“A rock pile ceases to be a rock pile
the moment a single man contemplates it,
bearing within him the image of a cathedral.”
Antoine De Saint-Exupery

See if you can imagine this: It is the year 2050, and predictions about people are perfect. They are made with a high-tech chemical test. You can accept a ride from a total stranger, you can ask a homeless person you’ve never seen before to watch your house while you are out of town. You can do this free of fear that they might harm you because predictions of intent and character are totally reliable.

You are skimming along in your hover-craft one afternoon, taking your six-year-old daughter to the park, when you are paged to come to an urgent business meeting. You go to the park anyway and look around for some stranger with whom you can leave your daughter. There is a middle-aged woman sitting on a bench reading a book, and as you sit down next to her, she smiles. Using a device nearly everyone carries these days, you conduct an instant high-tech test on her, as she does on you, and you both pass with flying colors. Without hesitation, you ask if she’ll watch your daughter for a few hours while you skim over to a meeting. She agrees, you exchange some information about how to reach each other, and off you go without any concern, because you have predicted to your satisfaction that this stranger is emotionally healthy, competent, drug-free and trustworthy.


The story sounds far-fetched, but in our time we already make every single one of these predictions about baby-sitters. We just don’t do it as quickly or as accurately.

With present-day technology, how much time would you have to spend with a stranger before she wouldn’t be a stranger anymore? How many of your low-tech tests would a baby-sitter have to pass before you’d trust her? We undertake this common yet very high-stakes prediction by reviewing an application and asking a few questions, but let’s really look at this prediction. For starters, we wouldn’t just interview a woman we met in a park. No, we’d want someone who was recommended by a person we know, because we like to rely on predictions made by others. Our friend Kevin is so bright and honorable, we think, that if he endorses somebody, well, that person must be okay. What often happens, however, is that we attach Kevin’s attributes to the person he recommended, and we don’t listen to our own uncertainty. As we drive away from home, leaving our child behind with someone we met just a half-hour ago, there is that tug that says, “You never really know about people.

In our interview with the baby-sitter, we watch her attentively for any signs of… of what? Drug use? Well, that can be tested with great reliability; tens of thousands of drug-screen tests are done every week by employers who have less at stake than parents do when hiring a baby-sitter. Though most people believe the drug question is a critical one, have you ever heard of a parent requiring a drug screen of a baby-sitter candidate? Or a Breath-alyzer test to see if she’s been drinking? Most parents don’t even contact all the baby-sitter’s references, so it’s no wonder they drive away feeling, “You never really know about people.”

I am not, by the way, suggesting drug tests or polygraphs for baby-sitters, but I am pointing out that we rarely bring even a tenth of the available resources to high-stakes predictions. For example, the question people really want answered by a prospective baby-sitter is: Have you ever mistreated a child? But they never ask it! Why not? Because people feel that asking a question so direct is rude, or ridiculous, since it wouldn’t be answered truthfully by someone who had mistreated children. Ask the question anyway, and how it is answered will make you more comfortable or less comfortable with that applicant. Imagine you asked, “Have you ever abused a child?” and the applicant responded with “Define abuse,” or “What have you heard?” It is entirely fair and appropriate to ask someone to whom you’ll entrust your child to discuss the very issues you care about most. Good applicants will certainly understand, and bad applicants may reveal themselves.


Having not sought any of the information he or she really wants to know, a parent might see the applicant stroke the family cat and think: “She likes animals, that’s a good sign.” (Or worse still: “Tabby likes her, that’s a good sign.”) People want so badly to get someone hired for a job that they spend more time qualifying a candidate than disqualifying a candidate, but this is one process in which it’s better to look for the storm clouds than to look for the silver lining.

Let’s go back to 2050 for a moment. Not only do you have no hesitation about accepting a ride from a stranger, but there’s a city-run transportation computer to facilitate precisely that. Rather than drive yourself from Los Angeles to San Diego, you enter your destination into the computer along with what time you want to leave, and it identifies several other people who are going from your area to San Diego at the same time. A perfect stranger will stop by your house and pick you up, and you’ll get back from San Diego the same way. That’s what could happen if predictions were perfect. Since they are not, a hundred thousand cars carry the passengers that twenty-five thousand could carry just as well. Fear of each other and lack of confidence in our predictions makes any alternative seem impossible.


But what if we had that transportation computer today, and in addition to identifying the people who are making the trip you want to make at the time you want to make it, it also provided some demographic information? You could ride to San Diego in an old van with two unemployed men in their thirties, or you could ride in a late-model station wagon with a housewife and her one-year-old child. You’d likely conclude that the ride with the housewife and her baby would be safer (noisier, perhaps, but safer). What else would you want to know about candidates for ride-alongs? Their criminal histories, driving histories, condition of their vehicle? The point is, if you could learn enough about each candidate, you would actually be comfortable relying upon your prediction because that’s exactly how strangers become people you trust anyway. You learn enough about them. They pass several of your tests and suddenly they aren’t strangers anymore.

Some animals perceive danger chemically—maybe even part of the way we do it is chemical; I don’t know. But will the day really come when we’ll be able to make predictions about people not by judging their appearance or clothing or smile or assurances, but by applying a chemical test? I believe the answer is yes, though I won’t still be around to say I told you so. In the meantime, since we have to keep doing predictions the old-fashioned way, it’s all the more important that we understand what’s really going on.

▪ ▪ ▪

Psychologist John Monahan is a pioneer in the field of prediction who has influenced my work and life a great deal. In his beautifully written book, Predicting Violent Behavior, he begins by asking the simplest question: In which direction would this book fall if you let it go?

The reader could technically state only that every other solid object he or she has let go of in the past has (eventually) fallen down rather than risen up or remained suspended. What allows for the prediction that this object, if released in the future, will also fall down is that we possess a theory—gravity—that can plausibly let us generalize from the past class of cases to the current individual case. The catch, of course, is that we understand gravity much better than we understand violence.

My friend John and I might have a spirited discussion on that, for I know much more about violence than I do about gravity. I believe behavior, like gravity, is bound by some essential rules. Admittedly they may not always apply, but remember, they don’t always apply with gravity either. Where you are (such as in space, or in water) affects how objects will behave. The relationship of objects to each other and to their environment (i.e., magnets, airplanes, etc.) also bears on such predictions. With behavior, as with gravity, context will govern, but there are some broad strokes that can be fairly applied to most of us:

  • We seek connection with others;
  • We are saddened by loss, and try to avoid it;
  • We dislike rejection;
  • We like recognition and attention;
  • We will do more to avoid pain than we will do to seek pleasure;
  • We dislike ridicule and embarrassment;
  • We care what others think of us;
  • We seek a degree of control over our lives;

These assumptions are hardly ground-breaking, and though we might expect something more esoteric about people who are violent, these mundane concepts apply to most of them, just as they do to you. You see, this list contains a few of the ingredients in the human recipe, and how much of one ingredient or how little of another will influence the final result. With the man who goes on a shooting spree at work, it is not that he has some mysterious extra component or that he necessarily has something missing. It is usually the balance and interaction of the same ingredients that influence us all. Am I saying the shooting spree at work can be predicted in part by weighing the balance of factors as common as the eight general assumptions listed above? Yes.


Certainly there are hundreds of other variables that my office considers in predicting violence, and I could present them here with charts and graphs and templates and computer print-outs. I could use psychiatric terms that would require a psychiatrist to interpret, but my purpose here is to simplify, to identify in your experience the factors that matter most.

As I discussed earlier, no matter how aberrant the person whose behavior you seek to predict, no matter how different from him you may be or want to be, you must find in him a part of yourself, and in yourself a part of him. When you undertake a high-stakes prediction, keep looking until you find some common ground, something you share with the person whose behavior you seek to predict—this will help you see the situation as he perceives it. For example, the anonymous caller may seem to enjoy the fear he is causing in his victim. Getting pleasure from the fear of others is something most of us cannot relate to, until we recall the glee of every teenager who startles a friend or sibling by jumping out of the dark. Anyway, with the frightening caller, fear may not be the issue as much as liking attention, which we can relate to. When the caller causes people to feel fear, they are very attentive. It might not be his favorite way of getting attention if he perceived better options or if he felt he brought other assets to his relationship with his victim, but it has likely worked for him in the past. I don’t mean to imply that the threat caller is so introspective that he consciously considered all this, but neither is our behavior usually the result of conscious decision-making.


Though it is true that people have more in common than in contrast, you will encounter some who have vastly different standards of behavior and vastly different ways of perceiving the same events. For example, some people operate without listening to their consciences; they do not care about the welfare of others, period. In the corporate boardroom we might call this negligence; on the street we call it criminality. The ability to act in spite of conscience or empathy is one characteristic associated with psychopaths. Robert Hare’s insightful book Without Conscience identifies several other features. Such people are:

  • Glib and superficial
  • Egocentric and grandiose
  • Lacking remorse or guilt
  • Deceitful and manipulative
  • Impulsive
  • In need of excitement
  • Lacking responsibility
  • Emotionally shallow

Many errors in predicting behavior come from the belief that others will perceive things as we do. The psychopath described above will not. To successfully predict his behavior, you must see a situation your way and his way. It will be easy, of course, to see it your way—that’s automatic. Seeing a situation from another person’s perspective is an acquired skill, but you have already acquired it. Imagine that you are about to fire someone whose behavior, personality, and philosophy of life could not be further from your own. Even with all the differences, you would still know if he’d view the firing as fair, completely unfair, part of a vendetta, or motivated by discrimination or greed, etc. Particularly if you worked closely with this person, you could recite his perception of events much as he would. Though you may not share his view, you can still bring it into focus.


Predicting human behavior is really about recognizing the play from just a few lines of dialogue. It is about trusting that a character’s behavior will be consistent with his perception of the situation. If the play is true to humanness, each act will follow along as it should, as it does in nature.

Imagine you are watching a bird as it floats to earth about to land. The sun is casting the bird’s shadow on the ground, and both bird and shadow move toward the landing point. We know that the bird cannot possibly arrive there before its shadow. L ikewise, human action cannot land before impulse, and impulse cannot land before that which triggers it. Each step is preceded by the step before it. You cannot shoot the gun without first touching it, nor take hold of it without first intending to, nor intend to without first having some reason, nor have a reason without first reacting to something, nor react to something without first giving it meaning, and on and on. At many points before aiming a gun and pulling the trigger, particularly if the context is not unique, there are thoughts and emotions which others in similar situations also experienced.


Think of any situation that many have shared, say, getting to an airport late (but not too late) for a flight. Based on your experience, you can predict some of the thoughts, emotions, and thus behaviors of a harried traveler. Is he likely to stroll? At the ticket counter will he cordially allow others to get in line ahead of him? Will he savor the interesting architecture of the airport?

Because we are familiar with the airport situation, we find it easy to predict what the traveler will do. It is precisely because some people are not familiar with violent behavior that they feel they cannot predict it, yet they daily predict non-violent behavior and the process is identical. In his book Information Anxiety, Richard Saul Wurman explains that “we recognize all things by the existence of their opposite—day as distinguishable from night, failure from success, peace from war.” We could add “safety from hazard.”

When a woman is comfortable with a stranger in her home, someone delivering furniture, for example, her comfort communicates that she has already predicted that he is not dangerous to her. Her intuition asked and answered several questions in order to complete that prediction. It evaluated favorable and unfavorable aspects of his behavior. Since we are more familiar with favorable behaviors, if you list them and then simply note their opposites, you will be predicting dangerousness. We call this the “rule of opposites,” and it is a powerful predictive tool.

Does his job and no moreOffers to help on unrelated tasks
Respectful of privacyCurious, asks many questions
Stands at an appropriate distanceStands too close
Waits to be escortedWalks around the house freely
Keeps his comments to the job at handTries to get into discussions on other topics; makes personal comments
Mindful of the time; works quicklyNo concern about time; in no hurry to leave
Doesn’t care if others are homeWants to know if others are home
Doesn’t care if others are expectedWants to know if others are expected
Doesn’t pay undue attention to youStares at you

All types of behavioral predictions, not just those about danger, can be improved by applying the rule of opposites.

Just as we can predict behavior once we know the situation or context, we can also recognize the context by the behavior. A man insists on being first in the ticket line at the airport, looks frequently at his watch, appears exasperated by the slowness of the ticket agent. After getting his ticket, he runs awkwardly along, carrying his bags. He appears rushed and stressed. He looks expectantly at each gate he approaches. Is he:

a) A politician seeking votes who will stop and chat with each passer-by?

b) A charity volunteer who will solicit donations?

c) A person late for his flight who will proceed directly to the gate?

A hostile employee is fired the day he returns from a leave of absence. He refuses to vacate the building. He tells his supervisor, “You haven’t heard the last of me,” and then recites the supervisor’s home address. He says, “I’ll be visiting you with my buddies, Smith and Wesson.” Security guards are called to remove him and the following morning, his supervisor’s car windshield is smashed.

Is this fired employee likely to:

a) Send a check for repair of the windshield?


b) Enroll the next day in medical school?

c) Start making late-night hang-up calls to the supervisor’s home?

A couple of days after the man is fired, his supervisor finds a dead snake in his mailbox at home.

Was it placed there by:

a) A neighborhood prankster?

b) A member of the Snake Protective League trying to raise social consciousness?

c) The man fired a couple of days before?

I’ve used these obvious examples to demonstrate one of the greatest resources for predicting human behavior: You will rarely fail to place people in the most likely category when you frame the choice between contrasting options. This may seem obvious, but it is a powerful assessment tool.

A woman in an underground parking lot is approached by a stranger who offers help loading groceries into her car. She could refine her predictions about the man, and enjoy a creative exercise, by asking herself, Is this man:

a) A member of a citizen volunteer group whose mission is to patrol underground parking lots in search of women to help?


b) The owner of a supermarket chain looking for the star of his next national advertising campaign?

c) A guy who has some sexual interest in me?

By the time you consciously develop even the first possible category for your multiple choice list, you likely already know the correct answer and you have already considered the level of immediate hazard intuitively. Intuition, remember, knows more about the situation than we are consciously aware of. In the parking lot, it knows when the woman first saw that man, as opposed to when she first registered seeing him; it may know when he first saw her; it may know how many other people are around. It knows about the lighting, about how sound carries here, about her ability to escape or defend herself should she need to, and on and on.

Similarly, when assessing the fired employee, intuition knows how long he held on to resentments in the past. It remembers sinister statements he made that were followed by some unsolved vandalism. It recalls his disconcerting story about getting even with a neighbor.


The reason for creating three options is that it frees you from the need to be correct; you know that at least two of your options will be wrong, and this freedom from judgment clears a path to intuition. In practice, this turns out to be less an exercise in creativity than an exercise in discovery; what you may think you are making up, you are calling up. Many believe the process of creativity is one of assembling thoughts and concepts, but highly creative people will tell you that the idea, the song, the image, was inthem, and their task was to get it out, a process of discovery, not design.

This was said most artfully by Michelangelo when asked how he created his famous statue of David. He said “it is easy—you just chip away the stone that doesn’t look like David.”

Well, you can know it will be a man long before the statue is complete. It is an irony of prediction that if you just wait long enough to commit, every prediction will be accurate. By the eleventh hour, most factors are apparent, and they are less likely to change because there is less opportunity for intervening influences. The key is to complete a prediction far enough in advance to get some benefit, in other words, while you still have time to prepare or to influence outcome.

Why, after all, do we make any prediction? To avoid an outcome or exploit an outcome. To do either, prediction must be followed by preparation. Prediction without preparation is just curiosity. Predicting that Lucky Dancer will run fastest is only valuable when you have time to exploit the outcome by placing a bet at the racetrack. Conversely, if you are standing in the path of the galloping horse, you use the same prediction to avoid the outcome of being trampled, and you get out of the way.


The amount of preparation appropriate for a given outcome is determined by evaluating the importance of avoiding or exploiting it and the cost and effectiveness of the strategies you’ll use.

In deciding what preparations or precautions to apply, one also measures the perceived reliability of the prediction. If I predicted that you would be struck by lightning tomorrow and said that I could ensure your safety for $50,000, you wouldn’t be interested. Though it is very important to avoid being struck by lightning, the reliability of my prediction is low, and thus, the cost is much too high. If, however, a doctor says you need immediate heart-transplant surgery or you will die, the $50,000 cost is suddenly reasonable. The outcome of death is the same with lightning or heart failure, but we perceive the reliability of the medical prediction as much higher.


This same process of comparing reliability, importance, cost, and effectiveness (which my office calls the RICE evaluation) is how people go about making many daily decisions.

Society makes its precautionary decisions using the same RICE evaluation. Avoiding the assassination of a big-city mayor is less important to society than avoiding the assassination of a presidential candidate. That’s why we spend more in a week on a presidential candidate’s protection than we spend in a year for most mayors. We may consider our prediction that a presidential candidate might be shot to be more accurate than that a mayor might be shot, but it isn’t necessarily so. In fact, mayors have been shot more often and more lethally than presidential candidates. When we add governors to the mix, we find that nearly all have protective details, some quite extensive, but I am unaware of any governor ever killed in office. (Two governors have been attacked while in office, but neither because he was governor: George Wallace, because he was a presidential candidate, and John Connolly, who was in the car when President Kennedy was shot.) So, though a mayor is more likely to be killed as a result of holding the office, I am sorry to tell my mayor-readers that we care more about avoiding that outcome for governors.


With societies, as with individuals, when the RICE evaluation is made irrationally, it is always because of emotion. For example, avoiding the outcome of hijacking is so important around the world that passengers are screened for weapons more than one billion times each year. Only a few hundred people are actually arrested on weapons charges, almost none of whom had any intention of hijacking a plane. Ironically, the only deaths associated with airline hijacking in America have occurred since we instituted weapons screening. So, given the effectiveness, we pay a lot to do something that might prevent just a few incidents. We do this because of emotion, worry specifically. To be clear, I support screening of airline passengers, but this is as much for the fear-reducing benefits and deterrence as for the actual detecting of weapons.

The likelihood of burglary in your area may be remote, but the importance of avoiding it makes the cost of having locks on the door seem reasonable. Some people consider burglary to be more likely, or consider avoiding it more important, so they purchase security systems. Others don’t feel that way. Our approaches to caution and precaution all boil down to a personal RICE evaluation. Ask yourself about the reliability of the safety predictions you make in your life, the importance of avoiding a bad outcome, and the effectiveness of available precautions. With those answers, you can decide what precautions to apply to personal safety.

When you are at imminent risk, intuition forgets about all this logical thought, and just sends the fear signal. You are given the opportunity to react to a prediction that has already been completed by the time it comes into consciousness. These intuitive predictions are involuntary, but often we must make predictions consciously. How can those be improved? Ingmar Bergman said, “Imagine I throw a spear into the dark. That is my intuition. Then I have to send an expedition into the jungle to find the spear. That is my intellect.”

Simply by throwing the spear, we greatly improve conscious predictions. By nothing more than the act of inquiry, or even curiously wondering about what a person might do, we enter into a conscious alliance with our intuition, an alliance with the self. Logic and judgment may sometimes be reluctant to follow that spear into the jungle, but the concepts in the next chapter should help persuade them.




Which book you would like to read next? Comment Below.

Don't forget to share this post!


Popular posts from this blog

Wealth is What You Don't See

The art of staying young while growing old

‘Making People Glad To Do What You Want'