Jeffrey LewisNational Security Presidential Directives

The National Security Presidential Directive process is essential to making and implementing foreign policy. Every Administration since the National Security Act of 1947 has made extensive use of such documents to coordinate and implement policy directives, as is evident for the following table.

Presidential Directives Since 1960

President Decision Number
John F. Kennedy National Security Action Memorandum 001 – 272
Lyndon B. Johnson National Security Action Memorandum 273 – 371
Richard M. Nixon National Security Decision Memorandum 001 – 264
Gerlad R. Ford National Security Decision Memorandum 265 – 348
James E. Carter Presidential Directive 001 – 063
Ronald W. Reagan National Security Decision Directive 001 – 243
George H. W. Bush National Security Directive 001 – 079
William J. Clinton Presidential Decision Directive 001 – 075
George W. Bush National Security Presidential Directive 001 – 041

Imagine my surprise, then, when the Washington Post quoted national security advisor Steve Hadley this way:

On Iran, for example, he says that the Bush administration reached consensus several years ago on a tough approach, but that when details of bickering over a formal National Security Presidential Decision leaked to the press, “we decided to shelve the NSPD and go ahead and implement the policy.”

This was just Hadley trying not to look like an ass in the Post. No NSPD means no policy. In another context, senior officials explained that Condoleeza Rice refused to answer whether the United States favors regime change in Tehran because the Bush Administration is still working on an Iran policy.

Policies are usually expressed in an artifact like a National Security Presidential Directive (NSPD), Jeffrey Richelson argued in his essay, “Presidential Directives and National Security Policy” (Presidential Directives on National Security from Truman to George W. Bush, Volume II), because:

Putting down on paper presidential decisions in the national security area has several virtues. First, it forces a single statement of policy in a particular area. Second, it provides a means to communicate that decision to agencies which will be responsible for implementing it. Third, in the case of disputes as to what the policy is the directives can be used as a point of reference.

Let’s look at the arguments in more detail. Formal Decisions …

  • force a single statement of policy. What Hadley dismisses as interagency “bickering” is actually the process of formulating a policy that reflects the range of advice available to the president. The creation of a document, then, is an organizing exercise. In the absence of a policy, internal divisions can immobilize a government—a situation perfectly captured by a pair of comments about arms control at the end of the Reagan Administration. “Even if the Soviets did not exist, we might not get a START treaty because of disagreements on our side,” one official told Strobe Talbott. Another official told John Newhouse, “If they came to us and said, ‘You write it, we’ll sign it,’ we still couldn’t do it.” (Way Out There in the Blue 2000, 445)
  • communicate decisions to agencies responsible for implemention. Both Warren Christopher (In the Stream of History: Shaping Foreign Policy for a New Era 1998, 9) and Henry Kissinger (American Foreign Policy: Three Essays 1969, 22-23) have noted that formal decisions by the President guide implementation by individuals not involved in policy formulation. Even participants in decision making process may find that the first draft of a presidential directive may differ from their recollections. Cyrus Vance, for instance, found that NSC documents “when the summaries or PDs—with the president’s marginal notes, or his initials or signature—arrived back at the State Department by White House courier, I found discrepancies, occasionally serious ones, from my own recollection of what had been said, agreed or recommended.” (Hard Choices: Critical Years in America’s Foreign Policy 1983, 37).

  • serve as a point of reference. Once a policy has been “decided”, dissatisfied parties will begin to work to undo the policy. George Shultz, for example, found that hard-liners on the NSC staff were attempting to block an initiative to improve relations with the Soviet Union. Obstruction, Shultz concluded, “would be difficult to revolve unless President Reagan paid continuing attention once he had made a decision.” Formal written decisions allow a President to do just that: On one occasion, Shultz “reminded the president that in his formal directive on Soviet Relations, National Security Decision Directive 75 (right), issued on January 17, 1983, he had endorsed the exchanges idea, arguing that an adequate formal framework for exchanges was the only way to … penetrate the Soviet Union with our own ideology.” (Turmoil and Triumph, 275-276).

Concern about the use of National Security Presidential Directives has focused on the practice of classifying the documents, which confines coordination to parties with access to the decision and diminishes the involvement of Congress. GAO conducted a pair of studies in 1988 and 1992 to consider the impact of reliance on presidential directives in formulating national security policy.

The lack of an NSPD on Iran is a disturbing sign. In the absence of a formal statement of policy, the President has very little control over the various bureaucracies that will formulate and implement their own policies—often at cross-purposes.

There are some positive signs that an ongoing trio of intelligence reviews concerning Iran’s clerical regime, economy and nuclear issues will lead the Administration to decide on a policy. But I wouldn’t hold my breath.

Arms control advocates favor laborious treaty negotiations for precisely this reason: the process of negotiations can force a reluctant government to initiate internal deliberations that might otherwise have remained neglected. The Bush Administration’s evident inability to coalesce behind a coherent Iran policy underscores the need to bring the United States into formal negotiations regarding Iran’s nuclear programs.


  1. Brad (History)

    Great post; thoughtful and informative. Thanks and keep up the good work.

  2. Cheryl Rofer (History)

    I’ve just started reading Jack Matlock’s Reagan and Gorbachev. The beginning of the book is mostly about this process, a nice example.

    The virtue (to some minds) of not writing things down is that you can change easily, which seems to be an obsession of this administration. Look at the continuing comments (particularly by Rumsfeld) on not giving estimates of the costs of the Iraq war to Congress. The excuse is that they don’t know what the eventual cost will be.

    Business and government deal with this kind of eventuality all the time. Responsible people make an estimate, then come back and say they were wrong. You get a little beaten up when you are wrong, but that usually depends on how wrong you were and what the political axes are.

  3. JLo (History)

    I’ll echo the previous comments here: this is a post that has managed to transcend the form cogently and elegantly, and it is a forceful indictment of the current administration’s use of its organizational structures and mechanisms. Cheryl’s observation about the flexibility or fiat afforded by refusing to put a decision into the public record is perhaps the only reason one could find for such a move. Democracy works best with transparency and clear communication, but that doesn’t appear to be the cause being advanced here.

  4. EARL (History)

    To me, the lack of a formal policy document is in perfect synch with the style of this admin. It gives them room to do anything, to deny everything, and to say anything. All the details coming out now about the intel interrogation ‘standards’ just plays to the same tune. I really think we are turning into the Soviets, and will come to the same end.