"Repugnant Philosophy":
Ethics, Espionage, and Covert Action(1)

by Dr. David L. Perry

(Reprinted with permission from Journal of Conflict Studies, Spring 1995.  A slightly modified version was reprinted in Jan Goldman, ed., Ethics of Spying: A Reader for the Intelligence Professional, 2006.)

    There are great occasions in which some men are called to great services, in the doing of which they are excused from the common rule of morality. - Oliver Cromwell

    A real diplomat is one who can cut his neighbor's throat without having his neighbor notice it. - Trygve Lie

Introduction

The sources and methods of espionage, the goals and tactics of covert action, and the professional conduct of intelligence officers are matters typically hidden from public scrutiny, yet clearly worthy of public debate and philosophical attention. Recent academic studies of intelligence that have had any intentional bearing on ethics or political philosophy have largely focused on procedural questions surrounding the proper degree of oversight of intelligence agencies. But what is often missed in such examinations is substantive ethical analysis of intelligence operations themselves.

This gap in the literature may be due in part to the lingering influence of the idea that ethical principles are not appropriate to apply to "statecraft" or international politics, as if doing so one makes a kind of "category mistake." But an amoralist view of international relations clearly cannot be sustained. Whatever interests or rights that states can legitimately be said to have must derive from the interests and rights of their individual citizens. War waged in defense of the nation can be morally justified in ways analogous to and derivative of individuals' rights to resist an assault on their family. And because individuals have no right to murder and steal from their neighbors, there can be no morally sound raison d'état for waging aggressive war. The proper question, then, is not whether ethical principles apply to statecraft but rather how they should be validly applied to statecraft.

Most of us recognize certain basic moral rules to be binding in most cases: tell the truth; keep your promises; care for your family; avoid harming others; respect the rights of others; and so on. Rules like these cannot be ignored even when they prove to be personally inconvenient. But most of us would also agree that there can be legitimate exceptions to otherwise valid moral rules. Sometimes, for example, we may be forced to choose between two or more important moral principles when they can't all be fulfilled at the same time. Occasionally we even face truly tragic decisions, where each of the available options will result in serious harm and we must therefore choose the lesser of evils. Morally right actions don't always produce outcomes that are good without qualification.(2)

Moral conflict can arise in professional life as well as in private life. Professional roles sometimes involve special obligations which can outweigh other important ethical considerations. Investigative journalists may be permitted to misrepresent their identity in order to gain the confidence of a government official they suspect of corruption. Or defense lawyers may be expected to try to undermine the credibility of a prosecution witness in cross-examination even if they know the witness is telling the truth about the defendant.

Some of the virtues required for intelligence work, such as discretion, loyalty and tenacity, are also instrumental to professions like diplomacy, the military, law, business and journalism.(3) But many of the skills and character traits drawn upon and reinforced by the profession of intelligence are very different from those expected of the average citizen or other professionals.

The CIA was created by an act of Congress in 1947, and authorized by the National Security Council the following year to undertake "special projects" (i.e., covert action).(4) An impassioned report to President Eisenhower in 1954 argued that the U.S. faced "an implacable enemy [i.e., international communism] whose avowed objective is world domination by whatever means and at whatever cost," and urged the U.S. to "learn to subvert, sabotage and destroy [its] enemies by more clever" and "more ruthless" methods than those of its opponents. The report conceded that this entailed a "fundamentally repugnant philosophy" and contradicted "long-standing American concepts of 'fair play,'" but it insisted that such an approach was necessary given the grave international situation that existed.(5) More recent advocates of strong U.S. espionage and covert action programs have typically focused on the strategies and methods they deem essential to meeting various foreign threats,(6) from the KGB(7) to contemporary drug lords and terrorist organizations.(8)

I do not question the "just cause" underlying the creation of the CIA in the late 1940s, nor do I think that its legitimacy is up for grabs in the wake of the collapse of the Soviet Union. If the CIA did not already exist, it would be necessary to invent it. Furthermore, the fact that CIA is a lawfully authorized arm of a liberal democracy lends it a legitimacy--a "social contract," if you will--which could never be ascribed to the KGB, for example. Dismantling the CIA, as some in Congress have recently proposed, would endanger U.S. national security. Even as persistent a critic of government secrecy as Sissela Bok nonetheless grants that deception can occasionally be justified in national defense: "Honesty ought not to allow the creation of an emergency by the enemy, when deception can forestall or avert it. . . . Whenever it is right to resist an assault or a threat by force, it must then be allowable to do so by guile."(9)

In some essays written by former CIA officers, it has been suggested that the peculiar nature of the knowledge and expertise of intelligence professionals relative to grave external threats requires an extraordinary or specialized morality.(10) In other words, like politicians who "get their hands dirty" in the public interest, the role of intelligence officers is thought by some to warrant excusing them in that capacity from certain ordinary moral constraints. As Arthur Hulnick and Daniel Mattausch wrote:

Professional standards require intelligence professionals to lie, hide information, or use covert tactics to protect their "cover," access, sources, and responsibilities. The Central Intelligence Agency expects, teaches, encourages, and controls these tactics so that the lies are consistent and supported ("backstopped"). The CIA expects intelligence officers to teach others to lie, deceive, steal, launder money, and perform a variety of other activities that would certainly be illegal if practiced in the United States. They call these tactics "tradecraft," and intelligence officers practice them in all the world's intelligence services.(11)

But can the goals of intelligence truly justify ruthless methods? No doubt Machiavelli, Cromwell, Lenin and others would concur. But it would be highly problematic to grant the profession of intelligence a kind of "strong moral role-differentiation."(12) Of course, we do allow professionals to make certain ethical trade-offs. But professionals are mistaken if they consider their roles to render them immune to moral scrutiny. Physicians may not deceive patients into serving as unwitting subjects of medical experiments even though the knowledge derived from doing so will almost certainly benefit future patients. Soldiers may not maim or slaughter civilians in order to deter their compatriots from harboring guerrillas. Exceptions to prima facie ethical principles must be shown to fulfill more important principles, not simply be assumed to be acceptable due to their being professionally "expedient." An affirmation of the legitimacy of the CIA as an institution does not entail moral approval of every end it might pursue nor every method it might employ.

What follows is an exploration of selected ethical problems arising in the work of intelligence officers. I cannot claim to possess the level of knowledge and wisdom requisite to resolving all of these issues satisfactorily, but perhaps my "spadework" will at least serve to stimulate further debate and reflection.

Voluntary Agents

When U.S. intelligence officers recruit agents in foreign countries, they may have very specific offensive or defensive goals, or they may wish simply to build "assets"--human sources of information and influence--for future use. Richard Bissell, CIA's Deputy Director for Plans from 1958 to 1962, testified before Congress in 1975: "It was the normal practice in the Agency and an important part of its mission to create various kinds of capability long before there was any reason to be certain whether those would be used or where or how or for what purpose. The whole ongoing job of . . . a secret intelligence service of recruiting agents is of that character."(13) Bissell did not estimate what proportion of CIA agents take on that role freely. But that is a matter of significant ethical import.

Foreign citizens apparently become agents of the U.S. Government for a wide variety of personal reasons: ethical concerns about their own governments; the lure of adventure, excitement and secrecy; desire for money; sexual and other blackmail; agents' resentment and frustration regarding their overt careers; or some combination of these factors.(14)

Some agents require little or no persuasion on the part of intelligence officers to engage in espionage on behalf of the U.S.(15) Many agents are motivated by the sheer excitement of spying and the promise of steady extra income. But many others decide to engage in espionage out of deep-seated antagonism toward their native regimes.(16) This was true, for example, of a number of high-ranking Soviet military and KGB officials who either passed sensitive documents to the CIA or who defected when they no longer in good conscience could remain loyal to the Soviet government.

One such agent, a Soviet military intelligence officer named Pyotr Popov, supplied valuable information to CIA during the 1950s apparently out of repugnance toward the KGB's treatment of Russian peasants.(17) Another Soviet defector-in-place, Oleg Penkovskiy, fearing that Khrushchev intended to launch a preemptive nuclear strike, provided U.S. intelligence with thousands of pages of Soviet military documents, including information on Soviet nuclear weapon capabilities that proved vital to President Kennedy's actions during the 1962 Cuban missile crisis.(18) It is perhaps not too far-fetched to believe that in a state characterized by an oppressive political system, espionage intended to undermine that system's power and prestige can actually provide authentic hope to agents and dissident groups. Former CIA officer Harry Rositzke argued that although agents sent on missions against the Soviet Union in the late 1940s and early 1950s "knew from the beginning that the cards were stacked against them," they were nonetheless "highly motivated," having witnessed the effects of Soviet power in Eastern Europe, the Ukraine and the Baltic States.(19)

However, espionage against one's government is considered treason in every part of the world and, if exposed, frequently entails severe punishment for the agent. Both Popov and Penkovskiy (like the many agents betrayed by CIA officer Aldrich Ames[20]) were reportedly executed after their capture and interrogation by the KGB. Thus the fact that an agent is a volunteer does not thereby purge his or her CIA case officer of moral responsibility or liability.

Although witting agents usually have no illusions about the consequences of capture, their covert sponsors may ask them to accomplish tasks entailing greater risk than those of which they are aware or would agree to accept. Rositzke described how a nervous double agent was emboldened to meet with his KGB handler: CIA polygraphed the agent, but then showed him a different graph than his own to convince him that he could successfully withstand a KGB debriefing.(21) Another agent was asked by U.S. officials to organize "a small, tightly knit resistance group" of his military colleagues, but he refused out of fear of the KGB's wholesale infiltration of society. In fact, he wouldn't even provide CIA with the names of anyone who might be a Soviet dissident, fearing that a failed attempt by CIA to recruit any of them could easily "blow back" on him.(22)

A more recent case (1990) also illustrates the ominous consequences of agent exposure:

Two or three undercover agents believed to be working for Israel in a Syrian-based terrorist group were unmasked and killed last fall, not long after the United States gave the Damascus Government information about terrorist activities in the country. . . .

Officials said the Administration argued that [Syrian President] Assad should be given an unusually detailed briefing about the actions of Syrian-based terrorists to impress upon him the weight of the evidence against his Government. Intelligence officials are said to have warned that such a briefing would put undercover agents and methods of gathering information at risk.

"It was quite an argument," said one official who has been informed of the debate. "The intelligence guys finally told them, 'O.K., but the blood will be on your hands if something happens.'"(23)

Agents working against tyrannical regimes or terrorist cells have a compelling ethical claim to have information about their clandestine activities very closely guarded by their CIA handlers.

But some voluntary agents have apparently been regarded as "expendable."(24) James McCargar, a former operations officer, wrote that many American agents have been gratuitously slandered by the CIA upon their termination or "disposal" as agents, presumably to render them less credible should they attempt to publicize their former espionage work.(25) Alternatively, an agent who had stopped producing useful intelligence might be intentionally exposed in order to humiliate the target country's counterintelligence personnel. This is especially tempting when the agent is a mole, since such a revelation would cause agents employed by the target country to worry that their own safety was endangered by leaks. The aftermath of the arrest of CIA officer Aldrich Ames as a Soviet/Russian mole is instructive in this regard.(26)

British journalist Tom Mangold learned through extensive research into the long tenure of James Angleton as CIA's head of counterintelligence that a number of bona fide Soviet defectors and other CIA agents were grossly mistreated--some even betrayed to the KGB--due to Angleton's unfortunate reliance on the bizarre, self-serving opinions of one particular Soviet defector, Anatoliy Golitsyn. To the Agency's credit, though, following Angleton's forced retirement it made efforts to compensate some of the agents (and CIA officers) who had unjustly suffered as a result of Angleton's and Golitsyn's suspicions.(27)

The CIA has also been criticized for building up the hopes of agents beyond what the U.S. Government really intended to support. According to McCargar, U.S. intelligence developed a cooperative relationship with an unnamed Eastern European monarchist group, deceiving them into believing that the restoration of the monarchy was intended by the U.S. (it was not) in order to benefit from the "considerable intelligence" the group provided.(28) In addition, historian John Ranelagh has accused the U.S. of a "cold ruthlessness" in supporting partisans in postwar Ukraine and elsewhere when it had no intention to commit its military forces to save them from being annihilated.(29)

The culpability of U.S. officials has been mitigated, however, in regard to certain covert operations in Poland, Albania and Cuba, where U.S. long-term objectives were defeated by the compromise of its operations and communications by enemy intelligence. U.S. officials were unaware, for example, that British intelligence officers Kim Philby and George Blake were actually Soviet agents who would succeed in betraying numerous espionage and covert action projects and cause the deaths of hundreds of Western agents.(30) The temptation to exploit voluntary agents for purposes of Realpolitik must be considered as a plausible moral risk, though. One is reminded of the ways in which various governments have repeatedly inflated Kurdish nationalist hopes solely in order to place temporary pressure on Iran, Iraq or Syria, only later to abandon the Kurds out of expediency.(31)

Deception and Coercion

When the CIA is unable to obtain voluntary agents, it sometimes "recruits" them, so to speak, through deception. In some cases, people who wouldn't willingly work for the CIA are made unwittingly to do exactly that by passing information to a trusted friend or associate who happens to be in CIA employ but who presents himself as one with loyalties more congenial to the person being duped.(32) This method is sometimes called "false-flag" recruitment,(33) since the recruiter misrepresents the country that he or she is representing. It's essentially a con game, wherein one first ascertains the potential agent's basic loyalties and core values in order to concoct a scheme to persuade him to provide sensitive information without upsetting his conscience or arousing his suspicions.

Miles Copeland said that "[i]f the prospective agent hates Americans," for example, the recruiter "can tell him he is acting in behalf of the French--or the British . . . or some Senator or crusading newspaperman," whatever his conscience is assessed as most likely to tolerate.(34) David Phillips, another former CIA officer, attested that "there are unsuspecting zealots around the world who are managed and paid as spies; they sell their countries' secrets believing all the while they are helping 'the good guys.'"(35)

"False-flag" cases are odd from a moral perspective, since in one respect the agents willingly provide sensitive information, probably knowing that they would be punished if their activities were exposed. But of course the voluntary nature of such action is only superficial, since if the agents knew to whom the information was actually being passed they would most likely not provide it.

Two general types of coercive recruitment have been cited in the literature.(36) In some cases, knowledge of an agent's potentially embarrassing or patently illegal activities is used to extort espionage service. Prospective agents may be confronted with proof of their past crimes and blackmailed into working as spies in exchange for their covert employer keeping such evidence from their own country's police. As a former CIA officer indicated, in many instances the local police would already be aware of such crimes but would cooperate with CIA in not referring them for prosecution.(37) Since this method closely resembles that of the FBI in coercing criminals into becoming informers, it may be seen to be less objectionable than some other methods of agent recruitment. In cases where prospective agents' perpetration of crimes mitigates their right to be free from retributive coercion, the issue of their consent to becoming spies loses some of its force. But this cannot be said to provide a "blank check" to a secret recruiter to coerce a criminal to engage in espionage. (In addition, as I shall argue, there are ethical concerns regarding the agent's society which should not be ignored.)

In other cases, though, embarrassing situations can be created for previously innocent potential agents, and the threat of exposure used to extort their compliance. One technique involves first establishing a seemingly natural friendship with a prospective agent, and gradually "stretching" the person's conscience to the point of accepting tasks that he or she would previously have found unacceptable. Casual requests by the recruiter for seemingly innocuous data evolve subtly to more obviously illegal assignments, until the agent either makes a conscious decision to remain an informant, or continues out of fear of exposure.(38)

Those cultivating the spy will press favors upon him, without, in the initial stages, asking for anything in return. This is clearly a matter in which sensibilities must be catered to in order to avoid giving offense or having one's motives suspect. Reciprocity obliges most people to respond in kind; the trick is to escalate the exchange to the point where a more compromising engagement can be undertaken.(39)

Espionage activity that is initiated in a deceptive manner can thus at some point take on more obviously coercive characteristics. James Angleton reportedly described this method as "incremental entrapment in a subtle web of irresistible compromises."(40)

The degree to which CIA employs blatantly coercive methods in its agent recruitment and handling has actually been a topic of contention among former CIA officers. James McCargar said that since the case officer is dependent upon the actions of the agent, this naturally inhibits the degree to which an agent can be dominated: "To this extent every agent is a free agent." He also claimed that "compulsion is a very limited technique," since the agent thus "is in no frame of mind to exploit his own skills or possibilities to the fullest."(41) Arthur Jacobs similarly argued that "there is rarely to be found any effective means of exercising absolute control [over an agent], even by such lurid devices as blackmail, exposure of offensive relationships or personal habits of the source."(42) Note that neither Jacobs nor McCargar implied that coercive methods would be morally objectionable if they were effective.

If CIA officials indeed concluded that absolute control over an agent was impossible, this was not for lack of trying. For at least two decades the Agency funded experiments using mind-altering drugs, electroshock, hypnosis, sensory deprivation, and other techniques in an elusive quest to find foolproof ways to manipulate agents. Some of the motivation behind these efforts lay in fears that the Soviet Union and China had developed technical "brainwashing" methods that needed to be understood and countered by U.S. intelligence. But sadly little consideration was given to the rights of the largely unwitting human subjects of CIA mind-control experiments.(43)

Even if agents cannot be completely "controlled" by their covert supervisors, we may infer that espionage agents are regarded by their sponsors primarily as means to the end of collecting intelligence. The full range of habits, beliefs, virtues and vices making up the character of an individual agent are to prudent espionage officers merely helps or hindrances to the production of useful intelligence for their superiors.(44) Of course, instrumentalist relationships are common to a wide variety of human endeavors, such as business negotiations and contracts. But the element of crude manipulation that can apparently be present in espionage is what ought to elicit our heightened ethical scrutiny.

E. Drexel Godfrey, Jr., former Director of Current Intelligence at CIA, strongly criticized CIA methods of recruiting agents, stating that CIA officers are "painstakingly trained in techniques that will convert an acquaintance into a submissive tool . . . shred away his resistance and deflate his sense of self-worth."(45) Miles Copeland, expressing a more sanguine view, claimed that CIA uses coercion in agent recruitment "only when there is a good chance of converting it into positive motivation":

As quickly as possible, the principal [an intermediary between officer and agent] must enable the agent to deceive himself into believing that he would have become an agent even had he not been caught with his pants down, and that what he is doing is justifiable on its own merits.(46)

Moreover, Copeland said, the agent must be persuaded that the government employing him in espionage regards his safety as more important than any particular piece of information he might forward.

Maintaining such an attitude might occasionally mean passing up some item of tremendous importance, but in the long run it pays off because it keeps the agent feeling safe and happy and maintains his productivity over a long period of years.(47)

William Hood has written that an element of control is not simply desirable but imperative in agent recruitment:

No espionage service can tolerate the merest whiff of independence or reserve on the part of an agent. . . . With a new agent, the case officer's first task is to maneuver him into a position where there is nothing that he can hold back--not the slightest scrap of information nor the most intimate detail of his personal life. Until this level of control has been achieved, the spy cannot be said to have been fully recruited.(48)

James Angleton, Hood's former boss in counterintelligence, apparently held a similar view, according to Edward Epstein:

Whereas money, sex, ideology, and ambition provide the means for compromising targets, the lever used to convert a man into a mole tends to be blackmail. . . . Whatever lure is used, the point of the sting is to make it impossible for the recruit to explain his activities to his superiors. He is compromised, not so much by his original indiscretion, but for failing to report it.(49)

(Note that Angleton here was referring to a special type of agent, the "mole" or penetration agent within an enemy's intelligence service. Not all espionage agents would be necessarily compromised by failing to report certain activities to their employer, but an intelligence officer would.)

Another former CIA official, Howard Stone, admitted that CIA often recruits agents by bribery or blackmail. But believing that such methods often produce unreliable agents who only pretend to have access to important information, he urged CIA instead "to win over prominent foreign officials of sound moral character."(50) One former CIA officer who served many years in Latin America told me that none of his agent relations were based on blackmail or other coercion. He believed like Howard Stone that such methods invariably produced "servile" and unreliable agents who "don't exercise good judgment." But this officer went on to say, "[My agents] produced for me because they knew I was reliable and they could count on me in a pinch. They would and did risk their lives for me." He added, though, that different methods might be necessary in other countries where the stakes and pressures were greater, such as the Soviet Union or Cuba.(51) It seems likely that a CIA officer having qualms about deceptive or coercive recruiting methods would simply not be assigned to such countries or would not remain there very long (at least in an agent-recruiting capacity).

Manipulation

Perhaps most troubling of the professional skills of an intelligence officer is the ability to manipulate persons. The degree of manipulation can vary from the subtle blackmail threat latent in a financial relationship with an espionage agent to more obviously coercive and even violent measures. The element of control in intelligence operations is directly related to suspicion of the loyalty of the agent. Suspicion is a professional virtue for intelligence officers, especially for those who work in security and counterintelligence, since in theory anyone thought to be trustworthy may in fact be secretly serving the enemy.

The practice of interrogation is a significant component of intelligence work, but also illustrates manipulation in its rawest form. William Johnson, a former CIA counterintelligence officer, has offered a glimpse of the ethical risks involved:

Interrogation is such a dirty business that it should be done only by people of the cleanest character. Anyone with sadistic tendencies should not be in the business.(52)

We are reminded, though, of the ease with which ordinary people can come to rationalize callousness and cruelty in dealing with perceived enemies. Given the natural human capacity for aggression, combined with the right set of biases, incentives and peer pressures, many ordinarily decent people can succumb to sadism.(53) The line between interrogation and torture is perilously thin.

What is it about interrogation that is "dirty"? To Johnson, it is not the presence of inflicted physical pain, which he regards to be not only morally dubious but counterproductive.

[P]hysical pain is not relevant in interrogation. Anxiety, humiliation, loneliness, and pride are another story. . . . The person who enjoys hurting is a lousy interrogator in even the most human [humane?] situation. But the humane person who shrinks from manipulating his subject is also a lousy interrogator. . . . The interrogator, like a priest or doctor [!], must have a talent for empathy, a personal need to communicate with other people, a concern for what makes other people tick even when he is putting maximum emotional pressure on them.(54)

In everyday moral parlance, empathy is related to compassion. But in intelligence work, the other is considered to be a potential threat to persons and interests that the intelligence officer is sworn to protect. "Knowing one's enemy" in this role means understanding the other, but not in the interest of enhancing his or her freedom or well-being: on the contrary, empathy becomes a manipulative tool.(55) This altered meaning of empathy holds true beyond the practice of interrogation. It also characterizes a significant part of the professional skill involved in recruiting and handling agents, whose trust is often essential to gaining their control. Empathy is also useful in the creation of propaganda, since the power of propaganda to influence its intended audience is largely dependent upon how well it is crafted to address that audience's peculiar cultural milieu.

Covert Action

Harry Rositzke argued that the kind of agent-manipulation that frequently occurs in espionage and counterespionage operations may not apply to some types of covert action. Covert financial support for a political leader or dissident, for example, need not entail his or her coercion since it serves his or her interests.(56) James McCargar expressed a similar opinion:

In a political operation the case officer must have arrived at a clear and workable accommodation of interests with the agent. Control by the case officer there must be, but not duplicity. The purposes of case officer and agent must have been presented with the maximum permissible clarity, and then a reconciliation of conflicts and limitations negotiated. In brief, the outstanding characteristic of the political case officer-agent relationship is that it must be an alliance, not a utilization of the agent by the case officer, as often occurs in intelligence.(57)

But the fact that this state of affairs applied for a time to CIA relations with Panamanian dictator Manuel Noriega,(58) among others, indicates that these arguments do not dispel moral concern for the wider context of covert action. Knowing that a covert action coincides with the interests of particular foreign nationals is not sufficient to justify it ethically, since covert action may involve the violation of rights that ought to override those interests.

It is also likely that an intelligence officer would seek to "vet" (or test the authenticity of) an agent of covert influence against the evidence supplied by informers or espionage agents, hence the need to use some method of agent recruitment and handling having one or more of the attendant moral concerns previously identified.

Now since the "product" of a covert action agent is in some respects "public" (unlike the product of an espionage agent), it is perhaps more difficult to deceive a covert action agent than an espionage agent as to the real intentions of his or her secret employers. One can more easily conceive, though, of a covert action agent (such as a newspaper reporter or editorialist) being coerced through blackmail or other threats into engaging in covert action. Such considerations provide further qualification, then, to Rositzke's and McCargar's assertions of the voluntary participation of covert action agents.

Secret financial support for foreign political or labor leaders is morally complex. In some cases it can plausibly be considered a form of humanitarian intervention. For example, CIA aid to centrist political parties in postwar France, Italy and Japan helped to counter covert Soviet aid to communist parties there, while CIA assistance to Christian Democrats in El Salvador was intended to prevent an election victory by right-wing candidates tied to death squads. Almost certainly the political consequences in those countries would have been grave had no aid been provided by CIA. But often the efforts made by U.S. officials to provide such aid openly have been insufficient.(59) Covert political action has been employed despite the fact that foreign citizens would rightly suspect the objective judgment owed to them by their leaders to have been clouded by their acceptance of secret payments.

Former CIA Director William Colby, who oversaw covert aid to anti-communist politicians in Italy during the 1950s, claimed in his memoirs, "The program in Italy gave aid to the democratic forces to obtain their goals. It did not 'bribe' them to follow American direction. . . ."(60) But Colby exhibited a very different view when asked by journalist Oriana Fallaci, ". . . if I came [to the U.S.], as a foreigner, and financed an American party, and 21 of your politicians, and some of your journalists, what would you do?" Colby responded that he would report her to the FBI.(61)

We rightly condemn corporate purchasing agents who accept or extort lavish gifts from would-be suppliers, doctors who receive kickbacks from pharmaceutical companies for prescribing certain drugs, and legislators who seek political contributions from industries they are supposed to regulate. Secret payments to foreign government leaders, even when made with good intentions, create at a minimum the appearance of a conflict of interest on the part of the recipient. They undermine government accountability and the public trust, and should be avoided.

Use of Underworld Figures

Another significant area of ethical concern has to do with the obstruction of justice in sheltering criminals used as agents, as in, for example, the postwar recruitment by U.S. intelligence of a number of Nazi war criminals to engage in espionage and covert operations against the Soviet Union.(62) Christopher Simpson, a reporter who extensively researched this subject, quoted Harry Rositzke as explaining:

It was a visceral business of using any bastard as long as he was anti-Communist . . . [and] the eagerness or desire to enlist collaborators meant that sure, you didn't look at their credentials too closely.(63)

Simpson claimed, however, that U.S. intelligence did indeed know about the war-crimes "credentials" of many of its postwar recruits, as did the British, French and Soviets, who also employed suspected and proven war criminals in intelligence roles.(64) He also showed that this practice became risky to U.S. intelligence as well, when ex-Nazis threatened to publicize U.S. covert operations in which they had participated unless the U.S. helped them to escape abroad to avoid prosecution for their wartime atrocities.(65)

The CIA later involved similarly shady characters in plots to assassinate various foreign leaders. Underworld figures like Sam Giancana and Santos Trafficante were approached to kill Fidel Castro.(66) Another individual recruited to kill an African politician was described in an internal CIA memo in the following fashion:

He is indeed aware of the precepts of right and wrong, but if he is given an assignment which may be morally wrong in the eyes of the world, but necessary because his case officer ordered him to carry it out, then it is right, and he will dutifully undertake appropriate action for its execution without pangs of conscience. In a word, he can rationalize all actions.(67)

The Senate committee that investigated the assassination plots appropriately judged this type of rationalization to be "not in keeping with the ideals of our nation." The committee also observed that employing underworld characters "gives them the power to blackmail the government and to avoid prosecution, for past or future crimes."(68)

In hindsight at least, it seems obvious that espionage and covert actions relying upon criminals as intelligence "assets" bear a strong burden of moral justification, chiefly since the victims of their crimes cannot be assumed to give tacit consent to their shelter from prosecution, but also because they can pose a threat to the societies in which they are secretly sheltered. Furthermore, in cases where perpetrators of mass murder (or even "ordinary" murder!) have sought refuge in intelligence work, it is difficult to see how the practice could be justified at all, even under the pressures that CIA officers felt in the early postwar years to quickly develop an underground network in the event of war with the Soviet Union.(69)

Economic Intelligence

Recognition of the sometimes grave consequences of espionage and covert action at least ought to have a sobering effect on the consideration of the ends they are intended to serve. If national security can possibly justify deceptive and coercive intelligence methods, it is far from clear that lesser ends can.

To illustrate, no American corporation today could or should hope to achieve the influence that the United Fruit Company had on U.S. policy toward Guatemala in the early 1950s. Before Allen Dulles became CIA Director under Eisenhower, he worked for a powerful law firm that arranged profitable deals for United Fruit in Guatemala, where it owned extensive plantations and rail lines and regularly crushed incipient labor unions. Thus, when the company asked the CIA to overthrow the country's first elected president, its request fell on eager and familiar ears. Allen Dulles even promised the company that whoever CIA selected to be the next Guatemalan leader would not be allowed to nationalize or in any way disrupt the company's operations.(70) The interests and objectives of particular corporations are not necessarily identical to those of the United States, yet this perspective was lost on many U.S. officials during the Cold War. The same people whose prescience enabled the U.S. to meet the many real challenges posed by the Soviet Union and its allies were nonetheless capable of rationalizing the corruption of national ends to serve a scandalously small elite.

The breakup of the Soviet Union has not only resulted in large cuts in U.S. defense spending, but has also caused the CIA's overall mission and budget to be carefully scrutinized. Many people are asking whether CIA resources should now be focused in ways that more directly enhance American economic competitiveness. Some members of Congress and business leaders have called upon the CIA to spy on behalf of American corporations, much as the governments of France and Japan are doing for their native companies. Unless U.S. firms are able to "fight fire with fire," the argument goes, they will be at a decided disadvantage in the global market.(71)

Fortunately, many large U.S. corporations have said that they neither want nor need this sort of help from the Agency, and CIA officials have lobbied against it as well. Former CIA Director Robert Gates said in an oft-quoted April 1992 speech:

[U.S. intelligence] does not, should not, and will not engage in industrial espionage. . . . Plainly put, it is the role of U.S. business to size up their foreign competitors' trade secrets, marketing strategies, and bid proposals. Some years ago, one of our clandestine service officers said to me: "You know, I'm prepared to give my life for my country, but not for a company." That case officer was absolutely right.(72)

The moral justification of espionage (let alone covert action) is highly dubious in the service of preserving or enhancing the global competitiveness of U.S. corporations. Apart from the defense industry, which is uniquely tied to U.S. national security, the reasons that American companies might offer to persuade the CIA to spy for them almost certainly could not be weighty enough to override the rights of foreign citizens duped or coerced into committing espionage. Nor would voluntary agents incurring great risks to deliver secret intelligence to the CIA be amused to learn that their reports were being forwarded to U.S. corporations to enable them to tap previously untouched consumer markets or to gain an edge over their foreign competitors. Even if it were possible to "sanitize" this data in ways that did not jeopardize intelligence sources and methods, questions of fairness would still arise. For example, which American companies would be given that information? Should it be free (i.e., subsidized by taxpayers), or should companies pay for it?

On the other hand, the current practice of the U.S. Government of providing counterintelligence advice to American companies overseas (e.g., how to prevent company phones from being tapped) is in most cases morally acceptable. Such assistance has in fact been provided for many years through the State Department's Overseas Security Advisory Council.

Assassination

That the CIA developed the capability to disable and assassinate foreign leaders is not in dispute, though precisely when that capability was conceived is unclear. The record indicates that an internal CIA "Health Alteration Committee" existed as early as 1960, and that a CIA "executive action" capability, which included assassination, was authorized by the White House as early as 1961.(73) However, since OSS had developed drugs during WWII for the purpose of assassinating and incapacitating Nazi leaders,(74) it is entirely possible that CIA inherited this capability and maintained it from its inception. Some evidence exists to suggest that CIA was authorized to create a special squad in 1949 whose duties included kidnapping and assassination, though primarily of suspected double agents.(75)

As John Marks has detailed, CIA technicians developed drugs and stockpiled bacteriological toxins that could immobilize an individual for hours, days or months, or kill him in a manner that could not be ascertained by autopsy or that appeared to be the result of a deadly disease that the individual might plausibly have contracted naturally.(76)

Assassination has been prohibited by U.S. executive orders since the mid-1970s, but there are indications that it may have been authorized since that time at high levels within the U.S. Government. A manual developed for the Nicaraguan contras by one or more of their CIA advisers, for example, urged that Sandinista officials be "neutralized" as part of a "selective use of violence for propaganda purposes."(77) In addition, former CIA general counsel Stanley Sporkin reportedly concluded in the early 1980s that violent actions taken against terrorists would not constitute assassination under U.S. law,(78) and this opinion may have served as the justification for "sensitive retaliation operations" launched against those believed responsible for the 1983 bombings of the U.S. Embassy and Marine compound in Beirut.(79)

Neil Livingstone, an expert on terrorism and low-intensity conflict, argues that "state-sanctioned terminations" can be justified against terrorists:

Just as it is not a crime to kill the enemy during wartime, so too should it not be regarded as a crime or a morally reprehensible act when a nation, acting in concert with its obligation to protect its own citizens from harm, seeks out and destroys terrorists outside its borders who have committed, or are planning to commit atrocities on its territory or against its citizens.(80)

Livingstone adds, however, that assassination should be considered "only when the potential target cannot be brought to justice in a more conventional manner."(81) His caveat is an important one, in part because assassination by definition excludes due process of law in ascertaining the guilt or innocence of the "accused" as well as in applying an appropriate punishment if and when guilt is established. The assassin in effect acts as prosecutor, judge, jury and executioner combined; the target is precluded from being represented by counsel before an impartial court. These concerns suggest that assassination ought only to be used as a last resort.(82)

Other writers have favored the assassination of foreign government leaders in particular circumstances. Angelo Codevilla questions the sensibility of the U.S. legal prohibition of assassination, suggesting that the practice is morally preferable to accepting the consequences of aggressive war:

The military art, the very opposite of indiscriminate killing, consists of striking those people and things most likely to stop the enemy from continuing the war. Today, the specialization of weapons and tactics of war make it easier than ever to go after those whose death is most likely to stop the killing. Often, as in the Gulf War [1990-1991], there is no quarrel with the enemy country, only with its chief. In such cases, it is both futile and immoral to demolish a country in the hope that this will persuade the tyrant to give way. Why not kill the tyrant?(83)

Codevilla's assertion that it is unwise to prohibit assassination tout court is supported by numerous arguments in the Western philosophical tradition justifying tyrannicide as a permissible (if last-resort) means to a just end. But missing from Codevilla's argument are other considerations present in the tradition that weigh against employing assassination as an isolated act. One important concern is that unless tyrannicide is coupled with a wider effort to replace the entire regime, it will likely result in greater repression of the populace rather than less.(84) Many of the CIA assassination plots investigated by the U.S. Congress in 1975 seem to belie any real consideration for the well-being of their intended victim's fellow citizens. Too often the death of a foreign leader was an end in itself rather than simply a possible outcome of a comprehensive effort to replace a bad government with one that respected human rights. Removing Fidel Castro, for example, would not by itself have diminished the Cuban regime's oppression of its citizens, yet numerous CIA plots were devised against Castro (with vigorous White House support, to be sure) apart from any coordinated plan to replace him with a viable and more liberal government.(85)

Assassination is often thought to be capable of ending and preventing war and terrorist crimes. No doubt this is true in many cases. But like other forms of covert action, assassination has too often been proposed as an option well before other less morally objectionable measures have been tried. For example, among the covert operations considered by CIA in 1954 against President Arbenz of Guatemala was his assassination by means of a "silent bullet." It is a disturbing commentary on the quality of moral reasoning employed by CIA and other senior U.S. officials at that time that the only apparent reason why the assassination option was discarded was a desire not to make Arbenz a martyr.(86)

Professional Standards

The late Paul Seabury once said: "The exercise of power does not necessarily corrupt. The craft of intelligence can have as its practitioners those who were able to maintain their integrity while being liars and obfuscators."(87) I do not doubt the essential truth of that provocative thought. The United States must be able to depend upon its intelligence officers to be persons of high ethical standards.

But personal integrity can easily be undermined by the wrong kinds of incentives and pressures, and must therefore be cultivated, monitored and reinforced institutionally. Those who have the authority to establish objectives for intelligence operations must not only weigh the ethical justification of those ends but must also raise ethical questions about the various means being considered to achieve them. Intelligence officers in the field, in turn, must be trained to recognize ethical issues as such, and must be allowed to communicate their concerns to their supervisors without fear for their careers.

To some extent, the CIA already addresses ethical concerns in its internal training and communications. Hulnick and Mattausch claimed that honesty must apply to internal CIA communications and practices. Intelligence judgments must never be altered "to fit the desires of policymakers who might prefer different conclusions,"(88) and intelligence officers "must be scrupulous in managing funds or equipment with which they are entrusted," given the fact that many funds are not subject to outside audit and that certain equipment is designed to prevent its being identified with CIA.(89) They also noted an interesting implication of the oath of secrecy which intelligence officers must swear. Like military personnel, they cannot appeal to a "Nuremberg defense" in the face of a clearly improper order from a superior. But unlike other public servants, they do not have the option of "going public" with an issue or order they consider to be illegal or immoral if the internal "whistleblowing" procedure proves to be unsatisfactory to them: "public discussion is not possible without a gross violation of classification rules and the professional ethics of the intelligence officer."(90)

In January 1992, James Barry, then Director of the CIA's Center for the Study of Intelligence, responded to a series of questions I had submitted with an interesting letter cleared by higher officials for release. Arguing that "[p]rofessional ethics is a central component of [CIA's] training and career development process," Barry described in general terms the kinds of training provided for new CIA employees, mid-career employees, new supervisors and middle managers, and senior officials. He stated that high ethical standards play an important role in the evaluation and certification of case officers, and that those individuals receive specialized, tutorial training in ethical issues related to foreign intelligence, counterintelligence and covert action. Barry further noted CIA's standing policy that any employees having ethical concerns "may report them in confidence to the [CIA] Inspector General."(91)

There is further indication that CIA has maintained a form of "conscientious objector" status for its personnel relative to certain morally problematic assignments. Testimony of former CIA officers before the Senate committee investigating assassination plots in 1975 suggests that CIA employees were allowed to decline to participate in those plots without experiencing threats to their career. "Michael Mulroney" (pseud.), a former CIA officer, testified that in 1960 he refused on moral grounds to carry out a political assassination requested by the head of CIA's clandestine division, Richard Bissell, and that his decision was supported by Bissell's deputy, Richard Helms.(92) In addition, during the Vietnam War, American personnel involved with the Phoenix program were apparently granted a similar "conscientious objector" option.(93)

Conclusions

Although space does not permit review of all of the issues examined in this essay, some concluding reflections on selected categories of intelligence operations are warranted.

The use of secret agents--voluntary and non-voluntary--is intended to provide valuable information believed to be unobtainable through methods overt or technical. The risks inherent in all espionage activities suggest, though, that for the sake of the agent alone, efforts should be made to determine before the agent is recruited that the information needed cannot in fact be ascertained by less problematic methods. In addition, since after an agent is recruited the agent-officer relationship takes on a life and momentum of its own, care must be taken to avoid situations where innocent third parties would be harmed or justice obstructed in the interest of preserving the agent's identity and continued service.

Recruiting voluntary agents has the advantage of involving no deception about the identity and general motives of the recruiter. Furthermore, a just cause can be served by intelligence officers and voluntary agents working together to undermine an unjust regime. But such agents deserve not to be deceived about the risks involved in the operations they are asked to carry out. Nor should the fact that their work is secret tempt their handlers to treat them as expendable, to allow them callously to be sacrificed to Realpolitik or the shifting winds of diplomacy.

The chief advantage of employing a false-flag approach or blackmail in certain situations is that a just cause can be pursued even where foreign citizens are highly unlikely to serve voluntarily as CIA agents. But such methods raise very difficult questions. False-flag methods deceive the agent as to the identity of the recruiter, and thus hide from the agent the full risks inherent in his or her tasks as well as their true purposes. Blackmail is blatant coercion. It is difficult enough to justify its use against known criminals; all the more so when it arises out of the calculated entrapment of a previously innocent person who merely happens to have probable access to sensitive information desired by the CIA. Finally, to the extent that recruitment tactics seek to "stretch" the agent's conscience, they can result in the corruption of the agent in addition to his or her victimization. Deception and coercion in agent recruitment should certainly not be used as routine methods of obtaining "assets" whose future value as sources of vital intelligence is dubious. (Recall Richard Bissell's 1975 testimony about agents being recruited "long before there [is] any reason to be certain whether those would be used or where or how or for what purpose."[94])

These concerns about espionage are challenged, though, by the claim that if one rules out an espionage source or method one may thereby eliminate the possibility of knowing certain kinds of vital information. It's not difficult to construct hypothetical cases in which knowing the intentions of a tyrannical regime or a terrorist cell could mean the difference between life and death for many people, cases which would therefore question the validity of strict prohibitions on deceptive and coercive intelligence methods or the use of criminals as agents.

In addition, if we imagine a prospective agent who works in a sensitive capacity for the government of a manifestly tyrannical state, there is a sense in which, since that government itself is not and cannot be rationally willed by its oppressed citizens, neither can service to that government in ways that maintain its tyrannical nature be justified. Some regimes simply do not deserve the loyalty of their citizens. But given the fact that opportunities to persuade citizens and government officials in tyrannical states that they ought to commit treason are sometimes quite limited, the justification of coerced recruitment of agents to achieve this becomes more plausible, in spite of the fact that unless the tyranny poses a dire threat to other countries, coercive recruitment would appear to be a form of paternalistic intervention. Coercive recruitment of agents within a tyrannical state may become even more acceptable as that state's threat to other countries becomes more grave or imminent. Remembering Sissela Bok's assertion that "whenever it is right to resist an assault by force, it must then be allowable to do so by guile," (95) espionage and covert action can serve as effective ways to prevent a tyranny from launching an aggressive war or intimidating its neighbors.

Of course, to say that a decision or action is morally right does not necessarily mean that the outcome is unequivocally good. It may be, for example, that coercive recruitment of an agent can be morally justified in a particular situation, given the dire consequences of not having the information he or she can provide, say, plus a lack of morally acceptable alternatives. But since coercion involves an infringement of the agent's freedom (and conceivably other basic rights), the external good that may result from the recruitment cannot do away with the fact that the agent--a human being with emotions, hopes, and dreams, not merely an abstract "source," "asset," or "penetration"--suffers real harm in the process.

Tragic choices are inevitable to some degree in intelligence work. The challenge, then, is to specify intelligence goals and manage operations in ways that recognize the myriad ethical issues at stake, in order to minimize the occurrence of avoidable tragedy.

Endnotes

1. An earlier version of this essay was presented on 30 March 1994 at a meeting in Washington, DC, of the Intelligence Study Section of the International Studies Association. Thanks to Jefferson Adams and Abe Miller for encouraging me to submit that paper. I have drawn upon portions of "Covert Action: An Exploration of the Ethical Issues," my 1993 Ph.D. dissertation for the University of Chicago Divinity School. Dissertation research in Washington, DC, was enabled by a fellowship from the University of Chicago's Program in Arms Control and International Security (PACIS), funded by the MacArthur Foundation. I benefitted from the impressive range of intelligence literature in the Library of Congress and the Russell Bowen special collection at Georgetown University. Finally, I'm grateful to the following individuals for their insightful comments during various stages of my writing: Jim Barry, William Colby, William Hood, Robin Lovin, James McCargar, Peter Savage, B. Hugh Tovar, David Whipple, David Charters and two anonymous reviewers for The Journal of Conflict Studies, and some former CIA officers who spoke to me in confidence. I have not been privy to any classified information, and have therefore not been required to submit my work to the U.S. Government for review.

2. Cf. W. D. Ross, The Right and the Good (New York: Oxford Univ. Press, 1930). Ross thought it impossible to derive all valid moral duties from one fundamental principle, let alone to construct an absolute code encompassing every possible moral dilemma. He argued that human beings have numerous prima facie ethical obligations, i.e., duties that are strong enough to override less important preferences, but which are nevertheless not absolute, since they can be overridden by one another in particular situations. In some cases one's paramount moral duty will be to promote happiness; in others, to prevent or alleviate harm; in others, to protect rights; etc. The need for moral deliberation and wisdom is simply part of the human condition, in Ross's view. But it's important to note that his theory does not entail ethical relativism, i.e., the (illogical) view that two contradictory ethical principles can both be true at the same time. Nor is one permitted to treat similar cases in dissimilar ways.

3. Affinities between intelligence and the practice of law (e.g., in the importance of secrecy and aggressive tactics against opponents) were suggested in John Ranelagh, The Agency: The Rise and Decline of the CIA, revised and updated edition (New York: Simon & Schuster, 1987), pp. 27-31.

4. NSC Directive 10/2 (18 June 1948), in William M. Leary, ed., The Central Intelligence Agency: History and Documents (Univ., AL: Univ. of Alabama Press, 1984), p. 131.

5. James Doolittle, et al., "Report on the Covert Activities of the Central Intelligence Agency," ibid., pp. 143-145, emphasis added.

6. See, e.g., Peer de Silva, Sub Rosa: The CIA and the Uses of Intelligence (New York: Times Books, 1978), B. Hugh Tovar, "Covert Action," in Roy Godson, ed., Intelligence Requirements for the 1980's: Elements of Intelligence (Washington: National Strategy Information Center, 1979), pp. 67-79, and Donald Jameson, "The Clandestine Battlefield: Trenches and Trends," Strategic Review, Winter 1983, pp. 19-28.

7. Three excellent histories of Russian/Soviet intelligence are Ronald Hingley, The Russian Secret Police: Muscovite, Imperial Russian, and Soviet Political Security Operations, 1565-1970 (London: Hutchinson, 1970), George Leggett, The Cheka: Lenin's Political Police (New York: Oxford Univ. Press, 1981), and Amy W. Knight, The KGB: Police and Politics in the Soviet Union (Boston: Allen & Unwin, 1988). An impressive contemporary assessment is J. Michael Waller, Secret Empire: The KGB in Russia Today (Boulder, CO: Westview, 1994).

8. Some recommendations on future U.S. uses of covert action were made by Roy Godson et al. in Covert Action in the 1990s (Washington: Consortium for the Study of Intelligence, 1992).

9. Sissela Bok, Lying: Moral Choice in Public and Private Life (New York: Pantheon, 1978; Random House, 1979), pp. 149 and 151.

10. See Arthur Jacobs, letter to the Editor, Foreign Affairs 56/5 (July 1978), pp. 867-875, Miles Copeland, The Real Spy World (London: Sphere, 1978), pp. 282-283, and Arthur S. Hulnick and Daniel W. Mattausch, "Ethics and Morality in United States Secret Intelligence," Harvard Journal of Law & Public Policy 12/2 (Spring 1989), pp. 520-522.

11. Hulnick and Mattausch, "Ethics and Morality in U.S. Secret Intelligence," pp. 520-521.

12. Alan Goldman discusses that concept in relation to business, medicine and the legal profession in The Moral Foundations of Professional Ethics (Totowa, NJ: Rowman & Littlefield, 1980). Compare Alan Gewirth, "Professional Ethics: The Separatist Thesis," Ethics 96/2 (January 1986), pp. 282-300. The literature on political "dirty hands" is extensive. Two important essays are Michael Walzer, "Political Action: The Problem of Dirty Hands," Philosophy and Public Affairs 2/2 (Winter 1973), pp. 160-180, and Leslie Griffin, "The Problem of Dirty Hands," Journal of Religious Ethics 17/1 (Spring 1989), pp. 31-61.

13. Cited in Congress, Senate, Select Committee to Study Governmental Operations with Respect to Intelligence Activities, Alleged Assassination Plots Involving Foreign Leaders, Interim Report #94-465, 94th Congress, 1st session, 1975, p. 186, ellipsis in the original, emphasis added.

14. See Christopher Felix [James McCargar], A Short Course in the Secret War (New York: Dutton, 1963; Dell, 1988), pp. 54ff., H. H. A. Cooper and Lawrence J. Redlinger, Making Spies: A Talent Spotter's Handbook (Boulder, CO: Paladin, 1986), ch. 2, and Chapman Pincher, Traitors (New York: St. Martin's, 1987; Penguin, 1988).

15. In the interest of "compartmentation" (the restriction of information to only those who can justify their "need to know" it), even voluntary agents are rarely told how the information they provide is actually used.

16. Jacobs, letter to Foreign Affairs, p. 870; Joseph Burkholder Smith, Portrait of a Cold Warrior (New York: Ballantine, 1981), pp. 114-115.

17. William Hood, Mole (New York: Random House, 1982; Ballantine, 1983); Harry Rositzke, CIA's Secret Operations (Boulder, CO: Westview, 1988), pp. 67-69.

18. Oleg Penkovskiy, The Penkovskiy Papers, trans. Peter Deriabin (New York: Doubleday, 1965). See also Rositzke, CIA's Secret Operations, pp. 69-71, and Ranelagh, Agency, pp. 400-402.

19. Rositzke, CIA's Secret Operations, pp. 26-28.

20. I'm aware that those agents were not all "voluntary": some KGB officers were entrapped into working for the CIA and FBI. Cf. Tim Weiner, "Spy Suspect Betrayed 10 U.S. Agents, F.B.I. Says," New York Times, 24 February 1994, pp. A1 and A13.

21. Rositzke, CIA's Secret Operations, pp. 123-124.

22. Hood, Mole, pp. 96-97.

23. Michael Wines, "2 or 3 Agents Are Believed Killed After Rare U.S.-Syrian Contacts," New York Times, 7 February 1991, pp. A1 and A18.

24. Orrin DeForest and David Chanoff, Slow Burn: The Rise and Bitter Fall of American Intelligence in Vietnam (New York: Simon & Schuster, 1990); Felix, Secret War, p. 107.

25. Felix, Secret War, pp. 62-63.

26. Tim Weiner, "Mole's Damage to C.I.A. Could Take Years to Fix," New York Times, 23 February 1994, p. A12; Oleg Gordievsky, "Aldrich Ames, My Would-be Killer," The Spectator, 5 March 1994, pp. 15-16.

27. Tom Mangold, Cold Warrior: James Jesus Angleton: The CIA's Master Spy Hunter (New York: Simon & Schuster, 1991).

28. Felix, Secret War, pp. 112-113.

29. Ranelagh, Agency, pp. 137, 226-228, 287, 302-309. See also John Prados, Presidents' Secret Wars: CIA and Pentagon Covert Operations Since World War II (New York: William Morrow, 1986), chs. 2-3.

30. Pincher, Traitors, p. 24. On Albania, see Nicholas Bethell, The Great Betrayal: The Untold Story of Kim Philby's Biggest Coup (London: Hodder and Stoughton, 1984).

31. Michael Gunter, "Mulla Mustafa Barzani and the Kurdish Rebellion in Iraq: The Intelligence Factor," International Journal of Intelligence and Counterintelligence 7/4 (Winter 1994):466-469.

32. Copeland, Real Spy World, pp. 125-129; Felix, Secret War, p. 112.

33. David Atlee Phillips, The Night Watch (New York: Ballantine, 1982), pp. 263-264; Edward J. Epstein, Deception: The Invisible War Between the KGB and the CIA (New York: Simon & Schuster, 1989), pp. 89, 182-183. Phillips preceded his description of "false-flag" recruitment with this interesting comment (p. 263): "Most intelligence officers who set out to persuade someone to become a traitor have to reach an accommodation of some sort with the code of ethics and morality they have inherited or adopted. Sometimes dirty tricks are involved in the recruiting of spies."

34. Copeland, Real Spy World, pp. 128-129. Copeland also asserted (p. 129) that "most spies really don't know which espionage service they are working for"! But that hypothesis was deemed nonsensical by three former CIA officers in separate confidential interviews with me in 1992.

35. Phillips, Night Watch, p. 264.

36. I have seen no evidence to suggest that CIA has ever imitated the tactic of the Mafia, KGB or Viet Cong of threatening to kill persons or their families if they do not agree to cooperate. In fact, a former CIA officer told me emphatically that any CIA officer who made such a threat "would have been fired outright." B. Hugh Tovar, letter to David Perry, 25 February 1992, pp. 5-6.

37. Smith, Portrait of a Cold Warrior, p. 115.

38. Copeland, Real Spy World, p. 127-128.

39. Cooper and Redlinger, Making Spies, p. 108.

40. Epstein, Deception, p. 180. It is not actually clear whether these were Angleton's words or Epstein's only. Epstein conducted numerous interviews with Angleton before the latter's death in 1987.

41. Felix, Secret War, pp. 51 and 56.

42. Jacobs, letter to Foreign Affairs, p. 871. The ineffectiveness of blackmail in agent recruiting was also suggested by former CIA Director William Colby in an interview with me on 14 September 1991.

43. John Marks, The Search for the "Manchurian Candidate": The CIA and Mind Control (New York: W. W. Norton, 1979, 1991).

44. Cooper and Redlinger, Making Spies, pp. 10 and 19.

45. E. Drexel Godfrey, Jr., "Ethics and Intelligence," Foreign Affairs 56/4 (April 1978), p. 631.

46. Copeland, Real Spy World, pp. 150-151.

47. Ibid., p. 130.

48. Hood, Mole, p. 29.

49. Epstein, Deception, p. 183.

50. David Ignatius, "In from the Cold: A Former Master Spy Spins Intriguing Yarns of His Past Intrigues," Wall Street Journal, 19 October 1979, pp. 1 and 41. Ignatius added his own opinion that agents recruited by the CIA "can be a rather scurvy lot."

51. Confidential interview, 1991.

52. William R. Johnson, "Tricks of the Trade: Counterintelligence Interrogation," International Journal of Intelligence and Counterintelligence 1/2 (1986), p. 104.

53. One of the most compelling studies of this phenomenon is Christopher R. Browning, Ordinary Men: Reserve Police Battalion 101 and the Final Solution in Poland (New York: HarperCollins, 1992). See also Ari Shavit, "On Gaza Beach," New York Review of Books, 18 July 1991, pp. 3-6.

54. William R. Johnson, Thwarting Enemies at Home and Abroad: How to Be a Counterintelligence Officer (Bethesda: Stone Trail Press, 1987), pp. 33 and 32, emphasis in the original.

55. Ralph K. White, "Empathy as an Intelligence Tool," International Journal of Intelligence and Counterintelligence 1/1 (Spring 1986), pp. 57-75.

56. Rositzke, CIA's Secret Operations, pp. 185-186.

57. Felix, Secret War, pp. 144-145.

58. Frederick Kempe, "Ties That Blind: U.S. Taught Noriega to Spy, but the Pupil Had His Own Agenda," Wall Street Journal, 18 October 1989, pp. A1 and A20.

59. This was one of the findings of a panel appointed by President Johnson and headed by Nicholas Katzenbach in 1967 to investigate CIA political action. "Texts of Statement and Report on Covert C.I.A. Aid," New York Times, 30 March 1967, p. 30.

60. William E. Colby and Peter Forbath, Honorable Men: My Life in the CIA (New York: Simon & Schuster, 1978), p. 115, emphasis in the original.

61. Oriana Fallaci, "The CIA's Mr. Colby," New Republic 174/11 (13 March 1976), pp. 12-21.

62. See Christopher Simpson, Blowback: America's Recruitment of Nazis and Its Effects on the Cold War (New York: Weidenfeld & Nicolson, 1988), chs. 8-12, on "Operation Bloodstone." A major drawback of this book, though, is its sanguine view of postwar Soviet capabilities and intentions.

63. Ibid., p. 159, ellipsis and brackets in the original.

64. Ibid., p. 73. Rositzke mentions in CIA's Secret Operations, pp. 27-28, 166-173, that ethnic Russians, Balts, Ukrainians, Armenians and Georgians were recruited as agents for missions against the Soviet Union. He stresses their justified resentment against Soviet oppression, but does not discuss how any Nazi collaborators who may have been identified among them were handled.

65. Simpson, Blowback, p. 175.

66. U.S. Senate, Alleged Assassination Plots, pp. 43-48, 74-77.

67. Ibid., p. 46.

68. Ibid., p. 259.

69. Simpson, Blowback, pp. 159-160, quotes Franklin Lindsay, who apparently in the early 1950s oversaw CIA paramilitary operations in Eastern Europe that involved some former Nazi collaborators: "You have to remember that in those days even men such as George Kennan believed that there was a fifty-fifty chance of war with the Soviets within six months. . . . We were under tremendous pressure to do something, do anything to prepare for war."

70. See Stephen Schlesinger and Stephen Kinzer, Bitter Fruit: The Untold Story of the American Coup in Guatemala (Garden City, NY: Doubleday, 1982), and Richard H. Immerman, The CIA in Guatemala (Austin: Univ. of Texas Press, 1982). I discuss that coup as well as covert operations in Iran and Italy in chapter four of Perry, "Covert Action."

71. "Should the CIA Start Spying for Corporate America?" Business Week, 14 October 1991, pp. 96-100; Thomas McCarroll, "Next for the CIA: Business Spying?" Time, 22 February 1993, pp. 60-61; Peter Schweizer, Friendly Spies: How America's Allies Are Using Economic Espionage to Steal Our Secrets (New York: Atlantic Monthly Press, 1993).

72. George Lardner, Jr., "U.S. Demands for Economic Intelligence Up Sharply, Gates Says," Washington Post, 14 April 1992, p. A5. A view similar to Gates' was expressed by Randall M. Fort in Economic Espionage: Problems and Prospects (Washington: Consortium for the Study of Intelligence, 1993).

73. U.S. Senate, Alleged Assassination Plots, pp. 181-187.

74. Marks, "Manchurian Candidate", pp. 16-18, 119.

75. Simpson, Blowback, p. 153.

76. Marks, "Manchurian Candidate", pp. 80-81.

77. "Tayacán," Psychological Operations in Guerrilla Warfare, with essays by Joanne Omang and Aryeh Neier (New York: Random House, 1985), pp. 57-59. B. Hugh Tovar, in a 25 February 1992 letter to me, wrote (p. 9), "'Neutralization' was not authorized at high levels [within CIA], and the individuals responsible for the [Nicaragua] manual were censured severely."

78. Bob Woodward, Veil: The Secret Wars of the CIA 1981-1987 (New York: Simon & Schuster, 1987), p. 362.

79. See Russell Watson, Nicholas M. Horrock and Abdul Hajjaj, "Fighting Terror with Terror: An Abortive CIA Operation in Lebanon Is Uncovered," Newsweek, 27 May 1985, pp. 32-33, and William Cowan, "How to Kill Saddam," Washington Post, 10 February 1991, p. C2.

80. Neil C. Livingstone, The War Against Terrorism (Lexington, MA: Lexington Books, 1982), pp. 174-175.

81. Ibid., p. 175.

82. One alternative is to kidnap unsavory individuals and bring them to the U.S. for trial, a tactic which has been allowed by the U.S. Supreme Court. Ruth Marcus, "Kidnapping outside the U.S. Is Upheld," Washington Post, 16 June 1992, pp. A1 and A4.

83. Angelo Codevilla, "Get Rid of Saddam Hussein Now: The Moral Justification," Wall Street Journal, 25 February 1991, p. A8.

84. The theologian Dietrich Bonhoeffer, who participated in an unsuccessful attempt on Hitler's life in 1944, reportedly believed that "the act of assassination must be coordinated with the plans of a group capable of quickly occupying, or remaining in, the key organs of the totalitarian dictatorship," a principle which resembles the jus ad bellum criterion of probable success. Larry L. Rasmussen, Dietrich Bonhoeffer: Reality and Resistance (Nashville: Abingdon, 1972), p. 145.

85. On the various plots against Castro, see U.S. Senate, Alleged Assassination Plots, pp. 71-90, and Thomas Powers, The Man Who Kept the Secrets: Richard Helms and the CIA (New York: Knopf, 1979; Simon & Schuster, 1981), ch. 9.

86. Schlesinger and Kinzer, Bitter Fruit, p. 112.

87. Paul Seabury, quoted in Roy Godson, ed., Intelligence Requirements for the 1980's: Covert Action (Washington: National Strategy Information Center, 1981), p. 107.

88. This is an issue that was hotly debated during the 1991 Senate confirmation hearings of Robert Gates as Director of Central Intelligence. Some CIA analysts testified that Gates had slanted intelligence to fit the biases of former director William Casey, and that this violated a fundamental CIA ethical principle. See George Lardner, Jr., and Benjamin Weiser, "Spy vs. Spy: 4 CIA Veterans Criticize, Defend Gates," Washington Post, 2 October 1991, pp. A1 and A14.

89. Hulnick and Mattausch, "Ethics and Morality in U.S. Secret Intelligence," p. 521.

90. Ibid., p. 522.

91. The fact that most of CIA's training methods and materials remain classified makes it impossible for an outsider to come to any sound conclusion about their adequacy relative to the issues and concerns explored in this article. In January 1991, the CIA formally agreed to fulfill a Freedom of Information request I had made for materials related to its implementation of presidential orders on ethics in government. But the Agency has yet to provide me with any of those materials.

92. U.S. Senate, Alleged Assassination Plots, pp. 37-42.

93. See William Colby with James McCargar, Lost Victory (Chicago: Contemporary Books, 1989), pp. 247-248.

94. U.S. Senate, Alleged Assassination Plots, p. 186.

95. Bok, Lying, p. 151.