Essays by Herb Meyer
National Review Online — October 17, 2002
To really understand why our country's intelligence service failed so miserably before the September 11 attack, you must go beyond the bits and pieces uncovered thus far by the House and Senate Intelligence Committees. Yes, there were reports from agents out in the field which might have made a difference had they been acted on by headquarters-based officials. Yes, the various intelligence agencies failed to coordinate. Yes, they relied on outmoded and mediocre information-handling technologies. Yes, the CIA and FBI failed to connect the dots.
But why did they make all these mistakes, and by doing so leave us so vulnerable? Until we come to grips with this root question, we won't really get to the bottom of it. There is an answer. But saying it out loud is so impolite, so impolitic — so downright rude — that no one wants to do it. But if we hope to prevent another attack, we really must set aside all these considerations and talk bluntly. So, here goes: Our intelligence services failed because their leaders and their top-level analysts just weren't smart enough to figure things out early enough to make a difference. They had lots of energy and dedication, to be sure. What they didn't have enough of, was brains.
To grasp the nature of what really went wrong — of what led to the worst intelligence failure in our country's history — you must understand that in a very fundamental way, intelligence is like science: Success depends utterly on having the most brilliant people studying a problem. Only they will know how to go about finding the right answer.
A scientist does not simply walk into his lab each morning and start mixing chemicals, or scribbling equations on the blackboard. Instead, relying on knowledge, judgment, and instinct, he develops a working hypothesis — an idea of what he thinks is how some aspect of the physical world works. In other words, he makes an intuitive leap. Then he designs an experiment to test his thesis. For example, in 1915 Albert Einstein published his General Theory of Relativity, which held that light is affected by gravity. He postulated that light coming from the sun therefore would be "bent" as it passed by Mercury on its way toward earth. It was not until 1919 that Arthur Eddington — a British physicist armed with the right camera for the task at hand — was able to take photos during a solar eclipse that proved Einstein was right.
Moreover, as geniuses like Einstein and Jonas Salk remind us, in science there is just no substitute for brains — for sheer intellectual firepower. Let's face it: The cure for cancer won't be found by average researchers. It will be the exceptional ones who do it. That is why scientific research institutes seek out the smartest people they can find. And that is why they invariably seek out someone who is even more brilliant than these exceptional scientists to manage them — to decide which of their proposed experiments to back and which to stop. It isn't a coincidence that so many leading research institutes are headed by Nobel laureates. Of course, you want a Nobel laureate with good administrative skills. But it's his achievements — his track record, his proven instinct for what will turn out to be right — that matters most.
Put together a group of brilliant scientists, and have them managed by someone even smarter, and you are on your way toward solving the problem. They will know what equipment and lab tools to purchase. They will have good judgment about what information to focus on and what to ignore. They will know which conferences to attend and which to avoid. They will go about the business of coordinating with other institutes doing similar research. Will they make mistakes and reach dead ends? Of course. But if a solution is there to be found, they will find it — perhaps not as quickly as we would like, but sooner than it could have been found by less-brilliant people.
It is precisely the same in the intelligence business. You don't "do" intelligence by generating a huge flow of information, then reading through it until your eyeballs bleed, hoping to find something — anything — that will be useful. In the real world there is so much information flowing in — so much noise — that your chances of finding something useful, or of recognizing something you do find as useful, are hopelessly slim. You will miss important items, or not see them until it's too late.
In fact, you "do" intelligence by taking precisely the same crucial steps before you start sifting through the information that a scientist takes. First, using your own experience, judgment, and instincts, you adopt a working hypothesis; you decide what you think is really going on. Second, you figure out what sorts of things you would expect to find if your hypothesis is correct. Only then do you start sifting through the information. And you had better be sure you have the world's best information-handling technologies, because without them you haven't got a chance of sorting through the torrent of data. Then, if your hypothesis is correct, you will quickly find the kind of information you expected to find, not only because it will be there but, more importantly, because you were primed to be on the lookout for it. (And if you fail to find any of the things you expected to find, that's an indicator your hypothesis was wrong.)
THE REAGAN MODEL
Anyone studying the history of U.S. intelligence will find dozens of examples of how all this has been done successfully. Cheap shots aside, we have had brilliant analysts at the CIA over the years, and from time to time the agency has been led by directors with first-rate minds. Allen Dulles, for example, or John McCone, Robert Gates, and James Woolsey. During the Reagan administration I worked for William J. Casey, and even Casey's worst enemies conceded that his intellectual firepower was awesome. Here's one example of how, under Bill Casey's leadership, the dots got connected in time to make a difference:
When President Reagan took office, the CIA's hypothesis was that the Soviet economy was growing at a three-percent annual rate. To anyone who had seriously studied the numbers — or who had spent an hour walking the streets of Moscow — this was nonsense. Part of my job was to set out and explore an alternate hypothesis: namely, that the Soviet economy had begun to shrink and was on the verge of collapse. More precisely, Casey ordered me to ask this question: If in fact the Soviet economy has begun to shrink, what should we expect to see?
The career analysts responded by digging in their heels; by insisting that the Soviet economy was growing steadily and dismissing the alternate hypothesis as unworthy of serious attention. So I wrote a lengthy "think piece" memo that simply made the assumption that the Soviet economy was shrinking, then outlined what the downward spiral would look like. Casey made sure that just about everyone inside the CIA and elsewhere in the intelligence community read that memo — and knew that he, personally, thought its radical hypothesis might be correct.
Now, of course, the very analysts who had scoffed at this new hypothesis started jumping in with their own ideas — often quite insightful, by the way — of what we might expect to see if the hypothesis were correct. For instance, we would see signs of discontent among the population. Moscow would start shifting military spending to the civilian sector in ways that helped the economy the most but reduced military power the least, such as by transferring steel from tank production to manufacture of (badly needed) locomotives. We would see increased efforts to purchase Western technology for civilian factories, to boost productivity without adding labor, which was increasingly scarce as birth rates plunged. We would see increased sales of Soviet oil and gas to Europe, to increase hard-currency earnings with which to purchase this technology. And so forth.
Literally within days, analysts throughout the intelligence services — not just at the CIA — began coming to me with bits and pieces of information that had never before surfaced. One afternoon an analyst handed me a report on the growing number of strikes — strikes! — at Soviet factories. I asked why this information hadn't surfaced before. He shrugged and replied, "No one here was interested. It just didn't fit." That report was on Casey's desk in five minutes, and on the president's desk later that same afternoon. Likewise with an astonishing report that another analyst brought to my office one day recounting an episode in which Soviet workers literally stopped and surrounded a train that was carrying meat. Troops arrived and surrounded the workers, and the standoff had to be resolved by the Politburo itself — which decided to allow the workers to offload and take the meat, rather than risk a shootout.
Casey delivered this knockout report to President Reagan in person. Then he ordered me to meet with analysts throughout the intelligence community, not only in Washington but overseas, to make sure they knew the director and the president wanted anything that might provide more evidence of the Soviet Union's economic troubles — and we wanted it now. And Casey ran interference by making sure that no deskbound bureaucrat would sit on whatever came in, whether from some analyst in the basement or from one of our clandestine agents abroad. Moreover, he brought in new analysts to replace some of those who just didn't "get it." It made for some nasty episodes in the office, but the raw stuff started reaching us fast and we started pulling the dots together into a pattern.
As the evidence accumulated, we weaned the CIA from its original hypothesis to the new one. Well armed now with solid and growing evidence that the Soviet economy was imploding, President Reagan had the confidence to move forward with his strategy of pushing the Soviet Union to its breaking point.
DEALING WITH AL QAEDA
Of course the threat from al Qaeda was different, but the CIA's approach should have been the same. Given all that had happened during the years preceding September 11 — the World Trade Center bombing in 1993, the attacks on our embassies, the attack on the Cole, and all those bits and pieces that have leaked out about al Qaeda's interest in killing Americans with airplanes — the CIA should have had as its working hypothesis the idea that al Qaeda had developed the strength, the sophistication, the planning skills, and the financial muscle to organize and execute a major attack here in the U.S., possibly by hijacking flights. It didn't. Indeed, it seems the agency had no working hypothesis at all — which is a management failure of the first degree.
If the CIA had developed the correct hypothesis, it could have outlined a list of things that might be expected to happen, such as men from the Mideast enrolling in flight schools or paying cash for one-way tickets. And it could have passed word to everyone at the agency — and at the FBI, and at friendly intelligence services abroad — to be on the lookout for anything that might be relevant, with orders or requests to pass that information up the line immediately. And to be absolutely sure to stay on top of all this, the CIA and FBI should have kept abreast of the information revolution. More precisely, they should have brought into their agencies the kind of information-handling technology that now is available in the private sector, and which is several generations ahead of what our intelligence services still rely upon.
It's clear that all this never happened, either. Nor could it have happened, because in the years before September 11 analytic brilliance was not among the qualities of the CIA or FBI directors. It was not sought by the president, and it was not demanded by the congressional-oversight committees. And because of that, these decent but merely average officials failed to bring on board the kind of brilliant analysts capable of developing a correct hypothesis, selecting the most advanced equipment to handle the information flow, coordinating with other agencies in Washington, throughout the country, and overseas, and in the end connecting all the dots early enough to — perhaps — prevent the attack.
None of the proposals currently bouncing around Washington to fix our intelligence services by reorganizing them, by shifting collection and analysis functions from one agency to another, or even by setting up a new intelligence unit within the proposed Department of Homeland Security, will solve the problem that led to the September 11 failure. Intelligence is people, not org charts. The only solution will be to locate and recruit the most brilliant analysts our country can produce, and then to assure that they are led by someone even smarter.
Happily, finding the analysts we need is easy. Our country is fairly bursting with them. They work on Wall Street, in Silicon Valley, and everywhere in between in industry, in academia, at think tanks, in journalism. More often than you might imagine, they are working by themselves or in small companies that develop and market niche technologies to a select group of clients and customers. In normal circumstances people like this would never be willing to take government jobs. Moreover, any agency that hired them would soon be driven nuts by their energy, their drive, their seemingly off-the-wall ideas, their sometimes bizarre work habits, even their tempers.
But war changes everything. The smartest Americans are just as patriotic as those less intellectually gifted, and they want — no, they crave — an opportunity to put their talent to work. Indeed, during World War II it was precisely this sort of creative, high-energy, sometimes oddball patriot who fairly jumped at the chance to use his or her talents when sought out by William Donovan, head of the Office of Strategic Services. Under Donovan's remarkable if sometimes erratic leadership — not for nothing was he known as "Wild Bill" — the OSS not only helped win the war but created a legend that no intelligence service, anywhere, has yet to match.
Finally, to point out the obvious, none of this will work unless the president chooses as his intelligence chief the kind of man or woman who can manage all these hotshots and get the job done. The problem, once again, isn't any shortage of such people. Rather, in this case, it is the natural reluctance of the president — any president — to choose one. By nature, the kind of intelligence chief capable of really doing the job — a Bill Donovan or Bill Casey, if you will — is hard to handle. They tend to be pushy, arrogant loners who don't play nicely in the bureaucratic sandbox that is our nation's capitol. Dealing with this sort of intelligence chief will require a president to be firm and patient. Above all, he will need to be tolerant of all the yelling and screaming he will get from those members of Congress outraged by what they will perceive to be a lack of respect and deference, but which — as anyone who has served on the board of a top-flight research lab can ruefully attest — is what always happens when people who are "merely intelligent" and experienced in the ways of the world try to oversee the work of people who are truly brilliant in one specific subject.
Well, too bad. This is what it takes to really do intelligence — which means to project the future accurately enough, and soon enough, so that policymakers have the option to change the future before it happens. If the president and Congress aren't willing to pay the price — if their egos are too delicate to cope with a brilliant intelligence chief and a first-rate team of analysts — then our next intelligence failure is just a matter of time.