Compared to the tensions and diplomatic activity
surrounding the Sudeten issue it was a more subdued, understated, muted affair;
hardly a crisis at all but a series of flickering, fizzling scares throughout the
summer, and with no direct diplomatic exchanges between Hitler and the Poles
before the outbreak of war or with the British until the middle of August. Each
of the participants “shrank from raising the question of Danzig,” giving it the
quality of a phantom crisis – “Danzig was not there; and if all the Powers wished
hard enough it would go away” (p. 307). Taylor’s explanation for this is that
the Poles felt they could only lose by entering negotiations, as the western powers
would force conciliation upon them. Hitler pursued his usual stratagem of waiting
for events to take their course. The west, meanwhile, did nothing, because it
recognized that the intransigence of both Berlin and Warsaw might result in war
if the Danzig question were raised: tension in the free city and its surrounding
territory meant that it was always highly combustible. But, as Taylor says, it
was a crisis that never broke. Hitler was indeed playing a characteristic game,
though not the one which Taylor suggests. Once more, Hitler was claiming the
moral high ground in order to divide his opponents. But, as he frequently stated
to his entourage, Danzig was not the issue. His reticence to discuss it was not
because he waited for others to force the Poles to give it to him, but because he
was apprehensive that if he stated his terms some unwelcome form of conciliation
would be foisted upon him by the west. Similarly, as Taylor states, the motive
of the Poles was to avoid pressure from the west for a new version of the
Runciman mission – mingled with a suspicion, which he only partly
acknowledges, that behind German pressure was a threat to the existence of
Poland. Fears of a third party conjuring up a deal were not unrealistic. To the
British, as well as to the French and Italians, Danzig was not worth a war.
Though they were not entirely convinced that a settlement over Danzig would
end Hitler’s belligerence, neither could they be certain that it would not. They
were torn between a recognition of the logic of the German case over Danzig
and a perception that it was another pretext hiding Hitler’s real objectives. This
gave a “dual nature” to British policy, making it “as much concerned to moderate
the Poles as to restrain Hitler,” and, though this is overstated, “ready, even eager,
to give way over Danzig” (pp. 270, 272). Nonetheless, there was no intention to
backpedal over the promise to support Poland’s independence. Co-existing with
the view that Danzig was “a bad wicket on which to make a stand,” and that “if
an impartial Martian were to act as arbitrator…he would give judgment…more
or less in accordance with Hitler’s offer,” was the tart assessment that “if Danzig
were to be restored to Germany in virtue of the right of self-determination, Hitler
should be required to evacuate Prague.”64 The problem remained one of deciding
where legitimacy ended and an unacceptable extension of German power began.
Because their judgment as to Hitler’s more likely purpose inclined toward the
latter, the British played down Danzig from fear that the issue might run out of
control.
240 The Origins of the Second World War Reconsidered
The British held to the illusion that they could be the ones who decided
whether Polish independence was at risk, not the unstable Hitler nor the
unreliable Beck. In the attempt to avoid a provocative action either from Berlin
or from Warsaw the British, while attempting to avoid undercutting the resolution
of the Poles to stand up to Germany or to allow Hitler to cry “encirclement,”
hoped – in a rather indolent way – that a solution which just might draw the
fire from the Danzig question would turn up. The problem was that this raised
the politically unacceptable vision of a new Munich. In Taylor’s words, “the
British government were trapped not so much by their guarantee to Poland, as
by their previous relations with Czechoslovakia. With her they had imposed
concession; towards her they had failed to honour their guarantee. They could
not go back on their word again, if they were to keep any respect in the world
or with their own people” (p. 263). The quandary in which the British found
themselves did not mean that they were “in effect, leaving the decision on war
or peace to Colonel Beck.”65 Nor was it the case that Danzig had become some
kind of symbol of Polish independence. Indications suggest that had a crisis
erupted over Danzig alone the British would have found excuses for backing
off. In the high summer of 1939 Halifax pointed out to the cabinet that although
there was an assumption that Britain was committed to fight for Danzig, the
reality was that “Danzig of itself should not be regarded as providing a casus
belli. If, however, a threat to Polish independence arose from Danzig, then this
country would clearly become involved.”66 In the end, this was never put to the
test, for Danzig never came under isolated attack. On September 1, fifty-two
divisions of the Wehrmacht smashed into Poland along four fronts.
The culmination of Taylor’s argument is that war broke out in September
1939 because of “a mistake, the result on both sides of diplomatic blunders” (p.
269); or, more famously, because of Hitler’s “launching on 29 August a
diplomatic manoeuvre which he ought to have launched on 28 August” (p. 336).
Only in this sense should we see it as “Hitler’s war.” This is no longer
controversial. It is simply wrong. To explain the actions of his singular version
of Hitler, a complex but coherent series of events is strained to breaking-point.
It appears beyond much doubt that by the end of April in Hitler’s mind the
military destruction of Poland had replaced the intention of drawing the Poles
into the German ambit. Convinced that the Poles would not play his game and
that they had been encouraged in this attitude by the British guarantee, this was
viewed as the prelude to an eventual conflict with the west. On August 12 Hitler’s
generals were instructed to bring forward the September 1 timing and to prepare
themselves for an attack on Poland in a fortnight’s time – on 26 August. Hitler’s
conviction of his own historic mission to lead Germany in a war of domination,
his frustration at being cheated out of war in September 1938, pressed him
forward. At a meeting on August 22, and with von Ribbentrop now in Moscow
dealing directly with Stalin, his commanders were told that a war with the
western democracies was inevitable. Poland, which had failed to respond to his
offers of “an acceptable relationship,” must be neutralized first. Neither Britain
nor France were expected to intervene.67
The Phantom Crisis: Danzig 1939 241
of settlement sought was one which would include a German guarantee for
Poland into which the Poles should enter freely. Lack of trust in Hitler’s word
meant that there was no attempt to coerce the Poles into this, as there had been
with the Czechs a year earlier. This was the legacy of Prague. No close
contingency of view existed between London and Berlin. Hitler did not leave
himself short of a crucial twenty-four hours of negotiating time (such precise
measurement for such an imprecise activity!), on the last day of August, which
would have opened the breach between the Poles and the west if only he had
timed things right.
There were obvious failings on the British side. Both the hiatus between the
guarantee and the Anglo-Polish alliance and indications that they were prepared
for a deal on Danzig no doubt fed Hitler’s belief that, in the last resort, the British
would betray the Poles, just as they had abandoned the Czechs. The shuttling
of Dahlerus and the intervention of other semi-official intermediaries may, as
Taylor suggests, have added to this belief. But if the British took such jockeying
seriously, and there are good indications that they did not, it was just as easy
for them to interpret these as a mark of Hitler’s failure of nerve.70 Possibly this,
and the clear evidence available to them that Hitler had indeed backed down
on the August 26, filled some in London with undue optimism that an acceptable
solution to German–Polish antagonism might yet be found.71 Essentially, however,
the purpose of British negotiations at this stage was to ensure that every possible
avenue to preserve peace which did not violate the assurances given to protect
Poland had been tried. Impatience had impelled Hitler to act on the August 26.
It did so again on September 1, even though the diplomatic situation had not
altered in Germany’s favor. There is even an indication that his timetable for
action could have been stretched to September 2, but rather than waiting to the
last moment he attacked before it was technically necessary to do so. As
Weinberg trenchantly insists, “there is…no sensible person who today disputes
the fact that the Third Reich initiated World War II.”72
The avalanche of evidential material which has emerged since Origins was
first published over thirty years ago has inevitably added sophistication to our
understanding of why war came in 1939. That Taylor’s assessment would have
been substantially different, had this material been available to him, is, however,
questionable. Certainly he did not demonstrate much inclination to recant in
his later references to Origins.73 His notorious rejection or manipulation of
documentation inconvenient to what he admitted was an “academic exercise”
(p. 266) would likely have survived. As would his Namierite view of history, in
which ideologies and impersonal structures are considerably less important forces
in human activity than are the fallible attempts of individuals to pursue what
were perceived to be traditional national interests within the parameters of a 100-
year-old European state system. Indeed, while what seemed most shocking thirty
years ago, his interpretation of Hitler and the accidental nature of the outbreak
of war, continues to be the most inadmissible, many of Taylor’s general The
Phantom Crisis: Danzig 1939 243 conclusions have endured in the face of more
recent approaches and fresh source material. Taylor dismisses as crude Marxist
The Phantom Crisis: Danzig 1939 243
dogma the view that economic crisis propelled Hitler’s action in 1939. Despite
the heat which has sometimes been generated by this debate, it remains an
unsettled question. There is little in what has emerged from it which would have
caused Taylor to rethink his position. Indeed, his conclusion that “the economic
argument ran against war, not in its favour,” probably still represents majority
thinking on this. 74 On the other hand, Taylor’s relegation of the economic
dimension to an afterthought leaves a layer of gauze across our understanding.
As has already been suggested, economic considerations played some part in
the British attitude to backing Poland. A German–Polish conflict, provided
resources were not wasted on the Poles, would enhance Britain’s chances of
victory in a long war. Some indication of this was available to Taylor, though a
whole range of economic, political, and military intelligence material, of course,
was not. His account, one may cautiously surmise, would not have varied much
in its interpretation had he been able to work from intelligence records. It is far
from clear that such evidence has made any significant difference to our
understanding of these events, and its impact on those who received it appears
to have been indecisive. It tended to reinforce, rather than alter, predominant
inclinations. The economic intelligence available to the Chamberlain cabinet was
inaccurate, contradictory, and therefore confusing. Reports at the end of 1938
that the German economy was in good shape and ready for war tended to
produce the glum response that Britain could only match this performance if
“totalitarian” methods were adopted. More optimistic analyses, such as that of
the chiefs of staff in February 1939 induced complacency and the belief that a
war postponed could only increase Britain’s advantage over Germany. 75 Their
opinion that the British economy was better placed than the German for a long
war was contested five months later by the treasury. 76 The failure of British
political intelligence between the wars is widely accepted. The British had
accurate warning before Germany pounced on Bohemia, but they failed to act.
Secret reports revealing the true nature and ambitions of the Nazi regime between
November 1938 and March 1939 may have played their part in propelling the
British government toward resistance to Hitler on moral grounds. 77 But this
underplays the shock of the reality of events in Prague and Memel. After all,
public opinion responded with moral indignation to what happened there without
being privy to secret intelligence reports. By the same token, though military
and economic forecasts might, after Munich, have given some comfort to the
British cabinet, there is room for doubt that this was sufficient to encourage any
coolly calculated decision on the optimum time for Britain to go to war. In the
end it was public revulsion against Nazi ambitions as much as any careful
estimate of the current status of the balance of forces which pressed the British
government forward, though optimistic predictions would naturally give a fillip
to this more basic impetus. Faced with a mixture of the same weary view at
which the French had also arrived, that there must be an end to this, and a
recognition that, as Taylor points out, a failure to act decisively would mean
the collapse of the government, accounts for the final decision.
244 The Origins of the Second World War Reconsidered
Origins remains a monument to Taylor’s passion for argument. Despite its clear
signs of age, its insights still have the capacity to impress largely because of the
verve and audacity of the author’s style. Yet it is often this very stylistic
impudence, with its arrogant over-assertions, which grate to the extent of
detracting from much of its acuity. Sometimes it seems more one of Taylor’s
televisual tours de force than it does a work of scholarship. His treatments of Polish
policy and the transformation of appeasement after Prague still have something
valuable to tell us. His insistence upon Danzig as itself the pivotal issue is against
the evidence, while the immediate circumstances which produced war are simply
perverse. Integral to this is his idiosyncratic picture of Hitler. It is this flaw that,
like a kind of academic metal fatigue, brings the intricate structure falling to its
destruction – though some parts have survived the impact. Origins has stimulated
the work of a generation of historians. But its reign of influence is over. The
debts to it have been fully acknowledged, and there is, frankly, no reason to
resort to it now other than as a piece of historiography. To paraphrase the final
sentence of the book: The Origins of the Second World War has become a matter of
historical curiosity.
Notes
1 A. J. P Taylor, The Origins of the Second World War, Harmondsworth, 1964, pp. 302–
3. Page numbers in the text are from this edition.
2 J. Karski, The Great Powers and Poland, 1919–1945: From Versailles to Yalta, Lanham,
MD, 1985, p. 324. Public Record Office, Kew, PRO, FO 371, 21569, C11867/2319/
12. October 4, 1938.
3 Documents on British Foreign Policy (henceforth DBFP), 3rd series, vol. 5, no. 268.
4 Karski, Great Powers, p. 261.
5 A. Prazmowska, Britain, Poland and the Eastern Front, 1939, Cambridge, 1987, pp. 31
and 33.
6 J. Lipski, Diplomat in Berlin, 1933–1939, New York, 1968, pp. 503–4 and 566.
7 G. L. Weinberg, The Foreign Policy of Hitler’s Germany: Starting World War II, 1937–
1939, Chicago, IL, 1980, p. 480.
8 Quoted in A. J. Prazmowska, “Poland’s foreign policy: September 1938–September
1939,” Historical Journal, vol. 29, 1986, p. 858.
9 D. C. Watt, How War Came: The Immediate Origins of the Second World War, 1938–1939,
London, 1989, p. 67; DBFP, vol. 5, no. 274.
10 Weinberg, Hitler’s Germany, p. 465 and pp. 503–4.
11 Watt, How War Came, pp. 38–45.
12 DBFP, vol. 4, no. 515; Kennard to Halifax, March 25, 1939.
13 R. A. C. Parker, Chamberlain and Appeasement, London, 1993, p. 205.
14 See e.g. J. Charmley Chamberlain and the Lost Peace, London, 1989, pp. 164 and 167;
Parker, Chamberlain, p. 203–5.
15 PRO, CAB 23/98 11(39), March 18, 1939.
16 J. Harvey (ed.), The Diplomatic Diaries of Oliver Harvey, 1937–1940, London, 1970, p.
433. The phrase is Harvey’s.
17 Watt, How War Came, pp. 103–8.
18 Watt, How War Came, p. 108.
19 DBFP, vol. 4, no. 458.
20 DBFP, vol. 4, no. 551, Halifax to Lindsay, Washington, March 28, 1939.
The Phantom Crisis: Danzig 1939 245
21 S. Newman, March 1939: The British Guarantee to Poland, Oxford, 1976, p. 129;
Prazmowska, Eastern Front, p. 51.
22 DBFP, vol. 4, no. 395, Halifax to Hoare, Bucharest, March 17, 1939.
23 C. Hill, Cabinet Decisions on Foreign Policy: The British Experience, October 1938–June 1941,
Cambridge, 1991, pp. 27–8.
24 D. Dilks (ed.), The Diaries of Sir Alexander Cadogan, 1938–45, London, 1971, p. 157
(diary entry of March 16, 1939); Chamberlain statement quoted from Hill, Cabinet
Decisions, p. 27.
25 W. K. Wark, The Ultimate Enemy: British Intelligence and Nazi Germany, 1933–1939,
London, 1985, pp. 116–20; DBFP, vol. 4, no. 522.
26 DBFP, vol. 5, no. 361, foreign office memorandum, May 4, 1939.
27 DBFP, vol. 4, no. 523.
28 Newman, British Guarantee, p. 139.
29 DBFP, vol. 4, no. 523.
30 DBFP, vol. 4, no. 523.
31 DBFP, vol. 4, no. 518; Harvey, Diaries, p. 267 (diary entry of March 25, 1939).
32 I. Colvin, Vansittart in Office, London, 1965, p. 305.
33 DBFP, vol. 5, no. 281.
34 Weinberg, Hitler’s Germany, p. 555.
35 A. Roberts, “The Holy Fox”: A Biography of Lord Halifax, London, 1991, p. 148.
36 Watt, How War Came, pp. 185–6.
37 A. M. Cienciala, “Poland in British and French policy in 1939: determination to
fight – or avoid war?”, in P. Finney (ed.), The Origins of the Second World War, London,
1997, p. 429.
38 G. Bruce Strang, “Once more unto the breach: Britain’s guarantee to Poland, March
1939,” Journal of Contemporary History, vol. 31, 1996, p. 724.
39 Newman, British Guarantee, pp. 196 and 219.
40 DBFP, vol. 5, no. 274.
41 A. J. Foster, “An unequivocal guarantee? Fleet Street and the British guarantee to
Poland, 31 March 1939,” Journal of Contemporary History, vol. 26, 1991, pp. 33–48.
42 Quoted in Foster “Unequivocal guarantee?”, p. 43.
43 DBFP, vol. 4, no. 584.
44 Prazmowska, Eastern Front, pp. 56–7.
45 DBFP, vol. 4, no. 458.
46 PRO, CAB 23/98, 16(39), meeting of March 30, 1939. See also DBFP, vol. 4, no.
584.
47 See W. K. Wark, “Something very stern: British political intelligence, moralism and
grand strategy in 1939,” Intelligence and National Security, vol. 5, 1990, pp. 163–4.
48 Strang, “Britain’s guarantee,” p. 743.
49 PRO, CAB 23/98, 16(39), meeting of March 30, 1939.
50 C. Hill, Cabinet Decisions, ch. 3; Parker, Chamberlain, pp. 229–31.
51 Halifax’s influential role during this period is described in: Watt, How War Came;
Charmley, Lost Peace; Parker, Chamberlain; R. J. Q. Adams, British Politics and Foreign
Policy in the Age of Appeasement, 1935–39, London, 1993; Roberts, Halifax; and Strang,
“Britain’s guarantee.”
52 Roberts, Halifax, p. 143.
53 Charmley, Lost Peace, p. 167; Adams, Age of Appeasement, p. 140; Watt, How War Came,
p. 167. Roberts states that Halifax “inserted a passage into the Birmingham speech”
envisaging war as preferable to dishonor, but provides no evidence for such direct
intervention on the part of the foreign secretary; Halifax, p. 144. Also Hill, Cabinet
Decisions, p. 22.
54 Cmd 6106, p. 5.
55 The quotation is from Strang, “Britain’s guarantee.” Roberts and Charmley take
the same view. Roberts, Halifax, p. 148. Charmley’s assessment of the guarantee is
246 The Origins of the Second World War Reconsidered
Bell, P. M. H., The Origins of the Second World War in Europe, London, 1986. ——
France and Britain, 1900–1940: Entente and Estrangement, London, 1996.
Beloff, Max, Imperial Sunset; vol. 2: Dream of Commonwealth, 1921–42, London,
1989.
Bennett, E. W., Germany and the Diplomacy of the Financial Crisis, 1931, Cambridge,
MA, 1962.
—— German Rearmament and the West, 1932–1933, Princeton, NJ, 1979.
Benz, Wolfgang and H. Graml (eds), Sommer 19 3 9. Die Grossmächte und der
Europäische Krieg, Stuttgart, 1979.
Berger, Gordon M., Parties Out of Power in Japan, 1931–1941, Princeton, NJ, 1977.
Bernstein, Gail Lee, and Harahiro Fukui, (eds), Japan and the World: Essays on
Japanese History and Politics in Honour of Ishida Takeshi, New York, 1988.
Berstein, Serge, La France des années 30, Paris, 1988.
Bessis, Juliette, La mediterranée fasciste. L’Italie mussolinienne et la Tunisie, Paris, 1980.
Bialer, U., The Shadow of the Bomber: The Fear of Air Attack and British Politics 1932–
1939, London, 1980.
Birn, D. S., The League of Nations’ Union 1918–1945, London, 1981.
Bischof, E., Rheinischer Separatismus, 1918–1924, Bern, 1969.
Blackbourn, David and G. Eley, The Peculiarities of German History, Oxford, 1984.
Bloch, M., Von Ribbentrop, London, 1992.
Boemeke, Manfred F., Gerald D. Feldman, and Elisabeth Glaser-Schmidt (eds),
The Treaty of Versailles: A Reassessment after Seventy-Five Years, Cambridge, 1998.
Bolech Cecchi, D., L’accordo di due imperi: L’accordo italo-inglese del 16 aprile 1938,
Milan, 1977.
Bolloten, Burnett, The Spanish Civil War. Revolution and Counterrevolution, Chapel Hill,
NC, 1991.
Bond, B., British Military Policy Between the Two World Wars, Oxford, 1980.
Borchardt, K., Wachstum, Krisen, Handelsspielräume der Wirtschaftspolitik, Göttingen,
1982.
Bosworth, R. J. B., and G. Rizzo (eds), Altro Polo: Intellectuals and Their Ideas in
Contemporary Italy, Sydney, 1983.
—— and Sergio Romano (eds), La politica estera italiana (1860–1985), Bologna, 1991
—— Explaining Auschwitz and Hiroshima: History Writing and the Second World War, 1945–
1990, London, 1993.
—— Italy and the Wider World, 1860–1960, London, 1996.
Boyce, Robert and Esmonde M. Robertson (eds), Paths to War: New Essays on the
Origins of the Second World War, London, 1989.
Boyle, John H., China and Japan at War, 1937–1945: The Politics of Collaboration,
Stanford, CA, 1972.
Braunthal, J., History of the International;vol. 2: 1914–1943, New York, 1967.
Brenan, Gerald, The Spanish Labyrinth. An Account of the Social and Political Background
of the Civil War, Cambridge, 1967.
Broué, Pierre and Emile Témime, The Revolution and the Civil War in Spain,
Cambridge, MA, 1970.
Browning, Christopher, Ordinary Men: Reserve Battalion 101 and the Final Solution in
Poland, New York, 1992.
Brundu Olla, P., L’equilibrio difficile: Gran Bretagna, Italia e Francia nel Mediterraneo,
Milan, 1980.
Burgwyn, H. James, Il revisionismo fascista: La sfida di Mussolini alle grande potenze nei
Balcani e sul Danubio, 1925–1933, Milan, 1979.
—— The Legend of the Mutilated Victory: Italy, the Great War, and the Paris Peace
Conference, 1915–1919, Westport, CT, 1993.
—— Italian Foreign Policy in the Interwar Period, 1918–1940, Westport, CT, 1997.
Burk, K., Britain, America and the Sinews of War, 1914–1918, London, 1985.
Bibliography 249
De Grand, A., The Italian Nationalist Association and the Rise of Fascism in Italy,
Lincoln, NE, 1978.
Debicki, R., The Foreign Policy of Poland, 1919–1939, New York, 1962.
Debo, Richard K., Revolution and Survival: The Foreign Policy of Soviet Russia, 1917–18,
Toronto, 1979.
—— Survival and Consolidation: The Foreign Policy of Soviet Russia, 1918–1921, Montreal,
1992.
Deist, Wilhelm, The Wehrmacht and German Rearmament, London, 1981.
—— et al., Germany and the Second World War; vol. 1: The Build-Up of German
Aggression, Oxford, 1990.
Del Boca, A. (ed.), Adua: Le ragioni della sconfitta, Rome, 1997.
Deutsch, H., Hitler and His Generals: The Hidden Crisis, January–June 19 3 8,
Minneapolis, MN, 1974.
Retreat from Power, 2 vols, London, 1981. Dilks, David (ed.),
—— Neville Chamberlain, Cambridge, 1984, vol. 1.
Dingman, R., Power in the Pacific, Chicago, IL, 1976.
Dockrill, Michael L. and J. Douglas Goold, Peace Without Promise: Britain and the
Peace Conferences, 1919–1923, Hamden, CT, 1981.
——and Brian McKercher, Diplomacy and World Power: Studies in British Foreign Policy,
1890–1950, New York, 1996.
Döscher, Hans-Jürgen, Das Auswärtige Amt im Dritten Reich: Diplomatie im Schatten der
Endlösung, Berlin, 1986.
Doughty, Robert A., The Seeds of Disaster. The Development of French Army Doctrine,
1919–1939, Hamden, CT, 1985.
—— The Breaking Point. Sedan and the Fall of France, 1940, New York, 1990.
Douglas, R. (ed.), 1939: A Retrospective Forty Years After, London, 1983.
Dower, J. W., Empire and Aftermath: Yoshida Shigeru and the Japanese Experience, 1878–
1954, Cambridge, 1979.
Dreifort, John E., Yvon Delbos at the Quai D’Orsay. French Foreign Policy During the
Popular Front, 1936–1938, Lawrence, KS, 1973.
—— Myopic Grandeur. The Ambivalence of French Foreign Policy Toward the Far East, 1919–
1945, Kent, OH, 1991.
Duroselle, J.-B., La Décadence, 1932–1939, Paris, 1979
Dutton, David, Austen Chamberlain: Gentleman in Politics, New Brunswick, NJ, 1985.
—— Simon: A Political Biography of Sir John Simon, London, 1992.
—— Anthony Eden: A Life and Reputation, New York, 1997.
Duus, Peter (ed.), Cambridge History of Japan; vol. 6: The Twentieth Century,
Cambridge, 1988.
Edwards, Jill, The British Government and the Spanish Civil War, 1936–1939, London,
1979.
Eichengreen, B., Golden Fetters, New York, 1992.
Emmerson, James T., The Rhineland Crisis, 7 March 1936. A Study in Multilateral
Diplomacy, London, 1977.
Endicott, S. L., Diplomacy and Enterprise: British China Policy, 1933–1937, Vancouver,
BC, 1973.
Engel, G., Heeresadjutant bei Hitler, 1938–19 43; Aufzeichnungen des Majors Engel,
Stuttgart, 1974.
Enssle, Manfred J., Stresemann’s Territorial Revisionism, Wiesbaden, 1980.
Erdmann, K. D., Adenauer in der Rheinlandpolitik nach dem Ersten Weltkrieg, Stuttgart,
1966.
Favez, J.-C., Le Reich devant l’occupation franco-belge de la Ruhr en 1923, Geneva, 1969.
Feldman, Gerald D., Iron and Steel in the German Inflation, 1916–1923, Princeton,
NJ, 1977.
Bibliography 251
—— The Great Disorder: Politics, Economics and Society in the German Inflation, 1914–
1924, Oxford, 1993.
Felix, David, Walther Rathenau and the Weimar Republic, Baltimore, MD, 1971.
Ferris, John R., Men, Money, and Diplomacy: The Evolution of British Strategic Foreign
Policy, 1919–1926, Ithaca, NY, 1989.
Fink, Carole, The Genoa Conference: European Diplomacy, 1921–1922, Chapel Hill,
NC, 1984.
—— Isabel V. Hull, and MacGregor Knox, German Nationalism and the European
Response, 1890–1945, Norman, OK, 1985.
—— et al., Genoa, Rapallo, and European Reconstruction in 1922, Washington, DC,
1991.
Finney, Patrick (ed.), The Origins of the Second World War, London, 1997.
Fischer, Fritz, From Kaiserreich to Third Reich, trans. Roger Fletcher, London, 1986.
Fleischhauer, Ingeborg, Der Pakt: Hitler, Stalin und die Initiative der deutschen
Diplomatie, 1938–1939, Frankfurt, 1990.
Fletcher, William M., The Search for a New Order: Intellectuals and Fascism in Prewar
Japan, Chapel Hill, NC, 1982.
Floto, Inga, Colonel House in Paris, Åarhus, 1973.
Foot, M. R. D. (ed.), War and Society, London, 1973
Fox, J., Germany and the Far Eastern Crisis, 1931–1938: A Study in Diplomacy and
Ideology, Oxford, 1982.
Frankenstein, Robert, Le prix du réarmement français, 1935–1939, Paris, 1982.
Frei, N. and H. Kling (eds), Der nationalsozialistische Krieg, Frankfurt am Main, 1990.
Friedlander, Saul, Nazi Germany and the Jews, New York, 1997.
Frucht, Richard (ed.), Labyrinth of Nationalism/Complexities of Diplomacy, Columbus,
OH, 1992.
Fry, M. G., Illusion of Security, Toronto, 1972.
Fuchser, L.W., Neville Chamberlain and Appeasement. A Study in the Politics of History,
New York, 1982.
Funke, Manfred, Sanktionen und Kanonen: Hitler, Mussolini und der Abessinienkonflikt,
1934–1936, Düsseldorf, 1970.
Gannon, F. R., The British Press and Germany, 1936–39, Oxford, 1971.
Gates, Eleanor M., End of the Affair: The Collapse of the Anglo-French Alliance, 1939–
1940, Berkeley, CA, 1980.
Gatzke, Hans W. (ed.), European Diplomacy Between Two Wars, 1919–1939, Chicago,
IL, 1972.
Geyer, M., Aufrüstung oder Sicherheit: Die Reichswehr in der Krise der Machtpolitik, 1924–
1936, Wiesbaden, 1980.
Gibbs, N., Grand Strategy, London, 1976, vol. 1.
Gilbert, Martin, The Roots of Appeasement, London and New York, 1966.
——and R. Gott, The Appeasers, London, 1963.
Gillingham, John, Coal, Steel and the Rebirth of Europe, 1945–1955: The Germans and
French from Ruhr Conflict to Economic Community, Cambridge, 1991.
Giordano, G., Il Patto a Quattro nella politica estera di Mussolini, Bologna, 1976.
Girault, René and Robert Frank (eds), La Puissance en Europe, 1938–1940, Paris,
1984.
Gnedin, Evgenii; Vykhod iz labirinta, New York, 1982.
Goldstein, Erik, Winning the Peace: British Diplomatic Strategy, Peace Planning and the
Paris Peace Conference, 1916–1920, Oxford, 1991.
Gorodetsky, Gabriel (ed.), Soviet Foreign Policy, 1917–1991: A Retrospective, London,
1994.
Gordon, G. A. H., British Seapower and Procurement Between the Wars: A Reappraisal of
Rearmament, London, 1988.
Graham, Helen and Paul Preston (eds), The Popular Front in Europe, London, 1987.
252 Bibliography
Griffiths, Richard, Fellow Travellers of the Right. British Enthusiasts for Nazi Germany,
1933–9, London, 1980.
Grundmann, K.-H., Deutschtumpolitik zur Zeit der Weimarer Republik, Hanover, 1977.
Gunsburg, Jeffery A., Divided and Conquered: The French High Command and the Defeat
of the West, 1940, Westport, CT, 1979.
Haggie, P., Britannia at Bay: The Defence of the British Empire Against Japan, Oxford,
1981.
Haigh, R. H., Morris, D. S., and Peters, A. R., Soviet Foreign Policy, the League of
Nations, and Europe, 1917–1939, Totowa, NJ, 1986.
Halliday, Jon, A Political History of Japanese Capitalism, New York, 1975.
Hannah, L., The Rise of the Corporate Economy, London, 1983.
Hansen, E. W., Reichswehr und Industrie, Boppard am Rhein, 1981.
Haraszti, E. H., Treaty-Breakers or Realpolitiker? The Anglo-German Naval Agreement of
June 1935, Boppard am Rhein, 1973.
Hardie, F. M., The Abyssinian Crisis, London, 1974.
Harper, Glenn T., German Economic Policy in Spain during the Spanish Civil War, 1936–
1939, The Hague, 1967.
Haslam, Jonathan, Soviet Foreign Policy, 1930–33: The Impact of the Depression, New
York, 1983.
—— The Soviet Union and the Struggle for Collective Security in Europe, 1933–39, New
York, 1984.
—— The Soviet Union and the Threat from the East, 1933–41, Pittsburgh, PA, 1992.
Havens, Thomas R. H., Farm and Nation in Modern Japan; Agrarian Nationalism,
1870–1940, Princeton, NJ, 1974.
—— Valley of Darkness; The Japanese People and World War Two, Lanham, MD, 1986.
Heiber, Helmut, The Weimar Republic, trans. W. E. Yuill, Oxford, 1993.
Heinemann, J. L., Hitler’s First Foreign Minister, Berkeley, CA, 1979.
Hiden, John, Republican and Fascist Germany, London, 1996.
Hilberg, Raul, The Destruction of the European Jews, New York, 1985 edn .
Hildebrand, K., Vom Reich zum Weltreich, Munich, 1969.
—— The Foreign Policy of the Third Reich, London, 1973.
—— J. Schmädeke and K. Zernack (eds), 1939: an der Schwelle zum Weltkrieg, Berlin,
1990.
—— Das vergangene Reich: Deutsche Aussenpolitik von Bismarck bis Hitler, 1871–1945,
Stuttgart, 1995.
Hill, C., Cabinet Decisions on Foreign Policy: The British Experience, October 1938–June
1941, Cambridge, 1991.
Hinsley, F. H., British Intelligence in the Second World War: Its Influence on Strategy and
Operations, London, 1979, vol. 1.
Hochman, Jiri, The Soviet Union and the Failure of Collective Security, 1934–1938,
Ithaca, NY, 1984.
Hoensch, Jörg K., A History of Modern Hungary, 1867–1994, New York, 1996.
Hoepke, K.-P., Die deutsche Rechte und der italienische Faschismus, Düsseldorf, 1968.
Hogan, M. J., Informal Entente, Columbia, MO, 1977.
Hogenhuis-Seliverstoff, Anne, Les relations franco-soviétiques, 1917–1924, Paris, 1981.
Holtfrerich, Carl-Ludwig, The German Inflation, 1914–1923: Causes and Effects in
International Perspective, trans. Theo Balderston, Berlin, 1986.
Hölzle, E. (ed.), Die deutschen Ostgebiete zur Zeit der Weimarer Republik, Cologne, 1966.
Hoptner, J. B., Yugoslavia in Crisis, 1934–1941, New York, 1962.
Hosten, Germaine A., Marxism and the Crisis of Development in Prewar Japan,
Princeton, NJ, 1986.
Howard, Michael, The Continental Commitment, London, 1972.
Howson, S. and D. Winch, The Economic Advisory Council, 1930–1939, Cambridge,
1977.
Bibliography 253
Lukes, Igor, Czechoslovakia Between Stalin and Hitler, New York, 1996.
Lundgreen-Nielsen, Kay, The Polish Problem at the Paris Peace Conference, Odense,
1979.
Lungu, Dov B., Romania and the Great Powers, 1933–1940, Durham, NC, 1989.
Lyttleton, A., The Seizure of Power: Fascism in Italy, 1919–1929, 2nd edn, Princeton,
NJ, 1988.
Macartney, C. A., October Fifteenth. A History of Modern Hungary 19 2 9–19 4 5,
Edinburgh, 1956.
MacDonald, C. A., The United States, Britain, and Appeasement, 1936–1939, New
York, 1981.
Mack Smith, Denis, Mussolini’s Roman Empire, New York, 1976.
Maier, Charles S., Recasting Bourgeois Europe, Princeton, NJ, 1975.
—— The Unmasterable Past: History, Holocaust, and German National Identity, Cambridge,
MA, 1988.
Maksimychev, I. F., Diplomatiia mira protiv diplomatii voiny, Moscow, 1981.
Marcus, Harold G., Haile Sellassie I. The Formative Years, 1892–1936, Berkeley, CA,
1987.
Marder, A. J., Old Friends, New Enemies: the Royal Navy and the Imperial Japanese Navy,
Oxford, 1981.
Marks, Sally, The Illusion of Peace: International Relations in Europe, 1918–1933,
London, 1976.
—— Innocent Abroad: Belgium at the Paris Peace Conference of 1919, Chapel Hill, NC,
1981.
Marquand, David, Ramsay MacDonald, London, 1977.
Marrus, Michael, The Holocaust in History, Hanover, NH, 1987.
Martel, Gordon (ed.), Modern Germany Reconsidered, London, 1992.
Maurer, I. and U. Wengst, (eds), Politik und Wirtschaft in der Krise: Quellen zur Ära
Brüning, 2 vols, Düsseldorf, 1980.
May, Ernest R. and James C. Thomson, Jr, American–East Asian Relations: A Survey,
Cambridge, MA, 1972.
—— (ed.), Knowing One’s Enemies: Intelligence and Assessment Before the Two World Wars,
Princeton, NJ, 1984.
Mayer, Arno J., Politics and Diplomacy of Peacemaking: Containment and Counterrevolution
at Versailles, New York, 1967.
McDermott, Kevin and Jeremy Agnew, The Comintern: A History of International
Communism from Lenin to Stalin, New York, 1997.
McDougall, Walter A., France’s Rhineland Diplomacy, 1914–1924, Princeton, N J,
1978.
McIntyre, W. D., The Rise and Fall of the Singapore Naval Base, London, 1979.
McKercher, B. J. M., The Second Baldwin Government and the United States, 1924–
1929, Cambridge, 1984.
—— (ed.), Anglo-American Relations in the 1920s, Edmonton, 1990.
McSherry, James E., Stalin, Hitler, and Europe, Cleveland, OH, 1968, vol. 1.
Medvedev, Roy, All Stalin’s Men, Garden City, NY, 1984.
Megill, A., (ed.), Rethinking Objectivity, Durham, NC, 1994.
—— and R. J. Evans, In Defence of History, London, 1997.
Melchionni, M. G., La vittoria mutilata, Rome, 1981.
Meyers, R., Britische Sicherheitspolitik, 1934–1938, Düsseldorf, 1976.
Michaelis, Meir, Mussolini and the Jews: German–Italian Relations and the Jewish Question
in Italy, 1922–1945, Oxford, 1978.
Michalka, W., Ribbentrop und die deutsche Weltpolitik, 1933–1940, Munich, 1980.
—— and M. M. Lee, (eds), Gustav Stresemann, Darmstadt, 1982.
Middelton, R., Towards the Managed Economy: Keynes, the Treasury and the Fiscal Policy
Debates in the 1930s, London, 1985.
256 Bibliography
—— and Spencer C. Olin, Jr, Why War? Ideology, Theory, and History, Berkeley, CA,
1979.
Neufeld, Maurice F., Italy: School for Awakening Countries, Ithaca, NY, 1961.
Newman, Simon, March 1939: The British Guarantee to Poland, Oxford, 1976.
Niedhart, G., Grossbritannien und die Sowjetunion, 1934–1939, Munich, 1972.
Nish, Ian. H., Alliance in Decline, London, 1972.
—— Japan’s Struggle with Internationalism: Japan, China, and the League of Nations, 1931–
1933, New York, 1993.
Noel-Baker, P., The First World Disarmament Conference, 1932–1933, and Why It Failed,
New York, 1979.
Ogata, Sadako N., Defiance in Manchuria: The Making of Japanese Foreign Policy, 1931–
1932, Berkeley, CA, 1964.
Orde, Anne, Britain and International Security, 1920–1926, London, 1978.
—— British Policy and European Reconstruction After the First World War, Cambridge,
1990.
Ovendale, Russia., Appeasement and the English-Speaking World, Cardiff, 1975.
Overy, R. J., Goering: The “Iron Man,” London, 1984.
—— War and Economy in the Third Reich, Oxford, 1994.
Parker, R. A. C., Chamberlain and Appeasement. British Policy and the Coming of the
Second World War, London and New York, 1993.
Parrini, C., Heir to Empire, Pittsburgh, PA, 1969.
Payne, Stanley, A History of Fascism, 1914–1945, Madison, WI, 1995.
Peattie, Mark, Ishiwara Kanji and Japan’s Confrontation with the West, Princeton, NJ,
1975.
Peden, G. C., British Rearmament and the Treasury, 1932–1939, Edinburgh, 1979.
Pereboom, Maarten L., Democracies at the Turning Point: Britain, France and the End of
the Postwar Order 1928–1933, New York, 1995.
Peters, A. R., Anthony Eden at the Foreign Office, 1931–1938, New York, 1986.
Petersen, Jens, Hitler e Mussolini. La difficile alleanza, Bari, 1975.
Petricioli, Marta ( ed.), A Missed Opportunity? 1922: The Reconstruction of Europe,
Bern, 1995.
Phillips, Hugh D., Between the Revolution and the West: A Political Biography of Maxim
M. Litvinov, Boulder, CO, 1992.
Pimlott, Ben, Labour and the Left in the 1930s, Cambridge, 1977.
—— Hugh Dalton, London, 1985.
Pipes, R., The Bolshevik Regime, New York, 1994.
—— The Unknown Lenin, New Haven, CT, 1996.
Pitts, V. J., France and the German Problem, 1924–1929, New York, 1987.
Pohl, K.-H., Weimars Wirtschaft und die Aussenpolitik der Republik, 192 4–19 2 6,
Düsseldorf, 1979.
Poidevin, R. and J. Bariéty, Les relations franco-allemands, 1815–1975, Paris, 1977.
Pollard, S., The Development of the British Economy, 1914–1967, London, 1969.
Ponomaryov, B., A. Gromyko, and V. Khvostov (eds), History of Soviet Foreign Policy,
1917–1945, Moscow, 1969.
Porch, Douglas, The French Secret Services. From the Dreyfus Affair to the Gulf War, New
York, 1995.
Post, Gaines, Jr, The Civil–Military Fabric of Weimar Foreign Policy, Princeton, N J,
1973.
—— Dilemmas of Appeasement: British Deterrence and Defense, 1934–1937, Ithaca, NY,
1993.
Pratt, P. R., East of Malta, West of Suez: Britain’s Mediterranean Crisis, 1936–1939,
Cambridge, 1975.
Prazmowska, A. Britain, Poland and the Eastern Front, 1939, Cambridge, 1987.
258 Bibliography
Preston, Adrian (ed.), General Staffs and Diplomacy Before the Second World War,
London, 1974.
Proctor, Raymond L., Hitler’s Luftwaffe in the Spanish Civil War, Westport, CT, 1983.
Punka, George, Hungarian Air Force, Carrolton, TX, 1994.
Puzzo, Dante A., Spain and the Great Powers, 1936–1941, Freeport, CT, 1962.
Quartararo, R., Roma tra Londra e Berlino: Politica estera fascista dal 1930 al 1940,
Rome, 1980.
Raack, R. C., Stalin’s Drive to the West, 1938–1945: The Origins of the Cold War,
Stanford, CA, 1995.
Radice, L., Prelude to Appeasement: East European Central Diplomacy in the Early 1930s,
Boulder, CO, 1981.
Rainero, Romain, La rivendicazione fascista sulla Tunisia, Milan, 1978.
Raspin, Angela, The Italian War Economy, 1940–43, London, 1986.
Réau, Elisabeth du, Edouard Daladier, 1884–1970, Paris, 1993.
Reynolds, D., The Creation of the Anglo-American Alliance, 1937–1941, Chapel Hill,
NC, 1981.
Rhodes, Anthony, The Vatican in the Age of the Dictators, 1922–1945, London, 1973.
Rhodes James, Robert, Anthony Eden, London, 1987.
Rich, N., Hitler’s War Aims, New York, 1973, vol. 1.
Roberts, Andrew, “ The Holy Fox”: A Biography of Lord Halifax, London, 1991.
Roberts, Geoffrey, The Unholy Alliance: Stalin’s Pact with Hitler, London, 1989.
—— The Soviet Union and the Origins of the Second World War: Russo–German Relations
and the Road to War, 1933–1939, London, 1995.
Robertson, E. M. (ed.), The Origins of the Second World War: Historical Interpretations,
London, 1971.
Rochat, Giorgio, Il colonialismo italiano, Turin, 1973.
Rodder, A., Stresemanns Erbe: Julius Curtius und die deutsche Aussenpolitik 1929–1931,
Paderborn, 1996.
Rohwer, Jürgen and Eberhard Jäckel (eds), Kriegswende, Dezember 1941, Frankfurt-
am-Main, 1984.
Rosa, Arianna Arisi, La diplomazia del ventennio. Storia di una politica estera, Milan,
1990.
Rosenbaum, Kurt, Community of Fate: German–Soviet Diplomatic Relations, 1922–1928,
Syracuse, NY, 1965.
Roskill, S., Naval Policy Between the Wars, 2 vols, London, 1968 and 1976.
Rothwell, V. H., British War Aims and Peace Diplomacy, 1914–1918, Oxford, 1971.
Rovighi, Alberto and Filippo Stefani, La partecipazione italiana alla guerra civile
spagnola, 1936–1939, 2 vols, Rome, 1992–93.
Rowland, B. M., Commercial Conflict and Foreign Policy, New York, 1987.
Rupieper, Hermann J., The Cuno Government and Reparations, 19 22–1923, The
Hague, 1979.
Rusinow, D. J., Italy’s Austrian Heritage, 1919–1946, New York, 1969.
Saburô Ienaga The Pacific War, 1931–1945, New York, 1978.
Sadkovich, J. J., Italian Support for Croatian Separatism, 1927–1937, New York, 1987.
Sakmyster, Thomas, Hungary, the Great Powers, and the Danubian Crisis 1936–1939,
Athens, GA, 1980.
—— Hungary’s Admiral on Horseback. Miklós Horthy, 1918–1944, New York, 1994.
Salewski, M., Entwaffnung und Militärkontrolle in Deutschland, 1919–1927, Munich,
1966.
—— Die deutsche Seekriegsleitung, 1939–1945, 2 vols, Frankfurt am Main, 1970.
Sbacchi, Alberto (ed.), Legacy of Bitterness. Ethiopia, and Fascist Italy, 1935–1941, Law-
renceville, NJ, 1997
Schmidt, G. (ed.), Konstellationen internationaler Politik, 1924–1932, Bochum, 1983.
Schmidt, Royal J., Versailles and the Ruhr, The Hague, 1968.
Bibliography 259
Schmitz, D. F., The United States and Fascist Italy, 1922–1940, Chapel Hill, NC,
1988.
Schmokel, W. W., Dream of Empire: German Colonialism, 1919–1945, New Haven,
CT, 1964.
Schröder, H.-J. (ed.), Confrontation and Cooperation, Providence, RI, 1993.
Schuker, Stephen A., The End of French Predominance in Europe, Chapel Hill, NC,
1976.
—— American “Reparations” to Germany, 1919–33, Princeton, NJ, 1988.
Schwabe, Klaus, Woodrow Wilson, Revolutionary Germany, and Peacemaking, 1918–1919:
Missionary Diplomacy and the Realities of Power, trans. Rita and Robert Kimber,
Chapel Hill, NC, 1985.
Schwerin von Krosigk, L., Staatsbankrott: Finanzpolitik des Deutschen Reichs 1920–
1945, Stuttgart, 1974.
Segrè, C., Fourth Shore: The Italian Colonization of Libya, Chicago, IL, 1974.
Semiriaga, M. I., Tainy stalinskoi diplomatii, 1939–1941, Moscow, 1992.
Shai, A., Origins of the War in the East: Britain, China, and Japan, 1937–41, London,
1976.
Sharp, Alan, The Versailles Settlement: Peacemaking in Paris, 1919, New York, 1991.
Shay, R. P., Jr, British Rearmament in the Thirties: Politics and Profits, Princeton, NJ,
1977.
Sheinis, Zinovy, Maxim Litvinov, Moscow, 1990.
Shillony, Ben-Ami, Revolt in Japan: The Young Officers and the February 26, 1936
Incident, Princeton, NJ, 1972.
—— Politics and Culture in Wartime Japan, Oxford, 1981.
Shorrock, W. I., From Ally to Enemy: The Enigma of Fascist Italy in French Diplomacy,
1920–1940, Kent, OH, 1988.
Shunsuke Tsurumi, An Intellectual History of Wartime Japan, 1931–1945, London,
1986.
Silberman, Bernard, and H. D. Harootunian (eds), Japan in Crisis, Princeton, NJ,
1974.
Silverman, Dan P., Reconstructing Europe after the Great War, Cambridge, MA, 1982.
Sisman, Adam, A. J. P. Taylor: A Biography, London, 1994.
Skidelsky, Robert, John Maynard Keynes, London, 1983.
Smethurst, Richard, A Social Basis for Prewar Japanese Militarism: The Army and the
Rural Community, Berkeley, CA, 1974.
Smith, M., British Air Strategy Between the Wars, Oxford, 1984.
Smith, W. D., The Ideological Origins of Nazi Imperialism, Oxford, 1986.
Soutou, G., L’Or et le sang, Paris, 1989.
Spitzy Reinhard, How We Squandered the Reich, Norwich, 1997.
Steinhoff, Patricia G., Tenkô: Ideology and Social Integration in Prewar Japan, New York,
1991.
Steinmeyer, G., Die Grundlagen der französischen Rheinlandpolitik 1917–1919, Stuttgart,
1979.
Stern, F., The Politics of Cultural Despair: A Study in the Rise of the German Ideology,
Berkeley, CA, 1961.
Stevenson, D., French War Aims Against Germany, 1914–1919, Oxford, 1982.
—— The First World War and International Politics, Oxford, 1988.
Stoakes, G., Hitler and the Quest for World Dominion, Oxford, 1986.
Stürmer, M. (ed.), Die Weimarer Republik, Königstein, 1980.
Suval, S., The Anschluss Question in the Weimar Era, Baltimore, MD, 1974.
Takehiko Yoshihashi, Conspiracy at Mukden: The Rise of the Japanese Military, New
Haven, CT, 1963.
Tatsuo Arima, The Failure of Freedom: A Portrait of Modern Japanese Intellectuals,
Cambridge, MA, 1969.
260 Bibliography
—— “Strategies of French foreign policy after World War I,” Journal of Modern
History, vol. 55, 1983, pp. 78–95.
—— “The Soviet Union and Versailles,” in Manfred F. Boemeke et al. (eds),
The Treaty of Versailles: A Reassessment after Seventy-Five Years, Cambridge,
1998, pp. 451–68.
Jordan, Nicole, “Maurice Gamelin, Italy and the eastern alliances,” Journal of
Strategic Studies, vol. 14, 1991, pp. 428–41.
Jukes, G., “The Red Army and the Munich crisis,” Journal of Contemporary
History, vol. 26, 1991, pp. 195–214.
Kaiser, David, T. Mason, and R.J. Overy, “Debate: Germany, ‘domestic crisis’
and war in 1939,” Past &Present, no. 122, 1989, pp. 205–40
—— “Hitler and the coming of war,” in Gordon Martel (ed.), Modern Germany
Reconsidered, 1870–1945, London, 1992, pp. 178–96.
Kasza, Gregory, “Fascism from below? A comparative perspective on the
Japanese Right, 1931–1936,” Journal of Contemporary History, vol. 19, 1984,
pp. 607–30.
Keeton, Edward D., “Economics and politics in Briand’s German policy,
1925–1931,” in Carole Fink, et al. (eds), German Nationalism and the European
Response, 1890-1945, London, 1985, pp. 157–80.
Kennedy, P., “Idealists and realists: British views of Germany, 1864–1939,”
Transactions of the Royal Historical Society, series 5, vol. 25, 1975, pp. 135–56.
—— “The tradition of appeasement in British foreign policy, 1865–19 39,”
British Journal of International Studies, vol. 2, 1976, pp. 195–215.
—— “Appeasement and British defence policy in the inter-war years,” British
Journal of International Studies, vol. 4, 1978, pp. 161–77.
—— “Strategy versus diplomacy in twentieth-century Britain’ International History
Review, vol. 3, 1981, pp. 45–61.
Kettenacker, L., “Die Diplomatie der Ohnmacht,” in Wolfgang Benz and H.
Graml (eds), Sommer 19 3 9. Die Grossmachte und der Europaische Krieg,
Stuttgart, 1979, pp. 223–79.
Klein, Fritz, “Between Compiègne and Versailles: the Germans on the way
from a misunderstood defeat to an unwanted peace,” in Manfred F.
Boemeke et al. (eds), The Treaty of Versailles: A Reassessment after Seventy-Five
Years, Cambridge, 1998, pp. 203–20.
Knox, MacGregor, “Conquest, foreign and domestic, in Fascist Italy and Nazi
Germany,” Journal of Modern History, vol. 56, 1984, pp. 1–57.
—— “Il fascismo e la politica estera italiana,” in R. J. B. Bosworth and Sergio
Romano (eds), La politica estera italiana (18 60–19 8 5), Bologna, 1991, pp.
287–330.
—— “The Fascist regime, its foreign policy and its wars: An anti-Fascist
orthodoxy?’, Contemporary European History, vol. 4, 1995, 347–65.
Koch, H., “Hitler’s ‘programme’ and the genesis of Operation ‘Barbarossa’”,
Historical Journal, vol. 26, 1983, pp. 891–920.
Krüger, Peter, “A rainy day, April 16, 19 2 2: the Rapallo Treaty and the
cloudy perspective for German foreign policy,” in Carole Fink et al. (eds),
Genoa, Rapallo, and European Reconstruction in 1922, Washington, D C, 1991,
pp. 49–64.
Laffan, M., “Weimar and Versailles,” in M. Laffan (ed.), The Burden of German
History, 1919–1945, London, 1988, pp. 91–9.
Lammers, Donald N., “The May crisis of 1938: The Soviet view considered,”
South Atlantic Quarterly, vol. 69, 1970, pp. 480–503.
Ledeen, M., “Renzo De Felice and the controversy over Italian Fascism,”
Journal of Contemporary History, vol. 11, 1976, pp. 269–82.
Bibliography 267
Leffler, Melvyn P., “The origins of Republican war debt policy, 1921–1923,”
Journal of American History, vol. 59, 1972, pp. 585–601.
Lukowitz, D., “British pacifists and appeasement,” Journal of Contemporary
History, vol. 9, 1974, pp. 115–28.
MacDonald, Callum A., “Britain, France and the April crisis of 19 3 9,”
European Studies Review, vol. 2, 1972, pp. 151–69.
—— “Radio Bari and Italian propaganda in the middle east and British counter
measures, 1934–1938,” Middle Eastern Studies, vol. 13, 1977, pp. 195–207.
Manne, Robert, “The British decision for alliance with Russia, May 1939,”
Journal of Contemporary History, vol. 9, 1974, pp. 3–26.
—— “Some British light on the Nazi–Soviet pact,” European Studies Review, vol.
11, 1981, pp. 83–102.
Marder, A. J., “The Royal Navy and the Ethiopian crisis of 19 3 5–3 6,”
American Historical Review, vol. 75, 1970, pp. 1327–56.
Marks, Sally, “German–American relations, 1918–1921,” Mid-America, vol. 53,
1971, pp. 211–26.
—— “Mussolini and Locarno: Fascist foreign policy in microcosm,” Journal of
Contemporary History, vol. 14, 1979, pp. 423–39.
—— “Ménage à trois: the negotiations for an Anglo-French-Belgian alliance in
1922,” International History Review, vol. 4, 1982, pp. 524–52.
—— “Mésentente cordiale,” in Marta Petricioli (ed.), A Missed Opportunity? 1922:
The Reconstruction of Europe, Bern, 1995, pp. 33–45.
—— “Smoke and mirrors: in smoke-filled rooms and the Galerie des Glaces,”
in Manfred F. Boemeke et al. (eds), The Treaty of Versailles: A Reassessment
after Seventy-Five Years, Cambridge, 1998, pp. 337–70.
Martel, Gordon, “The meaning of power: rethinking the decline and fall of
Great Britain,” International History Review, vol. 13, 1991, pp. 662–94.
—— “The prehistory of appeasement: Headlam-Morley the peace settlement
and revisionism,” Diplomacy and Statecraft, vol. 9, 1998, pp. 242–65.
—— “Reflections on the war-guilt question and the settlement: A comment,” in
Manfred F. Boemeke et al. (eds), The Treaty of Versailles: A Reassessment after
Seventy-Five Years, Cambridge, 1998, pp. 615–36.
—— “Military planning and the origins of the Second World War,” in Brian
McKercher and Michael Hennessy (eds), Military Planning and the Origins of
the Second World War, Westport, CT, 1999, pp. 1–28.
Mason, T. M., “Some origins of the Second World War,” Past & Present, no.
29, 1964, pp. 67–87.
May, Ernest R., “Military and naval affairs since 1900,” Warren I. Cohen
(ed.), New Frontiers in American–East Asian Relations: Essays Presented to Dorothy
Borg, New York, 1983, pp. 107–28.
McCormack, Gavan, “Nineteen-thirties Japan: fascism?’ Bulletin of Concerned
Asian Scholars, vol. 14, 1982, pp. 20–33.
McKercher, Brian, “Reaching for the brass ring,” Diplomatic History, vol. 15,
1991, pp. 565–98.
Meyers, R., “Britain, Europe and the dominions in the 1930s,” Australian
Journal of Politics and History, vol. 22, 1976, pp. 36–50.
Michaelis, M., “World power status or world dominion?’ Historical Journal, vol.
15, 1972, pp. 345–59.
Mierzejewski, Alfred C., “Payments and profits: the German National Railway
Company and reparations, 192 4–1932,” German Studies Review, vol. 18,
1995, pp. 65–86.
Minardi, Salvatore, “Mussolini, Laval e il désistement della Francia in
Etiopia,” Clio, vol. 22, 1986, pp. 77–107.
268 Bibliography
Cadogan, Alexander 192, 195 Daladier, Edouard 77–83, 86–7, 117, 194,
Carley, Michael J. 85, 144 216
Carr, E. H. 116 D’Annunzio, Gabriele 58–9
Cavallero, General Ugo 68–9 Danzig crisis 96, 104–5, 118, 126, 228–9,
Chabod, Federico 58, 71 (n.9) 232–3, 235–6, 239, 242–3; Taylor on
Chamberlain, Austin 22, 24, 27–8, 49, 59 6, 48, 111, 226–7, 240, 245
Chamberlain, Neville 1, 41, 66–7, 75, 81, DAP [see Deutsche Arbeiter Partei]
83, 86–7, 117–23, 126–9, 136–7, 182, Darwinism, social 63–5, 70–1
191–2, 194, 206, 214–15, 217, 229– Dawes plan 43, 46, 135
38, 241, 243; Taylor on 3–4, 6–7, 10, De Felice, Renzo 60, 65 73 (n.55)
75, 79, 81, 117–20, 123, 128, 136, 230 Delbos, Yvon 208, 215–7
Charmley, John 120 depression, economic 27–8, 41, 49, 61,
Chatfield, Lord 236 82, 164; Taylor on 4
Cherry Society 161 Deutsche Arbeiter Partei 94
Chiefs of Staff 90 (n.7), 125–9, 231–2, Deuxiéme Bureau 195
236, 243 Dilks, David 120
Chomsky, Noam 172–3 disarmament 10, 79, 89, 95; conference
Churchill, Winston 4–6, 49, 57, 69, 120– 41, 49, 124; of Germany 26, 47–9, 107
1, 191–2 Dollfuss, Engelbert 61–2, 179, 181
Ciano, Count Galeazzo 66, 68, 100, 181– Dore, Ronald 160
2, 192, 212 Doughty, Colonel Robert 87
Cienciala, Anna 234 Dreifort, John E. 87–8, 207
Cole, Robert 1, 32 Drexler, Anton 94
collective security 10, 136–50, 236 Duroselle, Jean-Baptiste 55 (n.41), 84–7
Colvin, Ian 233–4, 238 Duus, Peter 165
Comintern [see also Anti-Comintern
Pact] 140–1 Easter Accords 66–7, 74 (n.64), 182, 192
communism 1–2, 5, 10, 14, 18, 22–3, 64, Eden, Anthony 6, 63, 119, 179–80, 191–
98–9, 103, 105–6, 128, 140–1, 143, 2, 196, 214–15
145, 149, 156, 159, 166, 168, 170, Edwards, Jill 207
182, 193, 208–10, 212–14, 220, 222 Emmanuel III, King 68, 181
(n.21), 224 (n.75), 233; Taylor on 31, Engel, Gerhard 97
40, 135–6 Ethiopian crisis and war 61–4, 70, 99,
Communist Party: Britain 40; Soviet 108, 121, 123–4, 178–84, 186–99,
Union 140–1, 145, 149; France 210 120, 122, Taylor on 4, 124, 182
Conservative party 119–21, 191 Eupen—Malmédy 19, 27
continuity in German foreign policy 4–5,
32, 49, 55 (n.42), 151 (n.11), fascism 1, 57–61, 63–4, 70, 72 (n.22),
Cooper, Duff 129 141, 157, 164–6, 172, 184, 198–9, 206,
Corfu crisis 59, 70, 193–4 210; Taylor on 10, 22–3, 39, 76
Cot, Pierre 79 Faupel, General Wilhelm von 218–19
Cowling, Maurice 120 Fischer, Fritz 19, 44, 57
Crispi, Francesco 8, 58, 65 Flandin, Pierre Etienne 79–82
Index 275
Fleischhauer, Ingeborg 142, 146, 148 240–5; and Mussolini 59–63, 65–70,
Fletcher, Miles 159 74 (n.77), 99, 179–80, 182–96, 198–
Four Power Declaration 231–3, 235 99, 218–20; A.J.P. Taylor on 1–10, 13,
Four-Power pact 61–2, 136, 183 16–17, 32, 39–41, 49, 75–8, 81, 89,
Four-Year Plan 109–10 93–6, 98, 103–6, 111, 117–18, 120–1,
Franco, Francisco 4, 6, 99, 181, 206–14, 128–9, 137–9, 184, 206, 226–30, 238,
218–19, 240–5; and Poland 32, 97, 102, 227–
Franco-Soviet agreement 136 35; and France 86–9; and Japan 99–
Frank, Robert 86 100; and bolshevism 99–100, 103,
Fritsch, General Werner Freiherr von 101 106, 108, 139–43, 145–6, 148, 179
Funk, Walther 101 Ho Chi Minh 169
Hoare, Samuel 4, 63, 119
Gamelin, General Maurice 79, 83–4, 87 Hoare—Laval pact 63, 80, 118, 121
Gatzke, Hans 44 Hochman, Jiri 142–3, 147–8
Gauché, Colonel Maurice 79 Holocaust 112, 119
Gentleman’s Agreements 63, 65–6, 182, Hossbach memorandum 3, 67, 103–4,
215 111
Gilbert, Martin 119 Hugenberg, Alfred 94
Giral, José 207–8 Hull, Cordell 214–15
Girault, René 84
Gnedin, Evgenii 141–2, 151 (n.23, 24) Iakovlev, Aleksandr 146
Goebbels, Joseph 104–5, 108 Ickes, Harold 215
Gorbachev, Mikhail 145 ideology 1, 23, 32, 47, 57, 64–5, 70, 73
Göring, Hermann 59, 96–102, 104, 109– n.44, 84–5, 88, 96, 101, 103, 105–6,
10, 146, 181, 210–11 124, 128, 136, 155 (n.61), 157–9, 161,
Gott, Richard 119 164–5, 169, 179, 197, 205, 207–10,
Grandi, Dino 68, 186, 191–4 212–14, 217, 220, 233, 237; Taylor on
Grew, Joseph 171 3–4, 6–8, 75, 93–5, 118, 135, 138,
Gromyko, Andrei 139, 146, 153 (n.44) 150, 243
Gunsburg, Jeffrey 87 Ienaga Saburô 172, 177 (n.35)
imperialism 6, 24, 58, 94–5, 103, 156–7,
Halifax, Lord 66, 119, 229–31, 233–8, 168–71, 173, 175, 176 (n.4), 197
240, 246 (n.53) Imperial Rule Assistance Association 162
Halliday, Jon 174 inflation 24, 26, 43, 102, 125–6
Hankey, Maurice 49 Inoue Junnosuke 161
Hardie, Frank 124 Inukai Tsuyoshi 161
Harper, Glenn 211 Iriye, Akira 1
Haslam, Jonathan 143–6, 148–9 Irvine, William 88
Havens, Thomas 160, 162 Italian-Yugoslav accord 182
Hayashi Fusao 173
Henderson, Arthur 49 Jackson, Peter 88
Henderson, Nevile 119, 229 Jacomet, Robert 79
Herriot, Edouard 26 Jiang Jieshi 169
Hildebrand, Klaus 98, 101, 104 Jordan, Nicole 85
Hindenburg, Paul von 5, 96 Jukes, Geoff 148
Hitler 28, 35, 55 (n.45), 57, 60, 206–7,
213, 220, 231; Overy on 94–111; Kaganovich, Lazar 144
Sullivan on 179–80; Habeck on 211– Kandelaki, David 140–2, 146–8
12, 219–20; Greenwood on 233, 235, Karakhan, Lev 140
276 Index
Nicolson, Harold 195, 213 revisionism 24, 42, 146; German 17, 23,
Nish, Ian 166 27, 49, 94–6, 100, 104; Italian 23, 60;
Noel-Baker, Philip 213 of A. J. P. Taylor 6, 58, 93, 117
Non-Intervention: agreement 207–10, Reynaud, Paul 79
215–6, 219; committee 207, 209, 214, Reynolds, David 123
217 Rhineland 48, 205, 219; occupation of
Nyon conference 182, 217 18, 21, 27–8, 44, 47; evacuation of
31–2, 41, 47, 95; remilitarization of 64,
Okamoto, Daniel 165 76, 80, 178–79, 183, 190
Overy, Richard 1 Ribbentrop, Joachim von 68, 96–7, 99–
102, 112 (n.18, 21), 187, 223 (n.94),
pacifism 49, 86, 121 226, 228, 240–1
Pan-German idea 61, 64, 95 Rice, Richard 162
Paul-Boncour, Joseph 79, 216 Roberts, Geoffrey 144, 146–49, 234
Pact of Steel 67–9, 178, 183, 186, 192 Robertson, E.M. 39
Parker, R. A. C. 121, 229 Rome-Berlin Axis 57, 65–7, 69–70, 100,
Peace Ballot 121 178, 181,183–4, 192, 194, 196, 217–19
Pétain, Marshal Philippe 85 Rome protocols 179
Phillips, Hugh 149, 153 (n.58) Roosevelt, Franklin D. 3–4, 217
Poincaré, Raymond 26, 28–9, 31, 35 Rosenberg, Alfred 96
Royal Air Force 125
(n.54), 40, 43
Royal Navy 126, 180, 191
Polish corridor 20, 23, 32, 47–8, 68, 96,
Ruhr, occupation of 26, 29, 31, 35 (n.54),
104–5, 225, 227, 241
40–1, 43, 46, 59
Politburo 144, 146, 148, 210
Russo-German non—aggression treaty
Popular Front 65, 87, 99, 141,190, 192,
[see Nazi—Soviet pact]
206–09
Prague crisis and occupation of 80, 119,
sanctions 26, 62–4, 180–1, 184–5, 190–1,
122, 129 137, 229–30, 234–5, 237–9,
197
242–4
Schleicher, Kurt von 49, 55 (n.42), 195
public opinion 16, 31, 64, 86, 118, 121–2,
Schacht, Hjalmar 96, 100–2, 140, 146,
129, 191, 214, 243 148
Schuschnigg, Kurt von 5, 7, 181, 190, 218
Raack, R. C. 140–1 Selassie, Haile 4, 180
Radek, Karl 141–2 Semeriaga, Mikhail 145
Ragsdale, Hugh 148, 153 (n.52) Seydoux, Jacques 46
Rapallo treaty 15, 21, 23, 31, 136, 140, Sheinis, Zinovy 148
146 Shillony, Ben-Ami 160, 162, 165
Raymond, Paul 144, 148 Shorrock, William I. 87
rearmament: of Britain 121, 125–6; of Simon, John 119
France 83, 86–8 of Germany 30, 49, Sisman, Adam 1
62, 96, 99, 101, 107–8, 110–11, 116, Smethurst, Thomas 162
137, 180, 183–4, 211 South Tyrol [see Alto Adige]
Red Army 31, 137–8, 140, 148 Spanish civil war 64, 99, 108, 121, 123,
Reichswehr 49, 55 (n.45), 59, 149 179, 182, 184–5, 188, 192, 206–7, 209,
reparations 19, 25–6, 28, 30, 45–6, 52 211–15, 217–20; non-intervention in
(n.19), 89, 95; Taylor on 14–6, 22, 79, 209–12, 214, 216–18, 221; Taylor
24–5, 30–1, 35 (n.47, 48), 40–7, 49 on 5, 7, 205, 218
Republicans, Spanish 190, 204–18 Speer, Albert 96, 104
278 Index