University of California at Los Angeles
January 23, 2013
Many scholars have argued that democratic institutions have a profound effect on the way democracies conduct themselves in the international arena, but few works in this area have generated as much interest as Dan Reiter’s and Allan Stam’s book Democracies at War.1 The core argument there was very clear. Democracies tend to win the wars they fight, according to Reiter and Stam, because their leaders would lose power if they did not do what the voters wanted; the voters would punish leaders who take the country into a war it would lose, or even into a war that might “drag on for too long,” so those wars tend not to get fought.2 “The will of the people,” Reiter and Stam argued, “restrains democratic leaders and helps prevent them from initiating foolhardy or risky wars”; the wars they do start, however, tend to be wars they can win, because they are inclined to “pick on relatively weaker target states.”3 And a big part of the reason why they are able to “select themselves” into wars where winning is relatively easy, the argument ran, is that they “produce better estimates of the probability of victory than their autocratic counterparts do.”4 Indeed, in democracies “the vigorous discussion of alternatives and open dissemination of information” make for better foreign policy in general, and even within the government, flawed policies are more likely to be exposed because toadyism is much less of a problem for democracies than it is for other sorts or regimes.5 But what goes on within governments is ultimately of secondary importance. Power lies essentially with the people, and a policy that looks toward war is viable only if the people support it: “Democratic decisions for war are determined and constrained by public consent,” and to “generate consent” the government has to make its case in the “open marketplace of ideas.”6 The ability of democratic governments to circumvent that process by controlling the flow of information is, in their view, quite limited. Indeed, as they themselves point out, their argument rests on the assumption that “consent cannot be easily manufactured by democratic leaders,” since if leaders could “manipulate public opinion into supporting military ventures, then of course public opinion would provide little constraint on democratic foreign policy.”7
Does this argument stand up in the light of the evidence? Soon after the book came out in 2002, a number of scholars criticized some of the key points Reiter and Stam had made, and this led to a lively debate conducted mainly in the journal International Security. But those critics were not primarily interested in examining Reiter’s and Stam’s assumptions about democratic governments not being able to pursue policies that the public does not support or about the non-manipulability of public opinion. Their real concerns lay elsewhere. Although they sometimes alluded to the arguments advanced by Reiter and Stam about the “marketplace of ideas” and about the government not being able to mislead the public on major issues, they seemed more interested in examining certain other issues that Reiter and Stam had dealt with, most notably the question of the relationship between military effectiveness and regime type.8
So it was not until John Schuessler published his article “The Deception Dividend: FDR’s Undeclared War” in International Security in 2010 that the whole question of a government’s ability to do what it wants without first making sure the public approves of its policy became a focus of discussion. Schuessler’s argument was framed in very moderate terms. He said he was not really challenging the basic Reiter and Stam “selection effects” argument, and that he was just interested in trying to understand what happens when the selection effect breaks down—that is, in what happens “when leaders are drawn toward wars where an easy victory is anything but assured” and many people would prefer not to go to war.9 His answer was that governments in those cases could resort to deception in order to get people to go along with a policy they would not have supported if the issue had been put to them in an honest and straightforward way. “Rather than press their case in the marketplace of ideas under such unfavorable circumstances,” he wrote, “leaders will be tempted to preempt debate by shifting blame for hostilities onto the adversary. The trick is to prepare domestic opinion for a possible, and even probable, war while providing firm assurances that it will come only as a last resort and only when the other side forces the issue.”10 A strategy of deception could thus allow a leader to bridge the gap between what is considered necessary for basic power political reasons and what the country as a whole is prepared to sanction.
To show that a strategy of that sort could sometimes work, Schuessler looked at U.S. policy in the period before Pearl Harbor. His analysis of American policy at that time, he said, supported his core argument. Although President Franklin D. Roosevelt “was sensitive to the domestic mood and waited until public consent was forthcoming to ask for an official declaration of war,” there was, he argued, “compelling evidence that Roosevelt settled on a war policy well before Pearl Harbor and that in the interim he engaged in a significant amount of deception, maneuvering the country in the direction of open hostilities while assuring a wary public that the United States would remain at peace.”11
Schuessler was thus taking issue with the Reiter and Stam argument in some fairly fundamental ways, and Reiter responded first with a long letter to the editor of International Security and then with the Security Studies article we are concerned with here.12 In that article, Reiter presents a detailed and systematic critique of the deception argument. His basic claim is that “elected leaders are deterred from deceiving because they recognize that if an attempt at deception is exposed, then they will suffer heavy domestic political costs.” Those attempts, he believes, are likely to be exposed because of the existence of certain institutions found in democratic countries: the “marketplace of ideas,” “a professional military and government bureaucracy,” and a political system in which two or more parties compete for power. And since “elected leaders are deterred from engaging in deception,” he argues, “they cannot circumvent public opinion constraints.”13 Those general arguments are supported by a detailed discussion of U.S. policy in the pre-Pearl Harbor period.
In developing that argument, Reiter was not just trying to refute the basic claims Schuessler had made. He also referred to various arguments I had made about U.S. policy in 1941, and which Schuessler in fact had drawn on.14 Both Schuessler and I had claimed, he wrote, that Roosevelt “wanted American entry into war in 1941, saw the public as hesitant, and secretly provoked Germany and Japan, hoping to circumvent the constraints of public opinion and cause war between the United States and these two powers.” With regard to Germany, we had claimed “that Roosevelt engaged in secretly provocative naval policies in the Atlantic, hoping to spark a naval clash between America and Germany, providing the American public with a casus belli.” With regard to Japan, we had claimed that “Roosevelt restricted the sale of oil to Japan in July 1941, knowing that such an action would trigger a Japanese attack.” But in neither case did he find much evidence to support those arguments. “Roosevelt’s policies toward Germany,” he believes, “were public, popular, and restrained” and Roosevelt’s policies toward Japan “were public, popular, and belligerent, and aimed to deter rather than provoke.”15
What is to be made of Reiter’s argument in this article? One can begin by looking at some of his more general claims. He is willing to admit that “leaders might be motivated to deceive,” but thinks they are “deterred from engaging in deception” because they know there is a good chance their deceptions would be exposed and they would “suffer heavy political losses.”16 But one has to look at both sides of the ledger. One has to look not just at the potential price, but also at the potential benefit; the decision about whether to adopt deceptive tactics would depend on how they stack up against each other. And potential costs might be more limited that Reiter would have us believe. Indeed, it is precisely because leaders know they might have to pay a big price at home if they get caught misleading the public that they have a great incentive to take measures that reduce the risk that they will get caught (or the price they would pay if they do get caught). They can, for example, limit the number of people who understand what the real policy is.17 And in fact a U.S. president can conceivably keep his real thinking to himself. Roosevelt especially tended to keep his own counsel, and, as many scholars point out, this is one of the reasons why it is so hard to know for sure exactly what he was up to.18
In any event, one cannot simply assume that “members of the opposition have access to information about secret foreign policy actions.”19 They may have access to some information, but rarely do they know everything—and key information is kept from them in large part because the government knows they may leak it. And one cannot simply assume that in democracies the civilian bureaucracy and the armed services are “more likely to be staffed by competent professionals” more loyal to the nation than to the particular government they serve, and thus are more likely than their equivalents in non-democratic regimes to speak out when they disapprove of the policies the government is pursuing.20 For one thing, in what sense are bureaucracies and military establishments more “professional” in democratic states than in similar countries with a different sort of political regime? In what sense, for example, was the French military more “professional” before 1914 than its German equivalent? Beyond that, do top civilian officials and high military officers in democracies think it is all right to leak to the press or publicly criticize government policy? Is that what they would mean by professionalism? And finally, one cannot simply assume that the best ideas prevail when issues are debated in public. As George Kennan once noted, “the truth is sometimes a poor competitor in the market place of ideas—complicated, unsatisfying, full of dilemmas, always vulnerable to misinterpretation and abuse.”21
But I don’t want to dwell on general issues of this sort. The whole argument here really turns on more concrete historical questions. So what are we to make of what Reiter says about America’s road to war in 1941?
Roosevelt and Germany in 1941
Reiter begins his section on “Roosevelt and Germany in 1941” by paraphrasing one of the basic claims Schuessler and I had made. “Some have argued,” he writes, “that Roosevelt attempted to provoke Germany secretly in 1941 through secretly belligerent naval policies in the Atlantic.” But the evidence, he says, “does not support this claim.” He agrees that Roosevelt wanted to take the country into the war, but says that public opinion placed a limit on how far he could go.22 Roosevelt could not pursue a policy that went beyond what the public would support—a secret naval policy aimed at provoking Hitler—“because of his fear of the domestic political consequences of exposed deception”; “the fear of public reaction to exposed deception deterred him from taking secretly provocative action.”23 The president, he writes, “understood the importance of remaining transparent and aboveboard”; Roosevelt “described major shifts in US policy publicly”; and the public essentially understood what the Navy was doing.24 Constrained by public opinion, and unable to evade those constraints by pursuing a policy in secret which he knew the public would not support, he therefore pursued a relatively moderate course of action in the Atlantic. “No major naval policies,” he says, “were adopted in secret.”25 The Navy was kept on a short leash. “Though the repeal of the Neutrality Act permitted Roosevelt to send armed US convoys escorted by US Navy vessels across the Atlantic,” Reiter says, “he hesitated even through late November, instead sending unarmed convoys to Britain and Russia.”26 The president thus “constrained rather than increased the provocativeness of US policy,” since he understood that he could not secretly pursue a policy aimed at provoking a naval incident with Germany, if that policy “exceeded what the public was willing to accept.”27
Schuessler and I would agree with much of this—with the idea that Roosevelt wanted to take the country into the war, and with the point that the president had to worry about public opinion and did not have a free hand to pursue whatever policy he wanted. As Schuessler himself wrote, “Roosevelt was certainly constrained by public opinion in the lead-up to World War II. Otherwise, he would have had no reason to resort to all the maneuverings he did to get the United States into the war.”28 We recognize that the constraints were real, but we think (and this is where we disagree with Reiter) that Roosevelt, to a certain extent, was able to maneuver around them—that he was able to pursue tougher and more provocative policies toward Germany than the public would have sanctioned if the issue had been put to them in an honest and straightforward way. He was able to pursue a policy that the public was not privy to, and said and did things which would have surprised and perhaps even shocked people if they had known about them at the time. And in making that argument we were thinking mostly of the undeclared naval war in the Atlantic and about what the President was secretly saying about it: Roosevelt, in our view, did “increase the provocativeness of U.S. policy” in that area in late 1941. He did want to provoke Germany into going to war with America—a policy which, however, thanks to Adolf Hitler, did not fully achieve its objectives.
There is only way to get at these issues, and that is to look at what was actually going on in the Atlantic at the time. The basic story here is fairly clear. In the early part of the war, American policy had been fairly passive. The assumption was that Britain and France would be able to hold the line against Germany, so the fall of France in June 1940 came as a shock. It was at that point that Roosevelt made the very fundamental decision, as David Reynolds says, “to back Britain as America’s front line”; the policy he pursued from that point on, which culminated in something “close to an undeclared naval war with Germany” in the fall of 1941, “rolled inexorably from the basic decision” he had made at that time.29 But the President could not move too quickly along that road. Most Americans wanted to keep out of the war and Roosevelt certainly had to take their views into account. This was particularly true in the run-up to the presidential election of November 1940. The Republicans, as Robert Dallek notes, had decided to scare people “with warnings that Roosevelt’s reelection would mean wooden crosses for their sons and brothers and sweethearts”; those charges were effective, and it seemed that the president might lose unless he could “demonstrate his commitment to peace.” The result was his famous declaration in his October 30 Boston speech, which Schuessler referred to in his article: “I have said this before, but I shall say it again and again and again: Your boys are not going to be sent into any foreign wars.”30 But his real thinking was evolving in a very different direction.
Indeed, even as Roosevelt was promising the country that he would not take it into war, one of his top military advisors, Admiral Harold Stark, the Chief of Naval Operations (CNO), was working on an important memorandum about strategy, commonly called the ‘Plan Dog’ memorandum. Stark’s basic argument was that America had a vital interest in making sure that Britain survived, that the survival of that country was by no means guaranteed, and that in the absence of “active American military assistance,” Britain might well collapse. Britain, Stark thought, “requires from us very great help in the Atlantic, and possibly even on the continents of Europe or Africa, if she is to be enabled to survive,” and by that Stark meant not just naval assistance. To win the war, a “land offensive against the Axis powers” was essential, and Britain alone could not mount a successful one. The Americans, he wrote, “in addition to sending naval assistance, would also need to send large air and land forces to Europe or Africa, or both, and to participate strongly in this land offensive.”31 Stark, it is important to note, was not just laying out a strategy for how the war should be fought if the United States entered it. He was also making an argument about how important it was for America to get involved, indeed ultimately in a fairly massive way. His views could scarcely have been clearer: “It has long been my opinion,” he wrote on October 8, 1941, “that Germany cannot be defeated unless the United States is wholeheartedly in the war and makes a strong military and naval effort wherever strategy dictates.”32
The important thing to note for our purposes is that the President by no means rejected that way of thinking. It seems, in fact, that Roosevelt, very secretly, made it clear to Stark that he and the Admiral were on the same wavelength on these issues. Roosevelt asked Stark to coordinate the plan with the Army and the State Department, but when the CNO wrote the Army Chief of Staff, General George Marshall, about this, he would not put down in writing what Roosevelt had said; he told Marshall that he was holding the president’s comments “very tight.”33 Stark would scarcely have needed to do that if Roosevelt had been evasive or non-committal or if he had simply endorsed the ‘Germany first’ strategy, the strategy for actually fighting the war if the United States got involved, which the Plan Dog memorandum had recommended. If Stark needed to be so secretive, this could only be because what the President had told him was extremely sensitive, and issues related to how the war would be fought if America came in did not fall in that category 34 And Stark’s impression in January 1941 that “we were heading straight for this war” and his comment that personally he did “not see how we could avoid, either having it thrust upon us or of our deliberately going in, many months longer” also suggest that he had reached the conclusion that the president was going to do whatever it took to keep Britain from going under; Stark would not have gotten that impression if Roosevelt had been determined to keep America out of the war.35 In short, one gets the sense that while Roosevelt might not have been totally open with Stark, the two men basically saw things the same way, and that their major difference can be explained by the famous aphorism about where you stand depending on where you sit: as President, Roosevelt naturally placed much greater weight on domestic political considerations than the Chief of Naval Operations did.
Roosevelt’s role in getting staff talks with the British going is another important indicator. Stark’s one explicit recommendation in the Plan Dog memorandum was that “as a preliminary to possible entry of the United States into the conflict, the United States Army and Navy at once undertake secret staff talks on technical matters” with their British, Canadian, and Dutch counterparts.”36 And in fact the staff conversations began in January 1941 and led two months later to an important agreement called ABC-1 on coordinated measures to be taken once America came into the war; in practice, however, ABC-1 also played a certain role in governing policy at the operational level during the pre-Pearl Harbor period.37 For years it was thought that in convening the talks, Stark had acted on his own initiative—Stark, in fact, told a Congressional committee that he had acted on his own—and it was only when the British records were opened in 1972 that scholars were able to see that it was the President himself who had authorized the talks.38 The talks themselves were secret— few Americans even knew about them until the war was over—and Stark actually warned at the time that it “might well be disastrous” if the press found out what was going on.39 “In this estimate,” as James Leutze writes in his important study of the subject, “he was no doubt correct.” The administration was trying at the time to get the Lend-Lease bill through Congress; its opponents doubted that Roosevelt was sincere when he said he wanted to keep the country out of the war. And Roosevelt’s credibility, Leutze points out, “would have been seriously undermined had it been learned that a few blocks from the Capitol British and American officers were making plans for a coalition war.” The gap between what the administration was saying in public and what it was actually doing in the staff talks was quite extraordinary:
While the debate over the bill proceeded in Congress and in the press, the ABC Conference was being conducted in camera at the Navy Department. Time after time the administration contended that Lend-Lease was essential to keep England fighting and America out of the war. On the floor of the Congress and in press conferences, commitments and plans to convoy or send American troops out of the hemisphere were denied repeatedly by administration spokesmen. But at the other end of Pennsylvania Avenue the military representatives of that same administration were conducting the first prewar staff conferences in American history and laying plans for convoying, building bases in the British Isles, and replacing British Tommys in Iceland with American GIs.40
The point about escorting convoys was of central importance because the warships doing the escorting might be attacked by, or might have to attack, German naval vessels (especially submarines) threatening the convoys. In January, Roosevelt said at a news conference (in Leutze’s paraphrase) that he had “no plans to escort supplies to England,” but the very “next day he stated—this time, privately—that escorting was probably necessary.”41 Torn between his desire to help the British and his sense for what the American people would accept, he went back and forth on this issue. In mid-March, for example, it seemed that the President was prepared to move ahead quickly: he “directed that the Atlantic Fleet be brought to wartime readiness.” A series of measures were to be taken in line with that order; as Stark told the Commander of the Atlantic Fleet, Admiral Ernest King, “this step is, in effect, a war mobilization.”42 By March 20, plans to escort convoys in the western Atlantic had been worked out. But Roosevelt was not sure that public opinion would support that move or whether Congress, if asked, would give him the authority to do it, so in mid-April he backed off.43
Then in June he decided to send American Marines to Iceland, a move explained to the public as a defensive act, designed to prevent the Germans from stationing forces there that could threaten America. But the U.S. garrison on the island had to be supplied and the supply convoys had to be protected, and Roosevelt was apparently now ready to escort not just American but also British and other allied merchant ships, simply by allowing them to join the convoys the U.S. Navy would now be escorting.44 The President, according Patrick Abbazia, the author of the most important book on the undeclared naval war, probably sensed “that the need to bring supplies to the occupation force in Iceland might provide a convenient and plausible rationale for escorting Allied merchant convoys in the western Atlantic.” And to protect the convoys, fairly aggressive action might be called for. As Roosevelt himself noted, the U.S. supply convoys needed to be able to protect themselves from “threat of attack”; “under conditions of modern sea warfare,” he pointed out, the term “threat of attack” could apply to forces “reasonably long distances” away from the convoys. It followed, as he put it, that the “very presence of a German submarine or raider on or near the line of communications constitutes ‘threat of attack,’” and “should be dealt with by action looking to the elimination” of such a threat. Admiral King implemented these instructions by authorizing the forces under his command to attack German naval vessels that threatened the convoys (using a fairly broad definition as to what constituted a threat).45 Germany had attacked the Soviet Union in late June; it seemed that the time for action in the Atlantic had come. Stark told the President that the United States should “start escorting immediately” and should protect “the Western Atlantic on a large scale”; “such action,” he recognized, “would almost certainly involve us in the war.” That, however, would in his view be a good thing, since “every day of delay in our getting into the war” was dangerous, and “much more delay might be fatal to Britain’s survival.”46
But in early July Roosevelt again drew back and canceled the escort plans, apparently because he was worried about the situation in the Far East. It was only in late July that he decided to take the plunge and authorize the escort operations in the Atlantic.47 Reiter points out quite correctly that Roosevelt made sure that a phrase allowing British and other ships to join the U.S.-protected convoys was deleted from the final version of the key document here. What he does not note is that that deletion had no practical effect, since more or less informal arrangements were worked out to allow British and other allied ships to effectively join the convoys; in deleting that phrase, Roosevelt, as was his practice, was just covering his tracks.48 The new policy was not to be publicized. Stark told his associates (and the British) that “‘the whole thing must be kept as quiet as possible.’ To this end, he urged that arrangements be made orally whenever possible.”49 Some of Roosevelt’s main advisors, most notably Secretary of War Henry Stimson, wanted him to be more forthright and make it clear to the nation what had been decided and why the government felt compelled to move ahead in this way, but the President “was definitely unwilling to lay the situation before the country so frankly.”50 Reiter asserted categorically that “no major naval policies were adopted in secret,” but this was not quite true.51 As one scholar put it, “Roosevelt refused to make a public announcement of this tidal shift in American maritime strategy.”52 And when the ‘shoot first’ policy was announced after the Greer incident, it was presented in a very different context.
The new policy was explained to the British at the Atlantic Conference, held in Placentia Bay off the coast of Argentia, Newfoundland, in early August; the two sides worked out arrangements there for the “American escort of British as well as American merchant ships as far as Iceland”; the new system was apparently “put into operation on August 20.”53 (The formal orders, with implementation instructions, were issued a little later.54) According to the British Prime Minister, Winston Churchill, Roosevelt explained what he was up to—that is, what the real point of the policy was. The President, Churchill told his Cabinet colleagues when he got back to England, “was obviously determined” that the United States should come into the war. But he was “skating on pretty thin ice” with the Congress, and felt that “if he were to put the issue of peace and war to Congress, they would debate it for three months.” He therefore had to proceed in a more indirect way: “The President said that he would wage war, but not declare it, and that he would become more and more provocative. If the Germans did not like it, they could attack American forces.” Roosevelt had ordered the American Navy to “attack any U-boat which showed itself, even if it were 200 or 300 miles away from the convoy,” and “Admiral Stark intended to carry out this order literally.” “Everything was to be done to force an ‘incident.’”55 In that way, the United States might very well be able to enter the war.
How is Churchill’s account to be taken? If he was reporting what Roosevelt had said accurately, and if the President’s comments reflected his real thinking, then this document would be something of a ‘smoking gun.’ This clearly was not the sort of line Roosevelt could take in public, and if it represented his real view, then that would show how averse he was to arguing out the great issue of war and peace in the ‘marketplace of ideas,’ and that he felt compelled instead to proceed in a more devious way. So it is scarcely surprising that Reiter plays down the importance of this piece of evidence. Indeed, he feels that Roosevelt probably did not actually say what Churchill had told the Cabinet he had said, and that Churchill was just trying to “persuade doubters in his own government that Roosevelt was serious about bringing the United States into the war.”56 But that argument, to my mind, is not very persuasive. Roosevelt, after all, had often spoken about waging an undeclared way and about provoking ‘incidents’ on the high seas. Reiter tends to minimize the significance of those comments, but it is hard to see why they should not be taken “as serious indications” of what was going on in the President’s mind or why they should not be taken into account when one is trying to figure out how genuine Churchill’s account was.57 But the main reason for taking Churchill’s account at face value is that Roosevelt did become “more and more provocative,” that he did begin an undeclared naval war shortly after the Atlantic Conference, that direct orders to the U.S. fleet to attack German naval vessels were issued not long after the conference was over, and that the president did use one incident that took place as a direct result of what the U.S. Navy was doing—the Greer incident—as the occasion for announcing the ‘shoot first’ policy.58 The description of his policy that Roosevelt gave Churchill was actually quite accurate. Indeed, he was far more open with Churchill than he was with his own people at the time.59
Reiter, of course, takes a different view, so to see whether the United States did become “more and more provocative” in the North Atlantic (Churchill’s paraphrase of what Roosevelt had said) or whether the President “constrained rather than increased the provocativeness of US policy” (as Reiter claims) it is important to look in some detail at what the Navy was doing in the Atlantic and at the sorts of orders that were issued. As late as July 3, policy was still relatively mild: American forces were to be on their guard but were to avoid “untoward incidents”; to that end—that is, to avoid the sorts of problems that could develop at night when it was hard to distinguish between British and American ships—U.S. warships were to “show navigational lights and normal deck and other outboard lights.”60 On July 15, the fleet was still to remain prepared to “repel hostile attack,” but was now ordered to “operate as under war conditions, including complete darkening of ships when at sea”—suggesting that it was less important now to avoid “incidents” than it had been. On July 28, U.S. forces were told to “destroy hostile forces which threaten shipping of U.S. and Iceland flag,” although what constituted a threat was still not clearly defined; U.S. naval units were simply told to trail German vessels and “broadcast in plain language their movements,” presumably to help the British move in to destroy them.61 A month later, after Argentia, the Americans were ready to take stronger measures. On August 22, Stark wrote the Pacific commander Admiral Husband Kimmel with regard to the Atlantic: “Thank God we should have things in full swing before long and with plans fairly complete. It has changed so many times—but now I think we at last have something fairly definite—may-be.”62 On August 25, Stark told King that his forces were “to destroy surface raiders which attacked shipping along the sealanes between North America and Iceland or which approached these lanes sufficiently close to threaten such shipping.” On September 3, King’s authority was expanded; his forces could now target all “hostile forces,” so submarines threatening those sealanes could also be attacked.63 On September 5, the fleet was told to “destroy potentially hostile surface raiders and submarines which attack or threaten. Threat exists when such potentially hostile vessels are found within one hundred miles of U.S. flag shipping or anywhere within proclaimed neutrality zone.” On September 12, the mission was defined even more broadly: the fleet was to “destroy potentially hostile forces which are encountered within the Western Atlantic Area.”64 The U.S. Navy, as one writer put it, was setting “the bar ever lower for all-out war at sea with Germany.”65
So the United States was certainly becoming more active in the North Atlantic in this period, and—to again quote Abbazia—the Greer incident, which took place on September 4, simply provided the President “with a convenient opportunity to announce what had already been decided, that the U.S. Navy would soon commence escort operations in the western Atlantic.”66 The basic order about escorting (Admiral King’s Operations Plan 7-41) had gone out on September 1, just before the Greer was attacked. And American destroyers attacked real or imagined German submarines with depth charges at least 80 times in the pre-Pearl Harbor period.67
But there were few major incidents, and not a single U-boat was actually sunk by American forces prior to Pearl Harbor. This does not mean, however, that the Americans were not serious about the naval war—that they could easily have opted for a more aggressive policy in the Atlantic but chose not to for political reasons. There were in fact a number of reasons why there were not more incidents, and one of the main ones was that Hitler had ordered his Navy to avoid attacks on American vessels.68 The principal goal of the naval operations, moreover, was to make sure that the ships got through, and it made sense, given that the German codes had been decyphered and that the naval authorities could listen in on German naval communications, to re-route the convoys so as to avoid the German submarines.
And there is a third factor that needs to be taken into account: the U.S. Navy at the time was simply not strong enough to pursue a really aggressive strategy. Reiter says that “Roosevelt did not unleash the US Navy in a brazen bid to provoke a naval incident,” but the Atlantic Fleet was still too weak to be “unleashed” in this way.69 Top Navy officers were frustrated by this situation, but for the moment there was little that could be done about it. Admiral King, for example, as Robert Love notes, “disliked the British strategy of evasive routing and hoped to defend the convoys with escort groups so powerful that they could defeat the strongest U-boat concentrations, but he did not have enough ships to do that.” “Owing to the shortage of escorts in 1941,” Love concludes, “very aggressive patrolling much beyond the perimeter of the convoy and the relentless pursuit of contacts—both highly desirable tactics—were just not possible.”70 Admiral Stark felt much the same way. As he wrote in November: “We are at our wit’s end in the Atlantic with the butter spread extremely thin and the job continuously increasing in toughness.”71
Even so, the Americans had not decided, as a matter of principle, to avoid major confrontations with German ships. As the distinguished naval historian Jürgen Rohwer has pointed out, when the Americans learned that a major German naval vessel—probably the battleship Tirpitz, they thought—would soon try to break out into the Atlantic from its base in Norway, a U.S. task force, consisting of two cruisers, two battleships and a number of destroyers, was sent out on November 5 to intercept and destroy the German ship. As it turned out, the Germans had intended to deploy not the Tirpitz but the pocket battleship Admiral Scheer. That vessel, however, was prevented from leaving its base because of mechanical problems. Rohwer speculates that had those problems not developed, a battle might have resulted which could have forced Hitler to declare war on the United States.72
How does all this bear on the main issues we are concerned with here? It is important to recognize, first of all, that in one major respect Reiter is right. We are not dealing here with a case of deception pure and simple. The public might not have understood in great detail just how the “undeclared war” was being fought. “The Navy is already in the war of the Atlantic,” Stark wrote on November 7, “but the country doesn’t seem to realize it.”73 But people certainly understood, if only from Roosevelt’s speech at the time of the Greer incident, that the United States had adopted a ‘shoot first’ policy and that something serious was going on in the Atlantic. And the public also had the sense the President was trying to take the country into the war. Reiter’s comment in his International Security exchange with Schuessler that Roosevelt’s “pre-Pearl Harbor speeches were quite bellicose” is clearly correct.74 Schuessler had claimed at a number of points in his article that the President “maneuvered the country in the direction of open hostilities while assuring a wary public that the United States would remain at peace,” but while that was certainly true of the period around November 1940 when Roosevelt was running for re-election, and was also true of the period in early 1941 when Lend-Lease was being debated, it was not true of the period immediately before Pearl Harbor.75 At that time, whatever his disingenuousness, and no matter how devious his tactics sometimes were, people basically understood what he was trying to do, and they were willing to go along with it.
This is essentially Warren Kimball’s argument, and I think he, more than anyone else, has put his finger on what was going on. The public, as Kimball sees it, did not want the unvarnished truth. People “understood quite well what was at stake,” he writes, “but refused to ratify the hard decisions.”76 What they wanted to hear “was not what they knew was the truth, but what they wished were the truth. In a sense, they wanted to be lied to. Public opinion, expressed in polls and Congress, saw Hitler as an ever-increasing threat, and someone had to deal with that.” That “someone” was the President. It was his job to “make unpleasant choices and take the blame.” But people had no real right, in Kimball’s view, to complain about the way he did things. “Whatever the tactics Roosevelt used—from disingenuousness to lies—Congress (and thus the public) had ample opportunities to say ‘no!’” But they chose not to go that route. “The entry of the United States into the Second World War, however halting and whatever Franklin Roosevelt’s deceptions and disingenuousness, came about because the American people—as represented by their Congress assembled—agreed to and accepted (no one ever desires) war.”77
But this does not mean that public opinion mattered in quite the way Reiter says it did. To be sure, Roosevelt was constrained by public opinion, but those constraints were not absolutely rigid, and to a certain extent he could work his way around them. The particular tactics he adopted—‘tailoring’ the truth, framing what was going on in particular ways, emphasizing some things and concealing or ignoring others, and so on—could make a real difference in terms of outcomes. Roosevelt did a lot more than just work “within the marketplace of ideas to influence public opinion through persuasion and public speeches”; he also tried to influence what would happen in other, less direct, ways.78 Instead of presenting his policies openly and forthrightly, he preferred sometimes, for example, to just let the truth seep out. A message he sent to Churchill in April 1941 is a good example. In that message, he outlined a series of naval measures he proposed to take, having to do mainly with the patrolling of the North Atlantic. He thought it was advisable, he wrote the British leader, “that when this new policy is adopted here no statement be issued on your end.” It was not certain that he himself would make a “specific announcement”; he might decide, he wrote, to issue the “necessary naval operations orders and let time bring out the existence of the new patrol area.”79 The assumption apparently was that it was much easier for people to accept a reality which they gradually became aware of than to have to make a choice.
Reiter himself gives another very nice example of the way Roosevelt did things. After giving his misleading September 11 speech on the Greer incident—the speech in which the ‘shoot first’ policy was announced—the President, Reiter shows, actually authorized Stark to tell the Congress what had really happened.80 In the speech, as is well known, the president had failed to mention that the Greer (in accordance with existing policy)had been tracking the German submarine and helping a British patrol bomber which had dropped some depth charges on the U-boat. Instead he suggested that the attack on the Greer had been unprovoked and emphasized the defensive aspects of his policy: “We have sought no shooting war with Hitler. We do not seek it now”; “The aggression is not ours. Ours is solely defense.” He also failed to reveal—and this is a much more important point for our purposes here—that the U.S. government had decided on, and had issued orders to implement, a ‘shoot first’ policy even before the Greer was torpedoed. The purely defensive and reactive part of the policy was emphasized, but at the same time the president did explain why the survival of Britain was essential to American security and why it was important to deal with the Nazi danger before the United States was directly attacked. He did make the argument that “the time for active defense is now”—that the mere presence of German submarines “in any waters which America deems vital for its defense constitutes an attack,” and that if German vessels entered that vast area, they would do so “at their own peril.”81 A week later, Admiral Stark, on Roosevelt’s instructions, as Reiter notes, gave Congress a more detailed account of what had happened, and in particular of the destroyer’s role in provoking the German attack.82 It was as though the president wanted to have it both ways. He wanted people to learn, in a relatively low-key way, about how active American policy was, but at the same time he wanted people to be able to view U.S. policy as essentially defensive in nature—to think of the new policy as a reaction to a German outrage and not as something the U.S. government was doing as a basic act of policy. And it is a measure of Roosevelt’s political genius that he was actually able to have it both ways—that the public was able to ‘tune in’ to what he was saying on both levels, that he was able to tell the people both what they wanted to hear (even if, in their hearts, they knew it was only a half-truth) and what they needed to hear.
What Roosevelt was doing was not deception exactly, but he was scarcely defending his policy openly and honestly in the ‘marketplace of ideas.’ Kimball’s discussion of the way the government managed the lend-lease debate is a good example. The administration, he writes, was “forced by tactical considerations to keep the actual subject matter of the debate as narrow as possible.”83 Key officials simply refused to be drawn into a discussion of broad issues of foreign policy. When asked, for example, if it wouldn’t make sense to just “go into the war,” Secretary of War Stimson replied: “I am not going to pursue this line of argument. We are not concerned with it in this bill.”84 “Examples of the Administration’s indirection and evasion in answering broad and speculative questions,” Kimball writes, “are innumerable.”85 Even on the specific issue of whether protecting convoys would be necessary “if Lend-Lease were to be effective,” both Stimson and Secretary of the Navy Frank Knox were evasive. They were both convinced that it would be, but to concede that point, Kimball notes, “might well defeat the Lend-Lease Act itself”—especially since Knox admitted in his testimony that escorting convoys would lead to war. So they simply finessed the issue by focusing on secondary points.86
So the whole idea that democratic institutions place sharp limits on how far the government can go in pursuing a policy that the public does not actively support is not really borne out by a study of Roosevelt policy toward Germany in 1941. It was not, of course, that the government, by adopting deceptive or evasive tactics, could do whatever it wanted, but the particular way it chose to proceed in 1941—the way it controlled the flow of information, the way it influenced the way the information that did come out was framed, and especially the way it at key points concealed its own basic thinking from the public—played a far more important role in this story than Reiter seems willing to recognize. The government was able to pursue a much more active policy than the American people, acting through Congress, would have supported if the arguments had been put to them in a more straightforward and honest way.
Roosevelt and the Road to War with Japan
Roosevelt was perhaps disappointed by the course that events took in late 1941. Hitler was not provoked into declaring war, and the naval incidents that did occur did not do much to mobilize opinion at home. The public would accept an undeclared naval war, but without a direct attack on U.S. territory it seemed the country would not make the massive effort required to win the European war and bring down the Nazi regime. That certainly was the conclusion well-informed British observers came to at the time. As Reiter points out, Lord Beaverbrook had felt even in August that the United States would not come in unless it itself were attacked, and Lord Halifax, the British ambassador in Washington, reported in November that the “present process of going gradually into war” would not produce the “outburst of public anger” which alone could bring about a full-scale U.S. intervention in the war.87
American leaders undoubtedly saw things the same way, and it was for that reason that the president and his top advisors did not view the Pearl Harbor attack as a total disaster. When news of the attack was received, they met “in not too tense an atmosphere,” since, as Roosevelt’s closest advisor, Harry Hopkins, wrote that very evening, “all of us believed that in the last analysis the enemy was Hitler and that he could not be defeated without force of arms; that sooner or later we were bound to be in the war and that Japan had given us an opportunity.”88 Looking back a couple of years later, Roosevelt himself felt that “if the Japanese had not attacked the United States, he doubted very much if it would have been possible to send any American forces to Europe.”89
Could Roosevelt’s policy toward Japan have been designed, at least to some extent, to achieve that result? The President probably sensed very early on that the strategy of provoking a German declaration of war might not work; if he was determined to bring the United States into the war in the very near future, then he might need a second string in his bow; the situation with Japan might be useful in this context. The British were certainly thinking in those terms. As early as October 2, 1940, Churchill “questioned the statement that it was not in our interests that the United States should be involved in the war in the Pacific.” Peter Lowe, the author of the most important study of British policy in the run-up to the Pacific War, quotes that sentence and comments that Churchill “was interested solely in the moment when the United States would become a full ally, believing that intervention in the Pacific would lead to participation in the European war.”90 Other British leaders also seemed to be thinking in ‘back door’ terms. If, however, they were capable of making that calculation (and thus pursuing a policy that made a Pacific War more probable), is it inconceivable that someone like Roosevelt, who shared their basic goal of bringing the United States into the war, might also have been thinking in those terms? But if the president was using the situation with Japan to bring America into the European war, that conclusion would have a direct bearing on the issue we are concerned with here, since this policy, if it did exist, was certainly not openly pursued.91
Most people, however, find it hard to believe that Roosevelt could have pursued a ‘back door’ policy. They tend to conflate that idea with the notion that the president knew in advance that the Japanese would attack Pearl Harbor, but kept that information from the commanders in Hawaii, in the hope that a devastating attack would bring an enraged, vengeful and united country into the war. That claim I view as absurd and baseless, but I think the general idea that Roosevelt was taking advantage of the situation with Japan to bring the United States into the European war deserves to be taken much more seriously. I think, in fact, that this is the most plausible interpretation of Roosevelt’s policy toward Japan in 1941 one can come up with, and the chapter on 1941 in my methods book outlined the sort of thinking that led to that interpretation. Schuessler accepts much of that argument, and Reiter takes issue with it in his article.
The ‘back door’ argument, as I developed it, had two parts. The first claim was that Roosevelt deliberately put the country on a collision course with Japan. The second claim had to do with why he did this: his goal in doing so, I argued, was to bring the United States into the European war. The idea that the policy was deliberate was itself based on two points: first, that Roosevelt was in control of U.S. policy, and that it was he who had decided to impose the embargo; and second, that in doing so he understood what the implications of the embargo were—namely, that it would in all probability lead Japan to seize the oil fields in the Dutch East Indies, and that this would lead to war with America and Britain. With regard to that first point, many scholars used to accept the argument that Roosevelt had lost control of policy, that policy had been hijacked by people like Assistant Secretary of State Dean Acheson, that neither the President nor Secretary of State Cordell Hull even knew for weeks a full embargo was in place, and that by the time they found out what was going on, it was too late to do anything about it, in large part because resuming oil deliveries at that point would look like appeasement. In 1988, however, Waldo Heinrichs argued compellingly on the basis of new evidence he had uncovered that Roosevelt, acting through Undersecretary of State Sumner Welles, with whom he was very close, had authorized the embargo; that interpretation is confirmed by British sources. Reiter does not seem to question this point about Roosevelt making the key decision here. He in fact drives another nail into the coffin of the old argument about the President not being aware until much later that the measures taken in late July were tantamount to a virtually total cut-off of trade by showing that from the outset newspapers “bluntly described” that move as a full embargo.92 If it was being reported that way in the papers, how could Roosevelt not have known about it?
What this means is that we need to concentrate on the second point: did Roosevelt approve the embargo, knowing full well what the consequences would be if that policy was maintained intact? It seems very clear that the President and his associates understood quite well what the implications of the embargo were. Up to the time it was imposed, he pointed out a number of times, both in public and in private, that an embargo would drive the Japanese down to the Indies and that that would mean war. This, he argued during that period, was the key reason for not imposing it. Reiter himself quotes one of those comments, made right before the tough economic measures were adopted in late July, and quotes it in a way that shows that he thinks it reflects the president’s real views. There was a reason, Roosevelt said, why America had so far allowed the oil deliveries to continue: if the oil had been cut off, Japan “probably would have gone down to the Dutch East Indies a year ago, and you would have had war.”93 This view was widely held in the government at the time. Welles, for example, thought that tough economic sanctions would “provoke Japan to war” with America “before long.”94 Stark had opposed the embargo for that very reason.95 And Admiral Richmond Turner, the Navy’s Director of War Plans, thought from the time it was imposed in late July that the embargo, if it were maintained, would make war between Japan and the United States certain.96
Reiter, however, thinks that “Roosevelt wanted to avoid war” with Japan, and that the goal of the tough policy he pursued in the fall of 1941 was “to deter and not provoke.”97 But there’s a basic puzzle here: if Roosevelt wanted to avoid war with Japan, why would he adopt a policy which he knew would ‘probably’ lead to such a war? To get at the issue, the first step is to look at what Roosevelt and his top associates were saying, and Reiter begins his section on this issue by quoting Welles as telling his British counterpart at Argentia that the President’s “chief objective in the Pacific for the time being was the avoidance of war with Japan” and goes on to point out that the president told Churchill at Argentia that “every effort should be made to prevent the outbreak of war with Japan.”98 Perhaps the phrase “for the time being” in Welles’s comment is the real key to U.S. policy at this point. Welles himself, in the discussion Reiter alludes to, referred to the importance of “dragging out” the talks between the United States and Japan “in order to put off a show-down (if such was inevitable) until the time that such a show-down was from our standpoint more propitious.” And Welles insisted that while the talks were going on, there “should not be the slightest relaxation by the United States Government of any of the economic or financial measures of sanction which it had now imposed upon Japan”—even though, as noted above, he understood that the sanctions would lead to war “before long.”99 The President (at least in the U.S. record of his meeting with Churchill) seemed to take a more moderate line, but if he had said anything about the importance of avoiding (and not just postponing) war with Japan, the British did not consider it worth noting in their own record of the meeting. In the British document, the President simply refers to the need to postpone the outbreak of war with Japan for a brief period. He also tells Churchill that he fully intends to maintain the “economic measures in full force”—a point not mentioned in the American document. But even in the U.S. record, Roosevelt made it clear that one of his main goals in conducting talks with Japan was just to postpone the outbreak of hostilities “for at least thirty days”—suggesting that he was not seriously interested in making a real effort to avoid war in the Pacific.100
The issue here turns, however, not on what was said but on what was done, and in their dealings with Japan at the time, U.S. leaders took a very hard line. The Americans were insisting that Japan pull out of China, and after four years of fighting this was a demand the Japanese naturally found hard to accept—and top U.S. officials understood that this was the case. Japan’s leaders, as Welles noted in November, needed to provide “some justification to their own people after four years of national effort and sacrifice” in China. He therefore “could not believe” that the Japanese would “agree to evacuate China completely,” but “nothing less,” he pointed out, would “satisfy [the] United States.”101 It was this policy that put America on a collision course with Japan. Given America’s position on the China issue, there could be no agreement with Japan that would allow the oil deliveries to resume; without oil, Japan’s military effort in China would grind to a halt; but if Japan tried to break the stranglehold by seizing the oil-producing areas in the Indies, that action would lead to war with the western powers. This was a very dangerous situation, and a government that was determined to avoid war with Japan would scarcely have allowed it to develop.
The point about China is of fundamental importance. It meant that the United States was not just trying to contain Japan or to deter that country from trying to take over new areas, but wanted to force Japan out of areas it already occupied. If the goal had been containment, the chances of avoiding war would have been much greater. The Japanese were deeply concerned about the prospect of war with America, and, bogged down in China as they were, naturally wanted to do what they could to avoid an armed conflict with the strongest country in the world.
One is struck, in fact, by the lengths the Japanese were prepared to go in order to accommodate to American power. On June 21 the Americans demanded that Japan’s pro-German Foreign Minister, Matsuoka Yōsuke, be dropped from the government; less than a month later, Matsuoka was forced out. There were, of course, other factors that contributed to Matsuoka’s fall; still, his ouster following that U.S. demand was quite revealing. Matsuoka himself thought that the Prime Minister, Prince Konoye Fumimaro, by forcing him out of the government, was “kowtowing” to the Americans; his dismissal, he said, was tantamount to a declaration that Japan had now chosen to break with the Axis and pursue a pro-western policy.102 And indeed Matsuoka’s successor as foreign minister, Admiral Toyoda Teijirō , did change Japanese policy. He and Konoye now sought to mend fences with the Americans. They made it clear that Japan was willing to forego further expansion, both in the south and in the north (against Russia).103 Konoye then made a major effort to meet with Roosevelt to try to settle the conflict, telling the U.S. ambassador Joseph Grew “with unquestioned sincerity” (in Grew’s view) that “he was prepared at that meeting to accept the American terms whatever they might be.” The U.S. government turned down the offer, and Grew thought a golden opportunity had been lost.104 That American refusal led to the fall of the Konoye government, which was replaced by a new cabinet headed by General Tojo Hideki . But even then the Japanese, spurred on by their new foreign minister Togo Shigenori, tried hard to avoid war with America. Their key proposal during this period was for a kind of modus vivendi—basically a plan to restore the status quo that had existed before the Japanese had moved into southern Indochina in late July and the sanctions had been imposed. They were willing to withdraw from the area they had occupied at that point in exchange for a resumption of oil deliveries. There were, to be sure, some problems with certain provisions in the plan relating to China, but people close to these issues at the time, and later scholars as well, thought those obstacles were by no means insurmountable. But again the proposal was turned down.105
What all this suggests is that the goal could not have been the containment of Japan; that goal could almost certainly have been easily achieved without war. But so many people have said the U.S. aim, in imposing the sanctions, was to deter Japan from making further advances (and in particular from moving north against the Soviet Union) that it makes sense to see what sort of evidence is given to back up that interpretation. In the present case, it makes sense to examine the evidence Reiter provides to support his claim that the purpose of what he refers to as Roosevelt’s “belligerent” policies toward Japan in this period was “to deter rather than provoke.”106
So let’s look at the first three documents he cites to support that point—or, more specifically, to support the point that although “before sanctions were imposed Roosevelt sometimes worried that sanctions might cause Japan to attack, he and other members of his administration also sometimes argued that sanctions might deter Japan.”107 The first document is a July 8 telegram from the British ambassador in Washington, Lord Halifax, to Foreign Secretary Anthony Eden, reporting a conversation Halifax had had with the president. Roosevelt raised the question of whether a full embargo “would work as a deterrent, or whether it would precipitate the Japanese into the Netherlands East Indies, which neither the United States nor Great Britain wanted.” He therefore wondered whether it would be a “good thing” to place Japan “under all possible economic pressure.” There is nothing in this document that shows Roosevelt thought that an embargo would indeed “work as a deterrent”; if anything, he seemed to think it would have the opposite effect.108
Reiter goes on to cite two letters, dated June 23 and June 25, from Secretary of the Interior Harold Ickes to the president. There are in fact three letters from Ickes to Roosevelt dated June 23 in the file Reiter cites, and two of them deal with the embargo issue. In the first, Ickes argued that an embargo would not provoke a Japanese move into the Indies because Japan at the moment was preoccupied with what was going on in Russia—a view the President, as one can see from the reply he sent Ickes the same day (included in that file), simply did not share. Ickes, moreover, went on to suggest in his letter that if an embargo did lead to an armed conflict with Japan, that might be a good thing, because it would bring America into the European war. “There might develop from the embargoing of oil to Japan,” he wrote, “such a situation as would make it, not only possible but easy, to get into this war in an effective way.” In the second June 23 letter, he also seemed to be suggesting that even if using the embargo was not in theory the best way to get involved in the war, it might make sense to go this route anyway, since delay could lead to disaster. “It may be difficult to get into this war the right way,” he told the president, “but if we do not do it now, we will be, when our turn comes, without an ally anywhere in the world.” This is as close to a ‘back door to war’ argument as you can find in the documentary evidence; and nowhere in either letter did Ickes argue that “sanctions might deter Japan.” Nor did Ickes claim in the June 25 letter that sanctions would “deter” Japan from moving into the Indies; his view was that it would not make a difference one way or the other.109
Other examples could be cited, but the only point here is that the evidence supporting the claim that Roosevelt was interested only in containing Japan is quite thin, and it really does seem that he was putting the United States on a collision course with that country with his eyes open. But this is profoundly puzzling. Why would he do so when his real goal was to defeat Hitler? Why would he want to get involved in a war with Japan when it obviously made more sense to concentrate America’s efforts on dealing with Nazi Germany? And why did his policy change when it did—in late July, a month after the Nazi invasion of the USSR? Up to that point, he had opposed sanctions with the argument that they would drive the Japanese down to the Indies and that would mean war in the Pacific, but now he was accepting a full embargo. Up to that point, the United States did not like what Japan was doing in China, but was not going to go to war with it to force it out of that country. Now the China issue was the real bone of contention between the two countries. Why such a dramatic shift in policy at this particular point?
The test of any historical interpretation of U.S. policy in late 1941 is its ability to answer questions of that sort, and this brings us to the whole issue of the ‘back door’ theory. Was it in order to bring the United States into the European war that Roosevelt deliberately put America on a collision course with Japan? Reiter, of course, does not think so. He argues that even though Roosevelt was pursuing what he calls a “belligerent” policy, his “hard line” had broad public support. “The public recognized,” he writes, “that the July oil sanctions were tantamount to an embargo, meaning there was no deception.” The policy was not secret, and the country actually supported it, knowing that it involved a certain risk of war. As Reiter argued in his International Security exchange with Schuessler, “whether or not Roosevelt actually sought war, his hard-line policy against Japan was popular. He was not out of step with public opinion by engaging in potentially provocative foreign policies, and therefore not engaged in deception.”110
But even if Reiter is right about what the public knew and was willing to accept, that would not mean that people understood what Roosevelt’s strategy was and supported his policy knowing where it would lead. The polls Reiter cites, for example, do not show that people understood that the embargo would probably lead to war and approved it anyway. They mainly show that the public was in favor of a containment policy—of saying to Japan, in effect, that further expansion would mean war. They do not show that people approved of the much more extreme policy of insisting that Japan withdraw from China or, in effect, face war with the United States; it is not even clear from the evidence Reiter presents that they understood that the U.S. government was making that demand.111 Reiter’s argument here, in other words, does not get at the heart of the ‘back door’ interpretation: if Roosevelt was essentially using popular American bellicosity vis-à-vis Japan to get the United States involved in the European war, then showing that the American people were bellicose on the Japan issue would not discredit the ‘back door’ theory. But if that theory is valid, then certainly the President Roosevelt was not being straight with the American public on an absolutely central issue of policy.
The fundamental issues here relate to the basic concept of an ‘open marketplace of ideas.’ Did people understand what the embargo meant? Did they understand that the administration was interested in a lot more than just containing Japan? Did they see how extreme U.S. policy on the China issue had become—that the Americans were, in effect, demanding a total capitulation on Japan’s part—and how sharply the policy diverged from what the policy had been before the sanctions were imposed in late July? Did they understand the connection between the Japan issue and the European war, and did they think it made sense to get involved in a war with Japan when the real problem lay in Europe? I don’t know the answers to these questions, but I do know they are studiable, and it would be very interesting to see what an analysis of the open sources from the period—newspapers, magazines, Congressional debates, and so on—would turn up.
But although these are all interesting issues, the main point to note here is that Reiter has not really shown that the ‘back door’ interpretation is baseless. He himself argues that Roosevelt wanted to intervene in the European war but was held back by public opinion at home.112 He himself notes that the public favored a hard line toward Japan and believes that the American people were not averse to going to war with that country.113 He is also aware of the fact that “if defeating Germany was Roosevelt’s top priority”—and he thinks it was—then it made little sense, in principle, to get involved in a second war with Japan if it was at all avoidable.114 Taken as a whole, doesn’t all this suggest that the President might have been thinking, very secretly, in back door terms?
To refute that interpretation, one has to provide an alternative and more plausible answer to the key question here: why, if in principle a second war with Japan was to be avoided, did the president choose nonetheless to pursue what Reiter himself refers to as a “belligerent” policy in East Asia?115 There has to be some explanation, and the ‘back door’ theory is one way of making sense of all this. But to get at the issue, you also have to look at whatever alternative explanations you can come up with and weigh them against each other in the light of the evidence. And that evidence has to be largely circumstantial, if only because leaders who use deceptive tactics have a great interest in covering their tracks—a point which, in fact, follows from Reiter’s basic premise about the price that would be paid if the deception were revealed. But if the leadership had a certain interest in suppressing evidence which, if revealed, would show that it was deliberately misleading the public, you cannot study the problem in the usual direct way. You have to use a more indirect approach. You have to figure out what the alternative explanations are and then make a judgment about which one is the most plausible. You can consider, for example, the idea that Roosevelt had no real strategy—that he was just playing things by ear, that he had no clear sense for where he wanted to go or how he proposed to get there—and then decide what to make of it. Given the extraordinary importance of the issues he was dealing with, isn’t it likely that he had developed at least a general sense for how he proposed to achieve his most fundamental goals, even if he believed he had to remain quite flexible in terms of how that policy would in practice be implemented? A judgment of this sort would not just be based on a study of Roosevelt’s policy toward Japan in 1941. It would also be based on something more fundamental: on your sense for what Roosevelt was like as a statesman, and indeed on your general sense for how international politics works.
The Bottom Line
So what’s to be taken away from this exercise? Much of Reiter’s argument rests on certain general premises about how foreign policy is made in democratic states. In his view, public opinion plays an absolutely fundamental role: as he and Stam put it in their book, “for better or worse, democratic foreign policy is driven by public desires.” To be sure, they admit, the leaders of democratic powers can engage in covert operations, use deceptive tactics, and “sidestep public debate,” but only when “the anticipated scale of an operation is sufficiently small.” When it is a question, however, of getting involved in a real war they cannot proceed in that way. They are “deterred from engaging in deception” in those cases because they understand that if an attempt at deception is exposed, “they will suffer heavy domestic political costs.” And that’s why, in Reiter’s view, the public really is in the driver’s seat: “Because elected leaders are deterred from engaging in deception,” he writes, “they cannot circumvent public opinion constraints.”116
Reiter makes a second argument about why openness is so important. Consensus-building is part of the process whereby a government prepares a society for war, and that process helps produce good policy. “An advantage that democracies enjoy,” he and Stam wrote, “is that the process by which the leadership seeks to gather and to generate consent necessarily involves public debate and engages the marketplace of ideas; this helps dismiss poorly considered ideas and bolster better ones.”117 The whole concept of a ‘marketplace of ideas’ plays a prominent role in Reiter’s new article as well, and by that metaphor he seems to have a number of things in mind: the notion that different groups (including the administration) put forth their ideas in an honest and straightforward way, that the public ‘buys’ the best arguments, that bad ideas are weeded out, and that the arguments that prevail in that competition essentially determine policy. And his claim, in dealing with U.S. policy in 1941, is that Roosevelt worked within that framework “to influence public opinion through persuasion and speeches.”118
My own view is of course rather different. I certainly would never argue that public opinion counted for nothing—that by resorting a strategy of deception, a president can do whatever he wants, no matter how the public feels about an issue. But I do think that the ability of the political leadership to shape the course of events is much greater than Reiter would have us believe. In the 1941 case, this was due in part to the enormous prestige of the presidency and of Roosevelt personally. The results of a public opinion poll conducted in late November are quite revealing in this regard. In that poll, when people were asked whether they would vote “to go into the war now” against Germany and Italy or “stay out of the war,” 44% said they would vote to go in and 48% said they would vote to stay out. But when they were asked in the same poll whether they would be “in favor of this country’s going into the war against Germany” if “our present leaders and military advisors say that the only way to defeat Germany is for this country to go into the war,” 70% said yes, and only 24% said no.119 What John Mueller calls the “follower effect” was quite dramatic in this case.120
But this was only part of the story. Roosevelt’s ability to influence the way things turned out also had a lot to do with the specific tactics he used—his ability to frame what was going on in certain ways, to emphasize some things and play down others, and more generally to control the flow of information and to dominate the political discussion. The way the President did things really did make a difference in terms of political outcomes—and not just in terms of his ability to draw America into the war, but perhaps even more in terms of his ability to bring it in as a country united behind its leader and willing to do whatever it took to defeat its enemies. Consensus-building was of fundamental importance—but a consensus could be built not just through honest debate but also in more indirect ways: by emphasizing the more purely defensive, reactive, and ‘hemispheric’ aspects of American policy, by making it seem that the United States was acting unilaterally in defense of its own national interests, by playing down (while by no means ignoring) considerations relating to the balance of power in Europe, and so on. The country could more easily come together behind the president if the issues were framed that way and if evidence to the contrary was suppressed or minimized. And in proceeding that way Roosevelt was astonishingly successful: America went to war in December 1941 as a united country, willing to make a massive effort abroad, and it is hard to see how he could have achieved that result if he had been more forthright and had not used the kinds of tactics he had.
Roosevelt was no dictator, and although the concept of an ‘open marketplace of ideas’ does not come close to capturing what was going on in 1941, it is important to recognize that these great issues were debated vigorously at the time, and that those discussions did play a certain role in the story. But that process did not quite work the way one might think; it was not as though strong anti-interventionist arguments simply strengthened the forces against involvement in the war. They could, in fact, have had exactly the opposite effect. The opponents of Lend-Lease, for example, wanted people to see that if one accepted the administration’s premises, the real conclusion to be drawn was that America should enter the war—that if one bought the administration’s argument, it would in fact be “cowardly” not to intervene militarily. The goal, of course, was to discredit the Roosevelt policy by showing where those premises led. But it is quite conceivable that those arguments helped crystallize the issue in people’s minds—that if they really did believe that “England [was] fighting our battle” then they should do the decent thing and put their own bodies on the line.121 Who knows? The point is simply that public debate can in principle work in some perhaps unexpected ways.
What, finally, does the whole story of America’s road to war in 1941 tell us about the general issue we’re concerned with here? We’re interested in the question of whether the political leadership in a democracy can pursue a policy that the public would not endorse, if the issue were put to them in a straightforward way. We’re interested in the question of whether its ability to pursue such a policy can make a real difference in terms of political outcomes—with whether the government of a democracy can take the country into a war that the country would not choose to go into if it had to make that choice in a more forthright way. How do the 1941 cases bear on these issues? Was Roosevelt’s behavior at the time typical of the way democratic governments behave? How generalizable are our findings here?
We can begin with the case of U.S. policy toward Japan in late 1941. Let’s suppose the ‘back door’ argument is valid. I think it’s the most plausible interpretation of U.S. policy at the time that one can come up with, but other people obviously do not share that view, and for that reason alone a lot more work should be, and can be, done on this issue.122 But for now let’s just assume it’s valid. That basic finding would be very interesting in its own terms, but it really would not tell us much about the larger issue at hand. That degree of deviousness is probably very rare in international political life—or at least that’s the impression I’ve gotten after studying these issues for almost half a century now. And I think part of the reason why people find the ‘back door’ theory so hard to take seriously is that this kind of thing is so uncommon in great power politics.
The way in which Roosevelt managed U.S. policy toward Germany at this time is probably more typical of the way things generally work. I think, for example, of a comment Lord Salisbury made in a June 1876 to Prime Minister Benjamin Disraeli: Salisbury took it for granted that it was important to “hide to the eyes of our own people . . . the nakedness of the sword upon which we really rely” in ruling India.123 I think of the way the British government played up the importance of the violation of Belgian neutrality in explaining why Britain should enter the European war in 1914, even though broader considerations relating to the balance of power on the continent and Britain’s relations with France and Russia played a more important role in shaping policy. I think of the way the French government in 1914 contrived to make it seem to its own people that French policy was purely defensive; the real policy the French government had pursued during the crisis was kept from the public for years afterward.124 Various U.S. cases from the post-World War II period are also relevant here. One thinks, for example, of the way Roosevelt misled the country about what had been agreed to at Yalta, or about the way the country was misled in 1946 about what had been decided at Potsdam the previous year.125 There are, in fact, many cases of official deception—the case of Secretary of Defense Robert McNamara’s Congressional testimony about the Jupiter missiles in February 1963 (rare because it involved outright lying) and the case of the U.S. decision in 1965 to go to war in Vietnam immediately come to mind.126 Indeed, one does get the sense that this sort of thing is a normal feature of international political life—that, as John Mearsheimer argued in his book Why Leaders Lie, governments tend to be more honest with other governments (including rival powers) than with their own publics. This is particularly true in the case of democracies: as I said in my review of the Schuessler article, “it is precisely because public opinion is so important that leaders feel under so much pressure to pander to it by framing their policies in a way that their own societies will find palatable.”127
And why is this kind of thing so common? Geopolitical considerations play a fundamental role in shaping policy, but when policy is presented to the public, governments have to take domestic political considerations into account and therefore need to engage in a certain amount of posturing. They have to emphasize the sorts of considerations that will appeal to people at home and get them to support, or at least accept, what their leaders are doing. That gap between appearance and reality is very natural; the real thinking is rarely revealed in a totally unvarnished way; key evidence about what is going on is often suppressed. Thus Grew pointed out that “the American public was totally unaware of the details of the proposals, efforts and assurances of Prince Konoye,” or of what he himself was reporting at the time, and that the public was therefore “hardly in a position” to judge what the policy should be.128 What goes on behind closed doors, or indeed even just inside the head of the individual at the top of the system, is normally of fundamental importance; historical work in this area, in fact, is based on the premise that there is a good deal to be learned by looking at sources that were secret at the time and were only released many years later—and that that sort of study can utterly transform our understanding of a period. The policy making process, in other words, is a lot less open and less transparent than many scholars would have us believe.
These are all major issues, and it’s important for us to discuss them in a serious way. That’s why I’m so pleased that H-Diplo/ISSF decided to organize a forum on the Reiter article. Debates of this sort are the lifeblood of the profession. There’s of course a lot more to be said about all these questions we’re dealing with here, the historical issues as well as the theoretical ones, but the kind of discussion we’re engaging in here really can play a fundamental role as we grapple with these problems and try to get some sense for how to get to the bottom of them. The truth may be elusive, but the whole process of trying to get at it can be of enormous value, no matter where it leads.
1 Dan Reiter and Alan Stam, Democracies at War (Princeton: Princeton University Press, 2002).
2 Ibid., 3-4.
3 Ibid., 4-9; Dan Reiter and Alan Stam, “Understanding Victory: Why Political Institutions Matter,” International Security 28, no. 1 (Summer 2003) (link), 168.
4 Reiter and Stam, Democracies at War, 23. For the notion of countries “selecting themselves into” wars, see ibid., 10, 11, 138.
5 Ibid., 23-24.
6 See ibid., 144 (for the quotation), 146, 160, 162, 193.
7 Ibid., 146.
8 See especially Michael C. Desch, “Democracy and Victory: Why Regime Type Hardly Matters,” International Security 27, no. 2 (Fall 2002) (link) and Michael C. Desch, Power and Military Effectiveness: The Fallacy of Democratic Triumphalism (Baltimore, MD: Johns Hopkins University Press, 2008). A reply to Desch’s article (link) by Reiter and Stam and Desch’s rejoinder (link) appeared in the Summer 2003 issue of that journal. Note also Risa A. Brooks, “Making Military Might: Why Do States Fail and Succeed? A Review Essay,” International Security 28, no. 2 (Fall 2003) (link), and Stephen Biddle and Stephen Long, “Democracy and Military Effectiveness: A Deeper Look,” Journal of Conflict Resolution 48, no. 4 (August 2004) (link). One of the most interesting contributions during this phase of the debate was also by far the shortest: Kenneth Waltz’s letter to the editor, International Security 28, no. 3 (Winter 2003/04), 181 (link). If the claim made by Reiter and Stam about democracies being good at choosing victims they can defeat were true, Waltz wrote, this would tell “us something frightening about the behavior of democratic countries: namely, that they excel at fighting and winning unnecessary wars.” Five years later another important article appeared: Alexander B. Downes, “How Smart and Tough Are Democracies?: Reassessing Theories of Democratic Victory in War,” International Security 33, no. 4 (Spring 2009) (link). A reply by Reiter and Stam and Downes’s rejoinder were published in the Fall 2009 issue of that journal under the title “Another Skirmish in the Battle over Democracies and War” (link). Many of these pieces were reprinted in Michael E. Brown et al., Do Democracies Win Their Wars? An International Security Reader (Cambridge, MA: MIT Press, 2011). Up to that point, the issues we are interested in here (i.e., deception, secrecy, the manipulation of opinion, and so on) were not the focus of discussion. But in a July 2011 H-Diplo/ISSF Roundtable (vol. 2, no. 12) (link), Reiter, Desch, and Downes did pay more attention to those issues.
9 John Schuessler, “The Deception Dividend: FDR’s Undeclared War,” International Security 34, no. 4 (Spring 2010), 133, 138, 142 (link).
10 Ibid., 143.
11 Ibid., 145; see also 149, 153.
12 For Reiter’s letter and Schuessler’s reply, see International Security 35, no. 2 (Fall 2010), 176-85 (link). Reiter’s new article, “Democracy, Deception, and Entry into War,” was published in Security Studies 21, no. 4 (November 2012) (link).
13 Reiter, “Democracy, Deception, and Entry into War,” 595 (link); see also 601-603.
14 Those arguments were laid out in Marc Trachtenberg, The Craft of International History: A Guide to Method (Princeton: Princeton University Press, 2006), chap. 4 (link). I had also discussed these issues in my article “Preventive War and U.S. Foreign Policy,” Security Studies, 16:1 (January-March 2007) (link), 22-29, and in a piece I wrote for H-Diplo/ISSF on the Schuessler article in 2010 (ISSF article review no. 3) (link).