HUMINT is far less expensive than the various technical collectors, although it still involves costs for training, special equipment, and the accoutrements clandestine officers need to build successful cover stories.
Like all the other collection INTs, HUMINT is susceptible to deception. Some critics argue that it is the most susceptible to deception. The bona fides of human sources are always subject to question initially and, in some cases, may never be wholly resolved. Many questions arise and linger. Why is this person offering to pass information—ideology, money, vengeance? The person will claim to have good access to valuable information, but how good is it? Is it consistent, or is this a single event? How good is the information? Is this person a dangle, offered as a means of passing information that the other side wants to have passed—either because it is false or because it will have a specific effect? Is this person a double agent who is collecting information on your intelligence agency’s HUMINT techniques and capabilities even as he or she passes information to you?
HUMINT officers must walk a fine line between prudent caution and the possibility that too much caution will lead them to deter or reject a promising source. For example, the United States initially rejected the services of Penkovsky, who then turned to the British, who accepted him. Only later did the United States take on this valuable spy. Deception is particularly difficult to deal with, because people naturally are reluctant to accept that they are being deceived. However, people might slip into a position where they trust no one, which can result in turning away sources who might have been valuable.
HUMINT’s unique sources and methods raise another issue. These sources are considered to be extremely fragile, given that good human penetrations take so long to develop and risk the lives of the case officers, their sources, and perhaps even the sources’ families. Therefore, the intelligence analysts who receive HUMINT reports may not be told the details of the source or sources. Analysts are not informed, for example, that “this report comes from a first secretary in the Fredonian Foreign Ministry.” Instead, the report includes information on the access of the source to the intelligence, the past reliability of the source, or variations on this concept. Sometimes several sources may be blended together in a single report. Although the masking of HUMINT sources promotes their preservation, it may have the unintended effect of devaluing the reports for analysts, who may not fully appreciate the value of the source and the information. This became an issue in the aftermath of the Iraq WMD experience, when it was recognized that some sources had been of questionable reliability and that analysts were not always given as much information as would have been desirable about the nature of some of the HUMINT reporting. It also denies all-source analysts the ability to make an independent judgment of the HUMINT source when compared with the other sources to which they have access. (HUMINT reports come with captions provided by reports officers as to the nature of the source: a reliable source, an untested source, a source with proven access, a source with unknown access, and so on.)
Also, as DCI Richard Helms (1966-1973) observed, most HUMINT sources are recruited for a specific assignment or requirement, based on their access to the desired intelligence. They cannot be assigned from issue to issue as they are extremely unlikely to have access to other intelligence. Helms also believed that spies who no longer had the desired access should not be held in reserve but should be dropped. He said that a well-run station (the base from which officers operate overseas) “does not cling to spent spies.” Thus, even successful HUMINT, although extremely valuable, is narrow in focus.
HUMINT also puts one in contact—and perhaps into relationships—with unsavory individuals such as terrorists and narco-traffickers. If one is going to penetrate such groups or develop other types of relationships with them, some may become recipients of money or other forms of payment. These types of relationships raise moral and ethical issues for some people (see chap. 13). In the aftermath of the September 2001 attacks, special attention was given to the so-called Deutch rules about HUMINT recruitment. In 1995, DCI John M. Deutch (1995-1997) ordered a scrub of all HUMINT assets, with a particular focus on persons who in the past had been involved in serious criminal activity or human rights violations. The scrub was the result of revelations that some past CIA assets in Guatemala had violated human rights, including those of some American residents in that country. New rules were promulgated, requiring headquarters approval of any such recruitments in the future. After the terrorist attacks, the rules were widely criticized, with many people asserting that they had limited the CIA’s ability to penetrate terrorist groups. CIA officials maintained that no valuable relationship was ever turned down because of the Deutch rules. Critics countered, however, that the very existence of the rules bred timidity in the DO, as officers would be more cautious about whom they recruited, running the risk of losing useful sources, instead of having these recruitments be scrutinized on the basis of changing standards. By the end of 2001, the Deutch rules were no longer considered an operational factor as field stations were told they could be ignored. In July 2002 they were formally rescinded. Writing in the aftermath of the September 2001 attacks, Deutch defended his rules, arguing that they allowed DO officers to recruit with clear guidelines and focused on acquiring high-quality agents.
In the United States, constant tension exists between HUMINT and the other collection disciplines. The dominance of technical collection periodically gives rise to calls for a greater emphasis on HUMINT. So-called intelligence failures, such as the fall of the shah of Iran in 1979, the unexpected Indian nuclear tests in 1998, and the 2001 terrorist attacks have led to demands for more HUMINT. There is something odd about this recurring call for more HUMINT in that successful HUMINT is not a question of the mass of agents being assigned to a target. Some targets, such as terrorist cells, or the inner sanctum of totalitarian regimes, will always be difficult to penetrate. There is no reason to believe that the twentieth agent who is sent will succeed when the first nineteen have not. It is not possible to swarm agents against a difficult HUMINT target in terms of the agents’ availability and. more important, the risk. Such an effort would be more likely to alert the target to possible penetration attempts, further hampering HUMINT.
Again, no right balance can be struck between HUMINT and the other collection disciplines. Such an idea runs counter to the concept of an all-source intelligence process that seeks to apply as many collection disciplines as possible to a given intelligence need. But not every collection INT makes an equal or even similar contribution to every issue. Clearly, having a collection system that is strong and flexible and can be modulated to the intelligence requirement at hand is better than one that swings between apparently opposed fashions of technical and human collection.
As with all other INTs, it is difficult, if not impossible, to put an ultimate value on HUMINT. It is one of the two most democratic INTs (along with OSINT), because any nation or group can conduct HUMINT. Clearly, it would be preferable to have good HUMINT access for key issues. But cases such as Ames and Hanssen raise questions about HUMINT’s value. These two spies provided the Soviet Union and post-Soviet Russia with invaluable information, largely about U.S. spy penetrations in that country but also, in the case of Hanssen, about technical collection operations and capabilities. When their activities are added to past espionage revelations—such as Kampiles (IMINT), the Walkers (SIGINT), and Pelton (SIGINT)—the Soviet Union and post-Soviet Russia gained substantial knowledge about U.S. collection capabilities. Yet the Soviet Union lost the cold war and ceased to exist as a state. One could argue, on the one hand, that all of this HUMINT ultimately proved to be of no value, thus raising questions about HUMINT’s utility. On the other hand, one could argue that no amount of HUMINT—or any other INT—can save a state that has profound internal problems.
Critics of HUMINT argue that the most important spies (Penkovsky, Ames, Hanssen, and many others) have tended to be walk-ins rather than recruited spies, which raises a serious question about HUMINT capabilities. If one accepts the idea that collection is a synergistic activity, then even the recruitment of lower-level spies adds to one’s overall knowledge. Also, even if the most productive spies are walk-ins, some sort of apparatus is needed to handle them, to get out the intelligence they provide, and so on.
One of the major concerns in HUMINT is the possibility that a clandestine officer will be caught and unmasked, with attendant personal risk for the officer and political embarrassment for the state that sent the officer. Even a successful long-term espionage penetration can prove costly. The case of Gunter Guillaume is illustrative. Guillaume was an East German spy who was able to penetrate the West German government, rising to a senior position in the office of Chancellor Willy Brandt. When Guillaume’s espionage was uncovered in 1974, Brandt was forced to resign. Many people believed that the political cost of the operation exceeded any gains in intelligence. Brandt’s
Ostpolitik
—or favorable policy toward East Germany—was never resumed by his successors, at great cost to East Germany, perhaps even greater than any intelligence that Guillaume produced over the years. Similarly, the fate of Jonathan Pollard (see chap. 15 for more details), who passed classified intelligence to Israel, became a constant irritant in U.S.-Israeli relations, again outweighing the value of the intelligence that Pollard provided.
The state of HUMINT remains a concern in the U.S. intelligence community. HUMINT suffered from budget cuts through the 1990s, as did all aspects of intelligence. Several officials have noted that the FBI had more agents assigned to New York City than did the DO worldwide. President Bush ordered a 50 percent increase in the number of DO (now NCS) officers. As noted, it will be seven years from their entry on duty (EOD) before these officers are considered fully operational. Porter Goss’s tenure as DCI and then DCIA saw the departure of many DO veterans, owing to friction with Goss’s staff. This seems to have eased under DCIA Michael Hayden but there have been press reports indicating that attrition rates in the NCS remained high, especially in the five- to ten-year cadre.
For the United States, at least, it remains important to view HUMINT as part of a larger collection strategy instead of as the single INT that meets the country’s most important intelligence needs. To place that sort of expectation on any one INT is bound to set it up for disappointment at best and perhaps even failure.
OPEN-SOURCE INTELLIGENCE. To some, OSINT may seem like a contradiction in terms. How can information that is openly available be considered intelligence? This question reflects the misconception that intelligence must inevitably be about secrets. Much of it is, but not to the exclusion of openly available information. Even during the height of the cold war, according to one senior intelligence official, at least 20 percent of the intelligence about the Soviet Union came from open sources.
OSINT includes a wide variety of information and sources.
• Media: newspapers, magazines, radio, television, and computer-based information
• Public data: government reports, official data such as budgets and demographics, hearings, legislative debates, press conferences, and speeches
• Professional and academic: conferences, symposia, professional associations, academic papers, and experts
In addition to these open sources, each of the classified INTs has an OSINT component. The most obvious is commercial imagery. One can also conduct a variety of SIGINT-type activities on the Worldwide Web, such as traffic analysis (the number of people who visit a Web site) or changes in Web sites. Given that some aspects of MASINT are related to geophysical phenomena, there are open aspects of MASINT. Finally, there is open HUMINT—the use of overt experts for their own knowledge or as sources of elicitation. This list is by no means exhaustive, but it does give a feel for the range of OSINT within the other INTs.
(See box, “Some Intelligence Humor.”)
SOME INTELLIGENCE HUMOR
In addition to GEOINT, SIGINT, HUMINT, OSINT, and MASINT, intelligence officers, in their lighter moments, speak of other INTs (collection disciplines). One of the most famous is PIZZINT—pizza intelligence. This refers to the belief that Soviet officials based in Washington, D.C., would keep watch for large numbers of pizza delivery trucks going late in the evening to the CIA, DOD, the State Department, and the White House as an indication that a crisis was brewing somewhere The notion was that, after seeing many trucks making deliveries, the officials would hurry back to the Soviet embassy to alert Moscow that something must be going on somewhere in the world.
Some other INTs that intelligence officers talk about are
LAVINT: lavatory intelligence, such as heard in restrooms,
RUMINT: rumor intelligence,
REVINT: revelation intelligence, and
DIVINT: divine intelligence.
One of the hallmarks of the post-cold war world is the increased availability of OSINT. The ratio of open source to classified intelligence on Russia has more than reversed from its 20:80 ratio during the cold war. The number of closed societies and
denied areas
has decreased dramatically. Many of the former Warsaw Pact states are now NATO allies. This does not mean that classified collection disciplines are no longer needed, but that the areas in which OSINT is available have expanded.