Denis Johnson - Tree of Smoke

The colonel, his chair cocked, his right hand on the railing, sat with the eastern half of Saigon spread out behind him and before him a long buffet, the Festival, apparently, of Hamburgers. His left hand gripped a cocktail. “The Agency is in a state of shock. The Kennedy thing and the Bay of Pigs business have left us quaking. We don’t know how to behave, how to carry out our mission. In Cuba we’re blundering around—we as an agency, and we as a nation. We’re the Russia of the Western hemisphere.”

Sands said, “And how do you see things working out for us here? At the moment?”

“It depends on the Vietnamese, Skip. We’ve been saying ‘It depends on the Vietnamese’ so long it sounds like bull, but it’s the truth. The question is, how do we help them? You, me, us, sitting at this table. I mean the three of us. I think we take a new approach. We’ve got to be more aggressive in handling the data.”

“Aggressive?”

“The three of us.”

“Us?”

“The question about intelligence-gathering is where do you stop taking the initiative? Do we get out there and beat the bushes aggressively, accumulate everything aggressively, and then passively leave it to others to sift? No. A sifting goes on continually, at every level.”

Jimmy: “A selection.”

“And I don’t like the goddamn selection, Skipper. What gets sifted out, among other things, is that one particular piece of information that’s going to make life unpleasant for us by troubling our superiors. And what’s left over is a lie that lands on the desk above, a happy lie, a monstrous lie.”

Jimmy: “A happy monster.”

“The lies go up, and what comes back down is poor policy, mistaken policy. Stupid ideas get generated out along the designated paths, and way out here, in the field, our limbs start jerking in a crazy way. Then when so ordered we file a report that says with care and deliberation we thrashed around causing havoc. You know how it works, Skip: Mindanao. We swing from being tepid and ineffectual to being ardent and silly.”

Jimmy: “Ardent—that’s a good word.” The colonel said, “Why should we wait for the silliness from the center of the hive? Why not generate our own scenarios?”

At this point Jimmy Storm took notice of a patron sitting down to another table, a rather tall young Asian woman, prepossessing, strikingly kempt, sheathed in a glamour of silk, and said, “I’d like to get into her groovy gravy.”

The colonel laughed. “HAH!” His jester picked a bit of meat from its sauce with his fingers and slurped it into his jaws. “Or maybe Skip wants to.”

… The article’s draft began with a handwritten note—the colonel’s block printing—photocopied:

WE DON’T HAVE AN INTRO YET

Want to revitalize the distinction between analysis and intelligence—clarity of thought, purity of language, correctness of speech, etc, clarity of fact—appreciating how a lack of clarity has led to the complete perversion of the intelligence function of our Agency. Its motives and its purpose. And its means. Its methods.

Let’s hit that as the main thing—the distinction between analy

sis and intelligence. Orwell—”Politics and the English Language” As far as intro— BASICALLY TO SAY HERE THAT WE’RE TALKING ABOUT TWO

FUNCTIONS OF THE CLANDESTINE SERVICES-INTELLIGENCE AND ANALYSIS. AND THE BREAKDOWN OF BARRIERS BETWEEN THE TWO ETC

On the next page began the typed material. Skip anticipated an embarrassing mess. By the third sentence he could see the colonel must have had assistance:

Cross-Contamination of the Two Functions

Our figures of speech with regard to the process of communication give us our model for this discussion. We speak of “lines” of communication and “chains” of command, reminding ourselves that data move in a linear and linked fashion through the ranks of those interpreting it. In the case of the functions of our intelligence services, we view this movement as originating in the field and terminating in archives, in plans, or in operations. Hard data collected by the officer in the field slows down as it trickles up the chain, and eventually finds itself stalled by considerations as to its impact—on other operations, on the goals of higher-ups, and even on the career-path of the person passing it along—until related data climbs up parallel structures to corroborate it, or—most unfortunately, perhaps dangerously—until command finds a need for it as justification for political policy, and those in possession of this data sense this need.

This hesitation and doubt is an indication that the intelligence function has been polluted by the analysis function. Data is being interpreted, albeit unconsciously, perhaps, and its effects on command anticipated. We speak of “command influence” on the intelligence function, and the fact that we possess a term for it acknowledges its existence; however, we have thus far failed to grapple with the operations, the mechanics, of command influence.

This paper suggests, in broad outline, that “command influence” operates through the cross-contamination of the two functions of the clandestine services: intelligence and analysis.

Cross-Contamination of the Two Categories

As data hesitates on the chain, awaiting (1) the accumulation of pressures to drive it upward and (2) the corroboration of related materials, the segregation of human intelligence from documentary intelligence is threatened and finally gives way. Simply put, the need to examine the veracity of sources yields to the pressures of process. The result is cross-contamination: data from human sources, notoriously undependable, become the support for doubtful interpretations of documentary sources, and these interpretations come to be seen as shedding light, in turn, on data from human sources.

The cross-contamination of these two categories, human intelligence and documentary intelligence, is a sub-process of the broader breakdown between the two functions of intelligence and analysis.

Cross-Contamination of the Two Waves

Meanwhile, the interpretive process, we remember, is always subject to appropriation and enlistment in the service of policy. Cross-contamination renders data vague, malleable, and eventually useless as anything but an ingredient of internal bureaucratic and political chemistries.

A detailed examination of the processes by which the needs of command are communicated downward along the chain must wait for another occasion. At this point let it be enough to acknowledge that a sense of the needs of command does travel downward through the chain in the same kind of wave action by which data are communicated upward. The result is cross-contamination of the two waves.

It is to be stressed that this process is of an entirely different nature than the intelligence-gathering process of our Agency in its earliest incarnation, the Office of Strategic Services (OSS). There the function of Intelligence remained almost untouched by policy, because policy is the game of peace, whereas the OSS served a command structure pursuing the objectives of war. From that era we have allowed to survive the old model of field-to-archive, field-to-plan, field-to-plan-to-operation. However, that model no longer serves us well.

The model of a chain on which two waves of data under pressure cross-contaminate one another is truer to the actual processes of our Agency today. The downward pressure derives from the needs of command, while the upward pressure derives from the need to satisfy command.

At this point in the discussion let us again acknowledge the process’s lack of utility, as we have now illuminated the category of service in which intelligence becomes useful, that is, in the pursuit of the objectives of war.

Cross-Fertilization of the Two Goals