The assessment by U.S. military intelligence of the threat capability of potential adversaries drives the procurement process. But with the Cold War long over and with many potential security problems outside the strictly military realm, loosening the connection between threat assessment and military procurement is worth considering. Aside from other benefits, doing so would remove pressure from the intelligence establishment to make our enemies seem larger and more fearsome than they are. This would produce better intelligence.
The greatest threat to the United States, its allies, and its clients are the movements, groups, and allegiance patterns identified as "terrorist." Terrorist threats are feared more for their immediacy, likelihood, and surprise than for their scope. The extent of the current threat is nothing like World War II, which left more than 50 million corpses, nor like the era of confrontation with the Soviet Union, which could have created far more. There is no reason to exaggerate terrorist dangers beyond those posed by conventional and nuclear forces.
After the Cold War ended, public officials quickly admitted that the former adversary's forces had been overstated. Former Defense Secretary Caspar Weinberger announced: "Yes, we used worst-case analysis. You should always use worst-case analysis in this business. You can't afford to be wrong. In the end, we won the Cold War, and if we won it by too much, if it was overkill, so be it." "Worst-casing" is justified by its practitioners as prudent: if one's own forces are sized and positioned to deal with the worst case, they presumably can accommodate all lesser cases. Another reason for this kind of thinking is that, although there is seldom any benefit to an analyst who produces a less than conservative assessment, there is real risk in underestimation.
Political authorities and policymakers identify enemies, adversaries, and rivals of the United States or they make, acquire, or attract them. Intelligence analysts take the current measure and estimate what dangers actual or potential adversaries might represent in the future. Separable in theory, the two functions are hard to keep apart in practice. For one thing, intelligence information is more useful politically than most other official documentation. This is because an intelligence connection to some story, with its implicit aura of secrecy and menace, seems to bestow a kind of credibility that the same information might not have were it derived from a wire service or released by the Department of Commerce. Various administrations have used unclassified versions of intelligence documents to inform, influence, and sometimes scare opinion formers. Such reports feed the engines of public discourse by providing seemingly authoritative sources for experts, the talking heads of television, and op-ed writers.
Exaggerate the Threat
Intelligence-based documents still exaggerate enemy forces much as they did during the Cold War, calling into question the value of intelligence analysis and, more serious, the value of judgments about force requirements that these assessments allegedly justify. China, Iraq, Iran, and North Korea now loom gigantic or dangerously asymmetric as presented. This is not only because there is no U.S.S.R. to compare them with anymore; it also is because of the way their forces are displayed in intelligence-data-based government documents, and in the political statements and press reports that draw on these publications. Threats get exaggerated by dramatizing the dangerous and minimizing or ignoring countervailing evidence and comparative considerations of scale, in numerous ways:
- Go Public. Partisans of a course of action usually can find some intelligence data to support their wisdom. When they cannot, or when the preponderant judgment seems to support an opposing view, attempts sometimes are made to alter intelligence estimates. During the Ford administration, a group of conservative thinkers from outside the intelligence community, led by Richard Pipes and including current Deputy Secretary of Defense Paul Wolfowitz, was given access to the data on the nature and significance of Soviet strategic weapon programs from which a national intelligence estimate (NIE) had been derived. (NIEs are coordinated, interagency efforts, prepared under the authority of the director of Central Intelligence. They deal with questions of national interest rather than with the concerns of the defense establishment or one of the armed services. Effort is made to make NIEs as objective and consensual as possible.) Not surprisingly, this conservative outside group concluded that the intelligence agencies had underestimated the threat posed by Soviet forces and misunderstood the intentions of the U.S.S.R.
Since then, congressional leaders, notably Senator Jesse Helms (R-NC), repeatedly have attempted to amend the Central Intelligence Agency (CIA) authorization bill in favor of independent assessments on dozens of topics, in the apparent hope of either deriving data for political use or casting enough doubt on some current estimate to devalue its unwelcome conclusions. Formally, Congress gets its intelligence information through briefings, notably to members of the House and Senate Intelligence committees. A prime source of intelligence information for Congress, however, seems to be analysts within the intelligence community who leak information to the press and perhaps to individual members or key staffers.
- Accentuate the Negative. During the Cold War, U.S. intelligence portrayed the U.S.S.R. (with its clients) as representing numerically superior and seemingly overwhelming forces, both in internal and published intelligence-based documents. The post-Cold War version consists of admitting adversary military weaknesses while stressing adversary intentions to exploit U.S. vulnerabilities (e.g., to terrorism, ballistic missiles, weapons of mass destruction, computer viruses, drugs). Having identified the many ways we could be hurt justifies tracking how well potential enemies are doing in each assumed threat area. Where we cannot follow such progress with precision, it is safest to assume the worst, so the threat is perceived to grow whether it does or not.
- Stress Motivations. Inferences about the U.S.S.R.'s evil intent, such as its inexorable reach for hegemony or its implacable threat to our allies or to principles we hold dear, supported the evidence presented about Soviet forces without adding to it materially. After the Cold War, our estimates—and, for that matter, public discourse—continue to suggest permanent, implacable attitudes of belligerence by new adversaries. Iran was a valued U.S. client before 1979; after the Iranian revolution in the mid-1980s, Saddam Hussein, who was at war with Iran, was a cooperative element in our own differences with the ayatollahs. Yet U.S. pronouncements and the intelligence documents that reflect them seem to have picked up the Marxist habit of not permitting an adversary any redeeming virtues or hope of redemption. This convenient assumption of semi-permanent enmity allows orderly force planning while perpetuating the adversary relationship.
- Claim Credit for What Doesn't Happen. Virtually no day passes without some official or politician claiming that some act of ours has sent or will send a "strong signal" to some state or other party we hope to influence. We have great expectations about how clearly our nonverbal signals get read, without much evidence to support assertions about accurate readings of our intentions. The concept of military deterrence is based on the idea that particular acts can discourage unwanted adversary behavior. It may be that the deployment of U.S. intermediate-range missiles in Europe deterred the former U.S.S.R. from attacking the continent; it also is possible that the Russians never intended to invade.
- Confuse Presumed or Announced Goals with Their Achievement. Analysts occasionally substitute what adversaries say about what they intend (goals and doctrine) with what they can do. This is either because highlighting these goals helps increase the perception of threat or because goals and doctrine are convenient substitutes for unavailable performance data. As long as forces are weak, under development, untrained, inadequately supported logistically, or uncoordinated with other force elements with which they must operate, most adversary forces are not a threat, only a potential one.
- Confuse Inventory with Capability. Intelligence analyses often compress distinctions between operating items (items deployed in ready military units) and those in inventory or in prospect, counting all such items as representing military capabilities. Yet transforming equipment into capabilities is especially misleading as pertains to countries such as Iran and Iraq, where prestige can be more important than capability and the militaries may lack the resources to acquire supporting equipment for their major weapon systems. No tank, ship, or aircraft represents an operational capability unless it has trained crew, maintenance, support, communications, and, when deployed, protection.
- Compare the Same Items. To anticipate how potential antagonists might perform in modern warfare, analysis should get beyond comparing numbers and characteristics of the same kinds of weapons. Because war is fought increasingly by the integrated effort of a large number of systems, comparing only the same equipment items on each side ignores related or complementary elements that might be more significant in combat than how the particular platforms perform. Pilot training and radar coverage, for example, can have as much or more to do with the outcome of aerial combat than performance characteristics or aircraft numbers. It is more important to know whether the items are in units, are manned by trained crews, have fuel available, enjoy adequate support, and, especially, are near enough to a potential enemy to be deployable.
- Let Precision Suggest Accuracy. Since the end of the Cold War, Congress has become a more active participant in intelligence analysis of China by requesting special reports that tend to focus intelligence analysts' attention on the potential Chinese military threat and on People's Liberation Army modernization. This reduces broader and probably more accurate intelligence coverage of China, notably its national priorities and investment patterns. Congress thereby helps distort the intelligence product much in the way our understanding of the U.S.S.R. and Warsaw Pact was distorted during the Cold War, when we had excellent coverage of force inventory and good warning indicators but failed to anticipate the collapse of the Soviet Union and the Pact.
Change the System
Current circumstances have important implications for military intelligence analysis. First, they imply a shift from bean counting to movement monitoring and resource accounting. Research suggests that economics is a useful indicator of future conflict: political or ideological dissatisfaction turns into conflict when combined with high unemployment, low education and external money, or lootable resources. Second, removing walls between police and intelligence agencies threatens civil liberties at home, as the distinctions between parties suspected of planning terrorism and those holding opinions unwelcome to the authorities erode over time. Dissent is not treason. Nevertheless, law enforcement agencies will need all the help they can get to identify terrorist threats quickly and accurately, including the behavior of terrorist sponsors abroad. This means better liaison between intelligence organizations and law enforcement and investigative bodies at home.
Exaggerating today's threats shortchanges the future. Being more balanced in what we publish about potential enemies, the threats they pose, and the forces they acquire would give journalists, politicians, and the public here and abroad a better understanding of how to compare our forces with those of others. Over time, there could be calmer assessments of other states' weapon acquisitions or other force changes, and less inclination to regard virtually all force improvements by possible adversaries as threats requiring immediate responsive force improvements lest the sky fall.
Interaction between intelligence analysts and academics and other experts outside the community enriches both. However, the politically useful sorts of second opinions about intelligence judgments that Congress seems to want may, in the long term, create the worst intelligence distortions and exaggerations. If second-guessing estimates becomes a habit, the best analysts and the most executive attention will shift from analysis of foreign threats to being right compared with outsiders' estimates, or to shading estimates in the direction of critics' views in the hope of avoiding criticism. Politicizing the intelligence product helps no one. Occasional second guesses that review the data the intelligence professionals used and then delineate to the analysts the points with which the outsiders differ can help. Second guesses that claim to improve the estimate by coming to a different conclusion and going to the less informed public for political vindication do not.
The time has come to decouple defense procurement from threat analysis. With the Cold War over and a completely different crisis in being, a tight connection no longer makes sense. Rather than trying to adduce various creative forms of larger-than-life menace, we must find another basis for force planning. If history is any guide, we again will have a serious, full-scale opponent, and it is best to be prepared. Advances in bioweapons, the vulnerability of the U.S. infrastructure, and the possibility of nuclear or radiological weapons are good reasons to concentrate resources on terrorism—more so than the prospect of North Korean or Iraqi missiles targeting San Francisco. Whether Russia and China evolve as threats should remain something we can influence. This suggests an approach like General Colin Powell attempted with his "base force" plan, which looked toward having the fundamental military capabilities for use against inevitable future adversaries in a post-Soviet world. General Powell believed military staffs would be capable of sizing such forces adequately without a specific threat to measure them against. Such a change could relieve much of the pressure for exaggeration under which intelligence analysts suffer and would improve the quality of the product.
Mr. Hirschfeld is a retired State Department senior foreign service officer and a former arms control negotiator. He was a member of the Secretary of State’s Policy Planning Staff in the Ford and Carter administrations and was senior analyst with RAND Corporation and the Center for Naval Analyses. The author is grateful to the Rockefeller Foundation, which supported preparation of this article.