jim.shamlin.com

Chapter 6 - Analyzing The Answers

In sizable intelligence organizations, the task of gathering content and analyzing content are separated to ensure there is no corruption of the two tasks. The moment that data collected begins to point to a given conclusion, or if the data collector has a foregone conclusion, than the task of collecting information will be biased (consciously or unconsciously) to seek only the information that is necessary to support the expected conclusion.

In addition to skewing the analysis, the combined interrogator-analyst may also influence the individuals he interrogates because his own practices - the questions he asks, tone and posture, etc. may disclose to the respondent what he expects to hear. A respondent who notices these signals may choose to tell the interrogator what he seems to want to hear, or can easily identify and be on guard about a subject he wishes to conceal. And this can be unconscious on both sides of the interrogation.

In smaller organizations that have fewer resources, the two roles are often combined. An investigator who must perform both roles, and should be aware of the conflict that arises, even (and perhaps especially) when has the potential to occur unintentionally.

Analyzing Content

Intelligence is gathered to support a decision to take action - and any information that supports that decision is called "actionable intelligence." Gaining actionable intelligence is the entire point of the investigative process - there are few organizations that want to know things "just because" or are willing to invest time and resources in gathering trivia that is of no specific purpose.

The ideal of investigation is to gather complete information - but this is an ideal rather than a goal because it is never literally possible to know everything. The data must be sufficient to provide a reasonable level of confidence that enough is known to support a decision. A perfect decision requires perfect information, and it is never possible to make one - it's a matter of "good enough."

Requirements

Data requirements can often be developed in advance, as those who are making decisions have a sense of the kinds of information they will need in order to make a sound decision - and will be aware of the kinds of information they do not have at their immediate disposal. Where all data that is required to make a decision is available, there is no need to conduct interrogation.

(EN: This seems self-evident but is often completely ignored. Claiming that there is not sufficient information to decide is often a stall tactic for an individual who is reluctant to decide or hasn't really thought through what he is trying to accomplish. I've found that needling someone about the precise kind of information they need is a good way to expose this and prevent being sent on wild goose-chases. "What do you need to know?" and then "How will that enable you to make a better decision?" often shuts down these nebulous requests - and it's good mental discipline to ask these questions of yourself when faced with a decision.)

In some instances, you can predict that the individuals who can provide the information may be reluctant to do so. The example given is that a person who is applying for a grant may be interested in knowing who are the decision makers, what criteria they consider, and the personal biases they may have in evaluating proposals. The person who knows this might be reluctant to disclose it, as it would give the applicant an "unfair advantage" or it may expose and embarrassing or inappropriate bias in the decision-making process.

Indirect questioning can be used in some instances to gather this information in a manner that does not arouse suspicion or cause people to raise defenses. The skill of "elicitation" involves asking indirect questions in a way that gets individuals to disclose information they might be reluctant to provide.

The author gives the example of a training exercise in which trainees were give a field exercise - to extract personal information from an unsuspecting stranger - and the trainee who got a woman to give him the PIN to her debit card, a detail most people know better than to share with a total stranger. Noticing her using her debit card to pay at a bar, he casually mentioned that he has trouble remembering his pin number. She confessed that she just uses her birthday. A little while later in the conversation, he suggested that her star-sign was Leo, and she corrected him - telling him she was a Pisces and was born on "the ides of March." From this, he was able to deduce her debit card PIN was 0315.

(EN: The author seems to favor direct questioning of a subject, and overlooks the wealth of information that could be gathered from other parties - such as asking a friend about her birthday - or even gathering information online - seeing if her birthday was on her Facebook profile, or even the date on which online friends wished her a happy birthday. It's likely learning to interrogate someone directly is the hardest, but alternate and easier routes are available.)

Reliability

Reliability pertains to whether the information that has been gathered is accurate. This pertains to the quality of the source and the quality of the information.

An individual who is a unreliable source is not necessarily intending to deceive: they may provide false information believing it to be accurate - so do not jump immediately to the conclusion that everyone is attempting to be deceitful. Other bad sources may answer to the best of their knowledge, or they may confabulate something just to give you an answer, and it will turn out to be false information - even if they state it with an air of confidence.

An unreliable source cannot provide good information: what they have told you is true to the best of their knowledge, and their knowledge will not improve during the process of questioning. (EN: My sense is that a person's memory may improve, or they may fill in details that were omitted, but the nature of the information they provided will not be more accurate.) At best, you may get them to admit that they fabricated details - not with the intention to deceive, but with the intention of telling a story that seemed more credible, more complete, or more significant than it actually is.

On the other hand, Information may be unreliable because a respondent is intentionally omitting or changing information in an attempt to conceal the truth or misdirect you, follow-up questioning can help to discover inconsistencies. Entire books have been written on the subject of detecting deception, but the author will consider a few techniques that can be used.

In many instances, a person who is deceptive is telling the truth, but is either leaving out or adding in details about some aspect of it that he wishes to conceal. An employee who stole merchandise can give an accurate account of how he closed down shop and left the premises, leaving out the fact that he slipped something valuable into his pocket - or a driver who hit another vehicle will given an accurate account of how he drove through an intersection, adding in the falsehood that he came to a complete stop before proceeding.

The author suggests there are a number of "text bridges" that should catch an interrogator's attention, which identify where a story has been altered:

The author doesn't provide a very good account of how these words are used by people to conceal or add information to an account, but merely provides a few examples of interrogations in which these words were used proximate to an omission or an added detail in a response.

Another technique, which the author suggests is a bit too complex to describe in much detail, is often useful in piecing together information about a sequence of events, and considers the way in which a speaker's description of events before and after a detail that was added or omitted changes. In particular, people will attempt to slow down while they are thinking up a lie, and speed up after its' been told because they are no longer investing the mental effort in devising the lie. Alternately, they may rush through details up to the point of a lie (because telling the lie is the most important part of their story) and then slow down and take a more relaxed pace after the lie has been offered (and they are no longer anxious). This can involve the pace of speech, the level of details, etc.

If you can invest sufficient time in an interrogation, you may become familiar with an individual's conversational style - and can identify instances in which they have changed. There are a myriad of patterns and habits expressed in the way people speak. Some people are casual and chatty, and when the become terse it's a sign something is off. Or if a person doesn't usually mention specific times when they describe a sequence of events, you should be concerned when they start being very specific about the time.

(EN: My sense is that all of these "flags" are likely very reliable indicators of anxiety - but anxiety does not necessarily mean that their account has been altered. It may simply be something that they are uncomfortable with talking about. Lie detector tests are often used to measure the same things: heartbeat and respiration, tone of voice, etc. The problem is the same: they cannot detect when a person is lying, only when they are anxious or uncomfortable - and it cannot be presumed that their discomfort is caused by being deceitful.)

Logic and Gut Feeling

There is a section in which the author extols "gut feelings" and "instincts" as a means of suggesting that the reader should not be dismissive of the things he "senses." He does not go so far as to say that gut instinct is better than logic, merely that it should not be dismissed - because human beings are social creatures, we have certain natural abilities to sense things in other people, and gut feel can be highly reliable.

(EN: I drop much of the detail here because some of what he has to say seems a bit suspect, and supportive of the notion that gut feel or instinct are mystical or inexplicable forces, which is a bit counterproductive. The theory I prefer is that "instinct" is simply a nonverbal form of logic - we know things without being able to explain, but just because we cannot use words to describe these sensations doesn't mean they are supernatural, merely that we lack the verbal skill to do an adequate job of describing them. You cannot explain the color yellow, but you know it when you see it, and things just "look yellow".)

Analyzing Nonverbal Communication

When questioning individuals, particularly those who are reluctant to share information, it's important to pay attention to more than the words they speak - there are many cues in their voice and body language that will help to indicate whether you are building rapport, getting complete information, and getting truthful information.

(EN: Here I would stress the first, as people very often assume that body language always indicates deceit and overlook the fact that it may simply be communicating discomfort with the interrogation.)

Of particular importance: there's nothing about a gesture or a tone that is indicative in all instances: you need to pay attention to a persons' normal pattern to know whether a given nonverbal is unusual, therefore worth considering.

Vocal Cues

There are various cues in a person's speech that will signal that they are putting inordinate effort into an answer, or attempting to conceal or misrepresent information:

Again, all of these changes may occur even if a person is telling the truth. Consider whether they are comfortable having a conversation: most people dislike making complaints, disagreeing with someone, or saying something negative or unflattering. In those instances, these behaviors indicate discomfort rather than deceit.

Physical Cues

In addition to vocal cues, there are a number of body-language clues: posture, facial expressions, hand gestures, and body movements.

As with verbal cues, it is important to establish a baseline of normal behavior before deciding whether a cue is significant, and to recognize the significance may be discomfort rather than deceit.

Follow-Up Questions

The level of detail in a person's response provides an opportunity for an investigator to cross-check by asking follow up questions that may contradict previous answers.

Consider the length of a person's answers: a short and direct answer is often taken as telling the truth, but it may also be the tactic of someone who is lying and doesn't want to give enough detail to give the lie away. Asking for additional details may cause them to slip as they try to invent a consistent story on the fly.

Longer answers are generally geared to convince the interrogator of the respondent's story - but this does not necessarily mean they are telling a false story, merely that they are nervous and expect follow-up questions, and mean to provide complete information to support their story.

An interrogator can ask for more detail in either case, in a series of follow-up questions that seeks to gather more detail and cross-check some of the earlier answers against later ones. Unfortunately, this can backfire - if the respondent senses you doubt what he says, he will be apprehensive. Again, it is just as likely that an honest person will be exasperated that you don't believe the truth as it is for a dishonest person to be exasperated that you don't believe in his lie.

For that reason, it's important to avoid playing the "gotcha" game in questioning - to point out consistencies immediately and accuse the other party of being dishonest. This can sometimes be useful in getting a liar to confess, but it invariably shuts down the respondent.

There's a brief mention of the "we already know" tactic used by interrogators, in which they tell the respondent a scenario that implies they already know all the details and are just confirming. This can only be done if you do, indeed, know a lot and have suspicions about a few loose details - otherwise it becomes clear to the other party that you don't know anything, in which case a liar will shut down to avoid giving you information that would uncover the truth you don't recognize, and a truthful person will be uncomfortable, expecting that you will regard what they say as a lie if the facts don't correspond to your story.

The best approach to interrogation is to ask questions in a curios manner, listening and taking notes, but not to share any suspicions or assumptions with the respondent unless you have a clear indication that they are not being truthful in their responses.