For Iraq the CIA spent a reported tens of millions to prepare an Iraqi covert unit called the “Scorpions,” who were to be a resistance front during the 2003 U.S. invasion. The agents never served in their intended role and instead became actors—guards or translators—in the sordid story of U.S. treatment of detainees. Indeed, CIA teams played a significant role in the abuses that rocked American and international public opinion when they became public. Again the risks came home to roost.

  Iraq led to George Tenet’s demise. The CIA’s flawed claims that Saddam possessed weapons of mass destruction, manipulated by the Bush administration to build political basis for war, became so controversial that the agency ultimately stopped defending them. Tenet left in July 2004, citing personal reasons. Porter J. Goss succeeded him but resigned himself in the spring of 2006. His main claim to achievement, refilling the ranks of the clandestine service, is really something that George Tenet began. Goss spoke of a new climate of risk-taking, but his haphazard methods were a disaster. Langley leaked talent as senior CIA officers departed or retired, impelled by the Top Floor’s animosity. Goss reopened stations, and defenders bragged of graduating the largest class of DO operators—again—in 2005. But the fledgling spooks have entered a world where experienced leaders are fewer than ever and much CIA capability is tied down in ongoing wars.

  Meanwhile a congressional-presidential commission that reviewed all aspects of the 9/11 attacks produced recommendations including the creation of a new post, director of national intelligence (DNI), to take over the community leadership role previously exercised by the CIA director. This became law in the Intelligence Reform Act of 2004. Porter Goss and his successors will now manage only the CIA. For the nation’s first DNI President Bush chose John D. Negroponte, U.S. ambassador to the UN and an NSC official in the first Bush administration. Negroponte’s chief experience with intelligence, as seen here, had been in the Nicaraguan secret war.

  Finally, political action has become a growth industry, but it evolves in new ways. Langley provided money, political experts, and propagandists to assist moderate politicians in the 2000 Serbian elections. Agency officers trained party activists in Bulgaria—and George Tenet made an unprecedented visit there to secure agreement of officials—but the political muscle was provided by semi-private groups like the National Endowment for Democracy, the National Democratic Institute, and similar entities sponsored by both political parties in the United States. That model also served in the 2005 elections in Iraq, where charges of a covert U.S. role have been made. The Venezuelan government has leveled similar charges concerning an election that took place in that country. This kind of activity harkens back to the CIA and the National Student Association, a recycling of the covert past.

  The new model for political action is disturbing from the standpoint of intelligence oversight. Ostensibly private political units have more flexibility than the CIA, as do private security services, which feature prominently in Iraq today. Both kinds of entities offer channels to bypass any legislative controls on covert action. As in Iran-Contra, the possibility of a White House–directed extra-legal covert operation arises.

  Political action has been a tool for secret warriors since the inception of the Central Intelligence Agency. It probably enjoyed its greatest sway in East Asia, where the United States essentially determined events in the Philippines for a decade during the 1950s and 1960s, and in Japan where CIA helped make the Liberal Democratic Party unassailable. Political action failed in Indonesia. In South Vietnam the agency innovated the “less is more” tactic of furnishing only cash plus general advice, and singling out a limited number of candidates in key districts; but the Saigon regime’s determination to retain power no matter what robbed these tactics of legitimacy, negating the purpose of CIA action. In Australia the CIA proved successful in at least one electoral intervention. At this writing, agency activities in South Korea, Singapore, and elsewhere remain shrouded in secrecy.

  Next in terms of the level of Langley’s involvement is probably Latin America in the 1960s, where this record shows the CIA determinate in Bolivia and successful in the 1964 Chilean election. Vietnam-style tactics succeeded in Chilean congressional elections but failed in the vote that Salvador Allende won. Agency activities in Ecuador, Paraguay, and Brazil during this period might comprise another detailed volume: suffice it to say that Washington involved itself in those places as well. In Guyana the CIA successfully turned the Jagan-Burnham election in the direction it desired.

  The most familiar CIA political actions are those in France and Italy during the agency’s first decade. Enthusiasts would insist these were vital missions, that America’s Cold War agency was the one to conduct them, and that U.S. interests could have been served in no other way. The counterfactuals are unanswerable, however. It is impossible to demonstrate whether, given the Western European tradition of parliamentary democracy, the French socialist or Communist parties, or the Italian Communist Party, if in power, would actually have led their nations into Moscow’s camp. When the socialists came to power in France in the 1980s they proved quite ready to continue their country’s alignment with the United States, but that was so late in the Cold War that no conclusion can be drawn. The CIA political actions were successful within their immediate parameters.

  After the question of how one judges success, a second drawback to political action is that it represents a wild card. That is, operators purchase victory for national leaders who may not be the people Washington thinks they are. The case of Guyana illustrates this point well. Having backed Forbes Burnham, the CIA found him unresponsive, and Johnson administration officials were reduced to begging the Guyanese leader to act responsibly. Burnham ignored Washington’s entreaties for almost four decades, periodically punctuated by elections in which he made liberal use of the manipulative techniques the CIA had taught.

  Political action is also a wild card in the sense that activities begun for stipulated purposes may catalyze quite unanticipated events. This occurred in both East Germany in 1953 and Hungary in 1956, when general programs for undermining Russian control of Eastern Europe became national upheavals, leaving the United States vulnerable to charges of instigation and intervention in the internal affairs of other states.

  Failed political action contains an inherent temptation to escalate, as tragically shown in Chile. In crisis situations, presidents can choose to intervene or resist doing so. But in a failed political action the intervention threshold has already passed, at a clandestine and relatively cheap level that obscures the gravity of the decision. When failure occurs it becomes easy to see the United States already “in” the situation rather than to accept the humiliation of defeat. Thus political action should not be viewed as the simple act of seeking political influence that is often portrayed. If considered at all, political action should be placed in a perspective that frames the decision in terms of the suitability of action at successive levels of involvement, not merely at the entry point.

  A related issue is that these kinds of actions exist within a shifting local and global environment. Political action is conceived for a specific purpose at a moment in time. Yet changing conditions and evolving problems lead to corresponding countermeasures, making political action much different than anticipated and more difficult to halt. Both Italy from 1948 to 1968, and Chile from 1962 to 1973 illustrate this difficulty.

  Political action is also irretrievable. Presidents who approve a commando mission or a paramilitary operation may decide to recall the operators before the point of contact, with no one the wiser. With political action, the first bag of cash handed to a foreign national, the first article planted in a newspaper, the first act in other words, bears the seeds of compromise. Even if the action is called off, the evidence of U.S. intervention remains—immersed in a foreign society and not subject to retraction. Since successful action requires sustained association with a broad range of foreign institutions, the evidence multiplies in a way that increases the likelihood of eventual revelation. And the pattern of dispersed knowledge makes it highly likely that the revelation of one piece of evidence, with attendant controversy, will lead to more exposures. Chile again illustrates this problem.

  Political action is frequently an element in paramilitary projects, a side activity intended to contribute to attainment of project objectives. This kind of activity is more bounded and more logically related to concrete goals. The appropriate question should be whether the project as a whole is necessary, and beyond that, whether political action and propaganda measures suitably contribute to it.

  In seeking influence the United States possesses a whole array of resources other than clandestine services. The Voice of America, State Department information activities, public diplomacy campaigns organized by the White House, student exchange programs, foreign aid—all contribute to American influence abroad. An entire range of public-private and private-sector entities now exist that furnish the kind of electoral assistance that was formerly the province of the secret warriors. A legitimate question is whether covert political action programs are either necessary or appropriate in this context, especially in a post–Cold War world. Nonintervention has real value as a principle in international affairs, and political action makes America herself vulnerable to similar ventures. To understand the sensitivity of these issues, one need go no further than the furor in the United States that resulted from reports that China had intervened in the 1996 election in this country, using exactly the same techniques as the CIA in Chile in 1964. It was Judge Webster, not any crank, who argued that covert action must be consistent with law, American values, and public mores.

  Political action is a first resort, not a last one, on any secret warrior’s scale of escalation. Intentions can be morally “just” only by means of excruciatingly devious manipulations of principle, and these operations by definition are not proportional, regardless of indigenous political tactics, because they involve CIA foreign intervention. Control is almost always vitiated since CIA tactics respond to a developing political situation, not to a menu of options. Finally, in almost all cases the probability of success is indeterminate at the moment of decision. By far the better course is not to become enmeshed in these moral ambiguities in the first place.

  America’s most valuable resource is the image and texture of its democracy, its example to the world. The worst aspect of covert political action is that the tool is a clear contradiction of democratic values. Manipulation of peoples anywhere runs directly counter to these professed values. Today, when the Bush administration has made an explicit goal of encouraging a democratic revolution in the Islamic countries, types of activity that subvert democratic values threaten to damage U.S. national interests. Political action should be avoided.

  Writing after the controversy over the mining of Nicaraguan harbors, McGeorge Bundy noted that “the dismal historical record of covert military and paramilitary operations over the last 25 years is entirely clear.” As the NSC adviser to Presidents Kennedy and Johnson and chairman of their Special Groups, Bundy had some basis for his observation.

  The truth is that the record of covert action is not without its successes. Notable among these are the partisan projects during the Vietnam War and Hmong efforts in Laos. A necessary qualification is that the programs were successful in mobilizing paramilitary forces but were not strategically decisive: neither in Laos nor in North Vietnam were adversary forces significantly hampered by the existence or effectiveness of U.S. paramilitary allies. It is also illuminating that in both cases the operations were wartime programs that could count on ample resources, including the expertise of military special warfare forces.

  The CIA programs most often cited as successes are Afghanistan, Project Ajax in Iran, and Project Success in Guatemala. Yet these victories produced only short-term benefits. The effort in Afghanistan, like the one in Tibet, had the characteristics of a Cold War spoiling operation. It lurched to success courtesy of policy entrepreneurs’ demands for escalation, and with the United States ignoring attempts at political solution. Had the operation not happened, there is a fair argument that the rise of Islamic fundamentalism could have been slowed, at least sufficiently to come to grips with the problem before the travesty of the 9/11 attacks. Washington lost a second opportunity to change the equation when it failed to act more openly in Bosnia in the early 1990s, when the United States could have earned credit with the Muslim world that would have further slowed the rise of terrorism, at a time when it seemed the Israeli-Palestinian problem might actually be moving toward a solution.

  Afghan covert action also had other costs for the United States. Those at CIA, State Department intelligence, and elsewhere who worried that Stinger missiles given to the mujahedeen would show up in the wrong hands proved exactly correct. Continuing inability to recover all these weapons led to efforts to negotiate international agreements pertaining to handheld surface-to-air missiles as well as consideration of a multibillion-dollar initiative to arm commercial passenger aircraft with anti-missile devices. Ironically, Washington has had to beg the U.S.-installed government of Nicaragua to perfect its own procedures for safeguarding such weapons.

  Action in Iran in the 1950s sowed the seeds for virulent anti-Americanism among the successors to the shah. Worse, the 1979–1980 hostage crisis ignited another open-ended subterranean conflict similar to the U.S. vendetta against Castro, except that it would be fought out on the field of terrorism, where Washington was far less equipped to deal with it. Iran became the equivalent of electing to fight on ground of the enemy’s choosing. The CIA operations of the 1980s, and the ham-handed tactics of the Reagan administration in Iran-Contra that reinforced Iranian perceptions of the United States as a perfidious adversary, further exacerbated a basically unstable situation. Washington missed a critical opportunity before the shah’s fall, when a chance existed to gain credit in supporting the creation of a moderate Islamic democracy. In Guatemala the overthrow of Arbenz turned the country away from democracy, the averred aim of the covert action. Neither victory materially affected the balance in the Cold War while, disturbingly, failure would have triggered shifts by forcing those nations into the arms of the Soviet Union. In all cases the benefits were short term, the costs long haul.

  A paradigm case for this sort of failure has been Cuba. The vendettas conducted by Eisenhower and Kennedy radicalized Fidel Castro, necessitated Cuban reliance on the Soviets, and converted a traditional U.S. friend into a foe. The covert action also backfired by leading the Cubans and Soviets into nuclear missile deployments, creating a direct military threat to the United States and a crisis that brought this country to the edge of nuclear war. The Cuban-American hostility became entrenched, with further rounds of proxy or direct sparring in lands as distant as Bolivia, the Congo, Angola, Grenada, and Nicaragua.

  Operations against the “denied areas” uniformly resulted in failures and have long since been abandoned. Had these been successful, the Soviets would have confronted the need for a response not confined to paramilitary conflict. The People’s Republic of China was one denied area where operations redounded to the detriment of the United States. Adventures with Muslim warlords and Li Mi served to identify the United States with Jiang Jieshi and the corrupt Kuomintang. The Chinese Communists were handed a propaganda tool, a “foreign devil” to use as a symbol in solidifying Mao Zedong’s control of the nation. The later operation in Tibet, a spoiling effort by definition, was merely incapable of a successful conclusion.

  Indonesia was not a denied area, but there the CIA obtained similar results. Sukarno used the apparent American threat as justification for eliminating vestiges of opposition to his own role. This disaster has escaped more intense examination because the military coup of 1965, with apparent CIA connections, turned Indonesia back toward the U.S. orbit. As with political action in Guyana, the coup installed a dictator who then governed without regard to democratic values or the interests of the Indonesian people as a whole.

  Most of the other paramilitary operations surveyed have been unalloyed failures, perhaps excepting the Congo, where the outcome is still disputed among former CIA officers. The ledger of failure includes Albania, Angola, and the Kurds in addition to efforts in the denied areas. Track II in Chile failed, mitigated by the success of economic pressures and political action that undermined the Allende government. The Chile case also illustrates what did work—manipulation of foreign aid and international cash flows. It should be quite clear that a paramilitary capability is not essential to that technique. The American action proved fundamentally anti-democratic, installing yet another dictator who reigned for decades. Pious theories about the worth of authoritarian dictators aside, the practical effect of American action ran directly counter to professed ideals of making the world safe for democracy.

  Support for military operations has involved a learning curve. At the time of Korea the military wanted to keep CIA out of its business. By 1960, when a joint study group under Lyman Kirkpatrick surveyed U.S. intelligence, the CIA exclusion seemed well established. The reversion of activity to the Pentagon that John Kennedy ordered supposedly closed the matter. But Vietnam drove the military and the CIA back into each other’s arms, and the Iran hostage crisis demonstrated the need for CIA support to military operations. Both Somalia and Bosnia illustrate the continuing demands. Robert Gates became the first director to have an associate DDO specifically for support to the military. Bill Clinton issued a presidential directive on the subject. George W. Bush has the CIA supporting his wars in Iraq and Afghanistan.