THE PHOENIX PROGRAM

CHAPTER 10: Action Programs

Before he bade adieu to Vietnam in November 1967, Nelson Brickham helped put together what was entitled "Action Program for Attack on VC Infrastructure 1967-1968." Signed by the CORDS assistant chief of staff, Wade Lathram, "Action Program" represented Robert Komer's administrative and operational directives for the ICEX program. It is the most significant Phoenix document, charting the program's dimensions and course over its first eighteen months. It set in place Brickham's reporting requirements, established tables of organization, identified major problems, and formed groups to find solutions.

"Action Program" consisted of twelve separate tabs, each addressing a separate mission or function to be accomplished by a specific deadline. First on the list, Tab 1, called for promulgating the ICEX mission directive, MACV 381-41. Tab 2 called for briefing all corps senior advisers, and Tab 3 directed the CIA region officers to designate corps and province ICEX coordinators, all by July 31, 1967. By year's end ICEX committees were operating in thirty-nine provinces, thirty-four of which were chaired by CIA officers. Most were meeting monthly and had initiated anti-VCI operations. Also by year's end twenty-nine Province Intelligence Operations Coordination Centers (the province equivalents of a DIOCC) were functioning and sending reports to the ICEX Directorate. In certain provinces, such as Vinh Long in the Delta, the PIOCC doubled as a Phoenix committee.

Tab 4 called for continuation and expansion of DIOCC development. At the time "Action Program" was issued, 10 DIOCCs were in operation; by year's end there were 103, although most were gathering tactical military intelligence, not infiltrating and attacking the VCI. In November 1967 more than half a million dollars were authorized for DIOCC construction, salaries of Vietnamese employees, office equipment and supplies, and transportation. "These were not operational funds in the sense of supporting anti-infrastructure activities." [1] Money for anti-VCI operations came from the parent agency.

To his credit, Evan Parker did not approve of the rapid pace at which Phoenix was expanding. "I didn't think we needed an elaborate structure everywhere in the country," he told me. "Some of the provinces didn't have enough people or activity in them to warrant it. I would have preferred to concentrate on the more populated active areas where you knew that you had people to work with and something to work against." [2]

There were too many variables, Parker contended, to have "a uniform program." The methodology had not been perfected, and too much depended "on the personal likes and dislikes of the senior Vietnamese people in the field ... and their adviser .... For instance, in I Corps there was a lot of activity, not so much concerned with the VCI as with the machinations of rival political parties -- the Buddhists or whatever .... These are things that were hung over from the French days .... This was always the problem with Thieu .... [it] was sort of open season on the enemy -- of settling scores."

Tab 5 of "Action Program" prescribed ICEX staff organization along the committee lines proposed by Brickham. In Saigon the ICEX board of directors consisted of the DEPCORDS as chairman, the CIA station chief, the MACV intelligence (12) and operations (13) chiefs, and the CIA chief of Revolutionary Development. In fact, the board met only once, and Robert Komer quickly assumed control of Phoenix, setting policy as he saw fit, with the directorate serving as his personal staff. "Komer or Colby [who replaced Komer as DEPCORDS in November 1968] said, 'You'll do it.' My job," explained Parker, "was to say, 'Okay, Colby says you'll do this, and this is how you're gonna go about doing it.' What I did was help people carry out what they were ordered to do. And I firmly believe in the soft sell."

In practice, Parker's CIA kinship with Komer and especially Colby enabled him to manage the Phoenix Directorate without having to consult agency heads. He had merely to state his wishes to the DEPCORDS in order to bypass the various chains of command.

"Colby was my division chief in the field, and in Washington also," Parker explained. "I served with him in World War Two when I was in England. I met him when we were both in a program known as the Jedburghs. He went into the field in Europe, and I went into the field in the Far East.

"Colby is a fine gentleman, I'll tell you. He was tremendously helpful to me. So was Komer. But their personalities were very different. Komer was essentially a rasping, grating sort of voice ... but he was consistently staunch in his support of the program .... He may have given orders, he may have been sarcastic -- all those things -- but at the same time he was not one to stand on ceremony, not one to do things because that's the way it's always been done. He didn't give a damn about that. He'd say, 'I want Parker's organization to get four trucks! I don't give a good goddamn where they come from, just give him four trucks!'

"Colby was quieter, more soft-spoken, but just as firm in terms of getting things done.... He would suddenly say, 'Let's go visit so-and-so,' in a province or region. That meant you would call up and get a helicopter or a plane, with no notice, and he would just go there and see them. That made it a whole lot more secure because we traveled without bodyguards."

Case in point: While serving as Phoenix coordinator in Quang Tri Province, Warren Milberg was visited by Colby, who was on an inspection tour. As Milberg recalled it, Colby decided to spend the night, so Milberg assigned a Nung guard to watch over him. That night there was a mortar attack. The Nung guard grabbed Colby by the scruff of the neck, dragged him backward down the stairs (Milberg arrived in time to see Colby's heels bouncing on the steps) into the basement of the building, threw him on a cot, and threw himself on top of the future director of Central Intelligence. Somewhat dismayed at the treatment the Nung had afforded the DEPCORDS, Milberg half expected the ax to fall when Colby and his entourage assembled for breakfast the following morning. But Colby merely thanked the earnest Nung for the gesture of concern.

The consummate insider, Colby would win many friends with his "just folks" management style, while using his considerable influence to refine and redirect the broad policies put in place by Komer -- the outside agitator who rode roughshod over everyone. Together, Komer and Colby were the perfect one-two combination required to jump-start Phoenix and keep it running for five years.

As of August 15, 1967, Parker's part-time staff had been replaced by three permanent CIA officers: Joe Sartiano as executive director; William Law as chief of operations; and James Brogdon as administrative officer; Colonel William J. Greenwalt had replaced Junichi Buhto as deputy director, and six MACV officers were assigned as full- time employees, along with a smattering of AID and State Department people.

"We set up a working organization built around agency people," Parker said, "with other individuals made available from the different agencies, but still paid for by the agencies they belonged to." By then there were American women serving as secretaries, MACV and CIA officers advising the Vietnamese, and others in the office keeping records. "There were probably three or four people I counted on more than anyone else," Parker remarked, but "in order to make this work, I would say that the core people were the agency people in charge of the special police -- the senior agency advisers."

***

Tab 6 provided for military augmentation of ICEX field units. As Parker put it, "Then you realize you're going to have a nationwide organization as well as a headquarters staff, and that you're going to need a lot more people than you envisioned. So the Army becomes the principal.

"In due course a table of organization was set up which assigned people to region, then to province, and most of them were Army. You'd have a captain at province and a major or [lieutenant colonel] at region with assistants -- corporals and sergeants and so forth. MACV took the bodies at first as they came in-country and assigned them regardless of the fact that they may have been intended for something else. For example, my deputy was going to a military unit but found himself in ICEX instead. Another fellow who was going to be assigned to MACV counterintelligence instead was assigned to an intelligence function in ICEX. That's where the first people came from."

The first MACV allotment to Phoenix was for 126 military officers and noncommissioned officers (NCOs), all counterintelligence specialists. One officer, one NCO, and one clerk-typist had been sent to each corps by September 15, and one officer and/or NCO to each province. By the end of 1967 one NCO had been assigned to each of the 103 DIOCCs then in existence. All military officers and enlisted men assigned to the Phoenix program in 1967 took orders from the CIA.

Tab 7 provided for briefing and coordination with senior GVN officials. While the groundwork was being laid on the American side of the program, Parker said, "we were working with the Vietnamese to sell them the idea. Although they were militarily assisting, the Vietnamese police had the major role because after all, you're dealing primarily with civilians. So the person who worked most closely with us was the director general of the National Police."

But General Nguyen Ngoc Loan was wary of the CIA, which was supporting Nguyen Van Thieu -- not Nguyen Cao Ky -- in the campaign leading up to the October 1967 presidential elections. And even though Ky was persuaded to run as Thieu's vice-president (they joined forces against "peace" candidate Tran Van Dzu), the two were bitter enemies. As Ky's enforcer General Loan opposed Phoenix not only because it infringed on Vietnamese sovereignty but because he believed it was being used to promote Thieu. Their opposition to Phoenix was to spell trouble for General Loan and his patron, Ky.

General Loan's opposition to Phoenix, however, did not mean that he refused to work with Americans on an equal basis. His support for CT IV disproves that. And Cong Tac IV "was a program that was doing well, too," said Tully Acampora, "until February 1967. Then Robert Komer arrived, grabbed the political implications, and, after returning to Washington and conferring with his boss, Walt Rostow, purloined it from the Vietnamese." [3]

CT IV differed, fundamentally, from Phoenix in that the U.S. military units it employed were not empowered to arrest Vietnamese civilians. Phoenix, on the other hand, relied primarily on the PRU, which operated under the exclusive jurisdiction of the CIA and thus were beyond General Loan's control. General Loan naturally preferred to work with General McChristian's Combined Intelligence Staff. But when McChristian left Vietnam in July 1967, Komer immediately exploited the situation. At Komer's direction, MACV officers assigned to CT IV were gradually withdrawn by McChristian's replacement, General Phillip Davidson, whom Tully Acampora described as "beholden" to Komer for his job.

"Komer was disastrous," Acampora stressed. "He more than anyone politicized MACV. He was forcing for a treaty, promoting Phoenix and promising Westmoreland the job of Army chief of staff, if he went along. In mid-1967 it was a completely political situation."

Indeed, by deducting more than a hundred thousand Self-Defense Forces and "political cadre" from the enemy order of battle, Westmoreland, Komer, and Hart were able to show success and in the process convince President Johnson that "the light" really was at the end of the tunnel. Meanwhile, having backed themselves into a corner, they decided to do the job themselves. So what if General Loan was resistant? As Nelson Brickham had said, "That's okay 'cause we're gonna do it anyway!"

Symbolizing this "get tough" policy was Phoenix, rising from the devastation of two years of a stalemated war. Phoenix in this hawkish manifestation represented the final solution to the problem of distinguishing between a covert Communist enemy and an inscrutable ally. Uninhibited by family ties, Americans in charge of irregular forces, or by themselves, began hunting the VCI in its villages, doing what the Vietnamese were reluctant to do -- even though they were never quite sure of whom they were stalking.

This desperate policy was not without its American detractors. Tempestuous Tully Acampora called it "detrimental and contradictory." Ed Brady, the Army captain assigned to the Phoenix Directorate as a cover for his CIA activities, concurs. "It's very hard to carry out secret covert operations and repressive kinds of things in order to separate guerrillas from people -- and then make a speech to them about how their individual rights are so important," Brady said in an interview with Al Santoli. [4]

But while Acampora and Brady believed the United States had no business preempting the Vietnamese when it came to the attack against the VCI, other Americans thought that the time for patience and cooperation had come and gone. From Evan Parker's perspective, the problem was competition between the Special Branch and the ARVN. "It involved one Vietnamese agency saying, 'Well, we can't give [information] to them, because they're penetrated by the VC.' That sort of thing. And in some cases undoubted it was true."

Parker raised a legitimate point. In order for an intelligence coordination  and exploitation program like Phoenix to work, institutional mistrust between the police and the military had to be overcome. But, Parker explained, "Having the Special Branch have such an active role made it difficult in many provinces and many of the more rural areas, because the special policeman was probably the equivalent of a sergeant. So ... he doesn't have much clout .... And the [outgunned, outmanned] police are pretty subordinate to the military, so you have all this business of army versus police. It's a wonder it worked at all."

Moreover, frustration with Vietnamese security leaks gave Americans yet another reason not to wait for the Vietnamese to throw their support behind Phoenix. As Evan Parker said, "One of the great problems with the Vietnamese in getting this started was that the classification of the directive was so high -- in order to prevent it from falling into enemy hands -- that it was very difficult to handle these documents in the field ... and tell people what they were supposed to do."

Typically, Tully Acampora refuted Parker's explanation and interpreted the emphasis on secrecy in political terms. According to Acampora, for whom the switch from CT IV to Phoenix meant a loss in status, Parker "always envisioned Phoenix as a wholly U.S.-promoted, -managed, and -supported program." Moreover, "Hart's one mission was to undermine Loan's influence, to reduce his power base, and to superimpose Phoenix on CT Four. They bought off the head of Special Branch, Major Nguyen Tien. Then Parker started suborning guys on the MACV intelligence staff. He seduced Colonel Junichi Buhto [MACV's chief of counterintelligence] by promising to make him a GS-nineteen if he went along with the CIA .... Davidson's mission was to destroy CT Four, and in August, Davidson and the CIA began withdrawing Americans from the Combined Intelligence Staff. This involves the election of 1967."

There is no doubt that Phoenix, in its fledgling stage, was conceived and implemented by the CIA. Furthermore, Ralph Johnson writes, "The results obtained by ICEX by the end of 1967 were primarily, if not totally, stimulated and supported by the Americans." [5] There was early acceptance of Phoenix by the Vietnamese in I Corps, but as Parker himself noted, much of that activity was directed against Thieu's non-Communist political opponents. Otherwise, the majority of Vietnamese hesitated to embrace a program as politically explosive as Phoenix. As Johnson observes, "most province chiefs were waiting for instructions from the Central Government." [6]

The first step in that direction was taken in late December 1967, two months after Thieu had been elected president and Ky had begun to lose influence. On December 20, 1967, Prime Minister Nguyen Van Loc signed Directive 89-Th. T/VP/M, legalizing Phung Hoang, the Vietnamese clone of Phoenix. However, the directive was not signed by President Thieu and thus carried little weight with cautious province chiefs hedging their bets while Thieu established himself more solidly.

It is also important to note that Prime Minister Loc's reasons for authorizing Phung Hoang were directly related to Robert Komer's attempt to undermine General Loan and Nguyen Cao Ky by ending support for CT IV. After December 1, 1967, when Komer managed to terminate Operation Fairfax, Loc had no choice but to support Phoenix. And, according to Tully Acampora, by withdrawing the U.S. units that shielded CT IV's Field Police, "Komer opened up all the avenues which led to Tet." Making matters worse, in an attempt to stimulate the South Vietnamese economy and, in the process, allow Thieu to reap the political rewards, Komer went so far as to remove police roadblocks and checkpoints around Saigon.

Meanwhile, Tully Acampora was pleading with as many American generals as he could find, asking them not to withdraw American forces from CT IV. "Loan was saying that there was a massive influx of VC into Saigon," Acampora recalled, "but Komer was calling it light, and Hart backed him. They wouldn't listen to Loan, who was trying to convince them for sixty days prior to Tet."

Nelson Brickham, for one, admitted to having been fooled. "The VC had pulled their good people out and sent them up North in 1966. We knew that. Then, in the summer and fall of 1967, they came back. But I misinterpreted it. In October 1967 I told Colby that we were in a position that no NVA or VC unit could move without us knowing it. We saw Loan's warnings as crying wolf." [7]

"We were picking up massive numbers of infiltrators," Acampora told me, "so Loan countermanded the Joint General Staff's orders to withdraw; he refused to pull out all of his people. He kept a paratroop unit and a marine unit in Saigon and canceled all police leaves. Those units, with the police, met the first assault in Tet. Then, of course, Loan was resurrected." But by then it was too late. In Acampora's judgment, Komer's machinations brought about Tet. "The fact is," he said, "that Parker contributed to that disaster, too. Parker said Phoenix was the only impediment, that it turned defeat into victory. But the embassy was attacked! How could that happen? The fact is, Phoenix was a failure, and it was only because of Loan that the VC suffered a setback."

"In any event, the prime minister said, 'Do it.' He gave the order," Evan Parker said, "and he wrote the letters to empower them to do it, and Phung Hoang came into being on the Vietnamese side .... A Phung Hoang staff was set up by the Vietnamese consisting primarily of people from Special Branch. Then they set up quarters for them " at the National Police Interrogation Center. "The two organizations had separate quarters," Parker added, "because we wanted the Vietnamese to feel that Phoenix was a Vietnamese program and that the Americans were simply advisers."

"So anyway" -- Parker sighed -- "we went through this organizational phase. The Vietnamese went through the same thing, pulling together the police and whatever, trying to set up staffs, finding places for them to sit, providing them with pencils and paper, and trying to get them to actually conduct some sort of operations. And here you come to the nitty-gritty."

***

Tab 8 of "Action Program" called for review of VCI intelligence collection requirements and programs, especially Project Corral, a unilateral American operation started in October 1966 solely to collect information on the VCI at province level. After completing their review, CIA officers on the Phoenix staff began to prepare a standard briefing on the VCI for incoming officers and interested officials. They also began compiling handbooks, interrogation guides, and "related materials" like most wanted lists.

Especially effective against the VCI, most wanted lists had been used for years by Special Forces when, in April 1967, Renz Hoeksema's deputy, Robert Brewer, initiated a Most Wanted program in Saigon and expanded it nationwide. "Every province was directed to examine its files for a list of ten," [8] Brewer explained noting that the object of the exercise was to show that the enemy was not "faceless." Soon most wanted "posters," replete with composite drawings (prepared by Special Branch officers using New York City Police Department makeup kits, of VCI suspects were being nailed to trees, DIOCC walls, and market stalls throughout Vietnam. The posters offered cash rewards and had a picture of the phoenix to catch people's attention. (See enclosure.)

In the spring of 1967 Komer appointed Brewer as senior adviser in Quang Tri Province. "When I got there, I got all the intelligence-gathering outfits together," Brewer recalled, "and we wrote up a list of the twenty-one most wanted VCI. One guy on my list, Bui Tu, had killed a district adviser's sergeant, and I wanted to get him. So I went to the high school and found his picture in the yearbook. That really paid off. On a sleepy afternoon in July the word came in from Special Branch that Bui Tu was in the area. The DIOCC notified district, district notified village, and the Marine combined action patrol went after him.

"Bui Tu had been spotted in a shelter on a rice paddy. Three guys jumped up and ran, and the Popular Force team and the Marines mowed them down. Bui Tu was number one. The top. He had captain's bars and a briefcase full of notes, with a quarter inch of papers on me! They knew where I slept in the compound and they were planning to kill me." Thanks to Bui Tu's documents and information provided by the defector, Brewer said, "We blew the VCI apart."

What Brewer described is a typical Phoenix operation: A most wanted poster led to a high-ranking VCI suspect's being spotted and killed, while his captured documents revealed the whereabouts and identities of many of his VCI comrades. Most wanted posters also served to inhibit the VCI. As Jim Ward explained to me, "All of a sudden this guy who used to travel from place to place begins to wonder who is going to turn him in! It begins to prey on him. We found out later that this really had a significant psychological impact on these guys, making them hide and becoming less effective." Said Ward: "It suppresses them." [9]

By the end of 1967 thirty-five provinces were compiling blacklists of VCI members, and twenty-two more had most wanted lists. [10]

Tab 9 of "Action Program" called for review and recommendations for action programs to exploit infrastructure intelligence. In theory this meant the training, direction, and coordination, by U.S. personnel, of Field Police and PRU in anti-VCI operations. Between the two, the PRU were more effective, accounting for 98 percent of all anti-VCI operations in I Corps alone. In November 1967, Ralph Johnson writes, "II Corps and III Corps reported that 236 significant VCI were eliminated by the PRU, which continued as the main action arm of the 'rifle shot' approach." [11]

"Basically the PRU were effective," Parker stated. "In some cases the police were effective. And in many areas more got done in capturing VCI in military operations. But I was interested in getting key people. You can arrest the little ones, but the operation goes on and on, and you haven't really hurt them. But it's very hard to get a really important man.

"I personally wasn't involved in any operations," Parker stressed. "Operational control was exercised at whatever level it was happening at, by the so-called action agencies. The idea was to use resources wherever they were .... If there needed to be cooperation, the Vietnamese would consult ... if they trusted the head of the other agency. Unfortunately the Americans would conduct operations without telling the Vietnamese. And vice versa."

By the end of 1967 the Field Police were conducting anti-VCI operations in twenty-six provinces; thirty-nine provinces were using systems taught by Phoenix staffers on how properly to "debrief" defectors, who were used as spotters, PRU, and interrogators. Included in the Phoenix arsenal were joint military-police search and destroy and cordon and search operations, population and resources control, and riverine and maritime operations.

Tab 10 charged the Phoenix program with improving the civilian detention system. About this subject Nelson Brickham remarked, "The one major element left out of all this was the civilian detainee problem. It starts with the Province Interrogation Centers, but the larger problem is, How do you screen detainees, and then what do you do with identified VCI?

"When you'd go through these village sweeps, you'd have whole corrals filled full with Vietnamese just sitting there looking at you all day long. In rural provinces you'd wind up with barbed-wire cages with tin roofs packed with people. It was a major problem basically because we were running a revolving-door operation. We'd capture VC; then a week later we'd capture them again ...assuming they were VC. The Vietcong always knew about these sweeps several days beforehand and always pulled out before we hit. In a lot of sweeps all you would get were the old men and women and kids. There were VC in there, too ... but nobody knows really who they are.

"There were legal questions. Do we reindoctrinate them? Do we shoot them? Do we put them back on the farm? It was just out of control. So one of John Hart's tasks on the original ICEX charge was, What to do with these civilian detainees? Do they have prisoner of war status? Remember, there's no war going on! But in Geneva Americans were saying, 'We're treating these people like POWs.' The Swiss were saying, 'Okay. We want a look into the prison system.' So Hart became concerned with the problem, and the reason it shows up in the ICEX proposal is at John Hart's insistence.

"It went 'round and 'round, and the long and short of it was, nobody wanted to get the name of the Jailer of Vietnam attached to them. USAID didn't want to touch the problem with a ten-foot pole .... Same with the military. Their attitude was 'He's a POW. Forget him. When the war's over, we'll ship him back to the farm.' And so one of our tasks was to investigate the problem and recommend a solution to it. But we never did. What we did was to beg the question. We tasked the problem over to the new plans and programs element of the ICEX staff. What they did, I don't know."

What the ICEX staff did was state the problem. As listed in Tab 10, the major issues were: (1) overcrowding, substandard living conditions, and indiscriminate crowding of POWs, common criminals, VC suspects, and innocent bystanders in ramshackle detention facilities; (2) lack of an adequate screening mechanism to determine who should be interrogated, jailed, or released; and (3) a judicial system (lacking due process, habeas corpus, arrest warrants, and lawyers, that might delay someone's trial for two years while he languished in a detention camp or else might release him if he could afford the bribe.

In seeking solutions to these problems, Tab 10 proposed: (1) the construction of permanent detention facilities; (2) a registration system, coordinated with refugee and Chieu Hoi programs, to eliminate the revolving-door syndrome; and (3) judicial reform aimed at the rapid disposal of pending cases, as devised by Robert Harper, a lawyer on contract to the CIA. In addition, a study team from the CORDS Research and Analysis Division (where Phoenix operational results were sent along with a weekly summary of significant activities, conducted "a comprehensive and definitive study of all aspects of the problems of judicial handling and detention of civilian infrastructure." [12] This three-man study team (John Lybrand, Craig Johnstone, and Do Minh Nhat) reported on apprehension and interrogation methods; the condition and number of jails, prisons, and stockades; and graft and corruption.

Regarding overcrowding, by early 1966 there was no more space available in the GVN's prison system for "Communist offenders." And as more and more people were captured and placed in PICs, jails, and detention camps, a large percentage was necessarily squeezed out. Hence the revolving door.

In the fall of 1967 the forty-two province jails where most VCI suspects were imprisoned had a total capacity of 14,000. Of the four national jails, Con Son Prison held about 3,550 VCI members; Chi Hoa Prison in Saigon held just over 4,000; Tan Hiep Prison outside Bien Hoa held nearly 1,000; and Thu Duc held about 675 VCI, all women. Approximately 35,000 POWs were held in six MACV camps scattered around South Vietnam. VC and NVA prisoners fell under U.S. military supervision while ARVN camps handled ARVN deserters and war criminals. [13]

***

As attorney Harper wrestled with the problem of judicial reform, a mild-mannered, medium-built, retired Marine Corps colonel, Randolph Berkeley, tackled the detention camp problem. Before retiring in 1965, Berkeley had been the corps's assistant chief of staff for intelligence. In 1966 he was hired by the Human Sciences Research Corporation to do a study in Vietnam on civil affairs in military operations, and in early 1967 he briefed Komer in the White House on the subject. Komer liked what he heard and hired Berkeley (who had no corrections experience) as his senior adviser on corrections and detentions) in which capacity Berkeley returned to Saigon in July 1967 as a member of the ICEX staff.

Upon arriving in Saigon in July 1967, Berkeley was assigned by Evan Parker to manage the SIDE (screening, interrogation, and detention of the enemy) program. Berkeley and five assistants -- all experienced corrections officers -- were listed on paper as employees of Public Safety's Department of Corrections.

"Shortly after my arrival," Berkeley recalled in a letter to the author, "I was called to report to General Westmoreland. I found him with staff members and Ambassador Komer, and it was explained to me that I needed to draft a plan, within a few weeks, which would make the prisons secure from attacks, as valuable lives were being lost in capturing VC who would then be sprung quickly to fight again .... The Westmoreland meeting turned me into an operator so busy with his requirements," Berkeley explained, "that my focus was more on prisons than detentions. [14]

"The CIA provided me space in one of their offices at MACV headquarters, and for several weeks I flew about in an Air America plane, scouting locations for attackproof detention facilities and prisons, taking aerial photographs myself, and developing the plan." While doing this, Berkeley learned: "There were over forty prisons nationwide, detention facilities [usually 'just a barracks surrounded by barbed wire'] in every province, and the GVN had neglected all of them in nearly every aspect, including protection from attack by the enemy.

"When my plan was presented on schedule, General Westmoreland approved it and directed that I execute it. In the next few months the prisons were provided defensive weapons and guards trained to use them, and ... attacks on prisons quickly lost their popularity. One other device we used was to fly VC prisoners to Con Son Island, which was secure from any enemy attack."

Having satisfied Westmoreland's requirement for prison security, Berkeley turned to the issue of detention facilities. "I visited Singapore and Malaya to look at prefab construction for possible use in detention camp construction but decided it was cheaper to do the job with local resources available in Vietnam. Meaning the detention problem was dropped like a hot potato, this time into the hands of the GVN." ICEX Memo No. 5, dated November 2, 1967, handed responsibility for the operation and security of detention camps to the province chiefs, with advice and some resources provided by MACV through Berkeley and the Department of Corrections.

On December 27, 1967, MACV issued Directive 381-46, creating Combined Tactical Screening Centers and stating: "The sole responsibility for determining the status of persons detained by U.S. forces rests with the representatives of the U.S. Armed Forces." Case closed. In every Combined Tactical Screening Center, the detaining unit did the screening, interrogating, and classifying of rows and civilian detainees, sending enemy soldiers to POW camps or to Saigon if they had strategic intelligence, to provincial jails if they were common criminals, or to PICs if they were deemed to be VCI.

"There were, in effect," Evan Parker explained, "two prison systems: "the civil one under USAID and the military one for POWs. PICs were separate and staffed as an agency program ... but there had to be a lot of understanding between us in order not to waste money." For example, the CIA would provide PICs with vans but not gas or oil or mechanics. The Phoenix coordinator would then have to persuade the Public Safety adviser to persuade the Vietnamese police chief to provide these materials and services to the Special Branch, which, considering the ongoing rivalries, got done grudgingly, if at all.

"The problem Phoenix dealt with," Evan Parker added, "was making sure that when a knowledgeable person got picked up, the right person got to talk to him and he just didn't disappear in the system." This weeding-out process happened in the PICs "because there you had the Vietnamese whose salaries were paid by the agency. They weren't beholden to the military or AID."

Ultimately Phoenix did nothing to alleviate the problems of civilian detainees. Rather, as Phoenix threw its dragnet across South Vietnam, tens of thousands of new prisoners poured into the already overcrowded system, and the revolving door syndrome was simply converted by province chiefs into a moneymaking proposition. Meanwhile, ICEX lawyers tried to paper over the problem by compiling a handbook on national security laws and procedures, which legalized the attack against the VCI by permitting the administrative detention of VCI suspects for up to two years without trial. No steps were taken to establish due process for civilian detainees.

***

Tab 11 called for the Phoenix Directorate "to conduct an on the ground review of interrogation facilities, practices and procedures, including coordination, exploitation, and follow through, with a view to optimum support to the attack on the infrastructure." The object was to focus interrogations on intelligence concerning the VCI at province and district levels and to improve coordination with other agencies. No report was required from the CIA compartment within the Phoenix Directorate on this sensitive subject.

Regarding the "practices" of the PIC program, what is known of official policy comes from Nelson Brickham. "I had an absolute prohibition in field operations activities toward conducting or sanctioning or witnessing any acts of torture," he said. "I said the same thing to my province officers from the third day I was in-country. My statement [which he never put in writing] simply was 'Any of you guys get caught in this stuff, I'll have you going home within twenty-four hours.' And there never was such a case that came into existence, although it's possible that there was and the reports never got to me."

Brickham also directed his province officers "to run the PICs from a distance. It's a Special Branch operation; Americans are not to be identified with the program. These guys were not to go near the PICs on a day-to-day basis. They were not to participate in interrogations there or anything like that."

Brickham's directive was ignored. Warren Milberg, for example, spent "15 percent" of his time in the Quang Tri PIC, supervising interrogations and advising on questions and topics to pursue. His experience is typical; an earnest Phoenix officer had to be at the interrogation center to obtain intelligence quickly. Indeed, in the final analysis, interrogation practices were judged on the quality of the reports they produced, not on their humanity. "Phoenix advisers who took an interest in PIC operations," Milberg writes, "normally attempted to improve the quality of interrogation techniques by carefully going over reports and pointing out leads that were missed and other items which should have been explored in greater detail." [15]

As for torture, "While the brutalization of prisoners did occur, interested Phoenix personnel could curtail support for the PIC unless such unauthorized activities ceased." However, Milberg adds, "Since most advisers were neither intelligence nor interrogation experts, the tendency existed to provide passive support and not to try and improve PIC operations." [16]

According to Robert Slater, director of the Province Interrogation Center program from July 1967 until April 1969, "The first thing the Vietnamese wanted to do was tie the guy up to a Double E-eight." As advisers, however, there was little he and his training team could do to prevent this use of an electric generator, other than to try to raise the professional standards of PIC personnel. Slater and his team (augmented and eventually replaced by a Vietnamese team) taught Special Branch employees how to track VCI suspects on maps, how to keep files and statistics on suspects, and how to take and process photos properly. They did not teach agent handling; that was done in Saigon by CIA experts imported from Washington. "The whole concept of the PIC," according to Slater, "was to get them in and turn them around. Make them our agents. It didn't work for us, though, because we didn't reward them well enough." [17]

The major "procedural" problem in the Phoenix interrogation program concerned the disposition of high-ranking VCI suspects. According to Parker, "High-level prisoners and Hoi Chanhs were invariably taken to higher headquarters and never heard from again." Milberg agrees: "People [at region or in Saigon] grabbed our best detainees on a regular basis, so you tended not to report that you had one. You'd keep him for two or three days," to get whatever intelligence he had on other VCI agents in the province, then report that you had him in custody." Milberg writes that when "prisoners of high position in the VCI were removed from local PICs for exploitation at other levels, morale of PIC personnel decreased. Often the result was that the PICs became auxiliary jails and were used to house common criminals." [18]

For Robert Slater, the transfer of important VCI prisoners to higher headquarters was merely standard operating procedure. "We trained Special Branch people how to properly keep statistics and files, how to use a board in the office to track cases, but most important, to send hot prospects from province to region to the National Police Interrogation Center [NPIC]." In other words, Phoenix interrogation procedures at the province (tactical) level were superseded by interrogation procedures at the national level -- the political-level Phoenix seeking strategic intelligence.

Having been the CIA's senior adviser at the National Police Interrogation Center, Slater had valuable insights into the interrogation system at its summit. His story began at Camp Pendleton in early 1967, when he was asked to join a presidentially directed counterinsurgency program that trained and sent fifty Vietnam veterans from the various military services back to Vietnam to serve as province officers and Phoenix coordinators. "But I was a separate entity," he noted in a conversation with the author, "... although we went over at the same time." A Vietnamese linguist with three years of interrogation experience in-country, Slater was assigned to the NPIC "on the basis of a decision made in Saigon. Dave West said he won me in the lottery, when the station people sat around and reviewed the resumes of the people coming over."

Slater's cover desk was in USAID II, where he sat beside his boss, a tall, muscular, blond CIA officer named Ron Radda, who served as an adviser to Dang Van Minh. Slater attended briefings given by Minh every morning at the NPIC on Vo Thanh Street, where he had his covert office. "When a prisoner came in from, say, Da Nang," Slater explained, "the reports would come over to my section. I'd put them on an eight-foot-long blackboard and report anything hot to Ron." At that point Radda and Minh's interrogators went to work.

Headquarters for both the Special Branch and the National Police, the NPIC was "a monstrous French compound with a separate, restricted wing for the Special Branch. We cleaned it up," Stater said. "Actually whitewashed it." After Tet, the CIA also built the Special Branch social club, the Co Lac Bo, on the gravesite of the VC killed during Tet. The NPIC held between three and four hundred prisoners, most of whom, Slater says, "were packed forty or fifty in little black holes of Calcutta."

The fact is that prison conditions and interrogation practices in Vietnam were brutal -- especially those taken out of sight. Case in point: "At a quarter after twelve on June 16, 1967, I was driving home from work to have lunch with my wife," writes Tran Van Truong in A Vietcong Memoir. Suddenly a car cut him off. Two men jumped out, pushed their way into his car, and told him that General Loan had "invited him to come in for a talk." Instead of going to the NPIC, however, Truong's captors took him to the old Binh Xuyen headquarters in Cholon. As he was led into the reception room, he found himself "face to face with a burly, uniformed man whose slit eyes and brutal expression were fixed on me in concentrated hatred ... a professional torturer who had personally done in many people." The interrogator said to Truong, "I have the right to beat you to death. You and all the other Vietcong they bring in here. There aren't any laws here to protect you. In this place you are mine." [19]

Truong describes this secret interrogation center. "Sprawled out on the floor the whole length of the corridor were people chained together by the ankles. Many of their faces were bloody and swollen; here and there, limbs jutted out at unnatural angles. Some writhed in agony, others just lay and stared dully. From the tangle of bodies came groans and the sound of weeping, and the air was filled with a low, continuous wail. My heart began to race. On one side of the hallway were the doors that apparently led to the interrogation rooms. From behind these came curses and spasmodic screams of pain." [20]

Later Truong was invited inside one of these rooms; it "looked like a medieval torture chamber," he writes. "Iron hooks and ropes hung from the ceiling, as did chains with ankle and wrist rings. These latter devices were well known among the activists and Front prisoners, who called them the Airplane. In one corner was a dynamo. Several tables and benches stood in the middle of the floor or were pushed up against the walls." What happened next, you can imagine.

The last tab of "Action Program," Tab 12, directed Evan Parker and his staff to establish "requisite" reporting systems, "for purposes of program management and evaluation, and for support to field collection and collation activities and operations against infrastructure." [21] At first, each agency used its existing system. Province officers gathered information on the VCI from the collation sections of PICs. They then sent this information to region officers, who used liaison branch reporting formats to relay the information to RDC headquarters in Saigon. There it was analyzed and plugged into a data base "against which future developments and progress may be measured." MACV sector personnel sent their reports on the VCI through military channels to the MACV Joint Operations Office in Saigon, which then coordinated with ICEX.

As MACV and CIA Phoenix personnel were gradually incorporated within CORDS province advisory teams and assigned to PIOCCs and DIOCCs, monthly narrative reports were sent directly to the Phoenix staff in Saigon; meanwhile, the Vietnamese used their own parallel, uncoordinated reporting systems.

Standardized reporting was fully authorized on November 25, 1967, and focused on three things: (1) the number of significant VCI agents eliminated; (2) the names of those eliminated; and (3) significant acquisition, utilization, and other remarks. Until mid-1968 reports about the DIOCCs would occupy as much time as reports generated by the 103 DIOCCs in business at the time. Ultimately information gathered on individual VCI suspects in the DIOCCs became the grist of the Phoenix paper mill.

Go to Next Page