Previous IFPA-Fletcher Conferences

National Security Strategy and Policy:
Planning for and Responding to Threats to the U.S. Homeland

October 28-29, 2004
Ronald Reagan Building
and International Trade Center
Washington, D.C.

Dr. Philip D. Zelikow, Director, The Miller Center and White Burkett Miller Professor of History, The University of Virginia, and Executive Director, National Commission on Terrorist Attacks Upon the United States (the 9/11 Commission)

INTRODUCTION BY: Dr. Robert L. Pfaltzgraff, Jr.

DR. PHILIP D. ZELIKOW: The approach I'd like to take in my remarks is a little different from the previous speakers. They’ve all spotlighted for you particular threats on which they think you should focus your attention. And by the way, I agree with them. But I'm not going to offer my particular threat that I'd like you to focus on; instead, what I want to do is focus on the methods by which this country and its government analyzes threats and prepares to meet them.

What I'd like to do is talk about four issues -- first, analysis; second, warning; third, net assessment; and fourth, joint planning.

First, analysis. The analytical approach of the intelligence community tends, as the Commission found, to be overwhelmingly oriented to current intelligence. This is a long-running trend that’s been evident to people who have been watching analytical products as I have for at least 10 to 15 years. It’s been commented on as a damaging trend in the intelligence community for some time. It became much worse during the 1990s.

I had the perhaps unusual opportunity of reading every single article in the President’s Daily Brief on the subjects of interest to us, from the beginning of 1998 until September 11 and its immediate aftermath, and was therefore able to step back and look at the entire corpus on a subject that was clearly important and on what was selected to tell the President of the United States.

We tended to provide the highly classified equivalent of newspaper articles. You get a tip from a source overseas, or you get something that comes out of signals intelligence, and then it’s glossed up with an analysis that provides some context of what this might mean, and that’s intelligence analysis.

We rarely produced synthesis documents. Before 9/11, the last national intelligence estimate done on terrorism at all was done in 1995. It was superficially updated in 1997. No national estimate on the terrorism problem was prepared thereafter, even as the threat moved to the top of George Tenet’s annual threat assessments. Yet, still no national estimate. Indeed, the intelligence community did not have a strategic estimative capability in the Counterterrorist Center. The Community began realizing that it needed to create one, and hired its initial director for that branch in the days just before the 9/11 attack.

The analytical establishment has very talented people overwhelmingly focused on current intelligence. The intelligence community competes with the 24-hour news cycle. Circumstances in a budget-starved intelligence establishment – before 2001 – made it harder to support long-term basic research that doesn’t put articles in the daily, classified “newspapers.” We also observed the tendency to send forward articles that are bureaucratically safe -- bureaucratically safe for the analyst and bureaucratically safe for top level decision makers who can’t be accused of politicizing news items. But the result is a dearth of wide-ranging, synthesizing, strategic analysis.

Second, warning. After Pearl Harbor and the creation of the CIA in the late 1940s, generations of effort were devoted to the craft of warning, especially the warning of surprise attack from the Soviet Union. That craft -- nurtured over the years by practitioners such as Sherman Kent -- distilled over time reasonable methods for how to try to predict surprise attack. The prediction of surprise attack can’t rely on genius. Qualities of imagination are not the qualities usually associated with government bureaucracies. Therefore, agencies must find ways of routinizing and institutionalizing the practice of imagination. All military organizations know this and do it. You can’t rely on “I've just got the brilliant guy who knows how to see around corners.” You've got to break this down into something you can do day in and day out, even with changes in personnel.

And one way of breaking down the craft is, first, to work through the most dangerous forms of likely attack; second, to identify telltale indicators that such an attack is being prepared; third, to collect against the indicators; and fourth, to design safeguards either to block surprise or to buy additional warning time.

What is remarkable about the 9/11 story is not that that long-honed, laboriously developed craft was tried and failed. It was that this craft was not applied to the enemy that in fact was most likely to launch a surprise attack against the United States before 9/11. That is what was so sad about the story. We were writing the estimates applying that kind of craft, say, to North Korea. We just weren't writing them about the enemy that was getting ready to launch the attack.

Did we do the analysis on what were the most dangerous ways we could be attacked? No. Suicide bombers use vehicle bombs. The notion that they might move from truck bombs to other vehicles is not a revolutionary insight. Yet we were then hit by a boat bomb in the case of the Cole; and you look back and find no analyses at the national level about how boat bombs might be used to attack us. Then in the year between the time the Cole is hit and the time we’re hit by plane bombs, there are no analyses of that either.

That’s not really quite right. There were some analyses done on how aircraft could be used as weapons. What you have on the side of the CIA is they will say, “Look, we gave you the following particular reports in which we mention that terrorists are talking about using aircraft as weapons.” Yes, they did. Did they ever write an analysis, then, working through how that might work? No, they didn’t.

Actually, the finest analysis we found prepared in the Executive Branch of the United States government before 9/11 of how aircraft could be used as guided missiles against the United States, and it’s a fairly substantial analysis, was prepared by a trial attorney in the Department of Justice – he thought that the scenario seemed so obvious that he ought to go ahead and develop the scenario so he could work through the legalities of whether you could authorize US aircraft to shoot down commercial airliners. It’s a 30+ page study; this was clearly a trial attorney at Justice who had too much time on his hands. [Laughter] But I guess it does demonstrate the point that this was imaginable.

The FAA’s Office of Civil Aviation Security looked at this, canvassing the ways in which aircraft could be used to attack us, and they concluded that the primary danger was hijacking for hostages. They looked at suicide hijackings, and the analysis on this point, in the spring of 2001, discounted the threat of suicide hijackings because a suicide hijacking would not offer the attacker “opportunities for dialogue.” Because you're in the hijacking for hostages paradigm.

Now, you could have consulted experts on al-Qaeda and asked them what kind of opportunities for dialogue they thought al-Qaeda was seeking with the United States, and they could have given you an answer to that question, but that wasn’t the way those analyses were prepared; those experts weren't consulted.

So that’s point one, how did we work through means of attack from terrorists?

Second, what about telltale indicators? One of the really great breakthroughs on 9/11, the thing that had never happened before, is the terrorists actually flew the plane. So one telltale indicator might be any sign that terror suspects are trying to get trained to fly large jets; that’s a telltale indicator, that’s something that hadn't happened before, you could work that out. Can you collect against that indicator? Yes. Did we collect against that indicator? No.

Did we think about, then, what would be safeguards that might thwart the surprise, or at least buy us more time for warning? No, that effort was never undertaken. We were thinking enough about suicide aircraft attacks so that the Secret Service was planning for the scenario against, say, an attack on the White House; there were local plans of that kind. People could conceptualize it in doing their particular jobs, but as a country we didn’t think about it. NORAD was actually dreaming up exercise scenarios involving such things, but the aircraft always came from outside the United States -- remember Steve Flynn’s point. So therefore, NORAD always had plenty of time as they crossed the Atlantic or Pacific to get here. Of course, that’s just not the way it materialized on 9/11.

So that’s what’s really sad about the warning story, is that we have a craft, we’ve developed a craft; we just didn’t apply it.

Third, net assessment. What is net assessment? Net assessment is when you have analyzed how the enemy may attack you, then you compare that with where you're most vulnerable, and then try to do a net assessment in order to guide the allocation of resources and develop strategies.

As far as I can tell, the US government has not yet developed an effective capability to perform net assessment today. This is in part because we’re perpetuating the same weakness we had in doing net assessments during the Cold War. The chronic problem in the Cold War in doing net assessments of the Soviet Union was the CIA would say, “We only do the ‘red’ side. We’re not allowed to analyze US strengths and weaknesses, because that’s DoD’s province and we’re not allowed to analyze ‘blue’”. So the Defense Department would write the assessment of American strength, the CIA would do the estimate of the enemy, and no one would really do the net assessment problem, except of course in arguments in the Congress, the opinion pages of newspapers, academic journals, and the like.

We've perpetuated that same system. We have TTIC or other organizations that are doing the “red” side, and then we say, “You, Department of Homeland Security, you do the blue side because you're the guys who handle the domestic effort.” In fact, the Department of Homeland Security has the nominal responsibility for net assessment, even though they're not the predominant repository of expertise about the enemy.

This is precisely the bifurcation that Steve Flynn was talking about between domestic and foreign. It’s a telling indicator of how we still haven't taken the step of adapting to the kind of world we live in today. The Commission made some recommendations on this point. That bifurcation is reflected in the separation of the National Security Council and Homeland Security Council, which the Commission regarded as an artificial dichotomy that allows the marginalization of homeland security concerns.

Fourth, evolving joint plans. If I were in your shoes and I was listening to all these people getting up and telling you about all the things you're overlooking and all the things you're not doing, it would be wearying after a while. And frustrating. Because you might agree with a lot of the things you're hearing and you might think, “Yeah, but within the area where I work, I'm doing all I can.”

I'd like you to reflect a little bit about the enormous talent we have in our government, the conscientious people we have working on these problems, and ask yourself: Are we really getting an effort that’s equal to the sum of the parts, that is equal to the $550 billion we’re spending on national and homeland security? I don’t know that our country can afford a whole lot more money than that. Over and over again, the problems are problems of horizontal integration. You go to Steve Flynn’s analysis and look how disproportionate the spending is on Long Beach. But that doesn’t mean the people who are working on San Diego aren't doing their job; they are doing their job. There’s clearly some kind of horizontal integration problem that doesn’t link the DoD requirements to other agency requirements and form joint strategic plans.

The mechanisms of horizontal integration in our government are broken. We have a federal executive management system that operates according to the finest management principles of 1950. [Laughter] We have large, vertically integrated industrial behemoths; Alfred Sloan’s General Motors, if you will. We have not developed the kind of executive management techniques that have actually been developed to solve these problems for more than a generation in all the large conglomerates of the private sector. In general, the government simply doesn’t acknowledge it has this horizontal integration problem or, instead, says “Well, that’s the job of the National Security Council.” Which is roughly akin to saying, really, General Electric ought to manage its problems of day-to-day operational coordination among its major business units by asking the board of directors to sort it out, which then ought to get a very large staff to assist them. And indeed, the NSC staff is growing about as fast as it can. It’s not enough.

So when we talk about things like a national counterterrorism center, which I think should be joined by a national counter-proliferation center, these titles sound nice enough, but what we’re really talking about are prototypical innovations in federal executive management for the 21st century.

If that’s not a good enough idea, think of other ones. We need effective joint operational planning and horizontal integration at the level of day-to-day operational management. It should develop joint plans, continually update and refine joint plans, and monitor the implementation of joint plans across executive departments.

I’ll stop there. Thank you. [Applause]

Questions and Answers

DR. PFALTZGRAFF: Thank you very much, Phil. It occurred to me as you spoke that perhaps this panel represents a beginning for the greater mechanism for horizontal integration, and maybe we can make some contribution in this conference to the horizontal integration that has been so evident in each of the interlocking presentations of this panel thus far.

We now have a few more minutes in the time we have available for discussion, and given the limitations of time, what I would like to suggest, again, is that we ask people to pose their question, do so succinctly if you can, and then I will give each of the members of the panel a very brief opportunity to respond to any of the questions. So let’s move as quickly as we can to get these questions out. Who would like to be first? Please wait for the microphone, and please identify yourself.

CHUCK McCUTCHEON: Hi, Chuck McCutcheon with Newhouse Newspapers. A question for Mr. Zelikow: If there is not going to be any Congressional action prior to the election on the 9/11 Commission’s recommendations, what is your assessment of the prospects for action in a lame duck session, or even next year with a new Administration and new Congress?

DR. PFALTZGRAFF: There’s one more question.

__: Thanks again, gentlemen, for your time and your insights. One of the things you brought up was this problem, Dr. Zelikow, of horizontal planning. A lot of the focus and emphasis has been on federal agencies, government responses and what the government is and is not doing. But a lot of the problems you mention cut across from the government sector to the private sector, and the solutions will have to involve that private sector in what we do. What are your suggestions for getting them into some of this horizontal planning interoperability and finding a cohesive solution as opposed to a federal idea that gets promulgated?

DR. ZELIKOW: I don't know whether I'd like to or not. [Laughter] The question about what happens to the prospects of the legislation if there isn't action before the election, people on the Hill can evaluate that better than I can. It’s been the position of the former Commission that we need to get this work done now, but the two sides are very close, closer than you might be able to tell from the kinds of things you read in the newspapers. That’s why Tom Kean and Lee Hamilton said on Monday that this is doable, that it is actually realistic to get this done right now within days. It’s not clear to me that the prospects get better after the election if material progress is not made in the next several days.

As to the question about how you integrate the private sector into these horizontal plans, the answer is you develop strategic plans that deliberately indicate what the role is for the private sector, and then you develop federal organizations and plans that reach out to them. So, for instance, it’s very important to think about the way the Department of Homeland Security interacts with the private sector; what is the regional and local organization of the Department of Homeland Security; and how do you integrate private firms into different strategic plans. You overtly acknowledge the role that they need to play in your policy. Then you overtly identify which government agencies are going to have which role with respect to those private firms as part of those plans, and you develop a federal and public/private approach on a number of these issues.