Roosevelt and the Munich Crisis: Insights from Prospect Theory
Author(s): Barbara Farnham
Source: Political Psychology, Vol. 13, No. 2, Special Issue: Prospect Theory and Political
Psychology (Jun., 1992), pp. 205-235
Published by: International Society of Political Psychology
Stable URL: https://www.jstor.org/stable/3791679
Accessed: 18-04-2019 14:35 UTC
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide
range of content in a trusted digital archive. We use information technology and tools to increase productivity and
facilitate new forms of scholarship. For more information about JSTOR, please contact support@jstor.org.
Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at
https://about.jstor.org/terms
International Society of Political Psychology is collaborating with JSTOR to digitize,
preserve and extend access to Political Psychology
This content downloaded from 147.251.120.50 on Thu, 18 Apr 2019 14:35:43 UTC
All use subject to https://about.jstor.org/terms
Political Psychology, Vol. 13, No. 2, 1992
Roosevelt and the Munich Crisis: Insights from
Prospect Theory
Barbara Farnham'
The use of prospect theory to explain political decision-making challenge
claim of rational choice theory to provide a more convincing account of
behavior. Because President Roosevelt's decision-making during the M
crisis manifests a number of phenomena associated with prospect theory, i
ing a change in the decision frame and corresponding preference revers
acceptance to avoid loss, and the operation of certainty effects, it affor
opportunity to view these competing claims in the light of an actual his
case. An analysis of Roosevelt's decision-making behavior during the
shows that prospect theory does in fact explain it more satisfactorily tha
the theory of rational choice. In addition, the analysis suggests that affec
sometimes play a role in causing decision frames to change and raises qu
about Janis and Mann's account of the impact of stress on decision-maki
KEY WORDS: decision-making; prospect theory; rational choice; decisional conflict
Munich crisis; Roosevelt.
INTRODUCTION
Recently, there have been a number of attempts to use prospect
explain political decision-making behavior (e.g., Quattrone & Tversky
1988). Because such efforts challenge the claim of theories of rationa
provide a more convincing account of similar phenomena (e.g., Buen
quita, 1981), an attempt to examine these competing models in the l
historical case should have considerable theoretical interest.
President Franklin D. Roosevelt's decision-making behavior during the
Munich crisis in September 1938 provides a suitable vehicle for such an explora'Institute
of War and Peace Studies, Columbia University, New York, 10027.
205
0162-895X/92/0600-0205$06.50/1 ? 1992 International Society of Political Psychology
This content downloaded from 147.251.120.50 on Thu, 18 Apr 2019 14:35:43 UTC
All use subject to https://about.jstor.org/terms
206 Farnham
tion because i
pect theory,
sponding pre
certainty eff
conviction th
not only desi
at least two r
be ineffectiv
have unfortu
What is mos
seems to hav
crisis to hims
implications
pean crisis, ev
the United St
did pose such
objective situ
come of war a
take action to avoid it.
This behavior is puzzling from a rational choice perspective because a
preference reversal based on a different representation of the same decision
problem clearly violates the expectations of utility theory (Tversky & Kahneman
1986, p. S523). Prospect theory, on the other hand, by providing "an alternative
account of choice under risk" which holds that people are frequently less analytical
in making their decisions than expected-utility theory would predict (Kahne
man & Tversky, 1979, p. 263), offers the possibility of explaining such behavior
In addition to illuminating an important episode in American foreign policy,
however, the application of prospect theory to the Munich case offers at least
three theoretical benefits. For one thing, while several of the predictions of
prospect theory have received experimental support, there has been relatively
little work demonstrating their applicability to political decision-making. Direct
evidence of the ability of prospect theory to explain behavior in an actual politica
decision-making situation, therefore, could increase confidence in its ecologica
validity and extend its range (For experimental evidence in general, in addition to
the work of Tversky and Kahneman cited below, see Davis & Bobko, 1986, pp.
125-126; Bazerman, 1983, p. 214; Levin et al., 1985, p. 372; and Shapira,
1981, pp. 334, 344-348, 250. For attempts to use prospect theory to illuminate
2According to Tversky and Kahneman a decision frame includes the decision-maker's conception of
"the acts or options among which [he or she] must choose, the possible outcomes or consequences
of these acts, and the contingencies or conditional probabilities associated with a particular choice."
Taken together, these elements serve to define the decision problem (Tversky & Kahneman, 1981, p.
453).
This content downloaded from 147.251.120.50 on Thu, 18 Apr 2019 14:35:43 UTC
All use subject to https://about.jstor.org/terms
Roosevelt and the Munich Crisis 207
political judgment, see Quattr
relate psychological theories to
see Tetlock, 1985; and Farnham
A second benefit of using pros
in suggesting a possible cause of
an elaboration of the theory its
little on this subject. Stating onl
controlled partly by the formul
habits, and personal characterist
neither the issue of how a decisio
question of what factors might
(Tversky & Kahneman, 1981, p.
intriguing possibility that at least
Apparently, that is, President R
of direct concern to the United
become emotionally compeling
Finally, because the Munich c
for Roosevelt, it offers an oppo
(1977) decisional conflict model
many of the conditions under w
do so raises a question about t
making behavior in political set
In the discussion which follow
during the Munich crisis, an att
derived from both prospect theo
tions of this analysis for prosp
general, will then be explored.
THE COURSE OF THE MUNICH CRISIS
The Munich crisis began in mid-September 1938 as Great B
France sought to cope with German pressure on Czechoslovakia wh
war and preserving at least the appearance of honoring their commit
latter. The crisis was resolved, however temporarily, with the sign
Munich agreement on September 30.
The proximate cause of the crisis was the escalation of German
Adolf Hitler's claims on Czechoslovakia from autonomy for the Su
mans, to which the Czechs had already agreed, to outright ces
Sudetenland to Germany. Rather than give in to such pressure, Cz
turned to France and Britain for help. As a consequence of this ap
Chamberlain, the British prime minister, offered to conduct direct
This content downloaded from 147.251.120.50 on Thu, 18 Apr 2019 14:35:43 UTC
All use subject to https://about.jstor.org/terms
208 Farnham
with Hitler (b
pp. 7-11; Of
20-21).
On Septemb
demand for t
undertook to
ultimatum. H
esberg on Sep
the cession of
transfers be
and French p
Chamberlain
surrender co
At this poin
the states in
Hitler alone p
include all in
issued invita
September 2
The outcome
of appeaseme
territory in
Sudeten area
of Septembe
demand for
way to give H
for it, the cr
ROOSEVELT'S RESPONSE TO THE MUNICH CRISIS
From the perspective of Roosevelt's decision-making behavior, the Munic
crisis can be divided into two stages, with the Godesberg meeting serving as
line of demarcation. The first stage, which was characterized by his marked
disinclination to intervene in the crisis, extended from September 13 to Sep
tember 22, coming to a climax with his conversation with British Ambassad
Sir Ronald Lindsay on September 19. The second stage began on September 2
when the initial reports of the Godesberg ultimatum reached the president, a
ended with the signing of the Munich Agreement on September 30. From a
American perspective, the high point of this stage was Roosevelt's dramatic
change of policy about intervention in the crisis culminating in his two appe
on September 25 and 27.
This content downloaded from 147.251.120.50 on Thu, 18 Apr 2019 14:35:43 UTC
All use subject to https://about.jstor.org/terms
Roosevelt and the Munich Crisis 209
Stage I: September 13-Sep
During the first phase of the cri
of general war fluctuated with the
mination not to intervene, howeve
At the outset, both the presiden
preferences mirrored those of the
climate in Central Europe in the
concern about the possibility of wa
the major players owing to contrad
views of members of his administr
resolve of the British and French a
United States.3
The mixture of uncertainty and
diagnosis of the Czech situation in t
determination not to involve the U
racies, and frustrated by his inabil
them, Roosevelt refused to join ope
threat, notwithstanding the determi
like Ambassador Bullitt, to persua
The onset of the crisis itself bro
policy preferences. While his assess
stage of the crisis varied according
avoid intervening did not.
Initially, Roosevelt believed that
French and British would not resis
he advised Secretary of the Interio
3For a sample of reports from abroad du
Schewe, 1979, #1180; "Kennedy to Hull,
Foreign Relations of the United States, (
Bullitt, Ambassador to France, to Rooseve
to Cordell Hull, Secretary of State, Augu
sador to Italy, to Roosevelt, September 1
1938," FRUS, 1938, vol. I, p. 589; and "Ke
584-585. The views of various members of
from Adolf Berle to the President, Septe
1965, pp. 452, 457; Ickes, 1954, pp. 381-3
the British were trying to manipulate th
1938, vol. 138, pp. 33-55.
4See, for example, Bullitt's letter of May
Roosevelt's reluctance to support the dem
September 9, 1938, 5:30 P.M.," Schewe, 1
518; Langer and Gleason, 1952, pp. 32-33;
of Roosevelt's attitudes and policy preferen
pp. 385-398.
This content downloaded from 147.251.120.50 on Thu, 18 Apr 2019 14:35:43 UTC
All use subject to https://about.jstor.org/terms
210 Farnham
Chamberlain
sure from E
move in. . . .
Czechoslovak
mean a swif
prostrate. .
Judas Iscario
Given the
that there
he agreed
president c
democracie
to transpor
pp. 69, 115
September
tion betwe
599]. As As
was reaffi
Between S
Berchtesga
morning o
demands w
tary Morge
Secretary I
prepondera
(Ickes, 195
Berchtesga
17 account
on a world
mans [Kenn
Other repo
Kennedy t
to Hull, Se
18, 1938, I
munication
acceded to
Convinced that a defensive war based on a blockade offered Britain and
France their best chance of vanquishing the Germans, Roosevelt discussed with
Morgenthau how best to get this idea across to them. Because of his reservations
about the ability of Ambassadors Kennedy and Bullitt to convey his meaning
effectively, the president decided to do the job himself, making an appointment
with the British ambassador for that evening (Morgenthau Diary, 1938, vol. 141,
p. 115. Also in Blum, 1965, pp. 519-520).
This content downloaded from 147.251.120.50 on Thu, 18 Apr 2019 14:35:43 UTC
All use subject to https://about.jstor.org/terms
Roosevelt and the Munich Crisis 211
According to Ambassador Lindsay
declared that while he would like to b
no illusions as the effect of his previo
make any pronouncements about the p
Today he would not dare to express approval o
slovak Government. He would [also] be afraid
sion lest it might encourage Czechoslovakia to
anything and thought at his press conference
would confine himself to refusing to make an
What the president really wanted t
proper strategy for conducting the w
soon be forced to fight. Declaring t
beaten if they tried to wage war on c
mended that they should conduct the w
manner." He also indicated that, within
would give the democracies as much h
Clearly then, while the news from B
velt to take some sort of action, what is
chose. Roosevelt's heightened expectati
it, nor to alter his policy of noninter
responded to the deteriorating situatio
courage of the democracies in the pres
struggle. (That this was Roosevelt's
Dallek's view [1979, pp. 164-165].)
After the president's meeting with
several days of waiting. Ultimately, th
their position, "accepted flatly and un
posal," acceding to virtually all of Hit
tion of war had been based on the b
development reduced his estimate of it
Prime Minister Chamberlain and Hitle
matter of arranging the details (Bu
1938, vol. I, pp. 630-631. According to
5"Sir R. Lindsay (Washington) to Viscount Ha
and Woodward, 1951, pp. 627-629. On the m
Taylor, 1979, pp. 846-848; Lash, 1976, pp. 2
Other evidence of Roosevelt's disinclination
Berchtesgaden is found in a memorandum of
September 18 or 19 in which he "pointed out
initiative from him. Such an initiative, if not ac
was" ("Memorandum by J. Pierrepont Moffat, C
State, September 20, 1938," Schewe, 1979, #1
Hooker, 1956, pp. 206-207).
This content downloaded from 147.251.120.50 on Thu, 18 Apr 2019 14:35:43 UTC
All use subject to https://about.jstor.org/terms
212 Farnham
the Czech deci
the German a
However dram
may appear, th
he had been f
his assessment
vene openly in
Stage II: Se
For President
earnest on Sep
and Chamber
increased dram
detailed accou
819).
Ickes has described Roosevelt's initial reaction to the news from Godesberg
in his account of a cabinet meeting on the afternoon of September 23 at which the
European situation was "canvassed very fully" (Ickes, 1954, pp. 473-474).
Opening the meeting with a worst-case analysis of the possible consequences of
the latest developments, Secretary Hull declared that there were "undoubtedly"
defensive alliances between Italy, Germany, and Japan, and Japan would support
Germany and Italy in their bid to dominate Europe while itself trying to gain
complete control over Asia. Consequently, in Ickes' words,
... France may soon find itself to be a helpless country lying between an enlarged and
strengthened Germany and Italy. England might even be reduced to the status of a third- or
fourth-rate power with many of her colonies gone. If this should happen, there might
follow attempts on the part of Germany, Italy, and Japan to penetrate South America. This
would mean that the United States would have to go to the defense of South America, in
which event we would be called upon to defend both the Atlantic and the Pacific seaboards
from powerful enemies. (Ickes, 1954, p. 473-474)
Roosevelt, on the other hand, in line with his previously expressed views,
was considerably less pessimistic about the democracies' chances, repeating that
they should fight a strictly defensive war, and declaring that the British and
French would control both the Atlantic Ocean and the Mediterranean and would
thus be able to bottle up the Germans and the Italians. For the first time,
however, he did express some concern about German preponderance in the air,
labeling it "the worst thing about the situation." The meeting closed with the
president reading from an analysis of the situation from Ambassador Bullitt just
handed to him, which predicted "that Germany might be crossing the [Czech]
frontier this morning and that Poland might go across even before morning"
This content downloaded from 147.251.120.50 on Thu, 18 Apr 2019 14:35:43 UTC
All use subject to https://about.jstor.org/terms
Roosevelt and the Munich Crisis 213
(Ickes, 1954, p. 473-474; "Memo
Schewe, #1294).
Clearly, then, the news of Hitler's unreasonable demands at Godesberg
produced an immediate upward revision in Roosevelt's estimate of the likelihood
of war in Europe. It did not, however, lead to a change in policy. Although war
now seemed imminent, the president still had no thought of intervening to
prevent it. According to Ickes' account, as late as the afternoon of September 23,
there was "no doubt of the President's desire to avoid any embroilment in
European quarrels."6 That he had decided to intervene in the crisis by at least the
afternoon of September 25 is thus a puzzle which must be solved.
The First Intervention
Throughout the day on September 24, bad news poured into the White House.
Despite the six-day respite Hitler had generously granted the Czechs, the situation
remained menacing.' As Ickes reported in his diary, much of Europe was gearing
up for war (Ickes, 1954, p. 473. See also Czechoslovak Legation to the Department
of State, September 24, 1938, FRUS, vol. I, pp. 645-646; Haight, 1960, p.
352).
Moreover, the diplomatic traffic, in emphasizing Hitler's total unreasonableness,
left no doubt that responsibility for war would rest solely with him. For
example, Kennedy cabled that the British
have just received Hitler's answers and they are . . . preposterous. Hitler not only wants
what everybody was willing to give him but it looks as if he wants a great deal more. ...
Cadogan feels that . . . they have taken every possible opportunity to demonstrate they
believed there was some sanity in Hitler and to save the world from the horrible results of
war. . . . Hitler's answers prove there is no sanity left in the man. (Kennedy to Hull, Sep-
6In fact, the message from Bullitt, which Roosevelt read to the Cabinet, had suggested that he should
now act directly with respect to the crisis and referred obliquely to the Ambassador's previous idea
of calling an international conference. Roosevelt, however, apparently made no mention of this
proposal to his Cabinet. That immediate action was not at the forefront of his thinking on September
23 is also shown by his negative reply to Nicholas Murray Butler's suggestion that he make an
appeal to the parties ("Memorandum to Roosevelt, September 23, 1938," "Roosevelt to Nicholas
Murray Butler, President, Columbia University, September 23, 1938," Schewe, 1979, #1294,
1295; Ickes, 1954, pp. 473-474.).
70n September 24 Ickes reported that, while "the situation still looks threatening," it was "not as
critical as the message read by the President yesterday indicated," because "Hitler has given six
days for acceptance of his terms by Czechoslovakia." (Ickes, 1954, p. 473). Roosevelt himself had
this news by the afternoon of September 23, though not at the time of the Cabinet meeting where he
was nevertheless still clearly disinclined to intervene. Moreover, since the clock continued to run
and all the news from Europe was bad, it is doubtful that Hitler's gesture greatly reduced the pressure
("Memorandum by Henry M. Kannee, Assistant to Marvin H. Mc Intyre, September 23, 1938,"
and "Memorandum to Roosevelt, September 23, 1938," Schewe, 1979, #1293, #1294. Note that
Schewe reverses the order of these two memoranda. However, it is clear that # 1294 was received in
time to be read to the Cabinet which met at 2 p.m., whereas # 1293 bears the time 4.50 P.).
This content downloaded from 147.251.120.50 on Thu, 18 Apr 2019 14:35:43 UTC
All use subject to https://about.jstor.org/terms
214 Farnham
tember 24, 1
Carr to Hull
Meanwhile
desire "som
be a failure
Hague. Bull
such a conf
(Bullitt to
Hull, 1948
Despite suc
24. Apparen
some kind o
the impuls
Hooker, 19
p. 9).
By September 25 war clearly seemed imminent, as throughout the day Roosevelt
continued to receive reports about both the draconian terms Hitler sought to
impose and their complete unacceptability to any of the other parties, which made
war a virtual certainty. There was, for example, a "flood" of telegrams from
Ambassador Bullitt detailing Hitler's demands and emphasizing their extraordinary
harshness. According to Bullitt, the German note to Chamberlain was "totally
unacceptable. The terms asked by Hitler are virtually those imposed on a defeated
German Army for evacuation of northern France" (Haight, 1960, p. 352; Bullitt
to Hull, September 25, 1938, FRUS, 1938, vol. I, pp. 648-649, received 11:35
a.m. See also Bullitt to Hull, September 25, 1938, Ibid., pp. 646-648, received
9:15 a.m.; and Bullitt to Hull, September 25, 1938, Ibid., pp. 650-652, received
12:25 p.m.).8
While these alarming reports poured in, the planning for an American response
initiated in the State Department continued. Berle and Moffat had volunteered
to produce the draft of a presidential statement and, after considering and
ruling out Bullitt's conference plan, settled on a message which would be, as Berle
desired, "not merely an appeal but a definite suggestion that we would use our
good offices in a draft leading to the revision of the Versailles Treaty." Hull, while
disapproving of the idea of treaty revision as "too dangerous," took the draft
8Bullitt's reports were supplemented by a cable from Ambassador Carr transmitting the Czech
president's plea to prevent "the assassination of the state" by urging the British and the French not to
desert Czechoslovakia. Moreover, the American ambassador to Germany, Horace Wilson, telegraphed
later in the day to convey the opinion of the British Ambassador to Germany that unless the
British and French managed to pressure the Czechs into accepting the German troops, there would
almost certainly be war (Carr to Hull, September 25, 1938, Ibid., pp. 649-650, received at 10:20
a.m.; and Wilson to Hull, September 25, 1938, Ibid., pp. 654-56, received 7 p.m.. See also, Hull,
1948, pp. 590-591).
This content downloaded from 147.251.120.50 on Thu, 18 Apr 2019 14:35:43 UTC
All use subject to https://about.jstor.org/terms
Roosevelt and the Munich Crisis 215
statement to the president at si
211-212; Offner, 1969, pp. 262-2
Moffat diary]; Haight, 1960, p.
Clearly by this point Roosevelt w
influence the manner in which a
totally supplanted by a determina
focused on the imminence of war and the need to end the crisis before it could
occur (Hull informed Moffat on the night of the 25th that Roosevelt had gotten
such bad telephone reports from both Bullitt and Kennedy that he could no longer
keep silent [Hooker, 1956, pp. 212-213; Hull, 1948, pp. 591-592]. On Roosevelt's
anxiety to stop the war by this time, see also Morgenthau Diary, vol. 142, p.
342).
Thus, the drafting of the appeal went forward. Bullitt had suggested that it
should include an offer by the president to arbitrate. Hull disliked this idea. (In fact
he opposed sending any appeal whatever.) Along with Normal Davis, he was also
against the somewhat less extreme step of a tender of good offices (according to
Hull, he kept urging the president to go slow, while Welles egged him on [Hull,
1948, pp. 591-592. Hooker, 1956, pp. 212-213; Berle, 1973, pp. 186-187]).
In the end, Roosevelt acceded to these objections, and any hint of mediation
was removed from the appeal. According to Moffat, this was done both because of
his fear of "untoward domestic effects," and because he believed the notion of
good offices to be implicit in the appeal in any case (Berle, 1973, p. 187; Hull,
1948, p. 592; Hooker, 1956, pp. 212-213; Offner, 1969, p. 263).
The president finished revising the final draft by midnight, and it was sent to
Hitler, Chamberlain, Daladier, and Czechoslovakian President Edvard Benes at I
a.m. on September 26. In the message, Roosevelt declared that the "fabric of
peace . . . is in immediate danger," and stressed that, while the United States
eschewed "political entanglements," it could not escape the consequences of a
general war. He, therefore, reminded the parties of their obligation to settle their
differences peacefully, and called on them not to break off negotiations (Alsop and
Kintner, 1940, p. 10; Offner, 1969, p. 263; Roosevelt to Hitler, September 26,
1938, FRUS, 1938, vol. I, pp. 657-658. The responses to Roosevelt's appeal may
be found in Ibid., pp. 661-673).
Exactly what did Roosevelt hope to accomplish with this message? Hull, who
had serious reservations about the efficacy of such an appeal, has stated that Roosevelt
was motivated by the feeling that nothing could be worse than inaction:
"The President . . . believed with Bullitt that something should be done, even if it
were not successful. He said to me: 'It can't do any harm. It's safe to urge peace
until the last moment.' "
Welles, on the other hand, told the French ambassador that the message reflected
Roosevelt's attempt to balance his desire to affect the course of the crisis
with his concern about possible domestic repercussions. In Haight's words, Welles
This content downloaded from 147.251.120.50 on Thu, 18 Apr 2019 14:35:43 UTC
All use subject to https://about.jstor.org/terms
216 Farnham
gave Ambas
aimed at br
and at the
p. 591; Hai
cautious ab
Chamberlai
(memorand
26, 1938, 1
593).
In any case, the day following the president's appeal was marked by considerable
anxiety. According to Hull, they "waited almost breathlessly" for the replies,
especially Hitler's.
Unfortunately, no word from the Fuehrer was immediately forthcoming.
However, Roosevelt received ample evidence of the democracies' continued resolve,
as well as numerous expressions of their gratitude for his message and
assurances about its positive influence which he found particularly gratifying
(Hull, 1948, p. 592. Memorandum of a phone conversation between Welles and
Kennedy, September 26, 1938, 1:30 p.m., Ibid., pp. 660-661. For accounts of
Allied determination, see Kennedy to Hull, September 26, 1938, FRUS, 1938,
vol. I, p. 659; Bullitt to Hull, September 26, 1938, Ibid., p. 668. For expressions
of appreciation see Kennedy to Hull, September 26, 1938, Ibid., p. 659, received
8:30 a.m.; Bullitt to Hull, September 26, 1938, Ibid., pp. 659-660, received 2
p.m.).
Hitler's answer to the president's message, when it finally arrived, was not
nearly so pleasing. In a speech heard in the United States on the afternoon of
September 26, the Fuehrer was far from conciliatory, declaring at one point "that if
the Czechs did not give the Sudeten Germans immediate freedom, 'we will go and
fetch this freedom for ourselves.' " As Ickes described it, Hitler "ranted and raved
for over an hour. At times he seemed to be almost incoherent. He shrieked his
defiance to the whole world, bragging of the prowess of Germany. . . . War
seemed to be inevitable, with every tick of the clock bringing it closer" (Offner,
1969, pp. 263-264; Ickes, 1954, p. 477). For an account of Hitler's intransigence
during this period see Taylor, 1979, pp. 870-875).
Moreover, Hitler's formal reply to Roosevelt's message offered no more solace
than his speech. Disclaiming all responsibility should further developments
lead to war, he recited at length German grievances against the Czechs and ended
with the ominous declaration that "the possibilities of arriving at a just settlement
by agreement, are . . . exhausted with the proposals of the German memorandum.
It does not rest with the German Government, but with the Czechoslovakian
Government alone, to decide, whether it wants peace or war." As the president
reported to his cabinet on the following day, the tone of Hitler's reply was "truculant
and unyielding" (Hitler to Roosevelt, September 26, 1938, FRUS, 1938, vol.
This content downloaded from 147.251.120.50 on Thu, 18 Apr 2019 14:35:43 UTC
All use subject to https://about.jstor.org/terms
Roosevelt and the Munich Crisis 217
I, pp. 669-672, received 9:14
form in the White House unti
State George Messersmith phon
of the 26th ["Adolf Hitler, Cha
1938," Schewe, 1979, #1302,
The Second Intervention
On the morning of the 27th, a group of State Department officials met to
consider an appropriate response to the latest developments. Berle and Welles
were deputized to draft a second message to Hitler, which they did, opting for the
path of "boldness" by including in it a call for a conference at The Hague. Just
before lunch, Hull and Welles took this draft statement to the President (Berle,
1973, p. 187).
By this time, still more evidence of Hitler's refusal to cooperate had arrived.
For example, both Kennedy and Bullitt cabled that the Fuehrer's reception of
Chamberlain's latest notes had been "completely and definitely unsatisfactory."
In Bullitt's words, Hitler's reply "was the most violent outburst possible; ...
nothing could have been more unhelpful." By way of contrast, moreover, this
demonstration of Hitler's intransigence was accompanied by a number of reports
of continued expressions of gratitude for the president's message from Great Britain
and France (Kennedy to Hull, September 27, 1938, FRUS, 1938, vol. I, p.
673, received 7:05 a.m.; Bullitt to Hull, September 27, 1938, Ibid., pp. 673-674,
received 9:15 a.m., and pp. 674-675, received 11:10 a.m.. A cable from Ambassador
Carr also underlined the unreasonableness of Hitler's demands from the
point of view of Czech military security Carr to Hull, September 27, 1938, Ibid
p. 679, received 1:20 p.m.).
The news of Hitler's obduracy and aggressive posturing had two implications
for Roosevelt's diagnosis of the crisis. First of all, it underlined th
Fuehrer's sole responsibility for the continued slide toward war. Secondly, it
increased expectations that war would actually occur unless something happened
to prevent it.9 Apparently the confluence of these two perceptions shaped Roos
velt's decision, at some point during the course of the morning of September 27
to take further action.
9Supporting the first point is Welles's statement to Bullitt that any action by the president would be
directed solely at Hitler (Memorandum of telephone call between Welles and Bullitt, September 27,
1938, 2:40 p.m., FRUS, 1938, vol. I, pp. 675-676). The second point is corroborated by Welles's
later statement that on September 27 " 'Information of unquestioned authenticity' had come through
that at 2:00 p.m. on the next day, September 28, Hitler would march his armies into Czechoslovakia
unless the Godesberg terms were met" (Haight, 1960, p. 355). This account was confirmed by
Roosevelt himself when he told Arthur Murray on October 14, 1938, that he had received news on
the 27th that Hitler would take action the next day (Murray, 1946, p. 95).
This content downloaded from 147.251.120.50 on Thu, 18 Apr 2019 14:35:43 UTC
All use subject to https://about.jstor.org/terms
218 Farnham
At the presid
discussed earli
support the A
Mussolini) wer
Roosevelt shou
"depressed" b
opposed it. In
Ultimately,
Daladier's and
approved, the
neutral Europe
be omitted (B
While Welles
message from
session. He had
and, as he told
When it was fini
four Members of
Monday, and so h
the shouting and
'Krieg, krieg,' an
p. 95. According
Murray himself.
Berle noticed lat
berlain's speech
For the rest
clusive subject
reply to his f
He then led a
Chamberlain's
alternatives. T
democracies w
mop up the It
Franco in Spai
England, Fra
other anti-Ger
vol. II, p. 481
confident that
Sometime dur
was held, Roos
possible. As h
I had intended to
Chamberlain's ra
This content downloaded from 147.251.120.50 on Thu, 18 Apr 2019 14:35:43 UTC
All use subject to https://about.jstor.org/terms
Roosevelt and the Munich Crisis 219
action at two o'clock on Wednesday aft
the draft of my message, and Hull cam
nine o'clock we had hammered out th
(Murray, 1946, p. 95. Apparently, Roos
Chamberlain and Daladier. In fact, by
Chamberlain's opinion, Welles was able t
message tonight without fail." [Mem
Kennedy and Welles, September 27,
Following his conversation wit
draft of the message to Hitler w
already approved. After dinner, t
work on this draft. There follow
ly depicted by Alsop and Kintne
smoking incessantly and shooting
near him, Berle fidgeting, Hull s
once almost out of countenance."
A moment of considerable tension occurred when "a report came in that the
Germans might march in the night, forcing a war to no purpose. For a moment
the President showed real anger." However, as the report was not confirmed, the
drafting continued. By 9 p.m. the draft was well enough in hand to allow the
secretary of state to go home to bed, and at 9:30 it was signed by the president
(Alsop and Kintner, 1940, pp. 10-11. See also Berle, 1973, p. 188, and Murray,
1946, p. 95).
Roosevelt's second message to Hitler was sent at 10:18 p.m. Recalling his
earlier emphasis on the peaceful settlement of disputes and the complete lack of
justification for any threat of force which might result in general warfare, the
president pointed out to the German chancellor that these considerations were
even more relevant now that agreement in principle had been reached between
Germany and Czechoslovakia. He therefore urged continuation of the negotiations,
raising the possibility of widening them to include "all the nations directly
interested in the present controversy." While reiterating that the United States
had "no political involvements in Europe," Roosevelt nevertheless closed by
stating that "the conscience and the impelling desire of the people of my country
demand that the voice of their government be raised again and yet again to avert
and to avoid war" (Roosevelt to Hitler, September 27, 1938, FRUS, 1938, vol. I,
pp. 684-685. Farnham [1991, pp. 431-435] analyzes several opposing views of
Roosevelt's intentions in sending the second appeal to Hitler and concludes that it
was a genuine attempt to prevent war by pressuring Hitler into peaceful settlement
of the dispute along the lines which had already been agreed upon.).
The period of anxious waiting for Hitler's reply following the dispatch of
this message ended on the morning of the 28th. As Berle dramatically recorded
in his diary, Hitler had invited Britain, France, and Italy to discuss the Czech
crisis ("The 'break'! Thank God."). Roosevelt was also relieved by the news and
This content downloaded from 147.251.120.50 on Thu, 18 Apr 2019 14:35:43 UTC
All use subject to https://about.jstor.org/terms
220 Farnham
at 1 p.m. s
"Good Man
FRUS, 1938
debated, m
uance of ne
One who d
lieves that,
Daladier as
tiate at Mu
p. 34; Off
1970, p. 2
Thus, altho
until Septe
announcem
the settlem
mained abo
ing over a p
war as the
widespread
over which
would give
EXPLAINING ROOSEVELT'S BEHAVIOR
During the first stage of the Munich crisis, Roosevelt's a
focused primarily on the general political and military problem
they related to American concerns and he seemed only minimally
the crisis per se, contemplating even an outcome of war with som
Moreover, although he was beginning to develop some notio
possible consequences for the United States, he still had no sense t
be directly involved in its resolution, viewing overt interventio
which involved considerable risk. Nor did Roosevelt alter hi
respect even after the news from Berchtesgaden had caused him
war was virtually inevitable.
After Godesberg Roosevelt's attitude changed, but not at on
although reports of that meeting on September 23 led him to belie
imminent, he still had no thought of intervening. Sometime dur
24, and almost certainly by the morning of the 25th, however,
underwent a dramatic change, and he decided to act directly. A
reported to his aides early on the morning of the 26th, the presiden
This content downloaded from 147.251.120.50 on Thu, 18 Apr 2019 14:35:43 UTC
All use subject to https://about.jstor.org/terms
Roosevelt and the Munich Crisis 221
become "very anxious to get in a
September 26, 1938, vol. 142,
Within less than two days, ther
from the general long-run impli
States and toward the crisis itse
panied by a change in policy fro
attendant risks notwithstanding.
was to advise the democracies abo
such considerations were overwhe
that had become the central prob
action on his part. Such a dramat
nation.
The Rational Choice Explanation
One interpretation of the president's preference reversal might be that it was
simply a rational response to new information which altered the subjective expected
utility of intervention. However, while one or more changes in the environment-in
the value of the outcome of war, for example, or its probability,
or the risks of intervention-could conceivably have triggered such a shift, there
were in fact no changes in any of these factors sufficient to justify Roosevelt's
reversal of preferences.
If, for example, Roosevelt's post-Godesberg belief that an outcome of war
would now be a loss had been based on new information about the costs of war to
the United States, it would have been rational for him to decide to intervene to
prevent war, even if doing so involved some risk. However, the crisis did not
provide any new information about the value of the war. No matter how certain
or imminent general war in Europe had become after Godesberg, it was in reality
no more of a threat to the United States than it had been when it was merely
probable. No one imagined for a moment that the United States would be directly
threatened by war, certainly not Roosevelt who firmly believed that Britain and
France would win. Yet despite his unaltered belief that the democracies would
emerge victorious, the president behaved as though the prospect of war had
become a direct threat to the United States. What is embarrassing to the theory of
rational choice is that after Godesberg Roosevelt redefined what was essentially
the same objective situation as a loss (The reports Roosevelt received emphasizing
the dreadful consequences of war for European civilization might possibly be
viewed as new information which changed his valuation of the outcome of war.
Such arguments were not new to Roosevelt, however, and he had ignored them
when they were made earlier [see, for example, Bullitt's letter of May 20, Bullitt,
1972, pp. 261-262].).
This content downloaded from 147.251.120.50 on Thu, 18 Apr 2019 14:35:43 UTC
All use subject to https://about.jstor.org/terms
222 Farnham
Not only w
have explain
perception
culations bo
utility in t
evaluated in
intervening
matters wo
French trad
tunate dom
With respe
opinion dur
crisis, he ac
steps to det
can people u
August 31,
what Amba
Pointe de G
concern abo
Hull told Bu
far as our
Roosevelt's
early Septem
intention
Schewe, 19
and Gleaso
Both befor
believed tha
such interv
frequently
activity wh
he was exce
public invol
sentiment i
activist cou
velt's habitu
sion of his
Brussels in
analysis of
opinion bef
At the outs
This content downloaded from 147.251.120.50 on Thu, 18 Apr 2019 14:35:43 UTC
All use subject to https://about.jstor.org/terms
Roosevelt and the Munich Crisis 223
might lead to failure, entail dom
ample support for the idea that
made the decision to intervene.
With respect to the risk of failure, nothing had occurred at Godesberg which
entitled Roosevelt to infer that he could now act more effectively than he might
have done earlier. On the one hand, while new evidence that Britain and France
were at last willing to resist Hitler might have made American action more
palatable to Roosevelt because he no longer needed to feel that the democracies
were trying to get the United States to do their dirty work for them, Hitler's
response to their firmness provided little reason to hope that United States intervention
could now succeed in preventing war. On the other hand, although
Roosevelt now thought that Czech agreement to the substance of Hitler's demands
had removed any rational basis for waging war, since Godesberg he had
had nothing but evidence that the Fuehrer was unreasonable in the extreme.
Thus, there was less justification than ever for believing that his appeal could be
effective.
Nor had anything changed on the American side of the equation to allow the
president to think that he had the means to act more effectively. Certainly, he had
no more ammunition than he had possessed before the Godesberg meeting.
Nothing that transpired there in any way altered the fact that the United States
would not, under any circumstances, actually intervene militarily in the Czech
crisis, or even threaten to do so.
Moreover, Roosevelt himself, despite his decision to intervene, seems to
have believed that it was still a risky choice in terms of effectiveness. The amount
of effort he put into learning whether his second intervention had been successful,
as well as the considerable anxiety this issue seemed to cause him,
indicates his awareness that the move could well have failed (Farnham, 1991, pp.
435-436). Clearly, the president seems to have been seeking reassurance that the
risk he had taken had succeeded. (This calls into question Secretary Hull's
assertion that Roosevelt decided to make his first intervention not because he
thought it would be successful but because he wanted to do something and
sending a message would at least do no harm. Of course, once war appeared
virtually certain, there is a sense in which nothing the president did could make
things worse. However, the evidence suggests that Roosevelt, while continuing
to fear that he risked failure, hoped for more.)
Not only did Roosevelt still worry about the risk of failure, however, he also
remained concerned about the domestic risks involved in intervening. That he
was worried on this score is demonstrated by his acquiescence to Hull's and
Davies's reservations during the drafting of his first appeal and supported by the
testimony of Moffat and Welles cited earlier. Moreover, during the planning for
his second intervention, Roosevelt displayed sensitivity to the possible risks of
This content downloaded from 147.251.120.50 on Thu, 18 Apr 2019 14:35:43 UTC
All use subject to https://about.jstor.org/terms
224 Farnham
intervenin
ticipation i
he also sho
Finally, no
to justify R
the probab
president's
tedly accur
not support
Rather, th
expectation
intervention
After Ber
quite relaxe
attempting
the post-Go
him to act.
democracie
neither a se
need to intervene in the crisis.
Thus, although Roosevelt had clearly assimilated the information that war
was imminent, he apparently did not believe that this news indicated the kind of
change in the environment which would have warranted an alteration in his
policy of nonintervention. o1 An explanation for his behavior must, therefore, be
sought elsewhere. One possibility is that, in accordance with the predictions of
prospect theory, Roosevelt's reversal of preferences about intervening in the
crisis resulted from a change in his framing of the decision problem.
The Prospect Theory Explanation
According to Tversky and Kahneman, the way a decision is framed in part
determines how people see the consequences of choice. It is entirely possible for
an individual to frame the same decision problem in different ways, and prefer-
10The slight increase in the probability of war between the afternoon of the 23rd and the morning of
the 25th is insufficient to account for the dramatic reversal in Roosevelt's policy preferences which
occurred that day. That small change in probability had a much greater impact on the president's
decision-making than rational choice theory would predict. (I am indebted to Eldar Shafir for
pointing this out.) Moreover, neither Roosevelt's perception of an increased likelihood of war nor
its imminence afforded adequate justification either for the feeling that the United States had
become directly threatened or the belief that American action might now be effective in ending the
crisis.
This content downloaded from 147.251.120.50 on Thu, 18 Apr 2019 14:35:43 UTC
All use subject to https://about.jstor.org/terms
Roosevelt and the Munich Crisis 225
ences between options have been
despite the fact that rational choice
not do so. In other words, the way
impact on the treatment of values
1981, pp. 453, 457; Kahneman an
Furthermore, according to prospe
comes against a neutral reference p
above the reference point are view
seen as losses. The importance of th
framing an outcome as a loss rathe
to it, even to the point of causing th
equivalent options. In particular, t
matter of achieving a gain, but ris
Quattrone and Tversky express it,
reference point induced by the fr
effects on people's risk preferen
1986, p. S258; Quattrone and Tve
Roosevelt's behavior during the M
of prospect theory in a number of r
the meeting at Godesberg, there w
the problem of whether or not to
led to his preference shift: after G
impending war in Europe as a loss
Not that he had previously seen
demonstrated repeatedly, neither d
tember 25, however, he unquestio
phe. 1 Nevertheless, as has just bee
there is not the slightest evidence
that such a war would threaten Am
are to some extent dangerous). Rat
loss without having changed his m
Moreover, Roosevelt's increased
consequence of perceiving war a
another change in his decision fram
have called the certainty effect. Th
to over-weight outcomes which are
merely probable (Kahneman & Tv
"In terms of prospect theory, Roosevelt's r
did not threaten the United States and he b
not constitute such a threat, i.e., it lay a
having reframed the outcome of war in Eu
upward, and he now viewed war as a depa
This content downloaded from 147.251.120.50 on Thu, 18 Apr 2019 14:35:43 UTC
All use subject to https://about.jstor.org/terms
226 Farnham
1986, pp. S
velt viewed
war seem c
growing con
it as a loss.
Moreover, the certainty effect may have combined with the so-called
pseudo-certainty effect to reinforce Roosevelt's sense that war would be a disas
ter. That is, people tend to treat extremely likely but uncertain outcomes as
though they were certain. Thus, as the probability of war rose after Godesberg
Roosevelt may first have converted it into a certainty in line with the pseudocertainty
effect and then overweighted it because of the certainty effect (Tversky
& Kahneman, 1986, p. S268).
The impact of Roosevelt's frame change, reinforced by the certainty effect
was to transform what had been merely a potential problem for American foreig
policy into a serious loss which was certain to occur. Furthermore, because this
reframing caused Roosevelt to view the crisis from a different point of reference,
it could also have been responsible for the reversal of preferences which led hi
to favor intervention in the crisis over inaction, despite the fact that the threat to
the United States had not actually increased. As Jervis points out, "losses which
are quite certain will be avoided even if they are relatively slight" (Jervis, 1989
p. 6).
As has been noted, framing effects on decision-making behavior occur when
the same alternatives are evaluated in relationship to different points of reference.
One way in which this may operate is that "the framing of an action sometimes
affects the actual experience of its outcomes" (Tversky & Kahneman, 1981, p.
458). Accordingly, framing the decision problem posed by the European crisis so
that the outcome of war seemed like the United States' loss, rather than merely
that of others, could have changed the way in which Roosevelt experienced the
consequences of nonintervention, which in turn altered his evaluation of the
alternatives for dealing with the crisis. Thus the president came to prefer acting
to halt the crisis before it ended in war to the passive stance he had adopted when
he regarded war as the loss of others. In a way, now that he viewed the impending
war as in some sense a loss for the United States, Roosevelt was no longer
evaluating the alternatives as merely an observer, but as a sort of participant.12
By reframing the war as a loss for the United States, he had also reframed the
crisis as an American crisis.
12This interpretation is supported by the fact that, from this point forward, Roosevelt's behavio
exhibited all the characteristics which, according to Raymond Cohen (1979, pp. 4, 24), identify a
decision-maker who has perceived a threat. The consequent change in the president's behavior al
accords with Hermann's (1972, p. 208) finding during a crisis simulation of differences in perce
tion and behavior between participants and observers.
This content downloaded from 147.251.120.50 on Thu, 18 Apr 2019 14:35:43 UTC
All use subject to https://about.jstor.org/terms
Roosevelt and the Munich Crisis 227
Moreover, as prospect theory pr
which he now viewed as a loss, R
showed a willingness to incur the t
that his intervention might be ine
crisis, and the danger that such
effects."
Finally, a prospect theory expla
by his apparently complete lack o
differently. At no time did the pre
to a change in his feelings about t
noticed that they had changed, let
not thought a European war suc
prevent it. This lack of recognitio
vation that "decision-makers are n
different decision frames on thei
line with their perception that, u
maker may not be able to antici
experience [Tversky & Kahnema
Tversky, 1984, pp. 349-350].)
Accounting for the Frame Ch
If Roosevelt did not initially con
the United States, why did he com
caused him to change his referenc
loss?
Unfortunately, prospect theory itself does not offer many clues about what
causes decision frames to change. Nor do the various laboratory experiments on
framing shed much light on this question. The experimenter alone manipulates
the frame, providing subjects with both its initial and its altered version, the
differences between the two being purely cognitive.
The Munich case, however, suggests that in the real world something more
than cognition may at times be involved. It would appear, that is, that the frame
change Roosevelt underwent in the midst of the crisis was triggered by the strong
emotions he experienced in the aftermath of the Godesberg meeting. In other
words, he seems to have reframed the crisis as a matter of direct concern to the
United States only after the idea of impending war had become emotionally
compelling to him.
This hypothesis is supported by several considerations. First of all, from a
purely cognitive perspective, Roosevelt's reframing of the problem posed by the
This content downloaded from 147.251.120.50 on Thu, 18 Apr 2019 14:35:43 UTC
All use subject to https://about.jstor.org/terms
228 Farnham
crisis is pu
tually certa
of threat or
Second, th
a time of g
war. Writi
The war scare
exclusion of e
Germany, Cz
and boxing up
transport to
concentrating
solini rattling
II, pp. 476Finally,
an
initial repo
numerous a
who, emoti
the crisis.
Before God
dwell on t
Rather, as
general wa
unappeasab
in focus m
Bullitt and
curiously d
emotional r
This sense
despite the
afternoon o
stopping th
26th (and w
Thus, the p
process of
13An indicatio
that it seems
rary radio dr
anything of t
affected nati
laboratory de
thunderclou
This content downloaded from 147.251.120.50 on Thu, 18 Apr 2019 14:35:43 UTC
All use subject to https://about.jstor.org/terms
Roosevelt and the Munich Crisis 229
in mood most probably induced by
was receiving from Europe.
In fact, from 23rd to the 25th of
increasingly dramatic messages wh
which would spell the end of Europe
and intransigence, graphically de
conveyed the heart-rending pleas of
Bullitt in particular, urged in the s
avoid the tragedy. Not only must t
tional impact of their own, but also
highly stressful atmosphere produc
It appears that these communicati
attention on the dreadful consequen
emotionally real to him. Thus, by Se
that war was bound to occur unless
begun to feel that it would be so ter
Strong emotion, then, was appare
from a detached observer of someon
language of prospect theory: as the p
emotional impact of the news from
of war as a loss. This in turn led him t
intervene in a fundamentally diffe
prevent the war rather than merely
This process continued as on the 2
velt underwent another day of grea
in his decision to intervene. Even
psychological pressure continued
anxious waiting for Hitler's respons
disappointment at his violent spe
feelings had by this time become ver
own account of his emotional react
tember 27 cabinet meeting and by r
drafting session for his second app
news that Hitler might be preparing
Hitler's behavior and great sympath
already considerable anxiety about
It may also be that Roosevelt's aw
his emotional distress in several way
ening event may generate greater em
the Munich crisis, the war which ha
Godesberg appeared imminent as we
cause Roosevelt to reverse prefer
This content downloaded from 147.251.120.50 on Thu, 18 Apr 2019 14:35:43 UTC
All use subject to https://about.jstor.org/terms
230 Farnham
Table I. Roo
Atti
Time Reference Point Frame of Choice Problem to Risk Preference
Pre-crisis No threat to U.S. Crisis (including outcome of Averse Do not inWar
not expected from Europe war) = within status quo tervene
9/16
Berchtesgaden
War expected
Allies -* win
9/23
Godesberg
War = certain
War = imminent
Allies -- win
FRAME CHANGE
9/24-25 " War = loss Acceptant Intervene
Affect-laden reports #1
War = certain
War = imminent
Allies -+ win
9/27 " " " Intervene
Hitler rebuff #2
War = certain
War = imminent
Allies -* win
exacerbated the effect of the emotions which did. That is, imminence increa
the salience of war by making clear that the war which seemed certain to oc
would do so now rather than later. Not only might this awareness have heigh
ened the painful emotions that Roosevelt was already feeling, it could also ha
intensified his stress by adding the pressure of time (Janis and Mann, 1977,
54, 59; I am indebted to Alexander George for pointing out that imminence m
have a different impact on a decision-maker's assessment of threat than do
certainty).
In any event, there is considerable empirical evidence pointing to the experience
of strong emotion as the crucial element in Roosevelt's post-Godesberg
change of frame and subsequent reversal of preferences. Moreover, the idea that
such emotion can mediate cognitive change has received ample theoretical support
in the literature dealing with the relationship between cognition and affect.
As Hoffman has pointed out, "cognitive psychologists often view affect as
providing the motivating force for initiating cognitive processes" (Hoffman,
1986, p. 260; for Hoffman's view of the ways in which affect may influence
information processing, see pp. 245-246).
Any account of what Roosevelt experienced after Godesberg, however, is
incomplete without an awareness that the reverse process can occur as well. That
This content downloaded from 147.251.120.50 on Thu, 18 Apr 2019 14:35:43 UTC
All use subject to https://about.jstor.org/terms
Roosevelt and the Munich Crisis 231
is, a change of frame can affect
Tversky and Kahneman themsel
sponse as the outcome of frame
Kahneman, 1981, p. 458; Kahn
Clearly, this process could also
framing the crisis as in some se
emotional involvement which ca
turn, could have had the effect of
of war was seen as a loss (It is a
personal experience of a crisis g
gan, for example, defines a natio
tant values which, for the decisi
sity. .. ." Morgan, 1977, p. 16
George, 1975).
Again, the literature on the rela
solid backing for the idea that
affect. Indeed, as Hoffman obser
been one "in which feelings a
(Hoffman, 1986, p. 24).
Moreover, evidence that the ca
runs in both directions has recen
appears to have been the case
mutually reinforcing. Thus, wit
cognitive and affective/motivat
tions," Sorrentino and Higgins a
motivation and cognition are, in fact
leads to motivation and motivation lea
of the other. They are synergistic in
effects. (Sorrentino and Higgins, 1986,
affect is, of course, the subject of a lon
no consensus. For opposing views on
Thus, the news from Godesbe
interaction between affect and c
Roosevelt from a somewhat detac
pant. He had come to feel that w
the United States as well, and, i
emotional terms, from being so
own. As a consequence of this
preferences and chose to interve
A final question about the rol
behavior during the Munich cris
decisional conflict model of dec
This content downloaded from 147.251.120.50 on Thu, 18 Apr 2019 14:35:43 UTC
All use subject to https://about.jstor.org/terms
232 Farnham
a number of
such as sign
experienced
pact on his
change his
What may
that not onl
Roosevelt d
Contrary to
the need to
emotion an
avoidance (J
is no eviden
biased infor
conflict mo
not appear t
rational man
Janis and M
while Roosev
ing," it was
pp. 50-52). N
question abo
degree of str
motivates h
the general
stress is act
question is a
individual d
for stress m
criticisms
Morgan, 19
CONCLUSION
This study has sought to evaluate the competing claims of prospec
and rational choice theory to explain political decision-making in the l
actual historical case. In applying the predictions of both theories to F
Roosevelt's decision-making during the Munich crisis, the study mak
that prospect theory provides a more satisfactory explanation of his b
that time than does the theory of rational choice.
In particular, the evidence uncovered by a detailed analysis of Ro
This content downloaded from 147.251.120.50 on Thu, 18 Apr 2019 14:35:43 UTC
All use subject to https://about.jstor.org/terms
Roosevelt and the Munich Crisis 233
decision-making behavior shows th
ability of American intervention i
of the expected utility of interven
the midst of the crisis, despite th
changed materially from an Ame
framed the outcome of war as a lo
point of incurring risks he had p
cannot be accommodated within th
With respect to the theoretical
least they support the demand
legitimate alternative for explain
context. They also show that in so
which is puzzling for theories of
Moreover, analyzing Roosevelt's
the predictions of prospect theory
would benefit from elaboration
change of frame was apparently tr
further research into the causes of
role of affect.
Finally, while supporting the id
decision-making in a number of a
decision-making problem and the
study raises a question about Janis
Roosevelt experienced a level of s
decision problem and reverse his
decision-making behavior not also
under conditions of high stress pr
model is to be of use in explainin
requires clarification in this area.
The attempt to apply prospect
theoretical benefits in several are
tion for advancing the theory itse
tions about alternative explanatio
ACKNOWLEDGMENTS
I would like to thank Alexander George, Robert Jervis, Jack L
McDermott, Eldar Shafir, and Janice Stein for their comments on th
earlier version was presented at the Thirteenth Annual Scientific M
International Society of Political Psychology, Washington, D.C., Ju
1990.
This content downloaded from 147.251.120.50 on Thu, 18 Apr 2019 14:35:43 UTC
All use subject to https://about.jstor.org/terms
234 Farnham
REFERENCES
Alsop, J., & Kintner, R. (1940). American white paper. New York: Simon and Sch
Bazerman, M. H. (1983). Negotiator judgment. American Behavioral Scientist, 27,
Berle, B. (Ed.), (1973). Navigating the rapids, 1918-1971: From the papers of Ad
New York: Harcourt, Brace, Jovanovich.
Bueno de Mesquita, B. (1981). The war trap. New Haven: Yale University Press.
Blum, J. M. (Ed.). (1965). From the Morgenthau Diaries (Vol. 1). Boston: Houghton
Bullitt, O. H. (Ed.). (1972). For the President: Personal and secret-Correspond
Franklin D. Roosevelt and William C. Bullitt. Boston: Houghton Mifflin.
Butler, R., & Woodward, Sir, E. L. (Eds.). (1951). Documents on British foreign
1939. (3rd Series, Vol. 3). London: His Majesty's Stationery Office.
Cantril, H. (1940). The invasion from Mars. Princeton, NJ: Princeton University P
Cohen, R. (1979). Threat perception in international crisis. Madison, WI: Universit
Press.
Cole, W. (1983). Roosevelt and the isolationists, 1932-45. Lincoln, NE: Universit
Press.
Dallek, R. (1979). Franklin D. Roosevelt and American foreign policy, 1932-1945, New York:
Oxford University Press.
Davis, M. A., & Bobko, P. (1986). Contextual effects on escalation processes, Organizational
Behavior and Human Decision Processes, 37, 121-138.
Divine, R. (1965). The reluctant belligerent. New York: John Wiley.
Divine, R. (1969). Roosevelt and World War 1H. Baltimore: Johns Hopkins Press.
Farnham, B. (1990). Political cognition and decision-making, Political Psychology, 11, 83-111.
Farnham, B. (1991). Value conflict and decision-making: Franklin D. Roosevelt and the Munich
crisis, 1938. Unpublished doctoral dissertation, Columbia University, New York.
Fischoff, B. (1983). Strategic policy preferences: A behavioral decision theory perspective, Journal
of Social Issues, 39, 133-160.
Haight, J. M. Jr. (1970). American aid to France, 1938-1940. New York: Atheneum.
Haight, J. M. Jr. (1960). France, the United States, and the Munich crisis, Journal of Modern
History, 32, 340-358.
Hermann, C. F. (1972). Threat, time, and surprise: A simulation of international crisis. In C. F.
Hermann (Ed.), International crisis (pp. 187-211). New York: Free Press.
Hoffman, M. L. (1986). Affect, cognition, and motivation, Sorrentino, R. M., & Higgins, E. T.
(Eds.), Handbook of motivation and cognition, New York: Guilford Press.
Holsti, O. R. (1972). Crisis, escalation, war, Montreal: McGill-Queens University Press.
Holsti, O. R. (1971). Crisis, stress and decision-making, International Social Science Journal, 23,
53-67.
Holsti, O. R. & George, A. (1975). Effects of stress on the performance of foreign policy-makers
C. Cotter (Ed.), Political science annual (Vol. 6, pp. 255-319). Indianapolis: Bobbs Meri
Hooker, N. H. (Ed.). (1956). The Moffat papers: Selections from the diplomatic papers of
Pierrepont Moffat, 1919-1943. Cambridge, MA: Harvard University Press.
Hull, C. (1958). Memoirs. New York: MacMillan.
Ickes, H. (1954). Secret diary (Vol. 2). New York: Simon and Schuster.
Janis, I. (1959). Decisional conflicts: A theoretical analysis, Journal of Conflict Resolution, 3, 6-2
Janis, I., & Mann, L. (1977). Decisionmaking: A psychological analysis of conflict, choice,
commitment. New York: Free Press.
Jervis, R. (1989). Political implications of loss aversion, Unpublished manuscript.
Kahneman, D., & Tversky, A. (1984). Choices, values, and frames, American Psychologist, 39,
341-350.
Kahneman, D., & Tversky, A. (1979). Prospect theory: An analysis of decision under risk
metrica, 47, 263-291.
Kahneman, D., & Tversky, A. (1982). The psychology of preferences, Scientific American, 246,
160-173.
This content downloaded from 147.251.120.50 on Thu, 18 Apr 2019 14:35:43 UTC
All use subject to https://about.jstor.org/terms
Roosevelt and the Munich Crisis 235
Kaufmann, W. W. (1963). Two American a
F. G. Gilbert (Eds.) The Diplomats, 1919Klass,
P. (October 30, 1988). Wells, Welles an
Langer, W. L., & Gleason, E. S. (1952). The c
and Brothers.
Lash, J. P. (1976). Roosevelt and Churchill, 1939-1941: The partnership that saved the West. New
York: W. W. Norton.
Lazarus, R. S. (1982). Thoughts on the relations between emotion and cognition, American Psychologist,
37, 1019-1024.
Levi, A. & Tetlock, P. E. (1980). A cognitive analysis of Japan's 1941 decision for war, Journal
Conflict Resolution, 24, 195-211.
Levin, I. P., Johnson, R. D., Russo, C. P., & Deldin, P. J. (1985). Framing effects in judgment tas
with varying amounts of information, Organizational Behavior and Human Decision Processes,
36, 362-377.
McDermott, R. (1990). Prospect theory in international relations: the Iranian rescue mission. Unpublished
manuscript.
Memo from Adolf Berle to the President. (September 1, 1938). President's Secretary's File, State
Department, Berle Folder, Box 93, Roosevelt Library, Hyde Park, NY.
Morgan, P. (1977). Deterrence: A conceptual analysis, Beverly Hills, CA: Sage Library of Social
Science.
Morgenthau Diary, (1938). Roosevelt Library, Hyde Park, NY.
Murray, A. (1946). At close quarters. London: John Murray.
Offner, A. A. (1969). American appeasement: United States foreign policy and Germany, 1933-
1938, New York: W. W. Norton.
Quattrone, G. A., & Tversky, A. (1984). Causal versus diagnostic contingencies: On self-deception
and on the voter's illusion, Journal of personality and Social Psychology, 46, 237-248.
Quattrone, G. A., & Tversky, A. (1988). Contrasting rational and psychological analyses of political
choice, American Political Science Review, 82, 719-736.
Shapira, Z. (1981). Making trade-offs between job attributes, Organizational Behavior and Human
Performance, 28, 331-355.
Schewe, D. B. (1979). Franklin D. Roosevelt and foreign affairs, January 1937-August 1939 (Vols.
4-8). New York: Garland Publishing.
Sorrentino, R. M., & Higgins, E. T. (1986). Motivation and cognition, In R. M. Sorrentino and
E. T. Higgins (Eds.) Handbook of motivation and cognition. New York: Guilford Press.
Stein, J. G. & Tanter, R. (1980). Rational decision-making. Columbus, OH: Ohio State University
Press.
Steinbruner, J. (1974). Cybernetic theory of decision, Princeton, NJ: Princeton University Press
Taylor, T. (1979). Munich. New York: Doubleday.
Tetlock, P. E. (1985). Accountability: The neglected social context of judgment and choice. Research
in Organizational Behavior, 7, 297-332.
Tversky, A., & Kahneman, D. (1981). The framing of decisions and the psychology of choice,
Science, 211, 453-458.
Tversky, A., & Kahneman, D. (1986). Rational choice and the framing of decisions. Journal of
Business, 59, S251-S278.
United States, Department of State. (1938). Foreign Relations of the United States (Vol. 1).
Zajonc, R. B. (1980). Feeling and thinking: Preferences need no inferences, American Psychologist,
35, 151-175.
This content downloaded from 147.251.120.50 on Thu, 18 Apr 2019 14:35:43 UTC
All use subject to https://about.jstor.org/terms