APPROVED FOR RELEASE
CIA HISTORICAL REVIEW PROGRAM
2 JULY 96
Some rules, ploys, and plays
BASIC PSYCHOLOGY FOR INTELLIGENCE ANALYSTS
Charles D. Cremeans
When Allen Dulles chose to have the words "For ye shall know the truth and the truth shall make you free," carved in white marble at the entrance to the Headquarters building he was giving expression to an article of faith in the intelligence profession. We must believe that knowledge of the truth sustains and supports our government or we couldn't justify what we are doing.
Working intelligence officers know, however, that it isn't always as easy as it sounds. "What is the truth? How much evidence do you have to have? how selected? how organized? how presented? how evaluated before we have the truth that will make our country free?—and free from what? We all know that good and true men disagree on these matters, as on the evidence on any given subject of intelligence concern. We also know that from time to time, every intelligence officer worth his salt wakes up with a shock to realize that he has been misreading the evidence on some familiar topic. This can happen because he has gone along with the common wisdom, accepted unexamined assumptions, or just plain gotten into a rut. It can also happen if preoccupation with success, or mere survival in the intelligence culture become more important than intelligence itself.
The sensitive intelligence officer becomes aware from time to time of the effect on our finished product of the interaction of personalities and institutions within the intelligence community. We are, after all, human beings; we have deadlines to meet; we tend to favor our own conclusions over those of others; and we all know that a little salesmanship here and there, a little blarney, a measure of cajolery, and some basic psychology can often get a paper agreed to and on its way to the White House, while without such inputs it might languish and spoil under the heavy hands of some well meaning but less subtle colleagues.
The object of this paper is to look at some of the ways in which we get our work done, ways that depend more on human psychology than on cold reason. The purpose in mind is not to collect a bag of tricks, a primer of intelligencemanship, but to focus a spotlight on one aspect of our craft which is usually ignored. The purpose in doing this is not to suggest that an end be put to this kind of thing. God forbid that we stop being human, that we coldly reject, as being unsuited to our profession, such phenomena as the well-known eloquence of the distinguished dean of photointerpreters. But we should be aware of ourselves as we really are and not be misled into thinking all our peccadilloes foster the rapid and certain discovery of the truth.
We may start with a look at some of the oft quoted laws of intelligence.1 The most famous of these is Platt's Law, which reads, as set down by its discoverer: "Whether or not the necessary explanatory details and pet phrases of an intelligence paper appear in the paper as finally published, depends entirely upon whether the number of higher groups which successively review the paper is even or odd respectively."2 In the Office of National Estimates this is sometimes rendered: "If the Staff writes it long, the Board wants it short—and vice versa."
Another famous principle is that of Excessive Approval. Every intelligence Indian—i.e., drafter—knows that when the review board or panel, or whatever the higher echelon is, responds to a request for comments with unstinted praise, there comes a point at which the drafter feels a sense of foreboding. It usually means that his paper is about to be torn to shreds.
All veterans of intelligence coordination are familiar with the law of Emphasis by Place. This law is often referred to in this manner: "I suggest that the item referred to at the end of the paragraph—or section, or paper—be brought up to the beginning in order to give it greater emphasis." It is equally often cited by urging that an item that appears at the beginning be put at the end "in order to give it greater emphasis." Adjudication on this matter usually depends on whether the Chairman wants to argue about whether emphasis is bestowed by early or late reference or whether he thinks the time is suitable for a throwaway concession in the hope that the gesture can be collected on at a later time.
Most notorious of the laws of intelligence is Murphy's Law: "When something can be misunderstood, it will be." The archives contain no record of Murphy. He may have been an honorable and well-intentioned man, but, sad to say, his law is more often than not cited by someone whose opinion of his boss is that he can and will read only one sentence at a time. The result of this assumption is that all the supporting calculations and data must be stuffed into the sentence in question, making it incomprehensible by the most intelligent reader, and probably to the boss for whose benefit the re-writing is being proposed.
A quick look at these laws of intelligence shows that they really are techniques of persuasion rather than laws the knowledge of which enables one to understand the behavior of phenomena in the real world. In fact it is in the realm of persuasion—of others, as well, sometimes, as of ourselves—that psychology most often obtrudes into intelligence.
Almost every intelligence analyst learns that if he wants to play it safe, or if he just doesn't know what is going to happen, an easy way out may be found through the Continuation of Present Trends formula. Unless he runs into really bad luck, an intelligence analyst of modest competence can usually go through a career with good marks simply by summarizing the evidence, and then pronouncing thus: "present trends are likely to continue." When this gets boring or too conspicuous, the More and More formula is often called into use. "King Hussein will find it more and more difficult to maintain control ...," or he "will find it increasingly difficult. ..." This gets to be a problem when he has been finding it more and more, as well as increasingly difficult for years and years and still hangs on. Then it becomes increasingly difficult for the analyst. The point is not that he should be ashamed of himself for being unable to find an answer to King Hussein's future in all that mass of paper that flows across his desk but that it should be quite clear to himself and to his readers that the evidence doesn't provide the basis for much of a judgment—which, of course, he should go on looking for despite the inadequacies of information and insight.
Perhaps the fundamental relationship among intelligence officers is that between the expert and the nonexpert. The former, of course, being the person who is supposed to know—although he doesn't necessarily really know all about Patagonia just because he is on the desk—and the latter being the person who reviews, edits, revises, or just approves his work. We are talking, of course, about the Indian and the chief in the intelligence analysis tribal culture. In real life the expert is usually comparatively young and the nonexpert or supervisor, comparatively old. The supervisor was probably an expert once but has to cover too wide a field, has too much administrative responsibility, or is too tired to be anything but a "generalist."
Actually, both the specialist and the supervisor have essential jobs to do, but the relationship is inherently a difficult one and, as a consequence, the ingenuity of man (real "intelligence officers") rises to the challenge with formulae that make life easier—sometimes for one, sometimes for both parties. For the expert the neatest solution is to know so much, to calculate so well the requirements and the quirks of the supervisor as well as the supervisor's supervisor, and to translate this into such a good end-product that the boss can only sigh and sign off. Unfortunately, not every supervisor knows when he is getting a perfect draft, and so even the best of the experts resort to certain stratagems to make their lives tolerable.
One approach, very often overdone, is that of laying on the expertise with a trowel: "Well, you know sir, unless you have lived with the Khmers as I have it is quite impossible to understand their reaction to the current situation." Another frequently used ploy is that of drowning your opponent, or boss, with facts. One famous "expert," who did know as much about the Arabs as anyone in town, insisted on going into the fine points of tribal differences, whatever the issue at hand, until in the end he had only to open his mouth to provoke groans and numerous visits to the washroom. In the first case by taking the line that only experts can understand, and in the second, by becoming irrelevant, the expert weakens his position and, indeed emphasizes the need for the intervention of a nonexpert, preferably one with good sense and judgment.
For his part the nonexpert (or no longer expert) supervisor can fall into equally dangerous traps if he tries too hard to compensate for his inadequacies. One of the most common dodges of the one-upped supervisor is the counterexpertise play: "Well, I don't know anything about the Khmers; I'll be the first to acknowledge it; but I remember a situation very like this Southeast Asia thing we are discussing which took place some time back when I was in Central America, and I can tell you. ..."
The old timing game, of course, is played by both sides in this contest of generations. How often has the drafter of a paper come rushing into a senior's office, saying, breathlessly, "Hope you can read this right away, sir. I spent all weekend on it and it's got to go to the DD this afternoon. Incidentally, the girls have started typing, so I hope you won't have too many suggestions." Of course, there have been a few times when a supervisor has stopped a staff man in the hall, saying, "By the way, I had lunch with the DD and he asked about that paper you gave me to look at. I thought I had better give it to him right away. Sorry I didn't have time to consult with you about it, particularly as I rewrote the last section and put it at the beginning."
Well, we're not all perfect, and this sort of thing goes on partly because in many cases things would not get done if it didn't. The point here seems to be that the better a man the expert is and the better a man the supervisor is, the less the need for stratagems. So, if you're an expert, get a good supervisor, and if you're a supervisor. ...
There are, of course, a good many pitfalls that specialists and nonspecialists together can get into. One of the worst, both from the point of view of the people involved and of the whole intelligence community, is a syndrome best represented by the famed "numbers game" The problem usually arises when there is a strongly felt need on the part of the top users of intelligence for a degree of precision which the evidence, or, indeed, often the subject, does not permit. When the top policy makers ask, for example, "How many Russians are there in Cuba anyway? Just give us your best guess." The people down the line ought to be very cautious, we all know now, about giving them a figure at all unless there is a certain minimum evidentiary basis for it. The consequence, of course, can be finding ourselves unable to change figures even when our intelligence improves, because of the difficulty of explaining how we got the original figures on the books anyway. All this adds up to one of the most important rules for the intelligence officer: Don't fool yourself into thinking that if higher authority demands it, it makes sense to put out something that is basically unsound.
The intelligence officer's working life is not spent only at his desk or in consultation with his supervisor. There is the group: the meeting, the committee, the task force, the discussion, the debriefing—all standard situations in the intelligence culture. These intelligence group's experiences might not seem to some to be as dramatic as what we are told goes on at the Esalen Institute in California where people grope, in the company of others, for self-understanding, but they can be pretty real and earnest. They probably produce as much self-realization and as much bloodshed as similar competitive situations anywhere. On any good workday one will find as wide a variety of successful personal styles on display in intelligence groups, as in a Madison Avenue idea session, in a back boardroom, or an academic committee.
Every experienced participant in group intelligence knows the country boy who talks of the inner mysteries of Soviet space technology with just enough of a southern drawl to add a human touch. There is the blustering Devil's advocate who specializes in outlandish and unanswerable propositions. There is the man with a cause who specializes in stripping the flesh off the proponents of a rival school of analysis. There is the specialist in the scathing personal attack at the right moment. (My favorite, and one done in good humor, is an instance where criticism of a sentence in a draft paper was conceded by the author to have been "ambivalent." "Sir," said the critic, "You do yourself too much credit. An ambivalent sentence has two meanings. Yours has none at all.")
Along with the bad guys, and the bores, the sycophants, and the fools intelligence officers may be carefully screened, but no foolproof battery of tests has yet been devised—there are, naturally a fair proportion of good guys of all sorts. Here, as elsewhere, the observer of the intelligence culture must conclude that the fact that intelligence people are people is all to the good, as well as being unavoidable. Furthermore, it does not obscure or change the fact that, whatever the style, the ability to produce sound intelligence is the payoff in the end.
There is still another situation in which intelligence officers interact and which gives rise to its share of specialized behavior patterns. I refer to the joys of coordination, particularly of that highest form of agony known as interagency coordination. Getting things done within an agency, as amply suggested above, is complex enough, but in an interagency situation, where the boss can't resolve the disputes, a very specialized form of interaction takes place. How it all works, I shan't pretend that I understand, though there are a few clues. When the Navy representative says, "Can't you please mention submarines in the section?" his colleagues are inclined to go along if it won't mess up the paper too much and if he can be expected to be agreeable when their turn comes. Perhaps the most time-honored and symbolic device of the interagency coordination process is the convention of bestowing "The Order of the Lion" on a representative who has done his duty and manfully presented his superior's case to an unreceptive audience. (The idea is that he can go back home and tell the boss that he fought like a lion but that the other agencies wouldn't have it.)
Nowhere else is the art of the tradeoff so highly developed. Nowhere else is such skill applied to the artful suggestion of a different form of words to say what is already in the text in order to save the face of a colleague who can neither withdraw nor make his proposal specific. The worst burden for coordinators is the colleague who insists that he, or his boss, doesn't like something but doesn't know why or what he wants to do about it. The greatest problem, of course, is the intervention of departmental interest, or policy commitment, into the discussion of an intelligence judgment. Most representatives realize that this is a high crime—or at least that it stultifies the process—but all tend to be sympathetic with the colleague who they know has to go back to a boss who doesn't know or care about the distinction between intelligence and policy. The miracle is that interagency coordination of intelligence works as well as it does, that the people who do it get along, and that the end product is almost always sound intelligence.
What is there to conclude from all this? That we are people, like other people, and that our personalities, our instinctive drives, and our subconscious minds get deeply involved in the process of "knowing the truth"? I believe so, and I believe it is essential that we acknowledge and take account of this while doing our best to create as much as possible of that marvelous stuff, objective intelligence, which is what Allen Dulles probably had in mind when he selected that quote from the Bible.
[Top of page]
1 We do not propose to set down all the "laws of intelligence," but only those commonly cited or applied in the production of intelligence. Kent's Law, for instance of any coup d'etat I have heard of isn't going to happen"—is a profound truth but not within the scope of this paper.
2 Studies Vol. 13, No. 4, pp. 89-90.
[Top of page]