Human-source intelligence

From Citizendium
Jump to navigation Jump to search
This article is developing and not approved.
Main Article
Related Articles  [?]
Bibliography  [?]
External Links  [?]
Citable Version  [?]
This editable Main Article is under development and subject to a disclaimer.

Human-source intelligence is the oldest form of intelligence collection. It is not limited to espionage; when a prehistoric tribe sent a skilled warrior to observe another tribe and tell his war chief of its strength, that reconnaissance was an example of human intelligence. Human intelligence often is abbreviated as HUMINT.

It includes all those forms intelligence gathering that involve means of interpersonal contact, as opposed to the more technical intelligence gathering disciplines such as signals intelligence (SIGINT), imagery intelligence (IMINT), and measurement and signature intelligence (MASINT). NATO defines HUMINT as "a category of intelligence derived from information collected and provided by human sources."[1] Typical HUMINT activities consist of interrogations and conversations with persons having access to pertinent information.

The techniques of HUMINT interaction can be used both as a means of collecting positive intelligence. As an intelligence collection discipline, it falls under intelligence collection management. The same or similar techniques, however, can be used for intelligence cycle security, in the discipline of counterintelligence. Indeed, a given interview can both give information on enemy penetrations into one's own service (i.e., material of counterintelligence value), but the questions the enemy is asking their agent to answer can tell the positive intelligence analyst more about the enemy's priorities.

The manner in which HUMINT operations are conducted is dictated by both official protocol and the nature of the source of the information. Within the context of the U.S. military, most HUMINT activity does not involve clandestine activities. Both CI and HUMINT do include clandestine HUMINT and clandestine HUMINT operational techniques.

It should be noted that the goal of coercing a false confession, to be used for psychological warfare, is quite different than the HUMINT goal of obtaining information. Coercing a confession is a process of putting information into the subject, not taking it out. [2]

Types of HUMINT sources

Sources may be neutral, friendly, or hostile, and may or may not be witting of their involvement in the collection of information. "Witting" is a term of intelligence art that indicates that one is not only aware of a fact or piece of information, but also aware of its connection to intelligence activities. Examples of HUMINT sources include, but are not limited to, the following:

It is necessary to know the people from whom information is being obtained, and then to obtain it. HUMINT can provide several kinds of information. It can provide observations during travel or other events from travelers, refugees, escaped friendly POWs, etc. It can provide data on things about which the subject has specific knowledge, which can be another human subject, or, in the case of defectors and spies, sensitive information to which they had access. Finally, it can provide information on interpersonal relationships and networks of interest.

HUMINT is both a source of positive intelligence, but also of information of strong counterintelligence value. Interviews should balance any known information requirements of both intelligence collection guidance and of counterintelligence requirements.

Basic HUMINT operations

Human source screening is the logical start of collection of HUMINT. This involves selecting people who may be sources of meaningful HUMINT, possibly positively identifying them, and conducting interviews of various types. Properly recording and cross-indexing the results of interviews is essential. No intelligence collection discipline is more likely to find meaning in apparently small bits of information than is HUMINT. Especially when there is reason to have additional interviews with the same individual, the subsequent interviews need careful planning, especially when the interrogator does not speak the language of the person being interviewed.

Intelligence preparation for working in cultures

As with other intelligence collection disciplines, intelligence analysis can play many supporting roles. An obvious one is biographical intelligence, to help identify known hostile undercover personnel, or individuals who will impartially mislead an assortment of national intelligence services for profit.

Equally important is the broader area of cultural intelligence, which draws heavily on the social sciences. In a book review in the CIA professional journal, Lloyd F. Jordan recognizes two forms of study of culture, both of which are relevant to HUMINT. [3] In the review, Jordan describes Patai's book[4] as an excellent example of a second type of cultural analysis. He reviews the first group of scientific analyses of culture and character as beginning with "cultural anthropology as early as the 1920's. During World War II, those methods employed earlier in the academic community in this field of research were brought to bear upon a variety of problems connected with the war effort.

"It was precisely the inaccessibility of the target country and the availability of only fragmentary information about it that made national character research relevant to intelligence analysis during the war. The cultural anthropologists had long been developing models of former and disappearing cultures from fragmentary materials. The anthropologists, joined by the psychiatrists, combined the use of psychoanalytic theory, interaction theory, child development theory, and learning theory with standard anthropological research methods to construct models of the contemporary cultures of wartime enemy countries, Japan and Germany." The classic work of this type is Ruth Benedict's study of Japan, The Chrysanthemum and the Sword.[5]

Jordan observes that Benedict's approach was the only one in use until the late 1950's. "National character studies" focused on the statistically most significant personality characteristics of the group (i.e., the modal personality), rather than the most common manifestation of the traits. "...modal personality construct[s] tended to be related to the total culture, or at least, its salient features."

The second class of studies, of which Patai's is an exemplar for Arab culture, had a more narrow focus. "...they concentrated on the relationship of personality traits to subsets of a given society or a given category of roles of that society, rather than on the identification of relationships between personality and the social structure as a whole." A third category, the comparative study, included Francis L. K. Hsu's Americans and Chinese. [6] Indeed, some recent and controversial works, such as Huntington's The Clash of Civilizations and the Remaking of World Order can be regarded as an extension of comparative study into the idea of conflicts among the groups compared. [7]

Basic differentiation by subject type

Different types of human subjects will share information voluntarily or involuntarily. The interrogator builds a relationship with the subject, a relationship that can be based on trust, fear, friendship, or any of a range of human emotions. Prisoners have an understandable fear of what may happen, and, contrary to "tough guy" images, it can be important to relax him and, to the extent possible, put him at ease/ [8] Some organizations teach their members that the other side tortures everyone, and, if that is known, that fear must be addressed; Japanese prisoners in WWII often attempted suicide for that reason [9] but were sometimes dissuaded, as by Guy Gabaldon.

"The question of torture should be disposed of at once. Quite apart from moral and legal considerations, physical torture or extreme mental torture is not an expedient device. Maltreating the subject is from a strictly practical point of view as short-sighted as whipping a horse to his knees before a thirty-mile ride. It is true that almost anyone will eventually talk when subjected to enough physical pressures, but the information obtained in this way is likely to be of little intelligence value and the subject himself rendered unfit for further exploitation. Physical pressure will often yield a confession, true or false, but what an intelligence interrogation seeks is a continuing flow of information."[8]


Especially when the subject is a prisoner, the screener, who need not be the main interrogator, should examine the Enemy Prisoner of War (EPW) captive tag or other basic information giving the circumstances of capture: when, where, how, by whom, and so forth [10] If the subject is not under any restraint, it is still quite useful for a screener to prepare contact information comparable to the information on the EPW tag.

When the subject is a POW, screeners should pay particular attention to rank insignia, condition of uniforms and equipment, and behavior demonstrated by the source. Screeners should look for things like attempts to talk to the guards, intentionally joining placement in the wrong segregation group, or any signs of nervousness, anxiety, or fear. Any source whose appearance or behavior indicates that he is willing to talk should be noted by the screeners. [10]

Assuming the subject has been under guard, the screener often can get valuable information about the subject's behavior from the guards. They can tell the screener how the source has responded to orders, what requests have been made by the source, what behavior has been demonstrated by the source, and so forth.[10] Along with the basic contact information, such observations can be extremely helpful to the interrogator, who can study the information before the interview. Having background on the subject helps the interrogator retain the initiative in an interview.

Again for prisoners, screeners should examine the documents captured with the source and any documents pertaining to the source.[10] If the subject is voluntary and providing documents, they may even be more valuable. Screeners may need to get help from linguists or document specialists in understanding the material. If the documents have insignia or other graphics, these should be compared with an existing graphics register, and added to it if they are new.

Documents captured with the source (identification card, letters, map sections, and so forth) can provide information that identifies the source, his organization, his mission, and other personal background (family, knowledge, experience, and so forth). This information can be used to verify information from documents captured with the source and further assess his willingness to cooperate. When examining documents, screeners should look for items that will indicate whether the source is cooperative or willing to cooperate based on any specific personal interest.

If the source has information pertaining to new foreign material, contact appropriate TECHINT specialists, and if the source has information of target exploitation interest, contact the appropriate staff members who deal with targeting. These specialists are not necessarily qualified interrogators and may need to meet jointly with the subject and interrogator, or pass questions to the interrogator.

Selecting potential sources

It involves evaluating an individual, or group of individuals, "to determine their potential to answer [intelligence] collection requirements or to identify individuals who match a predetermined source profile. The goal of the process is to identify individuals who may have information of interest, and are willing or can be persuaded to cooperate. Willing individuals will undergo debriefing, and, subject to the relevant laws, regulations and policies of the HUMINT organization, unwilling people may be interrogated.

Screening techniques also can select individuals who may be prospects to collect intelligence or work, in other ways, with the HUMINT organization or its sponsor. Individuals may also be identified who are of interest to counterintelligence or to technical intelligence specialists. [11]

System for Prioritizing Interview Subjects
Cooperation level Cooperation code Knowledgeability level Knowledgeability code
Responds to direct questions 1 Very likely to have pertinent information A
Responds hesitantly 2 May have pertinent information B
Does not respond 3 Unlikely to have pertinent information C
Priorities of subjects for interrogation

It may be necessary to screen individuals in a local area, as residents or possibly workers, to determine if they may be security risks.


Since the identity of individuals is very relevant to HUMINT, it makes extensive use of biometric data (e.g., fingerprints, iris scans, voice prints, facial/physical features) collected on persons of interest[12].

Types of interviews

Often not realized is that either a friendly or hostile subject interview has many of the same dynamics as a psychotherapeutic relationship, although obviously for very different, nontherapeutic reasons. In an extended interrogation session, or series of sessions, if there is dialogue between interrogator and subject, as opposed to a subject remaining silent, there will by dynamics closely related to the psychotherapeutic concept of transference and countertransference. In the former, the subject projects emotional experiences of his own onto the interrogator, and in the latter, the interrogator will start thinking of the subject in terms of his own life experiences.

There are other ways to use social science to understand interviews, in the broad sense of cooperative debriefing and hostile interrogation. Speaking of interrogation in the law enforcement context, "There is a vast literature on interrogation ... but hardly anything on it at all in the field of argumentation theory. There is, of course, a reason for this imbalance. Argumentation theorists have taken as their point of interest rational argumentation in which two parties reason together to try to get at the truth of a matter following collaborative rules of procedure. And the interrogation is scarcely a model of how to conduct balanced rational argumentation. Far from that, it seems to represent a coercive kind of dialogue exchange that is associated with intimidation..." [13] but a debriefing is not necessarily coercive, and, in some cases, even an interrogation becomes less coercive, if the subject begins to identify with the conceptual framework of the interrogator. The latter can be a false Stockholm syndrome context, or it can be a case where the subject actually sees that his belief system has fallacies, and he may start to share aspects of the interrogator's mindset.

One variant occurred with WWII Japanese prisoners who, perhaps having been captured while unconscious, believe that the shame of capture, in their native culture, permanently separates them from that culture. [14] If they are unable to commit suicide,[9] they appeared to see themselves as now part of the interrogator's culture, and became quite cooperative. "...there was one infallible method of convincing a reluctant Japanese prisoner to tell all he knows: to tell him, via interpreters, that they will send his name and picture back to Japan. ... We might note that few Japanese ever attempt to hold anything back. Being taken prisoner is not in their handbooks. No Japanese is ever taken alive. Thus, they are not drilled in the "Name- Rank- Serial No- nothing more" routine. They usually reveal everything easily without any persuasion and seem unhappy when their lack of information does not permit them to answer a specific question. [a captured officer]...doesn’t expect to return to Japan ever, has completely excluded his family and his wife from his thinking, and appears pleasantly resigned to going to the States and working there even as a prisoner for the rest of his life." [14] In argumentative theory, interrogation is a form of information-seeking dialogue, but also can include other kinds of dialogue, such as negotiation. In both police and intelligence interviews, there may be bargaining, where the interrogator offers incentives for the subject to reveal information. The revealed information, especially in a HUMINT context, is not necessarily about the interrogation and also may involve persuasion of the subject to speak.<Walton2003 />

Before going on to study the relationship of the interrogation to information seeking dialogue, some account must be given of other types of dialogue related to interrogation as well. For the interrogation has a way of shifting from pure information-seeking to other types of dialogue. As an example, argumentative theory recognizes ‘persuasion’, in the sense meant in the persuasion dialogue,[13] with certain strategic interrogations, the interrogator might objectively question some of the practices of the subject's side. This is not done for direct elicitation of information, but to set a new context for further interrogation in which the subject might question some of his own loyalties and assumptions. In the case of the Japanese prisoners of the US, they began to operate in that new context.[14]


Debriefing involves getting cooperating human sources to satisfy intelligence requirements, consistent with the rules, laws, and policies of the HUMINT organization. People being debriefed are usually willing to cooperate, although it is possible to obtain information through casual conversation (i.e., elicitation). Debriefing may be conducted at all echelons and in all operational environments. While it is usually face-to-face, it can be done by voice, video, or computer messaging.

Types of people being interviewed include both "tasked" and "non-tasked" individuals. Tasked individuals are, in some way, part of the interviewer's organization.

Categories of interviewees for voluntary debriefing
Tasked Not tasked
Military police and infantry patrols in nominally controlled areas Residents of nominally controlled areas
Special reconnaissance teams (see special reconnaissance) Nongovernmental organization workers in the area of operations
Diplomats of one's own country Friendly or neutral foreign diplomats
National or higher command level subject matter experts (e.g., intelligence personnel) Persons outside the area but knowledgeable about it (e.g., emigres)

Tasked personnel giving brief reports of the enemy use the SALUTE technique. More formal or extensive debriefing methods are used for obtaining specialized or complex information.

Other than talking to tasked personnel, there is a tendency for some HUMINT collectors to regard debriefing as a waste of time. The approach to a voluntary source needs to be quite different than even to a cooperative prisoner, especially if the interrogator has reason to believe the source is knowledgeable. While a subject may be a volunteer, a refugee or displaced person is likely to have some of the fears and uncertainty undergone by POWs. Active listening and sympathy can pay great benefits, especially in the areas of love of family, and anger at those who made them homeless. .

The HUMINT collector should allow specialized or senior sources more latitude to interpose their opinions and evaluations. Prior to the meeting, collectors need to examine all available information, to have an idea of the subject's personality and motivations when beginning to talk to them. It also may require unobtrusive observation of the subject to establish such things as patterns of activity and likes and dislikes. The closer the interview environment can be to the customary surroundings of the subject, the more comfortable and cooperative the source may be.

One example of source that should have latitude are trained foreign internal defense (FID) or unconventional warfare (UW) personnel that work with local residents, or military forces, on a routine basis. Such people may very well themselves have HUMINT and/or CI training; US Special Forces groups have two-man HUMINT/CI teams that can augment operational detachments.

Historically, after the WWII experience of resistance leaders such as the Jedburgh teams, occurred to various commanders that soldiers trained to operate as guerrillas would have a strong sense of how to fight guerrillas. Even before specialist FID/UW units were constituted, organizations such as Military Assistance Advisory Groups, as in Greece just after WWII, also had extensive local knowledge. Before the partition of French Indochina in 1954, French Groupement de Commandos Mixtes Aéroportés(GCMA) took on this role, drawing on French experience with the Jedburgh teams [15]. GCMA, operating in Tonkin and Laos under French intelligence, was complemented by Commandos Nord Viêt-Nam in the North. In these missions, the SOF teams lived and fought with the locals.

United States Army Special Forces, US Marine Corps Combined Action Program (CAP) program personnel (originally in Vietnam [16], but now in Iraq [17]) and various training detachments are apt to have valuable and structured information.

In all cases, the more knowledgeable the interrogator is to the volunteer, the better the result is likely to be. A collector does not need to keep the same level as control as in a hostile interrogation. Sometimes, admitting ignorance of a custom, and respectfully asking for explanation, will trigger a flow of information.

While it takes sophistication, the best general approach to willing subjects is a planned elicitation of information, always with a specific goal in mind. The key to elicitation is the establishment of a rapport between the elicitor and the source, normally based on shared interests. In the initial stages of an elicitation, the collector confines his conversations to innocuous subjects such as sports and social commentary. Dependent on the value of the source, the collection environment, and the security consciousness of the subject, the HUMINT specialist will then shift to a more focused topic.

Once in that mode, elicit the information by continuing to ask for clarification, with questions of the form "I agree, however, what did you mean by….?") or expressing a hypothetical situation. The focused discussion can involve mild flattery and interest in the conversation, or, in a much more delicate approach, selectively challenging statements or introducing new information to show knowledge and stimulate more responses.[18]


As opposed to debriefing, the subject of interrogation is not necessarily cooperating with the obtaining of information by the organization. The subject is normally in custody, although the legal circumstances may be such that an uncooperative subject may be able to leave.

Examples of subjects being interrogated include POWs, individuals detained by patrol as not being from the area, and a thief arrested by the civilian police.

Interrogation is a skilled technique, which often involves building rapport with the subject. In an intelligence context, interrogators should be trained specialists, although they may work with linguists and subject matter experts [19].

Regardless of whether the interview is voluntary or involuntary, the interrogator needs to keep the initiative. To keep the initiative, the interrogator may not need to be harsh. Indeed, the many successful interrogators are formally polite within the subject's cultural traditions.[10] If, in societies with a strong host-guest tradition, the interrogator takes the role of host, that can allow polite domination of the conversation.

The interrogation process itself is a HUMINT collection, not analysis, technique. It may well be that the interrogator, after the interview, does analysis, cross-checking statements made against name indices and "wiring diagrams" of social networks. [20] The interrogator checks his notes against the finished report to ensure that the report contains and identifies the information as heard, seen, or assumed by the source [10]

Planning the initial interview

While the actual interview technique will vary with the attitude of the subject and the needs of the interviewer's service, the interrogator should develop a basic plan for the first interview. This might be put into a paper folder along with documents (or copies) associated with the subject, or, if available, it can go into a HUMINT database so other interrogators, analysts, cultural and language specialists, can review it. Reviews can help plan the current interview, or give ideas to other interrogators for other interviews.

The basic elements to collect are:

  • Any urgent HUMINT collection requirements
  • Serial number of EPW/detainee to be questioned.
  • Location and time for the questioning.
  • Primary and alternate approaches.
  • Questioning plan including topics to be covered and the planned sequence of these topics.
  • Prepared questions for unfamiliar or highly technical topics.
  • Method of recording and reporting information obtained
  • Linguist support, if needed
  • Additional general or technical interrogators needed

"The preliminary interview is not intended to obtain intelligence, but only to enable the interrogators to make a firm assessment of the character and type of subject with whom they will have to deal." [8]

How much psychological pressure to use, how many symbols of dominance are appropriate to use, require great judgment. Some classic interviewing techniques, without threatening the subject, make him uncomfortable, as, for example, having "interrogators - preferably two of them - seated behind a table at the far end of a long room, so that the subject after entering will have some distance to walk before taking his chair in front of them. This device will enable them to observe his poise and manner, and may often quite unsettle the subject. In this pressure-oriented technique, the interrogators should sit with their backs to the main source of light in order to obscure their faces, veil their expressions, and place a strain on the prisoner. The subject can be placed under further strain by providing him an uncomfortable chair, say one with a polished seat and shortened front legs so that he tends to slide off it, or one with wobbly legs. [8]

"On the other hand, an opposite technique has sometimes been successful: the prisoner is made so comfortable, after a hearty lunch with beer, that he drops his guard in drowsiness. [8]

When conducting military source operations, the location of the questioning will have psychological effects on the source. The questioning location should be chosen and set up to correspond to the effect that the HUMINT collector wants to project and his planned approach techniques. For example, meeting in a social type situation such as a restaurant may place the source at ease. Meeting in an apartment projects informality while meeting in an office projects more formality. Meeting at the source’s home normally places him at a psychological advantage, while meeting in the HUMINT collector’s work area gives the collector a psychological edge. The HUMINT collector should consider the status and level of the source, security, the workspace available, furnishings, the amount of lighting provided, and the ability to heat or cool the room as needed. [11]

Characteristics of the interrogator(s)

Interrogation is an interaction, and, even before considering the different attitudes the subject may have, the interrogator needs to know his own style, strengths and weaknesses. He needs to judge if he needs cultural advice, how to handle language issues, and if he needs specialist help.

The HUMINT collector must also consider the physical conditions of the source and himself. After extended operations, there may be a limit on how long either the HUMINT collector or source can concentrate on a given subject.

Support needs

If the interrogator thinks that offering incentives will help, he must decide what they are and how they are obtained. If incentives were promised in earlier interviews, the collector must know if they were delivered. If they were not, the interrogator needs to plan how the failure to deliver may interfere with any trust built, and how to correct the problem.

The HUMINT collector must decide if he will need additional support including analytical, technical, or interpreter support. On rare occasions, it may be desirable for the HUMINT collector to seek polygraph support; intelligence services have varying levels of trust in the controversial device. In some cases, the fear of the unknown in a polygraph will have more to do with eliciting responses than anything the device actually measures; there are numerous anecdotal accounts of interrogators putting together random pieces of equipment, telling a subject it is a "lie detector", and breaking resistance. In other cases, it may be appropriate to get trained help from a Behavioral Science Consultant (BSC). BSCs are authorized to make psychological assessments of the character, personality, social interactions, and other behavioral characteristics of interrogation subjects and advise HUMINT collectors of their assessments, as needed.

Attitudes of the subject

The basic attitude of the subject will help define the approach to the interview. There are four primary factors that must be considered when selecting tentative approaches:

  • The source's mental and physical state. Is the source injured, angry, crying, arrogant, cocky, or frightened?
  • The source's background. What is the source's age and level of military or civilian experience? Consider cultural, ethnic, and religious factors.
  • The objective of the HUMINT collection. How valuable is the source’s potential information? Is it beneficial to spend more effort convincing this source to talk?
  • The HUMINT collector himself. What abilities does he have that can be brought into play? What weaknesses does he have that may interfere with the HUMINT collection? Are there social or ethnic barriers to communication? Can his personality adapt to the personality of the source?
Cooperative and friendly

A cooperative and friendly source offers little resistance to the interrogation and normally speaks freely on almost any topic introduced, other than that which will tend to incriminate or degrade him personally.

To obtain the maximum amount of information from cooperative and friendly sources, the interrogator takes care to establish and to preserve a friendly and cooperative atmosphere by not inquiring into those private affairs which are beyond the scope of the interrogation. At the same time, he must avoid becoming overly friendly and losing control of the interrogation.[10] The interrogator, however, may need to get expert advice on the customs of the subject's culture. While Americans might not spend time inquiring about the journey to the place of the interrogation, the health of one another's family, etc., not to do so is extremely impolite in some cultures. Not following cultural norms may transform a cooperative subject into a silent one.

For example, Arab cultural norms, reasonable to follow in a friendly conversation with a subject of that culture, do, often in ceremonial terms, speak of family. Handshakes are traditional at the start and end of the meeting [21].

Neutral and nonpartisan

A neutral and nonpartisan source is cooperative to a limited degree. He normally takes the position of answering questions asked directly, but seldom volunteers information. In some cases, he may be afraid to answer for fear of reprisals by the enemy. This often is the case in low-intensity conflict (LIC) where the people may be fearful of insurgent reprisals. With the neutral and nonpartisan source, the interrogator may have to ask many specific questions to obtain the information required.

Hostile and antagonistic

A hostile and antagonistic source is most difficult to interrogate. In many cases, he refuses to talk at all and offers a real challenge to the interrogator. An interrogator must have self-control, patience, and tact when dealing with him.

Inexperienced interrogators may do well to limit their interview, and pass the subject to a more senior interrogator. When handing off the subject, the next interrogator will find it very useful to have any clues why the subject is being hostile.

"The recalcitrant subject of an intelligence interrogation must be "broken" but broken for use like a riding horse, not smashed in the search for a single golden egg."

No two interrogations are the same... Each interrogation is thus carefully tailored to the measure of the individual subject. The standard lines of procedure, however, may be divided into four parts:

  1. arrest and detention
  2. preliminary interview and questioning
  3. intensive examination
  4. exploitation.

The first three steps, which emphatically do not involve torture, can be combined and called the "softening-up" process. If the subject is using a cover story, it may be broken by softening-up.

"Showing some subjects up as liars is the very worst thing to do, because their determination not to lose face will only make them stick harder to the lie. For these it is necessary to provide loopholes by asking questions which let them correct their stories without any direct admission to lying".[8]

The questioning itself can be carried out in a friendly, persuasive manner, from a hard, merciless and threatening posture, or with an impersonal and neutral approach. In order to achieve the disconcerting effect of alternation among these attitudes it may be necessary to use as many as four different interrogators playing the following roles, although one interrogator may sometimes double in two of them:

  1. the cold, unfeeling individual whose questions are shot out as from a machine-gun, whose voice is hard and monotonous, who neither threatens nor shows compassion.
  2. the bullying interrogator who uses threats, insults and sarcasm to break through the subject's guard by making him lose his temper or by exhausting him.
  3. the ostensibly naive and credulous questioner, who seems to be taken in by the prisoner's story, makes him feel smarter than the interrogator, gives him his rope and builds up false confidence which may betray him.
  4. the kind and friendly man, understanding and persuasive, whose sympathetic approach is of decisive importance at the climactic phase of the interrogation. He is most effectively used after a siege with the first and second types, or after a troubled sleep following such a siege.

Planning additional interviews

This can be part of softening-up, or part of the exploitation phase. "When the cover story and the will to resist have been broken, when the subject is ready to answer a series of carefully prepared questions aimed at an intelligence target, the exploitation can begin, often in a veiled spirit of cooperation and mutual assistance. At this stage the interrogation may for example be moved to an office assigned the subject, where he might even be left alone for a few minutes to show that he is being trusted and that there is something constructive for him to do. This feeling of trust and responsibility can be very important to a broken subject, because he may now have suicidal inclinations; he must be given something to occupy his mind and keep him from too much introspection.

Which course is better will depend on the subject's character, the way he was broken, and his present attitude toward those who have been handling him. Sometimes only a fresh interrogator can get real cooperation from him. Sometimes, on the other hand, he is so ashamed of having broken that he is unwilling to expose himself further and wants to talk only to his original questioner. And sometimes he has built up a trustful and confiding relationship with his interrogator which should not be destroyed by the introduction of another personality.

Principles of questioning

HUMINT specialists must take on a role appropriate to their understanding of the source, but be ready to change it if verbal and non-verbal clues show the role is not productive. In ideal situations, the more time spent in assessing the subject, the better, but real-world constraints will include the likely quantity and value of information the source possesses, the availability of other sources with knowledge on the same topics, and available time.

With due regard to the culture, the initial context should be formal. If the source becomes more cooperative, the interviewer may be more relaxed; if the source is hostile, the interviewer, subject to policy, may take on a more confrontational role. .

If a source cooperates, his or her motivations vary. They can range from altruism to personal gain; they may be based on logic or emotion. From a psychological standpoint, the HUMINT collector must remember that subjects: [22]

  • Want to talk when they are under stress and respond to kindness and understanding during trying circumstances. For example, enemy soldiers who have just been captured have experienced a significant stress-producing episode. The natural inclination is for people to want to talk about this sort of experience. If the EPW has been properly segregated and silenced, the HUMINT collector will be the first person the EPW has a chance to talk to. This is a powerful tool for the collector to use to get the subject talking. The desire to talk may also be manifested in refugees, DPs, and even local civilians when confronted by an unsettled situation.
  • Show deference when confronted by superior authority. This is culturally dependent but in most areas of the world people are used to responding to questions from a variety of government and quasi-government officials.
  • Operate within a framework of personal and culturally derived values. People tend to respond positively to individuals who display the same value system and negatively when their core values are challenged.
  • Respond to physical and, more importantly, emotional self-interest. This may be as simple as responding to material rewards such as extresponding to support in rationalizing guilt.
  • Fail to apply or remember lessons they may have been taught regarding security if confronted with a disorganized or strange situation.
  • Be more willing to discuss a topic about which the HUMINT collector demonstrates identical or related experience or knowledge.
  • Appreciate flattery and exoneration from guilt.
  • Attach less importance to a topic if it is treated routinely by the HUMINT collector.
  • Resent having someone or something they respect belittled, especially by someone they dislike.

Building rapport

Repeated experience shows that the most information comes from situations where the collector and the source establish some type of rapport. [20] This need not be friendship, but it does include some level of mutual respect, and confidence that the collector is professional. This does not necessarily equate to a friendly atmosphere. It means that a relationship is established and maintained that facilitates the collection of information by the HUMINT collector. The HUMINT collector may establish a relationship as superior, equal, or even inferior to the source. The relationship may be based on friendship, mutual gain, or even fear.

If he does introduce himself, normally he will adopt a duty position and rank supportive of the approach strategy selected during the planning and preparation phase. The HUMINT collector must select a rank and duty position that is believable based on the HUMINT collector’s age, appearance, and experience. A HUMINT collector may, according to international law, use ruses of war to build rapport with interrogation sources, and this may include posing or “passing himself off” as someone other than a military interrogator. However, the collector must not pose as—

  • A doctor, medic, or any other type of medical personnel.
  • Any member of the International Committee of the Red Cross (ICRC) or its affiliates. Such a ruse is a violation of treaty obligations.
  • A chaplain or clergyman.
  • A journalist.
  • A member of the civilian government, such as a Member of Parliament. [23]

"The attitude of the interrogators at the preliminary interview should usually be correct, studiously polite, and in some cases even sympathetic. It is imperative that they keep their tempers both now and throughout the interrogation. The prisoner may be given the true reason for his arrest or a false one, or he may be left in doubt, according to the circumstances of the case. The interrogators must try to determine whether his usually vigorous protestations of innocence are genuine or an act, but they should not at this stage give any indication of whether they believe or disbelieve him. A clever prisoner will try to find out how much the interrogators know; they should at all costs remain poker-faced and non-committal. [8]

"At this interview the interrogators should do as little as possible of the talking, however many questions they are anxious to have answered. The prisoner should be asked to tell his story in his own words, describe the circumstances of his arrest, give the history of some period of his life, or explain the details of his occupation. The object is to get him to talk without prompting in as much continuous narrative as possible; the more he talks the better the interrogators can assess his personality.

Use of recorders

Military interrogations are, by doctrine, recorded, by a stenographer or electronic equipment. Audio and video recording is particularly useful if the interview requires an interpreter. If the interpreter is not a regular member of the interrogator's forces, a recording, which can be reviewed by a staff linguist, can be an important part of quality control. The interrogator(s), however, cannot be distracted by doing the recording.

"Whether the stenographer or recorder should be concealed or visible depends on the subject's sophistication and the state of his alert. If the recording process is not evident some subjects may become careless of what they say when they see that the interrogators are not taking notes, whereas a visible recording would alert them to be more cautious. For others, consciousness of a recording going on in full view may be unnerving, and they may betray the weak links in their stories by showing signs of distress at these points.

"At a later stage of the interrogation it may be of value to play back to the subject some part of this recording. The sound of his own voice repeating his earlier statements, particularly any with intonations of anger or distress, may make a psychological breach in his defenses."

Working with interpreters

"Without doubt an interrogator using the subject's language is in a much better position than one who has to work through an interpreter. But the interrogation skill is infinitely more important than the language skill, and a good linguist should not be substituted for a good interrogator. In the absence of an interrogator who speaks the language, an interpreter should be used, preferably one with some training in interrogation techniques. It is very important that the interpreter not only report accurately what both parties say but also reflect as faithfully as he can their inflection, tone, manner, and emphasis. He should try to become part of the furniture in the room rather than a third personality, and the interrogator should act as though he were not there." [8]

Interpreters are frequently a necessary aid to HUMINT collection. There are certain advantages to using an interpreter. The most obvious is that without an interpreter, a HUMINT collector without the proper language or necessary proficiency in that language is severely limited. Furthermore, if properly trained, briefed, and assessed, the interpreter can be a valuable assistant to the HUMINT collector.

Interpreters may have more knowledge of the local culture as well as the language, but the HUMINT manager must be cautious that the interpreter is not, for example, a member of a subculture, religion, etc., of the area, that would be offensive to the subject.

Going through an interpreter will slow the process, and also increase the chance of miscommunication. HUMINT collectors also need to be sure that local or third-country nationals are aware of security, and indeed will be loyal to the interpreter's side.

The collector should be aware of potential problems in interpretation. Some warning signs include:

  • Long-to-short. If the interpreter delivers a long statement of yours as a short statement in the subject language, there are several possibilities. The interpreter may be trying to speed the interview, and the collector has to insist on accurate translation. The question may relate to a sensitive subject, perhaps even inducing fear in the interpreter.
  • Short-to-long. If a brief question of the collector turns into a lengthy statement, it must be resolved that the interpreter does not have his own agenda. It is possible that the interpreter is using honorifics, indirect approach, or other culturally appropriate phrasing. There are times, however, when a translation, to control the relationship, needs to be translated exactly and without indirect language.
  • Body-language shift. If the interpreter seems to change his body language or tone, the collector needs to learn the reason. This may be a justification to suspend the questioning until the interrogator can speak privately with the interpreter and find out if there is a problem. Sometimes, the change may be do to nothing more than fatigue or physical discomfort, but the interrogator must be sure that the native-language conversation has not gotten into dislike, threats, or personal connection that would bias the translation.

Recruited/trusted sources

Special Reconnaissance Patrolling

Special reconnaissance is done by soldiers, normally uniformed, who observe enemy activity deep beyond the front line of one's own side. Since these are highly trained specialists, they will usually have been communicating clandestinely to the HUMINT organization, and will be systematically prepared for debriefing. The debriefing may be done by HUMINT officers of their own organization, who are most familiar with their information-gathering techniques. Some of those techniques may be extremely sensitive and held on a need-to-know basis within the special reconnaissance organization. They operate significantly farther than the furthest forward friendly scouting and surveillance units; they may be tens to hundreds of kilometers deeper. They may enter the area of operations in many ways.

Their mission is not to engage in direct combat. It may be to observe and report, or it may include directing air or artillery attacks on enemy positions. If the latter is the case, the patrol still tries to stay covert; the idea is that the enemy obviously knows they are being attacked, but not who is directing fire.


Espionage is the collection of information by people either in a position of trust for the enemy, or with access to people with such access. The process of recruiting such individuals and supporting their operations is the HUMINT discipline of agent handling.

It may be possible for an agent handler to meet directly with the agent and debrief him. More commonly, agents send messages to the organization for which they work, by radio, Internet, or by leaving the messages in a hard-to-find place. The latter technique, called a dead drop, will have either a courier or the agent handler retrieve them in a clandestine manner. These are examples of espionage tradecraft.

Analyzing relationships among HUMINT subjects

After interviews, be they debriefings or interrogations, there is likely to be data about other people with whom the subject has had contact or knows about. These data are focused on human relationship networks, not, for example, on military information that the subject knows.

Once information is obtained, it is put into an organized form. Very frequently, information obtained at one interview may help structure the next interview with the same person, or with another subject.

Identifying other people of interest

During interviews, a subject is apt to mention things about other people, or be prompted in a seemingly conversational way.

Social networks

A first interview could begin with a villager named Anna. The interviewer notices that she is wearing a very attractive scarf, and inquires about it, careful to avoid customs that might obligate the wearer to give him the object he admired. She said "Thank you. I bought it from the silk man."

Human network information from first interview

The interviewer, really not knowing many of the villagers, guesses and prompts. "The silk man? Is that the son of the olive farmer married to...ummm...I will think of their names".

Anna responds, "No, the silk man is not of the village. I do not know his name."

Human network information from second interview

Knowing another subject to be debriefed is from the same village, the interviewer might arrange to have cheese and olives available as snacks in the interview room. During the interview with George the villager, the interviewer might say "these are poor olives, but the best I could find in the local market. They are not as good as Gregory, in your village, grows."

"Yes, Chino has excellent trees. It's too bad only one of his sons, Enzo, stayed on the farm. He misses his wandering son who sells cloth to women."

Confirming Anna's statement that the silk man is an outsider, the interviewer seeks confirmation. "Is that the silk man?"

"No, the silk man is Hassan, but only visits the village. Gregory's son Fritz sells wool."

"Do you know the sons of Diana and Chino?"

"Oh yes! I ran a long time to get Iliana the midwife to come. She delivered both his sons."

The interviewer sipped his tea, and asked "Midwives are so valuable. Yet what if a baby is coming in wrong, and Iliana cannot get it?"

"Iliana's husband, Justinian, has a truck. Iliana will decide whether he should go get the doctor, or bring the mother and herself to the doctor."

Human network information from third interview

The interviewer sees Anna in the street, getting out of Justinian's truck, and asks if all is well. "Oh yes, I was just visiting the doctor and everything is fine." Note that Anna has confirmed a relationship between Justinian and the doctor, whose name is not yet known.

"Anna, I forgot to ask. Who is your husband?"

"I thought everyone knew that Homer and I are such lovers. You must be blind!" The culturally sensitive interviewer laughed, and agreed he needed new glasses.

Human network information from repeat interview of first subject

The interviewer now knows that the Silk Man and the Doctor are outsiders that visit the village. If there is underground activity, they are possible couriers. Either the interviewer, or a HUMINT analyst, need to find other villages that are visited by the Silk Man and the Doctor, and see if any guerrilla events correlate to their visits.

Operational network "wiring diagram"

Much modern interest in tracking networks of people are relevant to guerrilla operations and terrorist networks, two loose categories that do not completely overlap. When examining the overall structure of terrorist groups, there are two general categories of organization: networked and hierarchical. A terrorist group may employ either type or a combination of the two models. Newer groups tend towards organizing or adapting to the possibilities inherent in the network model. Ideology can have an effect on internal organization, with strict Leninist or Maoist groups tending towards centralized control and hierarchical structure [24]. Whether the organizational model is hierarchical or not, the operational personnel almost invariably use the cell system for security.

One relevant study looks at modeling terrorist networks in a manner similar to other systems that "exhibit regularity but not periodicity (i.e., locally random, but globally defined) [25].

Their model focuses at the "mid-range", "not at the level of state leadership, and not at the level of mapping and predicting the behavior of each individual terrorist, but rather at an intermediate or organizational level"... Much as vulnerability analysis of connectionless packet networks such as the Internet [26] concentrates on the nodes whose loss would most interfere with connectivity, the study here looks for the "pattern of connections surrounding a node that allows for wide network reach with minimal direct ties. “Structural holes” at the intersection of flows across knowledge communities position unique and superior nodes. It is the individuals spanning these “internal holes of opportunity” that impact the network’s functioning and performance. The implicit corollary of this is that if a small number of these critical nodes can be identified and “clipped” from the network, then command signals will not be able to propagate through the system."

In the 9/11 attack case, the pilots were such key nodes, once the US operational groups were in-country and operating [27]. Taking the observation of centrality a step farther, COMINT can complement HUMINT in finding the nodes of a geographically dispersed human network.

Obstacles to development of HUMINT capabilities

The following observations are drawn from the article by Lawrence Wright who interviews Director of National Intelligence John Michael McConnell.[28].

Inability to recruit

To develop HUMINT agents it is necessary to recruit HUMINT controlling officers with native foreign-language skills. However, the U.S. security clearance process has several problems:

  • It is biased against first-generation immigrants with active relationships to their former country.
  • It is biased against non-heterosexuals. (For overview see Wikipedia article Sexual orientation and military service.) The U.S. Army, for example, his dismissed hundreds of gay translators during a time of critical national need for Arabic linguists.[29] This while, according to Wright's article, the FBI has gone from 8 Arabic-speaking agents to 9 in the seven years since the 9/11 attacks.
  • It takes up to two years to complete, longer than many immigrants are willing to put their careers on hold while waiting for a clearance.

McConnell's solution to this is to

  • Shorten the clearance process to a month or less.
  • Subject officers to "`life-cycle monitoring' -- that is, constant surveillance".

There have been other cases in the past where it has been necessary to balance security clearance policy against national needs, for example:

DNI McConnell notes also in Wright's article that the predominant risk factor which turns officers into traitors is not ethnicity, sexuality or politics, it is money:

"`Look back at all the spies we've had in our history', he said. `About a hundred and thirty. How many did it for money? A hundred and twenty-eight.'"

For example,

  • Philip Agee received $1 million from Cuban intelligence service to publicize the names of CIA agents as part of an active measures campaign by the KGB.[30]
  • Israel Defense Forces psychiatrist David Shamir "attempted to make contact with the Iranian Foreign Ministry, Hamas officials, and the Russian intelligence service, the FSB, with the intention of selling them classified information which he came upon during his military service".[31] While his claimed motivation was political, the bottom line was still money (emphasis added in quote):
"Shamir said his prime motivation for his actions was what he viewed as the continued breakdown of the state's social fabric. `The idea was to save my life and that of my son and the last years of my parents' lives, and to have things happen so that we can really be saved,' he said. `The thinking was that I would receive money which I would use to attain asylum for me, my son, and my family, and to enable them to live as reasonably as possible, not in Iran, not in Lebanon, but rather in a European country.' "

Inability to trust

Wright claims that

"Much of the intelligence community is technophobic and is also hamstrung by security concerns. Only recently have BlackBerrys made their way into some agencies, and many offices don't even have Internet connections."

Hence it is difficult to institute technical controls that would allow the organization to effectively compartmentalize the knowledge of officers in the organization so that even moderately untrusted people could serve. I.e., in many cases of "moles" or internal spies, low-level spies have had access to huge volumes of secret data. This would not be possible if adequate technical measures were instituted to effectively comparmentalize information access on a more finely categorized and motivated need-to-know basis. In other words, there is a so far lost opportunity to use technical measures to improve operational security while at the same time allowing a broader range of people to serve. Another aspect of this would be to automatically filter information "down to" and tailored to a specific subject-matter need-to-know characterization of the requesting individual.

Inability to share

A third obstacle to HUMINT and intelligence analysis in general is effective information sharing across the intelligence community. Wright's article notes that

  • "the community still relies on more than thirty online networks and eighty databases, most of which are largely inaccessible to one another."

Lack of information sharing has been partly addressed technically by adding new information-sharing tools

  • Intellipedia, using Wikipedia software with additional security features.
  • A-Space, "based on sites such as MySpace and Facebook -- in which analysts post their current projects as a way of creating social networks."
  • The Library of National Intelligence[32], which "is an online digest of official reports that will soon provide analysts who use it with tips, much the way Amazon and iTunes offer recommendations to their customers."

However, Wright notes that "These innovations have not yet made their way to the analysts and agents in the field", and "the intelligence community has only warily appropriated models whose usefulness is blindingly obvious".


  1. AAP-6 (2004) - NATO Glossary of terms and definitions
  2. Albert D. Biderman (1957 September), "Communist Attempts to Elicit False Confessions from Air Force Prisoners of War", Bull N Y Acad Med. 33(9): 616–625
  3. Jordan, Lloyd F. (8 May2007), "The Arab Mind by Raphael Patai. Book review by Lloyd F. Jordan", Studies in Intelligence
  4. Patai, Raphael (1973). The Arab Mind. Charles Scribners and Sons. 
  5. Benedict, Ruth (1989). The Chrysanthemum and the Sword: Patterns of Japanese Culture. Mariner Books. 
  6. Hsu, Francis L. K. (1953). Americans and Chinese: two ways of life. Henry Schuman. 
  7. Huntington, Samuel P. (1998). The Clash of Civilizations and the Remaking of World Order. Simon & Schuster. 
  8. 8.0 8.1 8.2 8.3 8.4 8.5 8.6 8.7 Compos, Don, "The Interrogation of Suspects Under Arrest", Studies in Intelligence. Retrieved on 2007-11-07 Cite error: Invalid <ref> tag; name "Compos" defined multiple times with different content
  9. 9.0 9.1 Gabaldon, Guy, A Lone-Wolf Marine: How One Man Captured 1,500 Japanese on Saipan. Retrieved on 2007-11-07
  10. 10.0 10.1 10.2 10.3 10.4 10.5 10.6 Department of the Army (8 May 1987), Field Manual (FM) 34-52 Intelligence Interrogation. Retrieved on 2007-11-07
  11. 11.0 11.1 US Department of the Army (September 2006), FM 2-22.3 (formerly FM 34-52) Human Intelligence Collector Operations. Retrieved on 2007-10-31
  12. US Department of Defense (22 June 2007), Joint Publication 2-0: Joint Intelligence. Retrieved on 2007-10-31
  13. 13.0 13.1 Walton, Douglas (2003), "The interrogation as a type of dialogue", Journal of Pragmatics 35. Retrieved on 2007-11-07
  14. 14.0 14.1 14.2 Frankel, Stanley A. (1992), Chapter 7, Japanese Prisoners: Docile and Cooperative, Frankel-y Speaking about World War II in the South Pacific. Retrieved on 2007-11-07
  15. Porte, Remy, Intelligence in Indochina: Discretion and Professionalism were rewarded when put into Practice.. Retrieved on 2007-11-26
  16. US Marines Combined Action Platoons (CAC/CAP) Web Site: Vietnam 1965-1971.. Retrieved on 2007-11-26
  17. The Combined Action Platoon in Iraq: An Old Technique for a New War
  18. FM2-22-3, p. 8-23
  19. Educing Information — Interrogation: Science and Art, Foundations for the Future, National Defense Intelligence College Press, December 2006. Retrieved on 2007-10-31
  20. 20.0 20.1 Tourison, Sedgwick Jr. (1990). Conversations with Victor Charlie: an Interrogator's Story. Ballantine Books. 
  21. Green, Eloisa (1990), Center for Army Lessons Learned, Newsletter No. 90-7, Winning in the Desert. Retrieved on 2007-11-07
  22. FM 2-22.3, pp. 8-3 to 8-4
  23. FM 2-22.3, p. 8-4
  24. U.S. Army Training and Doctrine Command, Assistant Deputy Chief of Staff for Intelligence – Threats (15 August 2005), A Military Guide to Terrorism in the Twenty-First Century. Retrieved on 2007-02-11
  25. Fellman, Philip Vos & Roxana Wright, Modeling Terrorist Networks - Complex Systems at the Mid-Range. Retrieved on 2007-11-02
  26. Cooperative Association for Internet Data Analysis (CAIDA), skitter. Retrieved on 2007-11-02
  27. Klerks, Peter, "The Network Paradigm Applied to Criminal Organisations: Theoretical Nitpicking or a relevant doctrine for investigators? Recent developments for the Netherlands", Connections 24 (3). Retrieved on 2007-11-02
  28. Wright, Lawrence (21 January 2008), "The Spymaster", The New Yorker
  29. "Army dismisses gay Arabic linguist", Associated Press, July 27, 2006
  30. "Counterintelligence/Espionage Case: Philip Agee",
  31. "IDF psychiatrist sentenced to 5 years for trying to spy for Iran", by Ofra Edelman,, January 17, 2008
  32. "Top Secret Only: The Library of National Intelligence (LNI)" September 7, 2007,