Presidential unilateral nuclear use authority under Truman

In the United States, the President of the United States, or a designated successor, is the unilateral authority on whether nuclear weapons are used. There are many extenuating circumstances that can transfer or expand that authority (such as predelegation and succession), but the reservation of this right for the President alone has been a hallmark of American policy since the early Cold War.

As this entire website is, essentially, dedicated to this topic, this particular Primer is focused on the historical origins of presidential unilateral nuclear use authority in the first presidential administration of the atomic age. It will not explore in detail all of the other possible angles involved with the question, either as a matter of history or policy.

Definition

Before starting, it is worth clarifying exactly what is meant here by “Presidential unilateral nuclear use authority,” because this is not a common term in the literature. By it, we mean the idea that authority for the use of nuclear weapons by the United States descends specifically and exclusively from the President of the United States. This does not mean that other people cannot be involved in the discussion (“consultation”), or that the President cannot delegate that authority downwards (“predelegation”), or that this authority cannot be transferred in the event of incapacitation (“succession”), but it means that President is the source of said authority for use (and not, say, the military), that the President must give some form (even delegated) of “positive” order for their use, and that no other authority’s policy consent is necessary for this order to be propagated (that is, there is no formal “veto” power vested in anyone else — with defiance of orders considered a separate issue). It also means that any use of American nuclear weapons that does not ultimately trace its authority back to the President is considered unauthorized use.

This might be contrasted with other forms of authority, for example, having a committee delegate that power to the military (such as with the Politburo in the Soviet Union), having a “two-man rule” with regards to authority (requiring more than one form of positive assent before order was valid, such as the Secretary of Defense, which is how many people incorrectly believe the US system is set up), or vesting that power with another individual or organization (say, the Joint Chiefs of Staff or its Chairman).

This idea is particular to the United States (though variations of it have at time been vested in other nations), and is the product of the specific historical circumstances of the early Cold War. This primer entry seeks, in part, to outline the circumstances that led to this particular historical construction.

Asserting control in the wake of Hiroshima and Nagasaki

Though there are sometimes appeals to the “decision” made by President Harry S. Truman with regards to the attacks on Hiroshima and Nagasaki as the establishing moment of this doctrine, this is not strictly the case.

For one thing, Truman’s role in approving the atomic bombing of Japan has long been understood by historians as peripheral: he verbally and informally assented to planning that was already underway, and may not have even understood exactly what was planned (it is unclear, for example, whether he understood that two bombs were to be used in such a short amount of time). As General Leslie Groves, head of the Manhattan Project and the actual author of the July 25, 1945, strike order, later put it, Truman’s decision was “one of noninterference — basically, a decision not to upset the existing plans.”¹ Truman gave no formal written approval of the strike order whatsoever.

The actual formal approval of the atomic strike plan was made by the Secretary of War, Henry Stimson, on July 25, 1945.² Separately, it is worth noting that the US definitely did not reserve unilateral strike authority to the President: under the Quebec Agreement of 1943, the assent of the United Kingdom was required before the United States could use nuclear weapons against a “third party.”³

A more persuasive original location of Presidential nuclear authority is in an order that Truman did give later, on August 10, 1945. On that date, the day after the Nagasaki bombing, General Leslie Groves, military head of the Manhattan Project, wrote to the Chief of Staff, George Marshall, with an updated schedule on future atomic bombing activity. Groves estimated that another atomic bomb could be dropped over Japan in about a week. The response from Marshall was swift: “It is not to be released upon Japan without express authority from the President.”⁴

Truman would tell his cabinet on a meeting that day, “he had given orders to stop atomic bombing.” As his Secretary of Commerce, Henry Wallace, recorded in his diary that day: “He said the the thought of wiping out another 100,000 people was too horrible. He didn’t like the idea of killing, as he said, ‘all those kids.’”⁵

In other words, it was only after the bombs were used that Truman implemented a policy of positive authority: he reserved for himself exclusively the possibility of future use of the weapon. This approach, moving the decision for use out of military hands and exclusively into his own hands, would characterize his attitude towards nuclear weapons use throughout his presidential administration.

Postwar codifications and contestations

The Atomic Energy Act of 1946, which while not specifying anything about the use of atomic weapons, reinforced this when it outlined that all such weapons would be produced by the civilian Atomic Energy Commission (AEC) and kept in their custody by default. The President was given the authority in the act to direct the AEC to transfer weapons “to the armed forces for such use as he deems necessary in the interest of national defense.”⁶

As the military could not use weapons that they did not have physical possession of, this invocation of custody as a control mechanism effectively meant that the President retained a de facto control of nuclear use. Subsequent attempts to get this arrangement changed, instigated by the military and the Department of Defense, are known by historians as the “custody dispute” over the atomic bomb.⁷

Over the course of Truman’s presidency, the US armed forces requested physical access to nuclear weapons many times in the interest of readiness. Truman resisted these requests for the most part, clearly indicating that he was suspicious that the military would treat the atomic bomb like a “military weapon,” rather than a weapon that “is used to wipe out women and children and unarmed people, and not for military uses.”⁹ The “custody dispute” was continued through Truman’s tenure, with Truman never allowing the transfer of more than nine complete nuclear weapons to the military (which were moved to Guam) starting only in 1951.¹⁰

NSC-30: Creating the formal policy of unilateral control

This assertion of Presidential control was not accompanied by a formal policy for some time — which is to say, it was assumed, and part of the “facts on the ground” because of the custody arrangement, in the early Truman administration, but not formally made into formal doctrine for some time. In early 1946, for example, General Hoyt Vandenberg, in outlining a plan for a “strategic striking force” (what would become the Strategic Air Command), noted that:

Without doubt the times of attack and targets for atomic bombs will be determined at a very high level — probably by the President — and the Strategic Air Force Commander will be charged with the responsibility of carrying out the operations. Actual operational instructions as to time and place will doubtless come from Washington.¹¹

So, while the military assumed that it would require policy recommendations from “a very high level,” it was only “probably” the case that the President would be the source of such orders — which is to say, there is inherent uncertainty in the phrasing. The fact that there was no formal policy on the matter is clear even by May 1948, when the Secretary of the Army, Kenneth C. Royall, requested that a formal study be made of the issue of the use of weapons. In the request, Royall noted that:

In addition to the basic question of engaging in or initiating atomic warfare there arises a question of what agent or agency of the Government shall be empowered to authorize actual employment. Employment might be undertaken by the Joint Chiefs of Staff on their own initiative, or the decision to authorize employment might be reserved to the President or to some other agent of the Government. It is important that this authority and responsibility be made perfectly clear in order that there will be no delay in acting in event the United States is subjected to sudden attack.¹²

The State Department, in a formal reply, noted that “The question raised by Secretary Royall… has never officially come before the Department.”¹³

The National Security Council issued a formal recommendation, NSC-30, in September 1948, which indicated that the United States should not have any automatic policy that involved the use of nuclear weapons in war (the question of their introduction would depend on the specific circumstances, in other words), and, importantly, that, “The decision as to the employment of atomic weapons in the event of war is to be made by the Chief Executive when he considers such decision to be required.”¹⁴

NSC-30 appears to be the first formal enshrining into doctrine of the idea that all authority for the decision to use nuclear weapons would originate with the President of the United States. That it took three years after the bombings of Hiroshima and Nagasaki for this to be formalized is not surprising given the context of the period, but it also points to the idea that this location of authority is not inherent to the weapons or the American political system, and that very context indicates the specific circumstances of the Truman administration that were responsible for its creation.

In November 1950, the Truman White House issued a press statement meant to clarify comments it had made about the possible use of nuclear weapons in the Korean War.¹⁵ The press released stated that:

Consideration of the use of any weapon is always implicit in the very possession of that weapon. However, it should be emphasized, that, by law, only the President can authorize the use of the atom bomb, and no such authorization has been given. If and when such authorization should be given, the military commander in the field would have charge of the tactical delivery of the weapon. In brief, the replies to the questions at today’s press conference do not represent any change in this situation.¹⁶

It is notable that the statement “by law” is incorrect — no legal statue directly regulated the authority for the use of nuclear weapons, as indicated already. Appealing to law (presumably this is a stretched interpretation of the Atomic Energy Act of 1946’s guidelines over “custody”) is one of several attempts to give this condition an appearance of being on a firm foundation.¹⁷

In December 1951, the Joint Chiefs of Staff and the Secretary of Defense attempted to formulate a more permanent solution the “custody” issue mentioned earlier. They proposed, in a September 1952 memo approved by Truman, that the procedure for the use of weapons be as follows:

In the event of a positive decision, the President would authorize the Secretary of Defense to use atomic weapons under such conditions as the President may specify. […] The Department of Defense should have custodial responsibility for stocks of atomic weapons outside the continental United States and for such numbers of atomic weapons in the continental United States as may be needed to assure operational flexibility and military readiness for use, subject to [Presidential authorization].¹⁸

Along with clarifying the question of “custody” — and making clear that possession of weapons does not grant the authority to use them — it also specifies a specific chain of command for use for perhaps the first time: from the President to the Secretary of Defense. This should not be confused with a “two-man rule” procedure, as the Secretary of Defense is not, in this arrangement, being given authority or veto power other than that conferred by the President (and in any case, the question of whether the Secretary of Defense could veto an order was not on the table at this time).

In summary: The origins of Presidential unilateral nuclear use authority can be found in the early Truman administration, dating to slightly after Hiroshima and Nagasaki, emerging as a response to Truman’s own desire to prevent the military from having use authority, first in the context of the custody dispute and later in the question of codifying nuclear use policy. Much of this early policy appears to be specifically rooted in Truman’s own conception of both the military’s eagerness to use nuclear weapons, and his own abhorrence, post-Nagasaki, towards their future use. All codification of this policy was done by guidance and directive, not statute.

Considering Truman in relation to later presidents

The Truman administration, as the first presidential administration of the nuclear age, understandably had a powerful influence on the nuclear policies of the presidents who followed. Several of his efforts were effectively reversed; as noted, Truman’s denial of physical custody of weapons to the Department of Defense was gradually eliminated by the Eisenhower administration, and the automatic transfer of new warheads to the custody of the DOD became standard under later administrations. This was not because later administrations were necessarily more eager to give up presidential unilateral nuclear use control, however: this change was due to changing technology (sealed-pit weapons) coupled with a changed sense of the global threat situation (in which the rapid mobilization of nuclear weapons would become paramount).

Under Eisenhower, there were similar “backslides” on the absolute control pursued by Truman, though the legal and policy framework remained essentially the same. For example, the policy of “predelegation,” in which under certain circumstances the members of the US military could be considered to already have presidential approval for the use of the weapons, maintained presidential control as a legal construct even while allowing the military increased autonomy towards use. Later presidents would constrict these possibilities and focus on methods to assure that the weapons could not be used in unauthorized ways.

Despite these changes, Truman’s essential post-Nagasaki assertion that it was the person of the US President who would make use decisions about nuclear weapons would become the defining control policy for the US nuclear arsenal. This was motivated not by legal or Constitutional considerations (these invocations would come after the fact) but out of a deeply-held fear that delegating these powers to other groups, notably the military, would result in the ill-advised use of nuclear arms.

1. Leslie Groves, Now it can be told: The story of the Manhattan Project (New York: Harper, 1962), 265.
2. “Reference your WAR 37683 [the strike order cable] of July 24, S/W [Secretary of War] approves Groves directive.” George Marshall to Thomas Handy, TERMINAL cable VICTORY 281 [CM-IN-24908] (25 July 1945), in Correspondence (“Top Secret”) of the Manhattan Engineer District, 1942-1946, microfilm publication M1109 (Washington, D.C.: National Archives and Records Administration, 1980) , Roll 1, Target 6, Folder 5E: “TERMINAL Cables.”
3. The Quebec Agreement stipulated, among other things, that: “…we [the US and UK] will not use it [the bomb] against third parties without each other’s consent.” “Agreement Relating to Atomic Energy [Quebec Agreement],” (19 August 1943), Foreign Relations of the United States. Conferences at Washington and Quebec, 1943 (Washington, DC: US Government Printing Office, 1970), 1117-1119. By late 1945/early 1946, this had been reduced to an agreement that weapons would not be used without “prior consultation” between the two nations. “Draft Report to the Combined Policy Committee by a Sub-Committee” (15 February 1946), Foreign Relations of the United States, 1946, Volume 1 (Washington, DC: US Government Printing Office, 1972), on 1209. The modus vivendi of 1948, renegotiated the American and British arrangement, rendering the Quebec Agreement “null and of no effect.”  to Lovett, “Draft Agreement Between the Governments of the United States, the United Kingdom, and Canada,” (6 January 1948), Foreign Relations of the United States, 1948, Volume 1, Part 2 (Washington, DC: US Government Printing Office, 1976), 683-685. In 1950, Truman and Attlee agreed to a further watered-down commitment: “The President stated that it was his hope that world conditions would never call for the use of the atomic bomb. The President told the Prime Minister that it was also his desire to keep the Prime Minister at all times informed of developments which might bring about a change in the situation.” “Communiqué Issued at the Conclusion of the Truman-Attlee Discussions,” (8 December 1950), Foreign Relations of the United States, 1950, Volume VII (Washington, DC: US Government Printing Office, 1976), 1476-1479, on 1479. There is much more that could be written about US-UK arrangements; see Paul 2000 for more detailed coverage.
4. General Leslie R. Groves to Chief of Staff George Marshall (10 August 1945), National Archives and Records Administration, RG 77, General Correspondence 1945-1950, File 25 Q.
5. Henry A. Wallace diary entry of 10 August 1945, in Henry A. Wallace, The diary of Henry Agard Wallace, January 18, 1935-September 19, 1946 (Glen Rock, N.J.: Microfilming Corp. of America, 1977).
6. U.S. Congress, Senate, Atomic Energy Act of 1946, Public Law 585, S. 1717, 79th Congress, 2nd Session (1 August 1946).
7. See esp. the discussion in Hewlett and Duncan 1969.
8. Data from L. Wainstein, et al., “The Evolution of U.S. Strategic Command and Control and Warning, 1945-1972,” Institute for Defense Analysis Study S-467 (June 1975), on 34.
9. Truman quoted by David Lilienthal (AEC Chairman) in Lilienthal’s diary entry of 21 July 1948, in David E. Lilienthal, Journals of David E. Lilienthal, Volume II: The atomic energy years, 1945-1950 (New York: Harper & Row, 1964), 390-391.
10. Office of the Assistant to the Secretary of Defense (Atomic Energy), “History of the Custody and Deployment of Nuclear Weapons, July 1945 through September 1977,” (February 1978), 18; exact number declassified only in L. Wainstein, et al., “The Evolution of U.S. Strategic Command and Control and Warning, 1945-1972,” Institute for Defense Analysis Study S-467 (June 1975), on 34.
11. Quoted in L. Wainstein, et al., “The Evolution of U.S. Strategic Command and Control and Warning, 1945-1972,” Institute for Defense Analysis Study S-467 (June 1975), on 38.
12. Kenneth C. Royall, “United States Policy on Atomic Warfare,” (19 May 1948), in Foreign Relations of the United States, 1948, Volume 1 , 572-573.
13. Undated notes from the State Department on Kenneth C. Royall’s “United States Policy on Atomic Warfare,” (19 May 1948), in Foreign Relations of the United States, 1948, Volume 1 , 570-572.
14. US National Security Council, NSC-30, “United States Policy on Atomic Warfare,” Foreign Relations of the United States, 1948, Volume 1, 624-631.
15. On the question of nuclear weapons in the Korean War, see Tannenwald 2007, chapter 4.
16. Harry S. Truman: “The President’s News Conference,” 30 November 1950.
17. It is worth noting the irony that most modern-day justifications do not appeal to law, but to the idea that it is rooted in Constitutionality, namely the Commander in Chief powers of the President. This argument did not emerge until sometime later.
18. Memorandum by the Executive Secretary of the National Security Council (James S. Lay, Jr.) to the Secretary of State, the Secretary of Defense, and the Chairman of the United States Atomic Energy Commission, “Agreed Concepts Regarding Atomic Weapons,” (10 September 1952), Foreign Relations of the United States, 1952-1954, National security affairs, Volume II (Washington, DC: US Government Printing Office, 1984).
19. Data for 1947-1959 is from L. Wainstein, et al., “The Evolution of U.S. Strategic Command and Control and Warning, 1945-1972,” Institute for Defense Analysis Study S-467 (June 1975), on 34. Date from 1960 is extrapolated from the stockpile size at the time, and the following information from Roman 1998: “When Eisenhower left office in January 1961, the nuclear stockpile consisted of over 18,000 weapons with yields ranging from a kiloton to several megatons. The military possessed physical custody of 90 percent of the stockpile, with some unknown number in the hands of operational commanders.”

Topics