DOCID: .

~6

ID:A523682

TOP SECRET

NO.

1666

UNITED STATES CRYPTOLOGIC

HISTORY

.,..

,
.'.
• '.'

d-t
.'.

,

- "'.
C"

'Ii!l:"
.,

,

•.•..

'.

.,~
."_.

t'

~&;-~.

American Cryptology during the Cold War, 1945-1989
Book II: Centralization Wins, 1960-1972

T KEYHOLE

TOP SECRET

CCH-E32-95-03 TCS-54649-95

DECLASSIFIED UNDER AUTHORITY OF THE INTERAGENCY SECURITY CLASSIFICATION APPEALS PANEL. E.O. 13526, SECTIO 5.3(b)(3) ISCAP No.z..~-o~\. .Document Date 1'" .•.ly U, 2.01
T

>

DOClD: 523682

REF lD:A523682

-

This monograph is a product of the National Security Agency history program. Its contents and conclusions aloe those of the author. based on original research. and do not necessarily represent the official views of the National Security Agency. Please address divergent opinion or additional detail to the Cen ter for Cryptologic Hi story
(E322)_

This document

is not to be used as a source

for derivative

classification

decisions.

DOClD: 523682

REF lD:A523682
TOP SeCRI:T UMSItA

UNITED STATES CRYPTOLOGIC HISTORY

Series VI

The NSA Period 1952 -Present
VolumeS

American Cryptology during the Cold War~ 1945-1~89 B.ookII: Centralization Wins~ 1960-1972
Thomas R. Johnson

CENTER FOR CRYPTOLOGIC HISTORY

NATIONAL SECURITY AGENCY
1995

lOP SE(REf tJMRFtA

DOClD: 523682

REF lD:A523682
lOP SECRET l:I MBRA

Table of Contents
Page

BOOK II: CENTRALIZATION WINS, 1960-1972
ChapterS: The Kennedy years.......... .. 289

The New Administration 289 McNamara at Defense 289 NSA and the Cryptologic System at the Beginning of aNew Decade .. 291 Enter the New Director 292 .People, Money and Organization 293 A New Reorganization 294 . Changing the Field Organization 296 The Cryptologic Map in the mid·1960s 297 Europe ·297 Turkey......................................................... 298 Pakistan . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 303 The Far East 306 .Back Home 306 New Collection Systems " , 308 NSA Gets Involved 310 The Airborne System 313 The TRS Program , . . . . .. 315 The Cuban Missile Crisis 317 The SIGINTEffort 318 The Berlin Wall 319 The Buildup to Crisis 320 The SIGINTContribution : , 323 The Crisis 324 The Aftermath :......................................... 330
Chapter9: The Post- Cuban Missile Crisis Reforms

The Dilemma of Centralization ANew Director NSA's Community Relationships ELINT(Again) .;................................................. DEFSMAC '.' . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. The Advent ofthe Command Center Centralization ofTheater Processing :................ CSOC SIGINTat the White House

338 340

341 343 345 346
348 350
352

HANDLE VIA~T~~:::~C~O~::::L~S~Y~ST~EMSJOINTL NOT ETO.FO ALS

'l

iii

TO~ !!CftfT UMBAA

DOCID: 523682
TOP $!CR!T tJMBRA

REF ID:A523682

~.O. 13526, section 1.4(c)(d)

Withheld from public release Pub. -';'.86-36

Carter Takes Command Mechanization of the SIGINT Process
1 I .............• .' ....•...•.....................•.......••..•...•.•••.....•..

I

AG-22 Changing the Communications
STRAWHAT

System

The Computer Industry at NSA IATS The Communications Solutions Automating the Collection Process

IBauded
1
COMSEC

Signals\
1

~.

The Bissell Study at Mid-decade The Development of American Secure Voice
TEMPEST ..•.......................................................................

357 361 362 . 364 . 364 366 . . 368 . 369 . 370 . 371 373 : : : : : : : : : : : : : : : : : : : : : : : 374 . 376 378 . . 379 . 379 381
. . : . . . . . . . . .
.
. : .

Geographical Retreat Turkey Pakistan Iran Airborne Collection The Wood Study .' The Harrogate Experiment Naval SIGINT Ships The EndofHF? : ;
1
STONEHOUSE . 1 I

382 .

:

1

'.'

:

Overhead : The Air Force ELINT Programs Program Management The Advent of Overhead COMINT TheC]Payloads Program C ~
RUNWAY RAINFALL
1 I

: . .
.

.

. .
.

....................................................•..

NSA's Foreign Relations Germany

.
.

.

D::::::::::::::::::::::::::::::::::::::::::::::::::::::::: NSA and CIA in the Third Party World :
NSA and GCHQ :

. .

383 385 388 390 391 393 395 397 398 398 401 402 404 404 406 406 407 408 409 411 411 412 413 414 415 415

HANDLE VIA TALENT KEY __ ~~EIl:EASABLE

CONTROL SYSTEMS JOINTLY TO FOREIGNr1N~'IOtHrl,s...... __

TOP SECRET UMBRlf<

iv

I E.O.

13526, section 1.4(c)

-------------_.

__

.

--

-_.

Withheld from public release .b:t:-86-36

DOCID: 523682

REF ID:A523682
f6P SeeIU!T HM BR"A

Chapter 10:

SIGINTin

Crisis, 1967-1969
425 426 432 439 439 443 445 449 453 454 462 462 463 464 467 470 470 472 473 474 474 475 476 477 479 481 482 485

War 0. • • • • • • • • • • • • • • • • • • • • • • • • • • The Cryptologic Posture . . . . . . . . . . . . . . . . .. The Attack on the Liberty :................................. The Pueblo Set-up..................................................... Capture Merma th Assessments Czechoslovakia The Prague Spring Romania-The Invasion that Never Happened The Shootdown of the EC-121 North Korea and the Aerial Reconnaissance Program The Mission ,........................................ The Crisis Security and the Work Force in the 19609 Dunlap :................................ Hamilton David Kahn and The Codebreakers :................. Cryptology is Legalized American Cryptology at the End of the Decade Relationships with the Military Marshall Carter Retires , Gayler Takes the Helm . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. The Eaton Committee '. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . The Eachus Committee The Creation ofNSOC :.............................. , SIGINT in the Nixon White House

SIGINT and the Second Arab-Israeli

Chapter 11: NSA in Vietnam: Building the Effort - The Early Years
Vietnam - The Country The Americans Enter the Fray Laos and the Beginnings of Direct American Involvement Hanoi Decides to Intervene in the South NSA Expands Cryptologic Involvement The Buildup ofCryptologic Assets DF Goes Airborne Into the Mire . . . . . . . ' 495 497 499 500

502
503 506 509

"
1 .•.......

I E.O.13526,

509 section 1.4(c)
1 The

Withheld from public release Pub. L. 86-36

Diem Coup The Cryptologic Expansion of 1964 AFSS Comes to Vietnam

. .

510 511 511

HANDLE V AsABLE TO ~OR v

NTROLSYSTEMSJOINTLY

TOP SeCReT UMBRA

DOClD: 523682
TOP SECRET UMBRA

REF lD:A523682

The Crisis The The The

in the Gulf ....................................•............. Desoto Patrols ;......................... 2 August MaddolPPatrol 4 August Patrol

.,

515 515 516 518

Chapter 12: From Tonkin to Tet.-: The Heart of the War
The President Expands the War Operation Starlight and the Ia Drang Campaign .. . . . . . . . . . . . . . . . . . The SIGlNTDeployment ARDFand the Two-Front War Search and Destroy :..... Predictions Infiltration· . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. The Dancers '. . . . . . . . . . The SIGINTRole in the American War The Air War '. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Fashioning the SIGINT Warning System - HAMMOCK •.•..••••••.•••.• The Border Violation Incident
. IRON HORSE ......•.•..........................•...•.......• ,.....

529 530 531 532 534 538 539 542 543 543 544 547
549

BIGLOOK .•.•.••.••..•••.....•....•....••.•..........•..•.••.•..• Weather and SAR Warnings PURPLEDRAGON The PURPLEDRAGON Task Force The Permanent Staff '

'. . . . . . . . . . . . . . . . . . . . . . . . .

550 550 551 553 555

Chapter 13: The Withdrawal
The Tet Offensive The Planning The Beginnings Khe Sanh NSA and the Impending Storm The Storm The Assessments The Warls Vietnamized The President Pulls Out Vietnamization American Special Operations The Cambodian Incursion Lam Son 719 The Son Tay Raid The Easter Offensive TEABALL The U.S. Moves Out of Vietnam TheSummingUp The Turn of the Wheel ' 559 559 560 ;......... 561 , 562 ;.......... " 563 564 565 565 566 " . . . . . . . . . 570 572 576 576 579 579 581 583 584 ,

;................................

'FOP SECRET "'MIRA

vi

DOCID: 523682

REF ID:A523682
TOP SECRET I;JMBRA

Glossary of Abbreviations Sources

. .

589 597

Index

.0

0.. ··

0·····.····

0

. 611

i

HA

CONTROLSYSTEMS JOINTLY

vii

lOP SECREll:lMBAA

DOCID: 523682

REF ID:A523682
TOP 5ECRR I:JMBRA

ChapterS The Kennedy Years
THE NEW ADMINISTRATION
In the long hiatory of the world, only a few generations freedom in ita hour of maximum danger. have been granted the role of defending - I welcome it. I

I do not8h.ri~k from this.responsibility places

do not believe that any of us would exchange generation.

with any other people or any other our

The energy, the faith, the devotion which we bring to this endeavor willligbt

country and all who serve it -and the glow from that fire can truly light the world. John F. Kennedy,lnaugural Address. 20 January 1961

John Kennedy came to the White House with an abiding interest in foreign affairs and defense policy. His politics, forged during formative years of the Cold War, were hard-line anti-Communist and anti-Soviet. But unlike Eisenhower, whose instinctive conservatism drove him toward small government and small defense budgets, Kennedy wanted a liberal remake of the world. Under the driving and optimistic Kennedy, it seemed that anything was possible and that John Fitzgerald Kennedy could make it happen. Kennedy knew little about intelligence when he arrived at the White House. He needed an interpreter but avoided the existing channels (DCI, secretaries of state and defense). Instead, he came to rely on an official on his White House staff who held the title of national security advisor. His choice for this relatively little-known office was McGeorge Bundy. Previous occupants of the position had been relatively obscure, but Bundy and his successors, Walt Rostow and Henry Kissinger, were to become household names. Power had shifted to the White House staff. MeN amara at Defense

For many years, the office of the secretary of defense had been weak and understaffed. The first secretary of defense had an office but little else. James Forrestal had no legal. deputy, no staff, a miniscule budget, and no tools to curtail the interservice feuding which had erupted after the war. In 1949 President Harry Truman got a reluctant Congress to create a Department of Defense, with a staff and a budget to go with the solitary office of secretary. The Defense Reorganization Act of 1958 accorded the secretary more staff and more power. Subsequent secretaries (the despondent Forrestal having committed suicide) battled the three warring services through the Eisenhower years, and each was driven .'nearly to distraction.

HANDLE VIA r AL

__--~~£D~~SABLETOFOREI

289

TOP SECRETUMBRA

DOClD: 523682
TOP U(RET tJMBRA

REF lD:A523682

/

John Fitzgerald Kennedy

H~~

~1\;)hI:JLETO

FOREIGN

.

'FOP SECRETUMlltA

290

OOClO: 523682

REF lD:A523682
TOP S!CR!T UMBItA

No one quite anticipated someone like Robert McNamara when the Defense Department was established. He had come over from industry. Brilliant and driven, he had become CEO of Ford Motor Company at the age of forty-four. McNamara was a Republican and had been so far from Kennedy's inner circle that the two had never met. He brought with him new techniques for managing large organizations. He was a centralizer par excellence, and he ruthlessly beat back internal opposition. McNamara resembled less a secretary than a cyclone. The new secretary brought with him a management team headed by Charles Hitch of Remington Rand. Hitch had had a hand in inventing a new discipline called Operations Research. Essentially, OR, as it was called, tried to quantify the basis for all manager-ial decisions. Using scientific methods, he would reduce all the variables Robert McNamara, of a decision to a mathematical quantity secretary of defense and choose the most attractive. Hitch under Kennedy and Johnson institutionalized the PPBS (planning, programming and budgeting system), a seven-year planning cycle which is still in use. As DoD comptroller, he scrutinized every element of the defense budget. The largest intelligence package was the newly created CCP, and Hitch and friends examined it rather thoughtfully every year.' Kennedy was not happy with the doctrine of massive retaliation. He was an activist, and MC 1412 (the document that codified massive retaliation in 1956) was essentially a defensive strategy. Instead, he opted for Maxwell Taylor's strategy of flexible response, which required conventional and unconventional forces to meet tacticalthreats. Finally codified in Me 1413 in 1967, flexible response in fact dominated the strategy of both Kennedy and Johnson throughout the decade. 2 .
NSA and the Cryptologic System at the'Qeginning of aNew Decade

Flexible response caught off guard an unsuspecting SIGINT system that had been optimized over an eight-year period to warn of, and support, total nuclear war. Not enough

HANDLE VlATA ~N[9l.UlI:;!1l~ABLE

. ONTROL SYSTEMS JOINTLY TO FOREIG

291

TOP SECRETUMBRA

DOCID: 523682
TOP S5CREl \;IMBRA

REF ID:A523682

Withheld from public release Pub. L. 86-36

I E.O.

13526, section L4(c)

attention had be~n paid to tactical SIGINT,not enough resources had been allocated. servicemen had flocked to large fixed sites and had learned how to work strategic SIGINT. problems. The weaknesses of the existing SIGINT system had been exposed I land the services were working on solutions. But no one was really ready for the decade of crisis and war that was to follow.

I

I

This became a decade ofSIGINT centralization. Just as the McNamara Defense Department strove to tighten the reins, so NSA, bolstered by repeated recommendations by high-level boards, commissions, and committees, drew SIGINTcontrol back to Fort Meade. True, there were countervailing forces, most notedly tactical commanders in Vietnam, who strove for a decentralized system. But at decade's end, the SIGINT system was far more tightly knit than it had been ten years earlier. Former deputy director Robert Drake once jokingly formulated a law that said, "Centralization is always bad, except at my level." NSAemployed Drake's Law to centralize its own system, but at the same. time fought a spirited rear guard defense against McNamara's people at DoD. Centralization was fine,unless it meant giving up any powers to the Officeofthe Secretary of Defense (OSD). Thus NSA tried to stave otTthe intrusions of Hitch's budgeteers. Succeeding directors fought the authority of the newly created Defense Communications Agency. The, creation of the Defense Intelligence Agency (DIA), too, represented a threat that NSA constantly crossed swords with. And NSA rejected the idea (pushed by Kennedy's PFIAB) that the DCI spend more time coordinating the entire intelligence effort, including the intelligence components of the Department of Defense. CIA was still regarded as a threat. Even to defense intelligence specialists, NSA was still an obscure agency in 1960. It entered the decade known primarily as a communications research organization which played with expensive toys and produced huge volumes of highly classified translations in a fairly leisurely time frame. Analysts still worked basically an eight-to-five schedule, and shift operations, when mounted, were highly unusual and tailored for specific crises. But pressure was mounting to change things .. SIGINT had proved to be of great utility on a widening variety of targets. It had become the most prolific producer of strategic warning information, and President Eisenhower had demanded that such information get to him faster. Kennedy was an activist president, who demanded even quicker and more accurate responses. He prodded the system, and NSA responded. By the end of the decade, NSA's world would change. Enter the New Director Vice Admiral Lauren~e H. Frost, who arrived at the end of the Eisenhower administration in 1960, was better prepared for the job than any other previous director. He had had three prior tours in intelligence, including a two-year tour as Canine's chief of

HANDLE VI ~-NnT

INTCONTROL
RELEASABLE TOFO

SYSTEMS JOINTLY

TOP SECRET UallBRA

292

DOCID: 523682

REF ID:A523682
TOP SECRET UMBRA

staff, and he had been director of Naval Intelligence. In addition, he had achieved distinction as a ship driver in two wars. The Army and Air Force had had their turns as . DIRNSA - now it was the Navy's turn. Frost contributed to SIGINT centralization by revoking the independence of the Soviet Navy problem at NSA. A compromise device instituted by Samford to bring the SCAs more fully into the N~A system, it had resulted in divided loyalties and jurisdictional disputes. In March of 1962 Frost resubordinated the chief of the Soviet navy problem to DIRNSA, removing him from the Navy chain of command where he had been directly subordinate to the director of the Naval Security Group. The independence of the Soviet ground and air problems lasted not much longer than that.3 But Frost himself lasted only two years in the job, and aside from that organizational change, left behind no distinctive legacy (for reasons which will be made clear on p. 340).
Laurence H. Frost

People, Money, and Organization By the time Kennedy arrived in the White House, cryptology had become the elephant in the intelligence closet. McGeorge Bundy discovered that of the 101,900 Americans engaged in intelligence work, 59,000 were cryptologists of one stripe or another (58 percent). Of those, about half worked in the Continental United States, while the other half plied their trade overseas at collection and processing sites. NSA had 10,200 assigned (17 percent of the total) but only 300 overseas billets. The field sites were still the domain ofthe SCAs. At NSA, the military filled 25 percent of the billets." Of the three services, NSG was still the smallest, with 6,900. AFSS, with 21,200, and ASA, with 20,400, dwarfed the Navy in size, although NSG made up quality what it lacked in quantity. Cryptologic manpower was projected to grow through the decade until it would-hit a peakof93,067 in fiscal year 1969.5 .

in

HANOL

EYHOLECOMINTCONT NOTRELEASAB

SJOINTLY ONALS

293

TOP5ECR!T UMBRA .

DOCID: 523682
I UP SECRET tJMBRA

REF ID:A523682

Within NSA's Production organization, fully 50 percent worked the Soviet problem. Another 8.4 percent worked in Acom (Asian Communist) while 7.6 percent.' were in Allo (all others, i.e., Third World). The remaining 35 percent was allocated to centralized technical or staff functions such as machine processing and collection support (including
ELINT).6

NSA's complex at Fort Meade underwent a building boom in the 1960s. Ground was broken for the nine-story headquarters building, and it was occupied in 1963. (General Canine attended the ceremony, and his wife cut the ribbon.) The new COMSEC building was dedicated in November 1968, and the quarters on Nebraska Avenue were finally given back to NSG. In the same year, owing to a moratorium on military construction, NSA began to lease three newly constructed "tech park" type buildings at Friendship Airport (which later changed its name to Baltimore-Washington International, or simply BWI). The complex was called Friendship Annex and came to be abbreviated as F ANX. In' 1961 NSA acquired the buildings that had housed the old Fort Meade post hospital and moved the training school from downtown Washington. The training component, newly renamed the National Cryptologic School, was one of the first occupants of the Friendship complex, gladly abandoning the antiquated hospital structure. A New Reorganization Following the Martin and Mitchell defection in 1960, the director established a management board to review NSA's organization. It was the first comprehensive review since the McKinsey study in 1956. This time, instead of an outside management team, Admiral Frost used home-grown talent. The board was chaired by Frank Rowlett (who had rejoined NSA during the Samford administration), Oliver Kirby from Prod, Brigadier General George M. Higginson, Maurice Klein (the head of personnel), and Dr. William Wray, with Dr. Milton Iredell as recorder." Its report, handed to Frost in July 1961, amounted to a reversal of the McKinsey approach. What was needed-was not decentralization (a key element of the McKinsey report) but centralization. The director's staff had grown too small, and too many functions had been farmed to Prod. "The Board found no effective mechanism within the existing organisation to exercise the strong centralized control of national policy, planning, and programming functions, which appears essential to insure concentration on and responsiveness to the Director's national responsibilities." Thus it ,created a policy staff to manage Second and Third Party affairs, to do central budgeting for the CCP and to effect systems planning and evaluation. It was similar in approach to that being used by McNamara's people in OSD (although probably no one at NSA would admit it).

MIN'r CONTROL SYSTEMSJOIN'ILY
~-TiruTRELEASABLE TO FOREIG

TOP SECRET l1MBRA

294

DOCID:523682

REF ID:A523682
lOP SECRET l:IMBRA
.. ':- .... ".

_ .....•

',.

Groundbreaking

(or the new headquarters

building

The Friendship Annex (FANX) complex

HANDLE VIA TALENT K

ONTROLSYSTEMSJOINTLY

T RELEASABLE TO FOREIGN NATI

295

. TOP SECRET UMBRA

DOClD: 523682
TOP SeeR!T tlMBRA

REF lD:A523682

The naming conventions for office designations was also tossed out the window. Martin and Mitchell had, at their press conference, reeled off a long list of NSA organizations, and it would be necessary to change to a new system. Out were the pronounceable syllables, in was the obfuscating alphanumeric system. Key components were to be designated by a single letter (R for R&D, P for Production, etc.), and subordinate elements would carry trailing numbers." PROD itself consisted of three key components: A B C the Soviet problem; everything else, including former ACOM and ALLO; and the

technical functions such as machine processing, central reference, former office of collection (including, for the time, ELINT processing).

Included on a central PROD staff would be a permanent watch office and an office of cryptologic research (an early version of PI). The board also recommended that the arrangement come to an end whereby the chiefs of the Soviet naval, ground, and air problems were subordinated to their SCA chiefs. Frost (as noted above) acted on this the next year." The board recommended that R&D be strengthened to handle increased responsibilities. (This was in accord with, and partly in response to, DoD-level recommendations that NSA take a more active hand in the development of cryptologic equipment across the board.) The R&D organization should assume policy direction on major new projects such as the Air Force's 466L collection system and the space collection (Spacol) systems. The COMSEC R&D function, which historically shuttled between COMSEC and R&D, returned to the research organization.'? Finally, the board took another swipe at the continuing lack of a career civilian cryptologic service. This had been a big issue during the Canine years, and fragments of the system had been put in place. But a systematic professionalization system, with . categories and criteria, had never been implemented. Under Samford the proposals had languished, and now another board made another recommendation. It was a continuing irritant.ll

Changing the Field Organization
While Europe remained stable, cryptologic organization in the Pacific was changing. The switch of NSAPAC from Tokyo to Honolulu, already mentioned, occurred under Frost in 1962. In the same year ASA and USAFSS moved their own regional headquarters to Hawaii to be in synch with military organization in the theater. This .was also a time when second-echelon processing in the Pacific finally came together in I I In the fall of 1961 a new processing organization, Joint Sobe Processing Center, opened its doors.

Ir-E-.-O-.-1-3S-Z-6-, -se-c-ti-on-1.-4-( c-)-'
HANDLE VIA CONTROL SYSTEMS JOINTL Y

TOP SECRET "UBRA

296

Withheld from public release Pub. L. 86-36

DOCID: 523982

REF ID:A523682
TOP S&C:RETYMIlAA

I

E.O. 13526, section 1.4(c)(d)

Withheld from public release Pub. L. 86-36

I

The first commander was an army colonel, Kenneth Rice of ASA, but there was also a large contingent of NSA civilians working LI ---' As time went on, it acquired processing responsibilities for North Vietnamese air, air defense, General Directorate of Rear Services (GDRS),and shipping.'!

I

Bucking the trend toward centralization, AFSCC remained operating in San Antonio. NSA wanted to move it to Fort Meade hut did not have the space. This problem would not be solved until the Friendship complex was leased in 1968. Meanwhile, AFSCC continued to work the third echelon aspects of the Soviet air problem, and it even acquired the I [problem under an agreement negotiated with ACOM early in the decade.UI In the meantime, NSA continued to set its own targeting priorities. Systems were devised throughout the 1950s and 1960s to allow for the expression of customer requirements, but none really had any teeth, and they were so general ("copy and report the world") that NSA was forced to prioritize for itself. The best indication of where NSA's priorities lay was the Agency's input to the new PPBS system in 1961. NSA thought that exploiting I Iwas Job One, followed in orde
'-----.1 It is fair to note that the Soviet problem encompassed four of the seven and that Cuba was not among the listed requirements. This omission would not last long. 14

THE CRYPTOLOGIC MAP IN THE MID·l960s By the time NSA was eight years old, the cryptologic map had exploded. NSA and the SCAs were in seventeen countries plus the Continental United States,' Alaska, Hawaii, and Puerto Rico. The biggest growth was in Germany. The three SCAs had major field sites in thirteen locations, and NSA had a theater headQuarters in Frankfurt. I

Europe Although the Robertson Committee warned against vulnerability to Warsaw Pact forces, collection sites were still strung out in a wide arc east of the Rhine.

I
HAN~LE~LY
NO LE ION _

E.O. 13526, section 1.4(c)(d) Withheldfrom public release

297

TOP SECRETUMBRA

Pub. L. 86-36

DOCID: 523682
TOP SECRET UMBRA

REF ID:A523682
Withheld from public release Pub. L. 86-36

I

E.O. 13526, section 1.4(c)(d)

, A8A began occupying the "Rubble Pile" late in 1962 or early in 1963. 15 Across Western Europe, new U.S. SIGINTsiteswere springing up. 1 I· I I ASA and AFSS were building 'sites and AFSS had occupied land on Crete (a Greek possession in the Mediterranean) an~ Wheelus Air Base in Libya. (Wheelus was deactivated in 1960 rather than pay additional rent to the increasingly nationalistic Libyan government, and the mission was moved to Crete.) All these sites were important because of the geographic cushion they gave from the potential advance of Soviet divisions.

c=J

Turkey

As a base of] I however, no country was more important than Turkey. The Soviet missile testing program drove the Turkey option, and in the 1950s the administration became increasingly close to the Turkish government. In 1955 Turkey joined the Baghdad Pact (a long-forgotten Eisenhower initiative to knit together the proWestern countries on the southern periphery of the USSR). Five years later a relatively antiseptic military coup placed,the pro-American General Menderez in power in Ankara and ushered in a period of harmonious relations between the two countries." The United States had been frantically building collection sites in Turkey in the 1950s. I I

I
HANDLE VIA A RELEASABLE TO FOREIGN NA TI

E.O. 13526, section 1.4(c)(d)

I

ROLSYSTEMSJOINTLY

TOP SECRET l;.IMBRA

298

Withheld from public release Pub. L. 86-36

IO~UEAsr' "'RIlA.

E.O. 13526, section 1.4(c)(d)

N IX)
It) ('t')

N
Li1 HANDU:V A

o H o

299

8

DOClO: 523682
TOP 5!CR!T UMBRA

REF lD:A523682

ABA's first collection effort on Teu(elsberg,

established in 1961, operated out of vans.

The Rubble Pile (Teufelsberg, West BerliD, as it looked when completed)

HANDLE VIA TALE ~.--!'tm~:[LEASABLE

ONTROLSYSTEMSJOINTLY TO FOREIGN

TOP SECREHIMBRA

300

1

'DOCID: 523682

REF ID:A523682 E.O. 13526, section 1.4(c)(d)

I

Withheld from public release Pub. L 86-36

I

E.O. 13526, section 1.4(c)(d)

Withheld from public release Pub. L. 86-36

N CD ID M N

~

r..

Q H

~

N CD ID M N LIl Q H

o

TOP SEeR&;~UIA'

302

o
Q

.DOCID: 523682

REF ID:A523682

I

E.O. 13526, section 1.4(c)(d)

I

Withheld from public release Pub~ L. 86-36

"fOP SECRET tJMBItA

Pakistan Like Turkey, Pakistan became geopolitically important to the Eisenhower administration because of its concern over the Soviet menace. Pakistan, Iran, and Turkey were lumped together by Secretary of State John Foster Dulles as the "Northern Tier of defense," and the administration cultivated all three. During the 1950s they were successful. Pakistan joined both the Baghdad Pact and the Southeast Asia Treaty Organization (SEATO). In 1954 Eisenhower announced that Pakistan would receive American military assistance.f \ John Foster Dulles had a friend in Karachi. His name was Mohammed Ayub Khan (normally referred to as "Ayub"), and he happened to be the military chief of staff. Ayub
T CONTROL SYSTEMS JOINTLY ELEASABLE TO FOREIGN N

HANDLEVIAT

303

TOP SECRET tlMBftA

---~_

..

__

.-

..

-

DOCID: 523682
1'8P SECRET YMBitt.

REF ID:A523682

worked consistently in the direction of close American-Pakistani ties. When, in 1958, he took over the government in a coup, the Eisenhower administration was hopeful that. relations would grow even closer. The signing of a mutual assistance agreement in 1959, whose wording appeared to leave no doubt that the United States would defend Pakistan against its enemies, seemed be a harbinger ofthe future. 2S

to

E.O. 13526, section 1.4(c)(d)

Withheld from public release Pub. L. 86-36

HANDLE VIA TA
OT RELEASABLE

CONTROL SYSTEMSJOINTLY
TO FOREIG

lOP SECRETtJMBRA

304

Withheld from public release Pub. L. 86-36

I

'fepSECRA" 'Y'p·

E.O. 13526, section 1.4(c)(d)

Q H

==:m!Etfir'c~
EASABLE TO YO

.

o

305 -

'Tm'SEcRef blMPRA

o
Q

DOClD: 523682
TOP SECRET UMBRA I E.O. 13526, section 1.4(c)

REF lD:A523682
.Withheld from public release Pub. L. 86-36.

The Far East Diplomatic problems of the magnitude of I I and I Idid not present themselves in the Far East. American SIGINT sites in the former American colony of the Philippines remained unqliestioned and unnoticed at Clark Air Base, north of Manila, and San Miguel near the giant Subic Bay Naval Base.
'--

Okinawa had become a virtual aircraft. carrier for American SIGINT collection, with sites at Sobe,l· I Hanza, and Kadena (wh~re the Air Force had begun an airborne intercept program). Processing operations were becoming centralized on the island, partly a result of the recommendations of the Robertson Committee in 1957.
--1

Japan was like Germany -close to the enemy, an ideal SlGINTplatform,and in a quasi. subordinate diplomatic status resulting from the American occupation. SIGINT sites at I IMisawa, and Wakkanai provided the Americans with excellent access to Soviet Far East, Korean, and Chinese communications,

I

I

I
Korea, still reverberating from the late war, remained heavily outfitted with SIGINT sites. An early plan to close sites after the war was over had been scotched, and the peninsula was still dotted with tactical ASA and AFSS units.

On the Pacific rim, Alaska, Hawaii, and Guam rounded out the SIGINTstructure. Hawaii was important as the headquarters of CINCPAC (with resulting SIGINT representation) and as a communications relay in the long HF hop across the Pacific. Alaska was far more important from the collection standpoint, frontihg as it did the Soviet Arctic. AFSS virtually took over the SlGlNT mission there, doing HF and VHF collection from various places, and flying ACRP aircraft. out of Eielson AFB. The most famous (or infamous) site was on Shemya, a miserable, wind-swept island known affectionately as "The Rock," ilJ"st occupied by SIGINTersin 1955.
Withheld from public release Pub. L. 86-36

Back Home

I E.O.

13526, section 1.4(c)

I

In the Continental U.S., ASA maintained major collection sites on both coasts, at Vint Hill Farms in Virginia and Two Rock Ranch in Petaluma, California. These had been important sites during World War II, but they were gradually losing their importance to the more far-flung European andPacific collectors. Navy SIGINT operations consisted primarily ofDF sites along both coasts and remained fully as important as during the w~r because their targets, being mobile, came to them rather than the other way around. In the Caribbean, the Army dominated the-Panama area with a site at Fort Kobbe, while the

Tep SECRET l:IMBRA

306

Withheld from public release Pub. L. 86-36

E.O. 13526, section l.4(c)(d)

N
(J)

ID M N III

HANDLE

VIA TA

__

~",.~LEASABLETOF 307

CONTROL SYSTEMS JOINTLY ALS

CI

H U

o
CI

DOClD: 523682
TOP SEeR!T UMBItA

REF lD:A523682

Navy maintained collection in Puerto Rico. The nice thing about collection close to home was that it minimized costs (collection from foreign locations being outrageously expensive), and it was not burdened with diplomatic problems. But the disadvantage was '-W-it-h-h-e-I-d-f-r-o-m-. --, hearability, and the U.S. collection base was slowly being eroded by the success of places like Peshawar, Wakkanai, I The future (at least the immediate future) was in public release Pub. L. 86-~6 exotic (and expensive) locations. I E.O. 13526, section 1.4(c) I

I

'---------=----'

New Collection Systems All three services modernized their field site equipment to equip the new sites being built around the world. But during the 1950s no SCA was as aggressive as AFSS. The 1950s marked the birth of a major new HF and VHF collection system whose trademark became a huge Wullenweber-style antenna called the FLR-9. Its distinctive appearance came to symbolize SIGINT to the outside world. The Navy was actually the first SCA to become involved with the Wullenweber design. NSG needed a worldwide DF system, and after having experimented with Wullenweber designs (chapter 4, p. 138), they settled on a system which came to be known as the FRD10. A large circularly disposed antenna array (CDAA), the FRD-I0 divided the HF spectrum into two bands, and thus it had double rings of antenna elements in a ring 873 feet in diameter. RF cables from the antenna elements were routed into an intercept building in the center of the array. This was a cheap and secure option but limited the size of the building. But DF, rather than collection, was the primary objective, and owing to an NSG strategy that scattered many small sites around the world (rather than concentrating into a few large ones), the size ofthe building was not a big issue. Beginning its systems R&D work in 1956, NSG fielded its first CDAA at Hanza, Okinawa, in 1962. By 1966 they had built thirteen FRD-10 sites in three foreign countries, the U.S., and its territorial possessions." ' Among the three SCAs, Air Force Security Service began life in the worst shape from an equipment standpoint because it simply inherited cast-off ASA equipment. But the Air Force emphasis on building its own, completely independent and self-sufficient SIGINT system resulted in very large amounts of money being poured into the USAFSS coffers. It also resulted in an AFSS R&D organization that waslarger and better funded than the other two SCAs. In the early 19505, AFSS set to work designing a new collection system from the ground up. The proposal went forward as a package under Gordon Blake, the new USAFSS commander, in March ofl957. It was called Project 466L, and included three components: a. GLR-1. a VHF system, optimized for processing.
ELINT

collection and first-echelon

HANDLE VIA TALE ~--NrlTRELEASABLE

CONTROLSYSTEMSJOIN~Y TO FOREIGN NATI

lOP SECRET l:IMBRA

308

IOPSECREI

UMBIbt

FRO-tO,

Hanu.

Oklru.w&

N CD 1.0 C"l N
L{)

HANDLE VIA TALE

NTROL SYSTEMS JOINTLY ASABLE TO FOREIGN NA 309
IOP.SECKEI UMBRA

g
Q

Q H

DOClD: 523682
Tap SECRET l:JMBRA

REF lD:A523682 I E.O.
13526, section L4(c)

I

Withheld from .public release Pub. L. 86-36

b.] theHF system, optimized for COMINT. The distinctive antenna was called FLR-9, but the package included more than just that. c. ~ a VHF airborne system. It never got past the prototyPe stage.

I

In addition, the 466L project came to include computers for second-echelon processing. It was a complete field system, minus the buildings. Sylvania won the contract to build the systems. 28 .. The most successful portion of the system was the FLR-9 component. With a circumference of 1,200 feet, it was the largest single antenna the U.S. ever designed for SIGINT. It was arranged in three circular rings, each with antenna elements-optimised for a certain frequency band, and a 120-foot-high reflecting screen. Antenna leads were routed into a central "roundhouse" where. complex beam-forming equipment and DF goniometers resided. From there a cable trench took coaxial cables outside the ring to the RF distribution room of the collection building. The distribution room looked a lot like the old manual "spaghetti boards" that predominated at standard sites, but without the people. An operator selected antennas by pushing a button on the position rather than calling to an RF'distribution operator on an intercom to reconnect cables. Early in its life someone called it an "elephant cage:' and the name stuck.29 The above-HF portion of the system, called OLR-l, was to be optimized for ELINT collection and first-echelon processing. Hof, Samsun, and Wakkanai. with partial systems at Misawa (processing only), Trabzon, Shu Lin Kou, and Northeast Cape. At a projected cost of I I a copy, GLR-l was hideously expensive. It was also fraught with technical risks which ultimately jeopardized the entire

I

I

project.so NSA Gets Involved

I E.O. 13526, section L4(c)

Withheld from public release Pub. L. 86-36

NSA watched from the sidelines in the mid-1950s as NSG and AFSS independently designed and fielded separate collection and DF systems. The Agency urged, with no result, that the two services compromise their differing requirements and develop a single system goodfor both tasks. Then in 1957 NSA.became directly involved when it was asked by the Air Force to review the AFSS 466L proposal. The level of involvement increased in 1958 when NSCID 6 gave the Agency a more explicit role in guiding and coordinating. service cryptologic R&D. NSA opposed the way AFSS was proceeding with the project. Apart from the lack of agreement between AFSS and NSG on harmonized development, NSA was concerned that:
8.

The project, especially the GLR-l, was far too expensive;

. b. Major components were overdesigned (Again, GLR-l was the culprit.);

I~~ TOP S!CRl!T tJM81tA

RELEASABlE TO FO

3H)

DOClD: 523682

REF ID:A523682
Tap SECRETUMBM

c. AFSS was proceeding with a generalized requirement, while NSA believed that AFSS should proceed with a "special purpose" approach, and that this would reduce costs; d. Sylvania, selected as the prime contractor several important areas; {or .the FLR-9, lacked experience in

e. AFSS had planned no test models of either system but had designated the initial sites (Hof and Samsun for the GLR-l and Chicksands and San Vito for the FLR-9) as "prototype sites." Nonetheless, AFSS planned to contract for the follow-on sites before knowing how things were working out at the prototypes." In 1960 NSA took its concerns about the 466L system to DDR&E and convinced him to freeze money for out-year funding. At this point the 466L prototype design was thoroughly reworked by NSA and AFSS, and many of the GLR-l "frills" were eliminated before the Wakkanai system was built. So extensive were the changes that the system was retitled and became known as FLR-12. The prototype sites were retrofittedto the new FLR-12 design.S2 . . Security Service planned originally for seven FLR-9 sites: San Vito, Chicksands, Misawa, Clark, Peshawar, Karamursel, and Elmendorf. As a result of experience with the prototype systems and NSA participation in the later R&D stages, the follow-on sites eliminated some of the features, such as automated DF flashing, that had made. the earlier sites so expensive." Owing to aforementioned diplomatic problems with Pakistan. the Peshawar system was never built. . Alone among the SCAs, ASA showed little initial interest in CDAAs. But by 1960 the command was looking more closely at the future of the FLR·9 and was' attending jointservice planning meetings at NSA. Soon thereafter ASA decided that its newly planned intercept site at Udorn in northern Thailand would be a CDAA based on the Air Force's .FLR-9 design. They named the project I l and the new site (called Ramasun Station) was opened in 196~. When ASA began planning the consolidation of its three largest German sites (Rothwesten, Herzo Base, and Bad Aibling) into a single super-site, the FLR-9 was again the option selected. By coming into the game late, ASA avoided the substantial development costs that AFSS had incurred. They simply bought "off-theshelf'designs.54

I E.O.

13526, section 1.4(c)

Withheld from public release Pub. L. 86-36

HANDL LEASABLE TO FO

TCONTROLSYSTEMSJOINTLY

311

lOP 5&(RET UMBRA

H U

c
fep

5EEREJ IoIII81U\

312

o
o

DOClD: 523682

REF lD:A523682
TOP !JEeREr tlMBRA

I E.O.

13526, section 1.4(c)

The Airborne System

Withheld from public release Pub. L. 86-36

USAFSS remained the biggest user of airborne collection platforms. Called the Airborne Communications Reconnaissance Program (ACRP), the program then consisted of a fleet of nineteen RC-130s configured with ten COMINT intercept positions each. The emphasis in those days was on VHF voice, especially Gel communications. Most of the routes were along the periphery of the Soviet Union and China, standing well back from the border to avoid another shootdown similar to the 1958 incident over Armenia. The command never experienced another shootdown. 35
In the late 1950s Security Service began working on a new program that would bring the RC-135 airframe into the ACRP program. It was developed from the KC-135 tanker used throughout SAC. Owing to the fuel capacity, the aircraft could routinely fly in excess of sixteen hours (the RC-130 was generally limited to an eight-hour mission) at altitudes topping 40,000 feet. USAFSS initially funded three airframes, packing f"llteen intercept positions into its innards. The flying partner was SAC, rather than a theater component command, and positions were converted to ELINT, to be manned by SAC electronic warfare officers. The program was called and it began flying out of Eielson AFB, Alaska, in early 1963. The RC:135 became the Cadillac of airborne collectors and eventually took over the entire job from the RC-130s.36

I

I

I

~

I

In the 1960s SAC continued its own SlGlNT airborne collection program. The SAC program initially used RB-47s with a limited EUNT capability. Later the program Iconverted to RC-135s with EUNT collection being the objective. COMINT positlons on board (manned by USAFSS operators, and I I Iserved for advisory warning." .

I

I

I

As for the Navy, it continued to rely on its fleet of seven EC-121s, although a newer and better aircraft, the P3 Orion, was first delivered in 1962. It would eventually replace the slower 121s, whose vulnerability was convincingly demonstrated when the North Koreans shot one down in 1969 (see p. 462). The Navy program also retained its specific fleet support role, and it was always regarded as something of a maverick by NSA because its tasking was entirely a Navy matter." In the rush to collect SOviettelemetry, the U.S. employed a wide variety of collectors. Ground-based sites could never be certain to collect all the telemetry available, the most significant gap being telemetry that was transmitted on the pad before launch and immediately on lift-off. The information from this stage of telemetry was critical to an assessment of missile capability, and the only way to get it (before the advent of overhead collectors) was through airborne collection along the southern Soviet periphery. \

I E.O.
.

13526, section 1.4(c)

I

Withheld from public release

~HA~N;DmL~E~V~IA~T!A~~~~~~~~~~T~R~O~L~SY~S~T~E~M~S:J~O~IN:T1LY L-_P_u_b_._L_._8_6_-_3_6~
ASABLETOFOREIGNNA

313

TOP SeCRET l:IMBRA

DOClD: 523682

REF lD:A523682

I E.O. 13526, section 1.4(c)

I

Withheld from public release Pub. L. 86-36

The airborne reconnaissance program occupied the thoughts of President Kennedy the early days of his administration. He had learned that Khrushchev was planning turn over the surviving RB-47 pilots (shot down in the White Sea in July 1960) as a kind diplomatic peace offering to the incoming administration. But nothing had been done avoid future incidents, and Kennedy was anxious to insure that Khrushchev not be able again hold captured fliers as diplomatic pawns. The White House demanded action."

in to of to to

At the time, six advisory warning programs were in existence in various theaters, all with different criteria and warning methods. Some airborne programs (the Navy being the most prominent example) still flew without any warning capability at all. In 1961 the Pentagon took two actions to try to establish a program that would satisfy the White House. First, it created the Joint Reconnaissance Center, which would be responsible for coordinating and approving all peripheral reconnaissance worldwide. Second, it directed that a USAFSS advisory warning plan be modified and adopted worldwide." . The USAFSS program, which had originated in the Far East in the early 1950s, ~ad received NSA blessing in 1961. The chief impediment to its adoption worldwide was lack of agreement on a standard communications system. The Pentagon finally settled on the SAC single sideband communications system, which was a worldwide HF system accessible to all parties. The Navy held out unti11962, but finally agreed to the standard plan, and the new advisory system, called White Wolf, was adopted the following year. 4S The shootdowns dropped to almost zero - the only notable exception was the 1969 shootdown of aNa vy BEGGAR SHADOW mission along the coast of Korea, an incident that precipitated the creation of NSOC: The danger of peripheral SIGINT airborne reconnaissance missions becoming diplomatic contests dropped almost out of sight, and a long-standing source of diplomatic embarrassment simply went away.

HANDLE VIA T RELEASABLE

TCONTROLSYSTEMSJOINTLY TO FOREIGN N

'fOp SECRET UM8RA

314

DOCID: 523682

REF ID:A523682
TOP SE(AET blMBRA

The TRS Program The Soviet SIGINTtrawler program has been of such long standing and so visible that it is often forgotten that the United States, too, at one time had its own SIGINTtrawlers. It was called the Technical Research Ship (TRS) program. c=Jwas the beginning. NSA had no collection] and, land-based sites being so difficult to acquire, possibility of building a floating collection sittl that the need could best be satisfied by taking (essentially, freight-haulers) out of mothballs and Bureau of Ships estimated that it could be done for require eleven to twelve months." 11956, that NSG look into the The Navy thought some World War II Liberty ships converting them to SIGINT use. The about $4.5 million per ship and would it requested
t-"j

Withheld from' public release. Pub. L. 86-36

I

c
~
Ul
N

~O\

Defense budgets were slim in the late 1950s, and the first money was not in the budget until fiscal year 1960. The first !;hip selected, the USS Oxford, put to sea in 1961. She could do eleven knots Not much was happenin at the time, so the Oxfort!s first cruise was set for the west coast of Mrica later in the year. Instead, in November it was diverted to the Caribbean to cover a burgeoning crisis between the United States and Cuba. Already, the TRS program, only one ship large, was showing how flexible it could be.~ Enthusiasm over the potential of such floating collection sites led NSA to cut corners in order to get a second ship on line quickly. In early 1961 the Agency, beset with insistent collection requests by the DCI, found that the Military Sea Transport Service (MSTS) had a smaller, slower vessel that could be converted in fairly short order for only $2.5 million. Despite being smaller, the Valdez was crammed with twenty-two positions, and began her first cruise, to Mrica, about the same time the Oxford was deployed to the Caribbean." There developed from this decision two sorts of TRSs. The first, of the Oxford class, . was a wholly Navy owned and manned ship, larger and faster by a few knots. The second, owned by the MSTS, was a coastal type vessel with a civilian crew to go along with the NSG people in the SIGINTcompartment. The Navy ships were designated USS vessels, and by mid-decade the navy component of the TRS fleet consisted of five ships: the Oxford, Georgetown,Jamestown, Belmont, and Liberty. The smaller maritime vessels were designated USNS and consisted of only two ships: the Valdez and Muller. In 1968 a third was added to this list:USS Pueblo. 47 As for intercept positions, the ships did not vary much. The Oxford class typically carried, when fully outfitted, between twenty and twenty-five positions, while the Valdez class had between eighteen and twenty-one. Where they differed was in speed and general sea worthiness. Clearly, the Valdez class represented a less capable, but cheaper, option.48

~ ~

HANDLE VIAT

INTCONTROL SYSTEMS JOINTLY OTRELEASABLE TOFOREIGN 315 TOP SECRET UMBRA

TOP SKAET UMlRA

USN8VoldLz

N CD III M N 10

HANDI..E VIA TALENT K _---nvn ••LEASABLETO

ROL SYSTEM'S

FOREIGN NAnu~"..,,..-.

JOINTI. Y _

H U

C

.-1'0'

SEEAI'

uMaliA"""

316

o
C

DOClD: 523682 I E~O. 13526, section
1.4(c)

REF lD:A523682 I
Withheld from public release pub. L. 86-36

TOP SECRET l:JMBRA

One variant of this prop-am was established specifically to monit9~ I ============,-1-= I:I:::n-=l::.at:::e:....1:.:9:.:6:.:1~t:::here arose an urgent requirement to monitor a I An MSTS charter vessel, the Robinson, was hastily converted in only a few days and sailed from New York in January 1962. Its SIGINT manning was unique for a vessel - it was a combination of NSG and ASA operators in a partnership similar to the I program at the time. In February the Robinson relieved the Valdez, which had been pressed into emergency service]

I

I

I

I

I

In May 1963 there was another urgent collection requirement. The Robinson was headed for port after a long cruise, and so JCS arranged for NSA to use ani I USAFSS provided an equipped van and ASA furnished EUNToperators' for the cruise. Istayed on station through July, when the Robinson returned. So began a collection program that was to result in the I Ivessel which became an important I Icollector in later years. 4\1 Withheld from

I

I

I E.O. 13526, section

1.4(c)

I

public release Pub. L. 86-36

THE CUBAN MISSILE CRISIS
We were eyeball to eyeball, and I think the other fellow just blinked. Dean Rusk, 28

October 1962

About the greatest crisis of the Cold War, three. things can be said that concern cryptologists: 1. It was very definitely not precipitated by SIGINT warning. It was, and always has been, regarded as a crisis initiated by photographic intelligence, and there is nothing in the historical record to alter this statement. It marked the most significant failure of SIGINTto warn national leaders since World War II.
2. SIGINT played a very significant role in the unfolding crisis, a role which subsequent publicity and declassification of documents ~ave not fully revealed. 3. It marked a watershed, like the 1956 event, in the way cryptologists do business.

HANDLE VIA TALE ASABLE TO

ONTROL SYSTEMS JOINTLY FORE}
'fE)P SECRET l:JMBRA

317

/'

DOCID: 523682
TOP SeCRET I:JMBA-A

REF ID:A523682

The, Cuban situation began on its own. Years of poverty and political repression on the island ended in a young revolutionary, Fidel Castro, marching into Havana in January of 1959. But hopes that it woulddevelop into a pluralistic, liberal-style government were quickly dashed, as Castro put in place more and more institutional trappings of a solid Communist dictatorship. Experts eventually conceded that he had probably not been driven into the arms of the Communists by American hostility, but had planned it all along. Diplomatic contacts with the USSR had begun almost immediately, with the arrival of Soviet foreign minister Anastas Mikoyan in February of 1960 to open a Soviet trade exposition. Formal diplomatic ties were established in May.

A young Fidel Castro only days alter his guerrilla arm marched into Havana in 1959 Withheld from public release section 1.4(c) Pub. L. 86-36

The SIGINT Effort

I E.O.

13526,

I

SIGINT also tracked burgeoning trade between Cuba and the Soviet Bloc. Although cargo manifests were rather vague, it was becoming clear through SIGINT (as with a variety of other intelligence sources)that muchof the trade was military. In July 1960the first substantial military aid arrived in Havana, and it included Czech small arms and ammunition and five MI-4helicopters. Soonthereafter Cuban pilots were noted in SIGINT training in Czechslovakia,originally on piston-enginefighter trainers. so

I

The tiny Cuban shop atNSA/ lived offintercept fromthe Navy site in Puerto Ricoand the ASAstation at Vint Hill, Virginia, and had virtually no traffic from Cuban internal net~. Requirements against Cuban military targets werealmost nonexistenn" Withheld from

I

I

I E.O.
HANDLE VIA
~-"'1'inT

13526, section 1.4(c) .

I

public release Pub. L. 86-36

CONTROL SYSTEMS JOINTLY RELEASABLE TO FOREIG

'Fep S!CRE I UMaRA

318

DOClD: 523682
I E.O. 13526, section 1.4(c) I

REF ID:A523682
Withheld from public release Pub. L. 86-36

TOP SECRET tJM11tA

NSA had indications through contacts in the commercial world that Cuban internal communications would eventually go to microwave. I

But the target, while audible, was too weak to be copied at that distance. A new approach was needed. and NSA requested that the Navy try to intercept the microwave system from one of its afloat direct support units (DSUs). The first hearability testing was done by NSG operators aboard the USS Massey, which circumnavigated Cuba in Juiy
1960.52

I

The Defense Department already had non-noD competition. Following Castro's successful revolution, it was used primarily to support CIA's covert operations in Cuba.53

I

I

I I

By the Bay ofPi~8 failure of April 1961, NSA's level of effort had ·increased people but was still not a large-scale effort. ·At that point the Kennedy administration began directing a. major concentration of intelligence assets against Cuba, and SIGINT resources increased rapidly. A year later I . 1 people were involved. and by October 1962.1 1were allocated to the Cuban problem." I E.O. 13526, section 1.4(c) I Withheld from The Betlin Wall
public release Pub. L. 86"36

I

Although it began as a uniquely Caribbean phenomenon, Cuba quickly became a part of the international struggle between the U.S. and the Soviet Union. ·It came to be a pawn in the Cold War. a piece of Communist real estate located within the American sphere of geographic influence. On the other side was Berlin, Western-owned property clearly located within Khrushchev's zone of contrcl. Khrushchev understood the relationship between the two territories and exploited them adroitly. Berlin as a crisis first erupted in 1948 when Stalin cut off land access to the city. The resultant Berlin Airlift lasted for just over a year and marked a significant test of American resolve. It remained a potential sore spot, and in 1958 Khrushchev announced that in 1959, lacking an overall settlement of the Berlin problem, he would give control of East Berlin to East Germany. Although the Eisenhower administration managed to talk the problem nearly away, it was clearly only a temporary respite. In 1961 Khrushchev again increased pressure on the city, and it seemed that Berlin, rather than Cuba, would be the flashpoint for war. At midnight on 11 or 12 August 1961, heavy trucks and troop carriers rumbled to the demarcation line between East and West Berlin. Construction crews jumped out and, under the guard of East German soldiers, began flattening a thin strip of land and

HANDLE ~_Ne'MtI1XEASABLE

ONTROL SYSTEMS JOINTL Y TO FOREIG

319

lOP SECRET UMBRA

DOCID: 523682
TOP SECRET tJMISRA

REF ID:A523682

stringing barbed wire in the middle of the zone. The Berlin Wall. SOOn to become a high concrete and cement block barrier, was begun. Kennedy was vacationing in his yacht off Hyannis Port. and he was not notified until noon on the 13th. He was reportedly furious. and he summoned CIA director McCone to examine the intelligence failure. CIA. in sifting through everything that had been available. did find one significant bit of information. A 9 August COMINT report of an East German Communist Party message discussed plans to begin turning all foot traffic back at the sector border, and the Watch Committee assessment had stated that this might be the first step in a plan to close the border." McCone could come up with no other predictive information; the Berlin Wall was still regarded as an intelligence failure, despite the existence offragmentary COMINT. Kennedy denounced the Berlin Wall, and American-Soviet relations worsened. On 1 September the Soviets ran their first nuclear test since 1958, breaking an informal moratorium that had been in place since the middle of Eisenhower's secondterm. But the one bright spot was in comparative strategic strength. The so-called Missile Gap, which had loomed so large in 1960, had become a proven chimera. In September 1961 Lyman Lemnitzer, the chairman of the JCS, briefed Kennedy that the U.S. enjoyed a 7 to 1 advantage in strategic nuclear delivery capability. The Soviets still had only ten to twenty-five operational ICBMs. and Kennedy could launch more than 1,000 delivery systems carrying 1.685 nuclear warheads. compared with 253 for the Soviets.56 The Buildup to Crisis In late 1961, as a result of the Kennedy administration's continuing concern with Cuba. the intelligence community was directed to increase its efforts against the island. NSA instituted a rapid buildup of the problem, almost certainly in response to this edict. 57 NSA's initial plan was forwarded to McNamara in November: It included manning additional positions at the Navy site in Puerto Rico, bringing TRS resources into the picture, and instituting a new program for translating Cuban communications. This and an augmented plan presented in February of 1962 were pushed rapidly ahead. Given the go-ahead, NSA assembled cryptologic resources with remarkable speed. The most significant addition was the Oxford. This first TRS had been launched in 1961, and the early plans were for an African coastal cruise. But NSA diverted the vessel to copy the new microwave communications in Cuba. I . I

I

I

I The Oxford conducted a hearability survey off the coast of Cuba in December 1961, and it soon began forwarding I Iintercept to
5&

I

NSA

.

I E.O. 13526, section 1.4(c)

Withheld from public release Pub. L. 86-36

HANDLE VIA T

T CONTROL SYSTEMS JOINTLY

~-1'Crn'1'RELEASABLE TO FOREIGN N

TOP SECRET I:JMBRA

320

TOP lfeREHIP'8A'

The Oxford

N 00 N
li1

The ftral TItS. lheOr{o,d.

"won III spurs"dul'lnc

the Cuban

Mi~&Il.Crisis .•

'"
(Y)

HANDLE VIA TA ___ ~ •• "."ASABl.E

NTROL SYSTEMS J01NTL Y TO FO

Q H U

321

lap SEEREf tlM!RA

o Q

DOClD: 523682
TOP SECRET UMBRA E.O. 13526, section 1.4(c)(d)

REF lD:A523682
. Withheld from
public release Pub. L. 86-36

The linguist project, calledl ~because it occupied quarters in the old Fort Meade Post hospital) employed native Spanish speakers in a semicleared status until their expedited clearances came through. They were employed translating the huge volumes of Spanish voice intercept being collected by the Oxford and the ACRP (see below)."

All this was accompanied by explosive growth of NSA's Cuban shop. At the time the Cuban problem was worked in an organization called Bl , whose chief, Juanita Moody, had arrived from the Soviet problem in July 1961. Moody would become a central figure in NSA's Cuban response effort, presiding over an effort that went from 0analysts in April 1961 ~ople in October 1962.61

E.O. 13526, section 1.4(c)(d)

Withheld from public release Pub. L. 86-36

MINT CONTROL SYSTEMS JOINTLY T RELEASABLE TO FOREIG

Tep SEERETUMBRA

322

\

DOCID: 523682

REF ID:A523682
Withheld under statutory authority of the Central Intelligence Agency Act of 1949 (50 V.S.C., section 403g)

The first . rtant SIGINTcontribution to the Cuban problem was th eporting of Cuban mercial ties with the Soviet Bloc in mid-1961. By early 19621 I Iwas reflecting extensive Cuban trade with the East Bloc and Canada .. Soviet communications revealed very large cargo shipments, but the cargo manifests were .conspicuously missing, and this, in and of itself, was an indicator of sensitive military cargo: SIGINT, photography, and HUMINTallcombined to form a very accurate mosaic of the increasingly dose commercial and arms ties.53 The U.S. government was kept fully informed of these developments through intelligence sources. I E.O. 13526, section 1.4( c) The Cuban military problem also began to take on distinctive East Bloc overtones. Intercepts of Czechoslovak communications showed, as early as the fall of 1961, that Cuban pilots were training in East Bloc fighters. Still later, Cubans were discovered I Ito be training in IL-28 light bombers in the North Caucasus Military District. It came as no surprise, then, that photography began showing various MIG fighters and IL-28 bombers in Cuba in mid-1962. 64

I
Withheld from public release Pub. L. 86-36

In June 196~ the first ELI NTintercepts from Cuba showed that they had Soviet radars, and before the end of the year there were both early warning and AAA fire control varieties. By May of 1962 Cuban air force communications reports I I I I Just a month later NSA reported intercept of the first airborne intercept radar in Cuba, definitely indicating the presence of MIGfighters on the island. Soviet controllers were being heard on VHF frequencies in heavily accented Spanish, instructing Cuban pilots and controllers in operational procedures. lIS The Soviets became progressively more active, both in numbers and in degree of control over the Cuban air defense system. USAFSS field sites intercepted the first Cuban grid tracking on 9 October - it employed the classic grid system used by the Soviet air defense system. After 27 October (the date the U-2 piloted by Rudolph Anderson was shot down; see p. 329), the Soviets virtually took over the air defense system, and Cubans, who had been in the center of things from the beginning, moved to the sldellnes." By mid-August reports 'began to refer to objects that sounded like SA-2s. On 29 August the first SA·2 construction was noted in U·2 photography. In September NSA confirmed operation of a SPOON REST,radar, often associated with the SA-2 system. At least one site appeared to be nearing operation."

I

I

I E.O.

13526, section 1.4(c)

I

Withheld from public release Pub. L. 86-36

. HANDLE VIATALE TCONTROL SYSTEMS JOINTLY _~~tE!:&EASABLE TOFOREIGN N

323

lQP SECRET UMBRA

DOCID: 523682
'fOp SECRET UMBRA

REF ID:A523682

TheCrisia

The crisis itself did not begin with the 14 October U-2 flight that found the missile construction sites, nor with the 22 October presidential broadcast announcing that fact to the world. It had been building all summer. and each escalation of Soviet assistance to Cuba brought the White House more directly into the picture. The president was deeply concerned about Soviet military assistance, and the reports he was getting (primarily CIA HUMINT sources) indicated that the technicians accompanying the military equipment were really Soviet troops disguised as civilians. The confirmed arrival and operation of SA-2s brought the crisis to a new level. CIA director McCone contended that theonly purpose he could see for such a modern defensive armament would be to protect something of very high value. and that something, he felt, would be offensive missiles. So from August on, the intelligence community focused quite specifically on that possibility. To try to head off a crisis. Khrushchev on 4 September dispatched Anatoly . Dobrynin, the USSR's ambassador in Washington, to the Oval Office to reassure Kennedy that offensive missiles were not in Cuba. On the basis of this reassurance, Kennedy authorized Pierre Salinger. his press secretary. to announce the arrival of the SAMs. but to stress that they were not offensive in nature. But, Salinger added. the gravest consequences would result from the introduction of offensive missile~. On 11 September the Soviet newspaper T(J$s buttressed Khrushchev's confidential communique on 4 September with a public announcement that the weapons in Cuba were defensive."

Jobn McCone, Kennedy's DCI, was virtuaDy alone in predicting that Kbrusbcbev would introduce offensive weapons into CUbL

On 31 August politics intruded. Senator Kenneth Keating of New York, a Republican. reported in the Senate chamber that he had evidence that there were 1,200 Soviet troops in Cuba, and "concave metal structures supported by tubing" that appeared to be for rocket

HANDLE VIA TALE OMINT CONTROL SYSTEMS JOINTLY ___ .-,.,.". RELEASABLETOFO NALS

TOP SECRET tlMBAA

324

OOClO: 523682

REF lO:A523682
TOP SECRET UMBRA

Installetion." To this day no one knows where Keating got his information, but CIA had at the time a profusion of unsubstantiated HUMINT reports dealing with such possibilities from their·HUMINT interrogation center at Opa Locka, Florida. 70 The overt result of Keating's charges was political. The congressional elections were due in November, and Kennedy obviously wanted to hang onto as many Democratic seats as possible. He was keeping his hands off Cuba with Soviet assurances that no such missiles-existed there, but the clamor for action on both sides of the congressional aisle was considerable. Any revelation that affected the equation could become politically explosive and might alter the balance of seats during the election. In this atmosphere the White House became extremely sensitive to any intelligence that might bear on offensive arms in Cuba. Meanwhile, on 7 September Kennedy was confronted with a new crisis. Major General Marshall "Pat" Carter, the deputy DCI (who would, three years later, become DIRNSA) showed the president U-2 photographs of a surface-to-surface missile complex under construction at the Cuban coastal town of Banes. The installation was for a short-range naval coastal defense missile, and Ray Cline, CIA's director of intelligence, speculated that it might be for the purpose of insuring that the Oxford stay well offshore. But in view of Keating's recent charges, any surface-to-surface missile might be misconstrued as offensive (as Kennedy at first did), and such information had to be held very closely. So Kennedy directed that any indication, however tenuous, of the introduction of Soviet offensive forces in Cuba, be kept tightly compartmented. Huntington Sheldon, the assistant deputy secretary for intelligence (and CIA's top liaison on SIGINTmatters) designed a eompartmentation system, which was subsequently approved by USIB. The result of this decision was an overly tight compartmentation at NSA. Information . on the subject was extremely limited in distribution, and SIGINT reporting on the subject was to be specially flagged "Funnel. n This was on top of an already rigid compartmentation system for U-2 photography, so secret that even Juanita Mo09Y, the chief of B1, and her chief of staff, Harry Daniels, were not brought into the picture (although Moody was told about the impending 14 October overflight by William Wray of NSA the morning that it happened). During the crisis SIGINT analysts were forced to work, in a vacuum. (However, some of the A Group analysts on the Soviet problem knew about the photography program.)" SIGINT was coming up dry. Intensive effort by both BI and A6 analysts revealed no
-indication whatsoever that the Soviets were bringing in offensive missiles. But unknown

to NSA, CIA, or the White House, the materials for the missile sites were already in Cuba. . Since the end of the Cold War, top Soviet officials have revealed that the decision to place offensive missiles in Cuba was taken in May, and this was followed immediately by the preparation and shipment of site construction materials. The first materials arrived in Cuba in mid-August, followed, the first week of September, by large pieces of equipment for the MRBM sites. The Soviets assessed that October would be the month of maximum
)

HANDLE VIA

T CONTROL SY~EMS JOINTLY
5

~-NOT1iiEfiLEASABLE TO FO 325

'FOP SECRET ~MBR-A

OOClO: 523682
TOil seCReT UMBRA

REF lD:A523682

vulnerability - site construction would be visible from the U-2, but the missiles would not be ready to fire, and Cuba would thus still be vulnerable to U.S. military action.72 NSA did not have the information, but neither did anyone else. The matter of the Soviets introducing offensive missiles in Cuba was considered by the intelligence community no fewer than four times in the first nine months of 1962, and each time the assessment was negative." On 19 September, during the middle of the building crisis, National Intelligence Estimate 85-3-62 assessed that such activity "would be incompatible with Soviet practice to date and with Soviet policy as we presently estimate it. It would indicate a far greater willingness to increase the level of risk in U.S.-Soviet relations than the USSR has displayed thus far .... " John McCone was out of town at the time, but indicated that he did not concur with the assessment of his own estimates shop. 74 In early October CIA got photos of crates on board Soviet ships bound for Cuba, which probably contained IL-28 light bombers. These were clearly offensive (if a bit deficient in real ofTensi~e punch), and Kennedy directed that the information be suppressed. McCone "stated that this was extremely dangerous," but he was overruled. He and Kennedy then agreed that such information be disseminated to the principals of USIB (which included NSA's director, Lieutenant General Blake), who would in turn restrict it "to their personal offices"~5 Since the first of August, CIA had mounted seven U-2 flights over Cuba, and it would have flown more but for Secretary of State Dean Rusk's constant protests that overflights were diplomatically risky. (Those protests were given additional weight when, on 8 September, a U-2 onloan to the Chinese Nationalist government on a special CIA program was shot down over western China.) Those that were flown carefully skirted Cuba's periphery, darting briefly into Cuban airspace for a quick overhead photo. Much of the island was thus going unphotographed. McCone persisted and finally got authorization for overflight of an area west of Havana which, according to some fairly coherent HUMINT reports, was undergoing construction for what looked like missiles. Bad weather forced several postponements, but the flight finally took ofTon 14 October and flew directly over the suspect area. The National Photographic Interpretation Center (NPIC) got a look at the pictures the af\ernoon of 15 October, and the CIA analyst, Victor DiRenzo, found what looked like six SS-4 MRBMs at a construction site. Looking at the photos on a light table in the Steuart Building in downtown Washington, NPIC's director, Arthur Lundahl. turned to the photo interpreters huddled around the light table and said, ~e are sitting on the biggest story of our time. n7e It was seven days before the presidentwould go before the world and announce the presence of the missiles and impose a naval quarantine around Cuba. Back at NSA, it was a frantic seven days. The Soviet and Cuban shops concentrated their resources on communications that bore on the problem. The A Group element that was working the

)

~~m~Y TOP 5!(ReT

LEASABLE TO FOREIGN

UMBRA

326 "

DOClD: 523682 I E.O.
13526, section 1.4(c)

REF lD:A523682 I
Withheld from public release Pub. L. 86-36
1'6' sEEREf tlMBItA

Cuban air defense system (controlled by the Soviets) physically moved into B1 spaces to facilitate interworking. A and B issued independent product reports, but they also issued periodic combined wrap-ups in order to tell a coherent story. Upwards ofc=JA Group analysts and linguists joined the new combined outfit.77

I

NSA needed a command center for the crisis. As it happened, A05, headed by Colonel I (USAF) and NSA civilian I Ihad recently taken over a small room across the hall from the A Group front office to receive and display compartmented information like photography (TK). During the crisis this became the new Command center. I Ihurriedly outfitted the room with telephones and employed A Group analysts to begin publishing a new product, thel Ia daily electrical report detailing the status oil L 78 The director, Gordon Blake, kept the Oxford on station throughout the crisis, and AFSS upped its ACRP flights off cu~ ~ Blake directed that ASA get its SIGlNTers as soon as possible and that the shipment of new equipments to the existing SCA intercept site~ be speeded Up.79

I

I

The most valuable intercept came fro in I I There being no processing capability in the field, all this was shipped back to NSA; there th~

L-

1 Throughout the crisis new and better equipments were added to the mix for faster and more complete processing.so
-J

The Soviets and Cubans had their own separate communications systems on the island. As the Soviets set up military operations (SAM sites, naval surface missile batteries, air defense networks, etc.), they maintained separate communications, supplying to NSA strong evidence that they were not integrated with the Cuban armed forces. NSA intercepted no cross-net communications. There must have been points at which the two sides talked - for instance, in Havana there was a command center housing both Soviets and Cubans, and it was served by communications of both countries. But there were no instances in which Soviets were intercepted talking to Cubans on the same communications facility. NSA concluded that the Soviets controlled all their own facilities,. including their SAM and air defense systems, and this conclusion was accepted at the naticnal level." The] I intercepts provided a wealth of, command and control information, and when married with photography, supplied a good picture of what was happening in Cuba. although. ~---------------------~-----~ photography showed

I

I E.O.

13526, section 1.4(c)

I
TO FOREIGN

Withheld from public release Pub. L. 86-36

HANDLE VIA TALE
~Jj:o:PitEtE:A~SABLE

ONTROLSY:n-EMSJOINTLY
'fe' SECRET UMBRA

327

Tap SEEtlET ~lIIIiRA

MRBM lite, Cuba CIA photo.li.ke tbit oneeon't'lnCH tbe pr •• ident to

act.

N \0
(Tl

CD

N III

HANDLE

VIA l' NOTRELtASABL.ETO

TNTCONTROLSYSTtWSJOlN"tLY FOREIG

o H
o

-TO;SfCkEI UMBRA

328

o

o

-----_

.._---.

---

DOClo.: 523682

REF lD:A523682

I E.O.

13526, section 1.4(c)

I

Withheld from public release Pub. L. 86-36

TOP SECRET !:IMBM

microwave radio relay equipment the three IRBM locations.8'l

being readied at three of the six MRBM sites and two of

Once Kennedy went on television (22 October), Soviet communications in Cuba lit up. A new air defense-associated net went on the air immediately. (This was what prompted the A Group processing element to physically move into space in Bl.) \
I

1\
The crisis continued to deepen over the next two days. Soviet merchant ships steamed toward Havana, heedless to the looming catastrophe. But early on 23 October the Navy I Iintercepted a broadcast from Moscow to all ships headed for Cuba to stand by for an extremely urgent cipher message. The message came through an hour later, and the intelligence community waited tensely for the reaction. Although undecipherable, it appeared to contain some sort of instructions. Late the same day NSG direction finding indicated that some of the Soviet merchant vessels heading for Cuba had stopped dead in the water, while others appeared to be turning around.. At this point, according to CIA's Dino Brugioni, the Office of Naval . Intelligence (ON!) felt that this information had to be verified before it was reported. John McCone was awakened in the middle of the night and informed that the Navy had unconfirmed information, but this was not passed to the White House or the secretary of defense until around noon of the following day, once ONI had "confirmed" the information. When he found out, McNamara was furious, and he subjected Admiral Anderson, the Chief of Naval Operations, to an abusive tirade. SO many years have passed that it is impossible to determine why the Navy held up information that seemed critical to the president's declsions.P" On 27 October the crisis reached its climax. At that point, Soviet ships had turned away from Cuba, a clear indicator that Khrushchev was wavering. But so far the two nations had not resolved anything. That day a U-2 piloted by Air Force major Rudolf Anderson (SAC had taken over U-2flights from CIA on 12 October) was shot down, and NSA reported that an SA-2 from the area around the naval base at Banes had been responsible. Based on COMlNT intercepts, the U.S. believed that the SA-2 sites were manned and controlled by Soviets.85 The shootdown of Anderson was a wide departure from the caution the Soviets had so far shown. Was ita major escalation? The shootdown of Anderson precipitated an ultimatum. In a meeting with Dobrynin that day, Kennedy told him that the United States would attack the missile sites in Cuba by Tuesday morning unless there was firm evidence that the missile sites were being dismantled. That gave the Soviet Union only forty~ight hours to resolve the crisis before air attack, which would be followed by a full-scale invasion. Khrushchev caved in, and he sent a frantic telegram to Kennedy that very night promising to remove the missiles.

HANDLE

NTROL SYSTEMSJOlNTL _.-bl:e'i'ftElCEAA:sJSABLE TO FOREIGN NATION

Y

329

"F9P SECRET "MBRA

DOCID: 523682
lOP SEeREl' tlMBRA

REF ID:A523682

The Aftermath

I E.O. 13526, sectIOn . 1.4(c) I

Withheld from public release Pub. L. 86-36

NSA learned two years later that Cubans migHt have been in control. of the site that fired at Anderson. In digging through the intercepts, NSA analyst I Ipieced together soine fragmentary SAM-associated multichannel communications from the Banes area, and discovered that the Soviets at one of the SAM sites were talking about a fire fight at one of the other sites on 26 October possibly involving invading Cuban military forces. Soviet security forces at neighboring SAM sites had been summoned, and it appeared to that the fight was over by the morning of 27 October when Anderson's U-2 was shot down. But he could not be absolutely, sure that the Soviets were back in control, and the possibility remained that Cubans had actually "pulled the trigger." This story created a sensation when, in 1987, investigative journalist Seymour Hersh published an account of the incident, as related to him from an unnamed analyst from an "intelligence agency." Internal evidence from Hersh's article points away from any NSA analyst as a source of the information, but the basic stOryline was correct," .

c:::J

The Hersh story appeared in conjunction with a series of conferences on the Cuban Missile Crisis, which came to include Soviet as well as American participants. During a conference in Havana in January 1992, a Soviet geqeral claimed that the Soviet commander on the island, one Issa Pliyev, had been given authority to launch nuclear missiles if Cuba were attacked. If true, this would have brought the world much closer to nuclear war than anyone suspected at the time. Robert McNamara, who had been secretary of defense at the time, uncritically accepted the Soviet's story, as did most other observers at the conference. The issue was sensationalized in the press." . It made good press, but it was not true. A search of declassified Soviet documents relating to the crisis showed that precisely contrCldictoryorders were issued to Pliyev. (Even the general who made the statements, Anatolii Gribkov, eventually backed away from his earlier assertions.) All evidence now supports NSA's long-held contention that Soviet forces were subject to monolithic central control and that local commanders, particularly in situations involving nuclear weapons, were strictly controlled through central release authority similar to that in the U.S. armed forces.88 The U-2 flights over Cub~ had not been receiving advisory warning support from the cryptologic community. It occurred in that interregnum between the JCS decision to impose a standard, worldwide warning system and the' actual publication and implementation of the resulting White Wolf plan. After the Anderson shootdown, Juanita Moody and Harry Daniels directed the hurried implementation of a warning system for the Caribbean area, and it was subsumed the next year under the White Wolfprogram.8lI . The shootdown undoubtedly increased pressure for the system that soon emerged. One of NSA's major jobs during the crisis was watching Soviet force readiness. On 11 September the Soviets suddenly went into their highest readiness stage since the

HANDLE VIA TALE

NTROLSYSTEMS JOINTL Y LEASABLE TO FOREIGN 330

TOP SECREt"YMBRA

DOClD: 523682 I E.O.
13526, section 1.4(c)

REF lD:A523682 I
Withheld from public release Pub. L. 86-36

TOP SECRET YMBfbOc

beginning of the Cold War. Although the units at highest readiness were generally defense-related. the alert included some unprecedented activity among offensive forces. too. Of greatest concern was a total standdown of the Soviet long-range air forces. It also coincided with marked standdown of activity in the GSFG (Group of Soviet Forces. Germany). a highly realistic major exercise in the Far East.· a major maritime communications exercise. a dispersal exercise by Baltic Se~ Fleet elements. a major exercise in the North Sea Fleet involving apparent nuclear dispersal actions, and the first ever western Atlantic patrol by a Soviet submarine. The alert may have been called because Moscow sus cted that Kenned had found out about the missiles. The 11 September alert was cancelled ten days later. but on 15 October Soviet forces went into a preliminary. perhaps precautionary. stage of alert. This was followed a day later by Soviet reporting of North American weather. Once again. this readiness was likely due to Khrushchev's supposition that the U.S. had discovered a missile site. (He knew the White House would find out; the only question was when.)!il Following Kennedy's Oval Office speech on 22 October. Soviet forces again went into an extraordinarily high state of alert, similar to the September event. This time. however. with nuclear war threatening. defensive forces were primary. Offensive forces avoided assuming the highest readiness stage. as if to insure that Kennedy understood that the USSR would not launch first. Long-range aviation units continued normal training, although some precautionary steps were taken, such' as insuring that the, Arctic staging bases could be used, (Bombers were not deployed to the Arctic.) PVO (air defense) units went into the highest state of alert ever observed. as did Soviet tactical air forces. In Although Soviet offensive missiles and IL-28 bombers were pulled out Cuba following the end of the crisis. a Soviet garrison force remained. The air defense system which the Soviets had imported to the island was slowly turned over to the Cubans. although during the crisis the Cubans had had no say whatever in its operation (which might in turn have led to the 26 October' attack at Banes). The SIGINTsiteat Lourdes was activated during the crisis] I I l The Soviets maintained their western Atlantic submarine patrols until the mid1980s. In later years Soviet TU-95s flew regul~r1y between the Soviet Union and Havana.

I

I

I

of

I

I

Cuba remained a bastion of Soviet influence and military force presence until the collapse of the Soviet Union itself. sa As for the eryptologic community, temporary sites became permanent.

I

It was a permanent Ldi"'·,....v-er-s"':'io-n-o-=r-S-IG-I-N-T-a-s-se-:t-s-. -co-n-t-ri::-b-u-:ti:-n-g-to-t7h-e-o-v-e-r-:al:':'I-:SI~G-I-N-:T-:ti;"'o-lrce buildup during the decade.94

I E.O.
HANDLE VIA TAL lN~~H:::!:XS:Ar;JBLE TOFOREIGN NA 331

13526, section 1.4(c)

I

Withheld from public release LSYSTEMSJOINTLY Pub. L. 86-36

"'fOPSECRET UMBnA

DOCID: 523682
Tap SEeRET UMBRA

REF ID:A523682

warning, so highly touted during the Eisenhower administration, failed in Cuba. Although siGINT detected some of the troops and equipment as they were moving, the key elements of the movement that would have given the Kennedy administration decisive information about offensive capabilities did not come from SIGINT. In a 1963 postmortem, the National Indications Center faulted the entire intelligence system for failure to detect those key elements. Soviet communications security was almost perfect.~
SIGINT

Although SIGINT failed in its job to warn, it was an integral link in the chain of intelligence that supported the administration during the crucial days of decision-making. It gave the United States its most timely and specific information about the movement of troops and supplies to Cuba. It provided the only information about force command and control - absolutely critical in making decisions about Soviet involvement. It gave the White House the only timely' information that it had about Soviet reaction and military force alert posture.' And it provided most of the hard information about the air defense system, should the invasion (set for 30 October) proceed as planned. 98 The response to the crisis at NSA was more coherent and orderly than in 1956. The six-hour SIGINT wrapups, including both Soviet and Cuban activities, were the first such attempt by NSA. Agency reporting gave a better overall picture to customers than it had in earlier crises.9'1 .' Within the intelligence community, the crisis precipitated a debate about NSA wrapup reporting. Roundly criticized in the fall of 1962 for exceeding its supposed reporting charter, NSA defended itself in USIB circles br pointing out that no other agency was performing the essential function of summarizing developments as seen through SIGINT. In the months following the crisis an unrepentant NSA began putting out a daily wrap-up of SIGINT events, called the SIGINT Summary, The name was customarily abbreviated to . the term "Sigsum," but many just caned it the "Green Hornet" (because it was distributed under a cover of dark green paper). It survives today as the SIGINT Digest. 98

Notes

1. Gregory Palmer, 1968 (Westport:

Th4 McNam4rG Strategy and tM VNtll4m War: ProflrGm B~eting

in the PenkJlfOIl, 1960-

Greenwood Press, 1978),38,66.

2. Michael Beschlosa, The CriBia Yea •.•: Kennedy and Khrushchev, 1960-1963 (Edward 1991), 23-4.
3. CCH Series VI.N.N.l.l.

Burlingame Books,

4. Kennedy Library papers, CCH Series XVI;Transcripts ofvideotapes offive former directors, CCH collectioD ". 5. ·CCP, FY 1964-1969," in CCH Series VI.A.1.7.;Kirkpatrick Report, in CCH Series VI.C.l.32. 6. NSAlCSSArchivea, ACe 39741, H03·03l1·4.

HANDLEVlA ~_N6'ft(]~EASABLE

ONTROL SYSTEMS JOINTLY

TO FOREIGN

TOP SECREt UMBRA

332

DOCID: 523682

REF ID:A523682
Withheld from public release Pub. L. 86-36 l=eP SECRET UMBRA

7. "Management Board Report,~ in.CCH Series VI.B.3.11. 8. Ibid. 9. Ibid. 10. Ibid. 11. Ibid. 12.1 I(NSA: U.S. Cryptologic History Series, Special Series, n.d.); "History of the Closure of PACSECRGN, 1972," in CCH Series VI.H.H.12.3. 13. Tranacriptofvideotapes 14.•... 1 of ..• ,~NSAICSSArchives, 39471, H03-031l-4.

...• 1"The Consolidated Cryptologic Program and its Predecessors, 1957-1965,~ in CCH collection.

15. CCH Seriee Vl.I.3.1. 16. Harris,83-84.

18. NSA Retired Recorda, 43852,73-252. 19. Harris,. 93-94. 20. Ibid. 87-97. 21. Ibid. 107-118. 22. Shirin Tahir· Kheli. TM UniUd Sm", and Pallist4ll: TM ElJOlutiDi&.of aIl1n/llUTI« RelatioMhip. Studies of Influence in International Relations, Alvin Z. Rubinstein (ed.) (New York: Praeger, 1982). 23. Ibid. 24. NSAlCSSArchivea,ACC 29543, CBTK 7l;ACC 31435N CBDB 62. 25. NSAlCSS Archives, ACe 9734X csos 42. 26. NSAlCSS Archives, ACC 28650, CBTK 51; ACC 29540,CaTK 71 27.

c::=J Manuacript.

28. James E. Pierson, "History of the United States Air Force Security Service, Fiscal Years 1960-61, Part IV: Systems Development," available at AlA, Kelly AFB, Texas; oral interview with Lt Gen (Rat) Gordon A. Blake, by RobertD. Farley, 19 April 1984, NSAOH7-84; NSAlCSSArchives, ACC 1l219,GII-0205-1; ACC 27741,GI40306-5;1 I"NSA's Participation in the Research and Development ofthe 466·L System, 1957-1964," in CCH Series X.H.8. 29. CCH Series VI.BB.L10;e::::J manuscript available in CCH. 30. NSAfCSSArchives, ACC 11436, G11~205:1; ACC 14947, HI8.0608-6; Pierson, "History." 31.c=]ACC 11436,Gll-0205-1;ACC 11218,Gll·0205-1;ACC 112i9,Gll-0205-1;Pierson. Withheld from public release Pub. L. 86-36

32. NSAlCSSArchives,11436,G11.0205.1;c=J 33. NSAlCSSArchiv8s, ACC 11219,Gll-0205·1;CCH Series VI.BB.1.l.

HANDLE V ---~'OT

INT CONTROL SYSTEMS.JOINTLY RELEASABLE TO FOREIGN

333

1'6P SECRET UMBRA

DOClO: 523682
·1=OPSECRET l:IMBRA

REF IO:A523682

34. "Historical available

Swnmary

of United States Army Security Ageilcy and subordinate

Unita, Fiscal Years1968-70,~

at Hq INSCOM, Ft. Belvoir, Va.

35. CCH Series VI.CC.!.1. 36 .. USAFSS,"A CCH Series X.J. 37'. USAFSS," A Special Historical Series X.J.3.1. 38. CCH Series VI.CC.1.!. 39. "A Special Historical Study ... ," NSAlCSS Archiv8S, ACC 26289,CBOL 78;ACC 39741, HO·311·4. Study of the Advisory Warning Program, July. 1961-December 1964," inCCH History of the USAFSS Airborne SIGINT Re<:onnaissance Program (ASRP), 1950-1977," in

40. NSA Retired records, 43852, 73·252; 42068, A66-77; NSAlCSS Archives, ACC 30932, CBOD 68. 41. Kennedy Library papers; "A Special Hiatorical Study .... " Program ••. ,"

42. "A Special Historical Stlldy ..• ,""A Hiatory of the USAFSS Airborne SIGIN'!' Beecnaatsaance ACC 15157, CBRE 67. 43. "A Special Historical Study ..• ," ACC 15157, CBRE 67. 44. George F. Howe, Teel&nic41 Reaearch Ships, 1956-1959. Meade: NSA, n.d.). 45. Ibid. U.S. Cryptologic History,

Special Series,

#2 (Ft.

46~Ibid.
47. Ibid. 48. Ibid. 49. Ibid.

50. CCH Series VIII 11, Cuban Missile Crisis: NSA/CSS Archives ACC 9650X, CBDB 43.
51. Donald Wigglesworth, "Cuban Missile Crisis," CryPtolollicQUGrierly,Spring 11; CCH SerieaVlII 11. ACC 97 46X, CBDB 42. 94, Vol. 13, No.1.

52. NSAlCSSArchivesACC24907,CBOK
53.1

lII,

74; NSAJCSS Archives "Cuban Missile Crisis."

ACC 39471, H03·0311.4;

54. Wigglesworth,

55. NSA Retired records, 42068, A·077 •.
56. Beschlosa, 310. 57. Frank Sieverts, "The Cuban Crisis, 1962; in the Kennedy Library papers, CCH Series X VI. 58. Wigglesworth. 59. Wigglesworth. 60. Wigglesworth. 61. Series VIII.11.;oral history interview with Juanita M. Moody, 16 June 1994, by David Hatcb'L..1

Withheld from public release Pub. L. 86-36

...J

I

Land Colin Burke, NSA OH 32-94.

HANDLE VIA ~_..M~!t!1[]EASABLE

CONTROL SYSTEMS JOINTLY TO FOREI

lOP SECRET UMBRA
.\

334

-_.,

--'-----

DOCID: 523682

REF ID:A523682
1012 SiiCRET l::IMBAA

;. Oral history interview

with David D. Lowman, by Robert Farley, NSA OH 13-80.

J. CCH Series VIII.H.
~. Serin . 992). 65. ClAlORa, 0011 Staft'Study. ·Cuba 1962: Khrushchev's Miscalculated Risk," 13 Feb 1964, available Miscalculated through VUI.ll.; Mary McAuliffe (ed.), CIA DocurMnU on the Cuban M;',ile Cm;.. 1962, (Washington: CIA •

the CIA history ata1f.CIAJORR, 0011 StatfStudy. 66. Series VlII.l1.; 67. Serie. VIII.l1.;

"Cuba 1962: Khrushchev's

Risk." 13 Feb 1964.

NSAlCSS Archives, ACC 9750X, CBDB 43. Dino A. Brugioni, Eyeball to Eyeball: The IMide Story of the Cuban Missile CrUis (New York:

Random House,1990). 68. Series VIII.U. 69. Robert A. Divine (ed.), Tile CubonM;'sile CriB;' (Chieago: Quadrangle 70. Brugioni. 71. Oral interview Harold E. Daniela. with! l3 April 1980, by Robert Farley, NSA OH 10-80; Oral interview in CCH; Brugioni, 123-27. with Books,1971), 7.

16 AllI'Jat 1988, NSA OR 10-88; discuaaiona with Juanita

Moody, 9 October 1992, by David A.

Hatch; NSAlCSS Archives, ACC 38753, 102-1; George Howe notes available 72. Brugioni, 84; Volenick interview.

Withheld from public release Pub. L. 86-36

73. Sieverta, 17-18.
74. Quoted in McAulift'e, 93. 75. McAuliffe, 124-25.

76. 8rugioDi, 201: Sieverts. 77.1 1interview. wit~ ~15 January 1987, by RobertFarley and Tom JohnBon, NSA OH 3-87.

78. Oral interview 79. Kennedy

Library

papera, CCH Seriel XVI. ACC 24907, CBOK 11. telephone in~rview with!L. ....IlofNSA.29 December 1993.

80. NSAICSS Archives, 81. Dinterview; 82.

Withheld from public release Pub. L. 86-36

83. CIA, "Cuba 1962: Kruahc:hev's Miscalculated 84. Brugioni, Tnaa, 391, 399-400; Oral interview

Risk." City, •

with Lt OeD (Ret) Gordon A. Blake, 19 April 1984, Universal

by Robert Farley. NSA OH 7-84; Howe notes in CCH.

85. 8rugioni,462. 86. Telephone interview withL.1 --'129 Octobe~ 1993;1L..-_---'1 interview; Hersh article in Washington

Post,11 Oct 1987. HI.
87. See Mark Kramer, "'ractiul Nuclear Weapons, Soviet Command Authority, and the Cuban Missile Crisis,"

Withheld from public release Pub. L. 86-36

Cold War Int.rnationaIHi.tory
88.1 1Series

Projei:tBulktill, Fa1Il993, 40-51.
Archives ACe 9760X, CBDB 43.

vm.u, NSAlCSS

ass

TOP SECRE:r -

tlMBKA

---_

.

DOCID: 523682

REF ID:A523682
TOP 5&CRET UMBRA
89. Daniela interview;
interview.

LoWU\8.11

90. Series vm.l1; NSAlCSS Archive8, ACC 24907, CBOK 11. 91. CCH Series VlIl.l1. 92. Ibid.
93. NSAlCSS Archives, 97S0X, CBDB 43; Oral interview Farley,NSAOH with Harold L. Parish, 12 October

1982, by Rot:

20-82.

94. George Howe'8notee in CCH. 96. Seriea vrn.ll.: NSAJCSS Archives, ACC 24907. CBOK 11: KenDedy Library Papers in CCH Series XVI.
96. CIA, "Cuba 1962: Krushchev'a Miscalculated Rislt."

97. Morrison letter to Carter, 31 Jan 1969, in CCH Series VI.C.1.27. 98. Moody interview.

Tap SECRET I:JMBItA

336

DOClD: 523682

REF lD:A523682
-TOP SECREI UMBRA

Chapter 9 The Post-Cuban Missile Crisis Reforms

TheCCP reviewprocessbaa.in the judgment'ofNSA officials,becomea vehicleforvarious OSD and outaideDoDelementsto manipulatsresources assignedtothe Director,NSAanda forumfor the encouragementofopponentsofe centralizedSIGINTstructure.... NSAinternal memorandwn,1967

IntelIigencereform did not, of course, begin after the Cuban Missile Crisis ~ significant soul-searching had begun after the Bay of Pigs. But the events of 1962 made the matter more imminent. Kennedy demanded a responsive intelligence system to get him information when he needed it. The emphasis was on speed. At CIA, the Bay of Pigs e~ded the intelligence careers of both DCI Allen Dulles and Richard Bissell, who had supervised the invasion attempt. Owing perhaps to the rather small SIGINTinvolvement, it did not end careers at NSA, but it definitely hastened the pace of centralization. PFIAB, which had been told to get the intelligence house in order by a disturbed president, reported in June of 1962. Its S1GINT emphasis was on further centralization of the system under NSA, PFIAB wanted NSA to corral fugitive SIGINT efforts and to exercise strong central management over those it already headed. Noting that ELINT centralization directed in the 1958, NSCID 6 had been a failure, it suggested ways that NSA could gain control of the process. It specifically wanted a National ELINTPlan with stern NSA management.of resources under: the plan.! In 1964 it reported on progress over the two-year period. The board was intensely unhappy about ELINT, which remained frustratingly decentralized. As for internal NSA management, PFIAB made several technical recommendations for strengthening the research and development process, for rationalizing SIGINT requirements, and for establishing an operations research discipline at NSA similar to that which existed at the DoD level. PFIAB especially wanted NSA to expand its influence over the cryptologic research and development process then performed by the services. The SIGINTeffort was expensive, and PFIAB felt that a stronger NSA could reduce duplication and bring down

~~~

,

Studies of the cryptologic system: in the 1960s by the PFIAB, by DoD-level committees, .and by the Bureau of the Budget all came down heavily on a more centralized process. The emphasis was always on doing more with less, but in fact, cryptologic budgets increased steadily during the decade. What happened in practice was that NSA did more with more.

HANDLE VIATALE

L SYSTEMS JOINTLY ASABLE TOFOREIGN NATIONA

337

-

TOP SFCRiT l:;IMBRA"
,

DOCID:523682\
1'8'

REF ID:A523682

'!eRn UMBRA

The National Security Agency was only too happy to oblige. Beginning in the early 1960s, NSA management began working·on a plan to centralize cryptologic operations in the United States. Field operations would be reduced, especially at the theater level; SCA processing centers would be phased out; and, using the new digital data links sprouting up in the DoD communications system, data would be brought back to the States for· processing. Using the PFIAB's recommendations as a hammer, NSA eould achieve a degree of centralization dreamed of, but never achieved, in earlier years. S The Dilemma of Centralization Whenever there is a major foreign policy crisis, the response of an administration is usually to tighten up. The Kennedy administration responded to the Bay of Pigs and the Cuban Missile Crisis with a series of actions which resulted in an ever-tighter centralization of the intelligence mechanism. The effect on the SlGINT system was to further centralize a process which had been on a course toward centralization ever since World War II. But centralization meant the same both upwards and downwards. As NSA further strengthened its hold on the eryptologic system, McNamara got a firmer grip on the Defense Department. including NSA. The Agency had never had to answer in detail to anyone e.boutits program - certainly Graves B. Erskine's miniscule staff in OSO could not police a system composed of tens of thousands of cryptologists working in over twenty countries, with a budget of hundreds of millions of dollars. But McNamara did away with. 080 in 1961, and in its place he put the director of defense research and engineering (DDR&E), Dr. John Foster, in charge of cryptologic matters. (The post of DDR&E had been created by the Defense Reorganization Act of 1958, as a response to the Sputnik crisis.) Foster in turn delegated the job to his deputy, John Rubel. The reform measure was accomplished without even contacting Admiral Frost at NSA.' McNamara brought with him a team of "whiz kids» and a whole new management superstructure. Instead of dealing with just Graves B. Erskine or just John Foster or just John Rubel, Frost suddenly found himself talking to all sorts of subalterns like an .aeststant secretary for comptroller, an assistant secretary for management, an assistant secretary for international security affairs, ad infinitum. Each one felt he owned a piece of . NSA. None was experienced in cryptology, and few managed to attain any appreciation for the arcane business of breaking and protecting codes: and the flip side of the coin was increasing OSD control over NSA. McNamara's staff bore down hard on the Agency's programs, placing each one under a microscope. As the CCP made its annual pilgrimage through the OSD machinery, increasing numbers of officials came to question cryptologic

lw.nLE':"~'~Y
__
TOP SECRET \;IMBM

LEASABLE TO FORE
338

DOClD: 523682

REF lD:A523682
TOP SECRE'f tJMIRA

programs. NSA's existence became noncryptologists on MeN amara's staff.

a constant

battle

to educate

the legion

of

Cost control was a dramatic example of the dilemma that successive directors of NSA administration had always found themselves in. Late in the 1950s the Eisenhower introduced the concept of centralized cryptologic budgeting, in which the SCAs would send their annual budget recommendations to NSA, which would consolidate the inputs, add its own, and produce what came to be known as the CCP. This changed NSA's role from that of coordinator to centralizer. The SCAs were now beholden to NSA Cor their very livelihood. When the Agency looked down its nose at a major SCA procurement, as it had with the Air Force's 466L program, that program was in trouble.5 The new CCP was not fully implemented until fiscal year 1961, but in the two years in which it was being phased in it had already changed the landscape significantly. 8 ' McNamara arrived with a new cost management system called the Planning, Programming, and Budgeting System (PPBS). There were, under PPBS, nine major military programs. Cryptology, which began in Program Seven (general support), was soon switched to Program Three, general-purpose forces, where it stayed. Within each program there were five cost categories: R&D, procurement, personnel, a&M (operations and maintenance), and military construction. The cryptologic budget itself was in turn divided into fUty-six cost categories, called subelements. All eryptologic expenditures, both for NSA and the SCAs, had to fit into one of the flfly-six. This new process gave NSA substantial power. The subelements were managed at NSA, and the SCA budgets had to be structured and submitted to the subelement managers for their review .. After DDR&E. and the secretary of defense approved it, the plan became the approved cryptologic force level. NSA could then change the mission of each cryptologic component, right down to the collection site, to fit the program. The entire process resembled a gigantic funnel, in which the most significant narrowing took place at NSA. It effectively ended SeA independence. NSA's influence came to extend even to the equipment on collection positions. In a spate of technical control never before achieved, NSA wrote a document (TECH INS 1037) which dictated what equipment must be on each position to make it confcrm to the' program. It was up to the SCAs to get their positions in line with the edict. _ Most directly involved were Jack O'Gara, who managed the cryptologic program at the aSD level, and Dr. Eugene Fubini, who became deputy director for research and engineering under McNamara. O'Gara had a cryptologic background, but Fubini was a scientist. For the first time, the director's cryptologicstaff found itself arguing individual line items at the OSD level with people who wanted to know why it was necessary to have more than one position targetted on the North Vietnamese Navy or why two positions at different locations remained targeted on the same case notation. NSA was forced to provide proprietary personnel and facilities information to GSA (General

HANDLE VIA TALENT SYSTEMS JOINTLY ~NIQO!IJl:£.bB1l:lS1lJ:IILE TO FOREIGN NATION

339

DOCID: 523682
TOP SECRET !:IMBRA

REF ID:A523682

Services Administration) and the Bureau of the Budget, and the Agency frequently discovered that outside organizations were auditing NSA's operations without its concurrence, or even, in some cases, its knowledge. In 1967, Director Marshall Carter charged that "... the CCP. review exercise became a means for various DoD elements to manipulate resources assigned to the Director, NSA .•. an undesirable feature of this Office of the Assistant Secretary of Defense for Administration (OASD [A] ) review is that these officials are not SIGlNT-oriented and they frequently make unrealistic comparisons of agency positions to those in the Defense Agencies." Each director in the 19605, from Frost to Blake to Carter, claimed that McNamara's OSD staff was micromanaging NSA.7 Everywhere NSA turned, there were new restrictions on its independence. Allen Dulles's replacement as DCI, John McCone, did not share Dulles's aversion for centralized management of intelligence resources. McCone moved aggressively to place the extensive Defense Department intelligence assets under CIA's general coordination. His newly created National Intelligence Programs Evaluation (NIPE) office was an early attempt to establish an intelligence community staff; it gave the DCI a way to inventory and evaluate all intelligence programs. He never achieved control of DoD intelligence budgets, but under him CIA was clearly headed in that direction," A New Director

The hard-driving McCone was partly responsible for the relief of Admiral Frost as director. Frost was not a driver. His soft-spoken manner and laid-back style were not for McCone. He did not have Canine's "presence," and at USIB meetings would speak in a voice so low that he could scarcely be heard. One very senior NSA official who worked directly for Frost said, "He was a professicnal SIGlNTer, he knew about SIGINT,but somehow or other he did not project that he was a knowledgeable, dynamic leader for the SIGIN! effort ." Nor did he fare well with McNamara and his staff. People like McNama~a and Fubini expected clipped, precise answers to specific questions, and when they did not get them, beganto look

Gordon Blake

HANDLE VIA LEASABLE TO FORE

ROL SYSTEMS JOINTLY

TOP SECRET lIMBRA

340

DOClD: 523682

REF lD:A523682
lOP SI!CRET YMlRA

elsewhere for a director. Frost was relieved on 30 J,une 1962, more than a year before his term was up, was reduced in rank by one star, and was placed in charge of the Potomac River Naval Command. Such was the ignominy that Robert McNamara could visit on someone in his personal doghouse. 9 Frost's relief, Lieutenant General Gordon Blake (USAF), had shuttled between air operations (he was a command pilot) and communications assignments his entire career. His only intelligence assignment had been as commander of the Air Force Security Service from 1957 to 1959, but that had at least given him an introduction into the field which Canine. for one, had lacked. Blake, like Samford, was exceptionally good at personal relations and was very highly regarded in Washington. He had been in the job only three months when Cuba erupted, and he established high marks in the White House during the crisis. It has been saidthat no one disliked Gordon Blake, but even as smooth an operator as he still acknowledged difficulty getting along with McNamara's staff.IO
NSA's Community Relationships

USIB, which in 1958 had become preeminent in intelligence affairs with the disappearance of the Intelligence Advisory Committee, became honeycombed with committees in the 1960s. Instead of dealing solely with COMINT, as had USCIB, it dealt with general intelligence matters, and it assigned SIGIm to the dual COMINT and ELINT committees. By the time Kennedy took office, USIB already had twenty-six committees, and most of the work was done there rather than in a committee of the whole. In 1962 John McCone combined the COMINT and EUNT committees into a new SIGINT committee and chose John Samford to head the new panel. Samford was an ideal choice; he lent prestige to the committee - never before had such a senior person been chosen to head a USIB committee. Samford spent a lot of time trying to rationalize SIGlNT requirements, and it was he who first proposed that COMINT requirements be related to CCP line items. His overhaul of the antiquated requirements system in place paved the way for a new system introduced in the mid-1960s, the Intelligence Guidance for COMINT Programming." Throughout this period the day-to-day influence of USIB became more pervasi ve, and it operated as yet another check on NSA's iridependent authority. The dark days of the Canine-Dulles feud were over, but that by no means ended the problems between the two agencies. CIA still had intercept operations spread throughout the world, and by 1970 it was reputed to have
Withheld from public release Pub. L. 86-36

I E.O.

13526, section 1.4(c)

I

341

'F9P SECRET YMBRIe

------,--

-

-

--

DOClD: 523682
TOP SECRET UMBRA

REF lD:A523682

I E.O. 13526, section 1.4(c) I
In 1966 Huntington Sheldon of CIA studied CIA SIGINT operations to determine the proper size and to allocate funds. He found·that CIA had people doing SIGINT; with a budget of The result, which bec e n as the Sausage Study, was t Irst to document the truly significant CIA stake in SIGINT.1S 1961 a new competitor arose. The Defense Intelligence Agency (DIA) was created to centralize defense intelligence matters. DIA began life with a headquarters in the Pentagon but with subordinate offices scattered all over Washington. Arlington Hall's A and B buildings housed much of the efi'ort. The fragmented physical situation in which DlA found itself came to symbolize its participation in the intelligence business. DIA had stepped into a department whose intelligence was fragmented and decentralized and whose intelligence programs were managed under feudal baronies with great power and internal cohesion. None was more powerful than NSA. DIA began churning out intelligence reports and estimates in competition with the existing organizations. But ultimately the organization had to carve out, its own unique turf, and one of the first areas it chose to invade was the private game preserve of SIGINT. In 1963 DIA proposed that it, rather than NSA, should run the COMINT dissemination system. The next year it wrote a draft directive which would have the director of DIA become the principal advisor to the secretary of defense "concerning the security, use, and. dissemination of COMINT." DIA would take over the SSO system, including the communications apparatus. McNamara accepted the proposal, and the SSO systems of the SCAs were turned over to DIA in 1965.14 The post-World War II SSO systems managed by the SCAs had long since become more administrative than substantive, and by the time DIA got hold of them, they were serving as little more than communications and security managers. In their place, NSA was in the process of establishing a network of SIGINT representatives. This network consisted of two components. The first was the official representation system, which NSA managed at Unified and Specified levels, and the SCA's represented SIGINT to the component commands. This system took some working out, and resulted, especially in the early (Post1958)years, in turf battles between the SCAs and NSA. The second type of organization was the CSG (see p. 264). This was where the interpretive function was performed, and it closely resembled the functions performed by the World War II SSO network, minus most of its dissemination control (Le., housekeeping) features. DINs demarche into the SSO field accelerated the creation of esGs. 'The rust CSG. called NSAEURIISS, had been around since the late 1950s, and it served as a model for

TOP SECRET t;J MIRA

342

OOClO: 523682

REF lD:A523682
+QP SECRET tJMBRA

others. In 1964 Brigadier General John Morrison, NSA's representative in Hawaii, heard about .NSAEURIISS and journeyed to Paris to see how it worked. He liked what he saw and created what he called the NSAPAC NOG (NSA Pacific operations Group). The idea of having CSGs spread quickly and was incorporated into JCS Memo 506-67, which became the bible for SIGINTsupport to military organizations. By 19'74 there were eight CSGs, with two additional CSGs in the process of being formed.'! CSGs became effective because of the access they had to the SIGINTsystem:. To a great extent they depended on the growing network of Opscomms to get them that access. Every CSG began life with an Opscomm Circuit to NSA. With it, the CSG could get quick and accurate information to t~e supported commander .18
ELINT (Again)

While COMINTwas coming under increasingly centralized control, ELINT was still fragmented. A study commissioned by McNamara in 1961 concluded that little real control over EUNT had been instituted in the three years since NSA had been given the charter. Theater commanders were still running their own ElJNToperations, and in many cases they were proliferating processing centers without coordination or control. Their Third Party EUNT relationships continued unabated, and their collection assets were pumping low-quality and often inaccurate ELINTinto the processing system, unaffected by any sort of quality control. The study group concluded that there should be a strict apportioning of ELINTassets between the U &S commands and NSA, and. that the Agency should institute stringent technical controls over all DoD assets. NSA should take control of all Third Party ELlNT arrangements. Theater-level EUNT processing centers should not be established willynilly, but should conform to some overall plan. That plan should be coordinated by NSA, which would accept inputs from the military commands and crank out the final product. It would be called the National ELlNTPlan (NEP). But the bottom line was that it would have no teeth.' Coordination, not direction, would be the mOdus operandi. 17 A National EUNT Plan finally emerged in 1966, after several years of bureaucratic struggle and false starts. It marked the first real attempt to organize and control ELINT; but since it was not directive, it had only a minimal impact on the actual course of DoD) EUNT. Meanwhile, NSA and DIA tried to negotiate a system of ELINTtasking which would conform to DIA's new charter to centralize all DoD intelligence requirements. They worked out a complex system in which all parties to the National ELINTPlan (including CIA) would forward ELINT requirements to DlA for registry. NSA would maintain a complete list of all ELlNTcollection assets (including those that the Agency did not control) and would assess the capability of relevant assets to satisfy each requi,rement (called a

HANDLE VIA TALENT BLE

YSTEMSJOINTLY TOFOREIGN NATIO 343

TOP 5!iCRiTUMBR'

--------_._-

--

DOCID: 523682
TOP SECM,. UMBRA

REF ID:A523682

SICR, Specific Intelligence Collection Requirement). NSA would then return the . requirement to DlA, which would task the appropriate U&S command, while NSA would task assets under its own control.1a Attempts to rationalize theater-level EUNT processing centers were only semisuccessful. Proposals for NSA control were opposed by theater commanders and thus went unimplemented. The best NSA could achieve was to appoint a technical assistant to the director of the theater processing center and to transfer CCP billets and NSA people into the center to help maintain quality control, as was done in Europe, in the Pacific, and in the Atlantic Command. a Successive directors felt that the job of managing ELINTwas simply too much for NSA. General Blake felt that "a National ELINTPlan [was} neither desirable nor practical." Given the job of writing the plan, General Carter found that NSA was not set up iritemaUy to manage such an effort, and he had to create an ad hoc group, which he called Dagger, to write it. Looking back in later years, Carter called the NEP "unworkable." Difficult relationships with the Unified and Specified commands, disputes over ownership with DrA and CIA, and internal dissension over how the efTort should be organized within NSA all contributed to the sense of frustration. 20 News from the ELINTfront continued to be gloomy throughout the decade. In 1964 PFIAB launched a rocket at theater ELiNTcenters: "Meanwhile new centers from ELlNT analysis are being established without coordination, terms of reference, or technical guidance from our proven competency in established programs." CIA, which had retained a tenacious hold on telemetry, opened a new telemetry center called FMSAC (pronounced "Foomsack": Foreign Missile and Space Analysis Center), which became, as was intended, a direct competitor with NSA's efforts. EUNT requirements were in a chaotic state, and local commanders were constantly confusmg the situation with overlapping demands."
.
\'

The 1968 Eaton Committee (see p. 479) found that the NEP was a marginally effective document negotiated to compromise among various competing power centers. NSA had never been given tasking authority over many ELINTcollectors - SAC airborne assets came immediately to mind. There was no central budget review process tor ELINTand no way to deconflict competing assets. There was no effective quality control, resulting in parametric garbage cluttering disparate databases managed by widely separate organizations that did not talk each other. Despite the 1961 recOmmendation that NSA . should take over Third Party ELINT. nothing of the kind had taken place, and those relationships were still being managed by CIA and the .theater-Ievel component commands, as well as by NSA.22 No wonder NSA directors were so ambivalent about the task which NSA had shouldered for ten years running.

to

HANDLE VIA TAL
--~'flT

CONTROLSYSTEMSJOlNTLY

RELEASABLE TO FOREIG 344·

l8P SECRETtlMBItA

DOClD: 523682

REF lD:A523682
Tap SECRETtJM11tA

DEFSMAC

I E.O. 13526, section

1.4(c)

I

Withheld from public release Pub. L. 86-36

Occasionally the demands of centralization resulted in measurable steps forward, relatively unaffected by bureaucratic rivalries. The 1964 creation of the Defense Special Missile and Astronautics Center (DEFSMAC) was such a moment.

I

A41 had two round-the-clock operations centers. The A41 Operations Center (Opconcen),located next to the A41 officeson the third floor of the operations building, was the nerve center. It had Opscomma to the primary warning sites and had established a tipoff system so that warning information could be flashed back to A41. That organization, in turn, alerted I that were standing by. By 1962 the Opconcen had six Opscomms to collection sites. It was further linked by Opscomms to customers, notably NORAD (North American Air Defense Command, which had responsibility for tactical warning of missile .launches) and the Washington-area organizations. -

I

I

I

Downstairs in the computer complex was the Sigtrack center. The Sigtrack center was in close touch with the Opconcen, but, although there were plans to consolidate the effort, they were stiUphysically separate." I E.O. 13526, section 1.4(c) When the consolidated facility, the Space and Missile Analysis Center (SMAC), was created in January' 1963, it had Opscomms to sixteen facilities, plus the customers. Severaldifferent organizations had mounted twenty-four-hour operations, but SMAC and NORAD were far and away the major players - others simply fed off the information generated through the air defense and SIGlNT warning systems.24 The disorganization in the missile warning business led, in 1963, to a full DoD-level review. The team surveyed the entire problem, talked with every organization involved, and made field trips to warning facilities like SMACand NORAD (in Cheyenne Mountain, outside Colorado Springs). They found that NSA had the only coherent, centralized program, and, at the suggestion of A4, they took SMAC as the organizational model for.a new, combined facilty. It would be called DEFSMAC, would be located at NSA, and would be jointly staffed by NSA and DIA people. The chief and deputy chief would be selected jointly by DIRNSA and the director of DIA. Because most inputs were SlGlNT-based, NSA
Withheld from public release Pub. L. 86-36

345

lap SECRETYMBRA

DOCID: 523682
TOP SECREI UMBRA

REF ID:A523682

~ '-' ~ ~

possessed virtually the sum total of technical expertise. / DIA was charged with integration, 'reviews, and nontechnical analysis of findings. DEFSMAC would have the same inputs, through the same Opscomm net, that SMAC had had. But because its official charter was established at the Department of Defense level, it carried with it far more authority than had SMAC.DEFSMAC had tasking and technical control of all DoD intelligence collection activities directed against foreign missile and space activities. It provided technical support, including tip-offs, to all DoD missile and space intelligence collection activities. The only exception to its virtual blanket authority was that it could not launch airborne collection platforms on its own - that required a JCS go-ahead. 25 At its creation in 1964, DEFSMAC INSA billets, to twenty- . three for DIA. Its first director (and all thereafter) was an NSA official, Charles Tevis, while the deputy was a DIA official. 28 The Advent ofthe Command Center

.§ . had I

'E

~ ~ ~

Withheld from public release Pub. L. 86-36

Present-day NSOC and the. plethora of round-the-clock watch operations that Agency workers know evolved slowly over a.long period of time. The key date in its evolution was October 1962 - the Cuban Missile Crisis. But the development began years before that. AFSA had had a shift operation, established originally to monitor developments in the Far East during the Korean War. It was part of AFSA-25, the organization that dealt with Charles Tevis customers, and, within that organization, the publications and distribution branch. Manned originally by a staff of two junior officers and several analysts and enlisted communicators per shift, it scanned outgoing messages for release and maintained a liaison group to answer requests for information. Aft.erNSA was created, it became known as the Prod Watch Office, or PWO, but proposals to give it executive powers were scotched whenever they came up. In 1954 it became responsible for the director's daily inte1ligence briefing, and when the Critic program was created in 1958, the PWO insured that all Critics had the correct external and internal addressees. But when real horsepower was needed, the PWO called in day workers.

H~~~;=:=eY
TOPSECRET UMBRA 346

OOClO: 523682
· I

REF lO:A523682
. 1

E.O. 13526, section 1.4(c)

Withheld from public release . Pub. L. 86-36

TOP SEERET UMBRA

The COMSEe organization also had a watch office, charged specifically with responding to reports of compromise. Although small, it did a good job of quick response, and over the years kept potential compromises from becoming major hemorrhages." Through a succession of reorganizations, the PWO became the PIWO (PROD Intelligence Watch Office), and more civilians were added. In 1962, the last year orits life, the PIWO consisted of! I people, ten of whom were civilians. But its functions still remained procedural rather than substantive. NSA's method of handling round-the-dock responsibilities bespoke the way that the organization viewed itself. NSA thought of itself as a long-term reporting shop, a concept which had become completely outmOded by the Soviet strategic threat and the role ofSlGINTin warning of that threat. The vision of NSA as Sleepy Hollow ended abruptly in October 1962. The new director, Gordon Blake, realized that he did not have a command post, and his assistant director for operations, Major General John Davis, created one during the middle of the crisis. The chief of the new shift operation was known as the SNOO (Senior NSA Operations Officer), and he hadDanalysts on duty. The original command post was located close to the PIWO and the communications center and had telephone connectivity to both. Z8 After the dust settled, General Davis decided that he could not continue to operate on an ad hoc basis, and early in 1963 the Command Center was made permanent. With eight bays of space and ,~O,OOO, the reporting staff headed by I I and I Ifashioned a command post look-alike, with situation maps, multicolored telephones, and pony circuits from the communications center. (This came to include a KY-3, which permitted secure voice contact with the White House, CIA, DlA, and several. other Washington consumers.) The PIWO was wiped out and the bodies transferred to the Command Center. Although the Command Center became a nerve center of sorts, it never became what its creators had hoped. To begin with, the SNOO did not represent the director; he only represented the assis.tant director for production. Executive decisions above Production. required that other deputy directors be called in. Second, even within PROD the Command Center was to some degree emasculated. This owed to the refusal of the analytic groups to contribute skilled analy~ts. The Command Center wound up with a personnel cadre, but the real power remained within the analytic groups themselves, each of which, over a period of years, established various ~atch operations. These "puddles" (as they were called) tended ~ arise during crises and simply continue. Thus it was that the B Watch Office was set up in 1965, when Vietnam heated up, and the Bl Watch was established as a result of the Pueblo capture. G Group established no permanent watch but continued to call analysts to duty during crises. 29 Regulations governing the Command Center carefully circumscribed the authorities of the SNOO who, after all, was only a grade 13 or 14. He monitored the Critic program,

HANDLEVlA'l'
BLETOFORE

mROL

SYSTEMS

JOINTLY

347

. TOP SECRET l;JMBAA

DOClD: 523682
1'0' SECRETtJMBRA

REF lD:A5236~2

and could change distribution, but he could not change the text or issue a new report. He could not call a SIGINT readiness, did not have direct connectivity to field sites, and could not modify field site collection instructions. A and B Groups had "coordinators" in the Command Center, but whenever a problem arose, either referred the matter to one of the "puddles" or called someone in.so Centralization of Theater Processing

As the Vietnam War heated up, Robert McNamara began looking for money. He put considerable pressure on all DoD elements to become more efficient. In the early 1960s Gordon Blake was under considerable pressure from McNamara's staff'. According to them, the SIGINT system was too big, too costly,too spread out, and inefficiently organized. If McNamara needed money, they thought they could sweat some of it out of the SIGINT budget. And anyway, they believed that centralization was inherently good as well as cost-effective. MeNamara's point man in this effort was Dr. Eugene Fubini, In 1964 Blake was directed to take a close look at theater processing. Fubini believed that there were too many theater processing nodes, especially in Europe,and so NSA turned its attention to the European theater. Studies in that year turned up quite a complex of centers spread acro~s Germanyl

I

The Air Force had centralized SIGINT processing at Zweibrucken, which by 1964 had become a complex of over D people, IBM 1401 processors, and Opseomm eonnectivityD over Europe I I The reporting operation alone was the busiest and largest reporting center ever put together up to that time. It was the hub for timely reportingl Ian absolutely irreplaceable asset.

I
Withheld from public release Pub. L. 86-36

The Army operation, centralized in Frankfurt, had a very different focus. Its COMINT Processing Center (CPC) concentrated on preliminary processing of the increasing volumes a ~ __ ..L..:.A.:::;SA refused to join ~ and it maintained its own development effort in .NSA's theater focal point was also in Frankfurt, where NSAEUR had put together a processing effort called JNACC (Joint Non-Morse Acquisition Control Center). I I.

I E.O. 13526, section 1.4(c) I
HANDLE VIA TALENT CONTROL SYSTEMSJOIN'I'L Y ~-"'r 'T RELEASABLE TO FORE1GN NA Withheld from public release Pub. L. 86-36

l=OPSiCA'T UMBRA

348

.DOCID:

523682

,---_-=--=RE=.:F=------=;ID : A52 3682
Withheld from public release Pub. L. 86-36
'FOP SE(RET tJMBRA

E.O. 13526,section lA(c)(d)

In July 1964, under continuing pressure from Fubini, Blake named Benson Buffham to chair an ad hoc committee to produce an austere SIGINT posture in Germany. This group wrestled with the problem of the competing power centers in Frankfurt and Zweibrucken, and it finally came down on the side of Frankfurt. But the committee went much further. It decided that ultimately much of what was going on in Germany would be done at Fort Meade. The interim European architecture would close Zweibrucken and create two separate but closely related organizations I I in Frankfurt. The first, I I would take over theater processing operations The second, called L...---J, would take over the timely reporting functions then exercised at Zweibrucken. Manning for the new facilities would come directly from the hides of ASA and AFSS, with a significant NSA admixture.

I

I

The panel was looking at far more than reorganizing theater assets, however. It began to consider a longer-range plan of closing theater operations and moving them to Fort Meade. NSA would establish a high-speed (2400 baud, high speed for the mid-60s) data link from Frankfurt to Fort Meade. Frankfurt was clearly a way station on a much longer journey.34 The plan to close theater functions also included JNACC. NSA decided to establish a worldwide printer steering group at Fort Meade. Called the COC (Collection Operations Center), it functioned much like JNACC, interacting with field sites through.a network of Opscomms, When opened officially in 1969, COC began using a new reporting system, called The basis of reportin was a short reformatted re rt resembling a ----J The reports were formatted for computer input and formed a database on all printer intercept wo~ldwide. COC adjusted collection 'ofl llinks . based on the I Ireporting and daily contact with cryptanalysts in A5, the office o~ ~ It was not finally phased out until 1993. 56
L..-

Back in A Group, the planning committee came up with two schemes: Plan A and Plan B. Plan A assumed that precessing functions would be moved to Fort Meade but that basic timely reporting would remain in the theater, a~ and! I Plan B assumed that these centers would eventually be closed and the functions moved to Fort Meade. General Carter favored Plan A, but his staff favored Plan B. Ultimately; the reluctant director was persuaded to sign Plan B, and the residual organizations in Frankfurt were doomed.sa ' .

I

The adoption of Plan B required drastic changes in A3, the analytic organization responsible for the Soviet problem. A3 was basically a term reporting organization, but
r-~------'

E.O.13526,sectiO~~~
. HAm .••.••••••.•.•~LEC .

~

~
LSYSTEMSJOINTLY BLE TO FO .

Withheld from public release Pub. L. 86-36

349

Tap SECRET ijMBRA

OOClO: 523682 I
I uP SECRET tlMBIb\ E.O. 13526, section 1.4(c)

REF lD:A523682
Withheld from public release Pub. L. 86-36

under the new scheme it would have to split into two camps, the term shop (A7, material older than seventy-two hours) and the current shop (AS, material not yet seventy-two hours old). The current shop, A8, would have to pick up responsibility for a number of daily summary reports produced by Zweibrucken. More significant, it would have to create a shifi effort to monitor timely reports like spot reports and Critics. It would interact closely with the c=I which would retain some of Zweibrucken's reporting functions. Thd Iwouldbe an emasculated 1 1 retaining substantial authority for coordinating timely reporting on U.S. reconnaissance flights, but without the reporting or collection management authority that Zweibrucken had exercised. A3 would pick up some 1 1billets in order to mount the required reporting effort.S'7

csoc
The ABlA7 split was the genesis of a new organization, called the ,Current SIGINT Operations Center. CSOC, as it was usually referred to, was formed by Walter Deeley of A05 from a group of AGroup analysts and reporters who had been in proximity to, but not an integral part of, the Command Cen~r. Deeley believed that, by integrating processing computers with communications systems, he could create an analytic and reporting center .In which all activity was electronic. He later popularized this as his"paperless environment," a concept that was adopted when NSOC was created. Deeley planned toreterminate the! [reports from Zweibrucken to CSOC, but instead of the reports being dumped onto a Teletype Corporation printer, they would appear on computer screens, where analysts could manipulate them. A communications interface computer would be required to receive the incoming I I reports, sort them according to type of activity ; and route the sorted reports to analysts who were trained to watch different types of activity. CSOC would have the same reporting and collectionmanagement authorities that Zweibrucken had. Deeley wanted a new name for the tip-off reports, and he came up with the name KLIEGUGHT, which would be used into the 1990s. The computer Deeley selected was a Univac product, which was the best machine at the time for communications interface. The TIDE software system, which managed the KUEGLIGHT database and routed reports throughout CSOC, was written for the Univac computer." A8 was established officially in June of1967. CSOC guaranteed that would die. It was put into operation a year prior to 'I ----,1 and by the time Frankfurt was ready to assume Zweibrucken's reporting responsibilities, C50C had already proved it could do them. Real authority thus bypassed Frankfurt and went directly back to Fort Meade. Moreover, CSOC proved the feasibility of a global SIGINT view. Now there was a reporting center that had inputs from all SIGINT sources on the Soviet problem. Army, Navy, and Air Force data flowed into the new center, and CSOC could see the

c=J

I E.O.

13526, section 1.4(c) -

I
~
•.•••• -. _

,-W-I-·t-hh-e-I-d-f-ro-m------, public release Pub. L. 86-36

HA~EC~STEMSJOINTLY ~5ABLETOFO._._

TOP SECRETl:IMBItA

350

DOClD: 523682

REF lD:A523682
'fOp SECRET UMBRA

Walter Deeley He W8.& tbe driving force bebind cryptologic centralization and the automation ortime\y reporQn,.

HA~DLE VIA

NTROL SYSTEMSJOINTL Y lN~:JU!ii:;E:1fJS)UlBLE TO FORE[GN

·351

'fOP SECRET UMBRA

OOClO: 523682
TOP SECRET UMBRA
Withheld from public release Pub. L. 86-36

REF lD:A523682

interrelationships between activities in differing Soviet military forces and theaters of operation. The· idea that SIGINT might get a handle on Soviet force posture by such an across-the-board look took hold, and AS analysts William Black, and others began looking at activity level indicators from various areas of the Soviet problem.

I

I

Just asc:=Jwas in its death throes,1 Iwas under threat. The high-speed data link, called the DLT-5, permitted SIGIN'T to flow back to Fort Meade at the then-incredible rate of 2400 bauds per second. Cecil Phillips, who was placed in charge of processing operations in C5, was told to try to duplicate, as near as possible, the operations then existing at I I Phillips even used the same computer, an IBM 1401, to receive the data and format them for follow-on processing on the IBM 7010, which was an upgraded version of the 1410 used atl I Originally he used the same software package in useJ J As long as the DLT-5 was operating,l Iwas superfluous. NSA had succeeded in duplicating the field processing center."
SIGINT at the White House

All presidents since Pearl Harbor had a mechanism for timely notification of crises. In the 19509 intelligence warning was funneled through CIA, which was responsible for alerting the president through his military advisor. The Army ran the White House communications center, which in turn served the military advisor. This placed CIA in the position of deciding what the president saw and when he saw it. By the time of Kennedy's inauguration, the alerting mechanism in the White House had come to be called the White House Situation Room. It was basically a communications handler - no substantive analysis was performed in the "Sit Room ." 4G Following the Bay of Pigs incident, Kennedy decided to put some teeth into the Situation Room. I ICIA was brought in to create a truly round-theclock intelligence center. The Situation Room began taking a more active hand in crisis alerting and in keeping the president informed. It was basically an arm of the CIA, however.·1 All SIGINT product of interest to the president and the National Security Council staff passed through CIA, which forwarded key items after it had taken off the NSA header. SIGINT reports arrived in fairly significant volumes, but NSA was not directly involved. It produced only "information," not "intelligence." Some of the products got to the White House because they related to impending or ongoing crises. Other reports were forwarded simply because the intercepted messages mentioned political figures by name." During the Cuban Missile Crisis, the "White House" (presumably National Security Advisor McGeorge Bundy) was unhappy with the delay experienced in getting certain SIGINT reports. The incident involving McNamara and the DF of Soviet merchant

HANDLE VIA

NTROLSYSTEMSJOINTLY OT RELEASABLE TO FOREIGN NA

TOP SECRET UMIItA

352

DOClD: 523682

REF lD:A523682
TQP S'ECRET UMBRA

ships (p. 328) was emblematic of the problem. But CIA remained the choke point as long as Kennedy lived. 4S ThingS began changing under President Johnson. In late 1965,1 began meeting with Deputy Director Louis Tordella and Chief of Policy John Connelly. along with representatives from CIA and State. The president wanted direct distribution of'certainSIGINT,and he wanted it immediately. CIA and State protested that NSAdid not produce "intelligence" and that it should not send things directly to the White House. was adamant - they could protest all they wanted, but the president had already decided. A direct circuit to NSA was already being installed, and I Iand TordeJla had developed a procedure to courier espe~iallysensitive' material to the Situation Room."

I

I

I

The White House wanted direct distribution for Critics.' Moreover, it wanted to see product reports that quoted or named White House people, including the president, his key advisors, and cabinet secretaries. (This was the material that Tordella was having couriered to the White House.l Late in the year, Tordella appointed Edward Fitzgerald as the first NSA liaison officer to the White House.f The White House concern may have been spurred by SIGINTproduct reports detailing ,...., ~
,...;

'-;:-----::---:-:~--::-~:___.:____:_:_---:__-_:__-,_;_-_::_____",..J Placing the White House on direct distribution for these reports, and cutting off other addressees from normal distribution . I
It is difficult to know what John Kennedy thought about SIGn~T,if he ever thought about it at all. His national security advisor, McGeorge Bundy, seems to have used it as part of a larger intelligence mosaic, and he acceded to the CIA method of organizing intelligence, in that it came to him only after it had been massaged. Bundy appeared to . violate this scheme near the end of his stay at the White House by demanding direct infusion ofSIGINT.This was partly to keep a better handle on late-breaking events, but it was also

.S: .~ ~ ~ ~

=

c
~ Withheld from public release Pub. L. 86-36

~--------~l
.

tol

~

I ------------------------------·

But Kennedy was assassinated in November of 1963, and the new president, Lyndon Johnson, replaced Bundy with Walter Rostow in 1966. Rostow had worked in England during World War II to plan the strategic bombing campaign. He learned not to accept filtered intelligence and worked directly with SIGINTevery day. 48 Lyndon Johnson was the most avid consumer of intelligence ever to occupy the White House. He consumed it voraciously, chewing through stupendous piles of intelligence reports every day. Johnson did not like to be briefed - as former DCI Richard Helms once said, "President Johnson, when he had something on his mind, simply wasn't listening to what one had to say to him. . .. But when he read, he read carefully, and he hoisted aboard what he read .... " 4? Johnson insisted on direct information. He had a great variety of

353

TOP $ ECRET tlMBRA

DOCID: 523682
TOP SEeRET tJMBRA

REF ID:A523682

direct information feeds, including a three-screen television set for all three networks, tickers, and other devices to stay on top of things.48 During crises (and his administration seemed to be one long series of crises), he would sidle down to the Sit Room and pour through the intelligence reports. If a key military operation was about to be launched in Vietnam, he might stay nearly all night, so that he could get the latest information, or he might come in early the next morning to read the latest news. He resembled no one so much as Abraham Lincoln in the telegraph office, waiting for the news of battle to come off the Wire. Even when he vanished to the Oval Office during the day, he would often call the Sit Room to receive updates, and he knew many ofthe officers by their rll'st names. He was totally absorbed in mili~y operations and intelligence reports." Under Rostow, the trickle of direct SIGINT reporting into the Sit Room widened to a freshet, then a flood. SIGINT reporting on Vietnam was highly regarded in the White House. Sometimes it was used to cross-check other sources, other times as a stand-alone source. During the secret negotiations with the North (which occurred more or less continuously through three administrations), SIGINT was a highly prized source of informationl

Withheld from public release Pub. L. 86-36

I The main target remained the Soviet Union
L..-

The Agency processed the material ahead of everything else and sent it directly to the White House. Rostow got the Wormation ra w, analyzed some of the data himself or employed members of his staff to do it, and sent the conclusions to the president.
---l

HANDLE VIA TALENT ASABLE TO FOREIGN NA

L SYSTEMSJOINTLY

TOP S!eKET tJMBIbtt

354

Lyndon John.on conCer, wiUl RolMrt McNama,. GO ID M

<'I

iD l •• '.dW'lD,

lb, bei,blottb.

UJecre&aqol8c.ae. Dean RuikUi

i.~"

w.,. in ViMGaaa.
backcround')

<'I

10

HANDLE

VIA TALE

Q

g
Q

H

355

IOPSECREI

UMBRA

DOClD: 523682
TOP SECRET l;IMBAA

REF lD:A523682

Having an avid SIGINT consumer in the White House had its drawbaeks. David McManis, who replaced Edward Fitzgerald as the NSA representative to the Sit Room, remembers having to explain the nuances of SIGINT reporting to White House staffers all up and down the line. During the height of the war in Vietnam, the National Security Council staff wanted an accurate count of North Vietnamese infiltration into the South, and they buried McManis under a snowstorm of questions about infiltration groups appearing in SIGINT (the only high-validity source on -infil trat ien). To some, he had to explain that there was no turnstile for infiltration groups heading south, but this just got into SIGINT intricacies that the questioners were not prepared to handle. McManis summoned battalions ofNSA briefers to the White. House to explain trail group S1 accountability in SIGINT.
David McMIlDia

The White House insistence on raw, unevaluated SIGINT ereated other problems. Johnson wanted to be kept in touch with every crisis, a~d he once told] Ithat he wanted to be called on every Critic, not realizing how many there were. SIGINT Critics on Soviet long-range bombers over the Arcti~ were fairly commonplace, and wisely decided not to call the president on them, lacking other indicators.

I

I

Most of the SIGINT reports flooding into the Situation Room were relatively low-level reports and translations, with very little analysis and even fewer assessments. Assessing things was still not NSA's job. This situation kept thevolume of reports up, but there was little analytic glue to fit the disparate pieces together. It was critical that someone be available to interpret and assess the SIGINT. Thus McManis found himself spending long hours in the White House.' Moreover, NSA began contributing other Situation ROomstaff members on a permanent basis, the better to minimize the misuse of SIGINT. (The arrangement continues to this day.) Very few people outside 'NSA liked the new, elevated status that SIGINT was getting. But it was a logical. progression of events. Presidents wanted to know, and to know

HAN NOT RELEASABLE

NT CONTROL SYSTEMS JOINTLY TO FORE

TOP SECRET UMiR6.

356

DOCID: 523682

REF ID:A523682
Tep SEeReT UMBRA

quickly. They tended to be impatient with bureaucracy, and when they found a spigot of critical warning information, they turned it on, no matter whose feelings got bruised. When Nixon entered the White House, his Situation Roomchief was an NSA official, and a major portion of the inputs to the White House was coming from the SIGINTsystem. Whatever anyone else ingovemment might think ofSIGINT, the White House was known to view it as the fastest and the most unimpeachable source. Through this reputation, the position ofNSA grew, until it was virtually coequal with CIA and had far exceeded the other intelligence assets of the Defense Department. Carter Takes Command Gordon Blake retired in 1965. He was replaced byMarshall Sylvester Carter, the deputy director of CIA, on 1 June 1965. Carter, a crusty Army general in the mold of Ralph Canine, presided over the stormiest period of NSA's history. "Pat" Carter (the name he went by was bequeathed him by a Japanese maid when the Carter family lived in Hawaii) was from a military family, his father rising to the rank of brigadier general. As a result, his growing up was itinerant, and he set his sights on a military career very early. He took a traditional path up the chain, graduating from West Point in 1931 and going into the artillery branch (specializing in' defensive artillery). During World War II Carter caught General Marshall's eye, and from then on he was a George Marshall protege, serving Marshall in various executive capacities when he was chairman of the JCS, representing Truman, in China, and secretary of state. Mer Marshall retired, Carter held a variety of positions in combat units and also served a tour as chief of staff ofNORAD.

Marshall S. "Pat" Carter

HANDLE VIA TA ~---!"I6'I~[iE~ASABLE

CONTROLSYSTEMS JOINTLY TO FOREIGNNATI

357

Tep SECRETliMBRA

DOClD: 523682
'fep SECRET UMBRA

REF lD:A523682

In his NORAD job he had a fairly detailed involvement with various intelligence sources, includingSlGINT, but had never had ajob directly in intelligence until 1962, when President Kennedy nominated him to become deputy DCI. Carter came upon the position in the wake of the Bay of Pigs fiasco. There had been quite a shakeup at CIA, and one of those to lose his job was Air Force general C. P. Cabell, the deputy director. Carter survived his trial by fire, the Cuban Missile Crisis, in good shape, and was generally regarded to have had a successful tour at CIA. He provided a human face to the Directorate, which was headed by the austere and remote John McCone. He became known as an inveterate prankster and became popular with the work force while handling day-to-day business for McCone, whose ties were to the Kennedy family rather than to the bureaucracy. One "Pat Carter story" that CIA employees loved to tell was about the door between McCone's office and Carter's. McCone was not close to anyone at CIA, and, as if to make the Point, one day he had the door between his office and Carter's walled over. Carter placed a false hand at the edge of the new wall, as if a door had shut'On it, and enjoyed a good laugh at McCone's expense. 52 John McCone was apparently not even a ware of the hand. Marshall Carter became DIRNSA almost by accident. When McCone. left CIA in 1965, President Johnson appointed Admiral Raborn to replace him. By law, CIA could not be headed by two military officers, so Carter was out of a job. He put his problem to General Johnson, the Army chief of staff. A few days later he got a call from the deputy secretary of defense, Cyrus Vance. Gordon Blake had decided to retire, and Vance wanted to know if Carter wanted the job. It took him only a few seconds to make the decision. He had been a deputy or chief of staff virtually his entire career - as DIRNSA, he would finally run his own show.53 . Carter knew a lot about NSA and had a high regard for the Agency. But he felt that NSA needed to be more forceful about its conclusions, more aggressive about carving out a place for itself at the intelligence table .. He made it his business to make NSA more aggressive. The days of reticence and retirement Under Samford, Frost, and Blake were over. Carter fell on a startled national defense community like a bobcat on the back of a moose. He began with a symbolic assertion of NSA's independence. He directed that the NSA seal, which had its Defense Department affiliation prominently displayed, be changed to a new seal which referred only to the United States of America. Carter seriously considered the possibility of requesting that NSA be removed from the Defense Department and set up as an independent executive agency along the lines of CIA. He often referred to the fact that NSA was for him, as it had been for all previous directors, a final stop in a long military career. He was not up for promotion, and he did not care whose toes he stepped on.54

:DLE

VIA TAL= ~

""'b."OMIr:_l~SYSTEMSJ~Y
ASABLE TO FOREIGN NA uNkLS

lap SECRET UMBRA

358

DOClD: 523682

REF lD:A523682

_e. _....__ .
'VI";)~ftC'

III •.••••••.•

U'V,DhA

Even when he was deputy DCI, Carter did not get along with Eugene Fubini. He made his acceptance of the NSAjob conditional on an assertion from Vance (which he got) that he would report directly to Vance, rather than through Fubini at DDR&E. He did not hide his disdain for the brilliant and opinionated Fubini, once calling him "a radar teehnician beyond his competence." But since DDR&E continued to exercise a major influence over NSA's programs,it did not matter much whether Fubini was in Carter's direct line of supervision or not. The two battled almost daily until Carter's retirement in 1969, to the ultimate detriment of NSA's programs. Carter's abysmal relationship with Fubini and the OSD stafi' was more than matched by his almost disastrous relations with the armed services. The assertive Carter was ever on the lookout for service encroachments on NSA's prerogatives, and he found them daily. The military were, he felt, constantly building up their intelligence staffs, adding more analytic capability than they needed, especially in the. SIGINT field, and doing more interpretation of NSA's information than they were qualified to do (especially at DIA). He felt that they were engaged in a continuing effort to redefine SIGINT as "electronic warfare," the better to take it out of codeword channels and build up their own tactical SIGINTcapabilities outside of DIRNSA control. The services, for their part; complained about perceived lack of NSA response to their needs in Vietnam. SIGINTwas too compartmented, NSA refused to clear field commanders for the information they so badly needed, NSA was overprotective of its resources and too quick to fence off new capabilities under code words and compartments. A battle royal . erupted during Carter's regime over the handling of SlGINTand the provision of siGINT support in Southeast Asia. It poisoned the atmosphere and led to a confrontational relationship between NSA and the military it was sworn to support. When Carter retired in 1969, NSA's relationship with the JCS was at an all-time low. Successive directors were 80 instructed by the experience that they never allowed relations to return to that level.~ To the SIGINTcommunity, however, Carter was a champion. Like Canine, he elevated the status and pay scale of the work force, obtaining more supergrade billets and a generally higher average grade .. Displaying his vaunted independence of action, he went directly to Senator Sam Ervin to get the billets and to make sure that the new billet allocation was designated specifically for NSA so that OSD could not co-opt some of them (as he suspected Deputy Secretary of Defense Cyrus Vance of planning). After years of struggle at the OSD level, NSA under Carter got the authorization to begin a career cryptologic service, separate and apart from the systems of any other agency. At the same time, Carter began the civilian intern program, starting with a small number of recent college graduates entering the NSA work force. In 1969 he extended it to the on-board population. He fended off proposals that NSA's cryptologic work force join a DIA-sponsored intelligence community career development program, carrying with it the

HAND~

KE"UQIsIi'"C'OUINT CONTROL SYSTEMS JOINTLY TOT RELEASABLE TO FOREIGN NATIO 359

TOP SE(RET UMBRA

DOCID: 523682
f6P SECRET t:JMBRA

REF ID:A523682

clear implication that there should be transferability between the general intelligence field and cryptology. 58 Internally, Carter wanted a strong central staff, and he created an executive secretariat to manage his staff and its activities. This reflected his Army background and his experience as staff chief for General Marshall ...He strengthened the training school by upgrading its staff to assistant directorship and calling it the National Cryptologic School. Frank. Rowlett was its first chief, thus bestowing a status and prestige which it had never had before. Carter was an Anglophile, and he worked hard to maintain the strong ties with GCHQ that had developed over the years. 57 Under Carter the centralization OfSIGINTIDoved quickly ahead. A Group implemented Plan B and closed the theater processing centers. In the Pacific, the decision to close JSPC, opened only in 1961, was made in 1965. JSPC was a victim of improved communications programs, especially the move to automatic forwarding of intercept trafli~ under the AG22/STRAWHAT program (see p. 366). At first, arrangements were made for the AG-22 traffic to be routed through Sobe, where data of interest were stripped off for computer processing. But likel JSPC could do nothing that could not be done at Fort Meade, and the center at Sobe was doomed. As in Europe, the theater military commanders fought the closure ofSobe energetically, but to no avail.iS8

I

Withheld from public release Pub. L. 86-36

It was also during Carter's tenure that AFSCC was finally closed. Though Closure plans originated as early as the AFSA period, AFSC.C was even stronger and more important when Carter arrived than when Canine became the director. But Carter signed a new closure plan in 1967 and made it stick. NSA had begun quietly transferring functions from AFSCC to Fort Meade in 1966, and after the closure plan this accelerated. First to go was the I I followed by larger efforts like the I I I I AFSCC officially went out of the COMINT processing business on 30 June 1969.1 were transferred to NSA,Dwere eliminated, andDremained in San Antonio, where they merged into a new organization called Air Force Electronics Warfare Center, which analyzed the effectiveness of military-wide electronics warfare efforts, based primarily on SIGINT inputS.59

I

NSA would have closed AFSCC earlier if space Couldhave been found, but the Agency was always chronically short of space. The dedication of the new nine-story headquarters building in 1963 just barely caught up with an expanding population, and there was still no room for the Center. The key event was the lease of the Friendship (FANX) complex (see p. 294). NSA moved into the first building, FANX I, in the fall ofl967, and as new buildings were completed, it occupied those also until by the fall of 1970 the Agency was the tenant in FANX I. II, and III. (NSA was the first and only resident of all the FANX and Airport Square buildings that it leased except for FANX I, whose lease has been given up.) It was not cheap - Carter once stated for the record that for four years worth of rent,

f.(Jp SECRET UMBR-A

360

DOClD: 523682

REF lD:A523682

'fe' SEERET

l:JMBR:A
..
/

NSA could have built its own buildings. But military construction money was carefully controlled by Congress.eo

MECHANIZATION OF THE SIGINT PROCESS
You people are doing a tremendous job producing history. You are not producing intelligence, Moody to the Bl

Juanita

workforce, 1961

SlGlNT had a reputation for being laborious and expensive. Intercept operations tended to be labor-Intensive, while processing was equipment-intensive. Of all Department of Defense organizations, the SeAs were the most far-flung, draining. the federal government of foreign currency in the attempt to maintain small sites in remote areas difficult and expensive to supply. Robert McNamara had a war to fight, and he exerted intense pressure on the SlGINT system to economize. This manifested itself in pressure to reduce the number of people involved 'in the system front end, both through field site mechanization, and through the transfer of operations back to the Continental United States. Along with the economic pressures came demands to speed up the system. Eisenhower's concerns over war warning information, far from disappearing after his administration ended, intensified under Kennedy, The Bay of Pigs and Cuban Missile Crisis instilled a sense oChurry-up. The twin demands of economy and speed pushed the cryptologic community into a thorough remodeling ofSIGINT.The result was the fashioning of a new syste~ drastically different from the one which had emerged from World War II and had stood relatively intact through the 19509. It had been the dream of cryptologists for years to modernize and automate manual . Morse intercept, the largest part of the front end. A first try at it was during World War II, when OP-20-G attempted to produce a punched paper tape from a manual typewriter. thus readying the intercept for introduction into a follow-on processor without further manipulation. The resulte of the experiment are lost. It was the last attempt at that sort of thing for at least ten years.81 In 1957 NSA began toying with the idea of copying Morse on a special typewriter that would do more than just copy alphanumeric characters. The Agency modified a Remington-Rand Synchro-tape typewriter by adding special keys at the top of the keyboard that designated tags, indicating such things as callsigns and frequencies. The project was called SPlT (Special Intercept Typewriter).02 '

-~~=.:~
361 la, seCRETI:IMBRA

OOClO: 523682
TOP SECRET UMBRA

REF lD:A523682

While technicians modernized the intercept operation, NSA began looking at processing techniques. Since the dawn of America's SIGINT system, intercept· sites had forwarded raw traffic to Washington for processing. While raw traffic went by courier and took weeks to arrive, traffic extracts, often called TECSUMS (technical summaries) were prepared at' the field site from the raw traffic and were forwarded electrically so that Washington had at least a summary of significant intercepted material. 'Prior to the late 1950s the TECSUMS went by formal message, but with the advent of Opseomms, more and more TECSUMS were put on Opscomm circuits. At the time, NSA technicians and analysts were engaged in a philosophical debate about mechanization. Should traffic be brought back in bulk to NSA, where machines could prepare it for computer processing, or should the mechanization c>ccurin the field, closer to the front end of the process? In the end the front-enders won, and NSA began designing equipments that would mechanize the intercept operation.

I E.O.

13526, section 1.4(c)

I

Withheld from public release Pub. L. 86-36

The experiment with the SPIT typewriter spawned a new project, called] lor the AFSAV 3110. The! ~quipment consisted of a modified Remington-Rand typewriter similar to the SPIT model, with special keys referring to such traffic components as callsigns and to traffic externals like start-of-message, end-of-message, and case notation. These features would speed the intercept process by relieving the operator from having to type them in manually. Butl ladded a new feature similar to the World War II experiment - the output was both page copy and a seven-level paper tape. The beauty of this modifleation was that the tape could be transmitted just like an outgoing message, and it could be input to a computer at the other end, providing that it was compatible with both.a
I [quickly became the focus of the Joint Mechanization Group (JMG). This ad hoc committee was the brainchild of Frank Raven and Juanita Moody. Raven, one of the leading cryptanalysts to emerge Irom the Navy in 1945, was at the time chief of GENS, while Moody was a division chief within AOVA. They were intrigued by the possibility of automating the front end of the system and pushed I [as a possible answer. Moody named her deputy, Cecil Phillips, to head the JMG}14 A field test performed at ASA's Rothwesten site in 1960 proved the intercept portion ofthe concept.

The next logical step would be to input intercepted traffic produced on an I " position into a computer and do some processing on it. Frank Pinkston, a US~SS staff officer, heard about the I [machines, which at the time (1961) were lying idle, and asked if Security Service could run its own test. The Air:Force liked the idea because it would facilitate the rapid transmission and processing of highly perishable air-related traffic. Pinkston designed a test in which I Ipositions would be located at the AFSS sitel Iwould produce communications-formatted tapes, and would forward the I E.O. 13526, section 1.4(c)
HANDLE VIA TALE OLSYSTEMSJOINTLY T RELEASABLE TO FOREIGN NA

-

Withheld from public release Pub. L. 86-36

lOP SECRET UMBRA

362

DOCID: 523682

REF ID:A523682
. 1'OP S!<!R!T tlMBItA

Frank Raven

Juanita Mood), receiving tbe Distinguis bed Civilian Service Award from then-DCI George Busb ill 1978. NSA director General Allen looks on.

HANDLE VIA TAL ASABLETO FOREIGN

OL SYSTEMS JOINTL Y

363

TOP SECRET UMBRA

DOClD: 523682
lOP SECRET UMBRA

REF lD:A523682
Withheld from public release Pub. L. 86-36

E.O. 13526, section 1.4(c)(d)

tapes via Opseomm tol I where they would be fed into the IBM 1401, which would produce an automated TECSUM. The JMG got a Bogart programmer to design the software, and in September 1961 AFSS ran a successful test. (Bogart was one of NSA's RAMsystems.)~ The project then languished, primarily because every field site would need a 1401. The 1401 was at the time part of AFSS's 466L system, which was under.intense fire from NSA because of its complexity and expense. But interest never vanished. ASA had embarked on its own project, ealledl I which was soon subsumed under the auspices oftheJMG. Meanwhile,c=:Jproelaimed the concept revolutionary and proposed that it be broken down into.component portions and implemented gradually. Rather than locate computers at each field site,c=:Jproposed that traffic be forwarded to centralloeations. This concept would reduce the number of computers required (computers were still regarded as exotic and outlandishly expensive), but it would also overload the communications system. Thereby hung the dilemma."
AG-22

While the policy people thrashed out the dilemma, the technical people continued working on improvements to the device. The Remington-Rand equipment was judged not sturdy enough and was replaced by a Teletype Model 35, extensively modified by the addition ofthe special tagging keys. The Agency named the device the AG-22 and changed the output to an eight-level tape. NSA also standardized the tagging and traffic formatting requirements into a new TECHINS (T-5004), so that Morse traffic intercepted anywhere would lookjust like any other Morse traffic. Computer formatting requirements were beginning to drive the SIGINT system. &7 Changing the Communications System

The communications system that AG-22 tapes were preparing to assault had become creaky and outmoded, and it was incapable of handling the new requirements. The Cuban Missile Crisis jammed the communications system as it had not been since the twin Suez and Hungarian crises of 1956. After the creation of Criticomm, NSA continued to try to develop a high-speed switch that wQuld improve reliability and reduce handling time. At first, technical hurdles delayed adoption of a new switch. But in 1962 a new, bureaucratic obstacle appeared with the creation of the Defense Commu~ications Agency (DCA). Such an agency was a logical outgrowth of McNamara's centralization strategy. but it confused the Criticomm situation. DCA took over the job of searching for a new switch, regardless of the feeling at NSA that this would slow the development process. There is little doubt that the_project was further delayed by hard feelings between the two agencies.611

HANDLE VIA TALENT KE T RELEASABLE

ONTROL SYSTEMS JOINTL Y TO FOREIGN NA

lOP SIiCRET bJMBAA

364

DOClD: 523682

REF ID:A523682
TOP SeCRET UMBRA

AG·" - Configured Morse Position ad I (R-390 receivers are In the lert·hand rack; MOD·35in the center; and tape unit on the right)

I E.O. 13526, section
HANDLE VIA TALENT KEYHO NO

1.4(c)

I

Withheld from public release Pub. L. 86-36

L SYSTEMS JOINTLY TOFOREIGNNATIONA

365

TOP SECRET l:JMIRA

DOClD: 523682
,

REF lD:A523682

lOP SECRET UMBRA In the mid-1960s, DCA decided on a new satellite communications system called Defense Special Security Communications System (DSSCS), and it decreed that the new Criticomm switch would have to be compatible with the rest of the system. The fact that operators in general service (Genser) communications centers were not Sl-eleared created more policy problems, and the search for a switch slipped further. Then in 1964 the picture was further clouded when DIA got approval to manage the SSO system. Part of the package was the creation of a separate communications system for the distribution of COMlNT, called Spintcomm. This introduced new bureaucratic conflicts over who would be the ultimate manager of the composite CriticommlSpintcomm system, and the edict that established Spintcomm further confused the picture by assigning significant responsibilities to all three participating agencies (NSA, DlA, and DCA). Gordon Blake strongly protested DIA management of the system •.but he was overruled at the OSD level. This set off new turf' battles and further complicated the technical design of a switch that would have to handle all communications requirementa." Meantime, more and more traffic flooded the system, largely because of the Vietnam War, and message throughput actually declined from year to year. while errors increased. To stave off disaster. NSA took various halfway measures. Much traffic was diverted to the. expanding Opscomm systems, and Criticomm was reserved mainly for formal messages. The Agency also designed terminal equipment which would speed and improve handling of traffic within the Criticomm centers. One such solution was the BIX (Binary Information Exchange), a high-speed local message switch which could operate at various speeds to handle traffic from many dif.'f'erentinputs. NSA awarded the contract to ITT, which delivered the first BIX in 1961. The principal improvement was in data storage (the BIX used magnetic tape to store large amounts of data) and in improved throughput (BIX could handle 100,000 words per minute). As an automatic switch. however, it failed, and messages still had to be processed manually." At the same time. the COMSEC organization was working on crypto that would handle the new circuit speeds. The KQ..13, which could encrypt circuits up to 2400 bauds per second (the speed of the DLT·5 from Frankfurt) went on line in 1965.71
STRAWHAT

NSA planned to install ~G-22s in virtually every HF field site in the world, but the Opscomm system would not be able to handle the volume. Originally designed for analystto-analyst conversations, Opscomms were, by the mid·1960s, becoming overloaded with new TECSUM and! Iforwarding requirements. They were slow of'foot, either 60 or 100 words per minute, and barely able to handle current requirements. If AQ..22

I E.O.

13526, section 1.4(c)

-

Withheld from public release Pub. L. 86-36 .

Y

HANDLE VIA TAL", ~ CONTROL SYSTEMSJOINTL _ NOT RELEASABLE TO FOREIGN ~

TOP SECRET UMSRA

366

.

-

-

---_._-------------------------

DOClD: 523682

REF lD:A523682
lOP SECRETUMBRA
.. ..,.,. C"
"

....

',: '--:'~~":ll:'

1
;

S::~ . ,,' ,
:'~"' ..

'r'
,',

.. ~ .

".

'~',' t ' .J"1' '

f

'

! ••.

r 11.:i

i.'

}~ '1
.

t:
~

Comm Center, 1960s. Lackina a digital switch, Critlcomm centers continued to be overwhelmed by five-level tape and manual processing.

HANDLE VIA TALENT ONTROLSYSTEMS JOINTLY __ -NOT'R!iLEASABLE TO FOREIGN NA

367

Tap SECRET l:IMBRA

DOCID: 523682
19P SECRET lIMBRA

REF ID:A523682

data were diverted to Opseomm, it would expand the circuit requirements geometrically. Lacking a revamped Criticomm system, the solution lay in a separate, high-speed data system specifically for AG-22 formatted tapes. In 1967 NSA came up with the answer the Agency called itSTRAWHAT. STRA'YHATwas a 9600-baud data link system from field sites to processing centers. A time division mUltiplex system capable of up to eight-level forwarding, its equipment could be patched directly from the circuit terminal to a computer, bypassing .the person in the communications center. The first circuit became operational in December 1968, and NSA planned to wire up more stations with STRAWHATcircuits beginning in 1969. By mid1970, the entire SIGINT system would have at least anlnterln; STRA WHAT eapability." The Computer Industry at NSA By the mid-1960s mainframe computers had taken over much of the manual processing at N.SA. Although the dual tracks of scientific versus general-purpose processors were eontinuing, increasingly the Agency was focusing on the latter. It had to do so in order to handle the TECSUM data flowing into Fort Meade via the burgeoning Opscomm network. At that time, the computer of choice for this operation was the IBM 7010, an advanced model of the IBM 1410. IBM machines almost totally dominated the general purpose processing job, and the collection of 70105 was simply called "the IBM complex.'J'ls IBM was not the only company doing business with NSA. In 1963 the first minicomputer, the PDP-l, was delivered to the Agency. That, and its suceessor, the PDP-IO, were used for a wide variety of special-purpose processing jobs. That same year, NSA purchased the Univac 490, which had a capability of handling thirty remote stations simultaneously. The stations were equipped with both paper tape and Teletype Model 35 input devices. The software, called RYE, was developed at NSA and was ideal for handling simultaneous inputs from the remote stations. It was made to order for processing from communications terminals, and thus it fitted NSA's emerging needs for handling Tecsumized inputs from field sites, as well as a variety of other small-job applications." By 1963 NSA's computer collection was by far the largest in the country and probably the world. The value of its computers topped $50 million, which was greater than the Census Bureau, the Baltimore headquarters oftheSoeial security Administration, and all the field offices of the Internal Revenue Service put together. By 1968 General Carter could boast that NSA had over 100 computers occupying almost 5 acres of floor space.n NSA continued to do pioneering work in partnership with the commercial computer industry. One such innovation was the so-called Josephson Junction technology. This was a very-low-temperature phenomenon in which "switching an electron tunneling junction between two states is accomplished by means of a magnetic field."76. Discovered in the mid19605, the potential for speeding up computer processing was so attractive that NSA

HANDLE VIA TA NTROL SYSTEMS JOINTLY ~--fffl'I'1m:r::LE~ASABLE TO FOREIGN NA

lOP SECRET lIM8RA

368

-------'--

-

--

-_.

OOClO: 523682

REF IO:A523682
lOP SECRET UMBRA

funded about one-third of the IBM research on the Josephson Junction technology. Unfortunately, it didn't work, and IBM ultimately gave up on the Josephson Junction. The project illustrated both the need for research in advanced technologies and the risks involved. NSA also pioneered in techniques for mass storage. One such experiment was called TABLON, developed in concert with IBM and Ampex in the 1960s. Tablon used a photodigital process developed at IBM and a tape storage system developed by Ampex. The storage systems were internetted by means of two PDP-IOs. The philosophy was to have a central data storage system that could be used by the entire agency. But TABLON had serious technical problems. Ampex was unable to develop a tape drive that met system specifications, and too much software was required to run the PDP-IO-based star network. Ultimately TABLON was overtaken by new disk storage technology. 77 NSA programmers were in the forefront of special computer language development. Agency programmers created special languages for HARVEST (called Beta), for the IBM 1401 (called PAL) and punched card emulation language (Transembler) for the IBM 705. Still, the Agency was losing its edge in pioneering work, as the commercial world forged ahead with new innovations that owed less and less to the inspirations that had stemmed from cryptologic applications. It was an inevitable process. 78 IATS The new AG-221STRAWHAT marriage, innovative though it was, had some problems that could only be called "logistical." A large field site, with row on row of manual Morse positions, could produce a considerable amount of eight-level tape in a day. The process of accounting for, and carting to the communications center, long coils of tape cascading off collection positions was time-consuming, and an analyst (who had now become a communications tape handler rather than a SIGINT analyst) could literally become buried in tape before the end of the shift. In the mid-1960s K Group (the PROD organization responsible for interfacing NSA with the field sites) began working on a system for accepting manual Morse data directly onto a magnetic tape. After experimenting with several different computers, it settled on the Honeywell 316, which could accept data from 128 different sources simultaneously. (Thus, a field site would have to have more than 128 Morse positions .before it required more than one 316.) Honeywell, which sold the 316 at a very competitive $12,500, agreed to loan one to NSA, and a test was run at Vint Hill in Virginia. The test system worked, and the Agel)cy, which called the new system IATS (Improved AG-22 Terminal System), got $10 million in 1968 to install Honeywells at all AG-22 field sites. The AG-22 positions were wired to the on-site Honeywells, which packed the intercept files onto a magnetic tape. Periodically (usually every six hours) the tape was transmitted on a high-speed data link to NSA. 79

HANDLE VIA T TROL SYSTEMS JOINTL Y ..B£~Et1!~ABLE TOFOREIGN N 369 lOP $iCREl' UMBRA

DOCID: 523682
. T6P S!CREftJMIUtA I

REF ID:A523682
Withheld from public release Pub. L. 86-36

E.O. 13526, section 1.4(c)

At this point NSA embarked ona major software development effort to handle the expected influx oflATS data. Cecil Phillips gave the job to John W. Saadi, who was a team chief 'in Philiips's C Group. Saadi, writing in assembly language, created a series of processes (called I Iresident on a Univac 494, which accepted the data from the communications system. The 494 built batch files and passed them to the IBM 360 through a shared disk arrangement. This was a ground-breaking task because IBM machines were notoriously difficult to interface with the machines of ariy other company.
,

'i:i' '-'
""" ,..;

The IBM 360, the first third-generation machine. was introduced at NSA in the late 1960s to replace the 7010s.
i

= .~
...0
N

~ ~ ~

i

In

....
I"')

i Each production organization wrote applications programs for the 360 complex, so that its data, handed to the 360s from! I would be processed and ready for the analyst. The complex did its heaviest work at night, SO that the output would be ready for the analysts in the morning.80

Withheld from public release Pub. L. 86-36

c
~

Now that raw intercept files were available on computer, each production element developed databases. Some of the work in this area. especially that done by A Group to create a relational database for the Soviet problem, was on the leading edge of technology." The Communications Solutions

The impasse that had been created between NSA, DIA, and DCA lasted through the end of the Carter regime. By 1968 DCA had still failed to produce an adequate communications switch, and Carter felt that DCAfailed to understand SIGINT (despite the fact that. the director of DCA, Lieutenant General Richard Klocke, had been one of the founding fathers of the Air Force Security Service). But the next year brought a new director, Vice Admiral Noel Gayler, and a new approach to the logjam., Gayler moved quickly to iron out differences, and in August of 1969 he signed an agreement with Klocko covering management ofthe communications systems that supported SIGINT. The agreement was a carefully crafted compromise. DCA would manage the entire system, based on technical specifications submitted by NSA. DCA could satisfy communicatione requirements using any type of circuitry, as long as NSA technical specifications were adhered to. The next month DCA cancelled the automatic switeh contract withITT. Shortly thereafter, OSD decided that the new DCA communications . system, called Autodin, would be used for SIGINT traffic. This decision would result in NSA relinquishing a proprietary net that it had controlled since its birth. Some were not happy, but Gayler held to the compromise package, and an era of relative good feeling resulted between Gayler and Klocko.82

HANDLE~ilet.rrUPINT~NTROLSYSTEMSJOINTLY OT RELEASABLE TO FOREIG ,ur

3C>"",,,C; ,

VW'I01V"

37()

----------

..

--.

--

DOCID: 523682

REF ID:A523682
TOP S!(;RET UMBRltc

Lacking a DCA -automatic switch, NSA developed its own in-house version and hatched plans to use it in its own communications center at Fort Meade. The Agency decided to scrap the TeletyPewriter Distributions System in use since the new building had opened in 1957 and replace it with a new communications center based on the new switches. It was to be called IDDF (Internal Data Distribution Facility), and it opened its doors in early 1972 on the third floor of the Ops-l building. The year before, NSA introduced optical character readers in the message processing facility. an innovation which led to the elimination of the time-consuming step of teletype operators hand-poking every outgoing message. Called AMPS (Automatic Message Processing System), its rigid formatting requirements and special IBM Selectric typewriter balls were at first hard for secretaries to get used to, buta godsend to the communications center.as Withheld from public release E.O~13526, section 1.4(c (d) Pub. L. 86-36 New methods of forwarding data to NSA did not change the basic process of signal collection. Most of an operator's time was still spent searching for target signals. But with .the new digital technology and smaller on-site computers, it should theoretically be possible to acquire certain signals automatically. In the early 1960s, R&D began working· on the development process. The early development work was done in r963/1964 under a project calle<1

The production model ofl I It was a more sophisticated system. which had an automated digital front end connected to several back-end manual Morse collection positions. I

Digital computer-based collection systems eventually became the rule rather than the exception. Some. like the IRON HORSE system used in Vietnam (see p. 549), automated the collection of manual Morse signals. But Morse transmissions had a huge variety of r---.,---------, formats, and the length of the mark or space varied depending on the sending operator. Withheld from Computer-based collection was far more adaptable to baud-based signals'. An early success public release in this area was Flexscop, a digital collection system I I Pub. L. 86-36

I E.O. 13526, section 1.4(c)(d) I

~LEVlA~-===;;~:==="'JOINTL;
371 TOP SECRET tJMBM

lOP JI!Elte=r 1aI''''·

mnr
N (Xl \0 f'l N U'l Q H

HANDlLVlA LEASABLE TO

ONTROLSYSfEMSJOJN1LY

s
Q

372

DOClD: 523682 I E.O.
13526, section 1.4(c)

REF lD:A523682
Withheld from public release Pub. L. 86-36
'fap Se(RI!!T UMBRA

I I The on-site computer (a CP 818) land demodulated the signal, . then scanned the plaintext transmissions for key words. The system would alarm on . recognition of high-interest text, and the operators would react with special processing and forwarding routines. It replaced the "ancient" CXOF equipment which had been the equipment of choice I Isince the late 1940s.86 with its stable frequencies, plain text, and bauded structure, was especially suitable to automation, and NSA collection and processing systems for that efi"ortbecame among the most automated in the business.

I

D

In the 1980s NSA automated the collection of a very wide variety of signals.,

I The Agency employed a bewildering variety of ~~------~~----~~~ minicomputers for these specialized jobs, sometimes buying commercial computers from
outfits such as Honeywelland DEC, sometimes building its own computers in-house. 67 Bauded Signalsl
---J

In the late 1950s NSA was struggling to cope with the increasing use of bauded systems for record traffic. The trend toward the bauded world resulted partly from increasing traffic flow, which required faster circuit speeds that radioprinter made possible; it also had a corollary benefit of making possible. The field sites were collecting ever higher volumes of printer messages, most of which languished in N$A's warehouses on magnetic tape, waiting to be converted and processed. (For instance, the volume of enciphered communications collection increased I I from 1958 to 1968.88) By the early 1960s the volume of unprocessed magnetic tape was becoming difficult to manage technically and was embarrassing politically.

I

I

R&O's first approach was to build a general-purpose digitizer and diarlser for bauded . signals. ProjectC::Jwhich originated between 1956 and 1958, at first targetted the online was only part of the problem, and R&O; working with A Group, began working toward the on-line digitization . and diarization of the entire bauded signals problem. An ad hoe committee was elltablished in 1959 to study the problem, and R&D began designing equipment to digitize printer signals onto magnetic tape at the collection position. c=Jconsisted of a number of special-purpose components, which were designed to digitize, diarize, and format onto magnetic tape. It resulted in two parallel avenues,

I

I

I

I

.1E.o.

13526, section 1.4(c)

I

Withheld from public release Pub. L. 86-36

373

TOP SECRET UMBRA

DOCID: 523682
TO•• SeeRET UMBRA IE.O. 13526, section 1.4(c)

REF ID:A523682
Withheld from public release Pub. L. 86-36

While R&D experimented with general-purpose processors, DDO. was becoming overwhelmed by magnetic tape. During July 1961 NSA received 17,000 reels of magnetic tape, all of which required signal conversion prior to processin~. In fiscal year 1961 the Agency needed over J just to convert bauded signals for further processing. 90

I

To stem the tide, Operations initiated a QRC (Quick Reaction Capability) project called!. ,I which quickly changed its name tolland the various spin-offs of thec::J project were in full swing (and in direct competition with each other) when, in 1962, DDO initiated a crash requirement tct collect the burgeoning I Isignals. The urgency of the requirement vaulted it ahead of everything else. The new project, calledc=J would eventually result in the conversionofl Ito a standard position. The new positions would intercept, digitize, and record Everything would be processed at NSA in a standard format, thus simplifying the job of the processing organization and the task of designing proeessors."

I

I

I

I

_______

...• 1 The

Attack Continues

1 E.O. 13526, section 1.4(c) I.
HANDLE VIA TALENT ROL SYSTEMSJOINTLY T RELEASABLE TO FOREIGN NAT

Withheld from public release Pub. L. 86-36

Tap SECRET I;JMBRA

374

TOP S!CIt£T tlM11bIo

D
('oj

pcMlt.1on •• ltill ita distinctive

tuUlevend

scop.,.

overlookin,

Lb. tou.r-channel

dlJitben

CD
10

Withheld from public release Pub. L. 86-36
HANDLE VIA 'fALl:~T K __ -J.UT RELEASABLE TO FOREraN NTROL SYSTEM'SJOINTt. NATI Y

PI
('oj It')

C

H
{J

375

Tep SEERET

tlM&1bIo

o
c

DOClD: 523682
IOPSECREI UMBRA

REF lD:A523682 I
Withheld from public release Pub. L~86-36

I E.O.

13526, section 1.4(c)

The Bissell Study At mid-decade, CIA commissioned a study of the status of NSA attack on high-grade ciphers, the first since the Baker study in 1958. Richard Bissell, a top CIA official unhorsed after the Bay of Pigs fiasco, was named to head the committee. Bissell was a good choice. He had stubbed his toe on covert operations, but he was highly knowledgeable on technical intelligence and had in fact headed the U-2 development proj~t in the 1950s.ee

I

Unlike Baker, who had ranged all over the SIGINT landscape, Bissell coromed himself .""lus, •• I, to the project at hand. It was B;ssell who first noted I

I

-

I E.O.

13526, section 1.4(c)

I

HANDLE VIA TA •.•••• ". •.• ,"Vl' J.ROLSYSTEMS~OINTLY _ .NOT RELEASABLE TO FOREh.rn • r .,

19P 5EEREl UMBRA

376

-

Withheld from public release Pub. L. 86-36

DOCID: 523682
E.O. 13526, section 1.4(c)(d)

REF ID:A523682
l'6P SECRETUM8RA

Richard Bissell

. The draft of the Bissell report, which made the rounds of NSA seniors in December of 1964, generated a storm of controversy. The Agency believed that only cryptanalysts could make judgments about systems exploitability and that only NSA should make resource allocation decisions. Blake, at the urging of Deputy Director Louis Tordella, tried . to get Bissell to change the report draft, but did not succeed. Once the report was released early the next year, the new director, General Carter, launched a blistering attack on the specifics. Regarding the recommendations to reallocate resources, he said, "I am confident that our present mix is about right and shall ensure that appropriate changes in emphasis and use of resources are made as warranted." Basically, Carter folded his arms and did nothing." So it had finally come to the stone wall. The Agency firmly believed that it would eventually read enough] Itraffic to make a difference, but practically no one outside the headquarters complex at Fort Meade believed it. Carter, who had no basis for an independent judgment himself, believed what his deputies told him. He held fast, and in this case his independence of action and absolute refusal to brook outside . interference helped save the program.

I

E.O. 13526, section 1.4(c)(d)

I

HANDLE VIA TALENT LSYSTEMSJOINTLY JN~O;u:,J~~.sABLE TO FOREIGNNATI

377

+oP SiCR&T UMalt~

OOClO: 523682
TOP '&eRET UMBRA

REF lD:A523682
Withheld from public release Pub. L. 86-36

E.O. 13526, section 1.4(c)(d)

. HANDLE

-

vrA

l:~:;Z
.

E CQlRN'f CON~ltOL SYSTEMSJOI~ LE TO FORErGN AtION1tLS 378

TOP 'feRn

l:IMBRA

___

0

I

DOCID: 523682

REF ID:A523682

,

I

E.O. 13526, section 1.4(c)(d)

I

Withheld from public release Pub. L. 86-36

lOP SECRET I:IMIRA

-,

COMBEC

at Mid-decade

In the 1960s the KW-26, the equipment of choice for securing long-haul point-to-point record traffic circuits, dominated American COMSEC. But American involvement in Vietnam led to a new set of tactical encryption requirements. Typical of the new COMSEC demands was the need to encrypt record traffic. on low-level tactical nets in a combat environment. The KW-26 was ill-suited for this application, and to meet the demand. NSA developed the KW-7 to secure terminals which received traffic from multiple transmitters. This equipment added a unique indicator for each message. so that stations in a multiplestation net could correspond using a single crypto device. \OS The Development of American Secure Voice The big news in COMSEC in the 1960s, however. was secure voice. U.S. government users would use the telephone for classified talk. and the only solution was to provide them with a secure handset. Secure voice requirements spanned a broad swath from high-level . point-to-point conversations to tactical military applications in the jungles of Southeast Asia. Well aware of the vulnerabilities of voice, NSA approached secure voice cautiously, and for many years secure voice capabilities lagged behind record traffic. For strategic systems, NSA developed two devices in the 1960s. The KY-9 was a narrow-band digital system using a vocoder, and it was the first speech system to use transistors .. The advantage of the KY-9 was that it could be used on a standard Bell System 3 kHz-per-channel telephone system without modification. The disadvantages were many, however. It was big and heavy, encased in a safe that had to be unlocked every morning before the system could be activated. It was also expensive (over $40,000 per

HANDLE VIA TALE
LEi\SABLE

NTROLSYSTEMSJOINTLY
TO FOREIGN NA

379

TOP 5!CRE'f !;IMBRA

DOClD: 523682
lOP SECRETtJMBItA

REF lD:A523682

copy) and was a true "Donald Duck" system which required the users to speak slowly to be understood. Only about 260 sets were deployed, all to high-level users, mostly Air Force.lo. Far more significant was the X'I-3, developed about the same time. Built by Bell Labs under contract, it too was housed in a safe. It was big, klunky, and looked a lot like the KY-9, but without many or the drawbacks. The KY -3 was a broadband digital system, so voice quality was better, and it was not a push-to-talk system. But what brought it into wide use was its employment in the Autcsevecom network. Autosevocom was a secure voice . .network designed by NSA. Local networks consisted or KY -Ss, whose individual voice conversations were first decrypted, then reduced to narrow-band signals and digitized in the HY-2 vocoder, and finally re-.· encrypted for transmission. using a KG-13. The Autosevocom system achieved wide acceptance, and some 2,700 KY-3s were sold to users worldwide, including the White House, the Joint Chiefs of Staff, and the Strategic Air Command.l~ . ,---'---,

Kf-3

As Vietnam heated up, NSA's attention turned increasingly to tactical voice encryption. An early entry into the tactical arena was a set of systems called PARKHlLL. I,C An analog system, it was acknowledged to be vulnerable to exploitation and was not e ~ ~ '" authorized for conversations above the Confidential level. Knowledgeable COMBEe peo~le _ -e Q:j lo. QO • called it ~ .~ ~ 1....J1 But it was better than nothing, and NSA assumed that the Soviets, if they -5 ..Q were to exploit it at all, would have to devote inordinate resourc es.108

=~
fIj

~

I

I

.- ::c ==
~ Q..~

For digital encryption, the Agency first turned to the KY-8, whose development had begun in the late 1950s. The Air Force tested the KY-8 in its F-I00 series jet fighters, but found it heavy and cumbersome to key. (As former COMSECofficial David Boak once said, the Air Force would accept a device "only if it had no weight, occupied no space, was free, and added lift to the aircraft.") More to the point, if the KY-8 were to stay, the fire control

HANDLEVIATALE::;bF.l NOr

"OMJWC9fllKOL~YSTEMSJOI~ BLETO FOREIGNNAn NALS

lOP SECRET UMBRA

380

DOCID: 523682

REF ID:A523682
. 1'6P SECRET HMBItA

radar would have to go. The Air Force opted for the fire control radar, and American aircraf\ in Vietnam remained without voice encryption. The Army and Marine Corps, however, found that they could use the KY-8 in jeeps, and some 6,900 devices were eventually deployed. Meanwhile, NSA embarked on a whirlwind project to provide a KY-8 type of device, absent the bulk and weight. The result was two new tactical voice encryption systems, the KY-28 and KY-38. The former was ~developed for aircraft, while the latter was employed in man-pack radio systems. Weight in both was reduced by the use of integrated circuits. The three devices (KY-8,28, and 38) . were referred to as the NESTOR family. By the end ofthe decade, there were 27,000 NESTOR equipments in the U.S. inventory. lOT The next generation of voice encryption systems was called SAVILLE. Consisting of (KY-57/58) and BANCROFT (KY-67), they were smaller, lighter, and consumed less power than the earlier NESTOR sytems. They also employed updated keying systems and could actually be rekeyed from an aircraft, permitting the control station to remotely change the keys on a net in case a station were overrun by the enemy; BANCROFT was the first-ever combination radio and encryption device in a single unit. VINSON and BANCROFT were not introduced until the early 1970s.108
VINSON TEMPEST

TEMPEST standards had been set forth in the late 19505 in a document called NAG-I. Like other COMSEC policy documents, however, this one was advisory. What was needed was a directive policy and enforcement procedures. NSA spent the decade of the 19605 .working on that aspect of TEMPEST.

In September 1960 NSA briefed the USeSB on existing American TEMPEST vulnerabilities. It shocked USCSB into action; and at a meeting in October the board agreed on a crash program and established its first and only subcommittee, SeOCE (SubCommittee on Compromising Emanations). The first item on SCOCE's agenda was a request from USIB to evaluate the Flexowriter, which was being considered for almost universal adoption within theintelligence community as a computer input-output device. The Flexowriter, SCOCE found, was the strongest radiator ever tested, hardly a recommendation for its adoption within the intelligence community. With the proper equipment, an enemy listening service could read plain text as far as 3,200 feet. The subco~mit~ posted a series of recommendations that became known as the "Flexowriter policy," including recommendations that it not be used overseas at all. that in the U.S. it not be used for classifications higher than Confidential (and then only if the using organization controlled a space 400 feet in circumference); and that the Navy be tasked with a long-range technical fix, At the same time, SCOCE published two lists: one

HANDLE VIA TALE

STEMS JOINTL Y

----1~WiI::E:fti.S}rnr:LEETO FOREIGN NATIONA

381

Tap SECRET l:JMBRA

DOClD: 523682
TOP sFeRn I:JMBRA

REF lD:A523682

containing equipment that could not be used at all with classified information, and one . listingequipments that could be used only on an interim basis. USCSB took the issue to McNamara, whobecame an ally. In December 1964 he signed a directive imposing the policy DoD-wide. The reaction was consternation. Without waivers, some agencies would have to virtually close down. All would have to buy new equipment, that expense coming directly out of their O&M moneys. In many cases the cost of equipment would double - in some cases no fix at all could be designed, and the equipment would have to be scrapped or sold. The result was that many went straight for the waivers, and in the face of imminent operational shutdown, got them. Even most SIGINT sites had to operate under waivers for years as agencies scrambled to comply. 1011

GEOGRAPHICAL RETREAT
Certain reductions and consolidations in intelligence and communicationlrelectr~nics activ~tie8 in Turkey are feasible and desirable. Blanchard Study. 1963

The conventional collection system reached its point of maximum expansion in the early 19605. Then, like a star imploding, it began to shrink. The shrinkage was basically a product of two problems, one internal and one external. The internal cause was money. The Vietnam War, and President Johnson's domestic initiatives like the War on Poverty, began to squeeze the cryptologic budget (not to mention other DoD programs). By 1963 a serious international balance of paynients problem had already developed, and the far-flung conventional SIGINT collection system became a prime target for reduction. Directed to study the problem, NSASAB concluded in 1963 that technology to remote collection sites back to the U.S. did not yet exist, except for the technique of recording signals on wideband tape and transporting the tapes back to the CONUS for transcription. Since this did not in most cases meet timeliness requirements, overseas reductions would mean real reductions in srGINT collection capability. 110 The second problem was developing Third World nationalism. Many of the countries which hosted SIGINT collection sites were moving toward more independent foreign policies, and foreign troops on their soil did not play well in domestic politics. As the Vietnam War wore on, there was, in addition, a sense of diminishing American power in the world, and a feeling that it was better to move into a neutral camp, rather than to lean on weakening American military protection. These trends often manifested themselves in ademand that the Americans somehow "pay" for their rental offoreign space.

iOPSECkEi

UMBRA

382

DOCID: 523682

REF ID:A523682

I

E.O. 13526, section 1.4(c)(d)

I

Withheld from public release Pub. L. 86-36

I VI" ;)1:"'1\1: I UmDKA

Turkey In no country did these trends play out more forcefully than in Turkey. The Anatolian Plateau llad become·the ideal SIGINT collection platform. I

Turkey had been friendly to Americans since the end of World War II, and this friendship continued strong until the Cyprus crisis of 1963. Anti-Americanism rust made an overt appearance at that time, intergovernmental relationships were strained, and a Turkish mob burned the USIA library in Izmir.1l2

Leftist, anti-American factions, emboldened during the Cyprus crisis, became increasingly vocal in the National Assembly. By mid-1965these factions had succeeded in steering the pro-American government of Suleiman Demirel toward a reevaluation of the bilateral relationship with the United States.

was actually an airborne telemetry collection program using RB-57 aircraft newly available from the LlTI'LECLOUD collection program in Pakistan (see p. 386). The program was in its very early days, flying out of Adana, when, on 14 December 1965, one of the planes crashed over the Black Sea. The cause of the crash was (and is to this day) unexplained./ Weather was not the best, but did not appear to ~e ba' enough to cause the crash of a high-performance aircraft like an RB-57.
BIG RIB
j

I E.O. 13526, section

1.4(c)(d)

HANDLE VIA TALENT kE i dOU! cmeU! (:ON TROL SYSTEMS JOINTLY LEASABLE TO FOREIGN NATION

"---------" Withheld from public release Pub. L. 86-36

I

383

~

DOCln: .523682
lOPSECRETtJMIRA

REF ID:A523682

The unexplained crash resulted in a frantic American and Turkish search for wreckage, which the Soviets independently joined. Fragments of the plane were recovered. but nothing that would have provided clues to the cause of the crash. The incident hit the Turkish press and received wide play, amid leftist demands that the government throw the Americans out: Although the Soviets did not protest the crash itself, they called the search and rescue effort that followedit a "dangerous provocation." This merely oiled the flres of the Turkish nationalists, who contended that Turkey had becomea pawn in the chess game between the Americans and the Soviets. Following the Cyprus crisis by two years, and Kennedy's withdrawal of Jupiter missiles without consulting Turkey in 1962, the BIG RIB incident buttressed nationalist contentions that Turkey should draw away from American sponsorship.P!

I

E.O. 13526, section 1.4(c)(d)

I

Withheld from public release Pub. L. 86-36

lOP SECRET!:IMBRA

384

DOClO: 523682
E.O. 13526, section 1.4(c)(d)

REF lD:A523682
Withheld from public release Pub. L. 86-36
lOP SECRET UMBRA

Buffeted by rising nationalist sentiments in Turkey,c=:Jwas whipsawed by cost reduction efforts at home. A study by Lieutenant General W. H. Blanchard in 1963 had concluded that I

In July 1968 the DDR&E,I I informed General Carter that to meet McNamara's gold flow reduction targets, it would be necessary to close Trabzon and either Samsun or Sinop by fiscal year 1970. Carter chose Samsun, and soon Sinop was the only Black Sea collection site remaining.P' Pakistan To the east, Pakistan was an even more difficult case. The Pakistanis had drawn close to. the Eisenhower administration in hopes of getting the wherewithal to defend themselves against Hindu India. Eisenhower had a very different goal - to align Pakistan in an anti-Soviet alliance and, coincidentally, to obtain permission to use Pakistani soil for certain sensitive intelligence operadona. The Pakistanis did not much care about the USSR, but they cared very deeply about American military arms and agreed to all the conditions for purchase, I

I

I
Under Kennedy, relations between the United States and Pakistan plunged swiftly downhill. After the Sino-Indian War of 1962, Kennedy arranged to send-India military aid to help them defend against the PRC, but of course Pakistan felt the arms could be turned against them. Street demonstrations in Peshawar against the American presence did nothing to assuage fears for the safety of the Air Force people on the base. In March 1963, General Ayub (the Pakistani military dictator) began improving relations with the PRC , as a hedge against American indifference. Through the next two years it became increasingly obvious to the State Department that Pakistan was playing a double game and that it would accept aid from any quarter if it would improve its defensive position against India. 122

Withheld from public release Pub. L. 86-36

HANDLE

-

VIA!::r;~¥vOJ
If 1.0

I
'

E.O. 13526, section 1.4(c)(d)

E CO¥!NT Cgt=~:-rSTEMSJOI~Y ASABLE TOFOREIGN l\CS

385

. TOP SECRET t:JMBR:A

.-

'- -.-

-

-,-------------

-

-

DOCID: 523682.
1

REF ID:A523682
Withheld from public release Pub. L. 86-36

E.O. 13526, section 1.4(c)(d)

I_~:-called

I The new prognuu,l
was a unique international SlGIN'l' cooperative venture,

UTILE CLOUD,

Faced with increasing Pakistani nervousness about Soviet attitudes and an upsurge of militant Islam, the U.S. tried to make the arrangement more palatable to Ayub. To minimize the visibility of the base, NSA held up planned installation of an FLR-9·1

I
,.

·The India-Pakistani War erupted in September 1965, in the middle ofc=J precarious relations with Ayub. Indian air strikes hit near the city. Angry mobs roamed the streets of Peshawar, and American GIs, whose government was assumed by the Pakistanis to be in league with India, were restricted to the base. I

Nineteen sixty-seven was another bad year for American interests in Pakistan. Ayub regarded Lyndon Johnson as even less of a friend than Kennedy, and when the ArabIsraeli war broke out in June he offered aid to the Arab states. Once again militant Muslim mobs invaded downtown Peshawar, and Americans were restricted to the base.

r
I
1

Withheld from public release Pub. L. 86-36

-I
~DL! VflIl=::EXHOLE COW;::YSTEMSJOINTLY EASABLE TO FO 2 6NAbS

E.O. 13526, section 1.4(c)(d)

TOP SECRET l:IMBItA

386

DOCID: 523682 .

REF ID:A523682

I E.O.

13526, section 1.4(c)(d)

I

Withheld from public release Pub. L. 86-36

. TOP SECRET tJMBRA

By the end of 1967, Ayub had just about decided to dump the United States as a sponsor and go for either the l)'SSR or Communist China, depending on what kind of an aid package each could offer. In April 1968, Pakistan's minister of foreign affairs handed Ambassador Oehlert a note that Pakistan had decided not to renew the ten-year lease in Peshawar; thia gave the United States about a year and a half to get out.

Ayub would probably have reversed himself if the U.S. had provided Pakistan with a certain quantity of tanks and had downsized the Peshawar site to make it a less visible America, presere. This situation touched off a debate in the U.S. government over the value of vis-a-vis the tanks and overall U.S. policy toward the government of

I Pakistan. I

:=;J

The United States began a retreat from Peshawar that concluded ",hen the base was officially closed in September oC 1970. By that time. Ayub had been unhorsed by a new military dictator, General Agha Mohammed Yahya Khan, and Lyndon Johnson was no longer president. But neither Khan nor Richard Nixon was inclined to reopenl I

I

I
It had occupied the time of two presidents and dominated the attentions of the American ambassador in Rawalpindi. The issue had once again put NSA and CIA at sword's point.

Withheld from public release Pub. L. 86-36

HANDLE VIA TALENT YSTEMSJOlNTL Y ~NQOT~Wh\S:m:ETOFOREIGNNATION 387

E.O. 13526, section 1.4 c)(d

=r:gp SECRET l:JMBfbfc

DOCID: 523682
TOP SEC.a- unaat E.O. 13526, section1.4(c)(d)

REF ID:A523682

Withheld from public release Pub. L. 86-36

The loss of geography on the Asian subcontinent indicated which way the. winds of nationalism were to blow, and it gave a huge boost to the overhead collection program. In the long run it also gave impetus to efforts to develop remoting technelogiesj

Withheld from public release Pub. L. 86-36
Tap S&CRIiT

., E.O.13526, section 1.4(c)(d)
HANDLE vIA1:::

.

KUII9[$ (mum eON=SYSTEMSJOINTL ~
LEASABLE TO FOREIGN

.!

.

40Nkt;S

UMBRA

388

DOCID: 523682

REF ID:A523682
TOP SECRE I OMBRA

I E.O. 13526, section 1.4(c) I

Withheld from public release Pub. L. 86-36

RC·135

HANDLE VIA TALE
_Jl~~m:AiBLE TOFOREIGNA

L SYSTEMS JOINTLY

389

Tep SECRET UMBRA

DOCID: 523682

REF ID:A523682
Withheld from public release Piub.·L. 86-36

I

TOP SECRET I:JMBRA E.O. 13526, section 1.4(c)(d)

Airborne Collection The success of thel program in Alaska (first USAFSS use of RC-135s to collect CDMINT; see p. 312) prompted AFSS to ask for more RC-135s. After a lengthy struggle, six aircraft were added to the program, and all were initially ticketed for Kadena, Okinawa, to bolster a Far East collection program hard pressed to satisfy collection requirements in both Southeast Asia and the. Soviet/PRC/North Korean coastlines. The addition of the far more capable RC-135s pushed the RC-130 program farther down the priority list, and all eventually became strictly theater assets before they were phased out of the inventory in the early 1970s. It also meant that the airborne collection program would inevitably take on a stronger global connotation, with home basing at Offutt AFB in Nebraska and much less of a theater presence.l3S As collection requirements multiplied, so did AFSS airborne programs. Many . responded to the need to collect against] I and they were usually joint SAC-USAFSS operations. During ..the late1960s, airborne programs were pulled in different directions by conflicting requirements in Southeast Asia, I I Iand wars in the Middle East. For several years airborne SIGINT assets of the Air Force and Navy were frantically juggled to keep up with r_eq..:.u_l_r_em--,ents.
• L34

I

I
,-

I

I

E.O. 13526, section 1.4(c)(d)

I

Withheld from public release Pub. L. 86-36

HANDLE ~--l~rRl~L:EEASABLE

NTROLSYSTEMSJOINTLY TO FOREIG

lOP SECRET UMBRA

390

--------------------------------------~----------------------------

-

OOClO: 523682

REF lO:A523682
T-eP SEERET I:JMBItA

Many of the RC-130s were ultimately replaced by "mini-manned" U-2s. Receiver front ends were placed on a pallet that was loaded on board, and the I1ircraft served as a highaltitude intercept station, downlinking intercepted RF to operators on the ground. These programs were preceded, however, by an experiment using drones. Begun in 'Korea in 1971, the drone program (under a variety of names) never worked. The drones were vulnerable to antiaircraft fire, and it eventually became too expensive to keep replacing them.1S5 ,

The Wood Study

I E.O.

13526, section 1.4(c)

I

Withheld from public release Pub. L. 86-36

Budgetary pressures and the rise of nationalism in the Third World led to a series of high-level basing studies in the mid- to late 1960s. Aside from the NSA study that led to the closure o~ (see p. 349), the most significant was the so-called Wood Study, named after General Robert J. Wood, called out of retirement in 1968 to chair a Senior Interdepartmental Group (SIG) looking at the worldwide intelligence posture. The objective was to save money; the target was SIGINT .

I

. Wood felt that much of the expense of SIGINT was with the front end - the overseas bases. He put forth a litany of ways that SlGlNT could be done more cheaply, which would be repeated by future study groups. NSA should pour money into advanced technologies (such as sateIlites and remoting) that would reduce force posture overseas. It should place more reliance on Third Parties. It should develop transportable SIGINT assets. It should rely more on technical research ships (despite the relatively recent destruction of the Liberty and the capture of the Pueblo). And it should be much more aggressive about consolidating overseas field sites. There were very cogent reasons why SIGINT sites were spread so widely throughout the world; they related to propagation phenomena and a perceived need to diversify intercept in ease of attack. But these objections were drowned by the need to economize. The Wood Study increased pressure to "do something" about the huge number of sites, and the rust move was to further reduce assets in Germany. Thus the decision was made (it had been impending for several years) to close the three Army sites at Rothwesten, Herzogenau~ach

'~~~~
391
TOPSECR&TUMBAA

DOCID: 523682
lOP SeCRET UMBRA

REF ID:A523682

I E.O.
and Bad Aibling,I

13526, section 1.4(c)(d)

Withheld from public release Pub. L. 86-36

11--Country Ethiopia Morocco Taiwan Korea Philippines Thailand Vietnam Pakistan Turkey Greece Cyprus Iran

One interesting spin-off of the Wood Study was an assessment of political vulnerability in countries housing U.S. SIGINToperations. The chart rates postulated tenure (as measured by the Wood Study) and actual withdrawal dates.

_

Survivability

of SIGINTSites1S7 Actual Retention

Postulated Retention

indefinite 10 years indefinite. 10 years 10 years 10 years as long as war lasts 1year 5 years 5 years 10 years 5 years (depends on survivability of Shah)

6 years . 2 years 11years indefmite 13 years 8 years same 2 years indefinite 24 years indefinite 10years

To aSIGINTer used to an expanding SIGINTsystem, 1968 must have seemed like a shrinking world. General Carter, protesting late-decade cutbacks, protested "a pattern of subtractions from U.S. cryptologic strength."lS8 He fought reductions like ~ tiger. But the twin pressures of paying {or Vietnam and reducing the balance of payments deficit combined to trim the SIGINTposture no matter what Carter said. Thus base consolidations in Germany. Japan, and (to a lesser extent) Turkey tightened up the SlGINTwaistline. The pressure for this was budgetary, and it came from the top. Viewed {rom the standpoint of international geopolitics, however, the picture was a little different. Of the ten countries (above) that. the U.S. abandoned from an overt SIGINT collection standpoint, nationalist pressures were the clear culprit in seven cases and were at least partly responsible in two others. Thus, SIGINTreductions came from internal budgetary causes, while outright abandonment of a country resulted almost inevitably

HANDLE VIATALE ONTROL SYSTEMS JOINTLY OTRELEASABLE TOFOREIGN NA TOP SECRET UMBRA

392

DOCID: 523682
E.O. 13526, section 1.4(c)(d)

REF ID:A523682

I

Withheld from public release Pub. L. 86-36

1'8P SKRE'f HMIItA

from nationalist sensitivity. SIGINT sites were generally acceptable as long as they were invisible to the local population. ThuBthe U.S. was forced to close its site.in Thailand in . 1976, The lesson was clear, and it became a factor in the new remoting technology that was, even in 1968, picking up steam in NSA. . The Harrogate Experiment Manning the front end of the SIGINT system with civilians had long been an NSA goal. In the 1950s NSA sent integrees to SCA sites, but the numbers were never large, and as the decade wore on, the SCAs tended to get tougher on the idea ofNSA invading their turf. The CIA experiment in Cyprus (Project APPLESAUCE; see p. 92) was another attempt at civilian manning. But for an adequate rotation base, it would have succeeded. However, civilianization took on a life of its own, chiefly because of the advantages that could accrue. . . The most si collector . lcant advanta e was expertise. The SCAs had trouble training provide linguistic

Moreover, NSA could sometimes ~r-~~----~~--------~~~ ta ent that was hard to come by in the military world. Withheld from public release Pub. L. 86-36

I

A second advantage was retainability. Military retention rates, low "in the 19505, dropped even lower during the Vietnam war. NSA wanted tol I. employ civilian collectors and analysts at the front end of their system for many years. The Americans could not match the expertise found aq I 1r--E-.O-.-1-3-S2-6-, -se-ct-io-n-l-.-4(-C)-(d-)---';I.

I

The 1958 Robertson Committee initially considered a system of NSA-only collection sites, but withdrew the recommendation from the final report in the face of determined SCA hostility. Instead, the report recommended increasing NSA civilian presence in hardto-find skills and establishing roving NSA teams of experts to help out with special field site problems.' But even that proved difficult to implement, and civilianization appeared to be a dying concept. 138 This turf fight between NSA and the SCAs stopped civilianization cold until 1965, when a new factor emerged. The factor was Vietnam. By 1965 the drain on military manpower was becoming severe, In August, the Defense Department canvassed all its activities looking for jobs that civilians could do so that the military people in them could go to the war zone. The most severe pressure was in the Army, and Army stations were threatened with the most serious manpower cutbacks to support the war. Faced with rows or potentially unmanned positions. NSA proposed that it be authorized to coordinate a program of civilianization within the cryptologic community. After a heated internal debate at NSA regarding civilianization at Bad Aibling ~r Harrogate, NSA proposed the civilianization of Harrogate. 140

H~~

ABLETOFOREIGNNA~

393

lOP SECRET UMBRA

Tep SEEAETYUBA'

HatTop",
(PJc:luNd -_ain

Lathe earty day,
•••• and 9\1pport faclOO •• )

('oj

GO \D M
('oj

III Q H

o o
Q

DOCID: 523682 I E.O.
13526, section 1.4(c) Harrogate, ~

REF ID:A523682
Withheld from public release Pub. L. 86-36
'fOp SECRET UMIlItA

,.

I

I was an ideal candidate. \

~/~tedm

~I

the Yorkshire moors, Harrogate had originally been surveyed by ASA in the early 1950s. Construction had begun in 1956, and the site officially opened in 1960 as an ASA field station. A site in the United Kingdom was thought to be an attractive place for civilians to relocate. NSA moved rapidly forward, and the site converted to civilian status in August 1966, less than a year after it was originally proposed.i" Naval
SIGINT

Ships

The signal success of the Oxford against Cuban microwave communications during the . Cuban Missile Crisis resulted in a boom in the Technieal Research Ship (TRS) program. NSA's long-term TRS program included sixteen vessels. eleven Military Sea Transport Service (MSTS) charters and five of the larger Oxford-class Liberty ships. The Navy had an even more grandiose plan to build a TRS fleet from the keel up, at a cost of $35 million per vessel. They would have a cruising speed of at least twenty knots. But despite the giddy success of the Oxford, the numbers did not add up. For instance, it oost $13.5 million to convert a Liberty ship into an Oxford-class vesse1, but only $3.3 million to redo a Valdezclass MSTS ship.I"2 DoD was strapped for cash for the Vietnam buildup, and this kind of floating SIGINTpiatform,logical in theory, fell victim to the budget axe. Failing in the .big plan, the Navy opted for a far cheaper optio~. The idea was to convert some trawler-type vessels at very minor cost and outfit them for general intelligence collection, including (but not limited to) SIGINT. Their primary purpose would be naval direct support, with a secondary national tasking mission from NSA. They would call the vessels AGER (Auxiliary General Environmental Research). NSA opposed the program from the beginning. Some Agency seniors believed that it was an end run. around NSA's authority to control SIGINT. Nonetheless, the Navy converted the first AGER in 1965, calling it the USS Banner (AGER-!). The long-range program was to have twelve such vessels. When, in late 1965, the Navy went forward with a request to convert two more Banner-class trawlers, NSA opposed it, and Cyrus Vance, the deputy secretary of defense, sent the proposal back to the cryptologic community to resolve the conflict. .NSA and the Navy fashioned a compromise in which the vessels would sail sometimes on solely direct support missions, sometimes on hybrid national tasking and direct support orders. It would be a wholly Navy owned, manned, and protected program. The ships were smaller and less capable than the Oxford- or Valdez-class vessels, and as for speed, could not even make ten knots. They would be almost defenseless. but up to that time SIGINT ships had never been bothered by hostile forces. The Pueblo, which put out on its first operational voyage in December 1967, was an AGER-type trawler.l4S
I

~VIA

:=¥JIOI
e

ASABLE TO FOREIGN N ••• n l'IKLS

JmMUfFeo.r::SI'E~
395 'F9P SECRET I:JMBfU.

OOCIO: 523682
TOP SEERETl;IMBRA .

REF ID:A523682

TRS communications were, in the early years, bothered by crowding of the HF spectrum. To solve this problem, the Oxford, in February of 1964, demonstrated for the first time the feasibility of bouncing microwave signals off the moon from a ship at sea. This technique had been used first in 1959 between two stationary locations, Hawaii and Washington, but the technical problems involved in doing it from the deck of a pitching ship were daunting. Although the problem was considered essentially insoluble, Commander William Carlin White ofNSO managed to get the Naval Research Laboratory interested, and White, NRL, and NSA, all working together, gathered the equipment fora test. When the Oxford successfully communicated with the NSG site at Cheltenham, Maryland, a new era of naval communications was under way. Soon CNO-approved installation of this new gear (called TRSSCOM,or TRS Special Communication System) was programmed for the Belmont and Liberty, and plans were made to convert all TRSs to the so-called Moon Shot system.l44 I E.O. 13526, section 1.4(c) I TRSs became very popular substitutes for dry land SIGINT real estate, With nationalism on the rise and the 'United States experiencing declining popularity in the Third World, it was often the only platform available. A TRS was sent t?1 I TRSs were thrown into the Vietnam conflict, . essentially as augmentation for existing fixed sites. An Oxford-class vessel, the Liberty, was deployed to the Mediterranean durin the 1967 Arab-Israeli War.

/

I

I

Withheld from public release Pub. L. 86-36

c

~
~

In the flush of enthusiasm, the latent problems in the program remained hidden. Program flexibility led to scattershot deployments to areas where the technical database was nonexistent. Vessels were put against targets with exotic language requirements that the Navy could not meet. SIGINT crew training and expertise levels appeared to many NSAers to be declining in the face of so many short-fuse deployments to strange places. Command and control became convoluted, especially in war zones like Vietnam or the I I and at times it appeared that no one really knew who had control of. ,--------, TRSs in certain areas. Occasionally a TRS would wind up doing non-SIGINT work like Withheld from hoisting refugees aboard - this happened duri~g the Cuban Missile Crisis, and was public release ordered, but not done, during I I Further, TRSs had to compete, in '--_P_u_b_._L_._8_6_-_3_6----J essence, with even more rapid AFSS airborne assets. Often the airborne fleet won out because it could get there faster, and AFSS had better trained operators and linguists.148 Finally, and fatally, floating S~GINTplatforms proved to be not as secure as had been expected. The Liberty incident in 1967 (see p, 432) shocked a cryptologic community that had always assumed that American SIGINT platforms would be accorded the same courtesies that the U.S. gave to the Soviet S!GlNT trawlers. The incident was repeated (with variants) the very next year when North Korea captured the Pueblo. NSA support for the program was already crumbling because of the dispute over the control of AGERs. With the Pueblo, it completely died.

HANDLEVIATALENT ROLSYSTEMSJOINTLY ---MOT RELEASABLE TO FOREIGNNATION

TOP SECRET UM BRA

396

DOCID: 523682

REF ID:A523682
"FePSECRe:f tJM lltA

.

.

The program was goodin theory, and if the exeeution had been better, TRSs might still be around. It is still a goodidea today. but the Pueblo incident probably killed it forever. Withheld from

I E.O.

I

THE END OF HF1 public release 13526, section 1.4(c) Pub. L. 86-36 The decade of the 19608 led NSA inexorably into above-HF signal,s, more and more difficult to intercept, more and more exotic to process once intercepted. Fixation on the problem marked one v~ry difficult and expensive avenue, which would require complex intercept and processing gear and unconventional collection locations or platforms. The trend toward above-HF communications, especially microwave, radio relay, and communications satellites, marked another knotty problem for the cryptologic community.

I

During World War II, the Soviet Union's communications were estimated to be approximately 50 percent HF and 50 percent landline. I I

This pessimistic assessment of Soviet communications trends was not immediately borne out.

Still, all long-range forecasts agreed with the above-mentioned 1968 Eachus Report. NSA had been worrying about this problem for some years, and the Agency was in the process, in the late 1960s, of designing and fielding systems that would accommodate the expected surge in above HF communications.

The 19571aunch of Sputnik created an immediate requirement to track Soviet ESVs (earth satellite vehicles). The thought that the USSR might have an ESV in orbit whose

I E.O.

13526, section 1.4(c)


~
[EC~LY ASABLE TO FOREIG .

Withheld

from

~
1

public release
Pub.L.86-36

397

lOP SECRETUMBRA

-----------

DOCID: 523682
TOP SECRET YMBRA E.O. l3526, section 1.4(c)(d)

REF ID:A523682

Withheld from public release Pub. L. 86-36

existence and purpose were unknown was intolerable

,

Withheld from public release Pub. L. 86-36

STONEHOUSE

I E.O. l3526, section 1.4(c)(d) I
. specifically

".

Th e only site ever built for space collection (as opposed to missile telemetry) was STONEHOUSE, collocated with the ASA HE intercept site at Asmara. Set on the high equatorial plateau of Ethiopia, it was originally manned primarily by ASA people, with a small complement of NSA civilians and contractors. It sported two huge dish antennas 150 feet in diameter. In 1972 ASA got out of the business, and the site was left permanently for NSA to operate.lS1

HA~LE

~lA

;:;EiNT

ICFYHO! F eeMIN l' CONTROL SYSTEMS JOINTLY T RELEASABLE TO FOREIGNNAIIONAJ:;S

fe'SECR!TUMIUb\

398

Tep5EER£TY~'8M

I E.O. 13526, section 1.4(c)

Withheld from public release Pub. L. 86-36

N OJ
III
(Y)

N LO Q H

••.•• Nb~I.~ ASABLE TO FORE. 399

o

o
Q

OOClD: 523682

REF lD:A523682

STONEHOUSE Asmara, Ethiopia

'fap SEeR!T UMBRA

400

DOCID: 523682

REF ID:A523682
-

I

E.O. 13526, section L4(c)(d)

I

Withheld from public release Pub. L. 86-36

-....

'

.•......"..

I

I

By the early 1960s the United States had established that ESVs were potentially very' useful communications vehicles. On 31 August 1962, President Kennedy signed the Communications Satellite Act which sanctioned the Comsat Corporation to establish U.S. participation in a global network of communications satellites. Both Intelsat and Comsat were organized soon after to develop the systems to provide Comsat vehicles for international, as well as national. use. The feasibility of high-quality TV and voice transmission via satellite. was proved during the Tokyo Olympics of 1964, and the first .American Comsat, called Early Bird, was launched in April of 1965. It was so successful so fast that by 1966 the U.S. projected that Intelsat-assigned circuits would increase from 585 then to over 6,000 ten years later .152 The Soviets, too, understood the implications ofComsats. In 1966 they launched three satellites in elliptical orbit, which they called Molniyas, and began beaming multichannel and television signals to distant users. These early systems had sixty channels, but most were, in those early days, vacant. 53

Withheld from public release Pub. L. 86-36

I

E.O. 13526, section 1.4(c)(d)

I

HANDLE VIA 'l'ALEN".!' I(EVUO! E COMlN'l' S6I( 1ROLSYSTEMS JOINTLY NeT Ri::LEASABLE TO FOREIGN NA I JONAf;S

401

'fOP SECRET YMBRA

DOClD: 523682
1'6115I!CIU!T tJ MIRA

REF lD:A523682

E.O. 13526,section 1.4(c)(d)

Withheld from public release Pub. L. 86-36

But Army regulations required extensive s,upportfacilities for the troops, and the cost and visibility of the site quickly got out of hand. It died a sudden death at the hands of the budgeteers.lse

I E.O. 13526,sectiou 1.4(c)(d) I
Overhead
-:

Since the science fiction writings of Arthur C. Clarke in the 19308 and 1940s. it had been an American dream to place a reconnaissance satellite in orbit around the earth .. At the end of World War II, General Curtis. LeMay, then deputy chief of staff for Research and Development for the Army Air Corps, commissioned the Rand Corporation to do a study on the feasibility ofjust such a project, The Rand'study, dubbed Project FEEDBACK, proceeded in secret for eight years. It was (many turned over to the Air Force in 1954, coincident with the Eisenhower administration's thorough examination of the strategic warning dilemma under the Killian Board (see p. 229).158 •,

Withheld from public release Pub. L. 86-36

HANDLE
NOT RELEA$ABLE TO FOREIGN

NTROLSYSI'EMSJOINTLY

TOP SEC-RET YMBRA

402

DOClD: 523682

REF lD:A523682
lOP SECRET I:JMBRA

The Technological Capabilities Panel CTCP) on the Killian Board recommended that Eisenhower proceed with the highly compartmented U-2 project being developed by Lockheed. In addition, the Intelligence Committee of the TCP, chaired by renowned optics scientist Edwin Land, recommended that the United States begin to develop· reconnaissance satellites. This also got Eisenhower's approval, and it proceeded along a paralleltrack.P" The Air Force immediately began developing an intelligence satellite program. The prime objective was photoreconnaissance, but the initial operational requirement, published in 1955, also contained provisions for an ELINT package. 160 From the beginning, the program was beset by competing jurisdictions and security concerns, The Air Force, the Navy, and CIA (the latter by virtue of its domination of the U -2 program) all designed entries into this new intelligence sweepstakes. The prize for the most successful system was money and people, both on a very large seale. Overhead reconnaissance loomed as the biggest potential spender in ·the intelligence system. Once the Soviets launched Sputnik in 1957. American attenti~n focused on a competitor. Although the main objective would be reconnaissance, it would have been imprudent to be up front with this. So 'in 1958 Eisenhower decided that the Americans would publicize their satellite program as a purely peaceful program, with scientific objectives. The first program, called Discoverer, was pushed ahead as an overt "white" program. Reconnaissance would be a "black," covert program, with classified payloads attached initially to the Discoverer vehicles,181 The way Eisenhower created it, the new overhead program had a divided jurisdiction. The Air Force was to build and launch satellites, while CIA was to process. the photography. The first processing center was actually set up by CIA to process photos from the U-2. Called NPIC (National Photographic Interpretation Center), it was established in the old Steuart Motor Car Building at 5th and K St., N.W., in downtown Washington. The CIA's Richard Bissell was in charge of the program, and Arthur Lundahl headed NPIC.182 Meanwhile, the Air Force had set up operations on the West Coast. In October 1955, the Air Force moved its satellite development project from Wright-Patterson AFB in Ohio r to Inglewood, California, locus of their ballistic missile development. This, was done in order to insure that both programs remained in synch and that they would not compete for boosters. To control satellite operations, the Air Force chose to collocate with its prime contractor in California. UIS

HANDLEvIA

~~:ZrOJ
If

ISmW~T:r e61'1i*~SYSTEMSJOIN~ ASABLE TO FOREIGN NA:f:B

403

.TOP SECRET UMBRA
;

.

DOClD: 523682

REF lD:A523682

.. = '" ... ~ ...,.'E '0
,.c'""" .••• <::\
0<::\

1=OPSECRET !:IMaM The Air Force Programs

ELINT

]~
'; -;, ~ " c~

-==0
.-= ~

The rust SlGlNT packages were a product of SAC's desire to support the SlOP, or Single Integrated Operational Plan, the plan for nuclear war with the Sino-Soviet Bloc. For SAC ,S ~ to design penetration routes for its bombers, it had to know where the Soviet radars were ="" ~ ! ~ ~ and what they were capable of. At the time (the mid-1950s), ELINT was blissfully ",1:= ~ .~ ~ fragmented, and NSA was a COMINT agency. SAC proceeded with its program ; ~ . :: unchallenged.

-=..:~

1"

While all this was going on,l working in CIA's Office of ELINT, became concerned that the ELINT payloads might not be ready for the firsf launch of a ~ u;:5 photoreconnaissance satellite. c=Jconcluded that a small, interim, piggyback payload could be designed and ready for the first launch. Its only mission would be to detect threat radars The interim program was called k.and it became an end unto itself E.O.13526, section 1.4(c) Discoverer experienced all sorts of disasters, as payload after payload plunged into the Withheld from ocean, was fired into an unrecoverable orbit, or just exploded on launch. But when the rust public release photoreeonnaissance payload (Discoverer xun actually achieved its mission and was ~. L. 86-36 .snagged on reentry by elated Navy frogmen in August of 1960,\
vi .

~ ]j

I

Lj

V

I
Program Mana ement E.O. 13526, section 1.4(c) I Iremained an Air Force program, and SAC did the early signals processing. But in 1961 McNamara appointed Eugene Fubini to look into the proper relationships in the SIGINT satellite program. The Fubini committee concluded that the SIGlNT satellites had to be a partnership. The satellite payloads and their booster systems remained an Air Force and NRO concern, but processing and reporting became an NSA responsibility. This decision led to a series of fragmented agreements between NSA, on the one hand, and the various satellite operators on the other, regarding the precise terms of NSA's participation in each program .167 One beneficial result of the Fubini study was the signing, in September 1961, of a formal agreement between NSA and SAC regarding the processing of ELINT from the Air Force program. Essentially, they agreed that a certain amount of parallel processing would be done - NSA to benefit the intelligence community, SAC to support the SIOP.l68 In 1961, just before leaving office, Eisenhower set up a special compartmentation for overhead reconnaissance. Called Talent-Keyhole, or TK for short, it covered both the ongoing U-2 program and the nascent satellites. CIA, which exercised general supervision of

Tap SECRET "'MIRA

404

DOClD: 523682
_

REF lD:A523682
TOP SECRET "'MaRoA

,.c'-'

O,IlI'l

o

:: O~ ~

~ . the programs, controlled the clearances. The plan listed a total Of~ o :::: which NSA would have exactIyD (The Byeman compartment was set up two years later ~ ~ to handle technical aspects of the satellite programs.)l89 ~ N The next year the two main players in the satellite reconnaissance game managed an lI'l :;. accommodation. The CIA and Air Force agreed that a new multiagency program would be O· established, called the NRP (National Reconnaissance Program). The CIA component of ~ the NRP would be headed by Richard Bissell, who had managed the U-2 program from its infancy. The Air Force component would be housed in a new organization directly responsible to the secretary, called SAFSS (Secretary of the Air Force Space Systems), with Joseph Charyk as its head. The same directive established a joint agency, the National Reconnaissance Office, or NRO.170 NSA was still a minor player. It had very few cleared people, and its only . responsibility was to process and report ELINT data. Even though NSCID 6 gave it significant responsibilities in both ELINTand COMINT,NSA had no official role in the tasking of reconnaissance satellites.!" Satellite tasking was then handled by COMOR (Committee on Overhead Reconnaissance); a USIB subcommittee. COMOR was concerned at first only with PHOTINT,but as the EUNT packages broadened in function from purely a vulnerability assessment to wider intelligence applications, ELINTtasking came to be done by the SIGINT Working Group (SWG) of COM OR. 172· . SWG tasking tended to be very specific, and mission ground stations found it almost unworkable. NSA was used to having USIB set general collection priorities, which the NSA tasking messages would flesh out. One of the problems that bedeviled the overhead . program for years was the lack of sufficiently flexible tasking documents. 173 In 1962, reacting to this situation, NRO set up a Satellite Operations Center (SOC) in the Pentagon. NSA predictably saw this as another intrusion into its authority to task SIGINTcollectors, and it soon was sending representatives to the SOC to represent its interests.l'TS Tasking continued to be handled by COMOR until Huntington Sheldon of CIA became chairman of the SIGINTCommittee in 196'7. Sheldon lobbied USIB to split apart SIGINTand PHOTINTsatellite tasking and succeeded in getting COMOR divided into two pieces. A new USIB committee, COMIREX (Committee on Imagery Requirements and Exploitation) tasked satellites, while another committee, SORS (SrGINT Overhead Reconnaissance Subcommittee) tasked the ELINTand COMINT payloads. 175

I ~ =
-< >.
~ ~
" ~ o~
0,1

=<_
~ ~ ~ •.. ~...,. 0,1 ._ I:: ~ ~ .~
~

~]I ~
~ ~
~

~

::a!:: U ;:::~oo
U ~

HANDLE VIA ASABLE TOFOREIGN NATI 405

EMSJO.lNTLY

TOP SECRET I:JMBRA

DOClD: 523682
fep SEeKET UMBRA

REF lD:A523682

The Advent of Overhead COMmT I E.O. 13526, section 1.4(c) I

Withheld from public release Pub. L. 86-36

Although satellites were originally the domain of PHOTINT and EUNT, NSA was studying possible COlifINTapplications. A 1959 study by NSA analyst concluded that it would be feasible to collect COMINTsignals from the EUNT packages aboard Air Force satellites.!"

I

I

Beginning on thel

in the early 19608, experimental Isystems.

COMlNT-targetted payloads piggybacked

TheD

Payloads

\

In the early I Idays engineers designed a specialized payload that would do ionospheric mapping They realized during the development phase that the payload could be injected into an orbit different from the mother payload; Since the objective was independent of satellite electronic defense, there was no special reason for it to stay with the main payload. "This led to the development of a separate program,1

I

I
I

I E.O.

13526, section 1.4(c)
CONTROL SYSTEMS JOINTLY RELEASABLE TOFOREIGN

Withheld from public release Pub. L. 86-36

HANDLE VIA TA OT

'FeP SECRET YMILRA

406

DOClD: 523682

REF lD:A523682
lep SECRETtJMBRA

I E.O.

Program C 13526, section 1.4(c)

I

Withheld from public release Pub. L. 86-36

The Navy's share of the satellite pie was called Program C. (Program A was Air Force and Program B was CIA.) But, though it was last in the alphabet, it had the first successful launch of an ELINT payload on 22 June 1960. Moreover, the Navy designed a unique program that outIastedall the others. 180 The program was actually conceived early in 1958 by Naval Research Laboratory engineers. They designed a program to receive I I c=Jand transmit this intercept in real time to Navy ground sites I ~ I I These ground sites were self-contained units called ESV huts, mounted on vans that could be moved around quickly. The huts would be located primarily at NSG field sites, but because of geography it might be necessary to use sites owned by other organizations.181 Most sites acted as "dumb" terminals, receiving and recording the signals. Recordings were shipped to NSA for analysis. 182
. I

This early program, which was solelyunder the auspices of the Navy, was called OYNO, and was referred to in unclassified terms as GRAB. It was the first to document the extremely rich radar signals environment in the Soviet Union. But to some extent it was a targetting anomaly. The Navy was collecting signals of interest to all services and the CIA, but the program was not doing ocean surveillance. In 1962 the program was subsumed within the overall satellite collection system as Program C, and it was reriamed
POPPY. 183

In 1966, overhead photos of Soviet ABM installations showed considerable progress toward site construction, I This became a matter of grave concern to the President's Scientific Advisory Committee, and a study group was appointed. If ABM systems were not the highest priority target up to that point, the committee made them such. A series ofc::::::::Jpayloads was developed and launched rapid. fire to respond to the concern. l~

I

Program C was also affected./

Withheld from public release Pub. L. 86-36

I E.O.

13526, section 1.4(c)

I

407

T ONTROLSYSTEMSJOINTL Y

fOPSECRETtJMBRA

DOCID: 523682
Tap SECRET UMBRA

REF ID:A523682
Withheld from public release Pub. L. 86-36

I E.O. 13526,section 1.4(c)

IL...--_-=--RUNWAY

_

As for the control issue, that was solved by moving tasking control to NSA. NSA set up a new facility called SSSC .(SIGINT Satellite System Control) to provide technical support and tasking guidance to the program. Some non-NSA USIB members were less than pleased because SSSC amounted
,-----------,

Withheld from public release Pub. L. 86-36
HAND T RELEASABLE TO FOREIGN.

I E.O. 13526,section 1.4(c)
ROL SYSTEMS JOINTLY

laP SECRET I:IMBRA

408

DOCID: 523682 I E.O.
13526, section 1.4(c)

REF ID:A523682 I
TOP SEGRET UMBRA

Withheld from public release to a de facto"delegation of tasking controi to NSA. The direction was irreversible, however, Pub. L. 86-36 l88
and by 1972, representatives from the

sac in the Pentagon

had moved to SSSC.

I

The program was not popular. downtown, and it came under repeated attack. When this happened, Adniiral Gayler himself indicated that he wanted to attend the NRP Executive Committee meetings to defend the program. At his very rrrst meeting, Gayler went on the attack, not just defending the money that had been put into the system to date, but demanding more money to launch more satellites and to buy more processing

equipment.

I

)

RAINFALL

I

The RUNWAY program was encountering such ferocious opposition in Washington partly because CIA already had a competitor. The CIA project had been initiated by Albert "Bud" Wheelon, who had come to CIA during the early years of the Kennedy administration. A brilliant and aggressive administrator, as well as a top-notch scientist, Wheelon had been newly installed as John McCone's director of science and technology when he read about the Syncom II geosynchronous satellite. I from Soviet missile tests ~as the number one U.S. intelligence 'priority, Wheelon wondered ua geosynchronous satellite could be placed. in an orbit that would continuously look down on Tyuratam and Sary Shagan. Wheelon pressed his idea with McCone, who approvedl for a pilot study. ISO "

I

I'

I

The project was fraught with tremendous risk. It would be hideously expensive, the most costly intelligence system ever mounted. I I

I

I

L..----..,.....J I An immense antenna would be required - a scientist calculated that it would have to be at least seventy-five feet in diameter, the largest such object ever unfurled in space. The Department of Defense, wanting CIA out of the satellite business anyway, opposed it from the beginning.lel

Albert "Bud" Wheelon

Withheld from public release Pub. L. 86-36

I E.O.
OLE .,~ •••..••••EASABLETO EMS JOINTLY

13526, section 1.4(c)

-

-

."
TOil SEtRET YMBRA

409

DOClD: 523682
TOP SECRETtlM8M

REF lD:A523682
Withheld from public release Pub. L. 86-36

I E.O.

13526, section 1.4(c)

CIA cleared no one at NSA. Thus. CIA knew about NSA'g nascent plans for RUNWA'!, but NSA did not know about CIA's plans for a similarly disposed geosynchronous satellite system, I ~ This situation changed in the late summer of 1965. because General Marshall Carter migrated from the position of deputy DCI to director of NSA. When he arrived, he arranged to clear a handful of NSA people and sent them to CIA to learn about the RAINFALL program. 192 The road proved rocky in the extreme. CIA wanted no NSA partipation at all, and in the early months did a great deal to shut NSA out. But a breakthrough of sorts occurred .in December of 1965. when to c ear the air. Through these hig L.....re-v-e"'--co-n'""'ta---'cts""-, t""h-e-:t-w-o-o-r-ga-m ..• ·'za-t'"i-on-s ..• b'e-g-a-n"";'jo-=i,-n7"t p-,I'a ....• nning.l9S NSA immediately suggested that COMINT become an ancillary mission. After a period of hesitation, CIA accepted the proposal and gave NSA the job of collecting what COMINT they could from a bird whose job was TELINT,not COMINT. Through the Director's Advisory Group for ELINTand Reconnaissance (DAGER), headed by Charles Tevis, NSA negotiated the details of their participation in the RAINFALL program. NSA got a COMINT processing subsystem and an EUNT subsystem! and when the money for those systems was cut from the budget, NSA allocated CCP funds. DAGER was also instrumentall

I

1_.-----------'
'-and about half of the TELlNTcrew.19S --' Eventually

I

NSA provided all the COMINTstaff

SIGINTsatellites were the wave of the furtu:::r:...:e"'--=a:.::n:::d:....t=h:..:e"'-=~:..:....=..:===::.:a--=::..::.., opportunities for access to the Soviet Union. But it also offered a significant ne~ battleground for the control of intelligence resources. CIAAir Force conflicts over the control of imagery became well known to the American public through the publication of such books as William Burrows's Deep Black. Far more obscure, but just as fierce, was the competition between NSA and others (especially CIA) over the . ownership and control of SIGINTpayloads. It eventually settled down to a series of compromises based on the areas of respective technical competence. But the early years, when these compromises were still in the future, were not easy.

~--~--~----~------------------~------~~~----------~

Withheld from public release Pub. L. 86-36

I E.O. 13526, section 1.4(c)

RELEASABLE

TO FOREI

r

IlOlfA:hS

1'OP SiCRET !;IMBM

410

- -

-

,"

OOClO: 523682

REF IO:A523682
lOP SECRET tJMBItA

NSA'S FOREIGN RELATIONS
Tbey [Third Parties] should not be used for economy rea80nBto supplant vital U.S. capabilities. However, rapport with Third Parties should be developed bases in the future. Eaton·Committee, 1968
all

insurance against the

1088

of U.S.

With the cryptologic budget being cut back in practically every area except Southeast Asia, NSA in the mid-1960s gave a serious relook at what the Third Parties could do for the U.S. Every budget exercise resulted in an increased determination to bring foreign countries more fully into the process. By the late 1960s the budgeteers demanded that

I E.O. 13526, section 1.4(c) I

Withheld from public release Pub. L. 86-36

The Eaton panel in 1968 (see p. 479) backed NSA's contention and stated that Third Party collection should complement U.S. collection.lIIs General Carter, fresh from his stint at CIA, placed Third Party relationships on center stage, and he was reputedly the first NSA director to permit Third Party representatives into the NSA complex, But Carter's attention to foreign relationships brought NSA up against CIA's long-standing prerogatives in this area. Although NSA began to take a mor~ active hand in several of the relationships, the disputes were not resolved during the decade, and resol ution was put off until the late 1970s.m

J:lANDLE VIATA ~LETOFOREIG

ROLSYSTEMSJOlNTLY

411

1'OP SI!CK!T tlMBftA

OOClO: 523682
lap SEeIt!T tJMBRA

REF lO:A523682

I E.O.

13526, section 1.4(c)

I

Withheld from public release Pub. L. 86-36

Germany The Reinhard Gehlen organization (the BND) was one of NSA's most lucrative Third Party sources during the 19608. But there were serious problems within the organization itself which limited its utility and caused the Agency to keep it at arm's length. Most of the problems revolved around security. Basically the BND, like almost all West German governmental organizations, was penetrated and publicized. The problems began in 1952, when a leftist journalist named Sefton Delmer published a highly critical article in the London Daily Mail entitled "Hitler's General Now Spies for Dollars. Delmer appeared to get much of his material from one Otto John, who had headed the West German equivalent of the FBI until his defection to East Germany. John was, in 1952, engaged in a bitter bureaucratic struggle with Gehlen over the control of'intelligenee.P"
II

Thingsjust wentfrom bad to worse. In 1953 one Hans Joachim Geyer, a member of the Gehlen organization, fled to East Germany with the names of Gehlen agents. Within hours more than 300 Gehlen agents had been rounded up, and East Germany exposed the "spy ring" in a resonating press conference. Geyer had been passing classified documents to the KGB for several years, although it appears that he was not involved inSIGlNT.201 But the coup de grace was not administered until 1961, with the exposure of Heinz Felfe. A rising star in the BND, Felfe had worked for the KGB since the early 19508 and had passed thousands or documents. He worked in counterintelligence, not SIGINT, but his

HANDLE VIA TALENT KEY L SYSTEMS JOINTLY ASABLE TO FOREIGN NAnON

TOP SEeREl YMBAA

412

DOCID: 523682

REF ID:A523682
1'9P SEEREl' l:JMBAA

access was very wide, and nothing in the BND was really safe. The exposure of Felfe in November 1961 led to a prolonged and highly public spy scandal, during which it was J . revealed that the BND had been thoroughly compromised by the East Bloc. At the same time Gehlen himself was involved in a public row with Franz Josef Strauss, the minister of defense. His inflexibility in dealing with outsiders, and his lack of appetite to rid the BND of East Bloc agents, ended his effectiveness. Gehlen continued to head BND until 1968, but withdrew more and more from active management.202

I E.O.

13526, section 1.4(c)

I

Withheld from public release Pub. L. 86-36

This did not stop NSA-CIA competition. However, it did lessen the points of triction and charted the way for a gradual CIA withdrawal from the day-to-day intricacies of Third Party SIGINT exchanges. As Third Party SIGINT became more important and more timesensitive, this was a natural and evolutionary step.

I E.O.

13526, section 1.4(c)

I
Withheld from public release Pub. L. 86-36

HANDLEVlA TALENT
NOT RELEASABLE TO FOREIGN

NTROLSYSTEMSJOINTLY

413

TOil SeCRET UMBItA

DOCID: 523682
T9P SEEAET UMBRA

REF ID:A523682

[ E.O. 13526, section 1.4(c)

I

Withheld from public release Pub. L. 86-36

Tap SECRETtlMBRA

414

DOClD: 523682 I E.O.
13526, section 1.4(c)

REF lD:A523682

Withheld from public release Pub. L. 86-36
TOP SECR!T UMBRA

NSA and CIA in the Third Party World By the end of the 1960s, the control of Third Party quite muddled. I
SIGINT

relationships had become

Withheld from public release Pub. L. 86-36

NSA and GCHQ As for the American-British relationship, virtually inseparable by 1970./ the two

I E.O. 13526, section
SIGINT

1.4(c)

operations had become

I

~

HANDLE VIA TALENT KEYH LSYSTEMSJOINTLY LEASABLE TO FOREIGN NATIONA

415

lOP SECRET UMBItA

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

DOClD: 523682 I E.O. 13526,section 1.4(c) I
'fOp SEeRE1' UMBItA

.A523682 REF lD .

Withheld from public release Pub. L. 86-36

Notes

1. NSA Retired Records, 288Z, 199104; 2. CCH Series VI.C.1.27; KeDDeciy Library material in CCH Series XVI. 3. CCHSeriesVI.C.1.27. ~. CCH Series Vl.EE.l.25; V1.C.l.27; Frost interview at Kennedy Library, Bollton. 5. Anne S. Brown, "The Consolidated Cryptologic Program and its Predecessorll, 1957-1975,~ unpublished NSA history available in CCH. 6. CcHSeriesVI.EE.1.l2. 7. Brown, CCH Seriea VI.C.l.27. 8. S. D. Breckinridge, Th.e CTA 11M th.e U S.Tntclligeru:e System (Boulder, Co.: Westview Press, 1986), 58-9. 9. Kirby interview, Tordella interview, Ward interview, CCH Series Vl.D.2.S.
10. Blake biography in CCH Series V[D.2.S.

11.1

1 III,39,43.121-24.;

NSA Retired Recorda, 42068,A66-77.

12.1
13.1

1
111I,125.

14. "NSA's Telecommunications Problems ... Electronic Security Comm~nd•... "

,~v. III,

13, "Chronology of Signific:ant Events in the History of •

Withheld from public release Pub. L. 86-36

TOP SECRET.UMBRA

416

DOCID: 52r3~6~8~2~ __~
Withheld from public release Pub. L. 86-36

REF ID:A523682
TQP SECRET !:IMBM

15. Oral interview with MaJor General (USAF. Het) John E. Morrison. 10 Aug 1993. Charles Baker and Tom JOhnlOn, NSA OH 24-93; CCH Series X.H.26. 16. Oral iDterview with Milun Zaslow. December 1993. 17. CCHSerieaVl.0.l.8. 18. CCH Series VI.0.l.l0. 19. CCH Series Vl.0.1.2.; NSAlCSSAn:hives. ACC 10460, CBRI 51. 20. CCH Series Vl.D.2.6.; Vl.EE.l.12. 21. Memo. subject; "Material for Special Study Group Staff," 27 September 1967. in CCH Series VI.C.l.27.; oral history with RADM (Ret) Donald M.Showers, 5 May 1992, by

I

~

22. CCH Series VI.0.1.2. 23. "Plans for Establishment of a SIGINT Missile and Aatronautic:s Center." August. 1962. in WI files. 24. Ibid.;i rDEFsMAC - A Community Asset 1964-1989", in CCH Series VI.A.l1.

25. DoD Dir 805100.43. 1964; "DoD Review of Missile and Space .•• ," Oral interview with~Charle8 C. Tevis, 19

Aug87,byRobertFarley.1

land!

~NSAOH21.87.

26. DEFSMAC Memorandum 11,4. June 1964; ~DefenseJSMAC: Defenae Special Missile Astronautics Center," Cryptologic MilutoM •• February 1967. 27. Anne S. Brown, "The History oftha NSA SIOINT Command Center and its Predecessors, 1949-1969," in CCH Series VI.E.5.22; written comments submitted by David Boak, Oct 1994, available in CCH. 28. Brown,
"NSA

Commsnd Center:"

29. Brown, oral bistory interview with OH 31.87; Eastman interview, CCH Series Vl.E.S.I0. 30. CCH Series Vl.C.1; VI.E.5.10. 31. NSAlCSS Archives, ACC «073, H03·0602·5. 32. Ibid. 33. Ibid. 34. Ibid; CCH Seriea Vl.FF.l.14.

I

~15 Dec 1987, by Robert Farley andlL-

J~ NSA

35. Oral bistory interview with Jack Holley, 1993; NSAJCSSArchives, ACC 31039, G11'()202·3. 36. NSAlCSS Archives, ACC 44073, H03-0602·5.

Withheld from public release Pub. L. 86-36

37. Jack Holley interview; video at CACL60, BOX 536.

38.!

!interview.

39. NSAlCSS Arcbives, ACC.«073, H03-0602-5. 40. Kennedy Library mea in CCH Series XVI. 41. Ibid. 42. Ibid.

HANDLE YIATM.E;;=OLECOMINT~ NOT r BLE

-

ru FOKE
417

, l'gp SECRU OMBRA .

DOClD: 523682
lOP SECRETtlMBItA

REF lD:A523682

43. Oral interview with David Y. McManis, 18 November 1986, by Tom Jolmson and Gerald Haines, NSA OH 3486. 44. CCH Series XII.H.28. 45. McManiainterview. 46. Interview with Rostow, 22 March 1993, Austin, Texas. 47 Oral interview with Richard Helms, 4 April 1969, by Paige Mulholland ofLBJ Library, copy avail. at JFK Library. 48. &stow interview. 49. McManieinterview. 50. McManis interview; LBJ Library National Security File, Austin, TeUB. 61. 'Mc:Manisint.erview. 52. Brugioni, oral interview with Carter, 3~ October 1988, by Robert Farley, NSA OH 15~8. 53. Carter interview. 54. Carter interview; NSAJCSSArchives, ACC 37911, H03·0306·2. 56. Carter interview. 56. Carter interview; Oftlce of Career Development; "Review ofthe NSAlCSS Professionali2ation Program: June 1987; CCH Seriss VI.C.l.27.; QlI4rluly MaMgement Reu~w ,?T. 67. Carter interview. 58·1'-;:;--:-=--:---:,:=-:~:-:-:--:--.....-----;--:-Series, Special Series, (NSA: Ft. Meade n.d.).

1 Cryptologic

History
L-

Withheld from public -release Pub. L. 86-36

_

59. NSAICSSArchives, ACC 22636, CBJM 41; ESe, -A BriefHistoryofAFEWC,"1977,atAlA. 60. CCH Series VI.D.2.6.;interview with Mr. Hawes ofthe Airport Square Companies, September 1993.
61. Oral interview with Cecil PhiUips andl

~1 May 1993, by Charles Baker and Tom Johnson, NSA OH

14-93. 62. Ibid.

I E.O.

13526, section 1.4(c) I

63.1 ~•A Cue History of the Iproject and its Relationship to a Systems Approach," unpublished manuscript in CCH c:ollection; NSAJCSSArchives, ACC 32561, H01·010l·3. 64.0. Phillips andDinterview. 65. NSAlCSS Archives, ACe 37741, G14-0306.5; ACC 43367, CBOA 38;D 66. DPhillips 67.0 68. "NSA's Te19COmmunicatiQosProblemli.1952-1968," NSA historical study in CCH Series X.H.4. andDnterview. Phillips, andD interview.

I

Withheld from public release Pub. L. 86-36

1=eP SECRET I:JMBRA

418

DOCID: 523682 I E.O.
13526, section 1.4(c)
Problems

REF ID:A523682
IOPSECREIOMaRA

I
.Withheld from public release Pub. L. 86-36
"Evolution

69. "Communications

•.. ," NSAlCSS Archives, ACC 24188, H02·0207·4.

70. NSAlCSS Archives, ACC 33705, H01-0108·6; ACC 24188, H02-0207·4. 71·1 ~videotape lecture on NSA communications history.

72. NSAlCSS Archives, ACC 24188, H02·0207 ·4; ACC 33'107, HO 1·01 08·6; ACC 32432, H01.0411.1; ofSlGINT CommWlicatio~ Support,1949-Present \l978r in CCH Seriea Vl.~1.10.

73. Phillips interview. 74. Samuel Snyder, "lDfluence of Douglas Hogan, "General May 1986, unpublished manuscript

U.S. Cryptologic
in CCH meso

Organi%ations

on the Digital Computer

Induatr'y;

SRH-003i 23

and Special Purpose Computers:

A Historical

Look and Some Lesaons Learned,"

75. Ibidi NSA oral history, circa 1968, with General Carter. 76. Hogan. 77. Phillips interview, 78. See DoD inspection 79.1 80. Phillips Hogan. report on NSA computers in CCR Series VI.C.1.27i Phillips interview.

Imanuscript.
interview, Oral interview with John W. Saacli, 19 Nov. 1987, by Robert Farley and Tom Johnson,

NSA OH 29·87; NSACSS Archives, ACC 43067, G 10·0306·1. 81. Phillips interview. 82. NSAlCSS Archives, ACC 24188, H02·0207-4. 83. NSAJCSS Archives, Enderlin, Support •... " 84. Dmanuscript; VI.BB.1.l4. "Teleeommunications Problems, 1968-1972; ... 1974, draft manuscript " "Evolution of SIGINT available in CCH;

"NSA's Telecommunications

Problems.

1952-1968

CommllDications

85. CCH Series VI.BB.1.l4. 86. Ibid; CCH Series XI.,Dcollection. 87. Briefing on computer- based data acquisition Dcollection. systems, July 1968, byl . --_

ICh. K31; in CCH Seriea
•..•

VI,

~-------~.------------------------Withheld from
89. NSAlCSSArchivea, ACe 37741,Gl4-0306·5; . H01·0511·7. CCH Series Xl ..ocollection.

public release 90. CCH Series XI ..ocollBCtion. L...-_P_u_b_o_L_o_8_6_-_3_6-----' 91. NSAJCSS Archives, ACC 10847.
93. NSAlCSSArchives, 94. NSAlCSSArchives, 95. NSAlCSSArchivea,

I E.O.

13526, section 1.4(c)

92. NSAlCSS Archives, ACC 31065, CBDE 22. ACC 434097, G14·0602·2; ACC 31065. CBDE 22. ACC 31065, CBDE 22. ACC 434097,G15·0605·2;ACC 31065,CBDE22.

HANDLE VIA TALENT KEYHOLE COMINT CONTROL SYSTEMS JOINTL Y NOT RELEASABLE TO FOREIGN NATIONALS

419

fOP SECRET l:JMBfbIc

DOClD: 523682 I
TGP SECRET UMIRA E.O. 13526, section 1.4(c)
96. NSAlCSSArehives,

REF lD:A523682 .I
22.

ACC 31065,CBOE

Withheld from public release Pub. L. 86-36

97. NSAlCSS Archives, ACe 31065, eBOE 22 •

.98. Ibid. 99.1 OH4-93. 100. Ibid. 101.

I; and oral history interview with!

I,

an41=======;-'~ by Charles Baker, Tom Johnaon'r,I==========:::::;:12;;5~F;:e-;-b-:l::99;:;3:-.::N=:!SA

D. et at, interview; working papelll ofl•... __

----'t. in CCH collection.

~----------~-~~-----------------~
103. David Bosk, •A History of U.S. Communicationa Security," (The David G. Boak Lectures). 1973 (rev). 104. Beak;

I
-'

Ifile in CCH collection.

105. Boak; •... 1 __

106.1 I,oral interview with Howard E•.Roaenblwn, August 1991, by Robert Farley and Charlllll Baker. NSA OH 3-91. 107. BoskdL..--_---1. 108. I I papers, oral history interview,l Johnson. NSAOH 2·93. ------109. Boaklecture.

'.2

Feb 1993. Charles Baker

III1d

Tom

Withheld from public release Pub. L. 86-36 .
1=9P SECRET UMBRA

420

/

DOCID: 523682

REF ID :A523682
1=QPSECRET "MIRA

I E.O.

13526, section 1.4(c)

I
61;ACC 9734X,CBDB42.

123. NSAlCSSArchiV'es,ACC

28650,CBTK

124. NSA retired records, 43981, 74-295. 126. NSAJCSs Archives, 126. NSAJCSS Archives, ACC 28666, CBTK 52. ACC 28652. CBTK 51; USAFSS. -History 1965; ACC 28656, CBTK 52.

Withheld from public release Pub. L. 86-36
of the 6937th Communications Group

(USAFSS),l

July - 31 December

127. NSAlCSS Archives, ACC 28664, CBTK 54.

128. NSAlCSS Archives, ACC 28650, CBTK 51; Tordella interview.
i29. LBJ Library, 130.
L..I __

NSF,inCCHSerieaXVl.

----'I
Project: A SIGINT SUCC888 Story
,w

131. Corley WODua, -r~eTacksman 132.

Stud~8 in.lnlelligen.c4!, Fall 1991, 21-31.

NSAlCSS ArchiV'es, ACC 28650, CBTK 51; ACC 29842, CBOE 28. History of the USAFSS Airborne SIGINT Reconnaissance Program (ASRP), 1950-1977," in

133. USAFSS,"A CCH Series X,J. 134. Ibid. 135. Ibid. 136.

CCH Series VI.I.1.8;

VI.C.l.271:

NSAICSS Archives,

ACC 43981, 74-295: INSCOM, "lNseOM

and its

Heritage."

1985. avail at HqalNSCOM,

Ft. Belvoir, Va.

137. Wood Study in CCH Series vt.I.1.8. 138. NSAlCSS Arcbives, ACe 43852, 73·252. 139. George F. Howe, "A HiatAJry of U.s. Civilians in Field COMINTOperatioDa, Spectrum, Summer 1973.5-8. . 140. 1953-1970,"

part II. Cryptologic

I
141.1

I

I"The

Civiliallizatioll

of Harrogate;

Cryptowgic Spectrum, Summer available

1970, 8-16; Tordella interview. Archives, ACC

I ABA

FY 1967 Command

History,

at INSCOM; Ft. Belvoir; NsAJcSS

22885, HO-0504-6. 142. Howe, Teciaf'licalRe.eGrciaSiaip.:

143. Ibid; oral history interview with Eugene Sheck, 16 Dee. 1982. by Robert Farley and Henry Millington, NSA
OH26·82. 144. Howe. 146. Ibid. 146. Howe,Sheck. 147·1L...148. H.D. Wagoner, Space Surveillan.c4! SIG1NT Program, (Ft. Meade: NSA, 1980). 149. Wagoner; NSAlCSSArchives, ACC 37741,GI4-0306.5c::=J
...J

U.S. CryptAJlogic History,

Special Series, Number

3

I E.O. 13526, section 1.4(c)

Withheld from public release Pub. L. 86-36
'--------,----'

~

.

~

HANDLE vIA rAiJZWf!E¥UQJ ECONINTe6NIIWLSYSTEMSJOlNTU -e fffi r1tE'1:EASABLE TO FOREIGN NATiONALS

421

TQP SECRET I:JMBItA

DOClD: 523682
lOP SEeRET UMBRA

REF lD:A523682
Withheld from public release Pub. L. 86-36

I E.O.

13526, section 1.4(c)

150. Wagoner.

151. Wagoner; Oral history interview with David Williams, by Robert Farley, NSA OH 23·87. 152. NSAlCSS Arehives, ACC 25766, CBOL 17; ACe 4088, G12·0502·1. 153. NSA/CSS Archives, ACC 25766, CBOL 17. 154. Ibid. 155. Interview with! 156. N.a"LI Archives,ACC 4088, G12·0508·2. 157. NSA/CSSArchives,4088, G12'()508·2. INSAOR 54·94,13 December 1994, by Charles Baker and Tom Johnson. ---11[1990,1 in CCH collection; NSACSS

c=J TIu!SIGlNT
159. Ibid. 160. Ibid. 161.1 162. Brugioni. 163.1L... __ 164. Ibid. 165. Ibid. 166.1 167.1 168. Bradburn.
--I

158. .=1

=---

----,1Raymond B. Potts andc=J
SaleUite Story. Washington, D.C.: NRD,l995

1Burrows, Deep Black.

IBurrows, Deep Black.
11967SORS memo, subject: "Genes~," held in E322.

169. Eisenhower Library, Burrows. 17o'L...1 __ 171. Ibid. 172. Ibid. 173. Ibid. 174. Ibid. 175.1L...176'L..1 __ 177. Ibid. 178. Ibid. 179. Ibid.
--I --I

--'IStudy.

Withheld from public release Pub. L. 86-36
HANDLE VIA'
.1\.

I E.O.

13526, section 1.4(c)

YSTEMSJOINTLY

-

NOT RELEASABLE TO FOREIGN

TOP SeeKEr l:JMBR.A. .

422

DOClD: 523682
I E.O. 13526, section 1.4(c) I
180. 181. 182. Ibid. 183. Ibid. 184.. Ibid. 185. Ibid. 186. Ibid. 187. Ibid. 188. Ibid. 189. Ibid. 190. Ibid. 191. Ibid. 192. Ibid. 193. Ibid. 194.. Ibid. 195. Ibid. 196. Ibid. 197. Carter interview.

REF lD:A523682
TOP SECRET UMBRA

D

EisenhowerLibrarypapers.

Withheld from public release Pub. L. 86-36

Cl
199.c:::=J 200. Reese,GeNlralReinh.ardGehkn. ... 201. Ibid. 202. Ibid. 203·1 mte~~·e-w-.-----------------------------------------------------204~N52files,IL...-_---' 206. N62Dmes;PhilliPainte~ew. 206. historyBakerand interviewwith 1992,Oral by Charles TomJohnsol1,NSA OH7-92.

I Phillips

1========================================1
I E.O.

21 Dec.

Withheld from public release Pub. L. 86-36

13526, section 1.4(c)

I

423

16,. SECRETYMBAA

DOClD: 523682
TOP SECRET tJ MIRA
I E.O. 13526, section 1.4(c)
207. Ibid.

REF lD:A523682

208·1__ ---,-_-;::::::====:=::::;;-- __ ---:---' 209. Oral interview with 1 ~ 23 Dee. 1992, by Charles Baker and Tom Johnson, NSA OH 8.92;

L~.

--------~

~----~------------------------------------------------~
211. CCH Series Vl:..1.1.5 •. 212. CCHSeries Vl.J.l.5; ACC434097.GI5-0602·2.

Withheld from public release Pub. L. 86-36

"F9P SEC9ET I::JMBR-A

!

424

DOClD: 523682

REF lD:A523682
T-QPS5(Anl.lMlRA

Chapter 10 SIGINT in Crisis, 1967-1969
After the relatively placid decade of the 1950s, the 1960s produced a series of international paroxysms unmatched in post-World War II history. Although cryptology was involved in virtually all the events, four crises in late decade had particular impact on the cryptologic business. The Arab-Israeli War of 1967 was a defining moment in . cryptologic contributions to the intelligence picture. The Soviet invasion of Czechoslovakia in August 1968, and the accompanying crisis concerning Romania, helped shape SIGINT production and reporting in later years. The other two events, the capture of the Pueblo in 1968 and the shootdown of the naval EC-121 in 1969, were uniquely cryptologic in their origins and implications, and they changed the way NSA and the cryptologic community have done business from that day to this.

SIGINT AND THE SECOND ARAB·ISRAELI

WAR

The Suez Crisis of 1956 and the Lebanon Crisis of 1958 had turned NS~'s attention to the Middle East and had begun the buildup of American cryptologic capabilities in the region. This involvement was to grow steadily as NSA sought to keep track of the . situation and the intentions of the Arab governments . . On the Arab side, the late 19505 marked the height of pan-Arab sentiment. In 1958 Egypt's Nasser had convinced Syria to join Egypt in forming the United Arab Republic (UAR). But the idea never worked. Syrians chafed under heavy-handed Egyptian bureaucratic regimentation. In 1961 Nasser, believing that state socialism was the only true path. nationalized virtually aU manufacturing, banking, and utilities. He also reduced to 100 acres the amount of land that a farmer could own, and he put a ceiling on the amount of money that a citizen could earn. This was too much for the Syrians, and two months later a military coup in Damascus ended the Syrian involvement in the union. Nasser, hoping that another Arab state would take Syria's place, obstinately kept the name (UAR), bllt none did.' Three years later a new transnational organization emerged. The Palestine Liberation Organization (PLO) was formally established at a conference in Jerusalem in 1964 with Ahmed Shukeiri as its head. It formed a conventional army composed of Palestinians and their Arab sympathizers throughout the Middle East. The rea! power, however,developed around a guerrilla movement called al-Fatah, headed by Yasir Arafat.2 A low-intensity Fatah-Israeli conflict developed almost immediately. It was punctuated by cross-border raids and terrorist bombings, and each incident led to reprisals

HANDLE VIA ~~ENT

KB"IH9' E GOMlU'f CO~L

SYSTEMSJOINTL Y

·425

TOP SElkE I UMBRA

DOCID: 523682
Tap SECRET loIMBRA

REF ID:A523682

which created the foundation for the next incident. At the same time, the ambitious Nasser was becoming enmeshed in a civil war in Yemen in which the other proxy was Saudi Arabia. This created strains in the Arab world and accentuated the division between the so-called Nasserists and the more conservative Arab govern.ments like Saudi Arabia and the Arabian desert sheikdoms. By early ~967 the Middle East was'clearly about to boil over. Terrorism was at a high level, and Nasser seemed spoiling for a fight. Then on 14 May NSA detected UAR air defense forces going on full alert. Three days later, on 17 May, Nasser demanded the withdrawal of UN forces from Gaza, and UN troops immediately began evacuating what was obviously to become a war zone. On 23 May Nasser took the warlike step of blockading the Straits of TiraJ;l',and he announced that Israeli commercial shipping, . whether in Israeli or forei bottoms, would be stopped.s Withheld from E.O. 13526, section 1.4(c)(d) The Cryptologic Posture

public release Pub. L. 86-36

By 1967 the American cryptologic posture in the Middle East had improved dramatically. From a single station on Cyprus only recently taken over from CIA in 1956, the cryptologic community had collection sites Crete,c=J and Cyprus, as well as collection from Asmara, Vint Hill, and Cheltenham. Navy and Air Force airborne collection platforms flew regularly in the eastern Mediterranean, \

in,1

I

~--------------~/

Withheld from public release Pub. L. 86-36

E.O. 13526, section 1.4(c)(d)

TOP SECRET tJM1S1tA

426

IOPSfCBfT11Y'PA

.

Withheld from public release Pub. L. 86-36

Tb,MiddleEutiDlH1

N <10 \0 C"1 N III

~QLE~Y

N

au
427 ;oPUiCRi:r"u'RA

o
o

H

8

DOCID: 523682

REF ID:A523682

I

'fGp SECRE' YMIR A. E.O. 13526, section 1.4(c)(d)

Withheld from public release Pub. L. 86-36

Prior to Nasser's eviction of UN forces from Sinai, there was no consensus in the U.S. on the likelihood of war. A National Intelligence Estimate published in April assessed that there was no near-term likelihood of war in the region. In May, StatelINR assessed Egyptian military activities as defensive. Thomas Hu hes the to State De rtment intelli ence anal st, based much of his estimate

~-:-:--:-----------------.,-----_--.J

Walter Rostow, President Johnson's national security advisor, was hopeful that things could still be resolved by negotiation, and he noted that the Soviet Union did not seem to want to get directly involved.'

However. the cryptologic community had begun a series of SIGINT alerts as early as November.1966./ I I /NSA expanded the alert to include the entire Middle East. This was quickly elevated to a SIGINT Readiness Bravo when Nasser closed the Straits aCTiran on 23 May. A Bravo was as high as the SIGINT readiness system could proceed short of war.e By the accounts of all involved, it was . no longer a question of if. but tohen:'

Withheld from public release Pub. L. 86-36
HANDLE VIA

I E.O. 13526, section

1.4(c)(d)

1::: KE¥HOJ$

mwm e6N:~~:STEMSJOINTLY

TOP SeCRET UMBRA

428

DOCID: 523682
I E.O. 13526, section 1.4(c) I

REF ID:A523682
Withheld from public release Pub. L. 86-36
TOP SECRET UMBRA

To further bolster collection in the eastern Mediterranean, NSA decided on 23 May ,<the day Nasser blockaded the Gulf of Aqaba) to deploy a TRS. I I I I and realizing that even combined Air Force and Navy airborne collection could not produce round-the-clock coverage, NSA diverted the USS Li~rty to an eastern Mediterranean cruise. The Liberty was selected because of its superior cruising speed (18 knots, best of all the TRSs), its multichannel collection suite, and its availability. (It had just begun cruise and was fitted out for an extended voyage.)

a

Meanwhile,

SIGINT

indicators of impending war poured in. \

--=-:--::-- _ _:_~------:___:_---J The intelligence community had other sources ofinformation, but none was as timely or authoritative during an expanding crisis such as existed in May 00.967.10 In many ways the war preparations of 1967 resembled ,Japanese war preparations in 1941,/
L- __

I

The entire Middle East was on the brink when, at 0745 Middle Eastern time on 5 June, Israel launched a preemptive strike on Egyptian air forces. In what 'became one of the classic offensive attacks in the annals of warfare, the Israelis destroyed virtually the entire UAR air force on the ground. Within a few hours, 309 out of 340 combat aircraft were in smoking ruins, including all 30 of its long-range TU-16 bombers. Unaware of'how bad things were, Syria and Jordan jumped into the fray by launching attacks on Israel. But they were too late. No longer having to worry about the Egyptian air force, the Jewish state turned its attention to Syrian and Jordanian forces on its borders and to the Egyptia:n divisions massed in the Sinai. Having no protection in a desert environment, the ground forces were exposed and largely destroyed in three days. In all, 417 Arab aircraft were destroyed, 393 on the ground; only 26 Israeli aircraft were lost. 11 The White House first learned of the war (rom press sources. When the Situation Room called NSA for confirmation, they heard nothing for a time, but by mid-morning SIGINT reports were beginning to flood the wires. The Arabs' and Israelis were making charges and countercharges, and the president wanted to know who fired the first shots.
'--

reports were suffiCient for presidential advisor Clark Clifford to make an initial determination that the Israelis attacked first. This judgment was to be confirmed many times over when all the evidence was sifted through.
---I

Withheld from public release Pub. L. 86-36

I E.O. 13526, section

1.4(c)

E LSYSTEMSJOINTLY ASABLE TO FOREIGN NAT

429

TOP SECRET I:IMBRA

DOClD: 523682
lOP SECRET UMBRA

REF lD:A523682

War in the desert. Shattered Egyptian tanks smolder in the Sinai desert.

Amid the conflagration in the desert, .the Johnson administration Soviet Union. What would the Soviets do?

ke

es on the

I E.O.

13526, section 1.4(c)

Withheld from public release Pub. L. 86-36

:NDLE

yr~ :~;:
F

KFYHQI,E coum'%' eON I ;'~L SYSTEMSJOINTL Y LEASABLETO FOREIGN N'IONALS 430

'FOP SE<:R&T

U"IIBRA

----

-- --

--

_.

-

DOClD: 523682 I E.O.
13526, section 1.4(c)

REF lD:A523682
Withheld from public release Pub. L. 86-36
TeP SEfRET UMBRA

To White House analysts, it appeared that the Soviets were willing to fully support Arab governments with equipment but were not willing to send troops. The Arab governments misread the Soviet attitude ,I INasser jumped into war without understanding that he would have to go it alone. Once the war began, the Egyptians and Syrians expected intervention -what they got was an emergency shipment of equipment to replace that which the Israelis had destroyed. The arms deliveries began almost immediately I I

On 6 June, the Egyptians and Syrians claimed that U.S. and British forces had provided air cover for the attacking Israelis. This sensational charge, repeated and believed throughout the Arab world, was apparently intended to provoke Soviet intervention, an event that could have produced a dangerous American-Soviet confrontation. But Kosygin rejected the claim outright. I

L-.

--II Nasser was furious, but he did not succeed in egging the USSR closer to involvement. That same day, Kosygin contacted Washington on the hotline and pledged to work toward peace. As the succeeding days unfolded and Israel pressed toward the Suez Canal, Kosygin's talks with the Johnson administration over the hotline became more testy, but direct negotiations played a key role in American and Soviet abilities to avoid military involvemen t. 1e
Fighting finally terminated on the tenth.

I

Withheld from public release Pub. L. 86-36

I E.O. 13526, section 1.4(c) I
1-

.", .v\. 1••••OL SYSTEMS JOINTLY HANDLE VIA TA1.f;N"1. NuT RELEASABLE TO FOREIGN N " 431

-

-

rop SECRETUMBRA

DOCID:523682
lOP SECRet UMBRA I E.O. 13526, section 1.4(c) I

. REF ID'A523682
Withheld from public release Pub. L. 86-36

the Middle East War,

.

I The U:S. and the USSRhad narrowly
1-1 -----'

avoided involvement in

Missing from the postmortems were the usual breast-beatings about why intelligence failed to warn. In 1967 it did not fail. '-;-_---1 American intelligence generally downplayed the possibility of an Arab attack; the best possibility, and the one which actually played out, was an Israeli preemptive strike like the dash to the Suez in 1956.2~ The 1967 war was the closest that the United States and the Soviet Union came to war between the Cuban Missile Crisis and the end of the Cold War. I

Withheld from public release Pub. L. 86-36

/ 13526, section 1.4(c) I The Liberty, NSA's choice as the TRS deployment to the Middle East, was a reconditioned World War II Victory ship, converted to an AGTR in 1964. The vessel already had five cruises under its belt. It had 20 intercept positions, 6 officers, a SIGINT crew of 125 and an overall complement of 172 men. With TRSSCOM, ship-to-shore radiotelephone circuits, and two receive terminals for fleet broadcasts, the Liberty was one of the best equipped ships in the TRS inventory. The Navy approved NSA's request, and the Liberty, off the west coast of Mrica,steamed for Rota, where it took aboard an additional 9 linguists, including 3 NSA civilians, and more keying material for its communications circuits. On the second of June, it set offfor the eastern Mediterranean. 21
\

The Attack on the Liberty

I E.O.

The Liberty's sailing order specified that it was to stay at least 12.5 miles off the coast of the UAR and 6.5 miles from Israel. When war broke out on 5 June, the Sixth Fleet, to which the Liberty had been temporarily attached, was directed to remain at least 100 miles

'fOp SECRET l:otMBRA

432

DOClD: 523682

REF lD:A523682
TOP S!(RET I:IMBRA

off the coasts of Lebanon, Syria, Israel, and the UAR, but the Liberty's instructions were not changed. When it arrived in its operating area late on 7 June, Captain McGonagle, the vessel's commander, still had written instructions that brought the Lil1erty close into the coast.22 Nasser's charge on 6 June that the U.S. and Britain were providing air cover for the .Israelis, and the possibility that the Soviets might intervene, brought new orders to the Sixth Fleet to stand off at least 200 miles from the eastern Mediterranean littoral. The next day the JCS decided to pull the Liberty, the only U.S. naval vessel still in the far eastern Mediterranean, back to at least 20 nautical miles from the UAR and 15 from Israel. Later that day JCS changed again, this time to 100 nautical miles from both countries.23 The first JCS message never reached the Liberty - an Army communications center misrouted it to a naval communications station in the Pacific. When, an hour later, the. Joint Reconnaissance Center of tJieJCS decided to pull the Liberty back to 100 nautical miles, a series of communications fiascos occurred which stretched on into the night.Message misroutings, delays occasioned by the press of other business, refusals by the Navy to transmit based on a verbal order, all combined to delay the message receipt until after the attack. It was a repeat of the warning message to Pearl Harbor on 7 December 1941, and there was blame aplenty." The Liberty was reconnoitered by several unidentified aircraft during the morning hours of 8 June. That afternoon it was about twenty-five nautical miles north of the Egyptian city of Al Arish when, at about 1400 local, two French-built Israeli Dassault fighters veered toward the ship and began strafing it with cannon and rockets. The attack put some 821 rounds into the hull and superstructure, wounded McGonagle, and killed 8 crewmembers. The Liberty managed to get off a desperate message to Sixth Fleet before the power to the radio equipment went out, and Admiral Martin, the Sixth Fleet commander, launched 4 armed A-4 Skyhawks for air cover. Since his flagship was 450 nautical miles away from'the Liberty, however, the aircraft did not arrive before 3 Israeli torpedo boats launched 2 torpedoes at about 1430. The torpedoes tore through the SIGINT spaces, killing 25 men and putting a hole in the hull 39 feet across. As the crew of the Liberty scrambled to keep the vessel afloat, one more crewmember was killed by machinegun fire from 1 of the torpedo boats." Once the torpedo boats departed, McGonagle directed his vessel to Malta. Sixth Fleet escorts reached the Liberty sixteen hours after the attack and trailed the vessel, picking up . classified and cryptographic keying material escaping from the hole in the hull. The Liberty limped into Malta on 14 June after a heroic struggle to stay afloat that eventually earned McGonagle the Medal of Honor. In all, thirty-four crewmembers were killed, including one NSA civilian Arabic linguist, Allen Blue. The men lost their lives.in a war

)

.

W

ABLE TO FOREIGN NA u ~

-

433

1=9P SECRET l:IMBRA

OOClO: 523682
TOP SECRE r UMBRA

REF lO:A523682

The Liberty at Malta alter the ~ttack

RELEASABLE TO FO

0 AI:oS

lOP SECRET UMBRA

434

DOClD: 523682

REF ID:A523682
TOPSECREi OMBRA-

The Liberty SIGINT compartment

Another view

HANDLE VIA TA ASABLE TO FOREI

SYSTEMS JOINTLY

435

TOP SECR~f UMBRA

---------'

-------

--------

---

DOCID: 523682
'fOP SEEREl' t:JMBItA

REF ID:A523682

in which the was not a combatant because of errors in a military communications system that, by 1967, could no longer do thejob. At NSA, word of the attack reached Director Marshall Carter at 0915 Washington time. The telephone began ringing almost at once, as word of the attack spread through Washington. While Carter was directing intercept coverage reallocation, Secretary of Defense McNamara called him (at 1015) to ask for details on the vessel and the voyage so that he could make a statement to the press. Deputy Director Louis Tordella took charge of devising a cover story. Carter diverted many of the queries to NSG. At one point during the day the director got a call from the Joint Reconnaissance Center suggesting that the vessel be sunk. Carter replied that this was the worst thing they could do - heaps of classified documents and equipment would end up in shallow water. He was right, and McGonagle's heroic piloting of his vessel to moorage in Malta saved what could have become a much worse situation.28 Lyndon Johnson got word at 0949. At the time the U.S. still did not know the identity of the attackers, but the White House soon found out through a Defense Attache Office message from Tel Aviv that the Israeli navy had admitted the error. This presented the president with a very touchy dilemma. Because of Arab charges that the U.S. had assisted the Isra.'elis, the Sixth Fleet was standing far away from the conflict in the central Mediterranean. 'fet here, unannounced, was an American naval vessel only a few miles off the coast of'Israel, in the middle of a war zone. Johnson's first concern was about Soviet reaction. He had Walt Rostow send a message to Kosygin stating that the Israelis had apparently fired on a U.S. ship in error and that the Sixth Fleet was sending ships and planes to investigate (he repeated it twice). Kosygin replied that he had passed the message to Nasser. 27 . Meantime, the Pentagon had released a statement about the attack, indicating that the Liberty's mission was to "assure communications between U.S. Government posts in the Middle East and to assist in relaying information concerning the evacuation of American dependents and other American citizens from countries in the Middle East."28 This was the cover story that NSA had devised under hurried circumstances. It didn't work, but like the U-2 incident in 1960, no cover story would have worked in the situation. The press very quickly sniffed out the truth, which was attributed to an anonymous military officer that the Liberty was a "spy ship." According to this source, "Russia does the same thing. We moved in close to monitor the communications of both Egypt and Israel. We have to. We must be informed of what's going on in a matter ofminutes."29 The . assertion was denied by official sources, but the true mission of the Liberty was never in doubt again. (The vessel did not, in fact, have an Israeli mission, because linguists were too scarce.) How did the the incident happen? Was it a deliberate attack by Israel, as has been alleged countless times by many people? (Even General Carter believed it to have been deliberate.) If it was an accident, how could the Israelis have possibly misidentified the

u.s.

I

lap SECRET~MBR:A

436

DOCID: 523682

REF ID:A523682
lOP SECRET l:JMBItA

ship? The Liberty was flying an American flag, was clearly marked on the hull •• AGTR-5," and when the first flag was shot down by the attacking Ilighters, McGonagle hoisted the largest flag he had aboard, a holiday ensign seven by thirteen feet. This enormous flag was flying above the Liberty when the torpedo boats executed their attack." I-E-.-O-.-1-3-S-2-6-, s-e-c-ti-o-n-l-.4-(-d-)SIGINT

The idea that the attack was deliberate turned out to be wrong. Although there was no ,--, bearing directly on the attack, there was a I I report shortly after the Withheld from incident dealing with the aftermath, It reported air/ground conversations between a public release ground controller at Hatsor and two Israelihelicopters which reconnoitered the Liberty as Pub. L. 86-36 it was turning toward Malta. Hatsor first identified the vessel as Egyptian, "but later'------;-------' became unsure, and requested that the helicopter crews "verify the first man that you [bring up] as to what nationality he is.". A few minutes later Hatsor instructed: "Pay attention: if they speak [B-val Arabic] and are Egyptians take them to Al Arish. If they speak English and are not Egyptians, take them to Lydda ... the first thing is for you to clarify what nationality they are." Two minutes later Hatsor asked, "Did it clearly signal an American flag?" And a minute later, "Requesting that you make another pass and check again whether it is really an American flag." One can imagine the panic at Israeli naval headquarters at the time. apparently attacked a vessel of their closest ally. They had

Based on this report, Rostow told Johnson that the Israelis appeared to be confused about the nationality of the vessel, and he suggested that there might have been some breakdown within the Israeli military which resulted in the attack." The official Israeli court of inquiry concluded on 21 July that it had in fact been an identification error. When the Liberty was first discovered by an Israeli spotter plane on the morning of the eighth, it was unidentified but possibly hostile, and a red marker was placed on the map in the naval war room. Later in the morning, the identification was tentatively changed to friendly (American), and a green marker replaced the red one. But the Israeli navy then went a period of time without a location, and someone, instead of retaining the green marker with a question mark, pulled it otTthe map entirely.52 The shift changed at 1100 Israeli time, and the new shift knew nothing about the American vessel, which was no longer designated on the map. What they did know was that Israeli army units in the Sinai coastal town of Al Arish were reporting artillery bombardment from an unknown source. (It later turned out to be the explosion of an ammunition dump.) The Israelis began searching the sea for a possible hostile ship, and they found the Liberty. The crew of the vessel that did the identification claimed that its radar showed the ship to be heading at twenty-eight knots toward Suez (an impossible speed for the Liberty - an error by the radar operator), and Israeli naval control ordered an air attack. Two Mirage fighters on their way home from an air patrol over the Suez Canal were diverted to the spot where the supposed hostile was. After a quick pass, the pilots

"~LErp..u..eONI~
" ASABLE TO FOREIGN NA

437

'FOP SECRET l:JMBR:A

DOCID:

523682
TO' SECRET UMBRA

REF ID:A523682

claimed that the ship was not displaying a flag (another error) and were ordered to execute ail attack. The torpedo boats arrived in the area at 1418. A low-flying aircraft had just radioed to its controller that he had seen a marking ','CPR-S" on the hull. The naval controller told the torpedo boats to attempt a better identification, but the captain of one of the boats .claimed that when he requested identification, the ship requested him to identify himself first. Based on identifieation aids available on board, it appeared to him to be the Egyptian supply vessel El-Kasir, and with this information' in hand Israeli naval control again ordered an attack. After the first torpedo hit the boat, the markings "CTR-S" were observed on the hull. Control immediately terminated the attack, just before the torpedo boats were about to launch additional torpedoes that would have sunk the Liberty. An Israeli helicopter flying over the ship after the attack. finally noticed an American flag, and the Israeli navy realized what it had done.S3 An Israeli court of inquiry, whose findings were kept secret at the time (but which were uncovered and published by two Israeli journalists in 1984), condemned the confusion, incompetence, and interservice rivalry that contributed to the attack. There was no finding of a deliberate attack,' but there was plenty of blame for iln the Israelis associated with the incident. The Johnson administration was properly outraged. The State Department, in a scathing statement highly unusual for diplomats, called the attack "quite literally incomprehensible. As a minimum, the attack must be condemned as an act of military recklessness reflecting wanton disregard for human life." But Clark Clifford, who was appointed by the president to render a final judgment, called it an identification error. Clifford relied heavily on COMINT reports showing Israeli confusion about the identification; these would have been difficult to fake. Going into it-with a preconceived notion that the Israelis must have known, he concluded that what was involved was "a flagrant act of gross negligence ... " rather than a deliberate act.34 This did not, of course, quiet the press, Journalists, both reputable and disreputable, supported the "deliberate attack" theory, and the legend arose, without basis in fact, that the Israelis wanted to blind American SIGINT sensors to their communications, both to keep them from fmding out that Israel actually started the war and to keep secret a plan to launch an attack on Syria. (As was stated already, the vessel was not targeting Israeli communications and had no Hebrew linguists on board.) All these charges were repeated and embellished by James M. Ennes, a lieutenant aboard the Liberty' who published a book on the subject in 1980. Most of the crew still believes that the attack was deliberate." Many of the journalists properly questioned the position of the vessel at the time. Clifford, too, made a special point ofthis. The Liberty was clearly not where it should have been. The original plan was formulated before war broke out. Once the eastern Mediterranean became a battleground, it was decided to hold the Liberty out of the area,
J

HANDLEVlA TALENTKE

SYSTEMSJOINfLY ABLE TO FOREIGN NATlON

TOP SECRET

"MBRA

438

DOCID: 523682

REF ID:A523682
lOP SElkE I UMBRA

but the messages never reached McGonagle. The U.S. communications system was approaching breakdown; war sufficed to push it over the edge. The crew, on the other hand, performed magnificently, and they and their vessel deserved better. NSA wanted to refurbish the ship and use it again, but the price tag of over $10 million was too high. The Liberty was decommissioned a year after the attack, and in 1973 it was cut up for scrap in Baltimore's Curtis Bay Shipyard." An abashed Israeli government paid $13 million in compensation for the loss oflife and damage to the vessel. The attack on the Liberty should not be viewed as a bizarre, or even an especially unusual, identification error. Even in peacetime such errors are' made all too frequentlythe Soviet shootdown of KAL 007 and the American shootdown of an Iranian airliner are good examples. When a country is at war, the possibility of error is compounded by haste and fear. Losses to friendly fire always represent a substantial percentage of the casualties. And the Israeli agreement to compensate should not be taken as proof of guilty knowledge, but rather as an attempt to retain the friendship of a benefactor wronged.

THE PUEBLO
Any
WilY

you look at it this incident is a loser. We cannot come out even. We must cut our losses. Clark Clifford. 29 January 1968

Nineteen sixty-eight was a bad year for the United States. It started with the Tet offensive in Vietnam and saw the assassinations of Robert Kennedy and Martin Luther King and the Soviet invasion of Czechoslovakia. As disaster piled on disaster, the only people truly happy were the media. The very first disaster, however,was, for American cryptology, the worst. On 23 January North Korea captured a small srOINT trawler from the TRS program called the Pueblo. It was everyone's worst nightmare, surpassing in damage anything that had ever happened to the cryptologic community.
Set-up

After a long lull following the Korean armistice, Nor~h Korea had become more aggressive. A clarion can of sorts sounded from the convention of the Korean Worker's Party in Pyongyang in October 1966, at which Kim Il-sung announced a campaign of hostile acts aimed at the "liberation" of South Korea and unification of South and North. This was followed by a dramatic rise in North Korean infiltration, terrorist incidents, and . firefights along the demilitarized zone (DMZ). Between 1966 and 1967 incidents increased tenfold. On 21 January 1968 a group of thirty -one North Korean Infiltrators attacked the South Korean presidential palace in hopes of assassinating President Park Chung-hee.

ID~D~

N

TOFOru:.L •.•. !'<L"~

439

TOPSECReT tJ MBftt(

DOCID: 523682
lOP SECRET UMBRA

REF ID:A523682

This infamous Blue House incident raised tensions along the DMZ to their highest point since the armistice." Into this not very auspicious situation intruded the latest in a series ofTRS vessels. The Pueblo was first constructed in 1944 as an Army freight and supply vessel, and it was used to haul materials to South Pacific islands during the latter days of World War II. Decommissioned in 1954, it had sat in mothballs at Clatskanie, Oregon. In 1966 the Pueblo rejoined the Navy, this time as a TRS. It was recommissioned at the Puget Sound Naval Shipyard in Bremerton, Washington, and became the smallest version of the SIGINT ship, an AGER. The Pueblo carried just six positions and could make twelve to thirteen knots at top speed. Its new captain, Lieutenant Commander Lloyd M. Bucher, reported to take command in January 1967, while it was still undergoing refi tting. sa The captain and his crew were mismatched from the start. Bucher resented being jerked out of submarines to the surface navy. He knew nothing of electronic espionage and apparently learned little in his courtesy stop at NSA. His autobiographical account of the visit revealed considerable distaste for the mission and the people involved in it. Once on board, he found it difficult to get along with his executive officer, Lieutenant Edward Murphy. Moreover, he resented the operational control that Lieutenant Stephen Harris, the NSG-provided chief of the cryptologic spaces, had. To Bucher, not being in ful] control of his ship was Intolerable."

,-..,
'-' "":

~

.....

= .~ .•...
~ ••• '" ~ ~ .....
N

e
~

<l.i '-C o '" ~ _'-C
~ I

"O<l.iQO ~ lo.

..c:.::! ~ ..c: - .c..c• .•..• .
~ Q.~

~- = = capability."

The cryptologic crew was ill prepared for duty. Harris had a good background, including Russian language training and assignment on several NSG afloit detachments . But only two enlisted members had ever been to sea. The two Marine linguists who put aboard at Kami Seya[(USN-391were very green at Korean, and during the capture they could not understand the North Korean voice transmissions discussing the impending fate of their vessel. NSG had placed a vessel in harm's way without an advisory warning

c
~

The way the AGER program was set up, NSA had little influence on the mission. The Navy tasked the vessels, and NSA provided-technical support and suggested secondary tasking. Risk assessment for the voyage flowed through Navy channels up to DIA, which rendered the final judgment. By 1968 there were literally hundreds of missions worldwide every month, and there is no evidence that anyone put much thought into the Pueblo's first mission. The Navy assessed the risk as minimal, and DIA rubber-stamped it. The mission raised a few eyebrows at the 303 Committee (the organization that reviewed the monthly reconnaissance schedule), but the risk was not changed and the mission profile was not modifled." Since the risk assessment process occurred over the year-end holidays, it probably received less scrutiny than was normal.

HA==-=:~Q_"~
ABLE TO FOREIGN .

lOP SeCRET'

'MBRA

440

DOCID:523682

REF ID:A523682
l'6' 5!CIt!TtlMBRA

Lloyd Bucher (emerging from a hearing, witb Stephen Harris, after repatriation

in 1969)

1~:T~::::Wr~~:;:~JOINTLY
441 TOP SECRET UMBRA

DOCID: 523682
lOP SECRETIJMBR4

REF ID:A523682

CONPlDEN:rtAL

I E.O.

13526, section 1.4(c)

ORICnrAt
3l l)~;;r;llE;~ 3_%1

,

"o/!p._" ftoom

!':aroRCS NSA 2~
Fot'Qc<;1stlt

oss ?t;:t;3LO

C,aER-2)

Wit4held from public release Pub. L. 86-36

The Pueblo, betore its voyage

laP SEERH UMBRA

442

DOClD: 523682

REF lD:A523682
TOP SECRET UMBRA

In fact, it should have raised some eyebrows. The North Koreans had of late shown unusual sensitivity to coastal vessels. Just twelve days before they took the Pueblo, the small North Korean navy had chased 300 ROK fishing boats south of the Northem Limit Line (NLL - a geographical extension of the 38th Parallel into the Sea of Japan), capturing two and capsizing a third. On the 20th North Korea summed up its grievances about coastal vessels to the UN Command, claiming that the other side was dispatching "spy boats disguised as fishing boats and villainous spies together with fleets of South Korean . fishing boats. ,••2 Even prior to this.: however, NSA had dispatched a message to the Joint Reconnaissance Center discussing the recent increased North Korean sensitivity in relation to the upcoming voyage of the Pueblo. JRC simply sent the message to CINCPAC, which paid no mind.43 On 16 January, after putting out from Sasebo six days earlier, the Pueblo arrived at the northernmost point of its mission area and began slowly working its way south toward the port city of Wonsan. It had firm instructions to stay at least thirteen nautical miles off the coast, and there is no evidence to suggest that this order was ever violated. The crew was not having a happy trip, though. The seas had been rough almost every day since they had departed from San Diego in November, and the mission, which consisted of some very basic SIGINT sampling, had been dull and unproductive in the extreme.44

Capture. On the 20th, and again on the 22d, the Pueblo saw North Korean vessels that were close enough to note its position. Bucher was sure that he had been identified and broke mandatory radio silence to report this. At about noon on the 23d, a subchaser pulled up, . and after requesting that the Pueblo identify itself, the subchaser reported back to his. controller. Clearly, the North Koreans were by then certain that it was a surveillance ship, of some kind, and after some minutes, during which time it was possible that Wonsan control radioed instructions, the subchaser requested the Pueblo to heave to. The Pueblo turned to flee, and the subchaser gave chase, joined by three torpedo boats. The Pueblo radio room sent news of the incident to Kami Seya at Flash precedence. The Pueblo and the pursuing torpedo boats continued to playa game of tag, and for a time Bucher was successful in evading capture. But finally the subchaser got between the Pueblo and open ocean and opened fire. Almost simultaneously the torpedo boats opened up, and at this point Bucher very tardily ordered emergency destruction to begin. (One of the NCOs in the cryptologic spaces had already disobeyed an earlier Bucher order and had begun destroying things.) Finally Warrant Officer Lacy overrode a Bucher order and directed the ship to stop dead. The chase was over."

liANDLE VIAT _~DW'Ht!~:AASJSABLE

ROLSYSTEMSJOINTLY TO FOREIGN NATION

443

TOP SE(RH

UMSRA

.• 1

_

DOClD: 523682
Tap SECRETUMBRA

REF lD:A523682

CHINA

,
I I

,
,

I

I

" " i::~~;::~H~
,~

I ,

,
'/

"

TerritorialUmil

(

i')+
\ \

PCYONGYAK~'
.
.

-. +.£1+
WO~SA;\'

,
\

Sea

,, __,J
I r
I

""-,

of
, , , ,

Japan

,
\

SEOCL

\

\

, ,
I
I I I

r---------------~
~
I

LEGEND

SOCTH KOREA

'. i:~0
,
f f I

USS Pueblo

., \+ ~ ..... --- .. --- .... ---:

North Korean 1'1J7 Subchaser . I

North Korean

Torpedo Boats

-t'f6Etf)isposition

ofSorch Korean,nayaJ uniL5and Plltolodurinr attack and seiz.ure.23January

1968.

Map of the capture

HANDLE VIA TA
ABLE TO FOREIGN NATIONALS

MSJOINTLY
.

TOP SECR!T tI M~RA

DOCID: 523682

REF ID:A523682
faP SEERET HMBRA

As the Pueblo limped slowly toward Wonsan, escorted by the North Korean vessels, the crew was below decks desperately trying to get rid of all the classified material. It was a futile effort. This ship had far more classified material than it should have had, and it was not equipped to destroy in an emergency even that which it was authorized. Lack of adequate equipment, confined spaces which prevented use of the most effective destruction techniques, and an inexperienced crew that had never practiced emergency destruction aboard the Pueblo combined to virtually nullify. their' efforts .. When the ship was finaIly boarded, most of the material was still lying on the deck.48 The boarding took place at 1445, almost three hours after the first North Korean vessel had been sighted. One crew member had been killed during a volley, and several, including Bucher, had been wounded. The radioman had succeeded in apprising Kami Seyaoftheir predicament, and he keptthe station upciateduntil he had to go off the air to . destroy crypto material.' The Pueblo reached Wonsan at about 1900, after the harbor lights were already winking in the stillness. 'I'he crew was omoaded and placed in a captivity that would last almost a year."

Aftermath In Kami Seya, things were 'anything but still. The unit had been on the line with the Pueblo for the better pa~t of three hours, and it was frantically passing reports to Commander, Naval Forces Japan. But the initial reports failed to generate the appropriate concern there. Not until after hearing the phrase "we are being boarded" did the organization get itself mobilized. Mobilization, however, proved difficult. 'The quickest remedy would have been a flight of 5th Air Force fighters. But owing to the low risk assessment, no fighters were on alert, and it would have taken two to three hours to ready something. Adding flight time from Okinawa (where the aircraft were based), they could not have reached Wonsan before dark. Fifth Air Force F-4s in Korea were on SlOP alert and could not be rearmed in time. The carrier Enterprise was steaming south in the Sea of Japan on its way to Subic Bay whenit got the distress call. But the Enterprise F-4s wer~ armed with air-to-air missiles, and the time required to rearm and fly to Wonsan was too much. The Enterprise turned around and steamed toward Korea to rendezvous with other vessels headed for the same place, but none of them would be there in time. No help was available, and the U.S. military had to sit and watch." The middle of the day in Japan was the middle' of the night in Washington. Critic reports began arriving at NSA and the White House at about midnight. The senior operations officer called in. Major General John Morrison, the assistant- director for production, who hurried in to look at the traffic. Morrison called General Carter, who began directing the NSA response."

HANDLE VIA TALENT KEYHOLE COMINT CONTROL SYS'rEMSJOINTL NOT RELEASABLE TO FOREIGN NATIONALS

Y

445

TOP SECRET UMBRA

.

DOClD: 523682
TOP 5EeRET UMBRA

REF lD:A523682

At the White House, Walt Rostow, the national security advisor, came in first. After hasty calls to NSA and Hawaii to get more information, Rostow notified the president early in the morning.

I

E.O. 13526, section 1.4(c)(d)

I

Carter mobilised every SIGINTresourcehe could get his hands on, and assembled every scrap of paper that pertained. He called an Alpha Alert I Withheld from public release L--:-_---J1 So within theL-P_u_b_._L_._8_6_-3_6-----' cryptologic community, everyone was scrambling. But to the rest of the world Carter put up a stone wall. It was a NaVy mission, and he directed that most of the questions be diverted to naval authorities and the Joint Chiefs of Staff. Rather than spread his cryptologic authority to encompass the Pueblo, Carter found it useful in this case to put the pressure on the Navy.so

I

Now that the damage had been done, Carter wanted to assess what tne damage was. Regarding COMINT, NSA's initial assessment was equivocal. Assuming that most COMINT documents had been destroyed before capture, NSA focused on' the information that the crew might reveal under interrogation. It was potentially serious, but as yet unknowable. Regarding the COMSEe loss, however, NSA's conclusions, expressed initially only a day after the loss, were unmistakable: "The probable compromise of four major U.S. COMSEC equipments, including three of our modern electronic crypto-equipments, is a major intelligence coup without parallel in modern history." This was right on target as far as was known then, but the full extent ofthe loss was not known until the mid-1980s, as will be discussed below. 51' At the White House, the Pueblo capture was one of those transcending crises that occupied the president. Before the end of the month, Lyndon Johnson had participated in at least thirteen full-dress meetings on the subject, and Robert McNamara, Clark Clifford (McNamara's designated replacement; 23 January was his first day on the job), Secretary of State Dean Rusk, and Earl Wheeler (chairman of the JCS) were all fully engaged until 30 January at which time the Tet Offensive cornered their attention. The first meeting was the Tuesday lunch on 23 January. Discussions focused on where the Pueblo was when captured and what the United States could do about it. Inasmuch as it was too late to take the ship back, the group ran through several warlike options such as capturing a North Korean ship, hitting the North Koreans with U.~. forces, and augmenting U.S. forces in the Korean area. At this meeting the president articulated a feeling that came to dominate his thoughts - that the Soviet Union might be behind this and that it could be a "second front" designed to distract the U.S. from South Vietnam. There was no evidence to support this, just speculation." Later that day Johnson phoned the Soviet Union on the hotline to complain about it. He demanded Soviet intercession with North Korea, to which the Soviets replied that it

HANDLE VIA T
OT RELEASABLE

ROL SYSTEMS JOrNTLY
TO FOREIGN NATIONAL

TOP SECRET UMBRA

446

._-

..

_------------------------

POCID: 523682

REF ID:A523682
'FeP SECRET lIMBRA
;

was not their problem. Proof of Soviet involvement was lacking then and is still lacking today.5S Twenty-four January was the day which shaped the administration's response. In a series of marathon meetings which had come to define the White House in crisis, the "kitchen cabinet" 1. dealt with the problem of the ship's position. Not all the SIGlNT evidence was in yet, but there was enough to show that the North Koreans themselves knew the Pueblo was outside their territorial limits. This was confirmed through both intercepted voice and radar tracking which located the Pueblo approximately fUteen nautical miles offshore. The president decided to go on the air to reveal this information and to bring the evidence to the United Nations; .
)

2. determined, without evidence, that the capture was somehow related to Vietnam. All in attendance agreed that the Soviets must have known about it in advance. (Later that day CIA registered the only dissent.); '3. tentatively decided to move additional military aircraft into Korea, as well as station the Enterprise task force off the coast; decided to activate selected military reserve uni ts for the crisis. 54 That same day FBIS intercepted a Korean Central News Agency broadcast purporting to contain a "confession" by Bucher alleging, among other things, th~t the Pueblo had made a "criminal intrusion" into North Korean territorial waters. That very afternoon the Pentagon issued a rebuttal, stating that "the Pueblo's position as determined by the radar track of the North Koreans themselves ... " put the ship outside North Korean waters. NSA was not consulted on this release ofSIGINT.Carter was livid, but he was powerless to alter the administration's determination to publish SIGINT refutations of North Korean charges. 55 Simultaneously, the administration was working on a presentation to the UN, to be made by Ambassador Arthur Goldberg. As nothing appeared sufficient to head off this even more explicit release of SIGINT,Carter sent a team to New York to work with Goldberg and his staff on the statement. By cooperating closely, NSA had an opportunity to read Goldberg's statement before he went before the Security Council on the 26th. Goldberg presented both North Korean voice and manual Morse radar tracking to prove that the Pueblo was in international waters and that the North Koreans had known ""'-W-it-h-h-e-I-d-r-r-o-m--' Iit at the time. /

I E.O. 13526, section

1.4(c)(d)

I
L...-

(In1983, when the U.S. released SIGINTon the KAL 007 shootdown, the SIGINTrelationship with the Japanese was exposed by a blundering White House press secretary. The Japanese government was not pleased.)
,.

public release Pub. L. 86-36
--'

1.0

SABLE TO FOREIG

lONALS

447

TOP 5EC;R5T 1.1 M IRA

DOClD: 523682
'lep SEEREl' UMBRA

REF lD:A523682

Over the next several days, the White House continued to wrestle with all the ramifications of the Pueblo incident. One of the most difficult problems was that of protection of reconnaissance vehicles. The group concluded that it was impractical, given the number of such missions every year. The TRS Bartner was sent to Korea as part of the Enterprise task force, and when it patrolled the North Korean coast, it was under heavy escort. But this was more matter of showing resolve than of collecting intelligence, and the president recognized that it would be impossible to provide this sort or service to every ship and airplane engaged in peripheral reconnaissance. In an interview given to Hugh Sidey of Time magazine and Jack Homer of the Washington Star on 26 January Johnson made this point:

a

The Soviet Union and the United States have many suc:h ship. at aea and conduct literally
thousands of flights to collect intelligence by aircrat\. Neither currently provide [.icl protection .

. If they did so, they would require navies and air forces enormously greater than their present. forces.~7 .

During the various interviews and press conferences, the Johnson administration made a fairly clean breast of the peripheral reconnaissance program. During a meeting .with the National Alliance of Businessmen on the 27th, Clark Clifford explained that the United States had both SIGINT and photographic satellites in orbit, and the photo satellites "can see a tennis ball on a tennis court." Regarding SlGINT collectors such as the Pueblo, he said, "We have communication ships and very sophisticated electronic equipment .to intercept their communications. The Soviets have a number oC ships. And so do we ... The public has a bad idea about spying. However, we must do it. "sa The North Koreans continued to make propaganda hay. Several members of the
Pueblo crew were forced to make "confessions" similar to Bucher's which laid out the

effort against North Korea and sPecifically implicated NSA in the effort. SIGINT tasking documents were displayed on North Korean television, complete with the thencurrent SIGINT codewords, Trine and Savin. (This resulted in another codeword change, and the codewords adopted in 1968 have been used ever since.) In the end, there was little left to publicize that the North Koreans had not already displayed to a curious world. 58
SIGINT

The Pueblo incident also became stage to one of the biggest battles ever between NSA and the JCS. As a result of a number of developments in Southeast Asia, NSA and JCS staffers had crafted a compromise on the provision of SIGINT support to field commanders. Called MJCS 506·67, it set out new ground rules for deployment and operational control of tactical SiGINT units. When it was decided, in the middle of the Pueblo crisis, to deploy an AFSS Emergency Reaction Unit to South Korea, the JCS thought that operational control would automatically transfer to Fifth Air Force. Not so, said Carter. These resources simply augmented existing AFSS assets and were in a direct service, not a direct support, role. Therefore, operational control would continue with NSA. The JCS viewed

L SYSTEMSJOINTL HANDLE VIA __ .NE~tEL~AASSJABLE TO FOREIGN NATIONAL

Y

'FOP SECRET UMBRA

448

DOCID: 523682

REF ID:A523682
TOP SECRET UMBRA.

this as a betrayal of the compromise reached in negotiating the new document, and they ultimately prevailed, Operational control passed to Fifth Air Force on 19 February. Assessments Before the administration became caught up in a response to the Tet offensive in Vietnam, Johnson appointed a committee headed by George Ball to investigate the Pueblo incident. Ball and his committee concluded on 7 February that 1. the Pueblo had indeed been in international waters; 2. the mission had been a necessary one; 3. there had been no way of predicting the outcome, which might have been a spurof-the-moment decision by the North Koreans. "It was assumed on the principle of mutual tolerance that, 90 long as we paralleled the Soviet practice, our vessels would remain relatively free from danger .... "; 4. such missions should be continued, albeit with improved protection. Off the North Korean coast it would be necessary to provide escort vessels within a reasonable distance - aircraft on strip alert somewhere was not sufficient. Moreover, the design, armament, and equipment of the AGER-class vessels should be improved. and adequate destruction devices should be available. The rules of engagement should not bind the skipper to radio silence nor prohibit the use of defensive weapons until defense was impossible.80 In February Congress got involved. At least three different sets of inquiries were performed, including one by William Fulbright in the Senate Foreign Relations Committee. (Fulbright was acquiring an insatiable appetite for matters cryptologic, as would be revealed at the hearings on the Tonkin Gulf Resolution in August; see p. 522.) But by far the most intrusive was a subcommittee of the House Armed Services Committee, chaired by Otis Pike. On 10 March General Carter testified at length about the Pueblo in executive session. Twodays later Pike released some of Carter's information at a press conference, and Carter was furious. He had cultivated good relations with Congress and had occasionally provided sensitive information to members of certain committees when he thought it necessary." Pike's release set a very bad precedent and may have influenced NSA's response to that same congressman's far more extensive investigation of the intelligence community in 1975 - the so-called Pike Committee investigation. (At that time someone on the committee leaked the final committee report

HANDLE

-

'IA==:;===.":.'rSJO""L~
449
TOP SECRET I:JMBItA

DOCID: 523682
TOil SECREl' ~MBRA

REF ID:A523682

to the press, even though the House had voted to suppress it because it contained classified information, specifically eryptologic.) Assessments within NSA began almost immediately. Once the Agency had made its initial damage assessment (see above), Carter appointed a committee to do a more complete job. Through the spring and summer, the assessment became more refined, but a full accounting would have to await crew debriefing. To this-end the United States put on all the diplomatic pressure it could to secure the crew's release. In the end, however, the government had to sign a phony "confession" and apology at Panmunjom in order to get the crew back. They walked across the bridge at the truce village to freedom on 23 December, justin time for Christmas. The complete mishandling of the crew debriefing was emblematic ~f the entire Pueblo incident. Viewing it as an internal matter, the Navy kept NSA uninformed of arrangements for the debriefing and insisted, that NSG represent the cryptologic community. NSA viewed the assessment of cryptologic damage as their business, and finally got the Joint Chiefs to intercede with the Navy so that NSA could take its proper role. The debriefing process itself was typified by heavy friction between NSA's team and the Navy authorities ,---------, on the scene. The Navy even refused to Withheld from allow NSA's team chief, I I mblic release I to communicate with Carter, ,--P_u_b_. _L_. _8_6_-3_6----' except through him, and c:Jhad to resort to extraordinary methods to get his cables back to the Agency. reported that ••... we are encumbered by a totally uneducated admiral who has neither the rudimentary knowledge OfSIGINT, or for that matter, general intelligence, and who is in the position to edit our reports to the intelligence community." In response, Carter sent a bubbly message to Admiral Moorer, the eNO, complimenting the effectiveness of the debriefing team and the support received in San Diego (the debriefing site). Passed on to the Navy in San Diego, this message opened doors for

I

c=J

Withheld from public release Pub. L. 86-36

'fep SECRET I:JMaM

450

'.

DOCID: 523682 I E.O.
13526, section 1.4(c)

REF ID:A523682 I

..."
approach was not the smart way to
gO.82

I

ISometimes

the heavy-handed

Withheld from public release Pub. L. 86-36

Withheld from public release Pub. L. 86-36
HANDLE~MI~ ~l..I:.TOFO~·

I E.O.

13526, section 1.4(c)

I

451

TOP SECRET t;JMBM

DOCID: 523682
TaP SECRET UMBRA.

REF ID:A523682

NSA had always designed crypto devices under the assumption that the enemy would eventually capture the machine. In order to read any communications, it would also be necessary to get the keying material. This, said NSA, was the salvation of the Pueblo story. Assuming that the North Koreans turned over the material to the Soviets, they could be in position to read traffic through several crypto periods in late 1967 and early 1968, but nothing more. This was bad enough, but NSA's design principles had staved off . further disaster. Be

I E.O. 13526, section 1.4(c) I

Withheld from public release Pub. L. 86-36

Withheld from public release Pub. L. 86-36

I E.O. 13526, section 1.4(c) I

Was the Pueblo capture planned? Were the Soviets behind it? No direct evidence has ever been found regarding either charge. NSA's Robert Newton, who made the most intensive and incisive study of the incident, believes that it was planned. However, it could easily have been an extension of the on-going North Korean campaign to rid their waters of South Korean fishing boats, and there is evidence to suggest this. There is no evidence regarding Soviet foreknowledge, although their subsequent use of the captured materials is almost certain.

HANDLE VIA _

1:;:;;

ViIOT e91ftR;;;STEMSJOIN: . ASABLE TO FOREIGN NAALS

ECOMINT
452

lOll $iCRET UMBRA

DOCID: 523682

REF ID:A523682
Tep SECRET 1;.1 MBItA

It was a bad situation made worse by negligence. The crew was poorly trained, and its linguists could not even render advisory support to protect the vessel from capture. The Navy loaded it down with far too much classified material and equipment, some of it even beyond the clearance level of those aboard. The crew never practiced emergency destruction, which was next to impossible anyway given the inadequate destruction systems then available on board. There was evidence of poor coordination between captain and cryptologle crew. Following the capture, the Navy and NSA engaged in an unseemly jurisdictional battle over the debriefing process. On the Navy side, there was a lack of understanding of NSA'srole. Self-defense was only one of the problems besetting the TRS program. All the vessels had been recommissioned; most of them dated from World War II. They were becoming expensive to operate, and 1968 was to be the year in which NSA hoped to obtain money to refurbish and continue the program. Even while the Pueblo was being captured, NSA was working on an internal study of the future of the AGER portion of the TRS system. NSA felt that little was wrong with the AGERs that could not be fixed by a little redefinition of command relationships. But the Navy, strapped for cash to continue its presence in Southeast Asia, as well as elsewhere in the world, favored diverting the money to combat vessels. Both CIA and NSA put forth intelligence requirements supporting program continuation, particularly for Cuba, Southeast Asia, and the Mediterranean. But the Navy noted the difficulty and expense of protection. After a limbo period, during which each budget decision went against TRS, Deputy Secretary of Defense David Packard cancelled the program in October 1969. The last of the ships, the Belmont, was decommissioned just three months later." Surely the Pueblo and Liberty incidents were on his mind to the end.

CZECHOSLOVAKIA
Aa the U.S. tried to figure out whether or not the Sovieta would invade Czechoalovaltia in 1968, these [SIGINTI reports quite simply muddied tbe water and [chaUengedl;even conversations reported were relevant. There were just too many.
Angelo Codevilla,l nforming Stoiecro!t: InllJligefIU for
Q

the -most

-experienced all- source analystaearching for meaning and patterns in a mountain of material. The

New Century

The Soviet invasion of Czechoslovakia in August 1968 stands in history as one of the masterstrokes of the assertion of imperial control. It was masterful because of its speed, its surprise, and its brute force. It was hidden as part of a series of military exercises which,

HANDLE VIA TA ASABLETO

LSYSTEMSJOINTLY FOREIGN NATI

-

453

_TOP SECRET I;.IMBRA

DOClD: 523682
lOP SiCRE:r UMBRA

REF lD:A523682

like a tornado out of control, turned suddenly and savagely to stamp out a generation 'of new political leaders. And it allegedly took the West entirely by surprise. Viewed from a distance and as a whole, this analysis generally holds up. But viewed from up close, the generalizations begin to break down. They are simplistic and not entirely accurate. The reality is more complex. . The Prague Spring It began in October 1967. The old Communist order under Antonin Novotny was beginning to crumble. At home he had overcentralized the economic system, and in foreign policy his support of the Arab cause during the 1967 war grated on younger and more liberal colleagues. And he had dealt not very skillfully with the subsurface conflict between the Czechs and Slovaks. For all these sins Novotny confronted considerable unrest .72 I E.O. . 13526, section . 1.4(c) I Internal dissent erupted on the night of 31 October when a routine protest of the lack of electricity for their dormitories by students from the Technical College overflowed in a melee between students and lice. The t continued to bubble durin November and December

Withheld from public release

Pub.L.86-36

Novotny desperately clung to his position as first secretary of the Czechoslovak Communist Party until 4 January when the party leadership banded tog-ether to vote him out. In his place they installed an obscure Slovak nationalist, Alexander Dubcek, first secretary of the Slovak Communist Party. Dubcek was known as a good Communist, and at fll'st the Soviet leadership seemed to regard it as a routine and perhaps overdue unhorsing of a used-up Communist functionary. But Dubcek turned out to be anything but a routine Communist. Under his leadership, the Czechoslovak government quickly turned to market reforms and political liberalization which included press freedom and budding capitalism. Newspaper re rters be an calling it the "Prague Spring On 4 May according to press .reports, Dubcek and bis principal lieutenants made a hurried trip to Moscow. It was in fact a showdown with the Soviet Communist Party over the Prague Spring reforms and the general direction of Czech communism. The official communique spoke of a "comradely atmosphere," which one writer said "is Communist shorthand for cold disagreement ."75 This was followedby a series of secret meetings in the Kremlin, almost certainly onthe Czech "crisis."

Withheld from public release
-_. ..Nl~ItiH;Ej(:sAIBLE

I E.O.
TO FOREIGN NATION

13526, section 1.4(c)

L-P_U~b_._L_._8_6_-3_6~HH.A~ND~LZV~~~~~~~~~~~~L~S~Y~ST~E~M~S~J~Ol~N~T~LY~
TOI!SE(RETlJMB~
454

DOCID: 523682

REF ID:A523682

Withheld from lOP SECRETtJMIRA public release I E.O. 13526, section 1.4(c) I Pub. L. 86-36 Field reports (primarily from the ASA unit at Rothwesten) indicated that the Soviet troops were in a very high state of alert. But CIA, wading through the huge volume of reports, assessed the readiness as being related toa field exercise. This calmed the Whjte • forces did not House somewhat, and Walt Rostow told the president that Warsaw Pact appear ready to invade. In fact, it was very difficult to determine what the Soviets would

~ L
I

--..J/

I

This menacing troop buildup continued through the month, until there were some nine line divisions and three army headquarters just to the north and east of Czechoslovakia. I continued to track troop movements. (But the press also tracked the troop movements.) The situation in Czechoslovakia was tense; many believed that the Warsaw Pact would invade immediately." On 24 Maya joint communique was released announcing that Warsaw Pact exercises would take 'Place in Poland and Czechoslovakia in June. I

L-_

E.O. 13526, section 1.4(c) I Withheld from The exercise, called Sumava, played out from 18 to 30 June. Its scenario involved a three-prong invasion of Czechoslovakia, with Czech forces representing NATO as the sole public release defenders. Invading forces were Soviet, Polish, East German, and Hungarian, and the P_u_b_._L_._8_6_-_3_6---, exercise served as a dress rehearsal for the real invasion in August. At the termination, Warsaw Pact forces did not return to their bases - they ominously stayed in place until mid-July.80

I

Meanwhile, Dubce"k and the Czech leaders played a dangerous game with the Kremlin. Dubcek refused to retreat from liberalization measures and declined to attend a 14 July meeting at the Kremlin to discuss the situation. The meeting was held without him. With Soviet troops still on Czech soil, it took a great deal of courage not to back down.81

~LEVIA TALENT=:'U"ITCOWW" ....-JO,~TLy
NOT REI. FOREIGN NATIONALS

455

Tap SECREHIMBRA

TOP stele

I OMBRA

CMCbOaJovakia"

INS

HANDLE

VlA TALENT K N

MSJOINTLy

FOREIGN 456

o H
o

fQp

SEER! I UMBRA

8

DOClD: 523682

REF lD:A523682
TOP 5!C"!T UMBItA

On 23 July the Soviets announced yet another large-scale exercise, to be held along the Czech border and in western Russia, Byelorussia, and Latvia. The announced purpose was to work out rear services procedures. On 30 July they announced that the exercise would. be extended into Poland and East Germany. It did not include Czech troops." On 1 August Dubcek and his lieutenants attended an unprecedented face-to-face meating with Soviet Communist Party secretary Leonid Brezhnev and the Politburo leaders in the Slovakian town of Cierna nad Tisou. The proceedings are thought to have been acrimonious, but Dubcek did emerge f'r o m it with a "Declaration of Bratislava," a general statement of socialist principles which papered over the disagreements and preserved a measure of public agreement.83

Dubc:ek and Brezhnev in Bratislava, 4 Aug 1968, onl)' two weeks berore tbe invasion

On 20 July the control authority movedto Legnica, in Poland, and stayed there through the invasion preparations. During the last week of July, GSFG and NGF (Northern Grou of Forces) units moved to new positions closer to Czechoslovakia.

On 10 August Moscow announced the beginning of a communications exercise.

u
'---------------'' I E.O.13S26, section 1.4(c) I

)
Withheld from public release Pub. L. 86-36

I ~~--------------~~~--~
HANDLE VIA TA 5 JOINTLY

457

TOP SECRET UM!R-A

DOClD: 523682
tOP SECRET tlMBRA-

REF lD:A523682
Withheld from public release Pub. L. 86-36

I
1

E.O. 13526, section 1.4(c)(d)

_

On 18 AUlnlst, the same date that the command post exercise concluded

I

U

The welter of indicators was even more difficult to sort out in the United States. NSA was not making predictions or even doing a very good job of wrapping up the field site reporting. Since the White House had, in mid-decade, arranged for the input of SIGINT directly to the Situation Room, huge volumes of raw flowed in, but it was basically unmodulated from Fort Meade. As luck would have it, though, NSA's David McManis, the deputy chief of the Situation Room, was looking at the indicators and had established an easy dialogue with Walt Rostow, the national security advisor. He and Rostow privately agreed that an invasion was likely, although they did not have enough information to predict the date.

SIGINT

On 19 August McManis noted to Rostow that the invasion that they both thought would happen appeared to be imminentj The next day would be . time for Johnson's Tuesday Lunch with his key national security advisors. At the lunch, Rostow broached the subject of Czechoslovakia; it appeared to him that something was about to happen. In his planning notes for the president, Rostow noted: "You may wish to encourage the group to speculate about basic Soviet strategy in U.S.-So\ri.et relations at this stage, including the relationship to possible moves against Czechoslovakia .... "

I

Withheld from public release Pub. L. 86-36

I
Y1A

~OLE r;::;:A;:~~~~~~;~=:!=MSJOlNTLY
458

E.O. 13526, section 1.4(c)(d) --

I

TO!' SECRET UMBRA

OOClO:

523682

REF lD:A523682
Withheld from public release Pub. L. 86-36 TOP SECRET UMBRA

I

E.O. 13526, section 1.4(c)(d)

I

According to Restow, "We judged the Central Committee meeting as ominous, not hopeful," at the Tuesday Lunch. Richard Helms (DCI at the time) felt that the Soviets had decided to move."

\

1

Later that day, Anatoly Dobrynin, the Soviet ambassador to the U.S", called to say he would like to see the president that evening. The timing was almost unprecedented - the president knew immediately that the subject must be Czechoslovakia, and it must mean i~vasi()n.92 At about midnight, 20 August, Warsaw Pact forces, poised on the border, rumbled across. Some fifteen to sixteen Soviet divisions, augmented (for public relations purposes, no doubt) by three Polish divisions and smaller numbers of Hungarians and Bulgarians, attacked in three major spearheads. The largest contingent raced in from the north, along the East German border, toward the key cities of Prague and Pilzen, while smaller groups came in from the Soviet Union (Carpathian Military District) and north from Hungary. At the same time, airborne forces .launehed from bases in the Soviet Union (primarily Vitebsk and Panevezhis) to key nodes in Csechoalovakla."

I It was sudden, massive, and effective. ~~------~----~~--=-~~~ rolled over the almost defenseless Czech forces virtually unopposed.
95

They

~-------,

Once in Prague, Soviet troops arrested Dubcek and his liberal supporters in the National Assembly. There was little resistan,:e from the population, but the invaders, who

Withheld from .public release Pub. L. 86-36
HANDLE VIA TAL

I E.O. 13526, section
YSTEMSJOINTL Y ABLE TO FOREIGN NA

1.4(c)(d)

I

459

l'GP SECRETYMBIlt.

DOCID: 523682
TOP SECRET UMBRA

REF ID:A523682
Withheld from public release Pub. L. 86-36

had been told to expect a jubilant reception, were taken aback by the deep hostility of the Czech citizenry." No Soviet forces went on alert, and later postmortemsealled into question the validity of using alert status as an indicator of hostilities .. It was of a pattern with the tactical situation, which was evidently designed to be disguised as exercise activity.9T

1..--:,:,"::-:--"

,

The alert was probably precautionary ~--~----~~~~------------------" since the end of the Cold War the deputy commander of the Warsaw Pact invasion forces has written that the Soviets were confident NATO would not interfere, and they did not feel extreme measures were necessary.~8· . .

Following the invasion, a great national debate ensued about the Czech "surprise." Journalists were unanimous in condemning the failure of intelligence to warn. U.S. News and World Report reported that Johnson learned of the invasion from Dobrynin. Tad Szulc, in his history of Czechoslovakia since World War II, said that intelligence abounded, but "the recipients of all this intelligence input seemed unable or unwilling to interpret it adequately," and he noted that NATO did not go on alert all summer. Historian Walter Laqueur wrote that the West learned about the invasion from a radio broadcast in Prague. He Claimed that "technical intelligence [read SIGtNT] had the information, but did' not get it to decision makers in time. ,,100
• . I

They were all right, and they were all wrong. As with all intelligence analysis, success or failure depended on how you defined the two terms. Strategic warning was impeccable.

r-----------------------------,

~~=-~~------~~~~~--~--.~=-~~~--~~ .and Pact forces were poised on the border, the United States knew it.

when 20 August came,

One modern-day analyst has proposed that had DIA possessed the warning indicator system in 1968 that it later developed, it would almost certainly have published a warning report by 19 August. The case for this is good - Warsaw Pact force posture, reported I Iwas clearly at the highest level ever achieved; higher even than in May and July of the same year. The failure to publish a specific warning report. was due to the fact that the system for doing it had not yet evolved.101 Withheld from I-E-.-O-. -13-5-2-6-,-se-c-ti-on~l-.4-(c-)-

public release Pub. L. 86-36

HANDLE V4\ TA
SABLE TO FOREIG

ROLSYSTEMSJOINTLY

Tap SECRET blMSRA

460

----''--------_._.

.

-

DOClD: 523682 I E.O.
13526, section 1.4(c)

REF lD:A523682
lOP SECRET UMBRA

The president knew as much as was knowable by the afternoon of 20 August and was not, contrary to press reports, surprised by what Dobrynin had to tell him. What good would it have done to alert NATO forces? NATO could do nothing anyway. Better to stay cool and look surprised.

... .

public Pub.L

d from elease 86-36

Withheld from public release Pub. L. 86-36
HANDLE VIA _----1mrlt"ID:LEASABLE

I E.O.

13526, section 1.4(c)

I

LSYSTEMSJOINTLY TOFOREIGN NAT

461

TOP SEeR!! ijMBRA-

DOCID: ·523682
TOP SECRET UMBRA

REF ID:A523682

E.O. 13526, section 1.4(c)(d)

I

Romania - The Invasion That Never Happened

Withheld from public release Pub. L. 86-36

On the last two days of August,1 reports began to arrive at the White House concerning a possible Soviet move into Romania to bring the errant Communist regime of Ceaucescu back into line.\

I As it happened, the White House had been concerned about this possibility as early as the 23rd. Romania had pursued an independent foreign policy since 1964, and during the Czech crisis had pointedly supported Dubcek (alone within the Soviet Bloc). Soviet troop movements in areas peripheral to Romania could be interpreted as threatening to that country. too. Rostow contacted NSA; the Agency re lied that it did not look like an invasion t them and the White House calmed. Just to be on the safe side, however, President Johnson ~--~--~~--~--~~~ issueda public warning to the USSR on the first week of September. Romanian diplomats thanked the president for his support, and the crisis seemed to subside. 1M Rumors continued, but NSA stepped in again. In October the Agency again wrapped up recent activity, and it concluded that the Soviets were not about to move on Romania.105

,----------, Withheld from public release Pub. L. 86-36

In contrast to its performance on the Czech crisis, the cryptologic community was widely praised for its role on Romania. The difference appeared to be the active participation of NSA, which headed off speculation at every turn. Romania was the invasion that did not happen, and NSA's calming influence was noted at the White House. E.O. 13526, section 1.4(c)(d) THE SHOOTDOWN OF THE EC·121 The SIGINT crises of the decade came to a tragic end in 1969. The North Korean shootdown of a Navy EC-121, with the loss of all thirty-one men aboard, was one of those

lOP SECRET UMBRA

462

--------_

..

DOCID: 523682

REF ID:A523682
TOP SECRET 1:1 MaRA

. transcending events that precipitated drastic changes in the crisis structure Headquarters. The effects are still felt today. . North Korea and the Aerial Reconn8issance Program

at NSA

By taking the Pueblo in January 1968, Kim Il-sung's North Korea had once more branded itself as an international outlaw. As the United States redoubled its efforts to protect its peripheral reconnaissance missions, North Korea' continued its pattern of infiltration and subversion. In November 1968, a group of 120 well-armed commando infiltrators landed by sea on the east coast of South Korea and infiltrated villages in the area. It required 40,000 ROK militia and police nearly 2 months and the loss of 63 lives to clean out the groUp.lOO The situation on the ground was not necessarily mirrored in the air: Over the years there had been five incidents involving North Korean and American aircraft. Only two, involving RB-47 aircraft in 1955 and 1964, affected the peripheral reconnaissance program. In neither case was the aircraft shot down, so in reality North Korea had never shot down a reconnaissance mission, although they had tried twice. Considering the unsettled situation around the DMZ, and the hostility demonstrated 'by the Soviets and Chinese to this sort of electronic spying, this was not considered to be a very high number of incidents. 108 To see Soviet fighters in reaction to a peripheral reconnaissance mission was normal; often the Soviets would send fighters out in relays to pace the aircraft, staying between it and the Soviet coastline. By the mid-1960s, however, JRC had decided that the Asian Communist nations fell into a different category. When one of them launched a fighter in reaction, which was rare, they meant business. Because of this, two new conditions had been inserted into the White Wolf plan. Condition 3, which would be called any time a hostile fighter was seen headed over water within 100 nautical miles of the mission, required a heightened state of alert aboard the aircraft and diversion to a fallback orbit farther off the coast. If the fighter came within 50 nautical miles, this would be changed to Condition 5, which required an automatic abort. Since the institution of these new conditions, the U.S. had lost no missions to the PRe, North Korea, or North Vietnam.109 Navy and Air Force SIGINT reconnaissance missions were almost daily occurrences off the North Korean coast. One of the most frequent visitors to the area were the EC-121 aircraft, nicknamed BEGGAR SHADOW, from the VQ-l squadron in Atsugi, Japan. A large, slow, lumbering. Lockheed aircraft designed to haul passengers, the EC-121 had become the easiest target in the Navy inventory. But it was bigger than its sister collector, the EA-3B,1 I For this reason it was still the aircraft of choice for fleet support. 110..-,

I E.O.

13526, section 1.4(c)

I

;~~.~
N

Withheld from public release Pub. L. 86-36
LETOFuKJ<;It;N~

, 463

TO••5!CIU!T tlMSftA

DOCID: 523682
TOP UCRiT UMBRA

REF ID:A523682
Withheld from public release Pub. L. 86-36 .

I E.O.

13526, section 1.4(c)

And fleet support was the mission. BEGGAR SHADOW aircraft were Seventh Fleet assets. They were tasked and technically supported by] KKami Seya). NSA submitted secondary tasking, but the Navy jealously guarded operational control, and NSA's tasking often had little effect on the mission. 111
I

The week before the mission, General Charles Bonesteel, commander of U.S. Forces in Korea, warned of unusually vehement language and surly protests by the North Koreans at Panmunjom. The warning was sent to the V·Q-l squadron, which was advised to be extra cautious. But the North Koreans appeared to suffer through profound mood swings at the Armistice Commission meetings, and neither Seventh Fleet nor CINCP AC changed the risk category of 3 (hostile action unlikely). Conditions 3 and 5 appeared to cover any potential problems, anyway. 112 Despite the relative venerahility of the White Wolf warning program and its apparent good effect (there had been very few incidents since it had been instituted in the early 1960s), VQ-l aircraft were only loosely cobbled to the system. According to a senior NSA official involved with White Wolf, the Navy was an "unenthusiastic" player in White Wolf. Unlike the Air Force reconnaissance aircraft, the EC-121 had no secure method of contact with the ground. For warning, they relied on SAC HF broadcasts labeled "Sky King," which could not be acknowledged. Thus the ground station personnel issuing a condition did not know if a transmission had been received, or what the situation was aboard the aircraft. Moreover, the key Navy units involved in the mission (including at Kami Seya) were not on distribution for reports issued by AFSS sites watching the mission.

I

I

The Mission The doomed aircraft departed Atsugi at 0700L with a double load of thirty-one crewmembers - the excess members were in training status. It was to fly across the Sea of Japan to a point off the northern coast of North Korea, do two and a half orbits, and land at Osan Air Force Base in Korea. The EC·121 was reflected by both Soviet and North Korean radars,'

I
Withheld from public release Pub. L. 86-36

I E.O. 13526, section 1.4(c) I

HANDLE viA ~:;:?i'1? KiYtlOI 5 CQMIN'f CV;:OLSYSTEMSJOINTL

Y

lOP SECRET l:JMBAA

464

DOCID: 523682

REF ID:A523682
+9P SECRET tI MIRA

EC-121

I E.O.

13526, section 1.4(c)

I

Withheld from public release Pub. L. 86-36

The tracks

of BEGGAR SHADOW, 15 April 1969 .

HANDLE VIA TALENT K

YSTEMSJOINTL

Y

LEASABLE TO FOREIGN NATION

465

Tap SEeR!T UMBRA

DOClD: 523682
TOP S!CRET UMBRA

REF lD:A523682

E.O. 13526, section 1.4(c)(d)

I .

Withheld from public release Pub. L. 86-36

At about 1330, as the mission was nearing the topmost portion of its last orbit, two North Korean MIG-21s scrambled from the training school at Hoemun. The fighters had been there for about two weeks - it was unprecedented for MIG-21s to be at Hoemun, and their purpose there was never explained. As was customary, Osan waited for a second plot before issuing a Condition 3. They did not get one for eight minutes, at which time the fighters were reflected at about flfty-five nautical miles from the mission and closing fast. One of them peeled off to make a defensiveI patrol. but the other bore on straight for the . mission. At 1340 Osan issued a Condition 5, as the second MIG-21 was by this time reflected as well under fifty nautical miles from the mission. Only four minutes later I I the two aircraft merging. The shootdown probably came at 1347, while the mission was about eighty nautical miles from the coast. The tracks separated at 1349, and Soviet facilities ceased reflecting the mission two minutes later. The MIG-21 was headed home by that time. 114 AFSS reporters at Osan were concerned. The North Korean reaction was virtually unprecedented, and Soviet radar tracking was ominous. They were in close touch with 314 Air Division in Korea, and at 1345, two minutes prior to the shootdown, Brigadier General Arthur Holderness, 314 AD commander, directed that F-102s be launched in case of trouble. But, incredibly (considering the Pueblo incident the previous year), the Navy had not requested strip alerts, so no fighters were actually airborne until shortly after the hour. The analysts spent the ensuing forty-five minutes replotting the mission and communicating with I I in Misawa and 5th Air Force in Japan trying to see if anyone else had any information. The feeling was that the aircraft. must have "hit the deck" to evade the MIG-21.m

I

I

At the same time, Kami Seya was completely in the dark. They were making communications checks, but they were getting nothing in reply. I had issued a Spot Report, butl lviasnot on distribution. The VQ-1squadron was monitoring the SAC HF broadcasts, so they knew something was amiss, and they were making repeated calls to the air control facility at Fuchu asking for information. 116 I

r

Finally, at 1444, almost an hour after the shoctdownj issued a Critic. Still, no . one knew for sure what had happened until FBIS monitored a 1600 North Korean broadcast claiming to have shot down a "spy plane." By then the aircraft. was half an hour overdue at Osan. U 7 Fifth Air Force aircraft. swarmed to the spot,but debris was not spotted until the next day by a naval P-3. Eventually two bodies were recovered, along with some debris. Although Soviet vessels participated in the search and rescue (SAR) operations, compromise of classified material was never a significant issue, as it had been with the
Pueblo. us

I

Withheld from public release Pub. L. 86-36
HANDLEVIAT

E.O. 13526, section

1.4(c)(d)

I

fap SECRET UMBRA

466

DOCID: 523682
I E.O. 13526,section 1.4(c) I

REF ID:A523682
Withheld from public release Pub. L. 86-36
TOP S!CIt!T tJMBItA

Whilel was trying to figure out if they had a shootdown or not, the Current SIGINT Operations Center at NSA had called Major General John Morrison, the assistant director for production .. Morrison began coordinating the NSA response, but found it almost impossible. A Group had a crisis response center (the CSOC) with analysts and reporters
L...---:---::---'

I

But B Group had nothing equivalent to it, and analysts had to be called to duty in the middle of the night. By 0330 Local, CSOC had fashioned a follow-up to the Critic,

'--~..,----J

----------,

Withheld from public release Pub. L. 86-36
L---'

Morrison wore out his shoes walking between the A and B Group areas to try to get a coordinated response. The follow-up finally went out at 0500, but not before a thoroughly frustrated Morrison had vowed to John Morrison consolidate his crisis and warning facilities into a single organisation.l'"

The C~~;13526;section 1.4(c~

I

NSA's disorganized response was reflected at the White House. I At the Situation Room, David McManis was trying to piece together the details, and he was.on the phone with several different NSA divisions. He finally found it necessary to drive to NSA and get together the materials that he would need to briefthe president. 120 The shootdown plunged the new Nixon administration into its first international crisis. During the campaign Nixon had criticized the Johnson administration's handling of the Pueblo capture, and he had vowed to demonstrate that the Republicans were made of sterner stuff. Henry Kissinger, the new national security advisor, prepared a list of options which included a B~52 strike (according to journalist Seymour Hersh), and bellicosity nearly carried the day. But in the end the solid opposition of the secretaries of state and defense (Rogers and Laird) and the DCI (Helms) wonout.l21

467

TOP SECRET tlMSItA

DOClD: 523682
TO' S!CRET UMBRA

REF lD:A523682

Instead, the administration launched a diplomatic offensive. The cornerstone of this offensive was a presidential press conference on 18 April. There. Nixon, using data supplied by NSA, stated that intercepts of Soviet and North Korean radar reflections proved that the aircraft had been in international waters. This second presidential release of SIGINT information in fifteen months (the first went out during the Pueblo crisis) . occasioned a very detailed damage assessment study at NSA. In the end, John Morrison's DDO team could find no evidence of drastic changes to either North Korean or Soviet communications.P" Whatever changes were needed by both countries had probably already been made after Pueblo. And exploitation of Soviet air defense communications had been a matter of public record since the release of tracking information on the 1958 RC-130 shootdown. By 1969 this exploitation was no longer a secret to anyone who could read the newspapers. The administration decided ultimately on a military show of force in the Sea of Japan, a move almost identical to that which Johnson had made in January 1968. A massive flotilla was assembled under the name Task Force 71. It included three carrier task groups and 250 aircraftJ On 24 April AFSS flew a special RC-130 mission off the North Korean coast, heavily defended by American military might. By then, however, NSA had concluded that North Korea had crawled back into its leathery shell and was no longer an immediate threat. Moreover, there was no evidence that the Soviets or Chinese Communists were in any way involved in the incident.l2S

rr+ __

--,

I

'-----'

.A Washington Post story on 17 April called into question the value of the peripheral reconnaissance program. It was a good question, and it got a thorough airing in the Pike Subcommittee, which was still investigating the Pueblo capture. House Armed Services Committee chairman Mendel Rivers simply added the EC-121 shootdown to the list of things that Pike was tasked to look into.1U While General Carroll of DIA came out four-square in favor of the reconnaissance program, John Morrison was not so categorical. Morrison, an Air Force general,could see the value of the Air Force program, which appeared to him to be better managed, used -, more capable aircraft, participated more fully in PARPRO (the Peacetime Aerial Reconnaissance Program) - and were, hence, safer - and were more fully under national control. The Navy program, Morrison thought, suffered from a lack of all these attributes. NSA was getting only minimal value and had no control at all. Morrison stood his ground before Carroll and the Navy on the issue. He commissioned an internal NSA study of the situation, which basically backed up his gut feeling. It was the second serious run-in between NSA and the Navy on peripheral reconnais~~mce.
I

The Post reporter, who seemed to have impeccable sources, also cited the extended delay in reporting the incident from the field. General Wheeler (chairman of the JCS) also raised questions, and NSA was called to answer. An internal investigation completely exonerated I Ifocusing on its performance of advisory warning functions (on which E.O. 13526, section 1.4(c) I ~~~~~Ce~~I~:S~
Pub. L. 86-36

HANDLE vIA
-

!=:::s:~~
468

~~::rE;~!;'~73;!~MSJOINT~

TOP SECRETI:JMBIbtc

OOClO: 523682

REF lD:A523682
lOP SECRET tlMBRA:

it did a credible job) rather than on the delay in issuing the Critic.l~ This approach seemed to quiet external criticism, but any goodfield reporter knew that the Critic should have been issued as soon as there was any considerable doubt as to the fate of the mission. The investigation begged the real question. The Pike Committee expressed disquiet about the real value of such airborne reconnaissance in view of the cost in dollars and lives over the years. Some of the committee's concern may have stemmed from NSA's unwillingness to defend the 'Navy's ,----------, programs. Pike recommended that the full Armed Services Committee take a more active Withheld from role in monitoring the programs.12e I E.O. 13526, section 1.4(c) I public release The committee was also very critical of interservice disconnnects. The members cited ,--_P_u--,b_o _L_o_8_6_-_3_6-.J failure of the VQ-1 squadronl to receive any information from the Air Force about the mission until they received the Critic, and they noted that this time delay. contributed to delays in launching the search and rescue effort. They were incredulous over the failure of the Navy to ask the Air Force for fighter strip alerts, especially so soon after the PW!bkJ incident. 127

I

I

.

The rivalry between the Navy and NSA was not defused until General Carter stepped down as director. The new director, Admiral Noel Gayler, had the contacts within the Navy to build bridges, and as the new director he took NSA's case directly to Admiral John Hyland, CINCPACFLT commander. Gayler wanted closer NSA involvement with Navy SIGINT reconnaissance, and the authority to task missions. 'He eventually got part of what he wanted - NSA began tasking a few VQ-1flights in the Pacific area.l28' The 1960s absolutely overflowed with SIGlNT crises. After the Arab-Israeli War of 1967 and the Pueblo capture of 1968, John Morrison proposed to General Carter that NSA .establish a single national SIGINT watch center. The proposal was still hanging fire four, months later when the EC-121 went down. Morrison pressed Carter for a decision, and on 17 July 1969 he got one. In the twilight of his term, Carter concurred with the establishment of a National S[GIN'I' Operations Center (NSOC). Morrison himself was charged with putting it t;Ogether.l29 As for the EC-121s, their time was almost over. A Navy Board ofInquiry, looking at the shootdown, noted the cumbrous nature of the aircraft (maximum speed 220 knots) and low headroom (maximum altitude 10-20,000 feet), and the board recommended that something better be procured. The replacement was the EP-3E Orion, which gradually took over all EC-121 orbits. The EC-121s were moved back to safer orbits until they could be mercifully retired.130 Was the shootdown a deliberate act? Conspiracy theories usually require wild flights of imagination, but in this case it was the only explanation that made sense. Like the Pueblo capture, it seemed to follow no known North Korean procedure, and it did not appear to have simply been a routine operation gone haywire. Instead; ,it appeared to be a carefully preplanned event, from the placing of two MIG-21s at a training base that had

~DLE VIA =::;;~~~;~~=~:;:;~;!~:MSJOINT~
469 lOP SEC;RiT UMBRA

DOClD: 523682
lOP SECRe=FtJMBRA

REF lD:A523682

\

never seen them before, to the flight pattern of the aircraft that allowed for little misinterpretation of intent. The shootdown happened to occur on Kim II-sung's birthday, which led to speculation that it was a planned birthday present. Of course; the North Koreans had to hope that the JRC reconnaissance schedule conformed with Kim's birthday, which makes this part of the theory rather tenuous. It was likely just another of North Korea's xenophobic strikes. reconnaissance aircraft was in the way. SECURITY AND THE WORK FORCE IN THE 1960s Success on the cryptologic front did not translate into the security field. A succession of security problems in the early 1960s, begun in the summer or 1960 with the infamous Martin and Mitchell defection (see pg. 280), rocked the NSA community. For the first four years of the decade, it must have seemed like the sky was falling. Dunlap The House Un-American Activities Committee investigation into the Martin and Mitchell affair ended in 1962 when a final report was issued. Legislation to give the director additional powers to dismiss personnel, which resulted from the committee recommendations, was still dragging through Congress when in July 1963 an Army sergeant named Jack Dunlap committed suicide. A month later his wife showed up at NSA with a pile of classified documents which, NSA's security organization discovered, Dunlap had been selling to the KGB. Sergeant First Class Jack E. Dunlap had first come to NSA as the driver for Major General Garrison B. Coverdale, the chief of staff, in 1958. Dunlap had up to that time served a. rather uneventful career in the Army, which included service in Korea as an . infantryman. While overseas he had worked as a technician and messenger for ASA, which got him close to the security business. But Dunlap was affiicted with serious character flaws. He liked money, lots of it, and when he had it, he spent it on yachts, fast cars, and faster women. Once at NSA, he discovered how to get it. Sometime in This time a U.S.

Jack Dunlap

TOP SECRET tJMBItA

470

DOCID: 523682

REF ID:A523682
'f6P SECRET UMBRA

May orJune 1960, Dunlap walked into the Soviet embassy in downtown Washington and offered to sell classified documents. He claimed he could get his hands on them.1S1 Dunlap smuggled classified documents out of NSA literally under his shirt. He did not work in a technical area, had no knowledge of cryptology, and probably did not steal documents in any organized fashion. But he knew that the documents were worth money. He was in and about Coverdale's office and just scooped up whatever became available. The FBI and NSA security people were never able to determine with any certainty just what Dunlap had sold.lS2 Twice the Army alerted Dunlap for overseas assignments. This represented a serious threat to his lifestyle, which by that time included two Cadillacs, a Jaguar, a thirty-foot yacht, a world-class hydroplane, and a blonde mistress. The first time, Dunlap evaded the assignment by pleading a bad back. The second time, he informed the Army that he intended to resign, and he applied for a civilian position at NSA.1SS He did not get very far. His initial polygraph turned up evidence of petty thievery, immoral living, and living beyond his means, and his second try did not go any better. NSA initiated an investigation and withdrew his access to classified material. The investigation began in May, and the FBI interrogated him on 17 July. Apparently convinced that he was about to be exposed, Dunlap committed suicide six days later by inhaling carbon monoxide. Later in the summer his wife turned up with the classified documents that were still in the Dunlap residence. 134 The Dunlap affair brought further unfavorable publicity to NSA, but it did represent a success of sorts. Had the polygraph not been in place, Dunlap might have have been hired in some capacity and would have continued his espionage. The incident renewed discussions about requiring military assignees at NSA to take the polygraph, but the armed services staunchly opposed it, and successive directors (Blake and Carter) made little headway. The custom of excluding the military from the polygraph did not finally end until 1985. Much criticism attended the revelation of Dunlap's lifestyle, which had gone unreported by coworkers. Further, the affair spotlighted the ease with which employees could spirit classified documents out of the Agency. The impact was the initiation of exhaustive exit inspections, which continued for thirty years (until 1993), and a continuing focus on employee lifestyle, a point that was hammered home to NSA employees again and again during security awareness sessions. Although Dunlap is deceased, his ghost has lived ever after in the halls of Fort Meade.

HANDLE VIA TALEN _...Nio:P-lrtEt;E1;:WLE

L SYSTEMS JOINTLY TO FOREIGN NA TI

471

TOP SECRET lJMBRA

DOClD: 523682
TOP SECRET lJMBRA

REF lD:A523682

Hamilton The same day that Dunlap committed suicide, the Soviet newspaper Izoestia published an article about NSA attributed to one Victor Norris Hamilton, a former NSA analyst. The third security crisis of the young decade had burst on the Agency. Hamilton, whose family name was originally "Hindali," was Lebanese by birth. He met and married an American working for Point Four (a foreign aid program) Libya in 1953, and emigrated with her to the United States. Hamilton's fluency in Arabic attracted the attention ofNSA, and he was recruited for employment in 1957.l.S5

in

He remained at NSA for only two years. In early 1959 Hamilton began evidencing psychological problems, and he was sent to the medical staff for an evaluation. He was diagnosed as paranoid schizophrenic, but refused hospitalization, and he ,was medically terminated in June. He visited Moroccobriefly but returned dissatisfied. He applied for employment at CIA, but there was no billet available for him. NSA tried to get him committed for psychiatric evaluation, working through his wife, but this ralled. In 1960 he wrote a letter to the House Armed Services Committee claiming that an agent had offered him money to do business with the Soviet Union. The matter was turned over to the FBI, which tried unsuccessfully to interview him. He worked briefly as a teacher in Iraq but was discharged, and he dropped out of sight from May 1961 until the Izvestia article appeared.

E.O. 13526, section 1.4(c)(d)

Withheld from public release Pub. L. 86-36

Hamilton brought more opprobrium to a besieged NSA security organization. Yet in his case, as in Dunlap's, it could be argued that the system worked. His initial hiring was, in retrospect, inopportune, but the internal screening system weeded him out before he progressed into more responsible positions, The severe embarrassment of the publicity surrounding the Izvestia article had less impact on NSA's posture than was predicted at the time. \
Withheld from

I

I. E.O. 13526, section

1.4(c)(d)

_

public release Pub. L. 86-36

'FQP SeCRET YMBRA

472

DOClD: 523682

REF lD:A523682
Withheld from public release Pub. L. 86-36
fOP SECRET !:IMBRIe

I E.O.

13526, section 1.4(c)(d)

In March of 1974 the State Department reported to NSA that Hamilton was being detained in a Soviet psychiatric hospital. A Jewish emigre made a positive identification of Hamilton based on a photograph, and NSA closed the case in June. 1S8 The Hamilton and Dunlap cases heightened the sense of urgency in Congress about NSA personnel policies. When in 1964 Congress enacted PL 88-290, giving the director more authority to hire and fire NSA people, the legislation owed much to the three security cases that immediately preceded it. David Kahn and The Codebreakers The wave of'publieity surrounding the Martin and Mitchell case interested a Newsday reporter named David Kahn. Kahn already had an active lifelong interest in cryptology sparked by his youthful reading of Fletcher Pratt's book Secret and Urgent. Subsequent to the Martin and Mitchell expose. he wrote an article for the New York Times Magazine on the influence of cryptology on current events, and this spawned a publishing contract with MacMillan. TM Code breakers, a monumental work on the history of cryptology. was published in 1967 to a good deal offaiUare. It was, and has remained. the definitive work on the subject in the open press. The publication was not a welcome development at Fort Meade. When NSA learned of the forthcoming book, it obtained a copy of the manuscript from the publisher, Without a reasonable hope of cooperation from either Kahn or MacMi11an, the Agency reviewed the manuscript and marked a few passages for modification or deletion. To NSA's surprise. Kahn, then in Paris, reviewed the changes.and agreed with virtually all of them. The material NSA wanted removed related to UKUSA collaboration and was not central to Kahn's thesis.139 Although Kahn was reasonably cooperative, many other journalists were not. Press leaks relating to American cryptologic efforts became more troublesome over the decade, as the interest of the American public in NSA increased. Beginning as early as 1961, for instance, the New York Times quoted the presidential press secretary about the launch of Soviet manned space vehicles which referenced "listening posts" in the Middle East intercepting traffic between the launch site and downrange tracking stations. The next year Newsweek published references to satellite intercept of Soviet microwave transmissions. In 1966 the New York Times published a series of articles on .SIGINT collection at the U.S. embassy in Moscow and on satellite intercept of Politburo-level limousine car phones.1'o A year earlier a press photo of McGeorge Bundy with President Johnson contained a copy of the CIA Daily Bulletin with a clearly visible "Top Secret Dinar" (the then-current Category III COltUNT codeword) stamp affixed. This produced

473

TOP SECRET \:JMBRA

DOCID: 523682
Tap SECRET l:JMBItA

REF ID:A523682

numerous press references to a "codeword so secret the very existence is classified." All the reporters seemed to know that the codeword referred to SIGINT, even at that relatively early date. The anonymity that NSA had enjoyed in the 1950s was slowly disintegrating.v" Cryptology is Legalized The legal existence of a COMINT effort, rendered precarious by the 'Federal Communications Act of 1934, was finally established in 1968. The Omnibus Crime Control and Safe,Streets Act of 1968 dealt specifically with the issue: While prohibiting' all wiretapping and electronic surveillance by persons other than law enforcement authorities (and even then under restriction), it stated that
Nothing contained in this chapter or in section 605 of the Communications Act of 1934 ... shall limit the constitutional power of the President to take sw:h measures as he deems necessary to ... obtain foreign intelligence information deemed essential to the security qf the United States. ... 14~

It did sojust in time; the Watergate period and the attendant Church and Pike Committee hearings called into question all that was illegal about espionage, and much that was legal, too. The 1968 legislation provided a much-n~eded defense for NSA and the cryptologic community.

AMERICAN CRYPTOLOGY AT THE END OFTHE DECADE

It is important that you recognize the systematic character of the cryptoJogic enterprtse; that ita integrity must be maintained because the challenge with which it is confronted cannot be met if that system is debilitated. fragmented, or destroyed. General Marshall S. Carter on the occasion ofhia retirement, 1 August 1969

By the end of the 1960s, cryptology had become big business. SIGINT product reports had become common paperwork in the White House and at every level down from that. NSA sent representatives to nineteen organizations, ranging from enormous military commands like CINCPAC tol 1 A study of strategic warning done in 1967 called COMINT ••the workhorse of warning intelligence; no other source can match its continuity, timeliness, and span of coverage. "f43

I E.O. 13526, section

1.4(c)(d)

I

,--------,

Withheld from public release Pub. L. 86-36

lOP SECRET IIMBRA

474

DOCID: 523682

REF ID:A523682
Withheld from public release Pub. L. 86-36
lOP SECRETUMBM

I E.O. 13526,section 1.4(c) I

The cryptologic community was at its height in terms of personnel numbers. NSA employed about 18,000 peopleD percent of them military), while the SCAs nadi The total, aboutc=:Jmen and women, was a strength that had never been reached before and has not been attained since.l« Relationships with the Military

I

Paradoxically, the relationship between NSA and the military commands had never been at such a low ebb. Strains in tailoring SIGlNT support had developed during the Vietnam War. A series of situation-specific compromises had papered over the differences, while leaving the underlying issues unresolved. At mid-war, 1966 and 1967, NSA and the JCS had tried to hack out a comprehensive agreement concerning the use and control of SIGlNT resources. The resulting document, called MJCS 506-67, left DIRNSA in overall control of all SIGINT assets but provided that under certain circumstances certain types of assets would be delegated to the tactical commander .: The memo carefully defined the procedures for doing this, and for the first time the role of the cryptologic support group was defined and standardized. 14$ The trick was in universal interpretation and smooth implementation. The first try, during the Pueblo situation, collapsed in howling controversy, and it colored relationships for several years to come. Although the agreement was employed more successfully in later years, difficulties persisted. In 1967, the same year that MJCS 506-67 was published, the Army convened a board under Brigadier General Harris W. Hollis to "examine cryptologic and related activities." . At the root of this study were deep-seated differences between NSAand the Army over the management of cryptologic assets. The Hollis Board recommended a series of steps which would have both pulled ASAresources away from DIRNSA control on the one hand, and on the other, given ASA a more favored seat at the cryptologic table. . Hollis made a pitch to transfer ASA direct support resources from the CCP to the Army general-purpose program. This proposed move would have fragmented cryptologic resources while divorcing the Army from the CCP system. NSA opposed it, while recognizing the tendency to fully fund big-site resources and programs at the expense of tactical assets. Hollis also recommended that ASA be given operational control of tactical SIGlNT resources at all times - the Army deferred thiS.I46

"RAND~=T N

KEYHOLECOM~~:~~SYSTEMSJOINTLY i;bEASABt£RiPo 4:0NALS

.---

.

475

TGP SE'RET UMBRA

DOCID: 523682
JOP S&CRETUMBRA

REF ID:A523682

,---------,

Withheld from public release Pub. L. 86-36 '----------'

Distressed at the increasing concentration of resources at Fort Meade, the Hollis Board made a number of proposals that would have strengthened in-theater ASA processing.· This move to improve SCA theater assets amounted to an attempt to halt the tide. The waves of cryptologic centralization continued to wash inexorably over the valiant Hollis Board, and nothing came of the attempt.':" I E.O. 13526, section 1.4(c) I Finally, Hollis proposed that the Army become more involved in centralized crypto\ogic activities, by taking a role in futuristic projects like I Iand by increasing its manning at Fort Meade. While pointing out that ASA had already been given a piece ofl I(a logistics piece, but nonetheless a piece), NSA noted deepening trends in the opposite direction. Army policy led in the direction of diversification, especially at the officer level, rather than toward the cryptologic specialization that was required for greater ASA participation in the centralized cryptologic system.148 It was an ominous trend which led ASA in a tactical direction and which eventually caused it to virtually abdicate its unique SIGINT expertise, established so laboriously by Friedman and others in the 19305. The debate over SIGINT control intensified in 1969 whenJCS promulgated a new policy document for electronic warfare, called MOP-95. Electronic warfare (EW) had always been outside the purview ofSIGINT, but MOP·95 broadened the definition ofEW to include a new category called Electronic Warfare Support Measures. The new category sounded just like SIGINT, but without the codewords or centralized control. General Carter attacked the new JCS document, to no avail. The armed services continued to develop EW capabilities, in league with the SCAs, which were happy to participate in a new effort . divorced from NSA control. 140 . . During the summer of 1969, as General Carter's term as director wound toward its end, the Joint Chiefs were cOnsidering a direct assault on NSCID 6. The objective was to expand JCS authority over cryptologic assets, at the expense of DIRNSA. . Carter found out about the draft, and in a phone call to General Wheeler (chairman of the JCS) he called it an "absolute monstrosity." The revision of NSCID 6 was going through coordination when it was halted by Admiral Johnson, director of the Joint Staff, to await the appearance of Admiral Gayler at Fort Meade.l50 Marshall Carter Retires Weary of.conflict with the services and debilitated by medical problems, General Carter retired in August of 1969. But before he did so he loosed one final blast. In a letter . to Secretary of Defense Melvin Laird the day before his retirement ceremony. he . characterized the state of cryptologic management as "diluted."

lOP SEeRETUMBRA

476

DOCID: 523682

REF ID:A523682
TO' !!CRET UMSRA

Despite the vigor. ingenuity, enterprise, and growing competence of the national cryptologic establishment which emerged almost seventeen years ago, subsequent administrative and organizational arrangements ... have diluted the original concept and clouded the original goals. More and more·commoD tasks have been assigned outaide the crYptologic community, with a corresponding !ossorefficiency and economy.!S!

He excoriated the legal hairsplitting that had been employed to shave cryptologic resources from the central system, to call a duck something other than a duck in order to free it from NSA's control. He was pessimistic about the future. Carter was asked to hold invitations to his retirement ceremony at the Pentagon to 150. He invited only 3 people and zipped through the ceremony in ten minutes. The Pentagon was as happy to see the last. of Marshall Carter as Carter was to leave the wars.l~3 Gayler Takes the Helm With Carter on the way out, the Department of Defense decided to experiment with a new kind of director. Instead of appointing an intelligence specialist on his final military assignment, DoD nominated an admiral with an operational background and ambitions to go higher.

Lt Gen Carter shows incoming DJRNSA VADM Gayler his office.

HANDLE viA IALEN! KE i ROLE COMtNT CONtROt 5 i SrEMSJOINTL NO! RELEASABLE 10 fOREIGN NAI'LONALS

1

477

TO!' SECRET UMBRA

DOCID: 523682
TOP SECRET tJMBRA

REF ID:A523682

Noel Gayler was untainted by the intelligence business. The son of a Navy captain, he had gone into naval aviation soon after his graduation from Annapolis in 1935. Gayler had served as a flyer in the Pacific in World War 1I,.following which he had had many years of both operational and stafi'experience with the line Navy. He had been only the third naval officer ever to fly ajet aircraft; and.when he was nominated to-fill Carter's job, he still held the record for the longest flight from an aircraft carrier. He was a known protege of Elmo Zumwalt, the new and reformist CNO.l5S Gayler was the most unusual director in NSA's history from many aspects. Personally, he was dynamic, mercurial, and high-strung. Gordon Sommers, a senior civilian at USAFSS, described Gayler's management style as all Navy.
Gayler c.ame from a Navy background. and his perception of command and control was .~e captain on the bridge of the ship with a speaker tube down to the boiler room yelling orders to throw more coal on the fire, and everybody down to the lowest level threw more coal on the fire.1M

His impatience with briefers was legendary, and he was known to throw things when especially agitated. He seemed to strike out in all different directions at once, and he moved with dizzying speed from one topic to another. Short, stocky and athletic, he resembled a fireplug in constant motion. Gayler was put in the job to repair the damaged NSA.JCS relationship. He understood that he was to open up channels of communication, that he was to talk to the operational officials on the Joint Staff and get things moving again. One of his first moves was to create a permanent NSA representative to the Pentagon, accredited to the JCS, the military departments, and the officeof the secretary of defense .15$ He was immediately confronted with theJCS staff papers, forwarded to him by Vice Admiral Johnson. The papers were more than just critical - they amounted to an . indictment. In his reply to Johnson, he said that the basic directives (i.e., NSCID 6) seemed to be sound and that "any difficulties have been occasioned by the attitudes of personnel involved" (a clear reference to his predecessor and his antagonists). He believed that he could patch things up through personal diplomacy, and he began calling people at . the Pentagon. Within weeks he had defused the situation.l56 Although he did put NSA back on speaking terms with the military. it is hard to see how he accomplished it. His personal relationship with most of the Joint Chiefs was cold to the point of hostility. But Gayler was politically astute, and he moved easily in Washington's power elite despite his mannerisms. When he departed, he was rewarded with the plum assignment of CINCPAC and got his fourth star, the first NSA director ever so elevated.

HANDLE V1A1:::

KE¥l:wr E ClWIl'7!l' SaN I~~;STEMSJOINTLY

.

TOP SECRET tlMRRA

478

DOClD: 523682

REF lD:A523682
TOP SecRET tJMBRA

The Eaton Committee
By 1967 the SIGINT budget passed $1 billion, and manpower stood at nearly 100,000. Officials at the Bureau of the Budget were already taking a close look at the CCP when General Carter sent over his CCP proposal for FY69, which added another $200 million to an already high figure. The CCP monitor, William Mitchell, went through the roof. He took the Carter budget to Charles Schultz, director of the Bureau of the Budget, and convinced Schultz that cryptology had to be "investigated." Schultz, who had worked in ASA earlier in his life and probably thought he had special insight, sent an unstaffed memo to the president proposing a national-level cryptologic review.157

Richard Helms, the OCI, found out about this invasion of his turf, and he called White House staffer Bromley Smith. Walter Rostow and Clark Clifford put a stop to the Schultz memo, but this did not solve the cryptologic budget problem. Ultimately Robert McNamara, whose empire included NSA, convinced the president that Helms himself should be charged with the job. The DCI was to appoint a high-level committee to investigate cryptology. The objective was to reduce the CCP, and it was to be a review to end all reviews.i~ Helms appointed a very high-powered group. Lawyer Frederick Eaton was chair, and the members were General Lauris Norstad (former SACEUR), Ambassador Livingston Merchant, and Dr. Eugene Fubini, the DDR&Eand long-time nemesis of Marshall Carter. A more influential foursome could hardly have been found for thejob.i~9 The Eaton Committee suffered from the hostility of almost every organization with any stake in the problem. Helms himself had been cool to the idea when it was first proposed. Regarding NSA and SIGINT satellites, for instance, he stated that NSA's relationship with the NRO was a matter for him and McNamara to sort out, and it should. not be discussed by a committee. He opposedany investigation of Third Party matters as intruding onto CIA turf. He demanded that the committee not interfere with CIA's independent SIGINT effort: "Relations between NSA and CIA on covert SlGINT collection activities have been the subject of exhaustive discussion and review and present working arrangements appear to me to be satisfactory."16o Helms suggested that the committee occupy itself with considerations of ELINT management and reduction or consolidation of SIGINT field sites in vulnerable overseas areas. But DIA and the services opposed any look at EUNT, and NSA viewed the idea of reducing field sites with suspicion.'!' The appointment of Fubini to the committee was, to Carter, the last straw. He determined to have nothing to do with the effort, and his appointees to the committee staff (Walter Deeley and Gerald Burke) defended NSA interests at every turn. The investigative effort was so fragmented by staff bickering and external hostility that Eaton was able to accomplish little. It was hardly a review to end all reviews. 182

479

TOP SECRET tlMBRA

DOClD: 523682
TOP SECRE I UMBRA

REF lD:A523682

The conclusions of the Eaton Committee, especially in the area of COMINT, tended to support NSA objectives. Eaton was a centralizer, and he proposed that NSA obtain more' control over the cryptologic process. In his view, parts of the SCA staffs should be integrated with the director's staff. The committee recognised the central dilemma of resource control which was bedeviling SIGINT, and it viewed askance service attempts to flake off various parts of the process through inventive definitions of EW and increased control of cryptologic field sites .. Service complaints about lack of SlGINT support should not be used as a lever to fragment the eryptologic effort: "The tendency on the part of the military, unilaterally, to remove essential resources, both men and equipment, from the approved Consolidated Cryptologic Program is detrimental to the entire effort and should be resisted:'IBS Regarding ELINT, however, the panel proceeded in the opposite direction. Stating that "over the past ten years, it has become apparent that the decision to place ELINT as a whole within the COMlNT structure has not proved workable," the committee recommended that ELINT remain decentralized. NSA's proper role was to exert technical control, to collect and process signals of national strategic importance (like Anti-Ballistic Missile [ABM] radars), and to maintain a central database for the intelligence committee. On overseas basing, the committee simply repeated shopworn platitudes about the n~ed to reduce bases without hurting the effort. Eaton and company seemed to understand that overseas real estate must sometimes be retained in a less-than-productive status to preserve options against future targets. The Eaton members also felt that the SIGINT targets would increasingly become high-tech problems which required huge amounts of money, I land the overhead SIGINT satellite program. The committee cautioned against rushing in too fast, but recognized that increasing amounts of money would have to be funneled into those efforts at the expense of conventional collection. 1M On the critical issue of assessing the effort against I Ithe committee admitted that it had not been able to gather enough information to make a recommendation. There were telltale signs that NSA had decided not to unburden itself of its most closely guarded secrets to a group which it did not trust and that Eaton recognized a stone wall when he saw one.l6S The only Eaton recommendation that had any long-range impact on inteIligence was one which strayed beyond the borders of cryptology. The committee recommended that the DCI exert stronger direction over the overall intelligence program by creating a National Intelligence Resources Board (NIRB). This emphasis on centralized direction harmonized with the philosophical bent ofthe committee, and at CIA it fell on fertile ground. 1M

HANDLE VI

STEMSJOIN1iL L.EASABLE TO FORElGN NATlONALS

y

TOP $ECREI

"YRRA

480

DOClD: 523682

REF lD:A523682
-Tap SECRETUMBRA

The Eachus Committee Following the failure of the Eaton Committee to resolve the central problem of the worth of the effort against Soviet cipher systems, the NIRB prepared to take on the problem. But in the fall of 1968, before the NIRB could get moving, NSA itself established a panel for the I . I effort. The Eachus Committee was headed by Dr. Joseph Eachus ofMIT, a former Navy cryptanalyst during World War II and one of the leading civilian authorities on the Soviet cipher system problem. Eaehus was known to NSA and was a trusted friend. Carter placed his bets on a friendly assessment. In contrast to the Eaton fiasco, NSA revealed all to Eaehus. The .Eaehus report was the most thorough assessment of the NSA position on Soviet enciphered systems ever done.

Joseph J. Eachus

Eachus enumerated the systems that were defying attack - the prospects for many of them were dim. But he assessed prospects on other systems as good, as a result of a confluence of factors. I

HANDLE VIA TA

OL SYSTEMS JOINTLY

__

~HmgssiAABLE

TO FOREIGNNATI

481

TOP SlitRIiT UM8RA

DOClD: 523682 I E.O.
---------------------,
13526, section 1.4(c)
lOP SEOlEY &;IMBM

REFlD:A523682 I
Withheld from public release Pub. L. 86-36

Although Deputy Director Louis Tordella tried to justify the expense Eachus's .L-~----------------------------------------~--~ validate the effort and urge that it be-pursued with increased intensity. The Creation of NSOC Although the EC-121 shootdown pushed the NSOC cart over the crest of the hill, more than three years were to elapse before an organization actuallytook shape. NSOC's creation was delayed so long because of internal bureaucratic wrangling and logistics problems. The first problem was space. Initial planning assumed that NSOC would physically move into spaces contiguous to csoe, but it became clear fairly early that such a large organization would require its own spaces. Roomcould be made when the communications center (Tcom)moved to a new location on the third floor ofOps I, but NSOC would have to wait for Tcom to move out. The Second-floorspaces were to be available in 1971, but the calendar for the Tcom move kept slipping, and ultimately the area was not freed up until a year later. Meantime, the formation of NSOC was on hold. 169 The second problem revolved around what NSOC was to look like. In his initial NSOC concept paper, Major General John Morrison (the ADDO) described NSOC as a center that "would provide NSA with a single facility from which to conduct the production and dissemination of current SIGINT information .... " It would track ongoing events, but it would also produce reports and direct activities. It would comprise A Group's CSOC, Band G Group's crisis centers, elements of Kl associated with tasking mobile SIGlNT elements, P()4 elements involved in reconnaissance missions, and the Command Center. Shift operations would be headed by the SNOO (Senior NSA Operations Officer). Manning would come from csocs Dworkers,Dpeople from P04,Dfrom the Command Center, and unspecified numbers from B, G, and W Groups. Its communications would be primarily via Opscomms them, a huge number at the time). Morrison named Air Force colonel I ~to head the planning effort., I fresh from Europe, knew exactly how the operation at Zweibrucken functioned, and could get his hands on the people who had made it successful. 170 role was to

<r-bf

The operating concept that Morrison envisioned was basically csoe with other Agency elements grafted on. At the time esoe controlled European field site reporting. It could direct reporting and could issue its own reports (although as time wenton that function became almost the exclusive domain of the day shop). The day effort put out periodic summaries and wrap-ups, while events more than seventy-two hours old were turned over to A7, the term analysis shop. esoc still lived in the days of the Teletype Model 28 Opscomm terminal, and analysts got their traffic delivered in paper copy from the Opscomms that resided in a separate room. Even so, things moved very fast in esoe -

.

LEASABLE TO FOREIGN NATJO

Tep SECRETUMBRA

482

DOCID: 523682

REF ID:A523682
lQP SEERET l:I MlltA

it was closer to near-real-time than any other organization in the Agency. clearly modeled NSOC after CSOC.171

Morrison

And that was where the trouble began. CSOC might have been ahead of the competition, but it just wasn't the model that non-A Group organizations wanted to use. Morrison's concept paper raised a storm of controversy. Frank Raven, chief of G Group, agreed to place a desk in NSOC, but insisted that G Group operations were much too diverse to be amenable to centralization, and the G Group desk would be a watch desk only, with no production functions attached. lof B Group took basically the same tack, and he agreed to relocate certain B functions only to lessen the physical distance between B Group and other Agency elements. W Group agreed to establish a desk in the new organization, but its focus was still in DEFSMAC. and the NSOC effort was perfunctory. responding for xi, adamantly opposed absorption of any portion of the K 1 mission (managing mobile collectors) by NSOC.172

I

I

I,

Morrison forged ahead anyway. In 1972 he appointed a planning group dominated by people with A Group experience, and he named a full-time NSOC staff headed by Richard "Dick" Lord, the former head ofeSOC. Although key members ofB and G Groups assisted Lord, the organization kept the A Group flavor. NSOC was being called "A Group and the Dwarfs."17S The new NSOC edict was finally fashioned in the summer of 1972. By charter, NSOC was to "act as an authoritative .and responsive interface on current SIGINT product and . service both between SIGINT users and producers and between various producer organizations." It would also function as the NSA command center, and the senior officer, now called the SOO (Senior Operations Officer) would have true command responsibilities for the entire SIGINT system. In that capacity he or she represented the director .174 operationally, it resembled CSOC and its predecessor, the Air Force center at Zweibrucken. It monitored ongoing events and could take a variety of actions, including redirecting coverage and steering field reporting. Its original charter included the authority to do its own independent reporting, but this function was never exercised. NSOC did not become.another Zweibrucken, except in the area of reconnaissance reaction reporting. But it did become the focal point for the release of all Agency electrical product reports. Finally, it did the daily director's brief and supervised the worldwide CSG system.l75

483

TOP SECRET tJMBItA

DOClD: 523682
fOP S!(RET UMBRA

REF lD:A523682

Richard "Dick" Lord Named by Morrison to put NSOC together, he later became NSA's deputy director.

HANDLEVIAT

ROL SYSTEMSJOINTL Y T RELEASABLE TO FOREIGN NATION

TOP SECRET l:JMBRA

484

DoctD: 523682

REF ID:A523682
TO' S!(ReT UMBRA

Withheld from public release Pub. L. 86-36
'-...J

The NSOC that went operational in December 1972 (though the official ribbon-cutting did not occur till the following February) was in a state of technological transition. During the CSOC days, Walter Deeley, who had been Colonel ] ideputy in AS (CSOC), had been working toward what he called the "paperless environment." He planned to electrically connect the field Opscomms with a computer so that KLIEGLIGHTs could be processed and distributed automatically to CSOCfloor analysts. A revolutionary concept at the time, Deeley pushed it with a dedicated singlemindedness. A Group selected the Univac 494 as the mainframe because of its communications handling capabilities. Software to manage the KLIEGLIGHTsystem was called TIDE. The concept was in only a partial state of existence when NSOC was created, but it soon became the dominant concept within NSA. It made near-real-time truly feasible. 176
SIGINT

in the Nixon White House

The decade closed with a new president, Richard Nixon. It also opened with a new chief of the White House. Situation Room. Whenl lof CIA departed the Situation Room at the end of the Johnson administration, General Alexander Haig was appointed to the job. But Haig was clearly destined for greater things, and soon NSA's David McManis was given the job. 177 The national security apparatus under the new administration was enmeshed in a rather strange structure. Henry Kissinger, a Harvard history professor, became the national security advisor, but he came to exercise power far beyond that. Kissinger was in effect Nixon's secretary of state (shoving aside the supine William Rogers), a DCI (moving into the turf of Richard Helms, whom Nixon distrusted) and still later, a de facto chief of . staff for a president besieged by scandal and crime. Like Walt Rostow in the Johnson administration, Kissinger became the funnel for intelligence to the president. When someone had to be called in, McManis phoned Kissinger, who lived only a short distance from the White House in Rock Creek Park. He was, according to all contemporary accounts, a brilliant man, but not as experienced in SIGINT matters as Rostow had been. Moreover, he was inclined to shield the president from the details of intelligence, where Rostow shared all. Thus when did get to the Oval Office, it was generally subsumed into a mishmash of sources and not separated out and highlighted as it had been under Johnson. Nixon did not himself get involved in the details of intelligence, leaving those details to Kissinger.Y"

SIGIN!

:NDLE ~IA =::::=~: ~~~:;~~:~T:~;~u~=MSJOINT~
485
lOP $iCRET lIMBRA

I

DOCID

523682
TOP SECRET tJMBItA

REF ID:A523682

\

Henry Kissinger, May 1969,
in his office in the basement of the West Wing

TOP SiCRA' UM8RA

486

DOCID: 523682

REF ID:A523682
TO' SECItE'f I:fMBItA

To some extent been unique, on what being misused who had contact appears

this was an inevitable

development.

Johnson's

handling Hersh

ofSIGINT had have claimed, SIGlNT, was during
.

and it was not to be repeated. to have been good authority, for political purposes with the White House.

-Journalists like Seymour
that intelligence,

and especially

-. This has been confirmed

to some degree by SIGINTers
that was to emerge the

It

fell into a pattern

second Nixon term - the Watergate for the presidency. m

pattern.

It

was not good for SIGINT, and

it was

deadly

Notes

1. ArthurQQldschmidt, A Concise HiBtory the Middk Ea.t (Boulder: Westview Press, 1979).
2. (;Qldschmidt.

01

3. CCH Series VlII 16, The 1967 Arab-ISraeli War Crisis Files. •. -A Brief Review ofSIGINT and its Prospects," June 1963, in CCH Series VI.EE.l.ll; Series VIIU6. 5. NIE 30·67; Lyndon Baines Johnson Presidential Library, Austin, Teus, National Security Files, in CCH Series XVI. 6. SIGINT Readiness conditions proceeded from the lowest, Alpha (a state of increased watchfulness), through Bravo (a middle stage characterized by the substantial diversion of cryptologic resources and greatly increased reporting), to Charlie rus, involved in war - never invoked).
7. CCH Series VI.OO.l.4.

8. Ibid.; William D. Gerhard and Henry MiJlington, Atmck on a SlGINT Collector, TM USS Liberty, U.S. Cryptologic History, Special Series, Crisis Collection, (Ft. Meade: NSA, 1981).
9. CCH Series

vm.is,

10. CCH Series IX.16. 11. Chaim Henog, TM Arab-lsrceli War,: War andPeoce in the Middle East (New York: Random House, 1982).
12. LBJ Library, National Security Files. 13. Ibid.

14. Ibid.
15. Ibid.

I E.O.

13526, section 1.4(c)

16. Ibid.;j

\_-----~
17. CCH Series VIII 16. 18. LBJ Library, National Security Files. 19. CCH Series VI.C.1.27. 20. CCR Series VIII.16.

Withheld from public release Pub. L. 86-36

HANDLE viA ',,::::; _

!tSYWOT

E COMJl"!l' €aN''f'ItO~ 487

LEASABLE TO FOREIGN NAT u

:S:;TEMS JOINTL~
t8

_

WP 5ECftET UM8RA

DOCID: 523682
TO' SecRETUMBRA

REF ID:A523682
/

21. Gerhard and Millington; Howe, Techllkal ReseCJreA SAip,. 22. Gerhard and Millington. 23. Ibid. 24. Ibid. 26. Ibid. 26. CCH Series VIII.I6. 27. Gerhard and Millington. 28. CCH Series VIII.16. 29. Ibid. 30. Gerhard and Millington. 31. Johnson Library, National Security Files. 32. Gerhard and Millington, Hirsh Goodman, and Zeev Scbifl';"The Attack on the Liberty," September 1984, 7S-a4. 33. Ibid. 34. Johnson Library, National Security Files; Gerhard and Millington. 36. James M. Ennes, Jr., A"ault (In tM Liberty: tM True Story oftM l.raeli Attcc~ on an Americ~m lntelligenu Ship (New York: Random House, 1979); CCH Series VIII.16. 36. Gerhard and Millington. 37. Thomas P. Ziehm, TM Natw1UllSecurity Age~y and the EC·121 Shootdoum, NSA. U.S. Cryptologic HistOry, Special Series, Crisis Collection, V.3. (Ft. Meade: NSA, 1989). 38. Robert E. Newton, The Capture of the USS Pueblo and ill Effect Cryptologic History, Special Serie8, Vol. 7 (Ft. Meade: NSA,l992).
011

Tu AtlcnticMonthly,

SIGINT OperatwIII, United States

39. Lloyd M.Bucher, BucMr: My StOry, 1st ed (Garden City, N.J.: Doubleday,1970); Newton,Pueblo. 40. Newton.
\..

41. Ibid. 42. Ibid. 43. Ibid. 44. Ibid. 46. Ibid. 46. Ibid. 47. Ibid. 48. Ibid. 49. Ibid.

TOP SECRET UMBRA

488

DOClD: 523682

REF lD:A523682
TOP SECRET UMBRA

SO. Ibid. 51. Johnson Library, National Security Files; Newton. 52. Johnson Library, NSF.
53. Ibid..

54. Ibid.
55. Ibid.; Newton.

56. Newton. 57. Johnson Library, NSF.

58.

Ibid.

59. Newton.
60. Ibid.

61. Ibid.
62. Ibid. 63. Ibid. 64.. Ibic;t.

65. Ibid. 66. "CryptologicICryptographic Damage Aasessment, CCH Series Vm.18. 67. Newton. 68. Newton,158. 69. Ibid.
70. Ibid.

USSPueblo, AGER·2, 23 January-2~ December 1968," in

71. Howe. Technical Researcb: Ship •. 72. Tad Szulc, CzecMBlouakia since World WadI (New York: Viking Pre8B,1971).
73. Szulc; CCH, Series Vnr.17.

74. Szu1c; CCH Series Vlll.17, 75. Szulc, 313. 76. CCH Series VIII.17. 77. Series VIIl.17; Johnson Library, NSF. 78. Szulc; CCH Series VIII.17• 79. CCH Series VIII.17. SO. Ibid.

SABLE TO FOREIG

uNAbS_

489

10P SECRET UM8RA

--

.----------------------------------

DOClD: 523682
fOP SECRET UMBRA

REF lD:A523682

81. Szulc. 82. Ibid. 83. Ibid. 84. CCH Series VrII.17. 85. Ibid. 86. Ibid. 87. Ibid. 88. Ibid.

I E.O.
I ~

13526, section 1.4(d)

I

Withheld from public release Pub. L. 86-36

89. Oral interview withl ~ 17-18 Oct 1989. by Henry Schorreck, NSA OH 11-89; telephone interview with NSA. by Tom Johnson. October 1993; telephone interview withl NSA (PIS). by Tom Johnson. 12January 1993. 90.1

I.

Iinterview; McLean interview.

91. Johnson Library, National Security Files. 92. Interview with Walt Rostow. Austin. Texas, by Tom Johnson, 22 March 1993. 93. CCH Series VIll.17. 94. Ibid; "Strategic and TactiCal Warning and the Soviet Invasion ofCzechoslovaltia. 1968." MSSI thesis. July 1993. in CCH Series VIII.17. 95. CCH Series Vlll.17;c==:J. 96. Szulc. 97. CCH Series VIII.17. 98. CCH Series VIII.17; Mark Kramer. "Archival Research in. Moscow. Progress and Pitfalls •••·Cold War International History ProjeCt Bulletin. Fall 1993. . 99. CCH Series VIII.17. 100. Series VIII. 17;
of Secrets, 133.

1

I.

u.s. Nel//BGM

World Report article dated 2 September 1968;Szu.lc.Walter Lacquer. A World

10l.c=l. 102. CCH Series VIII.I7. 103. Johnson Library. National Security Files. 104. Ibid;c=J. 105. Johnson Library, National Sec~ity Files; CCH Series Vl.FF .1.9. 106. Johnson lJbrary, Nationa! Security Files. 107.c=:J. . 108. Ibid. 109. CCH series VIII.27.

I E.O.

13526, section 1.4(d)

Withheld from public release Pub. L. 86-36

TOP SECRET UMBRA

490

-'

_ .. -- --

----'------------

DOClD: 523682 I E.O.
13526, section 1.4(c)
110.c=J 111. [bid. 112. [bid.. 113. Ibid. 114. Ibid. 115.c:=J;CCH 116.c=J 117. Ibid. 118. Ibid. 119. Ibid. 120. McManis interview. 121. Books~ Series Vm.27.

REF lD:A523682
Withheld from public release Pub. L. 86-36
Tap SECRET I:JMBRA

II;

Seymour Hersh. TM Price of Power: Ku,inger .

iJJ

1M Nw,n WhiU House (New York: Summit

122.c=J. 123. Ibid. 124. Ibid. 125. Ibid. 126·0· 127. Ibid. 128. Ibid. 129. , "The National SIGINT Operations Center: Spectrum, Summer 1979, 4-15;c=:J ~The History ofthe NSA SIGIN! Command Center and it Predece880r8,1949·1969," NSA (P2217), March 1970, in CCH Series Vl.E.S.22. .

130·0· 131. NSAlCSSArchives, ACC 39292, GI8·0502·4; Kahn, TMCoclebn:akerll. 132. Ibid. 133. Ibid. 134. Ibid. 135. NSAlCSS Archives, ACC 27145, CBOI37. 136. Izoestia article in NSAlCSS Archives, ACC 34398, CBOI37. 137. Ibid. 138. Ibid. 139. Carter interview,Cburch Committee correspondence, in NSA retired reCords28794,80·079.

491

Tap SECRET I:JMBRA

-----------------------------------.---

.--

DOCID: 523682
TOP S!CRET tlMa~

REF ID:A523682

140. CCH Series VI.I.1.2. 141. CCH Series V£.I.I.1.2. 142. "Summary ofStatut.es Which Relate'Speciltcally to NSA and the Cryptologic Activities of the Government," undated manuscript in CCH collection. 143. CCH Series VIl. 1. 11.; VI.C.1.27. 144. Report of the Blue Ribbon Defense Panel (Fitzhugh Panel), 1 July 1970, in CCH Series VI.C.1.31. 145. See MJCS 606·67 in CCH Series VI.D.2.5. 146. See Hollis Board report in CCH Series VI.X.l.S. 147. Ibid. 148. Ibid. 149. 'Memo rue. "NSCID 6: Memoea and CorresPOndence SIGINT Sub-Panel. CIAiWbite House. JCS Policy Papars; DIRNSA Operationaltrechnical Authority: Concepts." Fiub.ugh Panel Report. 150. NSAlCSS Archives. ACC 31044. cssn 21. 151. NSAlCSS Archives. ACC 26457. CBOM22. 152. Cart.erinterview.

I E.O.

13526, section 1.4(c)

Withheld from public release Pub. L. 86-36
1990. by

153. Official DoDbiD, dated 1969. 154. ESC oral history interview withlL.. ~,available at AlA HQ. 155. Zaslow interview; CCH Series VI.D.2.7.
U56. CCH Series VI.C.l.27.

---'1. January

1'--

1andl'---_---'

157·1
158.

c=J.

111I,133.
Johnson Library, National Security Files.

159/ Johnson Library. NSF. 160. Ibid. ' 161.c=J.137-40. 162. Ibid. 163. Reportofthe Eaton Committee; 16 Aug, 1968. in CCH Series VI.C.1.24., 164. Ibid. 165. Ibid. 166. Ibid. 167.
1

168. Ibid.

TOp SECKEi UMBItA

492

DOCID: 523682
" -W-it-h-h-el-d-r-ro-m----, public release . Pub. L. 86-36
169.

REF ID:A523682
IU,. ~C,"nCI UMDKA

CcH Series

VI.C.8.

170. NSAlCSS Archives, ACC 18609, CBUH 48. 171.

Interview withl

1.30 Aug 1993. 18609,CBUH 28444.CBUJ 48. 27:

172. NSAlCSSArchives,ACC 173. NSAJCSSArchives,ACC

I

Iinterview.

174. NSAICSS Archives. ACC 28444, CBUJ 27. 175. Ibid. 116. NSAlCSS Archives. ACC 42165, H03·0407-1. 177. McManis interview. 178. Ibid.; Seymour Hersh. The Price o/Power. 179. Hersh; McMania

interview.

493

Tgp §ECAET YMBRA

DOClD

523682
T6PSECRETUMBAA

REF lD:A523682

Southeast Asia

~
.

"*==XHO)£rplll"'C~Y·
.

EASABLE TO FOREIGN NATION

.

TOP SECRET UMBRA

494

DOCID: 523682

REF ID:A523682
TOP SECRET UMBRA

Chapter 11 NSA in Vietnam: Building the Effort - The Early Years

Cochinchina

is burning, the French and Britiah are finiahed here, and we ought to clear out of

Southeast Asia.

Lt Col Peter A. Dewey (OSS) writing from Saigon, 1945

Much has been said about the American decision to become involved in Southeast Asia. The decision to "intervene was hotly debated and controversial from the first. Intervention res~lted ultimately in the nation's most humiliating aiilitary debacle (although by no means its first defeat). So many things went wrong that the failures obscured the successes, but successes there were. 'From both the military and the cryptologic standpoint, it was a learning experience.

VIETNAM-THE

COUNTRY

Actually, three countries were involved: Laos, Cambodia. and Vietnam. (Vietnam's political geography is complex, involving as it does .three separate areas: Cochinchina (presently known as Cochin China) in the south, Annam in the center, and Tonkin in the north.) But Laos was landlocked and primitive - it hardly counted - and Cambodia was little more than a "Sideshow to War" (to use British writer William Shawcross's phrase). Vietnam became the main show, the country where American lives and national prestige were put on the line. Vietnam (meaning, literally, "South Viet") had been settled by a Sino-Tibetan group called the Viet, who had been pushed by Mongolian population pressures farther and farther south. They finally wound up in the Red River valley. a broad and fertile plain suitable for wet rice cultivation. As they migrated ever farther south, however, they were hemmed in by mountains, which cascaded, like boiling water, into the South China Sea. The Viets picked their way along the coast, inhabiting isolated valleys, until they finally arrived at the broad Mekong delta. There were no mountains on the delta, and they quickly converted it to rice-growing. As a result, Vietnam became long and thin in the center, averaging no more than fifty miles wide along the Central Highlands, with two large plains attached to each end. It has been compared in shape to a pole across the back of a farmer, with a basket of rice on each end. Vietnam was a meeting place of disparate cultures - primarily Indian and Chinese. The Vietnamese warred fiercely with the armies of their neighbors, and they acquired a reputation for recalcitrance and military prowess. Chinese sovereignty over the region, strong during the Han dynasty (about a century before Christ), was reduced over time to a

HAND ELEASABLE TO FOREIGN NATIO

TEMSJOINTL

Y

495

lOP SECRET UMBRA

DOCID: 523682
TOP SECReT tJMBRA

REF ID:A523682

more or less nominal one. This was the situation when the French arrived in the midnineteenth century. France established a tenuous hold on the country - solid in Coehinchina, less sure in Annam, very loose in Tonkin. The French overwhelmed the Vietnamese with technology but had little chance to stay permanently. After all, the Chinese, who lived next door, had never completely subjugated the restive Vietnamese. French efforts were, in the long run, doomed by distance and the stubbornness of the Vietnamese.' French colonial rule came to an effective end during World War II. The Japanese retained a French colonial government, but it was only a puppet, and in 1945, faced with defeat, the Japanese extinguished even this shred of French dignity. The Japanese defeat left Vietnam without a government. What emerged was a government of sorts, effective only in the Red River Valley to the north, under a communist named Ho Chi Minh. The remnants of the Japanese war machine transferred formal power to Ho's organization, the Vietminh, on 18 August. On 2 September Ho declared the independence of Vietnam. The United States, mostly through OSS, maintained distant contact with the Vietminh during the war. The opportunistic Ho, apparently hoping for substantial American aid, even adopted phraseology from the American Declaration of Independence when he declared Vietnam a sovereign country.

Ho Chi Minh in Paris. 1946

~ANDLE

VIA \;~~:

I8!:YHOI E C0¥I"IT cerffn;~;:!TEMSJOI~ LEASABLE TOFOREIGNNA LS

TOP SeCReT UMBRA

496

"DOCID: 523682

REF ID:A523682
TOP S!eRET UMBRA

Occupied with larger matters, Allied leaders were not exactly consumed with worry over Vietnam. Roosevelt believed that colonial rule was finished everywhere, and that included Southeast Asia. But what to do with the former French properties was a more difficult question. He toyed with the idea of giving it back to the French under a trusteeship arrangement with independence guaranteed at a future date. He also offered it to Chiang Kai-shek, who did not want it. (He had enough trouble at home.) FDR died without resolving the issue, and Harry Truman had it on his plate. At the State Department, a stealthy battle was going on between the Asianists, who were promoting independence for all Asian countries, and the Europeanists, who did not want a dispute over the colonies to jeopardize postwar relations with Britain and" France. The Europeanists won, and the United States informed France in May 1945 that the U.S. recognized French claims to Indochina. It was decided that British forces would occupy the south of Vietnam, while Chinese forces under Chiang would occupy the north, until France could get some forces together to reoccupy its former colonies. French troops eventually regained a tenuous hold over much of Vietnam, especially the southern portion. Meanwhile, negotiations continued with Ho, who, it will be remembered,had already proclaimed independence and had effectively occupied much of the north. But negotiations broke down in 1946, and outright warfare began, This period of conflict culminated in the French defeat at Dien Bien Phu in"l954. " Having successfully ejected this latest occupying power from Vietnam, all that remained for the Vietnamese was to formalize a separation. Divorce court was held in Geneva. It resulted in an independent and neutralist Cambodia and Laos and in a Vietnam divided at the waist. The part north of the 17th parallel, effectively controlled by the Communist forces under Ho, would become the Democratic Republic of Vietnam, while the portion below the 17th parallel would"establish its own government. At some point the two would theoretically meet to hold elections of national reconciliation and reunite into a single nation. I The United States had by this time become deeply involved in Vietnam's troubles. American aid to the French mounted each year, and by the fall of Dien Bien Phu it came to about 80 percent of French expenditures for the conflict. There were behind-the-scenes talks of American air strikes to bolster the French position at the base, but at the last minute Eisenhower decided not to go ahead. At the peace conference, the Americans, frightened of communist encroachment, did everything they could to hem in Ho's government.

The Americans Enter the Fray
Once the war was over, the United States effectively assumed responsibility for the mess. When Ngo Dinh Diem, the new president in the south, refused to go ahead with elections for fear oflosing them, he had full American support. By early 1956 the U.S. had

HANDLE VIA TALENT ELEASABLE

SJOINTLY TO FOREIGN NATIONA

497

TOP SeeRET UMBRA

---------

-- ".-

-

OOClO: 523682
1'OP SECRET UMBRA

REF lO:A523682

E.O. 13526, section 1.4(c)(d)

Withheld from public release Pub. L. 86-36

assumed responsibility for arming and training Diem's army. According to historian George Herring,
The United
StatAla

inherited from France an army of more than 250,000 men, poorly organiuci,
2

trained; and equipped, lacking in national spirit, suffering from low morale, and deficient in officers and trained specialiata ...

A military assistance group in Saigon steadily expanded in surreptitious ways beyond the Geneva-imposed limit of 342 people, until it reached almost 700.

\

Then in 1960 CIA informed NSA that they had inked a Third Party srcixr agreement with the government of Vietnam. The Diem government maintained three fixed sites and was willing to trade raw intercept for cash and equipment./

E.O. 13526, section 1.4(c)(d)

Withheld from public release Pub. L. 86-36
HANDLE

VIA I:::;mn,ECQWNTCOWaQb8Y::;MSJOlNTLY
SABLE TO FOREIGN NATION

TOP SI:CRET UMBRA

498

DOCID: 523682

REFID:A523682

Withheld from public release Pub. L. 86-36
YeP SECRET UMBRA

I E.O.
Laos and the Beginnings of Direct American In volvement

13526, section 1.4(c)

When Kennedy arrived in the White House', Laos, rather than Vietnam, seemed like the crisis to watch. The 1954 Geneva settlement had initiated a period of tenuous teetering between pro-Western and pro-communist sympathies, with a neutralist group holding the balance of power. Eisenhower had tried to keep a pro-American party in power through lavish subsidies, but in 1960 a series of coups pushed the government first toward the East, then the West. Eisenhower a ministration succeeded in convincing demanded a favorable outcome," Wanting to appear firm, Kennedy had 500 Marines airlifted to the Thai side of the Mekong, which formed the border with Laos, while the carrier Midway moved into the Gulf'of'Siam." , But the Bay of Pigs fiasco brought Kennedy up short. If American military power could not secure a favorable outcome 90 miles from its shores, what might happen in an obscure, landlocked Asian nation more than 12,000 miles from Washington? The Pentagon estimated that at least 300,000 troops would be needed to maintain the proWestern government. So in late April Kennedy opted for a negotiated settlement and agreed to U.S. participation in yet another Geneva conference." A precarious coalition government emerged from the Geneva talks, but none of the three major factions was happy, and within a ·year the cease-fire was violated by the Pathet Lao. Once again Kennedy mounted a show of force , dispatching 5,000 Marines and infantrymen and two air squadrons to Thailand. Again a coalition government was formed, but its long-term chances for success were not bright, 10

The 1954 Geneva accords made it extraordinarily difficult to operate in South Vietnam. The Military Advisory Assistance Group (MAAG) staff was already bloated and· obviously in violation of the agreements, Thailand was the obvious choice. But the L~--~~~--~~~~--~~~~~ Thai, with a long tradition of independence (alone in Southeast Asia, they had never been a European colony), were skittish, and negotiations dragged oninconc1usively for years. Then the Laotian crisis served to pry open a crack in the door to Thailand,

I

I

I E.O.

13526, section

1.4(c)
HANDLE VIA TALENT KEYHOLE CO MINT CONT~L NOT RELEASABLE SYSTEMS JOINTLY

Withheld from public release Pub. L. 86-36

TO FOREIGN NATIONALS

499

top SECReT t:JMBRA

DOCID: 523682
1'611 SECKET UMBRA

REF ID:A523682

Withheld from public release Pub. L. 86-36
E.O. 13526, section 1.4(c)(d)

~
.

~In1965ASA

began building a major intercept site at Udorn, a Thai town in the far north, near the Mekong River. Called Ramasun Station, it became the location for an FLR-9 antenna, and at the height of the Vietnam War, it housed over 1,000 ASA and AFSS cryptologtsts." Hanoi Decides to Intervene in the South In 1954 Hanoi had decided to work on the infrastructure in the north and to put off' . attempted unification to a later date. But by 1959 the leadership decided that it must expand in the south or else its southern cadres would wither and die. In the spring of 1959, the leadership authorized resumption of armed struggle in the south, a decision that was ratified by a Party meeting in September 1960. At approximately the same time, Hanoi created a new group, MR 559 (so-named because it was created in May 1959), within the General Directorate of Rear Services (GDRS); to control infiltration into the south. Beginning with only 500 people, it eventually expanded into a network of 40,000-50,000 military and civilian workers. It was organized into sixteen units called Binh Trams, battalion-size units in geographical areas, each controlling the infiltration network through its region. This evolved into the Ho Chi Minh Trail, which provided the wherewithal for revolution and invasion. 1~

.

I

E.O. 13526, section 1.4(c)(d)

Withheld from public release ~ub. L. 86-36

TOP SECltfT UMBRA

500

a.maa..n "",d

Station.

196t. Conslstift6

oftcnle and aa y.t GOJ"LR..I.

vanl,

tb.re

wert f.w pen.paae. build\np

~M"~='.'~Y
~1'UFO~".81:'~

c
o

501

~TTCO~'~5if[eeftIllE:fT'ttl"'~"'B>RR>Ocf'<-

H

o c

E.O. 13526, section 1.4(c)(d)

DOClD: 523682
TOP SECRET l:JMBR-A

REF ID:A523682

Withheld from public release Pub. L. 86-36

NSA Expands Cryptologl.c Involvement The nascent Kennedy administration adopted an initially cautious line toward Vietnam. The U.S. government had troops in the South, but they were still called "advisors," and the numbers were limited. At the time, the only SIGINT involvement was the very limited CIA relationship with the South Vietnamese SIGINT service. There were no American cryptologists in the country]

I

But as the number of American "advisors" expanded, so did the cryptologic presence. In early 1961 the chief of the MAAG in Saigon advised Maxwell Taylor (chairman of the JCS) during one of his trips through Saigon that the ARVN (Army of the Republic of Vietnam) had no SIGINT capability. This touched off a debate back in the United States about the advisability of expanding in Vietnam.l" At NSA, Admiral Frost directed a complete evaluation ofSlGINT in Southeast Asia, and from that came a new plan to expand the cryptologic presence. Essentially, two plans were written. The flr'St was called SABERTOOTH, and it involved noncodeword assistance to the SIGINT services ofl Vietnam. The second, called WHITEBIRCH, would involve the esta6bshment of a mobile ASA intercept unit with Morse, voice. and HFDF positions. NSA was skeptical of the voice positions because ASA had few qualified Vietnamese linguists, but the Agency approved the plan despite the reservations."

I

I

I

I

I

The new NSA plan also envisioned a beefed-up collection posture. In addition to expanding the cryptologic presence in Bangkok, ASA would introduce people directly into Vietnam for the first time. The burden of field processing would fall most heavily on the sites in the Philippines. It also called for an "Evaluation Center" in Saigon to integrate SIGINT with other intelligence for the chief of the MAAG. When General Paul Harkins showed up in February 1962 to become the first COMUSMACV (Commander, U.S. Military Assistance Command Vietnam), this became the Current Intelligence and Planning Branch, J2, and was housed in the MACV building, originally located in downtown Saigon. 20 Before Harkins arrived, NSA interests had been served by a TDY arrangement. In April 1962, however, the first permanent NSA representative.l was on board. His arrival was accompanied by vigorous protests by the Army. Secretary of the Army Zuckert sent a scorching letter to Assistant Secretary of Defense John Rubel protesting the assignment. "This action," he said, "would result in removing these SIGINT . resources from the control of military commanders in the area. . .. Generally, responsiveness to intelligence requirements of CINCPAC and COMUSMACVwould be dependent

I,

E.O. 13526 section 1.4( c)(d)
,

I

HANDLE VIA TALENT KEYHOLE COMINT CONTROL SYSTEMS JOINTL Y NOT RELEASABLE TO FOREIGN NATIONALS

Withheld from public release .Pub. L. 86-36

TOP SECRET UMBRA

502

DOCID: 523682

REF ID:A523682
Tap seCRET UMSRA

upon the decisions of a national level agency, far removed from their areas of ; .. n. He proposed that all SIGINTassets in the area be placed under the responsibility. operational control ofMACV. It was the opening shot of a war within a war, the struggle to control SIGINTassets in Southeast Asia. 21 I-E-.-O-.-1-3-S-2-6-, s-e-c-ti-o-n-l-.4-(-d-)In April 1961 USIB took two giant steps toward direct involvement. One was a decision to a rove the new ended Third Part relationship with Vietnam, to be run by ASA. inasmuch as it was at the noncodeword level and involved a limited amount 0 L~~~--~--~~~~ targets. It went beyond the arrangement already operated by the CIA and involved AS~ advisors to train the Vietnamese. In return, USIB expected Vietnamese approval to establish American intercept facilities. 22 The second step was to approve an Army COMINTunit in Vietnam in support of counterinsurgency planning. The National Security Council then required that the results obtained by that unit be shared with the South Vietnamese to the extent needed to launch rapid attacks on the Viet Congo 23 Withheld from public release Pub. L. 86-36

The Buildup of Cryptologic Assets
The first ASA troops began arriving in May 1961. They were under cover, wore civilian clothes, and were prohibited from carrying military identification cards. They found spaces in an RVNAF hangar on Tan Son Nhut Air Base and lived downtown at the Majestic Hotel. Working areas were set up inside the hangar by piling boxes of C-rations seven feet high to make rooms. A few of the officers had desks, but the analysts worked at tables constructed of plywood and scrap lumber. Since there were few chairs, the tables were hoisted four feet off the ground so analysts could stand. Needless to say, there was no air conditioning, and the troops sweltered in the tropical heat. U The unit was called the 3rd Radio Research Unit (3rd RRU). Operationally it was called USM-9J, subordinate to USM-9 in the Philippines. The original processing mission consisted mainly of traffic collected by the SouthVietnamese SIGINTService, which was at the time composed of only about 100 officers and men. They had two collection sites, at Saigon and Da Nang, and soon established a third site at Can Tho in the Delta. They were operating with equipment left over by the French or provided by CIA. Among the asseta that they had inherited from the French were three DF stations and all the equipment, which happened to be of World War II vintage. In 1961 CIA gave them six AN/PR,D-1 mobile HFDF sets. When Srd RRU began processing, the main input was the DF bearings from the South Vietnamese. 2~ Meanwhile, ASA advisors conducted classes in DF, traffic analysis, and intercept for the Vietnamese under the SABERTOOTH program. They were supposed to hold the classification to noncodeword, but the line between SIGIN'l'and non-SIGINT was very shaky, and it was crossed regularly."

503

TOP SECREn,MBRA

DOCID: 523682
TOP SECRET tJMBItA

REF ID:A523682

I

E.O. 13526, section 1.4(c)(d)

Withheld from public release Pub. L. 86-36

The focus of the operation, though, was OF. ASA set up an HFOF net, called WHITEBIRCH. Because of availability, the ANtrRD-4 was the equipment of choice. Three. sets were mounted in vans and positioned at Nha Trang, Can Tho, and Bien Hoa, with control in Saigon. The Third RRU was also receiving bearings from an ASA site in Ubon, Thailand,l Isites in Vietnam, andi ARVN operated its own three stations at Pleiku, Da Nang, and Ban Me Thuot, and the results were supposed to provide direct support to the South Vietnamese Army. 27

I

The WHITE BIRCH net was a failure. It had the lowest fix rate iJ; the Pacific, and it was constantly short of manpower. This dismal state of affairs was due primarily to the circumstances surrounding its mission. In the dense and humid tropical jungles, the ground wave faded to imperceptibility in only a few miles. The sky wave came down at such a steep angle that the existing OF equipment (the ancient TRD-4s) could not cope with it. Moreover, the skip zone between ground and sky waves was almost ninety miles, meaning that most of the ASA sites were located in a skip zone. When inadequate maintenance and unreliable communications were added to the woes of WHlTEBIRCH, it was clear that the system would not do the job.:18 Frustrated, ASA turned to the mobile PRD-Is now owned by the ARVN. These were effective, but only if the DF set was within five to flfteen miles of its target. To be that close to a VC transmitter was often a dangerous proposition, but they tried it anyway. On 31 December 1961, they found out how dangerous it was. An ARVN DF operation returning to Saigon from the DF site at Ha Tien (on the southern coast) was ambushed by' VC. Nine ARVN soldiers were killed, along with Sp4 James T. Davis, the ASA advisor. Davis was later called by President Johnson the first American soldier to die in Vietnam. The 3rd RRU compound was named Davis Station, thus adding to the immortality of the unfortunate Davis.29 ASA had come to a full stop on the DF problem, and until they solved it, the amount of direct SIGINT assistance that they could provide to the ARVN forces was limited. The next group of SIGINTersto arrive in Vietnam were the Marines, who sent a training detachment from Fleet Marine Force in Hawaii. They originally set up next to the ARVN SIGINT operation in Pleiku, and as such were completely cut off from direct contact with other U.S. SIGINT units. This proved unsatisfactory, even for training. so In 1962, the cryptologic community decided to move its main base of operation to Phu Bai. A large ASA site was constructed, and it became the center of SIGINT operations for Vietnam. The Marines decided to move in with ASA, but the Air Force Security Service was more standoffish. Da Nang was the center of air operations, arid AFSS located its principal site there to support the Seventh Air Force.

STEMS JOINTLY
LEASABLE TO FOREIGN NATI

TOP SECRET UMBRA

504

lQP SEERU ~M'RX

.

!:n.&i'anc. ~ \lie WlUTEB1RCH

eo.apound

N
(X)

10 I"l N LO Q H

HANDLE viA lAtENt KEiHOUCOM1NI CONIROtSiSIEMSJOiNiLi NOt kEL£4:A:8t;e1'SP8RtiIGlfN'TIQt;I,' 'i 505 TeP

g
Q

SEeK!' UMBRX

DOClD: 523682
TOP SECRET YM8R.A.

REF ID:A523682

Withheld from public release Pub. L. 86-36
;

E.O. 13526, section 1.4(c)(d) At the time, the Vietnamese problem was entirely manual Morse. Rumors of VC voice swirled about, and in February 196~ ~ntercepted some voice traffic emanating from a low-level net in Vietnam. ASA tried but, right up until the Tonkin Gulf incident of 1964, had not intercepted any.31

L

I

As cryptolOglc resources expanded, the question of operational control occupied increasing attention both in Saigon and in Washington. The Army continued to insist that MACV should control all c:ryptologicresources in theater. During Admiral Frost's tenure as DIRNSA, a compromise of sorts was worked out. When the first ASA resources arrived in country, Admiral Frost delegated operational control to ASAand recognized the further delegation of control to the commander of the MAAG (later MACV). This gave MACV a handhold but kept the strings ultimately tied to DIRNSA.32 In 1963 General Wheeler (chairman of the Joint Chiefs) negotiated directly with General Blake. They arrived at a new compromise whereby NSA would continue to control major, fixed sites like Phu Bai, while operational control of ASA's direct support units (OSUs) would be delegated to ASA, andthence to the supported Army commander. This was actually more restrictive than the original decision, and it was made more onerous by the edict that when MACV wanted additional units under its control it would have to submit the request through the lengthy and cumbersome chain of command which ran through Hawaii." DF Goes Airborne The ambush of Davis and the ARVN DF team in December 1961 brought about a scramble for a better system. The safest thing would be to put. the mobile DF sets on airplanes. This technique had been tried as early as World War I. and the French had employed ARDF aircraft in their struggle with the Vietminh, with good results. But the technical barriers were serious. The problem was in the interference of ground and sky waves. Aircraft were up high enough to receive both, and the accuracy of the bearing was degraded because, while the on-board system tried to read the direction of the signal from the ground wave, the aircraft itself acted as a huge antenna for the sky wave, which arrived from a different direction." An ASA engineer, Herbert S. Hovey Jr., went to work on the problem and was joined by a team from the Army's laboratory at Ft. Monmouth, New Jersey. Knowing of the French ARDF effort but not knowing what technique they used. Hovey experimented with different techniques and various aircraft. He tried rotary-wing options. but found that the rotor blades created too much turbulence. Hovey finally settled on the U-6A, a singleengine fixed-wing aircraft widely available in Vietnam. Instead of using the almost universal (in DF arrangements) loop antenna, he used antennas fixed on the leading

TO •• SECRET I;JMBRA

506

DOClD: 523682

REF lD:A523682
TOP SECRET l:IMBItA

Herbert S. Hovey (second from rigbi) and an early U-6AROF -configured aircraft

A 3rd RRU ANIPRD.l sbort·range OF set

"":"" "'" T"
_

ENT

KEYHOLECOMI", CONTRO~~"","""
__ ALS

'OIN rL~
__

NOTRELEASU~bE'f6EOait6!UUTI

507

TOP iliCIlET

IIMIRA

DOClD: 523682
lOP 5ECRET IIMBRA

REF ID:A523682

Entrance to Davis Station, Saigon

edge of each wing, about forty feet apart, with the receiver in the center. This turned the aircraft itself into a large HF antenna. The aircraft had to be pointed directly at the signal, thus creating an aural null on the pilot's gyrocompass. To create the aural null, the pilot fishtailed the aircraft back and forth, going into and out of the maximum signal strength. He would then fly at the signal from three different angles, the three lines of bearing thus constituting a fix. This peculiar flying technique solved the problem. S6 ASA sent the first DF-equipped U-6 to Vietnam in March 1962. It was an instant success. In May 1962 the ARVN successfully struck a VC unit based on ARDF fixes.sa In December of that year, when an ASA ARDF fix located a VG radio transmitter in the northern Delta, American advisors under General Harkins used the intelligence to plan an assault on what they thought would be a communist unit of no more than 120 men. The ARVN 7th Infantry Division was employed in the action and swooped into the. area by helicopter early on the morning of 3 January 1963. Instead, they ran into a unit of more than three times that many, which stood and fought. The resulting battle of Ap Bac was a . turning point in the war for both the VC (which found that it could confront and defeat a main ARVN force) and for the Americans, who concluded that they would have to become more directly involved. The battle was initiated based on an ARDF fix.37

HANDLE VIA TALE NOT RELEASABLE

T CONTROL SYSTEMS JOINTLY TO FOREIGN NA

fOP SECRET UMBRA

508

DOClD: 523682

REF lD:A523682
TOP seCRET UMBRA

The value of ARDF was quickly recognized. It became the 'most important advance in the employment of SIGINT for, tactical applications in the war and the principal targetting tool for MACV. NSA boxed up this valuable technique within its own sphere of control by declaring that the ARDF aircraft were simply outstations of the WHITEBIRCH net, which was already a CCP resource. ARDF was to become the battleground on which the JCS and NSA fought for ultimate control of SIGINT in Southeast Asia. It was easily the most divisive issue in the entire intelligence community.38

INTO THE MIRE
The troops will march in; the bands will play; the crowds will cheer; ... and in four days everyone will have forgotten. Then we will be told we have t.o send in more troops. It'alike taking a drink.

The etrect wears

off, and yeu. have to take another. John F. Kennedy, 1961

While all this was going on, the Kennedy administration was assessing its chances in Southeast Asia. The first thing Kennedy did was to gather informatidn, using the timehonored technique of a fact-finding team. In the spring of 1961 he sent Walt Rostow and his personal military advisor, Maxwell Taylor, toSalgon. They came back very pessimistic. The Diem regime was crumbling and would require a large infusion 'of American troops and material. They recommended that some 8,000 American "advisors" be sent to Vietnam under the cloak of providing "flood relief." Averell Harriman, the long, time advisor to Democratic presidents, and Chester Bowles, a senior diplomat, both doubted that the corrupt and repressive Diem regime could be adequately shored up, and he urged Kennedy to call a new Geneva conference and negotiate a settlement. But Kennedy had just emerged from the disgraceful Bay of Pigs incident and was in no mood to be percei ved by ei ther the Soviets or the American public as a "negotiator. ml9 But he also rejected the Taylor-Rostow proposal compromised, increasing the size of the aid mission but make a big difference. All the while he was disturbed by of the Diem regime. To have a happy ending in Vietnam, more reasonable and competent government."
1 the

as transparent. Instead; he failing to increase it enough to the narrowness and inflexibility it would be necessary to obtain a

I E.O.
___________

13526, section 1.4(c) Withheld from public release Pub. L. 86-36

Crisis in Third Party Relations

509

fOP SECReT l:JMBRA

DOClD: 523682
l'6' SECMT tlM11tA

REF lD:A523682 I E.O.

Withheld from public release Pub. L. 86-36
13526, section 1.4(c)

USIB decided to back away from SIGINT collaboration with the ARVN, and USM-626 (the former ~SM.9J in Saigon) was instructed to stop providing certain technical data. At the same time, NSA made plans to move most SIGINToperationsto Phu Bai and to make it aU .S.-only site.4S The USIB decision, prompted by NSA, created an uproar in the field. Harkins protested and was backed up by Huntington Sheldon, the CIA official who watched over SIGINT for the intelligence community. Moreover, General Khanh, the RVNAF chief of staff. refused to authorize a solely American operation at Phu Bai, thus holding the superSIGINT site at Phu Bai hostage to a continued close 5lmNTrelationship. In the end, Khanh, Harkins, and Sheldon won. Admiral Frost issued a revised and liberalized interpretation of the USIB edict, and the Americans exited the controversy with as much grace as possible.r" The Diem Coup Riven by internal dissent, the Diem regime was tottering by 1963. The regime was controlled by Diem and his corrupt family, and no reform appeared possible. The last straw was a Buddhist revolt against the strongly Catholic Diem regime. The uprising began in May 1963 and became marked by self-immolations by Buddhist monks. When confronted by such opposition; no regime could last." Even Diem knew it and began exploring a negotiated settlement with the north. To the Kennedy administration, this looked like a way out. The JCS prepared a plan for a phased military withdrawal beginning later in 1963. The first 1,000 troops were actually withdrawn before the plan came to a halt.4e But negotiations were never begun. In early November the generals in Saigon rose against Diem, with the knowledge, ifnot the active connivance, of the American embassy .. Diem and his brother Nhu were captured and, in a twist which was not in the original script, executed, apparently on the orders of General "Big" Minh. Minh took over the government, beginning a series of revolving door regimes, each weaker and less popular than the previous one. The JCS withdrawal plans were shelved. Later in the month Kennedy was dead, and a new president had to look again at the morass in Vietnam.'7

TOP SECRET tJMBRA

510

DOCID: 523682

REF'ID:A523682
TOP SECRET UMBRA

The Cryptologie Expansion of 1984 With withdrawal plans on hold, the new:DIRNSA, General BI8ke, directed a relook at the American eryptologic posture in Southeast Asia. Blake decided to accept Phu Bai as the super-site for Vietnam, with major resource additions there and at other sites in the Philippines and Vietnam. Collection from Thailand would also increase, and Udorn was selected as the Thailand super-site. In early summer, with Maxwell Taylor (the new ambassador in Saigon) lining up behind it, Blake took the plan to Fubini. They agreed that most resources for the new effort would be transferred from existing SIGINT problems (primarily Soviet), except for some assets already targetted on Southeast Asia that would be moved to the mainland from Okinawa and Japan. 48
SIGINT resources would also be needed for a major new operation, under the general rubric ofOPLAN (Operation Plan) 34A. This was a JCS plan to support South Vietnamese infiltration and unconventional warfare operations. The SIGINT support for OPLAN 34A, Withheld from called KIT KAT, would come mainly from vans flown inl land the Philippines and public release located at Phil Bai. A new SIGINT Support Group in Saigon would provide MACV with Pub. L. 86-36 direct support to OPLAN 34A.~~ E.O. 13526, section 1.4(c)(d)

I

I

Communications still represented a sore point. SIGINT exited Vietnam through an Army communications center in Saigon that was known for its cramped quarters and ancient equipment. Worse, it was an HF shot to the Philippines; and in the heavy tropical atmosphere HF was even less reliable than usual. Reliability ranged from 30-75 percent, an unacceptable figure. 50 DCA came up with a solution. A submarine cable was installed between Nha Trang and the Philippines, and by the mid-60s all cryptologic communications were being routed through the cable (dubbed "Wetwash"). Circuit reliability leaped upward." This development had a major impact on SIGINT operations in Vietnam. The submarine cable could take higher circuit speeds, and it was possible to ship much more SIGINT back and forth. This led to the feasibility of sending encrypted traffic back to a central processing center - at first in the Pacific (Clark AB and JSPC in Okinawa) and later all the way back to NSA. It changed the way SIGINT was done in the theater, but it . also increased the suspicion of tactical commanders who preferred to rely on their own people from ASA rather than on some unseen computer far away.
AFSS Comes to Vietnam

The Air Force Security Service did not actually start its Southeast Asian operations in Vietnam. Like the Army, it arrived in Thailand in early 1961 to provide SIGINT support for the Laotian crisis. I RC~47 ROSE BOWL I I

I

I
I E.O.
13526, section 1.4(c)

I
511
TGP SECRET YMBAA

Withheld from public release Pub. L. 86-36

DOClD: 523682
1=OP SECRET I;IMBItA I-E-.O-.1-3-S2-6-,-se-c-tio-n-l-.4-(-c)-1

REF lD:A523682
Withheld from public release Pub. L. 86-36

I

The precessing was done on the ground in spaces occupied by the tiny AFSS intercept unit that had been there since the.summer of 1960. Spaces were so cramped that at one point a Russian linguist wound up transcribing his intercepted tapes in the shower room. But like the Laotian crisis itself, the SIGINT support operation lost steam, and by spring NSA had cancelled the deployment. 52

I

Vietnam was a ground war, and the U.S. Air Force did not get involved in a big way until 1964. The Air Force did, however, set up a tactical air control system beginning in January 1962. The unit was located atop Monkey Mountain near Da Nang, which would give American radars the longest possible reach. Along with the Air Force contingent of 350 people came an AFSS CCU (COMINT Contingency Unit), consisting of two H-l vans airlifted from Clark Air Base and a mobile AFSSO, also in a van. A smaller intercept and SSO effort was located at Tan Son Nhut, but the hearability was bad, and the intercept unit was soon relocated to Da Nang. The next year AFSS reorganized its Southeast Asian assets, designating Tan Son Nhut as the headquarters, with subordinates at Da Nang, Bangkok, and Ubon.ZIl

Monkey Mountain

~NbLE

v fA

:;;~"'~IT

KEYHOLE COMI~J!I'

T RELEASABLETOFOREI~Al:.S

e:::'~~!!.STEMS

JOINT~

.

'F6P SECRET I:JMBRA

512

"-r--.'-

DOClD: 523682

REF lD:A523682
TOP SECRET UMBRA

Da Nang remained the only AFSS unit of any size in the war zone. By 1964 USAFSS had two Da N ang sites, one atop Monkey Mountain and one at the air base below. SecUrity Service successfully resisted an NSA master plan to move the unit to Phu Bai, arguing that hearability was better at Da Nang and that they should be closer to the supported commander. U I E.O. 13526, section 1.4(c) In March of the same year, the ACRP returned to Southeast Asia. It arrived on the Withheld from heels of reports that PRC-North Vietnamese military relations were becoming closer. public release
Pub. L. 86-36
L..•••.• NSA initiated ACRP collection to follow this activity, and a new program, called QUEEN BEE CHARLIE, based at Yokota AFB in Japan, began fiying missions out of Don Muang. In July a follow-on operation, QUEEN BEE DELTA, consisting of two RC-130s, began flying missions over Thailand every other day. Initially processing was done at Don Muang, but plans were being drawn up to transfer the entire effort to Da Nang. That same year, the Navy began flying EC-121 and EA3B collectors in the Gulf of Tonkin. ss

Air Force ARDF trailed ASA into Southeast Asia. In 1962 AFSS tried out HFDF programs using two different platforms, a B-26 and a C-47. The ARDF effort had the strong personal support of General LeMay, then the Air Force chief of staff. From the beginning, however, the program was engulfed in controversy. The first problem was control. The Air Force wanted the ARDF program to be purely tactical, unattached to NSA, operating in a noncodeword environment. NSA, however, insisted that it come under the direct control of USM-626, as outstations of the WHITEBIRCH net. The program was thus placed under double ignominy - within the cryptologic system and under the thumb of the Army. Moreover, the Air Force insistence that it be noncodeword resulted in non-Slindoctrinated people being assigned to It, USM-626 was at first prohibited from passing technical data to support the AFSS effort. This was soon straightened out, and all the Air I Force people were S1 cleared, but it was a bad start for a program. Finally, the system did not work. It used larger aircraft but did not do well against low-power signals. The Air Force Security Service left the theater to do more research" . The next year AFSS was back, this time with a second ARDF system produced under a Navy contract and installed on an Air Force plane under Project HAWKEYE. It was more sophistieated than the Army system, using computers and larger, more capable aircraft. But. it, too, did not work, and at the end of the year ASA continued to have the only effective ARDF system in Southeast Asia." The small AFSS effort in Vietnam betokened the lack of an air war. They were not engaged in war - they were just waiting in case an air war happened. They hadn't long to wait.

HANDLE VIA
_ .

=;t:~{frlI1gbi

COMINT

QenTKO~;~:MSJOINTx:Y

EASABLETOFOREIGNNATJ

513

lOP SiCRET (;IMBRA

Tap "SiAE'

WM8A·

Tbl, C:ODlpl ••••.•.•

tJSA~2 Dperauoo. van•• 1986. Ioc.ted on 11MplaiD. at Uae fOGlotMClGker

Mount41:ln.

o

H

Top SElAE'

UMBRA

514

8

DOClD: 523682

REF lD:A523682
TOP SECRET UMBRA

THE CRISIS IN THE GULF
Well, I am the guy who rose from the ashea, and twenty years later telling you I saw it, and there were no boats. Adm. James B. Stockdale, Navy pilot, concerning the 4 August attack

In the many years of conflict in Vietnam, no single incident stands out as more controversial than the 4 August 1964 incident in the Gulf of.Tonkin. In it, two American destroyers patrolling in international waters were supposedly shot at by North Vietnamese gunboats. In retaliation, an angry president launched the first air raids on the North, and a few days later Congress passed the Tonkin Gulf Resolution, giving Lyndon Johnson a free hand to deal with North Vietnam in whatever manner he felt best suited the situation. For America, it was the beginning of an apparently irrevocable descent into the maelstrom.

The Desoto Patrols

I E.O.

13526, section1.4(c)

I

The attack on the destroyers originated with the Desoto patrols. These were begun in I 1962 as patrolling operations along the Chinese coast. There were three objeeti ves: intelligence collection, realistic training, and assertion of freedom of the seas. Naval Security Group detachments on board pursued the collection ofC] ELlNT and naval CO MINT. However, to naval authorities the mission of freedom of the seas clearly stood first, and training second; intelligence was the thirdpriority. By December, the patrols had been extended to the coasts of Korea and North Vietnam. ~ The rationale was to support special operations under OPLAN 34A.. OPLAN 34A stemmed from CIA covert operations which had been going on since the early 1960s under various names. Most of these involved the nighttime coastal insertion of ARVN commando forces, whose mission was sabotage. By early 1964 the Army had taken over most of the operations, under OPLAN 34A. The Desoto patrols were extended to North Vietnam primarily to provide SIGINT support to the commando raids.59 In addition to NSG afloat detachments on board Desoto craft, the Army was tasked with SIGINT support from positions at Phu Bai. 80 The operations got off to a very bumpy start in February 1964, but they eventually smoothed out. Although there was considerable behind-the-curtains controversy about their effectiveness, the raids were having at least harassment value by July 1964. The tiny North Vietnamese navy was beginning to pay them close attention.61 North Vietnam could mount only a modest defensive threat. Their first-line combatants were twenty-four Swatow motor gunboats acquired from the Chinese over a

HANDLE VIATA EASABLE TO FOREIGN NATIO

MSJOINTLY

515

lOP SECRET UMBRA

DOCID: 523682
1'6' SECREf UMBRA

REF 10 :A5·23682

period of years. More threatening, however, were twelve Soviet-built motor torpedo boats delivered to Haiphong in late 1961, capable offuty-two-knot speeds. These, in addition to a few minesweepers, subchaser and district patrol craft, represented the North Vietnamese navy.e2 The 2 Auguat Afadclor Patrol The increasing harassment value of OPLAN 34A was certain to make the North Vietnamese more belligerent. On 1 August NSA went on record as warning the Navy that their own Desoto patrols might be in danger of attack.as A day earlier., the destroyer Maddo" had begun a patrol in the Gulf of Tonkin. 64 On 2 August the North Vietnamese decided to attack the Maddox. During the morning hours, two SIGrNT units, a Navy intercept unit in the Philippines (USN -27) and a Marine detachment collocated with ASA at Phu Bai (USN-414T), reported that North Vietnam's naval headquarters had directed preparations ror attack. This series of reports was flashed to Captain Herrick, the task force commander on board the Maddox, as the morning wore on. The information was sufficiently unsettling that Herrick questioned the 65 day's patrol, considering it to be an "unacceptable risk .•• Just after noon, USN.27 intercepted a message from one of the coastal control authority at Port Wallut to one of the Swatows: "Use high speed to go together with the enemy following to launch torpedoes." USN-27 issued a Critic on this inflammatory declaration, and Herrick had it in hand almost an hour before the attack was launched. It was preceded and followed by other North Vietnamese messages leaving no doubt that they were headed for a major engagement. It could, of course, have referred to the 34A operations that had been going on earlier, but Herrick knew nothing of those operations. He had to assume that the North Vietnamese meant him - and he was right.66 . At about 1600 local, three PT boats launched .a high-speed attack on the Maddox. Herrick replied with surface fire, and within half an hour the torpedo boats withdrew. About that time air cover showed up, commanded by Admiral (then Commander) Stockdale from the carrier Ticonderoga. Stockdale's crew shot up the fleeing torpedo .boats, sinking one and putting another out of action. 67 Meanwhile, the two SIGINT stations continued to monitor North Vietnamese communications, keeping Herrick informed of what was happening on the other side. The patrol made for the mouth of the Gulf and withdrew. Back at Fort Meade, NSA declared a SIGINT Readiness Bravo.sa \ There was no doubt of the attack. Not only was it launched in broad daylight, but it .was preceded and followed by communications (intercepted by the Navy and Marines)

TOP SECREf UMBRA

516

DOClD: 523682

REF lD:A523682
1'6P SECRETUMBRA

,,,.

Gulf of Tonkin

".

Track orthe Maddo$, 31 July-2 August 1964

Captain . John J. Herrick. commander of Destroyer Div 192, ./ with Captain Herbert L. Ofi,er, commander of the Maddox

517

lOP SECRET UMBRA

DOCID:· 523682
TO•• SECRET ilM8AA

REF ID:A523682

niaking the entire attack procedure and objectives crystal clear. warning, and Herrick came to rely on it almost implicitly.

SIGINT

gave impeccable

The Johnson administration chose not to reply militarily to the attack. But at the White House the mood was grim, and there was' a feeling that they could not let another such attack pass unnoticed. . The 4 August Patrol After assessing the 2 August attack, the administration decided to keep the Maddo% in the Gulf at least through the 7th to assert ~reedom of the seas and to add a second destroyer, the Turner Joy, which had been part of the Ticonderoga task force. With two vessels, Herrick headed back to the Gulf on the 3rd.69 After spending the day near the coast of North Vietnam, Herrick withdrew both. vessels to the central Gulf of Tonkin for the night. Through intercepts lof Vietnamese radar transmissions, he knew that he was being silently shadowed by at least one North Vietnamese PT boat. Moreover, this tended to be confirmed by reporting from San Miguel that one of the Swatows involved in the previous day's activity (T-142)had been ordered by a naval authority to "shadow closely." During the night a 34A task force shelled a radar station and a security post. fleeing to Da Nang at daylight.To Herrick believed his vessels were in imminent danger, but the next morning he was nonetheless ordered back to the area of the previous two days' patrol. The Maddo:r. and Turner Joy loitered in the general area where the 2 August attack had taken place. At . about 1700 they turned b~ck toward,the central Gulf to spend the night.n .. At about the same time that Herrick was ordering his two-vessel task force back to the central Gulf, the Marine detachment at Phu Bai issued a Critic on an intercepted message from Haiphong ordering three of the boats involved in the 2 August attack to make ready for military operations that night. To Herrick this was very ominous, since he had been shadowed by a North Vietnamese vessel or vessels the night before. Based on this and follow-up messages from Phu Bai, he sent a message stating that he believed that the Vietnamese were preparing to attack, '12 At 2041, the Maddo:r. appeared to pick up radar contacts on North Vietnamese PT boats. For the next four hours, the MaddcJ:r. and Turner Joy zigzagged through the central Gulf, apparently pursued and attacked by unknown and unseen vessels. The crews of the two vessels claimed to have had radar and sonar contacts, torpedo wakes, ghn 'flashes. and searchlights, and fired repeatedly at whatever seemed to be attaeking them. When air cover showed up from the Ticonderoga task force (led by Stockdale), the pilots could not see any boats, but it was an unusually murky night with very low overcast and poor visibility.'3

TOP SECRET UMBRA

518

lOP SECRETbMIPiA

The American

deatro1el'

Tu,rlWrJIq

HANDLE VIA TALENT NOTRE

KEYHOLE

COMI

LSYSI'EMSJOINTLY

o

s

519

Ter SE£RiT

I

'''BRA

o Q

DOClD: 523682
. TOP SECRET l:JMBRA

REF lD:A523682

After the engagement, San Miguel reported that T-142 claimed to h8.veshot down two "enemy planes" and that "We sacrificed two comrades but are brave and recognize our obligations.,,74 Back in Washington, the events in the Gulf grabbed everyone's attention. The initial indication that something was afoot was the Critic and follow-up from Phu Bai. These were called over to DIA from NSA just after 8 A.M. By 0900 copies of the reports were distributed to McNamara and Wheeler, and McNamara called the president at 0912. This kicked off a long train of actions that spanned the entire day. 75 Thus forewarned, the president had no trouble believing that an attack had actually taken place once he received the first news at 1100. McNamara convened a meeting to discuss possible retaliation. At a lunch with Rusk, McNamara, Vance, McGeorge Bundy, and John McCone,Johnson authorized an aerial strike on North Vietnam~se targets. But soon thereafter, the White House was looking at a message from Herrick casting doubts about the attack. Adverse weather conditions and "overeager sonarmen" may have accounted for many of the alleged contacts. Based on this, Admiral. Sharp in Hawaii (CINCPAC) phoned McNamara to recommend that the air strike be delayed until they received more definitive information. At that time a retaliatory air strike, scheduled for 0700 Vietnam time, was only three hours away.1G . Soon after, Sharp received the new information about the supposed shooting down of enemy a4'craf't and the sacrifice of two vessels. Sharp, Admiral Moorer (CNO), and Johnson all became convinced that an attack had taken place, and Johnson authorized Pierce Arrow (the bombing attack on North Vietnam) to proceed. It was delayed almost three hours, though,and came very close to preceding Johnson's televised address to the nation announcing the Gulf incident and the American response. 77 The sequence of events at the White House was driven largely bySIGINT.The reliance on SIGINT even went to the extent of overruling the commander on the scene. It was obvious to the president and his advisors that there really had been an attack - they had the North Vietnamese messages to prove it. . But to the analysts working the problem at NSA, things did not appear to be so obvious. The preplanning messages could, after all, have been referring to reactions to the Desoto patrols. Or the entire series of messages might have been old traffic referring to the attack on the 2nd. NSA sent out frantic requests to the units involved (Phu Bai and San Miguel) to forward their raw traffic. NSA also requested verification from SIGINT intercept operators on the Maddox and Turner Joy. The-ships' operators had nothingtheir intercept capability (all VHF voice) was completely blocked by the ships' radios during the period of the incident. As for the mainland intercept, it took hours to obtain, and the fIrst NSA follow-up was issued without the benefit of the messages intercepted in the field. 78

TOP SECRET UMILRA

520

DOCID:

523.682

REF ID:A523682

20' Thanh Hoe. ....-,0700H

108'

109

0

20/1

~ I: •.. ~ ;;

Gulf of
Tonkin
-HI"

.
t
'" !::
~


~C
Ho·n Matt_

19'

t > c ••

( ~ ~

I

,

18'

18~-

i
0500H 3 Au!!
1080 10Q·'

III

3:

~

OOClO: 523682
lOP SElkE I UMBRA

REF IO:A523682

The first NSA report indicated that the vessels supposedly planning for operations on the night of the 4th apparently did not participate in the events regarding the Maddox and Turner Joy. A subsequent wrap-up on 6 August homed in on the 2 August attack (easy to substantiate). conveniently avoiding the direct issue regarding the 4 August , incident."
,

The NSA analyst who looked at the traffic believed that the whole thing was a mistake. The messages almost certainly referred to other activity - the 2 August attack and the Desoto patrols. The White House had started a war on the basis of unconfirmed (and later-to-be-determined probably invalid) information. so There had been no dlssemblingin the White House. The messages looked valid, and Lyndon Johnson had come to be a believer inSIGINT. When he ordered the attacks, he was sure he was right. He wasn't. and it was not until NSA analysts laboriously pieced together the SIGINTinformation over a period of days that it became obvious how big a mistake had been made. The Johnson administration defended its actions in public for years, but the reality eventually sank in. Even the president was heard to say in later years, "Hell, those dumb stupid sailors werejust shooting at flying fish.'>l!l Some months previously, William Bundy (deputy secretary of defense) concluded that Johnson would need some sort of congressional endorsement for the expanding American role in Vietnam. He felt that a declaration of war was too blunt an instrument, and its chances in Congress were slim. What was needed, he believed, was a joint resolution, similar to that which Congress had given to.Eisenhower during the Quemoy and Matsu crisis in 1955. Bundy drafted a resolution that gave the president the right to commit forces to the defense of any nation in Southeast Asia menaced by communism. 82 The resolution was ready by June 1964, a'nd the Pentagon hedalready identified some ninety-four targets in North Vietnam, in case the president should direct military retaliation. Everything was ready but was put on hold. Some sort of provocation would be needed. The Tonkin Gulf crisis was just such a provocation. The administration hustled -the resolution through Congress with only two dissenting votes. It was shepherded through the Senate by the chairman of the Foreign Relations Committee, William Fulbright.83 The Tonkin Gulf Resolution did not become a political issue until three years had passed. In July 1967, with antiwar passions heating, a reporter for the Arkansas Gazette quoted a former radarman on the Maddox as saying that North Vietnamese vessels had not been in the Gulf that night and that he believed his radar contacts had actually been reflections of the Turner Joy. This article came to Fulbright's attention. This appeared to wipe out the rationale for the resolution, and Fulbright, who was being gradually converted to the antiwar cause, felt that he had been hoodwinked, perhaps deliberately, by the White House in 1964. He began gathering the relevant material, including SIGINT reports obtained from the Department of Defense. When he felt he had enough, he convened a hearing on the Gulf crisis.M

-

I

fOP SElRE I UMBRA

522

DOClD:

523682

REF lD:A52'3682
'fOp SIiCRIiT UMBRA

The hearings, held in February 1968, made the Gulf of Tonkin Resolution infamous and converted it into a weapon in the hands of the antiwar activists. During the proceedings, Fulbright managed to cast considerable doubt that the 4 August attack ever took place. Inconclusive radar and sonar hits, mysterious weather conditions, the lack of a single verifiable ship sighting - all were used to beat down the Johnson administration's contention that the retaliatory action and the resolution itself were justified. But the central contention of the hearings became the SIGINT. When Fulbright brought McNamara to the stand, the secretary of defense kept referring to "intelligence reports of a highly classified and unimpeachable nature .... " He meant, of course, the SIGINT reports that, first, indicated that the Swatows should prepare for nighttime operations, and, second, contained the after-action reports alleging that aircraft were shot down and the loss of the two boats. The committee kept pressing McNamara and eventually dragged out of him virtually the full texts of the messages involved. McNamara resisted, but it was very hard to defend his actions without resorting again and again to his most convincing pieces of evidence. 65 These public disclosures damaged the SIGINT source - all the messages had been from decrypted North Vietnamese naval codeswhich were still in use in 1968. But it did not sell the case to the disbelieving committee, despite McNamara's contention that "No one within the Department of Defense has reviewed all of this information without arriving at the unqualified conclusion that a determined attack was made on the Maddox and Turner Joy in the Tonkin Gulf on the night of 4 August 1964.n88 In fact, not all DoD people were sold on this contention. NSA, for one, had failed to fully support the administration's position. It had confirmed the ~ August attack but had never confirmed the 4 August engagement. The Agency had concluded that the two Swatows instructed to make ready for action that night had never participated in the action with the Maddox and Turner Joy. The after-action reports could have referred to the 2 August engagement. But it didn't really matter. The administration had decided that expansion of American involvement Vietnam would be necessary. Had the 4 August incident not occurred, something else would have. Another expansion of the war occurred the following February, following the mortaring of an American installation at Pleiku. McGeorge Bundy said at the time, "Pleikus are like street cars. If you miss one, another will come along," He could have been talking about the Gulf of Tonkin crisis.

in

--""'OT

T CONTROL SYSTEMSJOINTLY RELEASABLETO FOREIGN

523

TeP !!CRE i UMBRA

I

DOClD: 523682
TOP SECRET UMBRA

REF ID:A523682
Withheld from public release Pub. L. 86-36

Notes

1. Histories of Vietnam and Southeast Asia abound. Three of the better ones, which were used in compiling this history, are Joseph Buttinger, Vietnam: A Political Hi.tory ~New York: Frederick A. Praeger, 1968); Stanley Karnow, Vietllam: A Hi.tory (New York: Ptlnguin Books, 1983); and George Herring, AmericlJ', Longut War: TM UniledStIlw, and Vietnal'll, 1960-1976 (Philadelphia: Temple University Preaa,1986). 2. Herring, 58. 3. Historical Study of U.S. Cryptologic Activities in Southeast Asia, 1960-1963, 1964, CCH Gerhard collection; Hiatory Series-Southeast Asia (Ft. Meade: NSA,1969). 4. CCH SeriesVl.HH.12.10.
5. VI.H.H.12.10.

William D. Gerhard, In tM SlIadowofWar,Cryptologic

6. Ibid.; oralinterview with
~86.

I
----------

" 23 Janua~

1986, by Robert D. Farley and Tom Johnson, NSA OH

7. Herring, Kennedy Library papers. 8. Be&ehlosB, The Criai. Years. 9. Herring, Be6ChIOlill. 10. Beschlosa. 11. Gerhard Collection. 12. William D. Gerhard,ln tM Shadow of War. 13. NSA Retired Recorda, 28515, 84-245; Gerhard,In the SNuiow of War; oral interview with " 23 December 1992, by Charles Baker and Tom Johnson, NSA OH 08-92. '-----------

I

f'

14. NSA Retired Records, 28515, 84-245; Thomas N. Thompson, Jesse E. Miller, and William D. Gerhard, SlGINT Appl~tiolll in U.8. Air OperatiDlII, Cryptologic History Series, Southeaat Asia, Part I: Collecting the Enemy's Signals (Ft. Meade: NSA,1972). 15. CCH Series VI.H.H.23.2-23.S. 16. Ibid. 17.Ibid.; V1.H.H.12.10;1 ~ftNSAin Vietnam: Proud BndBitter Memories," CryptJJtog, Oct. 1975, 3-6. c=Jinterview: Gilrhard Collection; X•• ellti41 Maltera: A HilltJJryof the CryptJJ8rap~ BrollCh oftA. People'. Army of Viet-Nam, 1945:-1975, United States Cryptologic History, Speeial Series, Number 5, translated by David W. Gaddy (Ft. Meade: NSA, 1994),93-111. 18. Herring; MSgt William E. Fleischauer, "History of Project 78," in ASA annual history, 1959, available at Hq INSCOM. 19. CCH Series VI.H.H.12.10. 20. Gerhard Collection. 21. Ibid.; CCH Series V1.H.H.6.22. 22. CCH Series VI.HH.6.22.; V1.HH.12.10;Gerbard Collection; Gerhard,ln the Sh.adow of War.

Withheld from public release "pub. L. 86-36
TOP SECRET I:JMIIR.6,

EASABLE TO FOREIGN NA u

524

·...

--.

-_

..

_---

DOCID: 523682

REF ID:A523682
tOP SECRETUMBRA

23. Ibid. 24. CCH Series Vl.H.H.lS.9. 25.1 93.

1, "Deployment of the First ASA Unit to Vietnam," Cryptolog~ quarWrly, FalllWinter 1991,77-

26. CCHSeriesVI.HH.l.10. 27. Gerhard Collection. 28. Gerhard,In~ Shadow of War.

Withheld from public release Pub. L. 86-36'

29. CCH Series VI.HH.16.12: Gerhard Collection. 30. James Gilbert, "The Beginnings ofAirborne Radio Direction·Finding" in U.S. Army Intelligence and Security Command.,HiI!torical Monographs, 1983 Series Vol. 1. 31. Gerhard Collection; CCH Series VI.HH.12.10. 32. Gerhard, In ~ Shadow of War, CCH SeriesVI.HH:12.l0. 33. CCH Series VI.HH.1S.1. 34. [bid. 35. Interview with David Gaddy, February 1993; Jules Roy, La Bat4ilk de Dum Bien Piau. (Paris: Rene Julliard, 1963);Gilbert. 36. Gilbert; Gerhard,ln tM Shadow of War. 37. CCHSeriesVI.HH.12.10. 38. Neil Sbeehan,A Bright Shining Lie: John Paul VaM and America in Vietoom (New York: Random House, 1988),201 If. 39. CCH Series VI.HH.12.10; Gerhard Collection. 40. Herring. 41. VI.HH.12.19;
{

D interview; Gen. Pharo Van Nhon, Manuscript history ofDGTS, available

in CCH.

42. Nhon, Manuscript history ofDGTS; 43. CCH Series VI.HH.12.l0.

0

interview.

«. Ibid.
45. Gerhard colleetion;CCH Seriea VI. HH.12.10. 46. Kamow. 47. Herring. 48. Kamow. Herring. 49. Gerhard,l n tM Shadow of War; Brown; "The CCP"; Gerhard collection. . 50. Gerbard,ln the Shadow of War. 5l. CCH Series VI.HH.l2.10.

Withheld from public release Pub. L. 86-36

525

fOP SECRET UMBRA

-

DOCID: 523682
UlP S&CREl' YMBRA

REF ID:A523682

52. Morrison interview. 53. CCH Series VI.HH.1.40. 54. Gerhard, In tIw SNuIoIII o(War; Gerhard Collection. 55. Thompson, et al, SIGINT Applic4tioM. 56. Gerhard,lntIwSMdDlllofWar. 57. Gerhard Collection. 58. Ibid. 59. Edward J. Maroldll and Oscar P.Fitzgerald, Tm United SlGleB Nauy and elw Vietnam Conflict, Vol. II: From to Combct,1959-1966 (Wuhington: Naval Historical Center. 1986).

Milit4ry AsaiBlG_

.60. Ibid. 61. Gerhard. I 11. ~ SNuiow of War, Gerhard Collection; Marolda. 62. Ibid. 63. Ibid. 64. CCH Series VI.HH.24.10. 65. Marolda. 66. Marolda;CCH Series VIII.I3. 67.. CCH Series CIll. 13. 6S. Marolda.
69. Marolda; CCH Series Vl.HH.24.10.

70. Marolda. 71. Ibid. 72. Ibid. 73. Marol~; CCfiSeries VI.HH.24.10.; VIII.Ill. 74. Marolda. 75. CCH Series VIII.13. 76. Ibid. 77. Marolda. 78. Marolda; CCH Series VIII.IS. 79. CCH Series VIII.13; oral interview Withl Johnson, NSA OH 33-87. '--------SO. CCH Series VIII.13.

1.22 Dee 1987, by Robert Farley and Tom

81. c=Jinterview. 82. Karnow, 374.

Withheld from public release Pub, L. 86-36

TOP SECRET tJMBRA

526

DOCID: 523682

REF ID:A523682
1=OPSECRET !:IMInA

83. Kamow. 84. Ibid. 85. David Wise, "Remember the Maddox!", E.qui,.e, April 1968; NSAoral history with Lt Oen Gordon Blake. by William Gerhard and Renee Jones, 5 June 1972. unnumbered. 86. CCH Series VIl1:13, containS a full text oftha hearings. 87. U.s. Congress, Senate, Committee on Foreign Relations, The Gulf of Tonkin, the 1964 Incidents, Hearings, 90th Cong, 2nd BeSS., with Hon. Robert S. McNamara, Secretary of Defense, on 20 Feb.1968, in CCH Series VIII.13.

HA~E

yiA

==

KFYHOI$ cmmf,!,:'~~SYSTEMSJOINTLY LEASABLETO FOREI . UAT:.8

527

+OR SECRET UMBRA

DOCID: 523682

REF ID:A523682
I

TOP seCReT tJMBItA

Chapter 12 From Tonkin to Tet - The Heart of the War
THE PRESIDENT EXPANDS THE WAR

Retaliation during the Gulf of Tonkin crisis was a one-shot affair, but it indicated that the administration was edging toward more active involvement.· It did not, of course, dissuade the North Vietnamese. In November the Viet Cong (VC) mortared the air base at Bien Hoa, only two days before the U.S. elections. Johnson regarded this as a bald affront. Then, on Christmas Eve, they bombed an American officers' billet in downtown Saigon in broad daylight, killing two and wounding sixty-three. This further hardened American attitudes and made direct intervention the following year more likely.' . Late in 1964, SIGINT began noting a strange communications pattern for the North Vietnamese 325th Infantry Division. The division headquarters at Dong Hoi opened communications with entities that controlled the infiltration routes into South Vietnam. Sometime thereafter, SIGINT(together with ARDF fixes) showed the 325th moving south, first into Quang Tri Province (just below the DMZ) and later all the way to the Central Highlands. It was the r11"st move of a regular NVA division into the South, and it pointed to a new and considerably more dangerous phase of the war. No longer were the ARVN facing an insurgent Viet Cong movement - they were up against North Vietnamese regulars." The 325th was in South Vietnam to prepare for the rainy season offensive, and it would create a bloody hell for the unlucky ARVN units in its path. The president now knew what the American people did not - that North Vietnamese regulars were in the South.· All that remained was for another provocation to take place. He had not long to wait. On 6 February 1965, the Viet Cong rocketed the American and . South Vietnamese facilities at Pleiku, killing 8 Americans and wounding 108, bringing newspaper headlines and extensive television coverage. At the time, press coverage had the effect of pushing the administration into retaliation. (A few years later it would have the opposite effect.) Twelve hours later American A-4 Skyhawks and F-8 Crusaders were launched from the 7th Fleet against Dong Hoi (whence the 325th and other units had staged on their way south). Twenty-one days later President Johnson institutionalized the pattern of isolated retaliation by starting daily bombings of the North and the Ho Chi Minh Trail in Laos. The operation, called Rolling Thunder, was planned to last eight weeks, but in April Earl Wheeler, JCS chairman, informed the president that it had had no effect at. all on the North. SoJohnson directed that it continue until it had an effect," The attack on Pleiku almost shouted out the vulnerability of American troops and equipment. With the initiation of Rolling Thunder, U.S. aircraft were at Da Nangalmost

HANDLE VIATAL::;::::eu _ N9

COWN! CO)J'!l'R6U::MSJOIN'I'LY BLETOFOREIGNNATIO

529

TOP S!C:ft!T tlMBRA

DOCID: 523682 -,
TOP SECRET UMBRA

REF ID:A523682

constantly, and they required protection. The U.S. commander, General William C. Westmoreland, asked Johnson for a defensive force, aDd the president obliged. On 8 March . the first MarinessplalJhed ashore at Da Nang, beginning the American deployment of ground combat troops to the theater." The commitment of ground forces, once begun, became an inexorable upward spiral. In May, Westmoreland asked for a total of 185,000 by the end of the year, and 100,000 in 1ge6. Johnson sent Secretary of Defense McNamara to Saigon to find out what was happening. The secretary returned with a gloomy assessment - Westmoreland was actually understating the need, and the U.S, wouldneed an additional 200,000 in 1966.~ Operation Starlight and the Ia Drang Campaign

Withheld from public release Pub. L. 86-36

SIGINT was still small-time in Vietnam, but it was growing. In August 1965, with new American troops swarming ashore almost every day. ASA SIGINT and ARDF located a new enemy communications terminal near the Marine base at Chu Lai. In Saigon. the NSA representative, I I.took the item to Brigadier General Joseph A. McChristian, the J2, who passed it to Lieutenant General Lewis Walt, who commanded the Marines in Vietnam. Walt discussed it directly with his SIGINT people at Phu Bai and became convinced ofits validity. He began planning a major entrapping operation. The VC forces, who had hoped to surprise the Marines, became themselves surprised and overcome in the operation, called Starlight. Starlight was a .turning point in the direct employment of SIGINT and ARDF in operational planning. 6

Ia Orang, the first significant campaign by a large force of NVA regulars, began as an attempt by the NVA 325th Division to cut Vietnam in half in the Central Highlands. In the process, the 325th attacked a Special Forces camp at Plei Me, about twenty-five miles south of Pleiku. ARVN forces attempted to rescue the troops trapped inside but were ambushed by two NVA regiments of the 325th, the 32nd, and 33rd, with heavy casualties." Following the engagement, the NVA retreated up the Ia DrangValley, with the First Cavalry (Airmobile) in pursuit. Owing to the recent success in Starlight. the American forces had five ARDF aircraft in support. Moreover, for the first time the ARDF crew had the capability to pass rlXeSdirectly to the ASA Direct Support Unit (DSU) supporting the ground forces. ARDF fixes followed the 325th elements retreating up the valley until they were cornered at the Chu Pong Massif. The 1st Cavalry, employing helicopters in pursuit for the first time, and supported by B-52 air strikes, devastated the NVA. The two. regiments suffered up to 60 percent casualties and were no longer an effective fighting force. The remnants retreated into Cambodia. During the action, the 33rd was so concerned about the Americans appearing to know their location that they concluded that they had spies in their ranks,"

TOP SECAEn.MBRA

530

DOCIO: 523682

REF ID:A523682
lG~ SECRETUMBRA

The SIGINT Deployment
To support American ground forces, ASA built Phu Bai into the largest ASA field site in the world, almost 100 positions. Together with the 3rd RRU in Saigon and the 9th in the Philippines, ASAhad substantial fixed site assets," The fixed sites were augmented by SIGINT tactical assets. ASA tactical units began to arrive with each incoming Army organization. Each unit normally had five manual Morse positions along with short-range DF and VHF intercept equipment." NSA's concept of direct support was that, since the problem was centrally controlled from Hanoi, the SICINT effort should remain centralized. NSA continued to exercise overall control from Fort Meade. In Vietnam. collection management authority (CMA) was divided into three areas, roughly corresponding to the division of American forces. USM-626 at Tan Son Nhut was CMAfor the southern part of the country, USM-808 at Phu Bai for the northern portion, and USM-604 at Pleiku for the central area.'! . .: Following its relocation to Phu Bai, the Marine SIGINT detachment became the DSU in support of the III MAF (Marine Amphibious Force) in the north. Eventually the Marines established DSUs like the Army and wound up with the same sort of a decentralized SIGINT support arrangement, with small detachments composed of only a few positions each collocating with combat units. Lacking their own ARDF assets, the Marines received ASA ARDF support." . Air Force Security Service SIGINT collection from Vietnam itself was more limited. The unit at Da Nang expanded quickly once Rolling Thunder began, but it never equalled the huge ASA contingent. This was not true, however, of the ACRP effort. USAFSS had a contingent of four RC-130s at Da Nang, which expanded to six in 1967, by the device of raiding airborne assets in Europe and Japan. IS Beginning in 1967, a new ACRP program began flying in Southeast Asia. This program consisted of the far larger and more capable RC-135s belonging to a new unit at Kadcna AFB, Okinawa, the 6990th SS (Security Squadron). With SAC front-end crews and USAFSS collectors, the RC-135s flew very long (often in excess of seventeen-hour) missions into the Gulf of Tonkin. The RC-130s continued to fly out of Da Nang until the end of the year, when the Kadena unit took over the entire mission. a Operational control arrangements continued to cause friction. NSA opposed fragmentation, while the Army insisted that field commanders should directly control all cryptologic assets supporting them. This became a critical issue when Army units began independent operations. . In mid-1965 a new arrangement was hammered out between Rear Admiral Schulz of NSA and Brigadier General Eddy, deputy commander of ASA. Under this Schulz-Eddy agreement, when DSUs were in active support of an ongoing tactical operation the field

531

1'6P SEER!T UMBRA

DOCID: 523682
lOP SECRET I:IMBItA

REF ID:A523682

commander would control them. When they were back in garrison, control would revert to ABA's designated field site (either Saigon, Pleiku, or Phu Bai). NSA continued to control all rlXed field sites, to the loud disapproval ofMACV.ls The second control issue to arise in 1965 concerned the air problem. Brigadier General Rocle "Rocky" Triantafellu, the deputy chief of staff for intelligence at 7th Air Force, proposed that an organization be established in Saigon which would produce a daily recap of the status of North Vietnamese air and air defense systems. But what Triantafellu wanted and what NSA was prepared to deliver were very different. Triantafellu had in mind an Air Force Security Service organization, all blue-suiters working for 7th Air Force. NSA countered by proposing an NSA unit, manned only partly by uniformed Air Force people. This nasty scrap continued until NSA won in March 1966. The resulting organization, called the SIGINT Support Group (SSG), consisted primarily of Air Force people, but was under NRV eontrol." The very next year, MACV itself got into a struggle with NSA over the positioning of cryptologic assets. In this case, MACV requested that a SlGINT processing center be established in Vietnam, to bring processing closer to the fighting. By 1967, however, MACV was swimming against the tide. NSA had moved processing back to Okinawa (JSPC) and Fort Meade and was not about to change directions. SlGINT centralization was "in," and MACVdid not get its processing center." ARDF and the Two-Front War In the beginning, ARDF was the exclusive domain of the Army. Starlight and the Ia Drang campaign had demonstrated the benefits of close ARDF support, and ASA expanded its assets rapidly. By the end of the year , there were four aviation companies in Saigon, Da Nang, Nha Trang, and Can Tho. The first two supported I FFV (First Field Force Vietnam) in the north, while the second supported II FFV. ARDF had clearly become a coveted asset. 18 In 1966 the ARDF picture became suddenly complicated. The Air Force deployed a new ARDF program, called PHYLLIS ANN. The Air Force considered ARDF to be an EW asset, and even in the test phase in 1962 had refused to submit to any sort of central control from the SlGINT system. The Air Force eventually conceded to bring its ARDF testing under cryptologic control, with USAFSS back-end operators and ASA technical support. (At the time, an ASA unit, USM-626 at Tan Son Nhut, was the tasking and technical support authority for Vietnam, and this made the pill doubly bitter.) But since the equipment was unsatisfactory technically, the issue of command and control became moot with the departure of the test aircraft.1S

°

~!

E VIA TALENT

KEYHOLE C~:::=:~::STEMSJOJNTLY

TOP SeCRET I:IMBRA

532

.::n
PKYUJS ANN Z:C.4'1ARDI'
aircraft

N CO ID M

N

LO Q H

I~MI:rre6"lRO~
ABLE TO FOREICN NA'I'lONA 533 Tep 5EEREt tlMBItA

U

o Q

·DOCID: 523682
TOP SECRET l:IMBRA

REF ID:A523682

PHYLLIS ANN was different. The equipment, mounted in C-47s, was good - just as accurate as the ASA systems, but because of technical factors, the C-47s (now called EC47s) could shoot more DF shots in an hour than the Army aircraft, The Air Force Security Serviceactivated the 6994th S8 on 15 April 1966, at Tan SOn Nhut, to man the ARDF positions. Soon they had detachments at Da Nang, Nha Trang, and Pleiku, A total of forty-seven EC-47s were deployed to the theater.2o

When PHYWS ANN aircraft arrived in theater,the issue of control and tasking of ARDF assets erupted into a three-eornered donnybrook. Seventh Air Force continued to regard them as EW assets and demanded complete tasking control. Westmoreland was equally insistent that all ARDF assets should be tasked centrally (i.e., by MACy). NSA was willing to see central tasking in theater, butinsisted that ARDF was a cryptologic asset whose ultimate owner was itself. In the Agency's opinion, it had simply delegated temporary operational control to the commanding general of ASA in 1961.21 By June of 1966, MACV had won the fight for in-theater control. EC-47s would be tasked by a central AROF tasking center called the ACC (AROF Coord\nation Center), collocated with Westmoreland's J2 in Saigon. Seventh Air Force continued the struggle throughout the war, but it could not get support from even PACAF (Pacific Air Force) for its position.22 The struggle for control went all the way to the deputy secretary of defense. In 1966, Cyrus Vance ruled that ARDF was anEW asset and would be controlled by Westmoreland through his J2. The victory was only temporary, however. Two years later, Deputy Secretary Paul Nitze reversed Vance, holding that ARDF was actually a cryptologie technique and that it would be placed in the CCP. In the meantime, the ARDF controversy had spawned a compromise document, MJCS 506-67, an effort to cut the SIGINT Gordian knot (see p. 475).2$

Search and Destroy
Westmoreland's strategy was to get American troops out of a defensive posture and out into the countryside on search and destroy missions. This placed a premium on unit mobility. The SIGlNT support for these sweep operations consisted basically of ASA tactical units with small numbers of Morse positions, supplemented by low-level voice and shortrange DF. To this mix was added the ARDF fixes flashed from aircraft to the ASA units on the ground and intercept from major SlOIN'! stations like Pleiku and Phu Bai. This pattern, initiated in 1965 during the Ia Orang campaign, became the dominant system in 1966 and 1967, during the height of tactical operations. i

T RELEASABLE TO FORE

nAbS

T9p SEC;R5:rUMiRA.

534

TOP SECRff tlMBIbli

The ARDF con&rol center

In SaifOD.

S~own In 1&11..* ••• ajolnt
('oj

Milly-All'

force facility_

CD \0
I"'l ('oj

10 Q H

U

o
Q

DOCID: 523682
lOP SECRET UMBRA

REF ID:A523682

ASA tactical SIGINT units provided direct support for a bewildering number of military operations during 1966. They came in all flavors: MasherlWhite Wing, Paul Revere, Nathan Hale, John Paul Jones, Geronimo, Attleboro, and many more. One was like the next. An example was Paul Revere II, an operation in the Central Highlands in July and August. SIGINT support consisted largely of ARDF fixes from aircraft that were, for the first time, allocated, based, 'and flown in a direct reporting, close support role from the command post of the supported commander.:UThe historical debate over the effectiveness of Westmoreland's strategy should not obscure the significant contributions of SIGINT. Some of the tactical operations were initiated based on SIGINT information, and most were prosecuted using updatedSlGlNT. A second type was the riverine. operation. Used primarily in the Mekong Delta and other low, marshy areas of the country, it was basically a waterborne search and destroy mission. But the difficult terrain, and lack of large-unit VC operations, made riverine operations frustrating and largely ineffective. This went as well for the SIGlNT support. Working with the Navy and Marines, ASA would deploy low-level voice intercept (LLVI) and short-range direction finding (SRDF) teams on boats. Because of a lack of good linguists, the LLVI teams were generally ineffective. The SRDF operations proved to be no more successful on water than on dry land. Bearings were divergent and frequently produced no intersection at all. 25
I

Army Security Agency was willing to go wherever it was necessary to collect and support. Sometimes units would be choppered to the tops of mountains. One such . operation placed an intercept team on top of Black Widow Mountain, an aptly named peak in a remote corner ofTay Ninh Province at the Cambodian border. This was VC territory, and it turned out to be one of ASA's most dangerous operations. As if enemy operations were not enough, the weather was atrocious - winds as high as eighty knots, heavy rain, low ceilings (which prevented helicopters from landing most of the time), and high humidity that would destroy intercept equipment in short order. But after only a four-day test cut short by hostile rll'e, NSA concluded that it was the only way to get Cambodian I IVHF air/ground communications aside from leaving a TRS iri the South China Sea. Since TRSs were on the way out, Black Widow Mountain was on the way in. So in May 1968 the ASA team was back, this time supposedly permanently·I-E-·.-O-.-1-3-S-2-6-,-s-ec-t-io-n-'--1-,4-(-c-)-' The second time around the team lasted two weeks. At that point, a VC attack killed one ASA operator and wounded another, and caused numerous casualties to the collocated Special Forces unit. The operation was withdrawn by helicopter at the first break in the weather.28

T9P SECRf!T UMBRA

536

DOCID: 523682

REF ID:A523682
TOP SECRET UMBRA

A riverine operation with an ANIPRD-l SRDF set

HANDLE

VIA
.

Z';=!~U CQMlNT
537

S9NIRuL SYSTEMSJOINTL~ BLETOFOREIGNNATIONALS

+oP SECRET l:IMBRA

I E.O. DOCID: 523682
TOP Sf!CRH UMBRA However, the value of operations like Black Widow Mountain spawned an effort to locate intercept equipment on mountain tops and to remote the signal to a safer location: That way, only the equipment would be exposed. The effort, called EXPLORER~ was developed at NSA in only three months, with I, Donald Oliver, and] [being the key players. The first EXPLORER operation lasted for almost a year before it was destroyed. But during its lifespan it was highly effective. In ideal conditions .ASA could intercept the traffic from the EXPLORER system, forward it to NSA for decryption, and have the decrypted text back in country in some four hours."

13526, section 1.4(c)

I

REF ID:A523682
Withheld from public release Pub. L. 86-36

I

Black Widow Mountain

Another successful technique was wiretap. NSA developed various wiretap systems, but they were uniformly dangerous to install. American or ARVN soldiers had to penetrate VC territory (especially risky in Laos, where most of the landlines were), find the land line, attach the tap to the line, and get out of the way. The VC would periodically sweep the line, and early wiretap systems required the Americans or ARVN to stay in the vicinity and, when a sweep came by, hurriedly detach the tap and get back into the bush. Later versions did not require a stay-behind person. Some taps looked like Vietnamese insulators and thus would not be viewed as possible taps. Sti1l1ater, the U.S. developed poles that could be dropped by helicopter into the jungle near a land line. 28 Predictions The highest intelligence art form is prediction. One of the most intensive activities in the war was the attempt to predict VCand NVA offensives-. when, where, and how many.

I E.O. 13526, section 1.4(c) I
HANDLEVIAT LEASABLE TO FOREIGN NAT L SYSTEMSJOINTL , Y

Withheld from public release Pub. L. 86-36

TOP SECRETtJMSIb\

538

DocrD:

523682

REF ID:A523682
'fap SEERR tJMBItA

But in 1964, concurrent with increased NVA involvement in the south, ASA began to intercept Morse communications pertaining to VC military operations. As the Morse nets expanded, NSA began to recover the VCINV A military structure through traffic analysis. The Agency identified the formation of five new organizations: MR TIH (Military Region 'I'ri-Thien-Hue), NVA 3rd Division, 83 Front, Headquarters Southern Subregion, and VC 9th Division. The Binh Gia campaign at the end of 1964 showed the first extensive use of Morse to set up and coordinate a local campaign." From then on, through painstaking traffic analysis and DF, the cryptologic community was able to discover communications patterns that indicated attacks. By 1967 it had become an art form, and many NSA seniors contend that past a point (probably in 1965 or 1966), the SIGINT system predicted every major VC or NV A offensive. This .included date, point of attack, and units involved. Indicators varied from battle to battle but almost always included the activation of a "watch net," contingency communications which indicated that the headquarters would soon deploy to a different location. Concurrently, a forward element would be activated, and would establish communications with the headquarters, which, un\il it moved. would become the rear element. It became important to locate the forward element and to track the movement of the headquarters. At a point in the operation, it would disappear from communications. When it reappeared, it would be in the area of the battle, and it would then be critical to locate it, usually through ARDF. I E.O. 13526, section 1.4( c) Other indicators would usually be present, including the use of unusual cipher systems, changes in message, volume, the appearance of operational planning messages indications of increased intelligence collection, and heightened logistics activity. Plain text and the decryption of low-level ciphers were important, but most of the work was done solely through a combination of ARDF and traffic analysis. Greatly aiding this effort was the fact that the VC and NVA used the same eallsign book throughout the war .' The U.S. had the book completely recovered and used this to identify the units involved."

I

I,

Infiltration A second resounding SIGINT success was in tracking North Vietnamese infiltration on the Ho Chi Minh Trail. Until the fall of 1967, this was done through a combination of photography, SIGINT (primarily traffic analysis), prisoner interrogations, and the like. It was a complex problem, which admitted of no easy answers. The U.S. did not, in fact,' ha ve a good handle on infiltration.
I

'--

Then, in October 1967 RC-130 intercept operators began picking up LVHF voiceD --'Ipassing logistics information. The messages emanated from Vinh,

I E.O.

13526, section 1.4(c)

Withheld from public release Pub. L. 86-36
539
lOll SI!CRET tJMBRA

DOCID: 523682 I D.O-.-1-3S-2-6-,
'fOp SECRET UMBRA -se-c-ti-on-1.-4-( c-)-I

REF ID:A523682

r----------,

!

a key logistics center on the Trail, just above the DMZ. Most of the messages pertained to which NSA decided must "represent groups of infiltrators on the Trail.

I

Withheld from public release Pub. L. 86-36

NSA eventually broke out the entire Trail group system and was able to determine with fair accuracy virtually every group moving onto the Trail, where it 'Yas headed, and when it would probably arrive. Some of the groups proved to be specialists like medics, while others were simply combat soldiers, augmentees for an offensive or replacements for casualties. Late in the war, infiltration numbers were assigned to integral units rather than individuals. The surprising bonanza came to be called the "Vinh Window.'132 The Vinh Window was very big news. MACV now knew where the biggest strategic push would come based on projected augmentees to a given frontal area or military region. The White House thought it had unlocked the key to the magic door, and David McManis, NSA's representative to the White House Situation Room, spent much of his time explaining the intricacies of trail groups. CIA cast aside much of its methodology of determining infiltration numbers and simply accepted the SIGINT numbers as virtually the final answer.3S In Asia, the ACRP program was swept up in a tidal wave of requirements relating to the Vinh Window. The RC-135 unit which had only recently formed at Kadena was pressed into premature service. The RC-130 program, which was eliminated in favor of the RC-135s by the end of the year, was replaced in the fall of 1968 by a new program called COMFY LEVI, RC-130s with roll-on SIGINT suites for the back end. The Air Force Security Service received authority to transcribe the most critical tapes in the aircraft and downlink the information to the Security Service unit at Da Nang in midflight. Untranscribed voice tapes began to pile up at Kadena, as demands overwhelmed resources. 54 The significance of the Vinh Window could not be overemphasized. Every intelligence agency adopted its own interpretation of the figures, and infiltration estimates varied to some degree depending on what agency one listened to. CIA's counts were probably the most accurate, but were not the only ones reaching the White House. The National Indications Center, in a 1968 study of the phenomenon, stated that "... the SIGiNT material which is now available is not only of value for estimating the strengths of Communist forces in South Vietnam, but also is a significant factor in assessing their future plans and intentions. ,~

HANDLE vIA l=:~i:YHor _!EASABLE.TO Tap SECRET UMBRA

E COMINT C9=~:=EMSJOIN~ FOREIGl:.S

540

DOClD: 523682

REF ID:A523682
1'eP SECRET l:IMBIlA,

.
"

>-,•.••..•4.:: MUON!:: NONS
'

..-.
" .:."

",

••••J, ••.•••••••••
• •••• It' •••~ _ ••••• "

, ., ••.••••

: ••.•;~: •••••.••.•• a..••.
.•

',.

CAMBOOIA

The Ho Chi Minh Trail

HANDLE viA

==:;01

E qOMJb!T GeUIRO~~YSTEMSJOINTL Y BLETOFOREIGNNAT 0 ALS
. I

541

TOP SECRET UMBRA

DOClD: 523682
TOP SECRET YMBRA The Dancers

REF lD:A523682
Withheld from public release Pub. L. 86-36

I

E.O. 13526, section 1.4(c)(d)

I

The cryptologic community in Southeast Asia had been overwhelmed with Vietnamese voice long before the Vinh Window. The problem began in late 1964, when the (U'Stvoice intercepts began to flood the SIGINTsystem.What had been entirely a Morse problem suddenly had a new dimension to it. The services had very few Vietnamese linguists, and those they had were little better than school trained. In 1964 USAFSS requested authority to establish native-born South Vietnamese as linguists to transcribe voice tapes to be collected at Da Nang in support of 34A operations. After studying the problem, NSA concurred with a Vietnamese transcription operation, but established it, not at Da Nang, but in Saigon. The DANCER project (as it was called) was established in January 1965, using 3rd RRU SABERTOOTH spaces, with three South Vietnamese linguists." By May 1965, USAFSS was processing Vietnamese voice offl nets being collected by the ACRP program at Da Nang. The program in Saigon was I Ithe onl)"female Dancer not productive, partly because ASA could hear no Vietnamese voice from that location. Since Da Nang was the ground processing point for ACRP intercept, it was decided to move the DANCER program north - ultimately it wound up at both Da Nang and Phu Bai (selected because NSA believed I Icommunications could be heard from that location). DANCE~ recruits came from the SABERTOOTH program and were vetted by General Nhon's South Vietnamese SIGINT organization. 97

I

Originally employed to transcribe voice tapes, DANCERs eventually became qualified in a wide variety of skills. They proved to be skilled at various traffic analytic recoveries, and they were soon an absolutely essential asset to any. SIGINToperation in' South Vietnam. By 1966, ASA units were intercepting LL VI communications and needed DA.NCERs to go to the field with them. This effort became Project SHORTHAND. Because the U.S. had run through the supply oflinguists available from the South Vietnamese SIGINT Service, ASA, under SHORTHAND, obtained authority to recruit from other sources within the South Vietnamese government.38 Withheld from

I

E.O.13526, section 1.4(c)(d)

I

public release Pub. L. 86-36

N'TROL SYSTEMSJOINTL Y . TRELEASABLE TO FOREIGN NATIO
TOP SECRETUMBRA

542

-

--.------

DOClD: 523682

REF lD:A523682
TO" SECRET tJM8RA

The

SIGINT

Role in the American War

During the period of maximum American involvement ~n the ground, SlGINT developed from an arcane art form to a day-to-day bulletin on enemy dispositions. Most commanders interviewed after the fact estimated that SIGINT comprised anywhere from 40 to 90 percent of their intelligence, depending on the availability ofPOWs. Every sizeable unit deployment had its ASA Direct Support Unit (DSU), which gave it access to ARDF and a pipeline into the national SIGiNT system. Many commanders used the information for daily battle planning.59 A properly employed DSU thus became an essential resource. But it had warts. As in Korea, the LLVI effort was sometimes fruitless because of the difficulty of getting good linguistic support; an insufficiently trained linguist was sometimes worse than no linguist at all. South Vietnamese linguists under the DANCER and SHORTHAND programs were spread very thin and were often not available. Moreover, short-range OF proved a dubious asset, especially in the Delta, where there were fewer targets. To the extent that DF was successful, it was generally ARDF. ARDF sometimes overwhelmed other intelligence sources. Tactical commanders used it for daily targetting,and it became the primary source for targetting information in the entire war. Used effectively, it was irreplaceable. But sometimes a commander would blast a patch ofjungle just because a transmitter had been heard there, The VC and NVA eventually became skilled at remoting their transmitters, just because of such American tendencies. There was still no substitute for understanding the source. And much of the difficulty that the SIGINTersfound themselves in stemmed from an unappreciative audience. Very few commanders had any training in SIGINT.In the 1950s it had been kept closeted.' a strategic resource suitable only for following such esoteric problems as Soviet nuclear weapons development. Now that it was "coming out of the closet," a generation of officers received OJT under fire. Some did well; some not so well. For every example of the proper use of tactical SIGINT, there was the opposite instance, where the source was either not believed or not used properly. No intelligence source was so technically complex or so difficult for the layman to understand. The lessons from the "American War" (1964-1968) were still being absorbed more than twenty years later. The Air War The air war began with the daily bombing of the North in March 1965. Like the ground war, the air war was a messy business organizationally. It involved three different air elements.

HANDLE VIAT ~-"'TnT

CONTROL SYSTEMS JOINTLY RELEASABLE TO FOREIGN NA TI
543

-

TOP SECRET tJMBItA

DOClD: 523682

REF lD:A523682

Seventh Air Force was the largest component. It had six tactical fighter wings and a tactical reconnaissance wing spread around Southeast Asia. Headquartered at Tan Son Nhut, 7th AF had a Control and Reporting Post on a hilltop called Monkey Mountain, near Da Nang. This was where command and control of tactical missions were executed, and this was where Air Force Security Service chose to set up shop." In the Gulf was Task Force 77, a carrier task force belonging to 7th Fleet. The Navy launched Rolling Thunder missions from the carrier decks, and it had its own control authority, called Red Crown." The First Marine Air Wing, under III MAF, operated out of airfields in northern South Vietnam. Although used almost exclusively for close air support in South Vietnam, they also flew some missions over the North.42 . Finally there was SAC. The Strategic Air Command launched B-52 strikes over both North and South Vietnam, flying out of Andersen Air Force Base, Guam; U-Tapao, Thailand; and Kadena, Okinawa.4S In response, the North Vietnamese. with a third-rate air force and practically no technological sophistication. had fashioned a competent if not overwhelming defense. Proceeding from the visual observer stage in the late 1950s, North Vietnam had introduced Soviet radar systems, and by the mid-1960s it had some 150 radar sites and 40 radar reporting stations. The North Vietnamese navy also had radar sites along the coast, primarily to keep track of enemy ships. They had a small group ofMIG-17s and MIG-21s which they carefully husbanded. They also introduced hundreds of AAA sites across the country and in late 19.65began installing SA-2 sites. American air strikes by no means went unimpeded." l'E-.O-.-1-3-S-2-6-,-se-c-tI-' o-n-l-.-4(-c-)Withheld from public release Pub. L. 86-36

Fashioning the SIGINT Warning System - HAMMOCK

SIGINT

Following the Gulf of Tonkin crisis. 7th AF (then called 2nd Air Division) requested support for air missions north of the DMZ. Security Service began planning an

fap SECRET UMBRA

544

DOClD: 523682

REF lD:A523682
TOP SECRET UMBRA

expansion of its unit at Da Nang (6924th SS, or USA-32) to provide some sort of Tactical Report (TACREP) service." What developed initially was a system called HAMMOCK, which became operational in December 1965. HAMMOCK consisted offive manual Morse intercept positions at USA-32, copying North Vietnamese air defense communications which reflected MIG activity. USA-32 could pass warning information to 7th AF when, and only when, the tracking fell within the theoretical range of American radar. (There did not actually have to be a radar located at the hypothetical point; the postulated existence of such a radar was enough.) The information was supposed to be validated at the Tactical Air Control Center (TACC) at Tan Son Nhut, which would convert the SIGINT plot to a geographical coordinate and then send it on to the Control and Reporting.Post (CRP) at Monkey Mountain. The CRP could warn the aircraft in jeopardy and would also pass the information via KW-26secured circuit to Red Crown in the Gulf. I! communications were down, USA-32 could go directly to a Security Service detachment at the CRP, where the information was converted from the grid system and passed to an uncleared CRP controller. This was much faster, but everyone was nervous about security because there were so many uncleared people in the facility." Needless to say, this convoluted system was less than satisfaetory, It relied, in the
first instance, on manual Morse tracking passed within the North Vietnamese air defense

system, which introduced a delay of several minutes. It was burdened by so many communications relays and authorization authorities that it had little chance to get anywhere in time. HAMMOCK plots generally reached someone who could warn a fighter pilot anywhere from twelve to thirty minutes after the fact. The average time of receipt to Red Crown was nineteen minutes. The Navy was profoundly unimpressed and chose to rely on its on-board cryptologic detachments. The Navy operators had little experience with North Vietnamese air defense systems, but at least they could warn within a few minutes of real time.48 Despite this, HAMMOCK was better than nothing. On 27 April 1966, the U.S. got its first confirmed MIG shootdown based on warning information provided by HAMMOCK. But the requirement to check everything with the TACC in Tan Son Nhut got the Air Force Security Service in the middle of a jurisdictional dispute between 7th AF and its subordinate CRP on Monkey Mountain. It was not the right way to run a war."
»<

The ultimate answer was not manual Morse tracking, anyway - it was intercept of VHF air/ground communications by the RC-130 QUEEN BEE DELTA aircraft flying in the Gulf of Tonkin. The ACRP often had the information that pilots needed to avoid being shot down, or to do some shooting down themselves. Security restrictions, however, prevented its use. The cropper came in April 1965, when two F-105s were shot down by MIGs. The orbiting ACRP had had information that would have been useful, and it was obviously

HANDLE VIATTA:~~::LE~C~O~M~l~NT~C::~~~MSJOINTLY NOT RELE ATIONALS

545 .

1f;1P SECRET b'MBRA

DocrD:

523682
TOP $I!CR!T tJMBItA

REF ID :A523682

Voice intercept operators at work. USA·32

HANDLE VIA TALEN'I

INT CONTROL SYSTEMS JOINTLY

~--r.v ~RELEASABLE TO FOREIG

TOP SECRET I:JMBRA

546

DOClD: 523682

REF lD:A523682
TOP$ECREII:JMBRA-

imperative that a system be devised to incorporate their intelligence. Pacific Security Region (the region headquarters for the Air Force Security Service) had devised a brevity code that could be used by the ACRP back-end crew to warn pilots in imminent danger, but it did not withstand COMSEC scrutiny. The only solution appeared to . be a secure link between the ACRP and USA-32. A device called a URC-53 already existed. Priority was so high that the installation and use of the URC-53 at Da Nang was approved the same day it was requested, and the circuit was installed and operational within a month.5O But this was still not fast enough .. General Moore, commander of 7th AF, proposed putting his own controllers on the QUEEN BEE aircraft, clearing them, and having them pass MIG alerts directly to Rolling Thunder aircraft, using the callsign of another aircraft in the Gulf (COLLEGE EYE; an EC-121) as cover. Reversing the normal procedure, Morse tracking would be passed uplink from Da Nang to the ACRP, where it would be integrated with the voice data. Moore's weapons controllers were flown to Bangkok (whence QUEEN BEE flights then originated), and three days later the ACRP issued its flr~t MIG alert." Then Moore tried to get control of the ACRPs themsel ves. He felt this was necessary to insure that there was always an ACRP aloft during ROlling Thunder missions. Here Moore ran into a buzzsaw. The aircraft he wanted control of were national assets. NSA successfully opposed 7th AFon this issue. Even PACAF refused to back 7th AF, stating at one point that there had never been an instance when the ACRP had failed to respond to a 7th AF request. 52 The autumn of 1965 brought a new threat - the appearance of SA-2 surface-to-air . missiles (SAMs) in North Vietnam. The North Vietnamese began employing SAMs against high-flying, nonmaneuverable targets like B-52s, while using AAA for the lowerflying Rolling Thunder aircraft. To counter SAMs, 7th AF introduced a procedure in which SAM activations acquired by the ACRP aircraft (now renamed SILVER DAWN) would be passed to 7th AF (through USA-32), which would direct Iron Hand (SAM suppression missions) against the offending SAM. At this point Security Service ran into an Air Force mind-set regarding the use of intelligence that proved to be destructive of its own interests. Air Force doctrine was to launch suppression only if the SAM site had been documented by photography, and 7th AF refused to launch Iron Hand in cases where this had not been done.ss The Border Violation Incident On 8 May 1966, a flight of RB-66s escorted by F4Cs strayed ov~r the border into Communist China and was attacked by four MI~17s. One of the MIGs was shot down in the engagement, which occasioned an impassioned diplomatic protest from the PRe. The communists released photos of the downed MIG well north of the international barrier. ~

HANDLE VIA TALE ONTROL SYSTEMSJOlNTI.. Y ~N!l:r.JiUSt~tABLE TO FORE

547

TOP SECRET UMBIU("

DOCID: 523682
lOP SECRET t:JM8RA

REF ID:A523682 I E.O. 13526, section 1.4(c)

Withheld from public release Pub. L. 86-36

The off-course Americans should have been warned. one, unfortunately, reached the American pilots. The Navy EC-121 that was supposed to act as a communications relay had aborted, and the warnings from Monkey Mountain went off into the ether.~ This incident led to a full-scale Pentagon investigation of command and control .procedures in Southeast Asia. The "Pearl Harbor question" kept coming up - why, if SIGINT was available, wasn't it used? The proceedings, headed by Marine brigadier general Robert O. Owens Jr. were marred by mutual recriminations between the SIGINTers,who were sure of their facts, and the operations people, who were determined to defend their ilots. ---l This claim was rejected by the full panel. In the end, Secretary of Defense McNamara reported to the president that "this account, derived from communications intelligence, is unequivocal. A thorough review of intercepted I INorth Vietnamese messages reveals no significant discrepancies .... I am convinced that our aircraft penetrated Chinese airspace before they were attacked by the
L---

MIGs."$I

The Owens report laid bare the inadequacies of command and control and the . disjointed way that SIGINT was introduced into the operational system. Owens demanded, and got, a thorough reorganization of the system in Vietnam. Authority to control operations was summarily removed from 7th AF in Saigon and placed where it should have been all along, on Monkey Mountain. The Tactical Air Control Center (TACC) at Tan Son Nhut was cloned on the mountain and called TACCINS (North Sector). The control facility on the mountain was upgraded from a CRP to.a CRC (Control and Reporting Center) and was given two subordinate CRPs at Udorn and

I

I

I

I

The Owens report also recommended that 7th AF have operational control over the ACRPs. This occasioned another huge fracas between the Air Force and NSA. The Agency won again, partly because it could certify that the ACRPs were already as responsive to 7th AF as they would be under that organization's direct control. 68 During the Owens deliberations, it·became clear that factors other than operational control affected ACRP capabilities. The biggest problem was fighter CAP (Combat Air Patrol). Many ACRP missions were scrubbed because of lack of fighter CAP, or had to abortin midmission because the fighters went home early. Following the Owens report, JCS approved unescorted missions in the gulf at night (because of known North Vietnamese reluctance to fly at night). As time went on, the rules were relaxed even more.$9

I E.O.

13526, section 1.4(c)

Withheld from public release Pub. L. 86-36

HANDLE

V'~IA~n:fm:.KE:~H[!O~L&L:Glf!tfN'lt'CCONTROL

SYSTEMS JOINTLY

NO

LE

NATIONALS 548

1'8' SEERETtJM11tA

DOCID: 523682

REF ID:A523682
Top·SFCaliT YMBItA

. The reforms permitted SIGINTto focus its input at one geographical point - Monkey ·Mountain. This shortened the chain of organizations through which a warning had to pass and simplified the task of the SIGINTers in Southeast Asia. It did not, however, provide a direct link-up between SIGINTand the operations people. That necessary step would not come for another five years.
IRON HORSE

In 1967 the SIGINTsystem improved the speed of its support to air operations by a quantum leap. The creaky manual system, HAMMOCK, was replaced by IRON HORSE, a flashy new automated system which could deliver information in seconds rather than minutes. Designed by NSA, IRONHORSEsimply linked the electronic output of an AG-22 intercept position, through a computer, to a radar scope. Instead of using a plot-tell system for calling aircraft positions to the T ACC or eRC, the computer would convert the grid plot to a geographical coordinate and display it on a radar scope. An Air Force Security Service analyst carefully selected the plots that were sent to 7th AF. Those that were passed went into the BUlC II air defense computer at TACCINS and were integrated with radar plots from the U.S. system. Plots from SIGINTthat went to the CRC. Task Force 77, and the Marines had a unique signature that identified them as not derived from American radar. USAFSS put a team of SIGINTexperts in the collocated TACC and called it the Support Coordination Advisory Team (SCAT) -in effect. a CSG to help 7th AF interpret the data. SCAT integrated manual Morse data as well as VHF reflections from the ACRP, the Navy's EC-121, and a variety of other sensors." IRON HORSE decreased throughput time from twelve to thirty minutes to anywhere from eight second~ to three minutes." It was state-of-the-art and about as fast as Morse tracking could be displayed. .

IRON HORSE consoles. USA-32

HA

r~»f'!~('1SYSTEMSJOINTL IONALS 549

y

IOP SECR&:TYMBRA

DOClD: 523682
TOP SECRET I;JMBM

REF lD:A523682

BIGLOOK The introduction of SAMs into Vietnam complicated the air warning picture. Special airborne warning systems to detect the SAM-associated Fan Song radars were thwarted when the North Vietnamese introduced the tactic of putting the Fan Songs on lower power except when they went into a track and destroy mode. Navy engineers devised counter for this, a system that could intercept and OF very low power signals. They mounted these systems on EC-121 airframes allocated to VQ-l for fleet support. The ELINT crews came - from the home squadron I I while the four voice intercept operators were supplied by USN-27 at San Miguel, Philippines.62 . I E.O. 13526, section 1.4(c)

a

BIGLook was supplemented by WEE LOOK, an EA-3B fleet support aircraft outfitted with ELINT positions. WEE LOOK was also used for threat emitter warning. Although the EA3B was designed to operate from carriers, WEE LOOK did not because of aircraft, weight. Like BIGLOOK, it launched from land bases." Weather and SAR Warnings One obscure but vital SIGINTcontributionwas weather. Early in the war, 7th AF flew weather reconnaissance missions prior to operational launches, but it was an Operational Security (OPSEC) nightmare. Weather reconnaissance was the surest indicator that the North Vietnamese could have that a strike was imminent. In 1965 NRV proposed to 7th AF that USA-32 at Da Nang begin furnishing "special weather" information intercepted on North Vietnamese nets. Da Nang initiated a twoweek test and within a month had become the sole source of CO MINT-derivedweather information on North Vietnam: Special weather was relayed to Task Force 77 as well as 7th AF, and an Air Force historian, with pardonable exaggeration, called this perhaps the -"premier contribution" of SIGINT in Southeast Asia.84 When the Air Force and Navy began losing pilots over Vietnam, SIGINT was once more called in. A special program was designed for reporting indications (through VC or NVA communications) of downed pilot locations and capture attempts .. The reports, called SONGBIRDs, were actually TACREPs, which went out at the noncodeword level to a wide group of organizations. Security Service averaged about ten SONGBIRD reports per month. There was very little feedback on SONGBIRD effectiveness, although .one historian estimated that, because of the time required to translate the Vietnamese voice transmissions, most SONGBIRDs did not arrive in time. 66

Withheld from public release Pub. L. 86-36

HAN
TOP SECRET UMBR.A

OLE ASABLE TOFOR

TROL SYSTEMS JOINTL Y

550

DOClD: 523682

REF lD:A523682
TOP SEeR!T UMBRA

PURPLE DRAGON
President Johnson •.. expressed concerns over the number of aircraft being lost on Rolling Thunder missions. Betwee"nJanuary and September 1966, a total of228 (!.Xed.wing combat and support aircraft had been lost during missions against North Vietnam. The answer was ye8, they did. Stephen J. Kelley in PURPLE DRAGON: The Origin and DewloprMnt of the United Stou, OPSEC Program The question in Washington was, did the enemy have prior warning of U.S. raids against North Vietnam? .•.

On Christmas Day 1969, a team of the First Infantry Division, on a sweep in Binh Duong Province near Saigon (part of Operation Touchdown), stumbled on an NVA COMINT unit. They captured twelve of the eighteen people assigned along with some 2,000 documents and the unit's intercept equipment. It was the COMINT "find" of the war. NSA sent in a TAREX team to evaluate what the soldiers had found. The result confirmed an earlier, and generally ignored, Agency assessment - that the NVA employed 4,000 to 5,000 COMlNTers and that this was their chief source of intelligence. Their intercept effort was targetted at ARVN and American communications, from which they could do fairly sophisticated traffic analysis, DF, and even some cryptanalysis." Brevity codes were especiaI1y vulnerable. But their main target was unenciphered tactical voice, and the easiest pickings were from the U.S. Air Foree." It was obvious from studying the Touchdown material that NVA COMJNTers were a source, probably the source, of predictive information on SAC Arc Light (B-52) strikes. But the Defense Department knew that already. 87 [-E-.-o-.-1-3-5-2-6-, -se-c-t-io-n-l-.4-(-c )---,

Withheld from public release Pub. L. 86-36

The ~tory had begun in 1965. NSA had uncovered a communications net supporting Chinese forces in Vietnam. analysts notleed that some or the messages contained an unusual Morse character - a barred echo. They remembered thatl used this character to flag uncommonly urgent messages. On a hunch, the division chief, suggested that they might compare barred echo messages with Rolling 'I'hunde r operations. The result was a direct hit. The barred echo message appeared almost every time a Rolling Thunder mission was flown over the northeast quadrant of North Vietnam. The PRC appeared to be obtaining predictive alerts on 80-90 percent of the missions in the northeast quadrant."

I

I

I

I

I

At about the same time, NSA found that ground control stations were alerting air defense forcesl las much as twenty-four hours in advance of SAC photo drone't:nissions. called (at the time) Blue Springs. As a result, approximately 70 percent of the drones were being lost to hostile fire. A check of existing traffic showed

I

I

I

I E.O.

13526, section

1.4(c)

I
LE'TO ALS"

,--------, Withheld from public release Pub. L. 86-36

~~,~y

551

'FGP SECReT UMBRA

DOCID: 523682
Tap SECRET l:IMBAA

REF ID:A523682
Withheld from public release Pub. L. 86-36

I E.O. 13526,section1.4(c)
that I

Ihad been issuing alerts on SAC reconnaissance missions as early as mid-

1965, and on Arc Light strikes, by late 1965.69

NSA released its report in May The effect was immediate and r--------------------, dramatic. Within days, NSA analysts .found themselves standing in the Pentagon briefing four-star generals. In August, after pulling together the Withheld from full story (including indications of public release foreknowledge of SAC operations), Pub. L. 86-36 General Marshall Carter briefed the JCS and, later in the month, the PFIAB.70
1966.

As a result, DIA was tasked to find the problems and correct them. The director, General Carroll, named Rear Admiral Donald M. (Mac) Showers to head the effort. Showers put together an interagency committee which included NSA, the JCS staff, and the . SCAs. The group was divided into two subcommittees, counterintelligence and communications security. 71

The counterintelligence group quickly concluded that the problem was enemy infiltration, but they could come up with no good way to stem the outflow of information. The COMSEC committee concluded that communications were the problem and that they were probably closer to the truth. But in addition, 'the COMSEC group came up with a methodology for investigating the problem and plugging the holes." I The COMSEC, committee adopted a multidisciplinary methodology for looking at the problem in which all facets, including communications, would be studied. NSA had been working on the methodology for several years, and the Navy had already tried it with some success in surveying maritime operations in the Gulf of Tonkin (called Market 'I'ime)." The committee also borrowed from a COMSEC study of Arc Light operations done in 1965, called the Guam Area Study. Although the Guam study looked at the communications of all three services, it concluded that most of the insecurities came from SAC communications. Traffic analysis of encrypted messages yielded much preoperations information, including probable launch times. They also discovered voluminous plaintext voice by logistics people an hour before the launch. Finally, they

TOP SECR!T tlMBItA

552

OOCIO: 523682

REF IO:A523682
TOP SECRET UMBRA

found that prestrike weather flights twenty hours before launch were dead giveaways (as they had been in World War 11). In July 1966, Admiral Sharp (CINCPAC) ordered a broader COMSEe study of the problem, encompassing operations throughout the Pacific.74 The
PURPLE DRAGON

Task Force

The CINCPAC and DIA studies joined in September. Sharp agreed to adopt the broader DIA multidisciplinary approach, and he named his J3 to head the effort. The new study, called PURPLE DRAGON, would encompass Rolling Thunder, Arc Light, and Blue Springs. Teams of experts would be dispatched throughout the theater. They would first interview all people involved in the three operations. They would then observe the operations, following that up with observations of support activities, including logistics and intelligence. They would build a database for their information and would build three profiles: operations, communications, and counterintelligence. An NSA person, Robert Fisher, served on the CINCPAC PURPLE DRAGON staff, and there was heavy infusion from the SCAs, primarily for COMSEC monitoring." The f11'St PURPLE DRAGON study concluded in April 1967. It had a bIg impact on operations in Southeast Asia, none more significant than Blue Springs. They discovered that the major leak was the encrypted single sideband messages from Bien Hoa to Da Nang prior to every mission. Usingtraff'lc analysis of that link alone, the team was able to predict eighteen of the twenty-four missions. As an almost direct result of introducing communications security on the link, drone recovery increased from 35 percent to 70 percent by November 1977.76 Arc Light was much more complex and harder to solve. One or the main culprits proved to be the information fed to the Manila and Saigon air control centers. This information was released all over Southeast Asia as NOTAMs (Notice to Airmen) giving flight routes, altitude reservations, and the estimated time of arrival at Point Juliette, the aerial refueling spot, hours in advance of the mission. SAC tightened up by curtailing much of the information in the NOTAMs and by delaying that which was passed until a time closer to takeoff.77 MACV had been passing warnings to villagers in the targetted area. This procedure was modified by simply declaring certain areas as free fire zones and discontinuing the advance notification program. 78 . Of the three, Rolling Thunder was the most difficult to plug. PURPLE DRAGON investigators found that many of the enemy's sources of warning consisted of tactical information obtained after the planes were launched. They determined that between 80 and 90 percent of the missions were being alerted, with an average warning time of thirty minutes for Navy missions off the carriers and forty-five minutes for Air Force missions

HANDLE VIA TALENT KEYH CONTROL SYSTEMSJOINTLY lU:~m~A:§jSABLE TO FOREIGN NA

553

TOPSECREI

UMBRA

DOClD: 523682
TOP SECRET UMBRA

REF lD:A523682

B·52

Air Force F·l05 fighter·bombers

on a Rolling Thunder mission

HAND

YHOL ABLET
./

NTROL SYSTEMS JOINTLY

NALS
554

TOP $!CR!T

tJMBRA

DOCID: 523682

REF ID:A523682
TOP SECRET tlMBRA

from airfields in South Vietnam. EB-66s accompanied many of the missions (those expecting hostile fire in particular), and those aircraft used distinctive call signs. Rolling Thunder frag (read "operations") orders were distributed to 120 different organizations, and those in turn often issued information that could be tied to the takeoff of bombing missions. MAC V cut down on the number of organizations getting gratuitous copies of the operations orders, and the Air Force changed callsigns for some of their operations." Much of what needed to be done simply could not be because of outside factors. MACV never did alter stereotyped operations (such as takeoff times, refueling points, and ingress routes) sufficiently to confuse the North Vietnamese. Tanker operations remained highly stereotyped throughout the war and in fact represented the most vulnerable aspect of Rolling Thunder. so

The Permanent Stafr
Following the initial blush of success, Admiral Sharp made a permanent place on his staff for the PURPLEDRAGON operation. He placed it in the J3 (operations) directorate, and NSA assigned a permanent representative (once again, Robert Fisber)." There was obviously a need to educate people about the concept and about the methodology and specific information that PURPLEDRAGON uncovered. This generated the first worldwide OPSECconference, hosted by DIA at Arlington Hall Station iIlMay 1968. Following the conference, General Wheeler directed that all Unified and Specified commands establish OPSECorganizations. He also created an OPSECorganization on the . Joint Staff. Meanwhile, OPSECconferences continued annually and helped to focus activity for the U &S commands. Cryptology continued to be a major player, and in 1988 NSA was given the job of worldwide OPSECtraining under the newly published NSDD (National Security Decision Directive) 298.82 The OPSECconcept in use in the defense department of the 1990s was largely an outgrowth of the PURPLEDRAGONstudy. It was a significant factor in prosecuting the air war in Vietnam, although neither it, nor anything else the United States tried in Vietnam, was a panacea. The CINCPAC OPSECteam would periodically resurvey operations in Southea'st Asia, and they found that, as the U.S. tightened up procedures, the North Vietnamese would find another leak, and their warning time would float back up to where it had been. Like cryptology in general, OPSECproved to be a constant struggle to stay

ahead."

HANDLE VIAT

SYSTEMS JOINTLY

555

l=QP SIiCRET UMBRA

DOCID: 523682
Tap SECRI!T U M81tA

REF ID:A523682

Notes

1. Eric:F. Goldman, The Tragedy of Lyndon. JpllnBon. (New York: Alfred A. Knopf, 1969),402. 2. CCH Series VI.HH.15.1. 3. Goldman,40~: KarnQw, ~lS. 4. Kamow,415-18. S. Karnow, 420-22. 6. CCH Series VI.HH.15.1. 7. Karnow, 479-80; CCH Series VI.HH.15.1 & 2. 8. CCH Series VI.HH.15.1 & 2. 9. CCH Series VI.HH.IS.l. 10. Ibid. 11. CCH Series VI.HH.15.1.

Withheld from public release Pub. L. 86-36

12. Ibid. ,13. Thomas N. ThompsonJ land William D. Gerhard, SIGINT Appliootio1l8 in. U.s. Air Opel'OtW1I8, Part]: Collecting the Enemy's Sig1l4l8, Cryptologic History Series, Southeaat Asia (Ft. Meade: NSA, 1972). 14. Thompson, et al., SJGINTApplicationll 15. CCH Series VI.HH.15.1·. 16. Thompson etal.,SIGINT 17. CCH Series Vl.HH.12.1. 18. CCH Series VI.HH.12.2.
19. Gerhard Collection. Applicalio1l8 •.. , VLHH.1l.3. ....

20. CCH Series VI.HH.15.12.2; ESC oral interview withl

I, San Antonio, TX,I Aug 1986.

21. VI. HH.15.12.2.; VI.HH.l 1.l.11.S;5MSgt Frank Whitacre, "A Historical Study oftM Drawdown ofUSAFSS Operatiol18 in Southeast Asia (SEA)M (San Antonio: USAFSS,1974). 22. Whitacre, VI.HH.15.12.2. 23. VI.HH.11.1.
24. CCH Series VI.HH.15.12.2.

25. CCH Series VI.HH.lS.1. 26.1 ~Focus on Cambodia, Cryptologic History Series, Southeast Asia (Ft. Meade: NSA.1974).

27. Gaddy interview.

TOP SECRET UM8R1<

556

DOClD: 523682

REF.ID:A523682
lOP SECR!T UMBRA

28. Gaddy interview. 29. CCH Series VI.HH.9.a.

Withheld from public release Pub. L. 86-36

30. Ibid. 31. Ibid.c=J illterview, Johnsoll Library, NSF.

32. "A History of the USAFSS Airborne SIGINT Reconnaissance Program (ASRP), 1950-1977," USAFSS history dated 20 September 1977, available atAIA, Kelly MB, San Antonio; CCH Seriea VI.HH.23.2-23.5. 33. McManis interview,Johnson Library, NSF. 34. "A History of the USAFSS Airborne Reconnaissance Program •.. ," interview with Major General (USAF, Ret.) Doyle Larson, 15 March 1994, by Charles Balcer and Tom Johnson, NSA OH 15-94. 35. CCHSeries VI.HH.23.26.
3G. CCH Series VI.HH.G.53.

37. Ibid. 38. CCH Series VI.HH.6.S8. 39. CCH Series VI.HH.16.1. 40. Thompson, et &1., SIGINT Application, ..•• 41. Ibid.
42. Ibid.

43. Ibid. 44~ Ibid. 45. Ibid. 46. CCHSeries VI.HH.ll.3. 47. CCH Series Vl.HH.11.1; James E. Pierson, "A Historical Study of the Iron HOI"5e System; 1960-1973," USAFSS history available at AlA, Kelly AFB, San Antonio, Teus. 48. Pierson, CCH Series VI.HH.15.9. 49. Pierson. 50. CCHSeriesVI~HH.ll.l.
·51. Ibid.

52. Ibid. 53. CCHSeries VI.HH.l1.l. 54. CCH Series VI.HH.ll.l. 55. Ibid. 56. Ibid.; Jobnson Library, NSF, in CCH Series XVI. 57. VI.HH.l1.l; 11.3.

HAND

T KEYHOLE COMINT EMSJOINTL Y NOTRE REIGNNATIONAI.S

557

DOClD: 523682
TO' SECRET tlMBRA

REF lD:A523682
"

58. VI.HH.1l.1. 59. Thompson, et al., SIGINT Applicationll .•..

60. CCH. Series VLHH.; Pierson, Iron Horse ....

Withheld from public release Pub. L. 86-36

61. Pierson, Iron Horse. '"

62·1

I,article in Cryptolog,3rd

issue, 1987, 7-9.

63. Thompson, et al., SlGINT Applicatioll8 •..• 64. CCH Series VI.HH.1l.3.
65. CCH Series Vr.HH. 66. Donald Betts, Hiram M. Wolfe Uf, Raymond P. Schmidt, and Thomas N. Thompson, Deadly TraTllfmislliollS (COMSEC MOllitori"ll alld AMlysis), Cryptologic History Series, Southeast Stephen J. Keney, PURPLE DRAGON: TM Origin anaDevelopIMN Asia (Ft. Meade: NSA. 1970); OftM United States OPSSCProgram, United David Gaddy and David

States Cryptologic
67. Ibid.

History, Series Boak, October 1994.

VI. Volume 2 (Ft. Meade: NSA, 1993); interviews with

68. KeJley,PURPLEDRAGON.
69. Ibid. 70. Ibid.

71.Ibid. 72. Ibid. 73. Ibid.
74. Betts, et al., Deadly TransmillBiollS. 75. Kelley. 76. Betts, et al., Deodly Tmnsmi8sion.. .

'1'1. Ibid.
78. Ibid.

'19. KeUey, PURPLE DRAGON.
80. Ibid.

Withheld from public release Pub. L. 86-36
L-________

81. Kelley,PURPLEDRAGON:~

I. ftThe Great

Conversation."

CryptxJlog,lst

Issue 1992, 2~.

. 82.1'-83. Kelley. PURPLE DRAGON.

lftOPSEC as a Management

Tool," Crypto/og, 1st Issue .1992,'1-9.

HANDLE VIA TAL BLE TO FOREIG T8P SECRET t;JMBR>A.

YSTEMSJOINTL

Y

558

DOClD: 523682

REF lD:A523682
TOP SECR!T UMBRA.

Chapter 13
The Withdrawal

THE TET OFFENSIVE
A~ericalUl do not like long, inconclusive ware - and this is going to be a long, inconclusive_war.
Thus we are sure to win in the end.

Pham Van Dong, North Vietnam'lIclIiefnegotiator at the Paris peace talks

In Vietnamese history there are many Tets. Like the American Christmas, the lunar New Year holiday is celebrated every year - one of the big events in the timeless cycle of Southeast Asian civilization. In American history there is only one Tet. It has become a synonym for defeat and withdrawal, the beginning of the great unraveling of American power in the region. Like many symbols, the characterization is desperately inaccurate in the military and cryptologic senses, but generally true from the political perspective. That is why Tet 1968 symbolizes the deep fissures about Vietnam within American society. The Planning It has become generally recognized that the communist strategy in Tet was to mount a sudden, massive assault, forcing the Americans to recognize the instability of their alliance with the South Vietnamese government and to realize the difficulty of ejecting the communists from their own country. It was to drive home to the Americans the long-range impossibility of surmounting a determined adversary on his own soil. Some say that it was a one-shot affair, but the weight of evidence is against it. Although the North Vietnamese leaders did call for a popular uprising against the Thieu government, there was no sense that, nit failed, they had come to the end. They would simply continue the struggle. Just as there would be lunar new years into the trackless future, there would be other times and other Tets. The tactic ofTet was to divert American attention to border areas, while building for a major assault on the urban populations. To do this, the North Vietnamese would have to mount a lIlIijor dry season offensive. By attacking In outlying provinces, Giap, the Vietnamese general, sought to make them magnets for American units, then hit the unguarded cities. He aimed for surprise, but he was confronted with the extreme difficulty otreadying so many people for such a herculean task without alerting the enemy.

-;~~'
.

N

LE TO FOREI

559

1QPSE(R&T UMlAA

DOClD: 523682
TOP '''RET l:JMBA-A

REF ID:A523682

The Beginnings The winter-spring offensive began, it is now believed, in September 1967 with a surprise attack on a small Marine fire base located on a barren hill south of the DMZnear the town of Con Thien. Westmoreland was delighted that the North Vietnamese appeared at last to be mounting major unit-level assaults. To defend Con Thien, he called In B-62 strikes, artillery, tactical air bombardment - anything at hand. Con Thien held. l The next attack was planned for Dak To, a provincial town northwest of Pleiku in the Central Highlands. But this time it was not a surprise. On 20 October the ASA station at Pleiku picked up indications that the B3 Front had sent a detached element toward Dak To, and two other NVA divisional organizations appeared to be concentrating in the Dak To area. Three days laterl referred to "combat reconnaissance," an almost certain indicator of offensive action. Dak To was immediately reinforced. Aerial bombing in the area of an ARDF fix brought secondary explosions, and American units airassaulted a hill near the town, encountering heavy enemy resistance. The resulting battle was one of the biggest of the war. It came to involve nine American battalions, an airborne brigade, and over 2,000 air sorties. Roughly 1,600 NVA troops were killed by ground action, and 500 more by aerial bombardment. 3
SIGINT picked up other indicators of major developments. In Nam se; the southern part of the country, changes to signal plans, accompanied by military reorganizations, long-distance unit moves, and the use of tactical signal plans appeared to presage some larger, ~ndermed development.!

The SIGINT indicators were accompanied by similar indications in captured documents and rallier interrogations. Something was afoot, and U.S. military authorities in Saigon had divined it by early January 1968.. On the 7th, Westmoreland cabled the White House that
We think that the enemy made a major decision in September 1967 to launch an all·out effort to alter the course of the war ... the Winter·Spring campaign which began in late October is It calls ror offensive in nature and exhibits a disregard for casualties heretofore unseen. popular revolt against the GVN (GovernmentotSouth Vietnam].'

con~uoua military offensives by large and small units, and concurrent political efforts to stir up

But then, in one of the most infamous miscalculations in American military history, Westmoreland focused his attention on the border areas. There, he believed, was where the major blow would fall, with attacks in the cities serving primarily as a diversion to military assaults on the exposed periphery.

TOP SECRET tlMBItA

560

DOCID: 523682

REF ID:A523682
TOP SECRET UMBItA

His assessment was supported by SIGINT indicators of a major buildup in the Central • Highlands (witness the assault on Dak To and the significant NVA concentrations still in that area) and far to the north, in Quang Tri Province. One of his area commanders, General Fredrick Weyand, did predict on 10 January that the main assault would come in the urban areas. Weyand was in charge of III CTZ (III Corp Tactical Zone), which included Saigon, so his warnings seemed to have something to do with his own responsibilities. Westmoreland did not disagree with him; indeed, he made major changes in his defensive and offensive deployments to support Weyand's defense of the Saigon area. Still, Westmoreland continued to be concerned primarily about the riorth and west. ~
/'

KheSanh The largest diversion was at Khe Sanh. Located on the Khe Sanh Plateau in Quang Tri, the northernmost province of South Vietnam, Khe Sanh was a key point if one were to defend the area immediately south of the DMZ,·Located astride major transportation links in the interior, some distance from the coast, it bore a superficial resemblance to Dien Bien Phu. Beginning in November 1967, S1GINT began tracking the concentration of NVA units in the Khe Sanh area. Two divisions began moving from the North into South Vietnam, the first time two NVA divisions had ever moved simultaneously. This caught everyone's attention and clearly pointed toKhe Sanh as the major battleground for the upcoming offensive. Everyone believed it, most of all Westmoreland. He began building up forces at Khe Sanh in anticipation. Westmoreland believed that Khe Sanh was to be the Dien Bien Phu of the American war, but this time the result would be reversed." The assault on Khe Sanh began on 21 January and did not end until April. It was defended by the Marines, assisted by a small Marine SIGINT detachment ranging from fourteen to twenty-four men. The Marine detachment had HF Morse, LLVI, short-range direction finding (SRDF), and access to the entire SIGINT system. This included ARDF support from the Air Force (EC-47s from two different programs) and links to the NSG . detachment at Da Nang. Technical support was provided from USM-BOB at Pleiku, which was collection management authority for the northern area. In addition, the ARVN had a small S1GINT detachment at Khe Sanh which was duplicating what the Marines were doing. When this was discovered, the American and ARVN SlGIN'! units were physically combined, and the ARVN were employed as linguists to transcribe tapeS.7 . The amalgamation was successful, and Khe Sanh became one of the greatest SIGINT success stories ever. The ground unit intercepted NVA artillery firing orders in time for the Marines to get under cover. They also collected ground assault orders, and one participant estimated that S1GINT predicted some 90 percent of all ground assaults during
.the siege."

-'=5;=:J?)~
NOT ETOF

S

561

TOPSECRET UMBRA

DOClD: 523682
TOP SecRET UMBRA

REF lD:A523682

Hovering ARDF aircraft passed fixes on NVA units, and artillery fire from Khe Sanh ;. was mostly directed from this source. Under good conditions, the elapsed time between obtaining a fix and "shells-in-the-air" was about ten minutes. At one point ARDF located Hanoi's forward command element for the Khe Sanh action, and tactical air strikes virtually obliterated it. COMINT was either the sole source of targetting information (30 percent ofthe time) or was married with other sources to produce what 7th AF intelligence chief, Major General George Keegan, characterized as the "best target database in the history [ofthe war).'t9 Khe Sanh cost the North Vietnamese about 10,000 killed, as opposed to 500 Marines dead." The level of effort at Khe Sanh, the time period it encompassed. and the casualties the North Vietnamese were willing to endure indicate that it was a military objective that stood on its own. Otherwise. Giap would have broken off the encounter far earlier.

NSA and the Impending

Storm

By mid-January, NSA analysts were becoming concerned, by NVA communications trends; This agitation began to show up in items in the Southeast Asia SIGINTSummary. One after another. the indications ofa major assault bobbed to the surface. Never before had the indicators been so ubiquitous and unmistakable. A storm was about to break over South Vietnam.'! Then on 25 January, NSA published a baldly predictive report. Titled "Coordinated Vietnamese Communist Offensive Evidenced in South Vietnam," it began in . unambiguous language: During the past week,SrGINThas providedevidenceofa coordinatedattack to occurin the near future in several areas ofSouthVietnam. Whilethe bulkofSIGINTevidenceindicatesthe most critical areas to be in the northern half of the country,there is some additional evidencethat Communistunits in Nam Bomay also be involved. Themajortarget areas ofenemyoffensive operationsincludethe WesternHighlands,the coastalprovincesofMilitaryRegion(MR) 5, and the Khe Sanh and Hue aeeas. . Details were most profuse in the northern areas, while Nam Bo got relatively short shrift. This appears to have been because SIGINTwas more voluminous in the north. rather than an attempt to steer the reader toward the idea that the north would be the major objective. American SIGINTattention had always been focused on the northern provinces. where the largest concentration of American troops was. Moreover • like the party organization itself, communist communications structures in the south had always been looser and less susceptible to intercept and analysis. 12 The report was succeeded by a series of follow-ups providing additional details as they unfolded. The reports grabbed a lot of attention at MACV, and by all accounts, deeply influenced Westmoreland's counterassault strategy. He continued to beef up American

NTROL SYSTEMS JOINTLY HANDLE VIA ~_-*eT"REiLWEfJASABLE TOFOREIGN TOP SECRET UMSIb\

562

--------------

DOClD: 523682

REF lD:A523682
I UP SECReT tJ MIRA

units in the north and the Central Highlands. He also cabled the White House to recommend cancellation of the Tet truce which was scheduled to take effeet for the duration of the holidays. He got a reduction in the number of days, but the truce itself was in effect when the offensive began. According to political scientist James Wirtz, the failure of the Johnson administration to cancel the truce in the face of overwhelming evidence that a conflagration was imminent was one of the major miscalculations of the war .13 reports began referring to "N-day" and ftG-hour," never-before-seen terms which seemed to refer to attacks of unprecedented magnitude. On 28 January, an NSA product report detailed the N-day for the Central Highlands - it was 0300 (local) on 30 January. The commonality of terms throughout the country clearly pointed to massive, coordinated attacks. (This was the first of the NSA report series to be addressed to the White House.) MACV was ready, but the ARVN were not. They took the Tet holidays quite seriously, and when the blow Cell, were generally in a holiday mood and a holiday deployment. The White House, too, seemed unprepared for what was about to happen. There was no mood of crisis at 1600 Pennsylvania Avenue. a The Storm The difficulty of coordinating such an unprecedented offensive proved insurmountable Cor the NVA. Some units in the Central Highlands attacked a day early, on 29 January. Pleiku and Kontum City, as well as smaller provincial towns, were assaulted in the early morning hours, and the attackers were not finally thrown back until four days had passed." The blow fell on the restofthe country twenty-four hours later. The coastal areas were hammered with coordinated attacks on 30 January. The major provincial capital of Nha Trang was occupied by the NVA for several days before being ejected with heavy losses. Quang Tri City was also attacked, but the most devastating blow fell on Hue. On 30 January, ARDF showed major NVA units clustering outside the city, and the next day the forces stormed into the city. American Marines finally completed the ~etaking of Hue on 24 February after a bloody struggle that left more than 2,000 NV A dead. The North Vietnamese captured and executed many of the leading politicians in the city, a tactic which caused them so much ill will that they pointedly avoid~ it in 1975. More than 3,000 civilian corpses were exhumed after the battle. It was one of the sorriest episodes of the war.t8 In the III Corps area (including the Saigon environs), attacks opened on 31 January. The largest assaults were against Saigon and the Bien Hoe-Long Binh complex, but attacks also includedTay Ninh City, An Loc, and many others. Vietnamese Communist forces entered Cholon (the old Chinese quarter) from the west, and a sapper battalion
SIGINT product

HANDLE VIA TALENT KEYHOLE COMINT CONTROL SYSTEMSJOINTL Y NOT RELEASABLE TO FOREIGN NATIONALS

563

TOP SECRET YMBRA

--

--_.--------------------------------------------

DOCID: 523682
TOP SECRET tlMBRA

REF ID:A523682

assaulted the presidential palace and the American embassy. Though costly unsuccessful. these attacks produced camera footage that horrified a nation undoubtedly produced the turning point in American attitudes that Giap was after. 17

and and

The Assessments The postmortems began even before the! last NV A troops were routed from Hue and . .. Saigon. CIA put together a study group. at PFIAB request. which included representatives from NSA and all the other Washington area agencies. Maxwell Taylor; the new PFIAB chair, requested that the DCI "ascertain to what extent, if any, our intelligence services and those of our allies were at fault in'failing to alert our military and political leaders of the impending large-scale attack on the cities and towns of South · t »18 -, ,----------, V ie nam. . Withheld The resulting
... larger

study stated that

I

E.O. 13526, section 1.4(c)
Considerable numbers

I
probably on a ofl in many areas of South and numerous than

communicationsin1;elligenc:e

was able to provide clear warning

that attacks,

from public release Pub. L. 86-36

I

scale than ever before, were in the offing.

I

I

enemy meseagea were read. These messages appeared to impending attacks,

Vietnam.

They included references Moreover,

more widespread

seen before. planning

they indicated

a sense of urgency, along with an emphasis ....

on thorough however,

and secrecy not previously

seen in such communications

The indicators,

were not sufficient to predict the exact timing of the attack.Ii

Aside from the last statement (invalidated by the N-day, G-hour warning that NSA issued on 28 January), the DCI assessment seemed pretty accurate. COMINT did indeed serve as i the main predictive element in the intelligence puzzle preceding Tet. The sense of foreboding that cryptologists felt throughout January 1968 was transferred to MACV and Westmoreland's staff. . That 'was about as good a prediction as could have been advanced. There was no precedent for the scope and ferocity of Tet, because it was a unique event in the war. But the military authorities in Saigon were as ready as they could have been under the circumstances. The sense of urgency did not appear to have penetrated the White House. This was unusual in Lyndon Johnson's administration. He and his staff were avid consumers. of intelligence in general and SIGINTin particular. But they did not seem to have been ready, What SIGINTwas criticized for was not the fault of the cryptologists. Owing to the . concentration of SIGlNTresources on the central and northern parts of the country, and to the historical ineffectiveness of SIGINT in the south, the product reporting drew the customer toward the northern and border areas. There were fewer SlGINTindicators in the south, and SlGINTcannot report what it does riot hear.

HANDLE

YHOLECO

YSTEMS JOINTLY NATIONALS

BL
TOP S!CR!T U MBItA

564

DOCID: 523682

REF ID:A523682
TOP SECRET UMBRA

What occurred was a phenomenon that became famous after the Battle of the Bulge in World War II. SIGINT had only part of the picture, and intelligence analysts relied too heavily on the single source. In hindsight, it is clear that too little attempt was made to flesh out the rest of the picture through rallier interrogations; captured documents, and the like. SIGINTbecamethe victim of its own success. The lesson was a moral in all-source analysis. In a far greater sense, however, it did not really matter. Westmoreland was ready for the major attacks, and he successfully countered them. The NVA lost 30,000 dead, an. immense military blow from which it recovered very slowly. The structure of the VC insurgency in the south was shattered forever. The White House, however, had the job of countering the political blows. It did a poor job ofit, and the sense of panic and disorganization was palpable.

THE WAR IS VIETNAMIZED
In the previous adminiatration, we Americanized the war; in this administration, Vietnamizing the search for peace... , Richard Nixon,1969 we are

The President

Pulls Out

Following Tet, the Pentagon decided that the time to win the war was now or never. General Wheeler, chairman of the JCS, sent Johnson a request for 206,000 more troops. This demand created a crisis within the Johnson administration's inner circle. It would require the call-up of reserves and would place the American people on an all or nothing track in Southeast Asia." Clark Clifford, the new secretary of defense, suggested that he form a group which had become known as the "Wise Men," long-time advisors to Democratic presidents. Reporting in March, ten out of the fourteen recommended against an increase in troop strength, and many felt it was time to begin a gradual disengagement. 21 The Wheeler troop demands, and the resulting debates within the Johnson administration; leaked to the press. The story played all through March, and toward the end of the month Robert Kennedy announced his candidacy for president. Johnson announced that he would go on television March 31 to make an announcement.f In a historic speech delivered to television viewers from the Oval Office, Johnson announced a halt to the bombing above the 20th parallel and the beginning of formal negotiations with the North Vietnamese. Long-time Democratic stalwart Averell

HANDL
N

OLE ABLETOF

TROLSYSTEMSJOINTLY

565

TO,. SECR!T UMBRA

DOCID: 523682
TOP SECRET l:JMBR:A

REF ID:A523682

Harriman was named to head the negotiating team. And in a surprise. announcement at the end of the speech, the president stated that he would not run again in 1968.23 For Americans, the war was only half over from a chronological standpoint, and more American soldiers were killed after Tet than before it. But the 31 March speech began a new phase. The United States was beginning a military withdrawal and would henceforth rely on negotiations to reach a peace accord." Vietnamization Almost immediately, the JCS set to work on a plan to gradually turn over military operations to the ARVN. When President Nixon took over, with the avowed goal of Vietnamizing the war, the JCS was already moving in that direction. A formal plan to support Nixon's version of Vietnamization waS firs~ drafted in late 1969, following his Vietnamization speech. Called JCSM 42-70, it contained a cryptologic tab written by NSA in collaboration with the SCAs. It was coordinated with the Vietnamese SlOlN'!' service (then called the SSTH, or Special security Technical Branch), but it was never offered for the approval or disapproval of the South Vietnamese government.~ NSA planned to turn over much of the SIGINT mission to the S5TB. In order to do this, it would be necessary to both augment its numbers and increase its competence. It had a long-range goal: "The RVNAF eventually will be capable of providing COMINT in satisfaction of its military requirements generated by the ground war in RVN. ,,28 At the time, S5TB consisted of about 1,000 people, three fixed sites (Saigon, Can Tho, and Da Nang), a small ARDF effort using U-6s, and a four-station DF net. It had no ELINT mission. It had plans for a major. expansion of its tactical capability, modeled after the ASA DSU concept, but as yet only one of the ten planned units was in existence."

H~L&~.TLY

ABLET

LS

TOP SECRET UMBRA

566

DOClD: 523682

REF lD:A523682
'FOP SECRET UMBRA

Richard M. Nixon

NT KEYHOLE COMINT CONTROL SYSTEMSJ NOT RELEASABLE

INTLY

567

JOp SE(;RE:r UMB!tA

DocrD: 523682
TOP SECRET tJMBRA

REF ID:A523682

In 1970, at the inception of the Vietnamization program, Admiral Gayler characterized the organization as "fairly effective" but In need of certain managerial and technical improvements. The ARDF effort was "considerably less than satisfactory" and the medium-range direction rmding (MRDF) net was "not accurate." Still, he concluded that "it is considered feasible for RVNAF to be able within the next three years to cover all Vietnamese Communist communications .... " Gayler felt the job was difficult but doable. %8 The South Vietnamese SIGINT system had been headed by General Pham Van Nhon since 1963. Nhon was considered by CIA to be a strong point, especially in the area of security. He ran a "tight ship," according to a CIA evaluation, and as a result, the SIGINT organization was a bulwark of security, especially when compared with the porous South Vietnamese government. Nhon reported directly to the J7 element of the ARVN Joint General Staff. COMINT was considered to be highly sensitive, and SIGINT matters would sometimes wind up in President Thieu's office.29 . To support the Vietnamese military structure as NSA understood it in 1970, SSTB strength would have to climb from about 1,000 to approximately 1,500 bodies. It would add one fixed site at Pleiku, collocated with the ASA unit there. This would bring the SSTB fixed sites to a total of four: Saigon, Can Tho, Da Nang, and Pleiku. In places like Can Tho, SSTB operators would sit side by side with ASA operators in order to enhance training." NSA maintained overall control of Vietnamization and established the training plan. NSA instructors taiight some of the higher-level training courses, but the execution of the . plan was decentralized. ASA and AFSS both got major training responsibilities." ASA was given responsibility for training the SSTB ground COMINT effort, including the ten tactical units. A team of advisors was attached to each of the units, called DARR (Division) and CARR (Corps) Advisory Radio Research units.32 Regarding ARDF, NSA decided to turn over twenty EC-47ARDF aircraft to the ARVN. Thus, to AFSS would fall the responsibility for ARDF training.33 Vietnamese SIGINi' communications security had to be improved. 'NSA initiated Project LACE BARK, which would upgrade crypto gear. The new COMINT network would internet the four fixed sites, EC-47 unit, and the tactical units." This was part of a larger project to upgrade South Vietnamese military communications in general. NSA intended to get rid of the obsolete Python tape system. The KL-7 off-line crypto equipment would be provided to RVNAF crypto nets. M-209s, of World War II vintage, affording minimal security, would be provided to the National Military Police, while NESTOR secure voice equipment would be provided to selected RVNAF combat units." .

NTROL SYSTEMS JOINTLY

ALS

TOP SECRET tlMBRA

568

Tap SEEAR WUIA '

I E.O. 13526, section 1.4(d) I

Withheld from public release Pub. L. 86-36

VletllUiMM

operakM'S

ill

walnin,

~~y H

N

TOroR1!IGN~

c

,

589

ropsFcPfIlfMRRA

g
o

DOCID:

523682
TOP SECRET tlMBAA

REF .ID:A523682

Nixon did not wait to see the results of the Vietnamization program. InMarch 1970 he announced a phased withdrawal of 150,000 U.S. troops over the course of the next year, despite the anguished protests of General Abrams, who had succeeded Westmoreland at MACV. The next year the president ordered the removal of another 100,000, and this continued until, by the beginning of the 1972 Easter Offensive, there were only 95,000 American troops in Vietnam, of whom only 6,000 were combat troops." This rapid withdrawal schedule was not reflected in the SIGINT plan. The 1970 cryptologic Vietnamization plan showed a phasedown from 8,500 cryptologic spaces in Vietnam in 1970, to 6.654 in 1973. The secretary of defense commented to the JCS that the cryptologic le~els did not seem in concert with the president's ideas about the pace of Vietnamization. It became characteristic of the cryptologic posture that it trailed rather badly behind the removal of combat troops. This undoubtedly reflected the long lead time required to get SSTB up to speed, in people, equipment, and expertise. Despite Admiral Gayler's initial guarded optimism, NSA and the SCA's all expressed ambivalence about the long-range capability ofSSTB to do thejob.s7 American Special Operations The slowness of the cryptologists to depart was reflected in the continuing vitality of American SIGINT operations in the theater. One manifestation was SIGINT support for Task Force Alpha. [E.O.

13526, section 1.4(c) [

i

Task Force Alpha, or TF A, was organized by 7th AF in the spring of 1968 and positioned atl ~ Its mission was to gather NVA infiltration data from such sources as IGLOO WHITE (the electronic sensor system in Laos) and SIGINT. A primary source was infiltration communications collected by the RC-135 inthe Gulf of Tonkin. This information was downlinked in near-real-time to a special USAFSS unit collocated with TFA. This unit also had available SIGINT collected by EC-47s from the ARDF unit, as well as information from USM-7 at Rainasun Station.38 Task Force Alpha. with its unexcelled access to the key intelligence systems targetted on the Trail network, was very successful. In the summer of 1968 it even directed aerial bombardment of the Trail. Although this authority was pulled back to Tan Son Nhut at the end of the summer. the long-range effect on the cryptologic community in the theater was considerable. It began a shift of cryptologic operations into Thailand and an increased focus on using SIGINT to try to choke off infiltration. rather than on supporting American ground combat forces. It was in line with the direction that the war was going. 39

Withheld from public release Pub. L. 86-36

iiAN~LE =:=~:::~==~='''''Ol~
VIA

TOP SECRETUMBItA

570

DOCID: 523682

REF ID:A523682
16P SECKff
UUBRA

Another special operation was COLLEGE EYE, an EC-121 that flew out of Korat, Thailand. COLLEGE EYE was an airborne radar station that was used to extend American radar coverage farther north. It was also used as a communications relay so that Monkey Mountain could still talk with its aircraft outside VHF communications range." Aboard ~e COLLEGE EYE aircraft were four SIGINT positions, codenamed RIVET GYM. Manned by USAFSS, the positions were used for COMINT tactical voice intercept. SIG(NT was passed directly to the on-board controller, who correlated it with the information that he got off his radar scope. Thus he knew not only where the North Vietnamese fIghters were, but what they were saying to their ground controller." In the Gulf, the Navy was going its own way on SIGINT. The larger vessels had small afloat detachments for direct SrGINT support. Among other things, they all copied North Vietnamese Air Defense nets, both radar tracking and VHF air/ground voice, to provide support to Task Force 77 air operations. At any given time there were four or five such detachments, each operating independently.Y . In 1969 the detachments were internetted under a project called CHARGER HORSE: Through the net they began exchanging information. ThIs allowed them to divide up the responsibility for air defense monitoring so that they weren't all copying the same nets, and to intercept lower level NVA air defense communications to reduce the lag time by several minutes. The information, which included both air defense tracking (considered sanitizable) and VHF voice (not sanitisablel, was exchanged over the Naval Tactical Data System. A second naval operation was called FACTOR, which was an attempt to use stop North Vietnamese maritime infiltration. It had a long history behind it ..
SIGINT

to

FACTOR's story stretched back to 1962. In November of that year NSG first isolated a communications net that supported NVN maritime infiltration. The North Vietnamese called it Group 125, and its mission was to load war material aboard steel-hulled trawlers and run them down the coast to South Vietnam. The trawlers would stand off in international waters until they felt they were not being watched, then dart into the coast to unload the goods.

At the time the cryptologic community was simply following the operation in SIGINT; no attempt was being made to tip off any counterinfiltration operations. But the longer they listened, the less activity they intercepted, and by July 1966 they had completely lost continuity on Group 125 communications. NSA· suspected that the vessels had been diverted to other operations, particularly escorting combat vessels to and from China.

HANDLE VIA

CONTROL SYSTEMS Jo,INTL Y ~Nl.U-ftf:l"~I\SABLE TO FOREIG

571

/

DOClD:

523682
l'6' seCRETYMIRA

REF

rn :A5236'82

After the 1968 bombing halt, Group 125 went back to maritime infiltration, and by November 1968 NSA had again isolated communications from a net that eventually proved to be continuity of Group 125. By 1970 maritime infiltration represented a significant problem, and NSA decided to see what it could do about designing a SIGINT tipoff system. A special position was designed under a new project, called FACTOR. The equipment maximized intercept of ground waves from the frequency range used by the trawlers. the equipment. was sent to Cam Ranh Bay. and from there it was loaded.aboard two P-3s being used for "Market Time," an interdiction operation. Success was immediate, and the P-3s intercepted trawler communications on their first mission. NSA designed a tip-off system to flash the.intercepts to Market Time operations. A CIA assessment later in the year waxed poetic about the success that Market Time was having,at least partly a result ofimproved SIGINT support." The Cambodian Incursion In the long story oCtIle Vietnam War. one military foray stands virtually alone in the extent and consequences of its. failure. The Cambodian incursion was an unmitigated disaster. The seeds of that failure were in the unstable political situation in Cambodia. The Cambodian leader, Prince Norodom Sihanouk, had lacked the political and military will to keep out NVA forces, which used the eastern section of his country virtually at will as a logistics and infiltration base. In March 1970. his chief lieutenant, General Lon Nol, and a coterie of his Army supporters overthrew him.44. While all this was going on, Richard Nixon was considering what to do about NVA domination of sanctuary areas in Cambodia. In February 1970 he authorized a secret bombing campaign which would target NVA base areas in Cambodia.45 Although supposedly secret, the bombing became known to many American correspondents in Vie!;nam. In Maya New York Times reporter, William Beecher, officially revealed it. Nixon'S reaction was rage, and he directed that the source of the "leak" be discovered. He ordered wiretaps on suspected journalists and eventually on White House staff members. Thus began a pattern of White House paranoia which led eventually to Watergate. It started with Cambodia. The pro-Western Lon Nol was no sooner in power than he launched his own campaign to evict the NVA and VC from Cambodian soil, and this was followed by a plea for aid from abroad." The White House responded almost immediately, announcing in late. April that the U.S. would provide military supplies and advisors to the new Cambodian government."

HAND~'~
lOP SECRElI;IMBRA-

BLE

NAl.S

572

bOCID: 523682·

REF ID:A523682
TOP S!CR!T UMBRA

On 30 April Nixon announced to a stunned American public that American troops had crossed the border into Cambodia in hot pursuit of NVA forces. The press denounced the move as a virtual renunciation of peace talks begun earlier by President Johnson. Demonstrations erupted, and on 4 May panicked National Guardsmen fired into a group of students at Kent State University." The incursion took three directions: one in the Central Highlands (Binh Tay, Peace in the West), another in the central border area around the Fishhook and the Parrot's Beak (Toan Thong, Total Victory), and the third in the Delta area (Cuu Long, Mekong River). American forces were heavily involved in the first two, but the only support rendered to ARVN in the Delta was riverine."
Q,l1,C)

= .::
0
"0

against Cambodia was good. disparate group of sites ranging from ASA l. ~ .~ ...:l sites at Ramasun Station and Pleiku to USM-7 at :Q.::i Ramasun was the principal in-theater processing site.so .
~ I

"'~ Q,l1,C)

The

SIGINT capability

I

I

~oo

I

I Collection was done from a

-= -=
~

I

I.

==

c.~

,

Unfortunately, the planning for the incursion excluded the SIGINT system,' allegedly for security reasons. The first word came to ASA lieutenant colonel James Freeze, commander ofASA's 303rd RRB at Long Binh. Freeze was tipped off on 28 April only two days before the operation began, by the G2 oUI Field Force Vietnam (FFV):51 This began a frantic few days of planning and assembling resources. Ultimately, an extensive network of ASADStJ s deployed, including sixteen intercept teams and various higher-level organizations. Low-level .voiceintercept was of greatest value, but Morse proved almost worthless. ASA instituted a complicated courier service which included hehcopters to get the traffic back to Quan Loi, where it could be forwarded electrically to Bien Hoa. In June, ASA deployed a team (with the interesting title RATRACE) to Quan Loi to process the take and return it to the units in Cambodia. This eliminated the requirement to get the material back to Bien Hoa.~2

CJ.

I

I

,

The most famous (or infamous) event of the incursion was the attempt to "get COSVN." Long known as the Central Office, South Vietnam, COSVN served as the VCINVA headquarters in the south. Situated just across the border from Tay Ninh province, its location was fixed daily by ARDF. It moved occasionally, usually to get out of the way of B-52 strikes (which, as we know, were predicted with great accuracy by the NVA intelligence people), and repeated air strikes over the years had never succeeded in doing any effective damage." . Creighton Abrams wanted to "get,COSVN." He had the ARDF fixes, and now he had the authorization to invade Cambodia. The timing seemed right. Whether the attack on COSVN was a primary objective of the incursion or an afterthought is no longer clear. But the press got hold of the COSVN story, and it became common knowledge to the American

HANDLE VIA~T~A~:~::L~E~C~O~MffiINT~::~:~TEMS NOTREL TlONALS

JOINTLY

573

TOil SECRE'F ijMBIbf<

DOCID: 523682
lOP SECRET YMBRA

REF ID:A523682

people. At that point, pressure from MACV to locate and overrun (or at least bomb) COSVN became considerable." SIGINT was mobilized. Ground positions placed COSVN communications on east-iron coverage. ARI)F flights over Tay Ninh and eastern-Cambodia darkened the skies. But the military system moved too slowly. COSVN was able to evade every B-52 strike and every ground maneuver. Abrams complained that he could have gotten COSVN had he not been forced to use the slow-moving ARVN5th Division instead of an American unit. ss But the fact was that MACV still did not fully understand the vagaries of SlGlNT. SIGINT.advisorsexplained again and again that they were only fixing an antenna and that· the transmitter, to say nothing of the headquarters itself, could be miles away. Moreover, the military targetting system seemed inflexible - SIGINT reports that COSVN had pulled up stakes from location A and was now at location B were not enough to get a strike cancelled or diverted. American bombs tore up miles of jungle, and ARVN troops floundered through a trackless quagmire of Cambodia in pursuit of COSVN. They never caught up with the headquarters, which moved safely to central C~mbodia ahead of the advancing Allies. 5&
-,

The best they ever did was to capture supplies. In early May, an ARDF fix located a base area of COSVN known as "The City" because of the extensive logistics depot suspected to exist there. Acting on this intelligence, an ARVN unit strut:k the complex and captured a vast store of material. It was enough to set back NVA offensive plans for a definable period oftime. But it wasn't COSVN.57 The incursion was a limited military success. American and ARVN troops proved
capable of capturing any territory that they really wanted. But the long-range results

were disastrous. The U.S.lARVN forces drove the NVA deep into Cambodia; where the NVA set up shop. By mid-May the major Cambodian provincial capital (and choke point on the Mekong) of Stung Treng fell, and within a month the NVA held every province in northeast Cambodia. Using this as a base of operations, their Khmer Rouge communist allies began an offensive against the Lon Nol government which ultimately led to the fall of Phnom Penh in April 1975, and began the great Pol Pot reign of terror. Few operations in American military history had such dismal consequences.

OMINT CONTROL SYSTEMS JOINTLY OTRELEASABLETOFO
/

TOP SECREf tlMSRA

574

Tlte SStb [lIJplly

DiYi&ioD II••••

into C••• bodi •• Ma,. lWlO.

co
C"l

N

\0 N
It) HANDLE VIA TALE TROL SYSTEMS JOINTLY

8LE TO FORl!JG.

o
H

676

l'eP§E~RIi~"

•• eRo

o
o

U

. DOCID: 523682
TOP SECRET U,MBRA

REF ID:A523682

LamSon 719 By early 1971 Creighton Abrams was becoming concerned about evidence pointing to a major NVA offensive during the 1972 dry season. The administration, as well, was concerned about the political consequences of a possible ARVN defeat so close to the November 1972 elections. Thus originated Lam Son 719, an attempt to invade Laos and disrupt the NVA logistics system that was being used to funnel record numbers of troops and supplies into South Vietnam. M As the Americans had correctly judged NVA plans, so too the NVA intelligence system sniffed out the American and ARVN plans for a preemptive strike. As early as October 1971, NSA reported that NVA communications were showing a heightened concern for the area that the ARVN planned to invade. Through November and December, NSA reporting showed increased NVA defensive measures along the Trail. Moreover, SIGINT was showing increased infiltration into the areas targetted for invasion. 59 Lam Son 719 was another disaster. The ARVN troops fought through to their major objective of Tchepone in Laos, but the going had been very tough and the troops were exhausted. Moreover, there was nothing remaining in Tchepone for them to take possession of. In the end they simply retreated. The retreat became a rout as large-scale NVA forces (shown by SIGINT to be massing for a counterattack) descended on unprotected elements of the retreating army.so
SIGJNT showed once again how flexible the Trail system had become. As the NVA lost sections of the Trail, .it simply diverted shipments to other sections not under ARVN control. In the end, Lam Son 719 scarcely interrupted the flow, and the NVA spring offensive of 1972 went off with hardly a hitch.

The Son Tay. Raid Son Tay, the infamous attempt to rescue American POWs, rescued no one. As a military operation, however, and as a way to set up SIGINT support, it was exemplary. Planning for the 1970 raid began in April. The SIGINT system was brought into the picture in August, which gave it time to react (as opposed to the Cambodian incursion, which did not). As briefed to a handful of cryptologists who were initially cleared for the operation, it would involve a wave of helicopters flying at low level to the prison camp at Son Tay, twenty miles northwest of Hanoi. It would also involve the participation of a diversionary attack by a naval forci in the Gulf, along with combat air patrols, fire suppression aircraft. and various logistics flights." .

HANDLE VIA
LEASABLE TO FO

NTROL SYSTEMS JOINTLY
LS

Tap

SECRET

"MBRA

576

l6,

3!Ck£1 UMBRA

Triple Canopy Ju.ngle UAder_r~ N <Xl \0
(Y)

t'irw. dur-inCthe Cambodian

incursion

N Ltl

H U

C

577

T"e:' Stc«trdNi&""

o
C

DOClD: 523682
lOP SECRIiT UMIAA

REF lD:A523682
Withheld from public release Pub. L. 86-36

. Brigadier General Manor, the overall operation commander, requested that SIGINT give him the best ingress and egress routes from Takhli AB, Thailand (whence the raiders came), and apprise him of all NV A capabilities to interfere with the operation. The NSA representative to Manor's staff was Lieutenant Colonel the chief of Pacific Air Defense Analysis Facility (PADAF) in Ha wail. PADAF's job was to do just that sort of analysis, arid people wrote a series of reports detailing to Manor the precise route that should be followed. Working with NSA analysts, people concluded that if Manor used their suggested route and went in at night, the NVA would have no capability to interfere. and his people were right, and the raiders entered and exited virtually undeteeted.62

I

~

I

I

I

I

I

I

put together a complex network for SIGlNT support. Working with people he could not clear for the project, he assembled RC-135 collection, COLLEGE EYE assets, and monitoring support from units all over the Pacific theater. He took extraordinary OPSEC measures. His biggest problem was that the RC-135 mission would have to fly at night, at a time when SIGINT reconnaissance missions never flew in the Gulf. He solved that by scheduling several nighttime missions in the weeks before the raid so that the North Vietnamese would get used to seeing them there.6S himself flew to Da Nang to watch the operation unfold. He had an Opscomm link that began at Oa Nang and was routed through NSA and ultimately to the Pentagon . • On the other end of the link was Milton Zaslow, the NSA representative who kept the JCS apprised of the raid's progress as reflected in SIGlNT.64 As the raid unfolded, it was being monitored by a select group in the National Military Command Center headed by the secretary of defense, chairman of the JCS, and certain three- and four-star officers. As Zaslow was briefing the group on NSA activity in support . of the raid, an officer broke into the room and announced that General Manor had declared a MIG Alert. Everyone turned to Zaslow, who had just stated that there was no threat fromMIGs. Zaslow stood his ground. "No MIGs," he said. He spent a very uncomfortable five minutes as the assembled Pentagon generals stared at him, wondering how he could be so sure. Zaslow knew that intensive SIGlNT analysis had identified all North Vietnamese night-qualified MIG pilots and at what airfield they were spending the night. Moreover, Zaslow's communications with [were the fastest at the Pentagon, and was reporting no MIGs, based on continuous monitoring of those airfields. Zaslow stuck to his story. A few minutes later another courier burst into the room crying, "Cancel MIG alert." Zaslow had been vindicated, and everyone.breathed easier.65

./

I

I

I

I

I

I

IlANDt~~~

TOP S!~RET tlMBAA

578

DOClD: 523682

REF lD:A523682
TOP SECRET UMBRA

NSA's assessment was confirmed completely, and the SIGCNT system worked as well as it ever had. No one ever found out for sure why the prisoners had been moved before the raid, but one HUMINT report said that about a month before the raid a Caucasian journalist had visited the camp and stated that the prisoners were moved immediately afterwards. Perhaps the North Vietnamese were "spooked" by the visit.6e . The Easter Offensive Lam Son 719 did little to slow down NVA plans for a great spring'offensive in 1972. NSA infiltration figures from the Vinh Window showed an unprecedented flow of supplies and a massing of forces in the border areas such as had never before been seen.. For the first time, intelligence showed NVA tank concentrations in the south, pointing to the employment of conventional forces in an attempt to overthrow the Thieu regime. er As the classic SIGINT indicators mounted, NSA reporting became more and more specific about the timing and objectives. When, at the end of March, the offensive finally broke, it had been more than seven months in the offing. This only increased its fury. The NVA concentrated on the areas thought vulnerable prior to Tet 1968 - the Central Highlands, Quang Tri Province, and the border areas near Cambodia in MRS. There was no. comparable assault on the cities, no appeal for mass revolution. This was a conventional attack with tanks and artillery. The ARVNbarely held, but in the end it looked like another Pyrrhic victory for the NVA. They lost 50,000 troops, almost as many as did the United States during the entire war. The attack failed all around.sa Nonetheless, it appears to have fallen on an unprepared Nixon administration. Several knowledgeable historians claimed afterwards that it was an intelligence failure. George Herring was extreme, stating that "American intelligence completely misjudged the timing, magnitude, and location of the invasion." Seymour Hersh, who is usually right, wrote that the offensive was so long delayed that the White House was focused on other things, and that Nixon claimed that the Pentagon withheld information from them; There is no SIGINT evidence to support the "surprise" hypothesis- perhaps there is other
evidence.f"

TEABALL One result of the Easter Offensive was the resumption of the air war. In early May 1972, Nixon ordered the bombing of Hanoi and Haiphong in an operation the Pentagon called Linebacker. Immediately, waves of B·529 roared over the North. It was the most intensive air bombardment of the war .70 :But the operation proved costly. The North Vietnamese adopted a new defensive strategy. Eschewing SAMs (which had proved ineffective and fratricidal in the face of

ROL SYSTEMS JOINTLY ONALS .

579

TOP

SECRET UMBRA

.DOCID: 523682
TOP SECRET tJMBItA

REF ID:A523682

American countermeasures), they launched pairs ofMIGs. The MIG pilots would home in on one of the flights of B-52s, would execute a single high-speed pass, launch missiles, and turn tail for home. By the first of July, the U.S. had already lost eighteenaircraIt to such tactics, with "only" twenty-four MIGs destroyed. The virtually one-to-one kill ratio had General Vogt, commander of 7th Air Force,looking for new tactics." It had long been the desire of the cryptologic community to pass MIG warnings directly to threatened pilots. The Air Force Security Service had set up a variety of operations over .the years, but all the warnings had had to pass through the filter of TACCINS, unless extraordinary circumstances intervened. Every request to pass warnings directly to operations people had encountered th~ implacability of the director of Air Force intelligence, General Keegan. . In 1967, Security Service had informally suggested a mechanism for passing warnings directly to operations, but Keegan would not hear of "raw SIGINT" going to a pilot. Two years later, the NSA representative to the Pentagon proposed a similar operation, only to have the idea die in staffing channels, once again a victim of turf protection, It appeared that direct warnings would never get through the bureaucratic thicket and that the Air Force would not get anything similar to what the Army already had from ARDF - tactical warnings passed directly to operations people." The Linebacker losses proved the undoing of the intelligence empire. In early July, General Vogt appealed to General Ryan, the Air Force chief of staff, for a new approach to the intelligence warning system. Ryan called Admiral Gayler, who already had the solution in his pocket. (It was the same solution that had died in staff'mg a year earlier.) He sent a team of SIGINT experts to Saigon, headed by Delmar Lang, who had been instrumental in devising a solution to a similar problem during the Korean War (see p.
49).

Lang knew that Vietnamese voice communications revealed the takeoff' of the MIGs and that the North Vietnamese controller revealed which B-52 sortie would be targetted (the so-called "Queen for a Day," after a 19505 radio quiz show ofthe same name). He also knew that the SIGINT U-2, called the OLYMPIC TORCH, was intercepting those communications and that the intercept operators were sitting at the 6908 SS at Nakhon Phanom (NKP) AFB in Thailand. He recommended that the takeoff and tar getting information be passed to a collocated 7th Air Force controller, who would alert the Air Force defensive patrol in the Gulf. When the MIGs arrived, theoretically the F-4s would be waiting for them." He caned the operation "TEABALL." Vogtestablished a new Weapons Control Center (WCC)in a van atNKP, right next to the vans housing the downlink for the OLYMPIC TORCH operations. S~curity Service operators had a hotIine from their intercept van to the WCC, where the information would be melded with other sources. In .practice, SIGINT was virtually the only source of information, and AFSSlinguists populated the WCC, sometimes passing information to

TOP SEeRET !::IMBRA

580

DOClD: 523682

REF ID:A523682
TOP SECR!T tJMIRA

the pilots when weapons controllers were not available. involvement in the air war that the SIGINTers hadwanted."

It was the kind of direct

The TEABALL operation got off to a slow start because of commu~ic~tions problems and lack of manning on the 7th AF side. But on 28 August, eighteen days after being declared operational, TEABALL got its first MIG kill. By the time Linebacker was cancelled on 15 October, American pilots had shot down nineteen MlGs while losing only five of their own. TEABALL was given credit for helping to vector U.S. pilots on thirteen of those nineteen kills." TEABALL became caught up in interservice rivalry. The Navy had its own control operation in the Gulf, a ground-coritrolled intercept (Gel) ship known as Red Crown (for its VHF callsign). Red Crown was supported by NSG afloat detachments, which claimed to be able to interceptMIG voice tracking on a more timely basis. Some of the MlG CAP operations got tangled up in jurisdictional disputes between the wee and Red Crown, and it was not clear which could provide the more timely warning information. The dispute was untangled in a joint 7th Air Force - TF 77 meeting in mid-September, at which a compromise over control offighter CAP in the Gulf was worked out. The WCC/TEABALL I operation relinquished control authority in certain situations, but not in ethers." When, on 13 December 1972, Le Due Tho, the North Vietriamese negotiator, walked out of the peace talks, Nixon turned to the B-52 operation again. Thi.s time the raids, under the name Linebacker II, were not confronted with MIGs, which had been chastened by the new American tactics. The North Vietnamese went back to using the less-thaneffective SAMs. One B-52 was lost, but it has never been shown that it was a SAM kil1. Lacking MIGs, TEABALL wasn't needed." Linebacker II was the most intensive aerial bombardment of the war. More than 36,000 tons of bombs were dropped, and though Americian pilots went to extraordinary lengths to avoid population centers, as many as 1,600 civilians may have died. Nixon and Kissinger claimed that it forced Le Due Tho to return to the negotiating table. Soon thereafter the truce agreement was signed." The U.S. Moves out of Vietnam Thecryptologists were still very active in Vietnam. There had been some changing around of people and positions; as some cryptologic operations' got bigger, others got smaller. One technique that prospered late in the war was remoting. After the early trialsen BlackWidow Mountain and others (see p. 536), NSA brought in permanent gear in a remoting system called EXPLORER.EXPLORERI, consisting of four VHF receivers, was placed on a hill near Phu Bai in June 1970. A year later it was destroyed to prevent capture and was succeeded by EXPLORER III, .destreyed under similar circumstances.

HANDLE

E

OLSYSTEMS JOINTLY

ABLETOFO. 581 lOP SECRET UMBRA

OOClO: 523682
lOP SECRET UMBRA. I E.O. 13526, section 1.4(c)
EXPLORER II

REF lO:A523682
Withheld from public release Pub. L. 86-36

was located on a remote hilltop I I It was controlled iby USM-604 at Pleiku and was withdrawn when U.S. forces left Vietnam in December 1972.79.

The last such operation in Southeast Asia was called SARACEN. Established in late 1972, SARACEN provided unique VHF collection primarily on GDRS communications. The remote location, on a hill south ofl Iwas almost inaccessible except by helicopter, and the security situation remained precarious throughout its existence, sitting as it did virtually overlooking the Ho Chi Minh Trail. Its collection station was the AFSS site at NKP, which also collected GDRS communications from the OLYMPIC TORCH U-2, until U.S. cryptologists were withdrawn

I

I

As diplomatic negotiations proceeded, the Nixon administration stepped up the pace of troop withdrawal. Status reports on cryptologic Vietnamization indicated that the SSTB was not yet ready to take on the load. The organization lacked people, needed more training in processing and reporting, and was short on good communications. NSA hurried the provision of communications and stepped up the training pace. NSA offered ten more EC-47 ARDF aircraft to help SSTB cope with the burden of supporting ARVN operations." In the fallof 1972, Nixon announced that American troops would be out of Vietnam by year's end. ASA operations were moved to Ramasun Station, while AFSS collection and processing were hastily removed from Da Nang to NKP, to be collocated with 7th Air Force command and control facilities. AFSS ARDF operations moved to Ubon and NKP, while the Army flight section transferred tol l The Dancer Vietnamese linguist operation moved to NKP, to provide assistance to 6908th linguists at the downlink end of the OLYMPIC TORCH.62 As with the negotiations in Korea prior to the 1953 armistice, NSA provided SIGINT support to the Kissinger-Le Duc Tho jace talks. N~AI I had been reading South Vietnamese diplomatic traffic throughout the war. The reactions of the Thieu government to the Paris peace talks were passed daily to the White House and influenced Kissinger's position on countless issues throughout." .

I

_

I

I

The cease-fire that took effect in February 1973 required that all U.S. military people be out of the country. The cryptologic withdrawal that had begun with the Vietnamization program proceeded very quickly, and by the implementation of the cease-fire the only American cryptologists left in the country were covert.

MIN
NOT RE~L~~r6t:;E"l'·O FOREIGN NAT

YSTEMSJOINTLY

TOP SECRET UMBRA

582

DOCID: 523682

REF ID:A523682
'fOp SECRET UMBRA

The Summing Up Vietnam was a rude education for the American military. It was also an education for" eryptologists. Cryptologists had forgotten how to do direct tactical support in an effective manner. It took the cryptologic system most ofthe war to relearn the lessons of World War II and the Korean War. The cryptologic community paid a high price for dismantling its taetieal support system. Meanwhile, a skeptical military, by then unlettered in cryptology, tried to pry the SIGINT system into pieces and fragment the effort. The struggle for control of cryptologic assets lasted the entire war. and the effects remained for years afterward. The SIGINT system was kept generally intact (with some significant exceptions). but it was not the same one that entered the war. Noone truly knowledgeable of U.S. intelligence could quarrel with the value of SIGlNT. It became the number one source of targetting information. An Air Force historian estimated that SIGINT provided 55 percent of all targetting information in Vietaam." It was the best method of predicting NVA offensives. Beginning with the VC offensive at Ap Bac in 1963 (made famous by Neil Sheehan's book A Bright Shining Lie, a biography of John Paul Vann), SIGINT tipped off virtually every VCor NVA offensive. M It was the predominant source of information on infiltration. Especially after the opening of the Vinh Window in 1967, SIGINT overwhelmed all other sourcesof intelligence on the subject. Its use, however, was very spotty. Some commanders, never having been exposed to it, did not know how to use it and either ignored it or misinterpreted it. Others, like Westmoreland, understood the source and used it to goodeffect. It was often misused, especially by intelligence people who did not understand it. ARDF fixes were espeeially prone to errant analysis. According to the last NSA chief in Saigon,

,---------,

'---------'

Withheld from public release Pub. L. 86-36

I

"

G2 and

J2 briefmp

aU over South Vietnam bloS$Omed with graphs, charts. plottingsyst.ema, between message flow and the number

and

mathematicians

trying to find the magic relationship

of

ARDF locations which, like the secret of the pyramids. thinking of the Communists.sa

could somehow shed divine light on the

Generally, the higher the echelon, the greater the dominance of SIGINT, in the intelligence picture. Sometimes,likejust before Tet 1968, the SIGlNT signals drowned out other sources. Sometimes, as in the Gulf ofTonkin crisis, it was flat wrong.

583

TOP SECRET tlM11tA

DOCID: 523682
lap SECRETUMBRA

REF ID:A523682

What came out of the war was a better SIGINT system; more attuned to the needs of field commanders, better able to render support. On their side, military people began to appreciate how the information could be best employed, how it fit in with their war. The fifteen years following the war represented, for the American military, a long slow road back to respectability and, eventually, dominance. As the military system went, so went cryptology. The ultimate payoff, Desert Shield and Desert Storm, was a model of what the new system was and how effective it had become. , The Turn of the Wheel Though cryptologists did administration would mark the was a period of almost unbroken 1972 had begun a retrenchment not know it at the time, the end of the first Nixon end of an era and the beginning of another. Behind them expansion. The cryptologic system peaked in 1969 and by the outlines of which could be' only dimly perceived.

The heyday of centralization, too, was over. The desperate in-fighting that marked the latter years of the war would contribute loa limited reversal of the engines of centralization. The wave was about to wash the other way. Ahead was a period of "downsizing," intensified by the Watergate crisis. The scandal that led to the president's 'resignation in 1974 would tar the intelligence system. It would not begin to recover until the last days of the Carter administration in 1979.

Withheld from public release Pub. L. 86-36
1. Karnow.
2.

Notes

r:::::::::==J

Glenn, "Uncertain

Origins," Drago", S~edB, Dec. 1972 .

. 3. CCH Series VtHH.9.6.
4. CCH Series VIII.19. 5. James J. Wirtz, The TeIO(ferr.siw: Cornell Univ. Press, 1991). InteUigerr.ce Failure in War. Comell Studies in Security Mairs (Ithaca:

.

6. CCH Series VI.HH.9.3; Wirtz; Johnson Library, NSF.
7. CCHSeriesVI.HH.9.3. 8. Ibid.

9. Thomas N. Thompson, "A Special Historical Study ofSIGINT Support to Air Operations in SEA, 1965-1971," San Antonio,Teus. USAFSS. November 1972.

HANDLE VIA TALENT KEY ~-NOT1mLEASABLE

TROLSYSTEMSJOINTLY TO FOREIGN NATION

lOP 5ic;RiT ua4BU.

584

DOCID: 523682

REF ID:A523682
T9P SEEAET UMBRA

10. Karnow. 11. CCH Series VIII.lS. 12. Ibid; Johnson Library, NSF. 13. Wirtz. 14. Wirtz; S. D. Breckenridge. The CIA and 1M U.s.lnteUigenu Sy.km (Boulder: Westview Press, 1986). 15. CCMSeries VIIJ.19. 16. CCH Series VIII.19.; Karnow. 17. CCH Series VIII.19.; Karnow. 18. CCMSeries VIII.lS. 19. (bid. 20. Karnow. 21. Ibid. 22. Ibid. 23. Ibid. 24. Ibid. 25. CCH Series VI.HM.18.9. 26. CCMSeries VI.HM.l.l0. 27. CCHSeriesVI.HH.18.9. 28. Ibid. 29. Ibid. 30. CCH Series VI.HH.l.ll. 31. Ibid. 32. CeM Series VI.HH.18.9. 33. CCH Series VI.HH.I1.1. 34. CCH Series VI.HH.1.l0. 35. Ibid. 36. Herring. 37. CCH Series VI.HH.1.l0; IS.9. 3S. Thompson, "Special Historical Study .••• " 39. Ibid. 40. Thompson, et al., SIGlNT Applicati.ons ...•

585

TOP 5l!CREHIMBRA

DOCID: 523682
Tap 5EERETI:JM81tA

REF ID:A523682

Withheld from public release Pub. L. 86-36

44. Karnow. 45. Ibid.

4.6.1 47. Ibid. 48. Karnow. 49.c=J

IFoclII on C4mbodia, Crypto\ogic

History Series, Southeast Aaia (Ft. Meade: NSA, 1974.).

.50.0

Gerhard,lntJu:ShaclDwofWar.

61r==J
52. Ibid.
63. Ibid. 54. Ibid.

55. Ibid. 56. Ibid. 57. Ibid.
58. Karnow.

59. CCH,NDUcollection,box210.
60. Karnow; CCH Series VI.HH.23.2-23.S.

61. CCH Series VIIl.21. 62. Ibid.
63. Ihid. 64. Ibid. 65. Zaslow interview. 66. Ibid. 67. CCH series VI.HH.9.2; 9.6.:!<-68. CCH Series Vl.HH.9.2-9.6; Ksrnow.
HOllie

-JI"Things that Go Clank in the Night," Dragon Seeds, Sept.1972.

69. Herring, 246; Seymour Hersh, The Priee of Power: Kissinger in the Ni:con White Books,1983),503-504. 70. Herring, Karnow:

(New York: Summit

71. "Historieal Study of TEA BALL. ..."

HANDLE VIA TA

L SYSTEMS JOINTL Y EASABLE TO FOREIGN N

lOP SECRET I;JMeltA

586

DOCID: 523682

REF ID:A523682
Tap SECRET I:J MIRA

72. Oral Interview with Doyle Larson, 15 March 1994, by Charles Baker and Tom Johnson, NSA OH 15-94. 73. "Hiatorical Study ot.TEABALL ... ~; Larson interview; Chancel T. French, Deadly AtlvcuIt4ge: Signals Intl/Uigenu in Com.b4t, V. II, Air University Research Report # AU·RRl·84-1 (Maxwell AFB: Air Univeraity Press, 1984). 74. Ibid. 75. ~Hi.storical Study of TEABALL ••• ~ ; Lang; Thompson, •A Historical Study ofthe ~l08ure Security Region .... " 76. Thompson. 77. Karnow; Thompson. 78. Herring. 79. ACC 16512, CRRG 36; "A Historical Study ofTEABAI,.L.~ 80. ACC 1651Z,CBRG35;Thompson. 81. CCH Series VI.HH.1.10. 82. CCH Series VI.HH.6.S4; 12.11; Thomp80n. 83. CCH Series VI.HH.26.S.

ot the

Pacific

Withheld from public release Pub. L. 86-36 '------------'

84. Ltr. Subj: Project Corona Harvest End·oC·Tour Report, 1 Auguat 1971.
8S.c=J 86. Ibid. "NSA in Vietnam .... "

HANDLE VIA TALENT

SYSTEMSJOINTL Y

o FOREIGN
587

TOP SI:CAET I:IMBRA

DOClD: 523682

REF lD:A523682
'fOp SEERET tJM81tA

Glossary of Abbreviations
ABM - Antiballistic missile ACC - ARDF Control Center AC&W -Air Control and Warning ACRP - Airborne Communications , Reconnaissance Program (or Platform)

AFEWC - Air Force Electronic Warfare Center AFSA - Air Force Security Agency AFSAC - Armed Forces Security Advisory Committee AFSAFE - AFSA Far East office AFSCC - Air Force Special Communications Center

AFSS - Air Force Security Service (See USAFSS) AGER - Auxiliary General Environmental Reserach

AMPS - Automated Message Processing System ANCIB -Army-Navy Communications Intelligence Board Intelligence Coordinating Group Committee

ANCICC - Army-Navy Communications

ANEEG - Army-Navy Electronic Evaluation ARDF - Airborne radio direction finding . ARVN - Army of the Republic of Vietnam ASA - Army Security Agency ASAE - ASA Europe ASAEUR - ASA Europe ASAPAC - ASA Pacific

AFSSO - Air Force Special Security Office (or Officer) AFSSOP - Air Force Security Service Office of Production ARVN - Army of the Republic of Vietnam ATIC - Air Force Technical Intelligence Center
I .

BIX - Binary Information BRUSA - British-U.S~ CAP - Combat air patrol CBNRC - Communications

Exchange

Branch, National Research Council Committee

CCC - Critical Communications

HAN~~DnL~E~VijI~~~~~~~~~::~~O~L~SY~S~T~EMSJOINTLY TIONALS

589

TOP SiCRET UMBRA

DOCID: 523682
TOP SiCRET l:IMBItA
CCP - Consolidated ~ CCU CO MINT

REF ID:A523682

Cryptologic Program

Contingency Unit disposed antenna array

!

.•..
~

I

CDAA - Circularly

I

.--W-I-·t-h-he-I-d-f-ro-m----, public release
Pub. L. 86-36

'" V:;

CHICOM - Chinese Communist CHINAT-ChineseNationalist CIA - Central Intelligence Agency CIG - Central Intelligence Group

'

::l c
~

M

I/'l

CINCEUR - Commander in Chief, Europe CINCPAC -Commander CINCPACFL~ in Chief, Pacific Command

- CINCPAC Fleet of Joint Operations Authority

CJO - Coordinator

CMA - Collection Management CNO - Chief ofN aval Operations

COMIREX - Committee on Imagery Requirements COMOR - Committee on Overhead Reconnaissance COMRADPAR - Combined Radio Printer Party

and Exploitation

COMUSMACV - Commander Military Assistance Command Vietnam COC - Collection Operations Center CONAD - Continental Air Defense Command .

COSVN - Central Office South Vietnam CPC CO MINT

Processing Center

CRC - Control and Reporting Center CRD - Communications Research Division

CRP - Control and Reporting Post CSG - Cryptologic Support Group CSOC - Current
SIGINT

Operations Center

CTAK - Cipher Text Autokey DCA - Defense Communications Agency

DCI - Director of Central Intelligence DDI - Delivery Distribution Indicator (000)

DDR&E - Deputy Director for Research and Engineering

HANDLE VIA TAL OT RELEASABLE

MINT CONTROLSYSTEMSJOINTLY TO FOREIGN

TOP SECRET UMBRA

590

DOClD: 523682

REF lD:A523682
TOP SECREl' UMBRA

DEFSMAC - Defense Special Missile and Astronautics DF - Direction finding DIA - Defense Intelligence Agency DIRNSA - Director, NSA DMZ - Demilitarized zone

Center

DSB - Defence Signals Branch DSD - Defence Signals Division DSU - Direct support unit

Withheld from public release Pub. L. 86-36

EAM - Electronic Accounting Machine ERA - Electronic Research Associates ESV - Earth satellite vehicle EUCOM - European Command EW - Electronic warfare F ANX - Friendship Annex

FBI - Federal Bureau of Investigation FBIS - Foreign Broadcast Information Seivice FCC- Federal Communications FFV - Field Force Vietnam FMSAC - Foreign Missile and Space Analysis Center FOIA - Freedom ofIniormation Act Commission

FRUMEL - Fleet Radio Unit, Melbourne FRUPAC - Fleet Radio Unit, Pacific GCI - Ground-controlled intercept

GDRS- General Directorate of Rear Services GMAIC - Guided Missile and AstrQnautics Intelligence Committee GSFG - Group of Soviet Forces. Germany lAC - Intelligence Advisory Committee IATS - Improved AG-22 Terminal System IDA - Institutes for Defense Analyses Facility

IDDF - Internal Data Distribution I FFV - First Field Force Vietnam

--.'~)~

.

LETOFORE

.

591

1=OPSECAETI:JMBRA

DOClD: 523682
TOP SECRET l::JMBAA

REF ID:A523682

II FFV - Second Field Force Vietnam IG - Inspector General
1 ------'

IRBM - Intermediate-range ISS - Intelligence JCEC -Joint

ballistic missile

Support Staff Eiectronics Committee

Communications

JCIC - Joint Counter Intelligence Committee JDAJE - Joint Development ActivitylEurope JMG -Joint Mechanization Group

Withheld from public release Pub. L. 86-36

JNACC -Joint
1

Non-Morse Coordination Center
-----

JSPC -Joint

Sobe Processing Center

LL VI - Low-level voice intercept LSIB - London Signals Intelligence LSIC - London SIGINTCentre MAAG- Military Advisory Assistance Group MACV - Military Assistance Command Vietnam MAF - Marine Amphibious Force MGS - Mission Ground Station MOU - Memorandum MPU -Main of Understanding Unit ballistic missile direction findings sErvice Board

Processing

MRBM - Medium-range MRDF - Medium-range

MSTS - Military Sea Transport MUSCO-

Manual of U.S. COMINT Operations of U.S. SIGINTOperations

MUSSO -Manual

NBS- National Bureau of Standards NCML - National Computing Machine Laboratory NCS - National Cryptologic School NEP-National ELlNTPlan Programs Evaluations Resources Board

NlPE - National Intelligence NIRB - National Intelligence

HANDLE

co 592

OLSYSTEMSJOINTLY

toe SEERET

UMBRA"

DOCID: 523682

REF ID:A523682
TOP SECRET YMBRA_

·NKP - Nakhon Phanom. NORAD - North American Air Defense Command NPIC - National Photographic Interpretation Center

NRL- Naval Research Laboratory NRO - National Reconnaissance NRP - National Reconnaissance Office Program

NRV - NSA Representative NSAAL - NSA Alaska NSAEUR - NSA Europe

Vietnam

NSAEURIISS - NSA Europe Intelligence Support Section NSAEUR OG - NSA Europe Office Germany NSAFE - NSA Far East NSAPAC - NSA Pacific NSAPAC NOG - NSA Pacific Operations Group NSASAB - NSA Scientific Advisory Board NSAUK - NSA Office United Kingdom NSC - National Security Council NSCID - National Security Council Intelligence NSG - Naval Security Group NSOC - National SIOIN'!' Operations Center NSS - Naval Security Station NTPC - National Technical Processing Center NV A - North Vietnamese Army Directive

NVN - North Vietnam or North Vietnamese OASD - Office of the Assistant Secretary of Defense OJT - On-the-job training ONI ~ Office ~£N avallntelligence OPC - Office of Policy Coordination OPCONCEN - Operations Center OPSEC - Operational security

OSD - Office of the Secretary of Defense OSO - Office of Special Operations

HANDLE~ATALENT~K~~::~~~~~~;Y~ST~E:M~S~JO~I:N:TLY NOTRELE~~.ti"'fU 593

DOCID: 523682
Tap SECRET l:JMBRA .
OSS - Office of Strategic Services OTP - One-time pad PACAF - Pacific Air Frace PACEXF AC - Pacific Experimental

REF ID:A523682

Facility Program

P ARPRO - Peacetime Aerial Reconnaissance PFIAB - President's

Foreign Intelligence Advisory Board

PIWO - Prod Intelligence Watch Office PLO - Palestine Liberation Organization and budgeting system

PPBS - Planning, programming PWO - Prod Watch Office

KAGFOR - Radio Analysis Group, Forward RAM - Rapid analytic maehine RGM - Radio Group Mobile ROK - Republic of Korea RRB - Radio Research Battalion RRU - Radio Research U ~it RSM - Radio Squadron Mobile· RVN AF - Republic of Vietnam Air Force SAC - Strategic Air Command SACEUR - Supreme Allied Commander, Europe SAM - Surface-to-air missile

8AR - Search and rescue SARC - Surveillance ~nd Reporting Center

SCA - Service Cryptologic Agency SCAT - Support Coordination SCOCE - Subcommittee Advisory Team

On Compromising Emanations

SEATO - Southeast Asia Treaty Organization . . SlOP - Single Integrated Operational Plan SMAC - Space and Missile Analysis Center SMTIG - Soviet Missile Technical Intelligence Group SN 00 - Senior NSA Operations Officer SOO - Senior Operations Officer .

HANDLE VlA

~e:~

It5YWO' E cgum'f ev;!:L

SYSTEMSJOINTLY

lOP SECRET l:JMBItA

594

DOCID: 523682

REF ID:A523682
TOl! SECRet UMBRA

SPACOL - Space collection SORS - SIGINT Overhead Reconnaissance SRB - Special Research Branch SRDF - Short-range direction finding Subcommittee

SSG- SIGINT Support Group SSO - Special Security Office (or Officer) SSSC - SI~JNT Satellite System Control SSSPB - Space Surveillance
SIGINT Planning

Board

SSTB - Special Security Technical Branch STANCIB - State-Army-Navy Communications Communications Intelligence Board

I

STANCICC - State-Army-Navy

I

Intelligence Coordinating
I

Committee

TACC - Tactical Air Control Center TACREP - Tactical report TAREX - Target Exploitation TDS - Teletype Distribution TEBAC - Telemetry System

and Beacon Analysis Committee

TECH INS - Technical Instructions TECSUM - Technical Summary TF - Task force TF A - Task Force Alpha

Withheld from public release Pub. L. 86-36

TICOM - Target Intelligence

Committee

TRO - Technical Research Office TRS - Technical Research Ship TRSSCOM - TRS Special Communications U&5 - Unified and Specified (Command) UKUSAUnited Kingdom-USA System

USAFSS - United States Air Force Security Service USCIB - United States Communications USCICC - United States Communications USCSB - United States Communications Intelligence Committee Intelligence Coordinating Security Board Committee

H

T KEYHOLE COMIl~~rTRRioDiLSYSTEMSJorNTLY IONALS

595

Tep SEeRn UMBRA

DOCID: 523682
TOP SECRET UMBRA

REF ID:A523682

USIA - United States Information Agency USIB - United Sta.tes Intelligence Board VC - Viet Cong VOA - Voice

of America
REGAL Center Communications Net

WA YES - Women Accepted for Volunteer Emergency Service WRC - Washington

ZICON - Zone oflnterior

HANDLE VI

OL ASABLETO 596

OL SYS'l'EMS JOlNTL Y

N'ALS

Tep 5&CRiT UMBRA

.. - - --------~~~-~-=---=~~

DOClD: 523682

REF lD;A523682
TOP SECRET lIMBM

Sources
Mostof this history was written from classified cryptologic another. The most useful document collections.are as follows: records of one sort or

Withheld from public release Pub. L. 86-36

1. The NSA Archives. This organization (currently E321) acts as the repository for retired NSA records. It is located in] I at NSA-Ft. Meade. Retired records remain the property of the donating office until they are screened and formally archived, at which time they become the property of the Archives organization. Thus, the organization has two collections: a. Retired records. Because these are stiJ1 property of the originating office, a researcher needs written permission to access the documents. Retired records are identified by a five-digit number representing the box number, followed by a shelflocation. An example is 43852, 73-252. b. Archived records. Documents in this area may be accessed by any qualified researcher without the permission of the originating organization. The collection is indexed by key words, .and trained archivists can search the collection for records responding to the query. Records are stored by Accession Number (ACC) and a location. An example would be ACC39471, H03-0311-4. 2. The historical collection of the Center for Cryptologic History (CCH), E322. This collection of historical documents actually predates the archived collections, and it contains records going back to the earliest days of cryptology. Records in this collection generally duplicate those in the Archives, but they are maintained as a separate file for ease of access by historians. The CCll collection is organized in series as follows: I. II. III. . IV. V. VI. VII. VIII. X. XI. XII. XIV. XVI. Pre-1915 1915-1918 (World War 1) 1919-1939 (Interwar period) 1939-1945 (World War II) 1946-1952 (pre-AFSA and AFSA period) 1952-present (NSA period) Special and miscellaneous collections Crisis files References Papers collected by NSA and pre-NSA officials Papers collected by NSA historians COMSEC documents Cryptologic papers from presidential libraries

Citations from this collection are by series number, followed by subseries designations, for instance, VI.A.1.9. Most of the Cell documents used for this history (not surprisingly) were from Series VI.

597

TOR SECRETua4BM

DOCID: 523682
'fOp SECRET l-JMBRA

REF ID:A523682

In addition, the CCH maintains the formerly DIA Vietnam document collection. For Vietnam, the DIA collection (which came to NSA through the National Defense University in serpentine fashion and is thus called the NDU collection) combines with CCH's own collection of mainly cryptologic documents collected. by William Gerhard in the 19705 to form perhaps the best collection of it~ kind in existence. 3. Oral histories. Compiled over a period of many years by various NSA organizations and individuals, the oral history effort has come to rest in the CCH, and the great preponderance of taped reminiscences were done by that organization and its predecessors. In addition, the CCH now has copies of most of the oral histories that were done before its time. Most are designated by an oral history number,e.g.,NSA OH 12-86.All are held in the CCH unless otherwise indicated. Oral histories which proved especially useful in this study were these: taken from videotaped discussions involving five NSA dlrectors associates (1969-1970 taping), no number 29-94 25-94 [ 31-87 ~ '::;Qo-rd-'-on-A-:--;:. B;::-;l-:ak"-e.....l, 7.84 r David G. Boak, 17-86 ~I Howard Campaigne, 14-83 Ralph J. Canine, no number Marshall S. Carter, 15-88 Herbert L. Conley, 1·84 Harold E. Daniels, 10-88 (videotape) I l8-85 Robert E. Drake, 18-83. ~.....I1~92 Transcripts their and

Withheld from public release Pub. L. 86-36

LI~-=~---------------------------------------.....Il4-83
John B. Eastman, 3-87 Henry R. Fenech, 8-81 Laurence H. Frost, by and held at JFK Library, Boston Charles L. Ga~nd 19-86

~_=====~:-:::::------:----------17.92 \ ~_824-86
Oliver .Doyle David 1 David R. Kirby. 20-93 E. Larson, 15-94 D. Lowman, 13-80 12.93 Y. McManis, 34-86

\

John E. Morrison, Jr., 24-93 Helen O'Rourke, ·11-8;:.;1:......;.._--. Withheld from CecilJ. Phillips and 1 114-93 public release Cecil J. Phillips, 23·93

~;!-87

Pub. L. 86-36

HANDLE·
__ --NU'rRELEASABLE TO FOREIGNNA

LSYSTEMSJOINTLY

lOP SECR'T UMIAA

598

DOClD: 523682
Withheld from public release Pub" L. 86-36
L-----'

REF lD:A523682
TOP SECRET UMBftA

.1

11-93

Howard E. Rosenblum, 3-91 I Ino number John W. Saadi, 29-87

I

I
Eugene Sheek, 26-82 Abraham Sinkov! 2-79 through 4-79 18-86 Kermit H. Speierman, 2-86 Earl K Stone, 3-83 Louis W. Tordella, 8-90 CharlesC. Tevis, 21·87

I

\

r;7;J

10-80

Milton Zaslow, 17-93 4. Internally published historical books and articles represented most valuable were as follows: a significant source. The (no year), 8Ft. Meade:

I

I"The Gulf of Tonkin Incident." Cryptoiog, Feb-Mar 10. (Located in CCH Series VIII.13.)
History of Venona. Security;

Benson, Robert Louis, and Cecil James Phillips. NSA,1995. Book, David G. A History of U.s. Communications Lectures.) Ft. Meade: NSA,1973. Boucher, Melville J. "Talomatry 1971, Winter 1972.

(The David G. Boak Fall

and How it Grew." Cryptologic Spectrum,

Burns, Thomas L. The Origins of tne National Security Agency, 1940-1952. U.S. Cryptologic History, Series V, Vol. 1., Ft. Meade: NSA, 1990. Campaigne, Howard H. "Lightning." NSA Technical Journal, July 1959.

Davidson, Max L. "The CRITICOMM Sy~tem." Cryptologic Spectrum, Spring 1975.
1 I "The National SIGlNT Operations

Center."

Cryptologic

Spectrum, Summer 1979. U.S. Cryptologie History Series- Special Series. Ft. Meade: NSA, n.d.

'";:=====:::!:"'--

I

Withheld' from 1 "BRANFLAKE.':. Cryptologic QUarterly, Winter 1994, Vol. 13, No.4. public release 1 "Glimpses of a Man: The Life of Ralph J. Canine." Cryptologic L-.P_u_b_" "_L_" _8_6_-3_6~

Quarterly, Summer 1987, 31-39. William D. Gerhard served as the general editor for a mid-1970s project to write the cryptologic history ofthe Vietnam War. The following volumes were published (all of them by NSA in the Cryptologic History Series - Southeast Asia) before the project expired:

599

'fep SECRET I:JMB~

DOClD: 523682
"TOP SECRET I:IMBAA

REF lD:A523682

Withheld from public release Pub. L. 86-36

'---

....J Deadly Transmissions

(COMSECMonitoring and Analysis).

1970.

Gerhard, William D. In the Shadow of War. 1969.

I
I , Applications

I Focus on Cambodia.

1974 Iand William'

D. Gerhard,

SIGINT

in U.S. Air Operations. 1972.

Gerhard, William D., and Henry W. Millington. Attack on a SIGINT Collector, the U.S.S. Liberty. U.S. Cryptologic History Series, Crisis Collection. Ft. Meade: NSA, 1981. ' I October 1975. I IHenry F. Schorreck, and Donald C.Wigglesworth. A Reference Guide to the Selected Historical Documents Relating to the National Security Agen.cylCentral Security Service, 1931-1985. Ft. Meade: NSA,1986.
J

I "NSA in Vietnam:

Proud and Bitter Memories."

Cryptolog,

Howe, George F. Technical Research Ships, 1956·1969; An Historical Study. Cryptologic History, Special Series, No.2. Ft. Meade: NSA, n.d. --. "A History of U.S. Civilians in Field Cryptologic Spectrum, Summer 1973.
COMINT

U.S.

Operations,

1953·1970."

'\

I
,I
1972.

I "OPSEC

as a Management Tool." Cryptowg ,1st issue, 1992. That Go Clank in the Night." Dragon Seeds, September ' of 1960." Cryptologic

I "Things

I I "Reflections on the Soviet Missile Threat Spectrum, Summer 1981.

I

I PURPLEDRAGON: The Origin and Deoelopment of the United States OPSECProgram. us, Cryptologic History, Series VI, the NSA Period, Vol. 2. Ft. Meade: NSA,1993.
Kirby, Oliver R. «The Origins of the Soviet Problem: Cryptologic Quarterly, Winter 1992, Vol. 11, No.4. A Personal View."

I

I NSA's Involvement in U.S. Foreign SIGINT Relationships through 1993. U.S. Cryptologic History, Series VI, Vol. 4, Ft. Meade: NSA,1995.

Moore, Elizabeth.' As We Were: An Informal History of Bad Aibling Station, 19361988. Bad Aiblinl;t: Englemaier Druckner, 1988. \ Newton, Robert E. The Capture of the USS Pueblo and Its Effect on SIGINT Operations. U.S. CryptologicHistory, Special Series, Crisis Collection, Vol. 7. Ft. Meade: NSA,1992. .

Withheld from public release Pub. L. 86-36

I

"Deployment of the First ASA Unit to Vietnam," Quarterly, FalllWinter 1991, Vol. 10, Nos. 3·4.

I

'Cryptologic

TOil SECRETl:JMBR6

600

DOCID: 523682
KO. 13526, section 1.4(c)(d)

REF ID:A523682
TOP SECREI UMBRA

Withheld from public release Pub. L. 86-36

I

I "Before BOURBON: American and British COMINT Efforts against Russia and the Soviet Union Before 1945." Cryptologic Quarterly, FalllWinter 1993.

---. "Early BOURBON - 1945: The First Year of Allied Collaborative COMINT Effort against the Soviet Union." Cryptologic Quarterly, Spring 1994, Vol 13, No. l. ---. "Middle BOURBON -1946: The Second Year of Allied Collaborative Effort against the Soviet Union." Cryptologic Quarterly, Summer 1994, Vol. 13, No.2. ---. "Old BOURBON - 1941: The Third Year of Allied Collaborative COMINT Effort against the Soviet Union." Cryptologic Quarterly, Fall 1994, Vol. 13, No.3.

I

I

I

"Early History of the Soviet Missile Program (1945-1953)."
Cryptolog, 1st issue 1992.

Cryptologic Spectrum, Summer 1975.

I

I "The Great Conversation."

Snyder, Samuel S. "Influence of the U.S. Cryptologic Organizations on the Digital ComputerIndustry." Cryptologic Spectrum, Fall 1911. . . "History of NSA General-Purpose Electronic Digital Computers." NSA Technical Literature Series. Ft. Meade: NSA,1964. [Wiley, Edward S.l On Watch: Profiles from the National Security Agency's Past 4(J Years (Ft. Meade: NSA,1986).

1Crypt()ZOglCSpectrum,

r=:-=-~=",,".......-:--:-=::-:-:c:-"[==-::-:::-

Summer 1970.

•.••••

....-_.....JI

"The Civilianiza tion of Harrogate. " June 1977.

____

----'1 "AG·221IATS:A View from the Bridge." Cryptolog,

Wigglesworth, Donald. "Cuban Missile Crisis: A SIGINT Prespective." Cryptologic Quarterly, Spring 1994, Vol. 13, No.1.

Wagoner, H.D. Space Surveillance SIGINT Program. Special Series, No.3. Ft. Meade: NSA,1980.

U.S. Cryptologic History,
in

Wonus, Corley. "The TACKSMAN Project: A SIGINT Success Story." Studies Intelligence, Fall 1991. (Also reprinted in Cryptologic Quarterly, Vol. 12,1993.) Ziehm, Thomas P. The National Security Agency and the EC·121 Shootdoum: Cryptologic History, Special Series, Crisis Collection, Vol. 3.

U.S.

5. Another collection is the vast array of informal, unpublished histories and summaries of historical events .. Most of these are held in both the CCH collection and in the NSA Archives. Bauer, Dr. Theodore W. "Historical Study: The Security Program of AFSA and . NSA,1949-1962." 1963.

~,~~

ASABLETOFvru,!\7~

601

TOP

S'CRET UMB~

DOCID: 523682
TOP SECRET UMBRA

REF ID:A523682

....•. 0.13526,

section 1.4(c)(d)
--.

I
"Training in AFSAlNSA,1949-1960." 1961. Intelligence during World and Its n.d.

Withheld from public release. Pub. L. 86-36

Benson, Robert L. "A History of U.S. Communications War II." Available in CCH.

I

I "The History of the NSA SIGINT Command Predecessors,1949-1969." 1970. .
"The National Security Agency Scientific Advisory&ard, "The Consolidated

Center

--. --. 1971.

1952-1963."

Cryptologic Program and Its Predecessors, .

1957-1965."

--. "NSA's Participation 1957-1964." 1968.

in the Research and Development of the 466-L System,

1__

-

[Drake, Robert and others.I "The COMlNT Role in the Korean War."· Enderlin, Arthur. [Enderlin.] "NSA's Telecommunications Problems,1952-1968." 1974. 1969.

"Telecommunications

Problems,1968-1972."

Fitzgerald, II." n.d.

Edward.

"A History of U.S. Communications

Security:

Post-World War

I

I "The U.S. COMuiT Effort during the Korean Conflict - June 1950August 1953." 1954.

I
I
1960."

I ;'Collected

Writings on NSA's R&D Effort."

I "The Early Structure of the National Security Agency, 1952-

"Historical

Study ofNSA Telecommunications,

Annual, 1973-1975."

Hogan, Douglas. "General and Special-Purpose Some Lessons Learned." 1986. Howe, George F. "The Narrative
--. "CO MINT Production

Computers:

A Historical

Look and

History of AFSAlNSA, Parts I-V." n.d.

in the Korean War: The AFSA/NSA Contribution."

HANDLE VIA T RELEASABLE TO FOREI

ROL SYSTEMS JOINTL Y

TOP SteRE r UMBRA

602

DOClD: 523682
~.O. 13526, section 1.4(c)

REF lD:A523682
TOP SECRET UMBRA

Withheld from public release Pub. L. 86-36

-. 1956.

"Centralized COMINT Communications Centers: The Historical Record." -----lI"RadioDirection Finding in the U.S.

... 1-,Navy: The First Fifty Years." n.d.
,;:1

:--;;~---::-:-:=---,I "History of HFDF in the Pacific Ocean Prior to the Advent of Bullseye." 1985.

NSASAB. "Technologyfor Special Purpose Processors." 1978. Page, Ryon A. "The Wired Rotor in U.S. Communications Security." 1980. I"History of Menwith Hill Station."· n.d. I;:::===:::::::;'::::::;I:-:':'T=-he Soviet Land-Based Ballistic Missile Program, 1945-1972: A Historical Overview." n.d. "Summary of Statutes Which Relate Specifically to NSA and the Cryptologic Activities ofthe Government."
L..-

---llftDEFSMAC - A Community Asset cl964-1989)." n.d. --JI"ConsumerLiaison Units, 1949-1957." 1957.

_______

Williams, Joseph L. "The National Security Agency's Gray Telephone System: Present and Future." 1982.
6. Certain documents are so important that they deserve separate mention, even though contained in the CCH and Archives collections above. Among them (in chronological order) are these: .

"Report to the Secretary of State and the Secretary of Defense by a Special Committee Appointed Pursuant to Letter of 28 December 1951." [Brownell Report]. CCH Series .v.F.7.13. "Report Intelligence Activities in the Federal Government, Prepared for the Commission on Organization of the Executive Branch of the Government by the Task Force on Intelligence Activities, App. I, Part 1: The National Security Agency." [The Hoover Commission report.] CCH Series VI.C.i.8. "The Baker Panel Report and Associated Correspondence, 1957." CCH Series VI.X.1.9. "Report of the Secretary's Ad Hoc Committee on COMINT/COMSEC,June 1958. [Robertson Report.] CCH Series VI.C.l.l1.

on

o

EASABLETOFO ~
603

[Q~AIS

lOP SECRET' 'MBRA

DOCID: 523682
'OP SECRETUMBRA-

REF ID:A523682

~

"Precis of the Bissell Report (Review of Selected NSA Cryptanalytic February 1965)." NSAlCSS Archives, ACC 290Z,199104. "Report of the Eaton Committee, 1968." CCH Series VI.C.1.24.

Efforts,

18

~ ,..; .S:
CJ Q,I

'" ...a
N
l/'l

-

c

Withheld from public release Pub. L. 86-36
7. Service cryptologic organizations all have collected a certain amount of material:

~

a. Air Intelligence Agency, formerly Electronic Security Command, Air Force Intelligence Service, and U.S. Air Force Security Service, has the best collection of official . histories. All are held at AlA headquarters at Kelly AFB, San Antonio; in addition, the CCH holds copies of many, if not most. Used in this study were the following: "AFSS-NSA Relations, October 1952-8eptember 1954, V. I." n.d.

"An Oral History Interview: The Electronic Security Command - Its Roots; Featuring the Founder of USAFSSI/ESC, Lt Oen Richard P.Klocko (USAF, Ret.)" Hqs ESC, 20 October 1989. . "Analysis of AFSS Effort in the Korean.Action," n.d .. Development of

Ferry, Richard R. "A Special Historical Study of the Organizational United States Air Force Security Service from 1948-1963." 1963.

French, Maj Chancel T. "Deadly Advantage: Signals Intelligence in Combat." Vol. II, Air University Research Report#AU-RRI-B4-l. Maxwell AFB: Air University Press, 1984. Available at both AlA and Air. University. [Harriger, Hop) "A Historical Study of the Air Force Security Service and Korea, . June 1950-0ctober 1952." 1952. "A History of the USAFSS Airborne SIGINTReconnaissance Program (ASRP), 195019'77." 1977. "Historical Data Report for the 6920 SG,l January 1953-30 June 1953." n.d.

"History of the USAF Security Service; Fiscal Year 1955." n.d. "His:orical Data Report for the 6901 SCG, 1956-1964." "A Historical Study of USAFSS SIGINTSupport to the TEABALL Weapons Control Center." 1974~ "Historical Resume: Development Pacific Area, 1949." 1957. and Expansion of USAFSS Capability in the

"Historical Report: The Developmentofthe

U.S. ELINTEffort." n.d.

Holub, Mary V., Jo Ann Himes, Joyce M. Homs and Ssgt Kay B.Grice. "A Chronology of Significant Events in the History of Electronic Security Command, 1948-1988." 1990. Larson Doyle E. ESC Oral History Collection interview, 1987.

TOP !!~It!T UMBRA

604

DOCID: 523682

REF ID:A523682
TOP SECRET l:JMBItA

~.o. 13526,

section 1.4(c)

I

Withheld from public release Pub. L. 86-36

I

I"History of the United States Air Force Security Service Fiscal Years 1960-1961," Part IV, Systems Development. 1962.
"A.Historical Study of the Iron Horse System; 1965-1973." IOral History, 1986. "Review of Reactions to Reconnaissan~e Flights Since 31 October 1958." 1960. Reporting 1974.

-.

I

'Rush, Robert. "AFSCC Tasking: Concept, 1949-1952." n.d. Sommers Gordon W. Oral History.

The Development of the Three-Echelon 1990. Warning Program,

"A Special Historical Study of the Advisory December 1964." 1965.

July

1961to Air

I
operations

"A Special Historical in SEA, 1964-1971." 1972.

I

Study of SIGINT Support ..

--. •• A Historical Study of the Closure of the Pacific Securitty Region and the Impact Upon USAFSS Operations in SEA." 1974. USA·36 Unit History. January-dune 1967. of USAFSS

Whitacre, SMsgt Frank. "A Historical Study of the Drawdown Operations in Southeast Asia (SEA)," 1974.

b. Compared with AlA, INSCOM has very little in the way of official histories, but its archives are more extensive .. The most useful items found in the archives were the unit histories, especially those of I Also used were unit histories of both AsAEUR, ASAPAC and ASAFE, the regional headquarters for ASA, as well as various individual unit histories Official histories included the following:

I

Assistant Chief of Staff, G-2, "COMINT Operations of the Army Security during the Korean Conflict, June 1950-December 1953." 1956. Finnegan, John P. "The Structure of Army Intelligence: "Beginnings of ARDF." INSCOM Historical Monographs. 1983.

Agency and

1946-1965"

c. Naval Security Group has the smallest historical program. There is a collection of archived documents that has recently been transferred from Crane, Indiana, to the new National Archives building (Archives II) in College Park, MD. There is also a collection of NSG command histories stored at the Naval Historical Center in Washington, D.C., which was consulted. However, since NSG did not become a "command" until 1968, there are no command histories prior to that date. The command has not had a program of preparing operational histories since shortly after World War II, and there is thusnothing similar to what AlA has available. The only "history" unearthed was "U.S. Naval Communication Supplementary Activities in the Korean Conflict, June 1950-August 1953," contained in CCH Series V.M.3.1. 8; CIA has an active history program and a large collection of official (classified) histories on various aspects of its operations. These histories can be consulted only at the CIA history office in Rosslyn, Virginia, and then only with permission of the CIA Historian.

_~OIEC~~

N~

b

BLETOrv."".'vn'~

605

TOP SECRET U MiRA

DOCID: 523682
TOP SECRt I UMBRA -

REF ID:A523682

I

E.O. 13526, section 1.4(c)(d)

I

Withheld from public release Pub. L. 86-36

In addition, there were three oral histories ofinterest: Richard M. Bissell, Jr. (separate interviews in 1976 and 1984). JohnA. McCone. 1989. James R. Schlesinger. 1982. significant number of meaning to and useful

9. Unclassified publications by outside scholars generally do not contain information about modern (post-1945) cryptologic history, but there are a exceptions. In addition, outside sources must be consulted to give context and cryptologic events. The following list contains a few of the more relevant outside sources used in this study. Ambrose, Stephen E. Eisenhower: Schuster, 1990. Soldier and President.

New York: Simon and

Andrew, Christopher. "The Growth of the Australian Intelligence Community and the Anglo-American Connection." Intelligence and National Security 4:2 (April 1989) 213-256. Appleman, Roy E. Disaster in Korea: The Chinese Confront MacArthur. Station; Texas: Texas A and M Press, 1989. Bamford, James. The Puzzle Palace. Boston: Houghton Mifflin, 1982. College

A Report on America's Most Secret Agency.

"""~
TOP SECRET tlMBItA

VIA";:;ENT <"JiIlI • "-:;:.~rmwLSYSTEMSJOI~ T RELEASABLE TO F ill N1t"le~r' 5
r

606

DOClD: 523682

REF lD:A523682
TOP SECRET tJMBRA

Barker, Wayne G., and Rodney E. Coffman. The Anatomy of Two Traitors: The Defectum of Bernon.F. Mitchell and William H. Martin. Laguna Hills, CA: Aegean Park Press, 1981. Ball, Desmond, and David Horner. "To Catch a Spy: Signals Intelligence and Counterespionage in Australia, 1944-1949." Pending publication from Canberra: Strategic and Defence Studies Centre, Australian National University. Bechloss, Michael. Mayooy: Eisenhower, Khrushchev and the U-2 Affair. York: Harper and Row, 1986. New

---,.

The Crisis Years: Kennedy and Khrushchev, 196G-1963.
Books, 1991.

New York: New York:

Edward Burlingame

Blair, Clay. The Forgotten War: Americans in Korea, 1950-1953. Times Books, 1987.

Breckinridge, S. D. The CIA and the U.S. Intelligence System. Bould, Westview Press, 1986. Brugioni, Dino. Eyeball to Eyeball: The Inside Story of the Cuban Missile Crisis. Edited by Robert F. McCort. New York: Random House, 1990. Bucher, Lloyd M. (with Mark Rascovich). York: Doubleday, 1970.

Bucher: My Story. Garden City, New Security. New

Burrows, William E. Deep Black: Space Espionage and Natio~l York: Random House, 1986. Buttinger, 1968. Joseph.

Vietnam: A Political History. New York: Frederick A. Praeger,
Acropolis

Cline, Ray S. Books, 1981.

The CIA Under Reagan, Bush and Casey. Washington:

Ennes, James M. Jr. Assal#t on the Liberty: The True Story of the Israeli Attack on an American Intelligence Ship. New York: Random House, 1979. Goldschmidt, Arthur. Westview Press, 1979. A Concise History of the Middle East. Boulder, .CO.:

Goodman, Hirsb, and Zeev Schiff. "The Attack on the Liberty." Atlantic Monthly, September 1984. Goulden, Joseph C. Truth is the First Casualty: tlie Gulf of Tonkin Affair - Illusion and Reality. Chicago: Rand McNally, 1969. .

---.

Korea: The Untold Story of the War. New York: Times Books, 1982.

Harris, George. Troubled Alliance: Turkish-American Problems in Historical Perspective, 1945-1971. Washington: American Enterprise Institute for Public Policy Research, 1972. Hermes, Walter G. Truce Tent and Fighting Front: United States Army in the Korean War. Washington: Office of the Chief of Military History, United States Army, 1966.

607

TOP SfCRET tJM81tA

DOCID: 523682
TOP SEeR!T tJMBAltc

REF ID:A523682

Herring, George. America's Longest War: The United States and Vietnam, 19501975. Philadelphia: Temple University Press,1986. Hersh, Seymour. The Price of Power: Kissinger in the Nixon White House. New York: Summit Books, 1983. Herzog, Chaim. The Arab-Israeli Wars: War and Peace in the Middle East. New York: Random Houae, 1982. Kahn, David. The Codebreakere: MacMillan, 1967. Karnow, Stanley.

The Story of Secret Writing. New York:
;

Vietnam: A History. New York: Penguin Books, 1983.

'Kramer, Mark. "Tactical Nuclear Weapons, Soviet Command Authority, and the Cuban Missile Crisis." Cold War International History Project Bulletin, Fa111993. -. "Archival Research in Moscow, Progress International History Project Bulletin, Fall 1993. Lamphere, R. J., and T. Schachtman. New York: Random House, 1986. . and Pitfalls." Cold War

The FBI-KGB War,' a Special Agent's Story.

Laqueur, Walter. A World of Secrets: The Uses and Limits of Intelligence. New York: Basic Books, 1985. Lewin, Ronald. The American Magic: Codes, Ciphers and the Defeat of Japan. New York: Farrar, Straus and Giroux,1982. .," Lewy, Gunter. The Federal Loyalty - Security Program: The Need for Reform. Washington: American Enterprise Institute for Public: Policy Research, 1983. Manne, Robert. 1987. Marolda,

The Petro» Affair: Politics and Espionage. Sydney:

Pergamon,

Edward J., and Oscar P. Fitzgerald. The United States Navy in the Vol. II, From Military Assistance to Combat, 1959-1965. Washington: Naval Historical Center, 1986.

Vietnam Conflict:

Martin, David. Wilderness of Mirrors. New York: Ballantine
.

Books, 1980.
i

McAuliffe, Mary S. (ed.) Washington: CIA, 1992.

CIA Documents on the Cuban Mi8sile Crisis, 1962.
.

Meilinger, Philip S. Hoyt S. Vandenberg: The Life of a General. Bloomington, Indiana: University ofIndiana Press, 1989. . O'Neill, William. AmericanHigh: The Years.ofCon{idence, 1945-1960. New York: Free Press, 1986. Palmer, Gregory. The McNamara Strategy and the Vietnam War: Program Budgeting in tbe Pentagon, 1960-1968. Westport, Conn.: Greenwood Press, 1978. Powers, Thomas. The Man Who Kept the Secrets: Richard Helms and the CIA. New York: AlfredA. Knopf, 1979.

HANDLE

ViA

IALSltT

KEYliOLECOMlNTCON'FReLSiSIEMSJOINTLY TOr'UREI6foi l>fA TIOliALS

neT-RELEASABLE

TOP SI:CRET tJMBAltc

608 .

DOCID: 523682

REF ID:A523682
TO~ECRET l:JMBRA

Radosh, Ronald, and Joyce Milton. The Rosenberg File: A Search for the Truth. New York: Holt, Rinehart and Winston, 1983. Randell, Brian (00.) The Origins of Digital Computers: Selected Papers. New York: Springer-Verlag, 1975. Ranelagh, John. The Agency: and Schuster, 1986. 2nd ed.

The Rise and Decline of the CIA. New York: Simon Fairfax, Va.: Allen and in Vietnam.
.

Reese, Mary Ellen. General Reinhard Gehlen: The CIA Connection. George Mason University Press, 1990.

Richelson, Jeffrey T., and Desmond Ball. The Ties That Bind. Boston: Unwin. 1985. Sheehan, Neil. A Bright Shining Lie: John Paul Vann and America NY: Random House, 1988. .I Shurkin, Joel. Engines of the Mind: A History of the Computer. Norton and Company, 1984.

New York: W. W.

Szulc, Tad. Czechoslooakia Since World War II. New York: Viking Press, 1971. Tahir-Kheli, Shirin. The United States and Pakistan: The Evolution of an Influence Relationship. Studies of Influence in Internal Relations; Alvin Z. Rubinstein (ed.). New York: Praeger,1982. Thies, Wallace J., and James D. Harris. "An Alliance Unravels: The United States and ANZUS. " Naval War College Review, Summer 1993. Willenson, Kim. The Bad War: An Oral History of the Vietnam New American Library, 1987. War. New York:

Wirtz, James J. The Tet OffenSive: Intelligence Failure in War. Cornell Studies in Security Affairs. Ithaca, New York: Cornell University Press, 1991. Wise, David. "Remember the Maddox." Esquire, April 1968. Wright, Peter (with Paul Greengrass). Spycatcher: The Candid Autobiography of a Senior Intelligence Offu:er. New York: Bantam Doubleday Dell Publishing Group, Inc., 1987. 10. Presidential libraries contain key documents and add insights into the cryptologic process at the executive level. All presidential libraries consulted .contained highly relevant information. They were Harry S. Truman Presidential Library, Independence, Missouri. Dwight D. Eisenhower Presidential Library, Abilene; Kansas. John F. Kennedy Presidential Library, Boston, Massachusetts. Lyndon Baines Johnson Presidential Library, Austin, Texas. The Nixon Library papers, which are presently stored at Archives II in College Park, were not consulted because the National Security Files have not yet been processed and made available tor research. Copies oCkey documents from the other libraries are available in CCH Series XVI.

HANDLE VIA TALENT Iff.IIQI EOOMtNTCONTROLSYSTEMSJOlNTLY ~ASABLETO FOREIGNN.

609

TOP SEOE i tJMBRA _

DOCID: 523682 I E.O.
13526, section 1.4(c)

REF ID:A523682
TOP SECRET UMISItA

Withheld from public release Pub. L. 86-36
Abel, Rudolph; 183
ABNER; 200

Index

Abrams, Creighton; 570, 573, 576 ACC - see ARDF Coordination Center Acheson, Dean; 39 ACRP; see Airborne Communications Adak, Alaska 1
1 _

Reconnaissance __

Program

..

1;29,0131,137,139,1

Adams, Sherman; 231

Advisory Council (CIA); 87, 88 Advisory warning; 143-149, 314,329 AFEWC - see Air Foree Eleetronics Warfare Center AFSA Far East (Tokyo; became NSAFE); 67 AFSAM-7 (AKA KL-7); 217-218 AFSA Y-816; 220 AFSAC -

see Armed Forces Security Advi~ry Committee
Center

AFSCC - see Air Force Special Communications

AFSSOP - see Air Force Security Service Office of Production

I
AG-22; 360, 364-365

I
536, 539,

Airborne Communications Reconnaissance Program (ACRP); 132, 139-148, 233, 313, 322, 327,331,390-391,426,428,463,513,531,539-540,542,547,548,549,570,578 Airborne Radio Direction Finding (ARDF); 506-509, 513, 529, 530, 531,532-534, 543,560,561,562,563,5.68,570,574,582, 583 Air Force Electronics Warfare Center (AFEWC); 360 Air Force Security Service Office oCProduction (AFSSOP); 76 Air Force Special Communications Center (AFSCC); 26, 30, 79, 82, 83, 258, 297, 360

Air Force Technical Intelligence Center (ATIC); 109, 176

I
OT RELEASABLE TO

I
Hhtl'l!lONAl S

. TOP SeCRET UM81tA.

611

DOClD: 523682

REF lD:A523682

I

lOP SECR!T tJMBRA E.O. 13526, section 1.4(c)(d)

I

Withheld from public release Pub. L. 86-36

1_Almond, Ned; 45 AMPS (Automated Message Processing System); 371 AMTORG; 158,160, 162

1'-----__
ANIFLR-9 - see FLR- 9

ANCIB - see Army-Navy Communications Intelligence Board ANCICC - see Army-Navy Communications Intelligence Coordinating Committee Anderson, George W.; 329 Anderson, Rudolph; 322, 329-330 ANEEG - see Army-Navy Electronic Evaluation Group

AN/FRD-I0 (CDAAsystem); 188,308,309

Ap Bac; 508, 583
APPLESAUCE

(project); 92, 393 396, 425~39,1,--- __

Arab-Israeli War of 1967; 2390386, Arafat, Yasir; 425

Arc Light (SAC bombing program); 551-52, 553 ARDF - see Airborne Radio Direction Finding ARDF Control Center (ACC); 534, 535 Ardisana, Benjamin; 266-67 Arnold, Henry H. "Hap"; 139 Armed Forces Security Advisory Committee (AFSAC) 30,35,67,68,102,241,243 Army-Navy Communications Intelligence Board (ANCIB);6, 7,15,159 Army-Navy Communications Intelligence Coordinating Committee (ANCICC);5, 6 Army-Navy Electronic Evaluation Group (ANEEG); 109-110 Army Security Agency Europe, Frankfurt (AKA ASAEUR, ASAE);D265, 266

ELEASABLE TO FOREI

.\i:a8

lOP SECRETUMBR4

612

DOClO: 523682

I E.O. 13526, section
268

1.4(c)(d)

II

REF lD:A523682 Withheld from I
public release Pub. L. 86-36
I ur .,L",,\I: I UMDJVo\

Army Security Agency Pacific, Hawaii and Japan (AKA ASAPAC); 40, 46, 52, 128, 131,

I

I
ASAPAC - see Army Security Agency Pacific

ASAE, ASAEUR - see Army Security Agency Europe

I
Asmara, Ethiopial

I
~29, 111,398,400,426

I
ATLAS; 198-200

I
ATIC - see Air Force Technical Intelligence Center Atsugi Naval Air Station, Japanl Attlee, Clement; 19
1142,463,550

[
Autodin; 370

[

Australian Security Intelligence Organization; 18 Automated Message Processing System - see AMPS Autosevocom; 219,380 Ayub (Mohammed Ayub Khan); 303-304, 385-388 Bad Aib1ing, Germany Bancroft; 381 Banfill, Charles Y.; 42

I

l311, 392, 393

I
Bangkok, Thailand (AFSS SIGINT operation); 512

I I
BACCHUS (COMSEC

I
system); 52

Bainbridge Island (Navy intercept site); 159 Baker, William, and the Baker Panel; 186,256-257,260,374,376,481 Ball, George; 449

I
I

Bassett, Hunt; 83
1

I

HAI~~

.

LE TO FOREIGN NA

613

Tep SECREH'MIIU

DOCID: 523682 I E.O.
lOP 5K1tET !;IMBM
13526, ·section 1.4(c) Becker, Loftus; 101 . Beecher, William; 572
/' BEGGAR SHADOW

·REF ID:A523682
Withheld from public release Pub. L. 86-36

program; 314, 463-470 198, 199,200,214,221,256

Bell Laboratories;

I
1

I
Bentley, Elizabeth; 164, 166
--------------------

Berlin Wall and SlGINT; 319 Bien Hoa, ASA tactical unit; 573 Big Look (Navy airborne reconnaissance project); 550

BIG RIB (airborne collection project); 303, 383-384, 386
1 1

Binary Information

Exchange (BIX); 366

Bissell, Richard (and the Bissell Study); 107,337,374,376-377,403,405

I

BI'l"l'ERSWEET(project);143-144 B1Xsee

Binary InformationExchange

Black Widow Mountain; Black, William; 352 Blake, George; 106

536,538,581

Blake, Gordon; 133,269, 307,326,327,344,347,348,349,357,358,36 . biography, 340-341 .

6,377,471,506,511

Bletchley Park; 1,2 Blue,' Allen; 433
BLUE SKY (ACRP project); 140

Blue Springs (SAC photo drone operations); 551, 553 Bohlen, Charles; 33 Bombe; 195-198 Bonesteel, Charles; 464 Bomber gap; 170, 177

HANDLE VIA T ASABLETO FOREIGNNATIO

STEMS JOINTLY .,

fOP SECRET' 'MBRA

614

DOCID: 523682

REF ID:A523682
Withheld from public release Pub. L. 86-36
TOP SECRET tlMBRA

I

E.O. 13526, section 1.4(c)(d)

Boucher, Melville; 262
BOURBON

(project); 15, 16, 18, 158-160,276

Bowles, Chester; 509 Bradburn, David; 408 Bradley, Omar; 161,244
1 _

Brezhnev, Leonid; 457

I
Brindisi - see San Vito;1

I

I

British Security Coordination (BSC); 13 Brownell Committee; 33, 34, 35, 54, 61, 62, 89,168,185,231 George A. Brownell; 33, 34 Brown, Harold; 216 Brugioni, Dino; 329 BRUSA (British-US) Agreement (and Conference); 16, 17, 18, 19,93,159 BrooksAFB, Texas; 11,28,30 Bucher, Lloyd M.; 440-441, 443, 445, 447, 448, 453 Buck, Dudley; 204 Buffham,Benson; 23, 91, 349 (project); 188

BULLSEYE

Bundy,~~rge;289,293,352,473,520,523 Bundy, William; 522 Burgess, Guy; 19, 165, 169 Burke, Arleigh; 46 Burke; Gerald; 479

HA~M1~

NOT

FOREI

615

TqP SKRET \;IMBM

DOClD: 523682
1'6' SECRETI:JMBR.A. E.O. 13526, section 1.4(c)(d) Burke, Joseph; 176, 262, 345 Burrows, William; 410

REF lD:A523682
Withheld from public release Pub. L. 86-36

11....-Bush, Vannevar; 195,204
1

_
---

Cabell, C. P.; 18,29,109,183,358 Callimahos, Lambros D.; 73 Campaigne, Howard; 199 Campbell, William B.; 206, 208
Camp des Loges (Paris); 68

Canine, RalphJ.; 30, 35, 61, 62, 63, 66, 67, 68, 72, 73,,74,77,78,80,81,82,83; 91, 93,101, 102,105-107,109,135,204,206-207,208,209,216,217,227,228,239,240,243-244,269, 279,293,294,296,341

Carroll, Joseph F.; 468, 552 Carter, Marshall S.; 325, 340, 344, 349, 359-360, 368, 377, 385, 387, 392, 410, 411, 436, 445-446,447,448-450,469,471,474,476-477,478,479,552 biography, 357-358 Castro, Fidel; 318

CBNRC (Communications Branch, National Research Council); 17, 208 CCP - see Consolidated Cryptologic Program CDAA (Circularly disposed antenna array, AKA Wullenweber); 138-139, 308-312 Ceaucescu, Nicolae;462 Central Bureau, Australia; 18 Central Intelligence Group; 87,162 Central Office,South Vietnam - see COSVN Chadwell, H. Marshall; 109 Chambers, Whitaker; 164, 166

LEASABLETO FOREIG

liOt(1j;bS

.

'fOP SECRET I:JMBRA

616

DOClD: 523682
E.O.13526, section 1.4(c)(d)
Chamoun, Camille; 237
CHARGER HORSE;

REF lD:A523682
Withheld from public release Pub. L. 86-36
lOP SECRI!T UMBRA

571

Charyk,Joseph;405

I
CHENEY

-:C=h-el:-te-nh:-a-m-,-=-M::-:d--;. 29, 329, 396, 426 (Soviet cipher system); 185

j===-I

Chiang Kai-shek; 38,43,99-100,178,497 Chicago Tribune; 275 Chicksandsl Chifley, J. B.; 18 1118,121,208,266,310,371,459

I
Chosen Christian Cho Yong II; 41, 42, 46, 49, 52, 53 Chou En-Lai; 44 Chun, Richard; 40, 41 Church Committee; 474

I
Cho-Do (island; AFSS intercept site); 50,51,140 College, Seoul (intercept location; AKA Yansei University); 49

Churchill, Winston; 1, 13, 157,214

I
I
Civop program; 69, 268

I
Circularly disposed antenna array - see CDAA

I
Clark Air Base, P.I. (USM79 and USA-57); 29, 91, 127, 128, 133, 138, 306, 311, 365, 498, 503,511,531 COMINT Comnet location; 208 Clarke, Arthur C.; 402, 408 Clarke, Carter W.; 4, 10,23,25,159,161,163,278 Clark,~ark,228-229 ClUIord,Clark;429,438,439,446,448,479,565 Cline, Raymond; 325 COC - see Collection Operations Center Codevilla, Angelo; 453

617

TOP SECRE I UMBRA

DOClD: 523682
I uP SECR!T tJMBRIc

REF lD:A523682

Withheld from public release Pub. L. 86-36

I

Coira, Louis; 387

--------

Collection Operations Center (COC); 349 COLLEGE EYE; 547, 571, 578 . Collins, BGen (Ch NSAEUR); 384

I
. Collins, Samuel P.; 29, 215, 384 Colossus; 197

I

COMFYLEVI;540 COMINTComnet; 207-211, 236, 253

COMIREX - see Committee on Imagery Requirements Command Center (NSA); 346-348, 350, 482 Committee on Imagery Requirements Committee

and Exploitation

and Exploitation (COMIREX); 405 (COM OR); 405

on Overhead Reconnaissance

COMOR - see Committee on Overhead Reconnaissance

I------------------~ Conley, Herbert; 23
Connelly, John; 353 Consolidated Cryptologic Program (CCP); 260,291,294,339-340,341,479,480,534 Control Data Corporation (CDC; successor to ERA); 205 Converter M-229 - see SlOCUM

I
Coordinator

I
of Joint Operations (CJO); 11, 12,25 Corderman, Preston; 12,159 Corderman-Wenger Agreement; Corry, Cecil; 414 12

-

HANDLE Vl:A'fAhE~T KEYHQbE eoMINTCONTROLSYSTEMSJOINTL~ NOT RELEASABLE TO FOREIGN NA I lOI{1tbS

1=OPSECRET tJMBItA

618

DOCID: 523682
.0. 13526, section 1.4(c)(d)

REF ID:A523682

I
SIGJNT

T9P SEtRET I:IMBRA

Withheld from public release Pub. L. 86-36

COSVN (Central Office, South Vietnam), Coverdale, Garrison B.; 470-471

attack on; 573-574

1----;=====;----__
Crete (AKA Irakll·on,·! ,...:::.::..:~:::.:::.:::=~.

==~ I;120 ,133 ,233 ,238 ,298 ,426

1

11...-

-

Critchfield, James; 98 Critic system/report; 253

Criticomm; 253-256, 364

Cryptologic career service; 67, 359
Cryptologic Support Group (CSG); 75, 264,265,342-343,461,475,483 CSE (Communications Security Establishment); 17

. CSG - see Cryptologic Support Group . CSOC - see Current SIGINT Operations Center
1 -'-----

Cudjoe Key, Florida (AFSS Current
SIGINT

SIGINT

site); 331,391 467, 482, 485

Operations Center (CSOC);350-352,

Currier, Prescott; 14 Customer liaison detachments; CXOF;373 Czech crisis of 1968; 453-461 Cyprusl 192, 233, 234, 238, 393, 426,D 7.5-76

DAGER - see Director's Advisory Group on ELINT and Reconnaissance Dak To (battle of, and SIGINT); 560 Da Nang, South Vietnam (USA-32); 504, 512-514, 531, 540, 542, 544, 545-547, 548, 550, 561,578,582 Dancers; 542,543,582 Daniels, Harold; 325, 329

Darrigo, Joseph; 40 Davidson, Max; 255

HANDLE VI

TROL SYSTEMS JOINTLY LEASABLE TO FOREIGN NATIONA

619 .

TOP SECRET UMBIb\

DOClD: 523682

REF lD:A523682

I
Withheld from public release Pub. L. 86-36

fep SECR!T l:IMSAA E.O. 13526, section 1.4(c)(d) Davis, James T.; 504, 506 Davis, John; 347 Davis Station, Saigon, South Vietnam (USM-9J, later USM-626, AKA 3rd RRU); 503, 504, 507-508,513,531,532,542

1'---

_
Agency Indicator

DCA - see Defense Communications DDI - see Delivery Distribution

DDR&E - see Defense Director for Research and Engineering Decentralization plan; 78-80, 135

Deeley, Walter; 217, 350-351, 479, 485 Defence Signals Bureau (DSB); 18, 19

~----------------------------------------~I
Defense Communications .Defense Intelligence

.
366, 552,655

Agency (DCA); 292, 364, 366, 370, 511 .

Agency (DIA); 292, 342, 343-346,359-360,
1

I
Defense Special Missile and Astronautics Delivery Distribution Delmer, Sefton; 412 Delta classification system; 276 Indicator (DDI); 209

Center (DEFSMAC); 345-346, 483

Demirel, Suleiman; 383

Deputy Director for Research and Engineering Desoto Patrols; 515, 520, 522

(DDR&E); 311,338

Dewey, Peter; 495 DlA - see Defense Intelligence Agency
DIANA (COMSEC one-time pads) 52

Dien Bien Phu; 497, 561 Dill, Sir John; 15 Director's Advisory Group on EUNT and Reconnaissance (DAGER); 344,410

I

I
HANDLE VIA TALENT KE -~""""'~E:r-:LEASABLE TROLSYSTEMSJOINTLY TO FOREIGN NATIONA

TOP seC:RET tJM8~

620

f

DOCID: 523682
E.O. 13526, section 1.4(c)(d)

REF ID:A523682
TOP S!C1t1!'f UMBRA

Withheld from public release Pub. L. 86-36

DiRenzo, Victor; 326 Discoverer program; 403-404

I
Donovan. William; 86-7 Drake, Robert; 292 Driscoll. Agnes; 7,276

I

Dobrynin, Anatoly; 324, 329,459, 460, 461

DSB - see Defence Signals Bureau

I__

--.,....--

---J

Dubcek,Alexander;454-456,459,462
Dulles.Allen; 106-107, 177,178,180,233,337,340,341

Dulles, John Foster; 147, 148, 178,233,303, 304 Dunlap, Jack E.; 470-471 Dupont, S.C. (USN-I8); 29 Dyno program; 407 Dyer, Thomas; 73, 241-244 Eachus, Joseph: 14,211
L-I

1

Eamons, Delos; 13 Easter Offensive (1972); 579 Eaton, Frederick, and committee; 344, 411,479-480 Eckert, J. Presper; 198,200 Eddy, Dayton W.; 531 'EC-121 shootdown, 1969; 313, 462-470, 482

Eielson AFB 4l....__

.....J!306, 313

HANDLE VIA TALENT NOT RELEASABLE

STEMS JOINTLY TO FOREIGN NATIONALS

621

TOP SECRET UMBRA

DOClD: 523682
TOP SECRET l:IMBItA

REF lD:A523682

Withheld from public release Pub. L. 86-36
Electronic Warfare; 476,480
ELlNT

I

.and the Baker Committee; 258-259 centralization after Cuban Missile Crisis; 337 and CIA; 109 collection; 112, 122,127. 139 and the Eaton Committee; 479-480 and NSCID 17; 110 and NTPC; 110 organization; 10~l09, 228, 343-344 origins and British organization; 108 and overhead satellite collection; 403-408 .

.

I

.transfer to NSA; 260-263

I
Elmendorf AFBLI __ Ely,R. R; 14

I
--11131; 132, 311

Enderlin, Arthur; 206:208, 241, 255 Electronic Research Associates (ERA); 197-198, 270 Engstrom, Howard; 184-185, 197-199 biography, 270-271 ENIAC; 197-199
ENIGMA; 1-2,14-15,135,195-196,257,276

Ennes, James M.; 438 ERA- see Electronic Research Associates . Erskine, Graves B.; 85,86,109,126,231,261,268,338 Erskine, Hugh; 68 Ervin, Samuel; 359
1 -

Examination Unit, National Research Council; 17
EXPLORER

(project); 538, 581

TOP SECRET UMBRA

622

'.

--

-

-

DOClD: 523682
E.O. 13526, section1.4(c)(d)

REF lD:A523682
TOP 5iCRET UMBRA

Withheld from public release Pub. L. 86-36

FACTOR (project); 571-572

Fairbanks,

Alaska (USM-7); 9, 71,131

Far East Combined Bureau - (FECB), Singapore; 18
FARMER (projected computer); 202
1 _

FBIS - see Foreign Broadcast Information Service FCC - See Federal Communications Commission

FECB - See Far East Combined Bureau Federal Communications Federal Communications
FEEDBACK (project); 402

Actof1934; 2~2, 273, 274, 474 Commission (FCC); 4,103; 107-108,275

Felfe, Heinz; 412-413 Felt, Harry; 269 Fenech, Henry; 183 Ferret flights; 139 Field Operating Manual; 77

1

---

Finlay, Richard E.; 450 Fisher, Robert; 553, 555 Fish', Hamilton; Fitzgerald, Fitzpatrick, 158

Edward; 353 Joseph; 92

Fleet Radio Unit, Melbourne (FRUMEL); 7 Fleet Radio Unit, Pacific (FRUPAC); 7 Fleming, Ian, 86 Flexscop; 371 Flexowriter; 381

FLR-9; ,304,308-312,86,387,500-501 FLR-12 (AKA GLR-1); 308, 310-311 FM~AC - see Foreign Missile and Space Analysis Center

HANDLE VIA TA ~_-NtlIT"Rl:L[iE~ASABLE TO FO

TROLSYSTEMSJOINTLY

623

TOP SECRETUMBRA

DOClD: 523682
TOP SECRET YMBRA

REF lD:A523682

~.O. 13526, section 1.4(c)(d) Foreign Broadcast Information Service (FBIS); 91, 102-103

I

Withheld from public release Pub. L. 86-36

Foreign Missile and Space Analysis Center (FMSAC);344 Forrestal, James; 23, 25, 289 Fort Knox (move to); 243-244 Fort Lewis. WA. (60th Signal Service Co.);40 Fort Meade (move to); 27,73,241-250 . new communications facility; 209 and the Yankee Alert; 236 Foster, John; 338 Foster, William C.; 244 r-F_ra.:..n_kf:_u:....rt~I~========:=;_.........Jh ~8, 83-84, 237, 2971L...-_

11....---__

FRD-10-see ANIFRD-10 Freedom ofInformation Act (FOIA); 167

. Freeze, James; 573 Friedman, William F.; 1, 10. 13.14. 15,67,71,73,158,246,259,274,276,476 Friendship Annex (FANX); 294-95, 297, 360 Frost, Laurence; 183,269,270,294,296,338,340-341,358,398, biography. 292-293 FRUMEL - see Fleet Radio Unit, Melboum~ FRUPAC - see Fleet Radio Unit, Pacific Fubini,Eugene;216,339,340,348,349,~59,404,479,511 Fuchs, Klaus; 19,164.167 Fulbright, William; 449, 522-523
1 ------

502,506

Galbraith, John Kenneth; 304 Gamma classification system; 276 Garafalo, Caterino; 544 Gardner, Meredith; 161-163 (}ayler, Noel; 370,402, 409, 469, 568,570,580

TOP SECRET UMBRA

624

DOCID: 523682
E.O. 13526, section 1.4(c)(d)

REF ID:A523682

I
biography, 476-471 97-98,169,412-413 of Rear Services (GDRS); 500

TOP SECRET "MIR
I

A

Withheld from public release Pub. L. 86-36

GDRS - see General Directorate of Rear Services Gehlen, Reinhardt; General Directorate

I
Geyer, Hans Joachim; 412 Ghormley, Robert; 13

I
Germany as a SIGINTpartner; 97-98, 412, 415 Giap, Vo Nguyen; 559,562,564

I

I
Intelligence Committee

GLR-1 (later changed to FLR-12); 308, 310-311 GMAIC - see Guided Missile and Astronautics

GOLDBERG

(rapid analytic machine); 198

Goldberg, Arthur; 447 Gold, Harry; 164 Goodfellow AFB, Texas; 133

Gouzenko, Igor; 18,161, 166 Grab program; 407

Gray phone system; 207, 209 Gribkov, Anatolii; 330

Gruenther,

Alfred; 25

GuamL...1 __

-ll 29. 111,

188. 306

HANDLEVlAT ~~ftTl'"R'ErLE::EASABLE

CONTROL SYSTEMS JOINTLY TO FO

625

rop SEeR&; lIMBRA

DOCID: 523682
TOP SECRET UMBRA

REF ID:A523682.

~.o. 13526, section
Withheld from public release Pub. L. 86-36

1.4(c)(d) .

I

Guam Area Study; 552-553 Guided Missile and Astronautics Intelligence Committee (GMAIC);177 Gurin, Jacob; 169 Haig, Alexander; 485 Hakata, Japanl Hallamaa, Reino; 162 Hallock, Richard T.; 160 Hamilton, Victor Norris; 472-473
HAMMOCK

l128, 260, 306

(project); 544-545, 549

I

I
Hanza, Okinawa

I

1111,127,139,139,306,308

Harkins, Paul; 502, 508 Harriman, Averell; 509, 565-566 Harris, Stephen; 440-441 Harrogatel
1 -

Harvest (general-purpose computer); 202-204 Harvey, William; 88,105 (project); 513

HAWKEYE

Hayes, Harold G.; 12, 2:17 Hebern, Edward; 212 Helemano, Hawaii (USM-5); 29 Helms,~chard;353,387,459,467,479,485 Henry, Father Harold; 42
1 -

Herrick, John J.; 516-518, 520
1 -

Herring, George; 498, 579. Hersh,Seyrnour;330,467,487,579

H~QI'O."';;~ LEASABLE TO FO
"FeP SE(ReT t:JMIR:A

626

.

-

----

DOCID: 523682
E.O. 13526, section 1.4(c)(d)

REF ID:A523682

I
Germanyl ----J129,69,80,83,91,112,

TOP SEGET l;JMBRA

Withheld from public release Pub. L. 86-36

Herzogenaurach,

11S; 311, 391

. Higginson, George M.; 294
Hillenkoetter, Roscoe; 89,102
----l

I

Hirota (Japanese General); 49 Hiss, Alger; 164, 167 Hitch, Charles; 291,292 Ho Chi Minh; 496, 497 Hodge, John R.; 36

Holderness,

Arthur; 466

Hollis, Harris W. and the Hollis Board; 475-476 Holtwick, Jack; 81,186

Honeywe11316; 369
Hoover Commission; 64, 71, 228-229, 257,276 Hoover,J. Edgar; 108,165,167

Horner, Jack; 448

House Un-American

Activities Committee; 283

Hovey, Herbert S. Jr.; 506-507
L...-

--.JI
revolt (and
SIGINT

i

Hughes, Thomas; 428 Hungarian crisis); 234-235,239,264,454

Hyland, John; 469 . LAC- see Intelligence Advisory Committee

1a Drang campaign; 530, 532, 534 IATS (Improved AG-2 Terminal System); 369-370, 371 IBM (International Business Machines); 195,198-199,204,368 700-series machines; 202, 204

HA

TKEYHO

627

TOP S!CREf UMBRA

DOCID: 523682
"FOP SECRET I:IMBRA

REF ID:A523682

IDDF (Internal Data Distribution
1

Facility); 371. 372
_

IGLOO WHITE;

570

INFOCON;77 Inglis. Thomas; 108

1
Intelligence Intelligence Internal Information

1'
Advisory Committee (lAC); 102-103 Steering Committee; 39 Facility - see IDDF

Data Distribution

International

Business Machines; see IBM

Intern program (civilian); 359 lraklion - see Crete

Iredell. Milton; 294 Iron Hand (SAM suppression
IRON HORSE

missions); 547

(project and equipment); 371, 549

,

.

1.----

__

1

Jackson, William H.; 33, 231

JCEC - see Joint Communications-Electronics JCIC - see Joint Counter-Intelligence

Committee

Committee

HANDLE VIA TALENT

KEyneS COf4m1'

(;ONTROLSYSTEMSJOINTLY

NOT RELEASABLE

TO FOREIGN NATIONALS

628

OOCIO: 523682
':.0.13526, section 1.4(c)(d)

REF IO:A523682

I

'---~---'-------------JMG - see Joint Mechanization Group
1 ------

Withheld from public release Pub. L. 86-36

JCS Dir. 2010; 26 JCS Memo 506-67; 343,448,475,534

I

JN25; 1 John, Otto; 412 Johnson, Harold; 358
Johnson, Kelly; 180

Johnson, Louis A.; 25 Johnson, Lyndon Baines; 231, 273,303,353-357,382,386,387,428,430,432,436,437,
446,448,455,460,461,462,468,473,479,485-486,504,515,520,522,523,529,548,564, 565,573

Johnson, Nels C.; 476 Johnston, Stanley; 275 Joint Communications-Electronics Committee (JCEC); 13,32,208,215 Joint Counter-Intelligence Center (JCIC); 87
1 -------

Joint Mechanization Group (JMG); 362 Joint non-Morse Acquisition Control Center (JNACC); 348, 349 Joint Sobe Processing Center (JSPC), Okinawa; 85,128, 296, 297, 360, 415, 511,532

I

I

Jones, R. V.; 108 Josephson Junction technology; 368 JSPC - see Joint Sobe Processing Center Kadena AFBI Kahn, David; 97,473 Kami Seyal--------1440, KAL-007 shootdown; 447
1128,306,390,531,540

443, 445,464,466,469

HANDLE VIA TALEN
j,I:.Q:f--R£1cEAASAS BLE TO FOREI

TROLSYSTEMS J<?INTLY
S

629

TOP SECRET UMBRA

DOClD: 523682
TOP SECRET I;IMBRA

REF lD:A523682

.0.13526, section 1.4(c)(d)

I

Withheld from public release Pub. L. 86-36

~_---=-=--:---;:::::========::::;--:-:--:--:--:--:-::-~I
Karamursel, Turkey 1

---'1; 122, 125, 183, 208, 238, 260, 298, 311,1__

----'

I
Keating, Kenneth; 323-324 Keegan, George; 562,580

I~
~ennedy,Jack;578

---

Kelly AFB, Texas; 30, 31 Kennan, George F.; 157 Kennedy, John F. and the Kennedy administration; 149, 178,289-293,304, 314, 320, 352, 358,361,384,385,386,401,499, 502,509,510 Cuban missile crisis; 324-332 Kennedy, Robert F.; 565 Kenney, George C.; 48
1 ---

( '1-;

, \ l~-- Vi -

KGB cipher traffic; 160, 161-168 KG-13;219,366,380
I 1

Khan, Agha Mohammed Yahya; 387 Khan, Mohammed Ayub - see Ayub Khe Sanh and
SlGlNT;

561-562

- c

/"

Killian, James R. and the Killian Board; 179-180,229-230',403 Kim Il-sung; 38,439, 463, 470 Kim Se Won; 41, 42,52,53 Kim, Y. P.;40,41 King, EarnestJ.; 5, 6 Kirby, Oliver; 83,183 Kirkpatrick, Lyman; 263-264, 276-277

I
Kit Kat; 511

1

Kissinger, Henry; 289, 467,485-486, 581, 582

___ -I'RH

TALENTKE RELEA

MlNT CONTROL SYSTEMS JOINTLY NATIONALS

TQP SECRET UMBRA

630

DOCID: 523682
E.O. 13526, section 1.4(c)(d) Withheld from public release Pub. L. 86-36
Klein, Maurice; 284, 294
KLIEG~GHT; 267, 350

REF ID:A523682
TOP SECRET UM8RA

Klocko, Richard; 370

I

KL-7; 212, 568

.

Korean War (and SIGINT); 30, 32, 33, 36-56, 61, 63, 64, 69,77, 78,140,227 Kosygin, Aleksey; 431--433, 436 KO-6;221 Krivitsky, Khrushchev, Walter; 164, 166 Nikita; 148-149, 281, 303, 313, 318, 323, 328, 330

Kullback, Solomon; 10, 67

KW-7; 379, 451 KW-26; 209-210,219,222, 255,379

1__

-

KY-l; 221 KY-3; 221, 347, 380 KY-8I28138; 380-381 KY-9;379 KY-11; 220-221 KY-57/58; 381 KY-67;381
LACE BARK (project); 568

Lacy, Gene; 443 Ladd, Mickey; 163 Laird, Melvin; 467, 476 Lamphere, Robert; 163-63,166'

Lam Son 719. and S[GlNT support; 576, 579 Land,Ed~n;180,230-231

HA

ONTROLSYSTEMS JOINTLY ALS

631

fOP

SECRET IIMB8A

DOClD: 523682
TOP SECRE'f I:IMBU

REF lD:A523682

Withheld from public release Pub. L. 86-36
1 _

Larsen, Finn; 216, 385

Lebanese crisis of 1958; 237-238, 292, 425 Le DucTho; 581, 582

~]day,Curtis;402,513 Lemnitzer, Lyman 319

Liberty (TRS); 391,396,429,432-439

I

I

lJGHTNING(projected computer); 204, 257 Linebacker (operation); 579-581 lJTTLE CLOUD (airborne collection project); 383, 386

I------------------------~
London SIGINTCentre (LSIC); 16

1-----------:-----Lon Nol, General; 572, 574 Lord, Richard "Dick"; 483, 484 Lothian, Lord (Phillip Kerr, 11th Marques of loW an); 13 . Lourdes (Soviet SIGINTsite in Cuba); 331 Lovett, Robert; 35 Low-level voice intercept (LLVI); 46,47,48,48,54,536,542,543,561,573 Lowman, David; 322

LSIC - see London Signals Intelligence Lundahl, Arthur; 326, 403 Lynn;

Center

Roy; 29, 30,72

MacArthur, Douglas; 2, 36, 40, 41, 43,44,45,46,68,99,268

I

I
HANDLE VIA TALENT K
~-NO'T

OLSYSTEMSJOINTLY TO FOREIGN NATIONALS

RELEASABLE

TOP SECRET U MIRIe

632

J,

DOClD: 523682
E.O."13526, section 1.4(c)(d)

REF lD:A523682

I

TO' SkRET I:JMBI:lA

Withheld from public release Pub. L. 86-36

Maclean, Donald; 19, 165,167, 169 MacMillan, Harold; 237 Maddox (U.S. destroyer); 516-518, 519-523

I
Mao Tse-tung; 38, 43, 44, 178 " Manson, Grant; 16

I

,-----------Manor, LeRoyJ.; 578
Manual of U.S. COMINT Operations - see MUSCO Manual ofU .S. SIGINT Operations - see MUSSO

Marine Guard detachment at NSA; 73, 247 Market Time (Navy maritime operation); 552, 572

Marr-Johnson,

Patrick; 17

Marshall, George C.; 5, 6, 43, 44, 357 Marshall-King Agreement; 5, 6 Martin, William; 74,280-284,294,296,470,473 Martin, William 1.; 433

I
Mathews, Mitford; 378, 410 Mauchly, John; 198,200

Withheld under statutory authority of the Central Intelligence Agency Act of 1949 (50 V.S.C., section 403g)
167 "

McCarthy,Joseph,
1__ -

and McCarthyism;

McChristian,

Joseph A.; 530
I

McCone, John; 183,319,324,326-329,

340,358,409,520

HAI5o==ii"~Y
633

ABLE TO FOREIGN ~

TOP 5ECRiT lJMBRA

DOCID: 523682
E.O. 13526, section 1.4

REF ID:A523682

mpr~D~rMBBA

c (d

.

Withheld from public release Pub. L. 86-36

McCormack, Alfred; 4,15 McGonagle, William L.; 433, 438 McKinsey Study; 239-240, 270, 294 McManis, David; 356, 458, 461, 467-468,485,540 MeNaDnara,R4bertS.; 291, 292, 330, 338, 340, 341, 342,343, 348, 352,355, 361,364,382, 404,432,436,446,479,520,523,530,548 McNarney, Joseph T.; 25 Meademobile; 24S-249

I
Menshikov, Mikhail; 146 MeJiwith Hill- see Harrogate Menzies, Robert; 19 Menzies, Stewart; 158 Merchant, Livingston; 479

MESSINA;

171
1..1

Microwave intercept; -:-=~=-=in Cuba; 319, 320,327 Mikoyan,)unastas;147,317 Minh, General "Big"; 510 Misawa, Japan

----l

I

1'80,127-128,133,143,306,311,466

Missile gap; 170, 177-178, 320 Mitchell, Bernon F.; 74, 280-284, 294, 296,470,473 Mitchell, William; 479 MI-8; 8, 99, 158

I~ __
Moo re, J••• ph

Monkey Mountain - see Oil N ang Moody,Juanita; 322,325,330,361,362-363

ASABLE TO FOREIGN GIONAtS

r-;·S47

TOPSEEAETt:JMBRA

634

OOCIO: 523682
E.O. 13526, section 1.4(c)(d)
r----------,

REF ID:A523682

I

Ioe SiCHET UMBRA

Moorer, Thomas; 450, 520 MOP-95;476 ~orrison,John;82,232,343,445,467,468,469,482-483
1 1

Withheld from public release Pub. L. 86-36

Mount Vernon Seminary; 8

Murphy, Edward; 440
1

:-:---:--=---_----...JI

Murphy, Robert; 146 Murray, Edward; 41, 42 MUSCO (Manual ofU .S. COMINT Operations); 77 MUSSO (Manual of U.S. SlGINT Operations); 77 M-209;213,218,568 Nakhon Phanom, Thailandl 1580,582 . Nasser, Gamal Abdel; 232-233, 237, 425-426, 429, 431, 433, 436 National Bureau of Standards (NBS); 199-200 National Cash Register Co.; 195-198 National Crypt.ologicSchool (NCS);27,294,360 National
ELINT

Plan; 337,343-344

National Intelligence Resources Board (NIRB);480, 481
1 I

National Reconnaissance Office(NRO);405, 407 National Security Council (NSC);33,35,56,102,253,261 National
SIGINT

Operations Center (NSOC);176,267,314,350,469; 482-483

National Technical Processing Center (NTPC); 110, 261 Naval Computing Machine Laboratory (NCML); 195, 197 Naval Research Laboratory (NRL); 138, 222, 396. 407 NRL Mixer; 222

I
Nave, T. E.; 18

I.

Naval Security Station (NSSor "Nebraska Avenue"); 8, 9,12,15,17,27,32,61,71,72,73, 74,81,87,109,110,187,195,198,206,207,~09,215,216,241,243,24~-246,294

'::::=:x.~~
. BLE.v, •..•
.I ••••••

v.~

.

635

tOP SECRET I:IMBRA-

DOCID: 523682
Tap SECRETtJMBItA b.O. 13526, section 1.4(c)(d)

REF ID:A523682

I

,-----------,

Withheld from public release Pub. L. 86-36

NBS - see National Bureau of Standards NCML - see Naval Computing Machine Laboratory Nebraska Avenue-see Naval SeeurityStation

I
NESTOR;

I
381, 568

Newton, Robert; 452 Ngo Dinh Diem; 497,498, 509, 510 Nhon, Pham Van; 542, 568 Nichols, Major; 41, 42, 49

.

i

Nimitz, Chester; 2, 68, 268 NIRB - see National Intelligence Resources Board Nitze, Paul; 534 Nixon, Richard M.; 147,357,387,467-468,485,565,566,567,570,572-573; 584
NOMAD

579, 581,582,

(computer); 201-202

Norstad, Lauris; 479 .

Novaya Zemlya (Soviet nuclear test site); 177 Novotny, Antonin; 454 NRL - see Naval Research Laboratory NRO - see National Reeonnaisance NSAAL (NSA Office Alaska); 68 NSAEUR (NSA Office Europe); 68, 264, 265 NSAEURIISS; 265, 342-343 Office .

NSAP AC (NSA Office Pacific); 68, 268, 296 NSAPAC NOG (NSAPAC Operations Group), 343

HANDLE V

CONTROL SYSTEMS JOINTLY

T RELEASABLE TO FOREIG

TOP SECRET tJMBItA

636

DOClD: 523682
.LO. 13526, section 1.4(c)(d)
.----------,

REF lD:A523682

"FeP SECRET UMBRA
184, ~27, 382

Withheld from public release Pub. L. 86-36

NSA Scientific Advisory Board (NSASAB) and predecessors; NSAUK (NSA Office United Kingdom); 68 NSC - see National Security Council NSCID 5; 53,90,93,107 NSCID 6; 68,107,261-262,263,310,337,405,476,478 NSCID 7; 254-255 NSCID 9; 35,68,75,76,77,90, NSCID 17; 110, 228 NSC 168; 216 NSOC - see National SIGINT Operations Center NSS - see Naval Security Station NTPC - see National Technical Processing Center 107,109,216,261

I
Oehlert, U.S. ambassador to Pakistan; 387 Office of Policy Coordination (Ope); 101 Office of Strategic Services (OSS)O

I
86,1------

Office of Special Operations (050), 85, 86, 231, 271, 338 Offutt AFB

I

1390

O'Gara, Jack; 339 Ogier, Herbert L.; 517

1
Oliver, Donald; 538

_

Omnibus Crime Control and Safe Streets Act; 274, 474

OPC - see Office of Policy Coordination Operation Plan. 34-A; 511, 515.518,542 Operations Security (OPSEC); 555 Operation Starlight; 530, 532

KEYHO

CONTROL SYSTEMS JOINTLY

BLE 637

ALS

;gp SECRET t:JMBrtA

DOCID: 523682

REF ID:A523682

Withheld from public release Pub. L. 86-36
Osan, Korea (uSA-31); 464, 466, 467-469 OSO - see Office of Special Operations OSS - see Office of Strategic Services Overhead intelligence collection; 179-184,230-231,402-411,479,480 Owens, Robert G.; 548 Pacific Experimental Packard, David; 453 Facility
(PACEXFAC);

268

I~==================:::;---Panamal<-h 306
Panikkar, K. M:; 44 Park, Chung-hee; 439 PARKHI.LL; 380 PARPRO (Peacetime Aerial Reconnaissance Program); 468

I
PDP-1I10; 368, 369 Peacetime Aerial Reconnaissance Pearl Harbor hearings; 273· Pendergrass, James T.; 198-199

I
Program - see P ARPRO

I
c=Jawar, Petaluma, Pakistan

I

I

II

----J~304, 308, 311, 3~'--

__

1396,

c:J

California (USM-2, AKA Two Rock Ranch); 29, 306

Petersen, Joseph Sydney; 279-280 Petsamo (Finnish town; locus of captured KGB codebooks); 162 PFIAB - see ~esident's Pham Van Dong; 559 Foreign Intelligence Advisory Board

HANDLE T RELEASABLE

NTROL SYSTEMS JOINTLY TO
FOREI

TOP SECR!T tJMaRA

. 638

DOClD: 523682

REF lD:A523682

~.o.13526, section
Withheld from public release Pub. L. 86-36

1.4(c)(d)

I
"Kim"; 165-167,169

rap S!CftET UMBRA

Philby,H.A.R.

Phillips. Cecil; 352. 362. 370 Phu Bai. South Vietnam (USM-808, USN-414T)); 504, 506; 511, 513, 515, 516, 618, 520, 530-531,534,542

,'------PHYLLIS ANN; 532-534

Pike, Otis and the Pike Committee; 449, 468-469, 474

I
Pinkston, Frank; 362
PLANTATION;

I
202

PIWO- see Prod Watch Office

Pleiku, South Vietnam (USM-604); 504,531,534,560,561,568,573,581

I
Pliyev, Issa; 330 Pollard, Jonathan Pol Pot; 574 Polygraph; 73-74, 282, 283, 471 Poppy program; 407 Jay; 280

I

I'----_~_Powers, Francis Gary; 175,177,180-184

I
President's

I
Foreign Intelligence Advisory Board (PFIAB); 263, 292,337-338, 344, 4~2, 564 Watch Office, PIWO); 346-

Prod Watch' Office (PWO; later changed to Prod Intelligence 247 Professionalization program (civilian); 296 Program C; 407-408

I
~blic

I
Law 88-290; 473 396,439-453,467,468,469,475

Public Law 86-36; 272. 273

Pueblo (TRS); 217,391,395,

I
PURPLE

I
and
PURPLE

Analog; 1, 14

I

HANDLE VIA TALENT KE
RELEASABLE

OLSYSTEMSJOINTLY
TO FOREIGN NATIONAL

639

TOP SECRET UMBRA

DOClD: 523682·.
TOP SECRET tJM8RA E.O. 13526, section
,----------,

REF ID:A523682

1.4(c)(d)

I

PURPLE DRAGON; 551-555

Withheld from public release Pub. L. 86-36

PWO - see Prod Watch Office Python systems; 213, 568 Pyong- Yong-l?o(A.FSS tactical voice intercept operations); 49

I

I

Quarles, Donald; 254, 258, 271

IL..QUEENBEECHARLIEIDELTA~ 513, 545, 547

I'

Raborn, William; 358

. Radio Analysis Group, Forward (RAGFOR); 7 Radioprinterexploitation; 169-170,177,178, 184,185-186

Radosh, Ronald; 167
RAGFOR - see Radio Analysis Group, Forward
RAINFALL

program; 409-410 Station, Thailand

Ramasun 573,582

(AKA Udorn;

I

, I; 311-312,500,501,511,570,

Rand Corporation;'402

Raven, Frank; 362-363,483 RB-47 shootdown, 1960; 314 Re-IS0 shootdown, 1958; 144-147,282,313,468 Ream, Joseph; 231, 270 Red Crown; 581 Redman, John; 208 Jteceivers(radio); 134

IL---_
"FOP SECRET tJMBRA

640

DOCID: 523682
E.O. 13526, section 1.4(c)(d)

REF ID:A523682

I
Corp.; 198
(COMINT); 229, 341

TOP SECRET UMBRA

Remington-Rand

Withheld from public release Pub. L. 86-36

Reporting (SIGlNT); 69-71 Requirements

Reynolds, Wesley; 163, 278, 284 Rhee,Syngtnan; 38, 39, 41,43,53

Rice, Kenneth; 297 , Rivers, Mendell; 468
RIVET GYM

(ACRP project); 571

Robertson, H. P. (committee); 67,109, 227-228 Robertson, Reuben, and committee; 128, 179,206,253,259,270,271,297,306,393

___

---'I

Rogers, William; 467, 485 ROGUE (Remotely Operated General Use Equipment); 200-201, Rolling Thunder (USAF operation); 529, 553-555

Roosevelt, Franklin D.; 13, 157, 159,164,214,497 Rose BOwl (RC-47); 511 Rosen, Leo; 14 Rosenberg, Julius; 164, 167 Rosenberg, Ethel; 165, 167

I
Rota Spain! Rothwesten,Germany

I
Rostow, Walter; 289, 353, 354, 428, 431-432, 436, 437, 446, 455, 458-459,461, 485,509 1432 462, 479,

I

~311,

362, 391,455 186,271,294,360

Rowlett, Frank; 10, 12,23,67,87,88,89,96,93,95,105,159,161, Royall, Kenneth; 23 Rubble Pile - see Teufelsberg Rubel, John; 338, 502

641

TOP 5ECRETYMSAA

DOClD: 523682
I

REF lD:A523682

TOP SEC:Rel' tlMBRA' -E-.O-.-1-3-S2-6-,-se-c-ti-o-n-l-.4-( c-)-I

Withheld

from public release Pub. L. 86-36

RUNWAY program; 408-409

Rusk,Dean;

325,446, 520

Russell, Richard; 280 Ryan, John; 580 '
RYE; 368

Saadi, William; 370

1'--Safford, Laurance;

SABERTOOTH (project);1

----' ;-------,
7,13,158,271,276

~542

Salinger, Pierre; 323 Samford,JohnA.; Samsunl 107, 204, 209, 231, 254, 261, 268, 269, 271, 296, 341, 358 1122,174,310,311,385

Sangley Point Naval Station, Philippines;
1

142
--

San Miguel, Philippines San Vito (AKA Brindisi;1
SARACEN; 582 .

(USN-27; Subic Bay); 306, 498, 516, 520, 550 1120,133,311,371

1_--,-Satellite Operations Center (SOC); 405,409

.

I

I
1

SAVILLE; 381
------'

I

SCAT - see Support Coordination Advisory Team

Schultz, Charles; 479 Schulz, Lester R. and the Schultz-Eddy Agreement; Schukraft, Robert, 87 SCientific Research Institute 88; 171 SCOCE - see Special Committee on Compromising Emanations Second echelon system; 83, 265, 296, 348-349, 360 Security, cryptologic; 73-75 badge,73

531-532

HANDLE VIA TALE ~JI.(»il:ttl~;UBLE

OL SYSTEMS JOINTLY TOFOREIGN NATIONA

TOP SECRET tlMBItA

642

DOCID: 523682
E.O. 13526,section 1.4(c)(d) Withheld from public release Pub. L. 86-36

REF ID:A523682
TOP 5ECR!T tJMlfI1(

'------'---------'

classification system; 275-277 compromise cases; 274, 277-284 destruction of classified waste 74-75 and product dissemination; 229

Semipalatinsk

(Soviet nuclear weapons test site); 176

SHAMROCK

(Project); 29

Sharp, Ulysses Grant; 520,553, 555 Shawcross, William; 495 Shedden, Sir Francis; 19 Sheehan, Neil; 583 Sheldon, Huntington; 325, 342, 405, 410- -~~306

-=S=he=m:y~a:,~A=la:S=k:a~I============~~ __ ~

1__

-

Shockley, William; 200
SHORTHAND

(project); 542-,543 .

Showers, Donald M.; 552

Shukeiri, Ahmed; 425 Shu Lin Kou, Taiwan 1-------------------------1310 Sidey, Hugh; 448 SIGABA;212 SIGCUM (AKA Converter M228); 213
SIGINT SIGINT SIGINT SIGINT SIGINT

Committee (ofUSIB); 341 Digest (AKA SIGSUM); 332 Overhead Reconnaissance ~tellite System ControlSubcommittee see SSSC (SORS); 405

Support Group; 532

.HANDL RELEASABLE

NT CONTROL SYSTEMS JOINTLY TO FOREIGN

643

TQP S&CRiT UMBRA

DOCID: 523682
'Fep SECRE'FI:IMBRA

REF ID:A523682

E.O. 13526, section 1.4(c)(d)

I

Withheld from public release Pub. L. 86-36

SIGINTWorking Group of COM OR; 405 SIGSALY; 214, 220 SIGSUM-SeeSIGINT SIGTOT~213 Sihanouk, Prince Norodom; 572 Digest

Sillitoe, Sir Percy; 18 811.0;202

Silver Dawn (ACRP program); 547 SUikov, Abraham; 10, 14,67,215 Sinop,Turkeyl Skaggs Island, California; 29,159

f385

,c==J

I
1 _

I

SMAC - see Space and Missile Analysis Center

Smith, Bromley; 479 Smith, Walter Bedell; 33, 87,89,90 SMTIG - See Soviet Missile Technical Intelligence Group Snyder, Samuel; 199-200
1 -

Sobe, Okinawa L...I _----.II AKA Torii Station); 44, 69, 128, 306, 371

I

I

Songbird reports; 550

SOLO;200 Sommers, Gordon; 23, 478 Son Tay Raid and SIGINT;576-578 SORS - see SIGINTOverhead Reconnaissance Committee

I
Soviet Missile Technical Intelligence Group (SMTIG); 176

I
South Vietnam as a SIGINTpartner; 411, 415, 498, 502, 503, 509, 566, 568-570, 582

~DLE

\I LA =~!T

KFytlOl E C:Ql,UNI CONTROL SYSTEMSJOINTLY RELEASABLE TOFOREIGN NATtoNAIS

TOP U(RE'F UMBRA

644

DOCID: 523682
I E.O. 13526, section 1.4(c)

REF ID:A523682
Withheld from public release Pub. L. 86-36
TOP SECRET l.IM8AA

Space and Missile Azialysis Center (SMAC); 176,205,266, 345
1 _

Spacol (space collection); 296 Special Committee on Compromising Emanations (SCOCE); 216-217, 22~ 223

Special Research Branch (SRB) 32 Special ~urity Officer (SSO); 1-2,82,96,264,342, 366

Special Seeuzity Technical Branch (SSTB); 566. 568. 570. 582 Spinteomm; 366
SPIT (project; Special Intercept Typewriter);

361, 362

Sputnik and crytology; 126, 177,211,253,397 SM - see Special Research Branch . SSO - see Special Security Officer SSSC (SIGINT Satellite System Control); 408-409 SSTB - see Special Security Technical Branch Stalin. Joseph; 178 State-Army-Navy State-Army-Navy Communications Communications Intelligence Boa'rd(STANCm); Intelligence Committee 7,11,16,17,108

(STANCICC); 7, 11

Stella Polaris; 162-163

Stephenson,

William; 13

Stern, Sol; 167 'Stimson, Henry; 274 St. Lawrence Island (AKA Northeast Stockdale, James B.; 515,518 Cape.] __
--.J~

132, 310

Stone,EarIE.;23,24,25.28,29,61,66,67,102,206,241,243,278 Stone Board; 23. 25
STONEHOUSE (at Asmara, Ethiopia); 398, 400-402

.

Strauss, Franz Joseph; 413
STRA~AT;360,366-368

Strong, George V.; 13. 14

I

I
HANDLE VIA TALENT KEY SYSTEMS JOINTLY ~_-N6T"R1ffL:E:EASABLE TO FOREIGN NATIONALS

645,

lOP SECRET tJMSRA

DOCID: 523682
lOP SECRETUMlRA

REF ID:A523682

~.o.13526,

section 1.4(c)(d)

I

Withheld from public release Pub. L. 86-36

Subcommittee

on Compromising Emanations

(SCOCE); 381

Subic Bay - see San Miguel Sue~ Crisis; 232-235J
1

425,D

432

I

I
I .

Support Coordination Advisory Team (SCAT); 549

II~=======================;
Sylvania (contractor for 466L system); 310, 311 Symington,Stuart; TABLON;369

170, 177

Taegu (siege of, and

SIGINT);

40, 43, 52

I_-'----~
Talent-Keyhole

program; 404-405

-support; 570

I
Target Exploitation

I
(TARE X); 3, 551

Tan Son Nhut AFB, Saigon (AFSSARDF operations); 534

Task Force Alpha (TF A) and

SIGINT

Taylor, Maxwell; 53,291,502,509,511,564 Taylor, Telford; 15 TDS - see Teletype Distribution TEABALL(operation); 579-581 TEBAC - see Telemetry and Beacon Analysis , Committee . . System.

I

Technical Research Ship (TRS) program; 314-316,391.395-397,426,429,440.453 and Cuba; 320-322 collection; 1313,1 early ELINT mission; 110

I

Telemetry

I

_

ONTROL SYSTEMSJOINTL

Y

~.J;Ie't1tEl:EASABLE

TOFOREIGN NATIONA

TOP SECRET UMBRA

646

DOClD: 523682
E.O. 13526, section 1.4(c)(d)

REF lD:A523682

I

.+011 SECRET UMBAA

Withheld from public release Pub. L. 86-36

from ESVs; 397-398 organization and classification; 262 andoverheadcollection;406,407,409-410 use with Soviet rockets (AKA Messina); 171

Telemetry and Bearon Analysis Committee (TEBAC); 263

I
Teletype Corporation; 198,209,255 ~odel19;206,208,268 Model 28; 268 Model 35, 368 Teletype Distribution

I

System (TDS); 371
376, ~81-81

TEMPEST (COMSEC project); 106,217,221-223,

Tempo R (Trainj,ng School); 72, '14. Tempo X (Training School); 72

I~
Tet Offensive; 559-565, 583 Tevis, Charles; 346,410 TF A - see Task Force Alpha

---

Teufelsberg, Berlin (AKA Rubble Pile,!'-

I; 298, 300

1'--__
TIDE; 350, 485

Thebaud, Hewlett; 159 Thieu, Nguyen Van; 559, 568, 579, 582 TICOM - See Target Intelligence

Committee

Tiltman, John; 14,17,93 Title 18, U.S.C. 798; 273-274, 279 Tizard, Sir Henry; 13

1_Tonkin Gulfincident; Tonkin Gulf Resolution; 515, 522-523
HANDLE

506, 515-523, 529, 583

Tordella, Louis; 67,89,90,91,97,159,183,199,254,261,263,353,377,

387,436,482

L SYSTEMS JOINTLY

SABLE TO FOREI

647

lOP 5liCRET l:IMBItA

.1

I

_

DOClD: 523682
TO' SECRET l:JMBRA E.O. 13526, section 1.4(c)(d)

REF lD:A523682

I
biography, 271-272

Withheld from public release Pub. L. 86-36

Touchdown (military operation); 551 Trabzon, Turkey

I

l122, 174,298,310,385

Training (cryptologic); 71-73, 95 Travis, Sir Edward; 15, 17 Triantafellu, Roclde ("Rocky"); 532

TRS -:-see Technical Research Ship TRS Special Communications System (TRSSCOM; AKA Moon Shot); 396

Truman, Harry S.; 16, 19~33, 36,40,45,56,87,102,157-158,215,289,497 Truman Memorandum; 35, 61,272,274 Truman Doctrine; 12.2., 157
1 _

Tucker, Gardner; 216 .
1 ----,-_-

Turing, Alan; 15

Turner Joy

(U .S. destroyer);

518-523

Twining, Nathan;

148, 237

Two Rock Ranch - See Petaluma

Ubon, Thailand Udom, Thailand

I

~512, 582
- see Ramasun Station 17,19, 235

UKUSA (United Kingdom, USA) Agreement; Ultra (codeword); 276

United States Communications Intelligence Board (USCIB); 18, 19, 32, 35, 44, 53, 56, 87, 89,91,92,95,99,100,102,108,109,110,144,147,204,218,227-229,253,259,261,279 United States Communications United States Intelligence Intelligence Coordinating Committee (USCICC); 12

Board (USIB); 255, 262, 331,341,381,405,413,499-500,503 Board; 32

United States Joint Communications United States Communications

Security Board (USCSB); 215-216, 222- 223,381-382

I

I
HANDL NOT RELEASABLE TO FOREIGN ROL SYSTEMS JOINTI.. Y

1'61' SEeR!T UMBRA

648

DOClD: 523682
:.0.13526, section 1.4(c)(d)
,---------,----,

REF ID:A523682

I

Withheld from public release Pub. L. 86-36
L---'

Univac Corp. (AKA Sperry Rand Univac); 200, 204 4901494 series; 205, 368, 370, 485 URC-53; 547 USCIB _ see U.S. Communications USCICC - see U.S. Communications USIB - see U.S. Intelligence Board Security Board Intelligence Board Intelligence Coordinating Committee

USCSB - see U.S. Communications

U Street Facility (Cryptologic School); 27
1__

---:--;:::::=====::::::::...._----U-2 intelligence collection; 580,582

I

~391, 403,

Vance,Cyrus;358,359,520,534 Vandenberg, VanFleet, Hoyt S.; 10,26,28,78,87

James A.; 36

Vann, John Paul; 583 VENONA(project); 19,160-168,185,186,276,27.8,284 Verkhnyaya Salda (Soviet missile site); 178

Vinh Window; 539-540, 542, 579, 583 VlNSON;381 \ Vint Hill Farms (USM-l), Va.; 8,28,135,306,318,369.402 Vogt, John; 580
1 -

Voice intercept; 227, 506, 542 VQ-1 and VQ-2; 139, 142,463-470,550
, ..

Wakkanai,
1

Japan
-

Walker, Walton; 43. Wallace, Henry; 159 Walter, Francis E. and the Walter Committee; 283-284 Walt, Lewis; 530

HANDLE VIA TALE ~_N&T"RlrorAASA:s.

LSYSTEMS,fOINTLY BLE TO FOREIGN NATIO

649

TOP SECRET UMBRA

DOClD: 523682
TOP SECRETUMBItA E.O. 13526, section 1.4(c)(d)

REF lD:A523682

I

Withheld from public release Pub. L. 86-36

IL----_
Watergate; 474,487,572,584 code; 1 Water transport Weapons Control Center (WCC); 580-581

I
Weeks, Robert; 14
WEE LOOK

I
(Navy aerial reconnaissance program); 550

Weisband, William; 169, 177.277-278,284 Wenger, Joseph; 6, 12,29, 63,93, 158,243

I
Westmoreland,

I
William C.; 530, 534, 536, 560, 561, 562, 564,565,570,583

I

I

"Wetwash" circuit; 511 .VVeyand, Frederick; 561 VVheeler, Earl; 446, 468,476, 506,520,529,555,565 VVheelon, Albert "Bud"; 409, 410

----------------_1 WHITEBIRCH (project); 502,504, 505. 513
VVhite, VVilliam Carlin; 396 White Horse Mountain; 48 White House Situation Room; 352-354

I-----:------:~:::--::-::----:~and SIGINT during the EC·121 crisis; 467-468
and SIGINT during the Pueblo crisis; 446 White Wolf(JCS advisory warning plan); 147, 314, 330, 463, 464

Wiesner, Jerome; 204

I
VVilloughby, Charles; 45, 46, 99
WILLY (Project); 42

HANDLE VIA TALENT KEY YSTEMS JOINTL Y • OT RELEASABLE TO FOREIGN NATIONA

TOP seCR!T UMBItA

650

DOCID: 523682
I E.O. 13526, section 1.4(c)

REF ID:A523682
TOP Si(RET I;JMBRA 259-160

I

Withheld from public release Pub. L. 86-36

Wilson,Charles;

Winchell, Walter; 275

Winn, Roger; 14
1 _

Wired rotor technology; 211-212 Wirtz, James; 563
1 ------

Wood, Robert J.; 391-392
1 -

Wray, William; 294, 325

Wright, Peter; 235 Wright, Wesley A. ("Ham"); 67, 68 Wullenweber
1

- see CDAA
--

Yankee Alert; 23S-237 Yansei University - see Chosen Christian College

Yardley, Herbert 0.; 17, 55, 99, 158, 273,274 YOKE (ta~tical voice intercept operation); 48, 49
Of

I

I

Yur'ya (Soviet missile site); 178

Zaslow, Milton; 578 ZICON net; 207, 255 Zuckert, Eugene; 502 Zumwalt, Elmo; 478 Zweibrucken, Germany] ~348-350J 1482,483

1

-

3RD RRU - see Davis Station

HANDLE

NTROL SYSTEMS JOINTL Y

651

°

TOP 5ECft!T IjMBR:A

p
DOClD: 523682
TOP SECRE', uMBRA

REF lD:A523682

60th Signal Service Company, Ft. Lewis, WA; 40 466L system; 296, 308-311, 339, 364

690lst Special Communications

Center - see Air Force Special Communications

Center

HANDLEVIAT T RELEASABLE TO FO

ONTROL SYSTEMS JOINTLY

TOP 5iCRn YMIlRA

652

.. :.-..

F ID :A523696 TOP SECRET
UNITED STATES CRYPTOLOGIC
HISTORY

(U) American Cryptology during the

Cold War, 1945-.1989
(U) Book III: Retrenchment andReform, 1972-1980

Derived From: Declassify on:

CCH-S54-9!H)1

TOP SECRET

DECLASSIFIED UNDER AUTHORITY OF THE lNTERAGENCY SECURITY CLASSIFICATION APPEALS PANEL. E.O. 13526, SECTION 5.3(b )(3) ISCAP NO.1.6M-02.1 ,Document 7. Date ...}UL.'"1

z". ~,~

DOCID:523696

REF ID:A523696

.J

This monograph is a product of the NationalSecurity Agency history program. Its contents and conclusions are those of the author. based on original research. and do not necessarily represent the official views of the National Security Agency. Please address divergent opinion or additional detail to the Center for Cryptologic History (8542).

This document is not to be used as a source for derivative classification decisions.

DOCID: 523696

REF ID:A523696
TOP SECRET tfMBRA

UNITED STATES CR YPTOLOGIC HISTORY

Series VI The NSA Period 1952 - Present Volume 5

(U) A merican. Cryptology during

the

Cold War, 1945-1989
.(U)

Book Ill: Retrenchment and Reform, 1972-1980
. Thomas R. Johnson

CENTER FOR CRYPI'OLOGIC ffiSTORY

NATIONAL SECURITY AGENCY
1998

"
W"'N9t.E 'II'" ThbE~rr KEVH~LlS eOMIN'%' eOU'PROb S'lS'fI!lMSJOIU'fLY

TOP SECRET UMBRA

DOCID: 523696

REF ID:A523696 rop SECRET UMBRA

Table of Contents .
Page Foreword :......................................... Preface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. Acknowledgments vi vii
IX

(U) BOOK III: RETRENCHMENT AND REFORM, 1972-1980 (U) Chapter 14: Cryptologic Retreat from Southeast Asia
The War Is Vietnamized The Fall of Saigon Hanoi's Final Campaign TheFallofDaNang The Fall of Phnom Penh The Fall of Saigon The Summing Up I . . The Mayaguez ' (U) Chapter 15: Downsizing. The Great RIF Scare ; , :....... The Clements Cuts :............. The Field Sites ; , . .-................. Turkey :.................................................. Ethiopia ; : ..'......................................... Thailand '. . . . . . . . . . . . . . . . . . . .. Closures and Consolidations :.:............... TacticalSystems Remoting :............................ Tennis : . . . . . . . . . . . . . . . .. Ora wstring ; :..
1

1
,'" .' : : ,
'.

'

.'

3 , 3 5 7 9 15 , 15

21 24

26
27 31 34 37 38 38

38
41,,-' 44 1 45 49 .,.--------, Withheld from public release Pub. L. 86-36

AROF

,

BROF

:

50
51 55 57 57 58

Remoting the Small Sites : ~. . . . . . . . . . . . . . . .. Guardrail .. : : ,............................ Reorganization ,.,........................................................ The Fitzhugh Panel . The Schlesinger Study

CSS
The Murphy-Commission .'.........................

59
65

IIANBUl

;'fA 'f'AUllff

ItEYHObE

eOMm'f'
III

eON'fROL

SYS'fEMS ,JOINTJ;. Y

TOP. SECRET UMBRA

DOClD: 523696
l!OP SECRET UMBRA

REF lD:A523696

The Hermann Study The U rsano Study The Creation ofESC

.

.'. . . . . . . .

66 6~ 72

(U) Chapter 16: Cryptology and the Watergate Era
Background to Scandal NSA and Clandestine Activities " .. Shamrock , " Minaret Clandestine Methods The Huston Plan The White House Tapes ." " ".... The Allen Era at NSA ' ' '" ., ,. The Church Committee The Pike Committee The Abzug Committee , ,................. The Backwash The Revelations .. '. . .. .. . . . .. .. .. . . .. .. . .. .. . .. . . .. . . . . . . . . . . . Glomar Explorer Koreagate Executive Order 11905 :.......................................... Congressional Oversight ,. The Enabling Legislation " The Enigma Revelations : ~,...... The Impact of Watergate :. 79 83 83
84

86 87 88
89

91 95

98
100

100 101 102

105
107

108
109 112,

w

•....

(J)

~~ ~ '" n
~

N

(U) Chapter 17: The New Targets and Techniques
....:,~S:....:tr::.:a::.:t;.:::.e5l:Iti·:.:.c..:..A::.r..::m:;:s...:L=i=m::.it:::a:.::ti:=o.!!n--.:....:,.; ..:.,., ....:, ...:... ..:.,., ...!. • ..:.... :.,.,' •...:, • ..:... •• :.,., •...:, • ..:.... :..:' •..!,.,:... :..:'•...:,.,:... :..:' •...:.. • .:.. • .:...:'•...:.. • .:... :";' •...:.. • ...:,. :";' .:...:• ...:.. • .:..,. '-'-," 117'
Withheld from public release Pub. L. 86-36

g. f->-__

=

~~==============~==============================~I 125 119
~5
1~ 1W 132

e
~ ComsatlIntelsat The A Group Secession Cryptologic Communications in the Post- Vietnam Era COMSEC and the Secure Voice Problem The Soviet Threat The Solutions NSA Computers Enter the 1970s The Era of Mainframes Platform ' NSA's Foreign Collaboration Great Britain . . . . . . . . . . .

-

134
: 134 137 140 142 144

, : , ,

;

145 151
i51 155

,

155 157

HANDLE VIA TALENT KEYII6LE C6MINT CON'mOI:. SYSTEMS .JOINTLY

TOP SEEREHIMBRA

iv

DOClD: 523696

REF lD:A523696
leP SECRETI;JMBftA

Withheld from public release Pub. L. 86-36

I

Australia · Third Party Proll!)lms

. 159 . 162 164 165 165 166

(U) Chapter 18: The Middle East and the Yom Kippur War Background to War The Preparations The Bunker Briefing The Attack The Postmortems :........... 175 178
179

181 183

(U)Chapter 19: The Rebirth ofInteUigence during the Carter Administration The Inman Era The Carter White House The War between the Admirals Apex The New Executive Order : :........ Panama SALT II ,................................. . HF Modernization :...................... The HF Studies . Inman Comes In Kunia : ;. . . . .. . . ... Conventional Signals Upgrade , Bauded Signals Upgrade The Perry Study The Wagner Study Bauded Signals Upgrade - the Project The Third World Situation The Peace Treaty with CIA 'Poetic ;.......... The HAC Investigation and the Negotiation of a Peace Treaty .
t. • • • • • • • • •

Peace Treaty Public Cryptography

I

~

189 193 196 198 199 199 202 206 206 208 210 215 215 217 220 221 223 224 227 228 230 , 231 231

I

·(U)Chapter 20: The Foreign Policy Crises of the Carter Years .The Iranian Revolution The Soviet Invasion of Afghanistan The Sino-Vietnamese Dispute The Soviet Brigade in Cuba The Final Days 245 . . . . . . . . . . . . . . . . . . . . . . . . . 250 ; 254 ~.. .. .. . .. . . .. . .. . . .. . .. 256 259

I1ANBLi!l VIA 'Hdl!N'f

KEynOl::E eeltflU'f

eeN'fROL S'IM'EMScJOUR'LY

v

T6P SECRET UMBRA

DOCID: 523696
fOP SECRET I;fMBRA

REF ID:A523696

Foreword

(U) The publication in 1995 of Books I and II of American Cryptology during the Cold War by Dr. Thomas Johnson created the NSA equivalent of a "best seller." Books I and II were distributed widely to offices and individuals and have been used as textbooks in courses at the National Cryptologic School. These two volumes filled a great need in the U.S. intelligence community for a comprehensive treatment of'cryptologic history. (U) The first book in the projected four-volume series dealt with the origins of modern American cryptology , particularly its organizational struggles in the 1940s and the great debates over centralization. The second book resumed the narrative in 1960, showing how the great strides in communications and overhead technology changed, renewed, and energized the cryptologic organizations. In both volumes, Dr. Johnson analyzed the successes and failures of cryptologic activities as well as support to national decision makers. Book II also gave an overview of cryptologic operations during the Vietnam War. and analyzes cryptologic operations from the fall of Vietnam through 1980, promises to have an impact on our knowledge and eryptologie education equal to its predecessors. This was a period of retrenchment in budgets and personnel, a period of shocking public revelation of improper intelligence activities, the beginnings of declassification about intelligence activities, and a period of technology changes that rivaled those of the previous eras. (U) This is to say, Book III deals with the period of cryptologic history that, as much or more than previous times, determined the shape and capabilities of the cryptologic organizations of our own day. For this reason, the Center for Cryptologic History recommends Book III, American Cryptology in the Cold War: Retrenchment and Reform, 1972·1980, as especially important professional reading for all members ofthe intelligence community today. Plus, it'~ a darn good story. (U) Book III, which discusses

DAVID A. HATCH

Director, Center for Cryptologic History

llAHf)LE ,{IA 'fALE!.'f . KE'WOLE

COldU('f CONTROL SYS'l'EMS ofOUffLY

,

TOP SECRET UMBRA

vi

DOCID: 523696

REF ID:A523696
TOP SECRET I:JMBR:.tc

Preface

(TS ceO) Expansion and centralization dominated American cryptologic history from the end ofWorld War II to the end of the first Nixon administration. From 1945 through at least 1970, cryptolo.gyforged ahead in a virtually unbroken expansion of people, facilities and influence in the halls of government.'

, The paradox (true in general but not in particular instances) resulted from the exploitation of everything else that was important about adversary communications, and from the enforced centralization and modernization of the cryptologic system to milk everything possible from that which was exploitable. Successes were most pronounced on the SIGINT side but were also noteworthy in COMSEC. The decade of the 1970s is remembered by most cryptologists as a scarcely mitigated disaster. Expansion came to a halt, beginning with the withdrawal from Vietnam from 1970 to 1975. The cryptologic system contracted in every way possible: people, facilities and money. Through. the administration of three presidents - Nixon, Ford and Carter - the downsizing continued. (U) Nixon's resignation in August of 1974·was followed only five months later by exposure of CIA operations by journalist Seymour Hersh. The result was a thorough airing of intelligence operations, including some by NSA, before two congressional committees, and further ignominy and public suspicion of intelligence and cryptology. Jimmy Carter came to the White House with a mandate to clean out the intelligence closet .-----, Withheld from and a predisposition to do so. He set to it with a will. I E.O. 13526, section 1.4(c) public release
~·eC())

Withheld from public release Pub. L. 86-36

I E.O. 13526, section 1.4(c) I

..rer

But the days were not as dark as they seemed.

I

Pub. L. 86-36

,

1L......J1 Even with decreased money,.cryptology was yielding the best information that it had produced since World War II. Two strong directors, LewAllen and Bobby Inman, ably steered NSA through the post-Watergate mire. In the end, Jimmy Carter became a believer in intelligence, especially what was called in the White House "technical intelligence." It was he, rather than Ronald Reagan, who first arrested the decline in the fortunes of American intelligence. (U) Reagan, who never understood intelligence as well as Jimmy Carter came to understand it,still had his heart in the right place. He directed an intelligence rebirth that resulted in a bonanza of money. The new 'dollars were shoveled into .highly sophisticated technical systems rather than into more people (although cryptology did add

IIM,BLE VIA 'i'ALEN'f KEYIl6LE C6MINT C6N'fR6L S"l"S'ff)MSdOftl'FL'I

vii

TOP SECRET UMBRA

DOCID: 523696
lOP SECRET UMBRA

REF ID:A523696

some billets). By the end of the Cold War in 1989 the cryptologic system had lots of shiny new toys, and was using them to very telling effect. The decade of the 1980s marked the high-water mark of a cryptologic system that had been in evolution since 1945. And it had .a presidential administration that believed in it.

THOMAS R.JOHNSON

JIANBbE '!b\ ftd:J!lNT KEl'.l'1I0btl COMIN'f CONTROl:; SYSTEMS 1f00PiTL'f
I-Ur

~C ••• nC I U.V.DKA

viii

DOCID: 523696

REF ID:A523696
TOP SECRET UMBRA

Acknowledgments
(U) A work of this size cannot be produced without the labors of many people whose names are not on the title page. Among them are NSA's archival and records management staff, with whom I have worked closely over the past ten years to find the needed files of material. Two research assistants, Yolande Dickerson and Rowena Clough, helped ferret out materials .. My thanks also go to the librarians and archivists at the presidential libraries who spent long hours in dingy basement vaults while I toiled through the national security files looking for reflections of cryptologic information. I especially want to thank Donna Dillon at the Reagan Library, Leesa Tobin at the Ford Library, and Martin Elzy at the Carter Library, but others on their staffs also assisted.

was handled by 1 1 and 1 I· Numerous people in NSA's photo lab helped get the pictures ready for Iinalpublicaticn. The Center for Cryptologic History's editorial staff of Barry Carleen, Barbara Vendemia, and Jeannette Gannaway probably spent more time than anyone getting the book ready for publication.
(U) The mapmaking (U) The book had many readers and consultants. Most important were Milton Zaslow, Eugene Becker, and Richard Bernard, all retired NSA officials who volunteered their time to read, correct, and advise. (U) Countless people agreed to oral history interviews to further the progress of the book. I especially want to thank four former directors: General Lew ABell, Admiral Bobby Inman, Lieutenant General Lincoln D. Faurer, and Lieutenant General William Odom, all of whom were willing to dedicate their time to set the record straight on important issues with which they were concerned. (Admiral Inman has sat for no fewer than four interviews over the years.) George Cotter and Robert Hermann provided unique information and seasoned judgments. John Devine and Marlin Wagner both gave important testimony on Bauded Signals Upgrade. The list of interviewees in the bibliography section is filled with the names of people who have provided detailed information on a host of projects and organizations. Much of these two books could not have been written without oral histories.

Withheld from public release Pub. L. 86-36

-'S COat Finally, I was given access to two special collections which provided information of unique value. the last incumbent in the old Soviet analysis office, A2, provided the executive files accumulated by him and his predecessors over a number of years. Loyd Luna: of the Executive Registry loaned executive files of the deputy directors - this was generically the most valuable set of files used in the preparation of the four-book set.

I

I

Withheld from public release Pub. L. 86-36

THOMAS R. JOHNSON

HANBfaE 'HATALSN'f' KFJ¥1JOf;:8 eo MINT eON'fROL S¥S'fEMSJOIN'fI:.¥

ix

TOP SECRET UMBRA

DOCID: 523696
TOP SECRET UMBRA

REF ID:A523696

U •••• NBbE VIA 'lYcLEto."f KEl'm6LE

C6MIN'f CON'fROL SYSTeMS ofOlK'fLY

lQP SECRET UM8RA

DOC,ID: 523696

.REF ID:A523696
TOP 5!(;RET UMBRA

(U)Chapter 14 Cryptologic Retreat from Southeast Asia
Direct American involvement in Vietnam ended with the cease-fire of February 1973. The Vietnamese were left to struggle on alone.

.ro1

I

.
(U) THE WAR IS VIETNAMIZED

I

I

.

I

Withheld from public release Pub. L. 86-36

I

.{s ceer The cease-fire that took effect in February of 1973 required that all U.S. military people be out of the country. The cryptOlogic infrastructure was already safely in Thailand, but the NSA office in Saigonhad to remain to provide support to the ambassador Moreover, NSA was committed to advising the South Vietnamese SIGINT service, renamed the DGTS (Directorate' General of Technical Security). There were NSA advisors at each of the major DGTS field sites and as DoD people, they were technically illegal according to the peace accords.

I.

Withheld from public release Pub. L. 86-36

I E.O. 13526, section
.(S CeOr-As soon as Americans were out of South Vietnam, support for the military budget was reduced. The 1974 cryptologic budget almost dropped off the edge of the table, as major field sites as well as small covert operations took huge decrements. The Air Force . EC-47 operation was discontinued in May of 1974, replaced by the much smaller remnants of the ASA U-21 program. ACRP programs declined by 50 percent, as many programs were either canceled or reduced. SARACEN, the remoted intercept operation in Laos, was . closed in April, and the huge ASA station at Ramasun was ratcheted down by about 40 percent."

1.4(c)

(8 CeO) The actual effect of the cryptologic drawdown varied by entity. It was most . severe on North Vietnamese civil traffic, which could no longer be heard by reduced RC135 operations forced to fly south of the 17th parallel. NSA also reported substantial reductions in its capability to monitor GDRS (General Directorate of Rear Services, and . thus infiltration) traffic. On the other hand, the ability to report on North Vietnamese air defense traffic suffered little or no decline." (U) In Vietnam, South Vietnamese military capability did not toughen up as fast as the Nixon administration had hoped, but the picture was not entirely dark. With only partial U.S. support (mostly from the air), the 1972 Easter Offensive had been blunted. Once American troops had left Vietnam completely, American arms and supplies bolstered ARVN capabilities. Vast quantities of military hardware arrived at &uth Vietnamese ports. So many trucks and jeeps sat on the wharves at earn Ranh and Vung Tau that one

nAl~HLE VfA TALENT KE'I HOLE eOMlNTeONTROL

SYSTEMS oiaUoft'LY

1

TOP SECRET UMBRA

DOCID: 523696
TOP SECRET UMBRA

REF ID:A523696

congressman wondered whether the objective of Vietnamization was to "put every South Vietnamese soldier behind the wheel." 4 The ARVN became, by the end of 1974, one of the largest and best equipped armies in the world, and its air force was the world's fourth largest. -. c-)(-d-)-' E-.O1-3-S-2-6,-s-e-ct-io-n-1.-4-(

Ir-

INSA 1.4 (c)(d)

I c=J

.(Set The SIGINT situation was very complex. Although confronted with major deficiencies in manpower and equipment, General Nhon's DGTS had developed at least the rudiments of what NSA had hoped for when the Vietnamization program began. It did a goodjob of collectin~ Its performance in traffic analysis was spotty, mainly because the DGTS often did not see the value. It had an outstanding ARDF capability on paper, although that program was hindered to some degree by the reluctance of Vietnamese pilots to fly in areas of hostile fire. The EC-47 fleet that NSA bequeathed to Vietnam was aging and prone to mechanicalfailure, which drove aircraft downtime to unacceptable levels. The DGTS used ARDF results primarily for order-of-battle rather than for tactical targeting. ~

I

Withheld from public release Pub. L. 86-36

(u)

General Nhon atNSA with John Harney, then commandant ortbe National Cryptologic School

r:

II-I,NDYiJ '/I:A 'fJtt.aSN'I' KBYlIebB

GeMINT OeNTReL

S-YS'PSMSdOIfR'LY

lOP SECRET UMBRA

2

DOCID: 523696

REF ID:A523696
TOP SECRET U..,BRA

JS..Ge6rGeneral Nhon had picked his SIGINTers carefully, and DGTS dedication was very high. It was hindered by a corrupt and inefficient government and by declining American financial support. Moreover, NSA had been very slow to recognize the need to give OGTS first-class SIGINTtraining. The philosophy in the early years had been to "buy off" the government in order to develop political support in Saigon for the build-up of American cryptologic capabilities. NSA never permitted a level of SIGINTexchange with the ARVN SIGINTorganization that the wartime situation demanded, and its lack of technical expertise was consequently low. When the Americans left, DGTS had a long way to gO.6 (U) The improvements in overall ARVN capabilities had .reeutted in at least a marginal improvement in the situation in the countryside. Village security was better in many areas, and the government, still corrupt and oppressive, had nonetheless announced a new land reform program. At year's end, a shaky stalemate existed between the ARVN (Army of the Republic of Vietnam) and the NVA (North Vietnamese Army). Little had changed in the government's ability to control geographical areas since the cease-fire." .(SG)-But trouble was afoot. NSA reporting since the cease-fire documented huge NV A shipments to the South. Unhindered by American bombing, they brought in engineers and road-builders, and turned the Ho Chi Minh Trail into the "Ho Chi Minh Road," an allweather highway suitable for heavy transport. By early 1975, NVA forces were better equipped than at any time in the past.' They were obviously waiting for the opportunity to renew conventional warfare.

(U) THE FALL OF SAIGON
(U) Hanoi's Final campaign

(U) The final round of the Vietnam War was apparently August of 1974. With American support for the government to weaken, victory appeared to be just a matter of time. But it was 1976. No one in Hanoi really envisioned the imminent

planned by Hanoi as early as in South Vietnam beginning the timetable was not 1975 collapse of the opposition."

-'SeTThrough the fall, NSA was reporting infiltration figures unheard of except prior to the 1972 Easter Offensive. The NV A launched the fIrSt attack shortly after the first of the year against Phuoc Long Province in MR 3. After the seizure of the province, Hanoi sat back to judge the American reaction. There was none, so the NVA renewed the offensive in MR 1 and 2 in March .
I

Withheld from public release Pub. L. 86-36

I

.!SG)-About the first of March, SlGINTindicators pointed to a strong.NVA attack on Ban Me Thout in the Central Highlands. The NSA office in Saigon, however, believed that the real objective was Pleiku, and that Ban Me Thout was a diversion, albeit a significant one. the NSA representative, accompanied by General Nhon, the DGTS commander, briefed the ARVN MR 2 commander, who refused to believe them. The

~

w ~gU

'IlA 'Fldal5N1' KaYIiO ••! OOMlN'f OON'l'ROL 8¥S'FFlM8 cJOI~Fl'bV

3

TOP SECRET YMBM

DOClD: 523696
Jep SCERn I:JM81b\--

REF lD:A523696

. The Final Days
(U) Vietnam .

-fe, SECRET t;lMBRA

. 4

DOClD: 523696

REF lD:A523696
TOP SECRET UMBRA

commander reinforced Ban Me Thout, but it wasn't enough, and he still Meanwhile, just as SIGINT had indicated, NV A forces fell on Pleiku.10

lost it.

(U) On March 15, President Thieu made the "tactical" decision to abandon the Central Highlands. ARVN troops at Pleiku abruptly abandoned the city, and it was in NVA hands within two days. (U) This began one of the most awesome and tragic civilian evacuations in modern· times. Spurred by the military abandonment and the advancing NVA forces, hundreds of . thousands of refugees jammed the single road from Pleiku to the sea, Route 14. About a third of the way to their objective ofTuy Hoa, Route 14 met with Route 7B at a town called Cheo Reo. There, streams of refugees from other towns intermixed, creating gridlock. In the vicinity of the town, NVA forces attacked retreating ARVN forces, creating a. bloodbath in which thousands of refugees and soldiers were killed. NV A harassment . continued the length of the road, but Cheo Reo wasthe worst.'! . (8 GGO) The DGTS center in Pleiku kept operating until the final day, arid then the center's people joined the fleeing refugees. Of the 87 men and 120 dependents who took to . Route 14, no more than half ever reached the coast. The rest remained unaccounted for. 12 ~NSA was picking up indications that the North Vietnamese were moving reserve divisions south. The 968th, which had remained in Laos for its entire existence, showed up in the Kontum-Pleiku area, and there were indicators that divisions in the Hanoi area, which had never done more than train men for combat in other organizations. might be moving out. Still, CIA predicted that the South would hold through the dry season. IS (U) But military analysts in the Pacific were not so optimistic. USSAG (United States Support Activities Group), which was really MACV in Thailand under a different name, . pointed ominously to the movement of reserve divisions, and predicted an all-outeffort to take Saigon during the dry season. IPAC (Intelligence Center Pacific) hinted on March 17 that the entire country could fail.l~ . (U) There was no let-up. Quang Tri City, defended with such high casualties in 1968, fell to the NVA on March 20. At the same time, NVA units were besieging Hue. On March 22 they severed the coastal road between Hue and Da Nang. The old imperial capital was a captive. 15.

(U) The Fall of Da Nang

(U) With Hue cut off and withering, refugees poured into Da Nang,the last important city in MR 1 still held by the government. By March 25 the city was choked with pedestrian and cart, traffic. ARVN units had turned into an armed mob and were commandeering any form oftransportatiotl available to get out of the city. Mobs swarmed

RAN OLE

v IA

I ALEN

r

KE f ROL! eOMINT eON'PftOL SYS'fSMS cfOl~fTb¥

5

TOP SECRET I;JMBRA

DOCID:523696
'FOP SE(:RET I:JMBRA

REF ID:A523696

.across the airport runway, and each successive World Airways 727 landing there found it. more difficult to take Off.18 Nang, received , a call from the CIA station chief. It was time to get out. Cameron drove his jeep to the air strip, leaving his personal goods behind, and squeezed aboard a jammed 727. He rode the overloaded plane to Saigon with a Vietnamese child on his lap." (U) The next day the Shell Oil personnel departed, closing the airfield refueling operation. Mobson the runway made it.impossible to land, and that morning an American embassy cargo flight was completely stripped by the mob after it landed. At that point World Airways ceased service to Da Nang." (U) The next day the last Americans got out of Da Nang via ships in the harbor. On March 29 the owner of World Airways took three 7Z7s from Saigon to Da Nang without authorization from either the Americans or Vietnamese. According to the CIA description:
At Da Nang one 727 landed and was immediately mobbed, surrounded by trucks and was forceably boarded by GVN military on the airstrip. 1'he plane made emergency takeoff preeedures and was rammed by a truck at the lef\ wing or bit a truck on takeoff. The plane was unable to take offfrom the normal runway as the VN military had it completely blocked with trucks or other vehicles. Accordingly. the plane took oft on a taxiway. The pilot stated that once airborne he was unable to retract the wheels and assumed he had major hydraulic casualty. However; one of the otl:ler planes that took off (from Saigon) af\er him came alongside and reported that he had a body in the lef\ wheel well that was jamming the wheel doors.1i

..csrOn the 26th, Al Cameron, the NSA advisor ~ the DGTS unit atDa

The World Airways flight (the only one of the three that was actually able to land) arrived in Saigon with 385 passengers (about the right complement for a 747), of-whom four were women, three were children, and the rest were ARVN soldiers.

~eOl The Da Nang DOTS station, at 429 people, was one of the largest in the country. The DGTSmanaged to evacuate two planeloads of equipment and dependents before the city fell. The operators continued operating until the site was overrun. The day . before the end, the De.Nang communications operator told Saigon:
Only workers are !ef\ at the signal center and we wiil not be able to get out. We are just waiting to die. We will wait for the VC to come in. hold our handa over our heada for them to cut. We will be here until the last, but the go~ernment doesn't think about the workerS. Please say something too ease our fUlaIboura.20

Photos of Da Nang on March 30 (the day the NVAentered the city) showed only a smoking shell of a building where the De. Nang center had been. All the operators were reportedly either killed or captured. 21

BANgYI 'III.TAYINT KBYUebS

GeMINi'

eeNi'OOL

SYS'ftlMS oi9lNTL¥

TOP

,!at!, tlMBItA

6

OOClO: 523696

REF lD:A523696
, TOP SECRET tJMBftA

(U) Fleeing Da Nang

CU) The Fall of Phnom Penh

(U) NVA forces raced pell mell down the coast, gobbling up city after city. The advance was dizzying to hunters and hunted alike. Within a week of the fall of Da Nang, all of MR 2 was in NVA hands except for Nha Trang, which was abandoned to the enemy on April 7, but not actually entered until the 9th.22 (U) Then a brief quiet descended on the land. NVA forces had outrun their supplies and their military plans. Hanoi began collecting assault forces for the final push to Saigon,. and the Saigon government began steeling itself for what had clearly become inevitable. (U) At that point, American attention refocused on Cambodia. As the NVA advanced down the Vietnamese coast. the Khmer Rouge organization in Cambodia had quietly but effectively squeezed the Lon Nol government into a trap. All that the government held by. January of 1975 was a narrow water alley through the center of the country. The

HANDLE VIA TALENt KEYHOLE 60MIN'P eON'l'ROb SY~TBMS dOIN'fL¥

7

TOP SECRET lIMBRA

DOClD: 523696
TOP SECRET UMBRA

REF lD:A523696

(U) Cambodia

- tile Khmer Rouge tighten their grip on Phnom Penh'

communist forces held all the countryside, and began pinching off the Mekong waterway through which the capital obtained almost all its supplies. Each year the KC (Khmer Rouge) had done the same thing, but like a bulldog tightening its grip, each year they choked the river closer to the city. '

JSe1 The American mission there was very small, only 140 people. It was well organized under an experienced ambassador, John Gunther Dean. Moreover, it had outstanding intelligence support, almost all of it SIGlNT. Moreover, the small
'--------1

I E.O.

13526, section 1.4(c)

H.~N9bE

'JIATAhS~TT KEYHObS

60MINT CONtROl:. S-Y&FISMSoJO!N'FL-Y

Withheld from public release Pub. L. 86-36

TOP SECRET UMBRA

'8

}

DOClO: 523696

REF lO:A523696
TOP SECRET UMBRA

I E.O.13526,

sectionl.4(c)
With!Ield from public release Pub. L. 86-36

ASA ARDF effort out of Thailand showed the tightening of the vise as the various KC headquarters moved closer to the city. But without American commanders to act on the information, there was little the U.S. could do.

'-

---1. But, as it was New Year's Eve, they were all at parties, and. the army made no preparations whatever. Gas tanks weren't filled, guns weren't even loaded. 23

--'set On April ll,the AFSS unit at NKP (Nakhon Phanom air base in Thailand) intercepted KC .plans for an all-out assault on the city. Admiral Gayler, by then CINCPAC, called Ambassador Dean to say it was time to leave. Dean agreed with him, and Gayler implemented Eagle' Pull, .the dramatic rescue of embassy personnel by helicopter from a sport field in downtown Phnom Penh. By the end of the day on April 12 the entire operation was over, and Phnom Penh waited for the KC to march in. Most of the .cabinet refused evacuation and waited for the doom that would befall them. They were all .executed. 24
(U) The Fall of Saigon

~ As the NVA repositioned and refurbished for the final assault, an air of unreality settled on the American embassy. Ambassador Graham Martin believed that the government could somehow hold out until the rains began in June. SIGINT, both from the DGTS station in Saigon and from the U.S. SIGINT system, showed the NVA massing around the city. Thi~u, who knew the end was near, resigned. In Washirigton, the White House understood what was happening. But Martin refused to heed the signs. He and his CIA chief of station, Thomas Polgar, believed that the SIGINT was NVA deception. A bill was pending in Congress to send an additional $700 million in military aid to the government in Saigon, and they held out the hope that this would pass and that it would come in time. The regime in Himoi, Martin thought, was really getting in position to impose a coalition government, not a military victory. 2:5 ~NSA station chiefl Imain concern was his people. When the country began falling apart, he had forty-three employees and twenty-two dependents. The dependents he began evacuating on civilian commercial flights, along with the thousands of Vietnamese fleeing the country. Ambassador Martin put the evacuation of the government employees on hold. He feared that the SIGINT system would not support him if they left, and that the DGTSwould not work without NSA assistance." .

Withheld from public release Pub. L. 86-36

rei The signs of collapse became more ominous, andc:=Jmade almost daily trips to the ambassador's office, pleading for permission to get people out of the country .. The' exchanges became angry; andc=Jwent to the director of NSA, Lieutenant 'General Allen, for help. In mid-April, Allen sent a distressed cable to the DCI:

lb\iofBL~VIA"AL~N'''Kfil •••. ilau

GaMIN,.eaN'fRab S'IS'fflMSJ9lN1'f?f 9

TOP SECRETUMBR-A

DOClD: 523696
TOP SECRET UMBIbtc

REF lD:A523696
Withheld from public release Pub. L. 86-36

1am fully aware of the complex political isaues involved in any withdrawal of U.S. Government personnel from the RVN. 1 wish to reiterate, however, that the safety of the cryptologic personnel in the RVN is my paramoWlt concern. 27. .

Not even this was sufficient to IChanj minds in the embassy. 'smuggled" people out of the country by bu y ing- them commercial tickets, and his staff gradually shrank to just a few. Those who remained spent almost all their time at work, often sleeping in the office rather than .returning to the hotel where they were billeted.28 . ..{e1. The final assault began on April 26 with the attack and capture of Bien Hoa. On the 28th, Dmade a final <visitto Martin, with a message from Allen directing him to secure his communications and depart. Still,. Martin refused. The next morning, the NVA began rocketing Tan Son Nhut; and the airfield was closed to even military aircraft. The embassy and its people were now caught in a trap, and the only escape possible was by helicopter.29

Withheld from public release Pub. L. 86-36

..,;..;.;.;..;...;;;.;.;... •.•• ' ch.lasc NSA.representative

In Saigon

..(S.CeetThe evacuation plan was called Talon Vise (later changed to Frequent Wind).
It envisioned the evacuation of all Americans and almost 200,000 of their Vietnamese

allies. Evacuees would be airlifted by fixed-wing transport from Tan Son Nhut or picked up at the port ofVung Tau on the coast. Helicopters would be employed to ferry pockets of people from exposed locations to Tan Son Nhut. Politically sensitive Vietnamese, such as those who had participated in the Phoenix program, or SIGINT transcribers (the Dancers), and their families would be afforded special evacuation priority.30 (U) But with the ambassador bewitched by clouds ofintelligence opiates, there was no time left; to implement such an orderly departure. All that was left was to use the helicopter option to try to g~t the Americans out. Martin, debilitated further by walking pneumonia, stood alone. With shells landing on Tan SOnNhut, the president gave the order, and Admiral Noel Gayler directed the evacuation. Martin was obdurate to the end.

IIAliBLE

Tt'b\ 'fA;LEN'f KE'l"116t:E €6MfU'f

e8~ffR6L

B\'S'fI3MB d81't''fLY

lOP SECRET YM BRA

10

DOClD: 523696

REF lD:A523696
"fOP SECRE"f"'MBRA

(U) Graham Martin

(U) Gayler had been assembling a vast armada in the South China Sea. It contained seventy-seven vessels, including five aircraft carriers. On the morning' of th~ 29th, the principal carrier to be involved in the operation, the Hancock, downloaded fighters and uploaded choppers. Sl ..(SCCO~-AtNSA, Director Lew Allen had been putting together a SIGINT support effort since mid-April. Most important was the monitoring of North Vietnamese communications to provide warning to the evacuation aircraft, since the NVA had brought SAMs into the vicinity of Saigon. A special AFSS SIGINT support team was flown to Clark Air Base to brief MAC (Military Airlift Command) crews on warning measures, should . they be targeted by NVA antiaircraft units. As it turned out, MAC aircraft were not used in Talon Vise, although they did continue to fly into Tan Son Nhut until the morning of the 29th.S2 . !S CCE»The Olympic Torch U-2 collection (downlinked to NKP) served as the primary monitoring system for NVAcommunications, and also monitored U.S. communications to keep tabs on the progress ofthe evacuation, This information was passed to Gayler and on to the White House. In addition, RC-135 missions were tasked with both NVA and U.S. communications. se

fLUHlbB VIA 'FA-bBN'f J(SYII9bB G9M1NT GONTR9L S¥SHMB dOIHTLo¥

11

"fOP SECRETI:JMB~

DOClD: 523696
lQP SECRETUMBRA

REF lD:A523696

(U) When, on April 29, President Ford directed the implementation of the evacuation plan, military planes had already evacuated almost 40,000 Americans and South . Vietnamese over the preceding eight days. But since the plan called for over 200,000 to be evacuated, this was just a start."
.'

(U) The helicopters began flying from the deck of the Hancock on the afternoon of April 29. All through the night, the heavy thump of chopper blades was heard above the embassy. The operators at NKP monitored the voice frequencies used by the chopper. pilots, and sent their reports to Gayler in Hawaii.

(U) Americans and Vietnamese rusb for a waiting belicopter at tbe DAO comPound, 29 April

Withheld from public release
Pub. L. 86-36

..reTThe remaining NSA contingent found itself marooned at their offices in the DAO. compound at Tan Son Nhut. found that no provision had been made to get him and his people out. He contacted General Smith, the military attache, ,whoarranged for cars to takeDand his people to the embassy. There they boarded helicopters late on the 29th for the ride to the waiting ships."

D

HAHBbE

'}Ill: 'f"'bEN'F

I{SYllObE COMIN'FCON'FROb S¥S'F5MS 49fN'FbY

MP SECRET !:IMBM

12

·'6' 5EeRET~MBRA

\0 01 \0 M N II')

(U) Midnl,b& 2. April -

tbermia.ebar,.,

b~

down

Use root

oltb.

DAO compound

H U

o

II dlSb&

1 'f

btill'l'liS

119b1ii S9'HHT88!I'PRBb6

&i'SMfjoleUPI'b

8

13

TOP

5[(AEHmSRA

DOClD: 523696
TOP SECRET tJM8RA

REF lD:A52369~

1£'1 At about midnight, Pineapple 6-1, a chopper pilot in the embassy compound, reported that he was in contact with the ambassador, who still refused to leave until the last Americans were out. Four hours later, intercept operators heard chopper pilot Lady Ace 9 tell Martin that the president had directed Martin to leave forthwith. The chopper hovered above the embassy rooftop as smoke from fires in the building made his landing temporarily impossible. Six minutes later an RC-135 operator heard the pilot broadcast; "Lady Ace 9 this is Tiger Tiger Tiger." THis was the codeword indicating that the ambassador was on board. <

(U) Vietnamese wait outside the gates or the American embassy &sa helicopter approaches the compound.

I

JMNBbE VIii TAYlN1' KE¥UabE GaMiNT GamRa!> S¥STEM8cJaINTla¥

f.E)p SECRET I;JMBftA

14

DOCID:.523696

REF ID:A523696
TOP SECRET UMBRA

(U) The choppers continued to pluck people off the roof of the burning embassy for another three hours. The last to leave was not the ambassador - it was the ground security force.35 .£SCOat It had been the largest helicopter evacuation in history. Seventy Marine helicopters had airlifted more than 7,000 Americans and Vietnamese from the embassy and the DAO compound. Among those who did not get out, however, were the DGTS operators. Saigon Center operated to the end. and CIA evacuated only about a dozen highranking officers, including General Nhon. The Dancers. DGTS linguists on duty in Thailand, were evacuated from Thailand to the United States. Their families in Saigon had already left South Vietnam and were waiting for them on Guam.37

(U) THE SUMMING UP

~Not having time for an orderly departure, the Americans left behind vast stockpiles of military equipment. Along with the runways full of planes and parking lots full of trucks, there were large amounts of crypto gear. . Deputy Director Berison BufTham estimated that it was the largest loss of COMSEC equipment ever. In practical terms, however, it was not as great a blow as the capture of the Pueblo. The crypto principles of most of the equipment had been compromised earlier, and very little actual. key was known to be in Vietnamese hands. Spare parts would be almost unobtainable. and . BufThamexpected that the U.S. would intercept very few NVA transmissione." ~ eCet The DGTS organization was captured virtually intact. At the time it consisted of more than 100 manual Morse positions,2,700 people, and seventeen ARDF j aircraft. Many of the South Vietnamese SIGINTers undoubtedly perished; others wound up in reeducation camps. In later years a few began trickling into the United States under the orderly departure program. Their story is yet untold. •

JS.Ge6j Their leader, General Nhon, made his way to Washington, D.C., and was hired as a linguist by NSA. Helived a quiet life in suburban Washington until his retirement in 1994. He now lives with his family in rural Virginia.
(U) TH~ MAYAGUEZ

(U) As if Southeast Asia had not caused America enough heartache, one last chapter remained to.be written. The seizure of the Mayaguez had a murky beginning and to the end remained unsatisfying. It also had a cryptologic component which remains confused to this day .

...cser The Khmer Rouge regime which rolled into Phnom Penh in mid-March 1975 quickly turned vicious. By early May, the White House was receiving SIGINT reports of widespread executions, offorced exodus to grim countryside reeducation camps, offamilies separated and of retribution on an unbelievable scale. Secretary of State Henry Kissinger,
H/.NebE '1M, t'Al.ENT KEYH9b! C9M1Nl' CONTROL S¥S'fEMS,JOINThY .

15

TOP SECRET I:JMBRA

DOCID: 523696
TOP SECRET UM8AA

REF ID:A523696

commenting one such I IKC message, wrote to President Ford, "The magnitude of the KC liquidation effort has heretofore been unclear. It would appear that ' -----if similar efforts are being carried out in other parts of the country, this would involve a slaughter of immense proportions."

on

INSA 1.4 (c) I

3'

.(Be) The Cambodian government of Pol Pot took a very aggressive approach to foreign relations, too, Among the territories which KC forces invaded were several small offshore islands which Vietnam and Cambodia both claimed. Among those islands was one named { . Poulo Wai. SIGINT intercepts of KC communications revealed a determination to hold Poulo Wai and to spread out farther i~to the offshore waters.

(U) u.s. destroyer off Kob.Tang Island

JSe)Beginning on May 5, NSA began publishing reports of the KC seizure of Thai flShing vessels and attacks on Panamanian and Korean merchantmen' plying the waters in the Gulf of Thailand. But the intelligence community focused not on th,ese commercial depredations, but on communist attempts to intercept Vietnamese refugees escaping after the fall of Saigon. Moreover, the U,S. government organization charged with issuing notes to commercial shipping had no links to the intelligence community. No notes were issued.co . (U) Into this nest of small-time raiders steamed an American flag container ship, the Mayaguez, plying a regular route between Hong Kong, Thailand, and Singapore. The first maydays from the vessel, on May 12, indicated that they were being boarded by Cambodians, and later that they were being towed to an unknown Cambodian port. An

TOP SECRET UMBAA

16

DOCID: 523696

REF ID:A523696
Tep SECRET lJMBfbIc

, exploration company based in Jakarta received the broadcasts and notified the American embassy. 'the embassy issued the initial critic at 0503 EDT on May 12. . (U) The president was briefed on the seizure that morning. .It was not a military challenge and was scarcely an impediment to commerce. But the Mayaguez seizure clearly represented a political challenge. The evacuation of Saigon had been a profound American defeat in Southeast Asia. Here was a chance to prevent the tiny Cambodian navy from tweaking America's nose. Coming only two weeks after the fall of Saigon, it was an event which found American military forces still in place in Southeast Asia. The president directed that a .response force be assembled and the crew recovered.' The discussions with the president harked back to the disastrouspzieblo seizure. Ford was determined to prevent that scenario at any cost. U (U) Initial Navy aerial reconnaissance ordered by the Pentagon established that the Mayaguez itself was anchored a mile off Koh Tang Island, thirty miles off the coast of Cambodia. The central concern of the Ford admi;"istration became the location of the crew. Ifit remained on Koh Tang (where it was, presumably), one sort of rescue operation . would be mounted. If the crew was transferred to the mainland, a very different operation would be called for.·2 JS=,Ce6) Here was where good intelligence was required. NSA still had in place virtually all its intelligence assets from the war in Vietnam, and the Agency directed a total focus on Cambodian communications, which were all readable. NSA declared a SIGlNT alert. Meanwhile, aerial reconnaissance continued to blanket the area. In the early morning of May 14 (Cambodian time), an American patrol craft spotted a thirty-foot boat, accompanied by escort vessels, making a .run for the mainland, with eight or nine Caucasians on the deck. Since the least desirable option was for a mainland rescue, a tactical air strike was called in, and the escortvessels were sunk. But the main vessel continued on, and the attacking A-7s held their fire. ~ ~An early intercepted message indicated that the crew was to be taken to Koh Tang. This caused the administration to focus on the island. But that was it. There were, no subsequent messages about the location of the crew,their destination or the intentions of the Cambodian government, until the very end." (S-€eO) The fragmentary SIGINT, and the lack of anything more definiti ve, caused the administration to focus on Koh Tang. A complex rescue operation was hastily arranged, and on the morning of May 14, only three days after the initial seizure, 200 Marines assaulted the island. They were met by heavy resistance. The 150 Cambodians on the island were armed with 75-mm recoilless rifles, claymore mines, and rockets, in addition to small arms. Marine helicopters were cut down on the beach, and eighteen Americans ~ere killed. The Marines were pinned down on the island, and they themselves had to be rescued the next morning."
(Se)" Meanwhile, Navy F-4s struck Ream Airfield inside Cambodia, based on SIGlNT intercepted by the USAFSSunit at Ramasun Station that the KC planned to move

\

W"'Nl)bE Vb\ TAbEN'F KBYU9f:J5 69MI:PR' 60NTReLS¥S'tEMS"OIN'fbY

17

TOP SECRET lJMBRA

DOClD: 523696
TaP SEERef tJMBRA

REF lD:A523696

Cambodian combat aircraft there. They destroyed seventeen aircraft on the ground and put the airfield out of commission. 48 . .(SeT On May 14,as the Marine assault was going on, there was a flurry of messages from various KC entities referencing response to the American attacks. Early on the 15th (in Cambodia) a message (probably from Phnom Penh) ordered a KC operational authority to let the Americans "take the ship and leave" and to "let the Americans go." Soon thereafter a KC gunboat appeared near the north end of Koh Tang showing a white flag. Four minutes later the destroyer USS Wilson scooped up the entire crew, and l'affaire Mayaguez was over, except for the extraction of the Marines on the' beach, 'which was difficult and dangerous to the end." .(U) The Ford administration claimed credit for a win. The crew was back safe and sound, although at the cost of eighteen Marines dead. President Ford went on television to explain the American response, and a Gallup poll taken shortly after showed the approval rating for the operation at 51 percent; To an administration which had been badly battered by its handling of the pardon of President Nixon, this was goodnews. ~ A month later the Vietnamese completed what the Americans had started. Intercepts revealed that the Vietnamese had wiped out the Cambodian garrison onPoulo Wai.48 . .

..is-ceG}- Although the crew was recovered and the vessel released, the Mayaguez incident has been counted as an intelligence failure. DIA and IPAC intelligence estimates of KC strength on Koh Tang were accurate but did not reach the deployed forces. Although this deficiency was cited in report after report, no one seemed to know why the information did not reach the users." But since the only reliable information on Cambodia at the time was SIGINT, classification diffifulties are readily suspect.
(S-eeO) There were other problems relating to the affair. The response of intelligence agencies in Washington was slow, and the NOIWONsystem was not used. While SIGINT classification undoubtedly hampered the dissemination of critical intelligence, in the opposite direction tactical commanders refused to share details of the' military operation with NSA - details which would have improved intelligence responsiveness. ~ ~ Why didn't SIGINT reveal the location of the crew? Reviewing the action some weeks later, an NSA analyst came up with the answer. Simply put, the operation was carried out by a local commander, without checking with higher authority. Khmer Rouge local commanders had long exercised such authority, and it is reasonable to suppose that it did not halt simply because peace had broken out in Southeast Asia. The first high-level SIGINT came from Phnom Penh on the 15th and was passed to Ta Mok, the regional . commander, directing that the crew be released. There was no prior direction from higher headquarters because headquarters had not .directed the action in the first place, and it got . involved only when the military consequences had become serious. In a radio broadcast the following September, Ieng Sary, the Cambodian deputy premier, admitted as much." So in the end SIGINT, the only goodsource on Cambodia, came up short.

HANDLr;

~ fA !"ALENT K'E Y'HOLEeOMIN'I' eomito£;

S"I.'STBMS .t6m'lL

Y

'fOp SEeR!' UMIItA

.18

DOClO: 523696
Withheld from public release Pub. L. 86-36
1. (U) Illterview withl and Tom Jobnaoll, OH':'"6--=-S2=-,-=-N:::SA:-:-. -----------2. (U) CCH Series VI.HH.26.1 3. (U)lbid. 4. (U) George Herring, Temple University

REF lO:A523696
Tep SECRET I::JMBRA

Notes
~ 2 December 1992, by Charies Baker

AIMI'icG', Longest WGr: The UnieedSfcfu Gild Vief/IGm,1950.1915 (Philadelphia!

Press, 1986), 231.
Truths: '

5. (U) Pbam Van NOOn, Remini8cellceS about oGTS, CCH Serie. XI.s. Robert,J. Hanyok, "The Relevant SIGlNT wues " o(the Indochina War,19S0.7S;" Part 2, CryptologicQuarWrl)' (FaUI997) 16:3, 1·50.

I

6. (U) Interviewl
Relevant Truths.·

I. -NSA

123 January

1986. by Robert D. Farley and Tom Johnson, OH 4-86, NSA; Hanyok, "The

in Vietll&m: Proud and Bitter Memories," Cryptolo, (October 1976),3·6.

7. (U) Herrini,

AmericG', Longe.t War.
Defen •• University collection on Vietnam, bos 301.

8. (0) CCH, National 9. (U) CCH Seriee 10. (Uc::::Jinterview.

vm. 30.
ViefllGm: A Hietory (New York: Penguin Boob, 1983).
Hanyok, "Relevant Truths, ••41.

11. (U) Stanley Karnow.

12. (U)CCH Series VI.HH.9.1; 13. (U) CCH Series Vl.HH.26.2. 14. (U) Ibid. 15. (0) CCH Serin 16. (U) Ibid. 17. (U)C]. VIIl.30.

et aI .• illterview.

18. (U) CCH NDU collection, 80s 320. 19. <U) CCH NDU Collection. 80s 323. 20. (U) CCH Series VI.HH.I1.l9. 21. (U) Ibid.

c=J,

eta!., interview;

Hanyok. "Relevant

Trutha," 41.

22. (U) CCH Series VIIl.30. 23. (u) C]'et CBOG36. 24. <U) CCH Series VI.HH.27 .10; 26.4. 25. (U)D interview; Karnow. CCH Series VI.HH.9.1. al., interview; CCH Series VI.HH.26.10; NSA Archives, Accesaiol1 Number 23500. Lee,

26. <U>C)nterview;

2'1. (U) CCH Seriea VI.HH.9.l.

Withheld from public release Pub,. L. 86-:~6

19

TOP SECRET I:JMBRA

DOCID:523696
"fOP SECRET UMBRA

REF ID:A523696

Withheld from public release Pub. L. 86-36

29. (U) CCH Series VI.HH.9.1; Kamow. 30. (U) CCH Series VIIl.30. 31. (U)CCH Series VJ.HH.26.l1. 32. (U) CCH Series VIU.30. 33. (V)CCH series VIIl.30; VI.HH.26.11; 34. (V) CCH Series Vm.30. 35. (U)c=:Jinterview. 36. (mCCH Series VIII.30. 26.9.

37.(U~interview;

B-WAR,6·12 May 1975. Karnow.

38. (U) CCH Series Vm.30. 39. (U) National Security Files, ~rald
AeialVietnamlCambodia," 40. (Ul CCH Series Vm.25. 41. (U) CCH Series Vm.25;

R. Ford Presidential

Library, Ann Arbor, Michigan; ftSoutheast

CCH Series XVI.

John F. Guilmartin,Jr.,A Very ShorrWar: The Mayaguuand tlwBattuofKoil Tarl/l (College Station, Texas: Texas A and M Press, 1995),36-37.
Support to the Recovery ofthe 55 M'ayaguez."

42. (m CCH Series 25, ·USAFss

43. (U) CCH Series Vm.25; Guilmartin. 44. (U) Ibid. 45. (U)Guilmartin,A

AVery Short War.

Very Short War.

46. (u) CCH Series VIll.2S. 47.

«n Ibid.
Se~ure of the Mayague" -

48. (U) Ibid. 49. (U)Guiimartin,

A Very Short War; CeH Series VUl .25, GAD report en~tJed ~e A Case Study in Crisis Management,"
60. (U) CCH Series vm.ss, CIA postmortem.

51. (U) CCH Series VlII.25; Guilmartin,A

Very Short War.

HAI'IBI:.E Yh\ TALENT KEYU914J 09MIN'l' 09liTROlz S'l-JS'l'SMS cJOIN'iVl

TOP SECRET UMBRA

20

DOCID: 523696

REF ID:A523696
Tgp SECRET YMBRA

(U) Chapter 15
Downsizing
had waxed fat during the war years. It did not seem so to those who struggled for dollars and manpower to help fight the. war in Vietnam, nor to those in other parts of the cryptologic system who desperately tried to maintain their hold on resources that seemed inexorably to slip into the pit of Vietnam. But in fact, the peak of the cryptologic system was reached in the late war years. After that, there came the reckoning.
,-... ...(8) Cryptology

~ ~

~

.s:
~ ~ ~

~ vi

-

o ~

(8) The peak years in overall field deployment came from 1967 to 1970. After that, it looked like the cryptologic system was going off a ski jump (see Table 1). The downslide lasted for a decade - field site deployment did not finally level out until 1981 - and the loss of field sites was matched by an overall decline in manpower. The cryptologic system. began the 19705 at approximately 89,000 people; it ended at about 50,000, a drop of 44 percent. The funding profile, unlike that of personnel and field sites, remained fairly steady over the period and was actually higher in 197~ . than it had been in 19691 But the decade was one of runaway inflation, so a steady stream of dollars did not equate to the same level of resources as before.'

I.

I

(U) THE GREAT RIF SCARE

Withheld from public release Pub.L. 86-36

.(et At NSA, the work force shrank from 19,290 in fiscal year 1970 to 16,542 in fiscal year 1979,a reduction of 14 percent." Looking back, this doesn't seem so drastic, but in 1971 no one knew how far the cutbacks would go, just that Congress had decreed a huge cutback in the federal work force, called the General Austertty in Government Expenditures Act; and that the Department of Defense would absorb the brunt. To maintain some sort offairness, cuts would be across the board, and NSA would give up its "fair share" of manpower, regardless of'missiorror need.
~ Soon after Congress levied the cuts, in September of 1971 Admiral Gayler, the . DIRNSA, issued a memorandum to the work force confronting the rumors swirling through the halls. Yes, a RIF (reduction in force) might be necessary, and it was certain that promotions would get scarce. But a RIF would be an absolute last-gasp measure. He hoped that retirements and attrition would turn the trick. This was suspect, however, because NSA's attrition was notoriously low - about one-third of the federal average. With a closed-loop personnel system and unique, nontransferrable skills, NSA employees could not go out and look for other federal jobs. (By the same token, employees of other agencies could not come looking for jobs at NSA.) What finally forestalled the RIF, however, was a device, called "discontinued service retirements." NSA began offering these immediately, and they were hugely successful. In 1972 the retirement rat~ doubled that of the previous

flANBLf ViA 'fALf.N'f KEYH6tE C6MIN'f C6N'f~OI:iS'ISftlMScJOINTb¥

21

Tgp SECRET YMBRA

DOCID:

523696

REF ID:A523696

Total !lumber '01 Situ
'JO --'_

IOO~r~------------------------------------------------------'
10"

u

10 -'_
70---'

,."
....
... 11

_.,. " .. " "

•••••

11

•• 8.

.. ..
••

fO---'

••

..

30 .:

.•••••.

2O..:rn4'~II'
10 ~

.

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~.

DOClD: 523696

REF lD:A523696
'FOP SECRET I:IMBRA

year. In June of 1973, moreove~, the Civil Service Commission authorized DoD to offer immediate annuities to individuals with twenty-five years of experience, regardless of age, or who were at least fifty years old with twenty years of service. In addition, a 6.1 percent cost-of-living increase was offered to those retiring before July 1. This did it - retirements in 1973 increased by 45 percent over the already-high level of the previous year. In the end, the RIF was never necessary." NSA's manpower bottomed out in 1975, as Table shows, and remained s~ady through the remainder of the decade, except for the military component, which continued to shrink slightly. It began its upward swoop in 1981 and topped out in 1989, the nominal end of the Cold War. (.G) Table 2 ' NSA's Manpower History, 1973·1993
_ •• Civilian ~ .• Military

eel

2

Thousands

301~------------------------------------------------~

(U) However, promotions were difficult to get throughout the decade. The problem was, the grade structure. NSA's average grade had marched upward from 8.96 in 1965 to 10.2 in 1972 (see Table 3). NSA was advancing faster than the federal average. In 1965 its average tied it for ninth place, while in 1972 it was in fourth. The grade problem led to a promotion freeze. Though it lasted only a few months, it damaged work force morale almost as much as the talk ofRIFs.

JR/f While NSA experienced a modest downsizing, the Service Cryptologic Agencies
(SCA) were devastated. Of the 39,000 cryptologic billets lost, almost 36,000 were military. Somec=J military billets associated with direct support and training were transferred into non-CCP (Consolidated Cryptologic Program) areas, so the net loss to the cryptologic system was "only"c=J The Army was hardest hit, losingc=J billets from its CCP structure. Security Service lost percent of its billets, while NSG lost more than

0

0

percent.'

Withheld from public release Pub. L. 86-36
HANDLE VIA 'fALElt'!' KEYHOLE eOMfNT eONTftOL B-YSTSMStlOnt'PL¥

23

TOP SECRET tJMBRA

DOCID: 523696
TOP SECRET UMBRA

REF ID:A523696

(U) Table 36 NSA's Average Grade. 1965-1972 NSA's Average
8.96 8.67 9.0 9.2 9.7

Year
1965 1966 1967 1968 1969 1970 1971 1972 .

All-Federal Average
8.3 8.3 8.4

8.5
8.8 8.9 8.9 8.9
i

9;9
10.07 10.2

(U) THE CLEMENTS

CUTS

~ NSA was in the middle of a desperate downsizing effort when, in 1973, it was hit with a round of-budget cuts which became known as the "Clements cuts." The real author of the directive was one James Vance, who worked for Dr. Albert Hall, assistant secretary of defense for intelligence and DIRNSA's immediate boss. Vance contended that cryptology was overfed and underworked, and he embarked on a detailed study of the cryptologic system. The upshot was a recommendation to Hall that cryptology be hit with an additional three percent cut. The Vance recommendation wound up in the office of Assistant Secretary of Defense William P. Clements. Clements imposed a total CCPbillet reduction of 12.999 to be completed by fiscal year 1978.7 (Since the cryptologic budget already showed a large reduction during that period. the real additional manpower cut was "only" 5,110jobsJ .(e)Clements specified that reductions were to come from 1. Management efficiencies. The crux of the problem, as viewed from the DoD level. was a bloated management system with overlapping authorities - basically. "too many bureaucrats." The answer would be to squeeze out the fat. without cutting into bone. . 2. Technological efficiencies. As will be seen later; NsA was looking at araft modernization proposals. chief of which was remoting (see p. 38). that would reduce manpower without substantial mission reductions.

of

HANDLE Vh\ TAU5N'F KE¥lleLE

GeMIN'F GON'f'ReL S¥STEMS dOlN'Fb¥

lOP SECAElYMBAA

24

DOClD: 523696

REF lD:A523696
lOP SECREllJMBRA

3. Mission reductions. This was a last option. At Clements's level, people felt that NSA could cut without reducing the mission . Lieutenant General Sam Phillips, who would soon be leaving NSA, answered that NSA recognized the "bureaucracy problem" and had just completed an internal reorganization that cut 649 spaces. Phillips felt that further efficiencies could be accomplished, especially through technology, but he cautioned Clements not to be too hopeful that NSA could do it without any mission cuts. He convened a panel to work through the reductions and come up with a plan," (U) The study group had tough sledding. The first reaction was a decree from the production side of NSA that it would not take a reduction until all support billets worldwide had been cut, whereupon the support organizations replied that they could not cut support until they saw the operational reductions. The SCA representatives were similarly obdurate." It was enough to make a budgeteer tear his hair out. (8 CeO) They slugged away during the summer and fall of 1973. When. in October, the results were due to Clements, Lieutenant General Lew Allen had become director. By this time the committee had forged some numbers which sounded a little like a congressional budget-cutting exercise, but which were plausible on paper. Allen told Clements that 1. Managerial efficiencies could absorb some of the needed reductions. The committee recommended cutting alt-deputy jobs below division level, consolidating some organizations that were split (such as A7 and A8), restricting hiring to one third of projections. virtually eliminating the analytic effort on Southeast Asia,reducing staff functions, and slimming down NSA overseas liaison offices. Overseas, support and managerial billets could be deleted by forcing closer integration of collocated SIGINT sites under the Single Service Executive Agent concept. A new concept in position tasking called COPES (Collection Operations Position Evaluation Standard) could theoretically reduce manual Morse positions by 25 percent. Since there were more than 0 Morse positions worldwide, this would have amounted to a significant savings. The SIGINT system would have to rely more on Second and Third Parties. Worldwide logistics would " be shaped into a more efficient mechanism, and some logistics operations would be contracted out. Some sites. I I, could be staffed by contractors. Army Security Agency and USAFSS had both built up theater-level administrative headquarters that could be eliminated without effect on the mission. 2. Technological innovations represented a higher risk option. The remoting program, I ' I, was still unproven, but Allen banked heavily on its success to save cryptology from the worst of the Clements cuts. Only the first site,1 I, was far enough along to count on. Other new programs with interesting and obfuscatory names like offered potential savings, b~t their contributions remained to be seen.

.ret

I

I

J lMiBLE VIA ''''LEf,y KE'tll6LE et)MU,y e6N'ft6L SYS'f'EMS JOJU'PLY

25

TOP SECRET I;JMBRA

DOClD: 523696
1'6P SECRET !;IMaRA

REF lD:A523696
1.4(c)

I E.O.

13526, section

I

Withheld from public release Pub. L. 86-36

I

3. Despite opposition from Vance, Hall, and Clements, Allen relied on mission reductions to make the mythical Clements's manpower ceilings. Some stations, like the Navy site at Todendon, West Germany, would ~ closed outright. The ASA trio of Herzo • . Rothwesten and Bad Aibling would be closed and the mission transferred to a new The Air Force site at Darmstadt would be cut, the operators moved to] land Single Service Executive Agent management would be applied to the new triservice station. The border sites in Ge"hnany would be closed. Back at Fort Meade, NSA would stop doing Cuban internal, all sub-Saharan internal and Middle East internal communications, 10

I

(U) Some economies were logical yet unattainable. The creation of Central Security . Service (CSS) the year before had created duplicate staffs at the NSA level. General Phillips had quietly scotched the operational effect of CSS, and the vestigial staffs had quietly taken on dual functions for the sake of economy, but the whole CSS exercise had made it more difficult to slim down because of the perceived need to keep up the appearance of a functioning CSS. The most far-reaching CSS proposal had been to bring the SeA headquarters to Fort Meade and collocate them with NSA, where, it was assumed, economies in the billet structure would be easier to effect. It had not happened and was not likely to happen in the future. The SCAs had successfully fended off collocation with "Mother NSA." 11

(s 'CeO) Lew Allen had replied with some well-thought-out planning options. Some, such ~s th~ Single Service Executive Agency, and. heavier reliance on Third Parties, came to pass. The elaborate and expensive remoting option was implemented in later years, although not quite the way Allen envisioned But other options like major reductions in the Air Force's Rivet Joint airborne collection program fell to operational reality (and determined opposition within the' parent services). Still others, like ccntractorization, simply transferred the cost to another budget category while yielding only minor savings.

I

it.

(8 Ceo) While NSA struggled to protecUts emphasis changed dramatically. \ \

resources from the budget axe, its mission

==oJ

[The real cuts had come at the expense of other production elements. The effort on Southeast Asia declined from 13 percent to 5 percent, whileG Group positions were down from 15 percent to only 8 percent,"

(U) THE FIELD SITES .. ....(€fIn 1970 the collecti~n site system stood at its highest level ever. Ninety-one sites were scattered throughout the world fromc:=J to Ethiopia. But the impending withdrawal from Southeast Asia, and the budgetary pressures that were moving J?oD toward contraction, were about to hit. I E.O. 13526,'section 1.4(c)

Withheld from public release Pub. L. 86-36 .
rop SECRET Ua411AA

26

DOClD: 523696

REF lD:A523696
TOP S!CltH UMBRA

,.OOrrhe collection site posture went into sudden freefall, and by the end of the decade only rllty-one sites remained. ASA was particularly hard hit, contracting from nineteen sites to nine. The Air Force lost half its sites, while the Navy, with a small-site posture and emphasis on worldwide DF, lost only seven of its thirty-six sites . ..(s COot In Japan, each service lost sites to a base consolidation movement. By 1975 all Southeast Asia sites were closed except for Clark Air Base in the Philippines. In Thailand, the closure ofRamasun Station resulted from a political forceout by the nervous Thai government. Farther west, the Turkey sites, with the exception of Si,nop,were closed at the request of the Turkish government, while the Stonehouse facility in Asmara was victimized by the fallout from the Ethiopian revolution of 1975. The Navy site at Nicosia was converted to the first overseas remotingoperation in the middle of a civil war. Moving round to Germany, a massive base ~onsolidation movement, which hit cryptologic and noncryptologic units with equal fervor, resulted in the closure of Herzo, Rothwesten, Darmstadt, Bremerhaven, and Todendorf, and the collocation of mission at the new.Army FLR-9 site at Augsburg. !.G')The closures resulted from a complex of budgetary pressures from Congress and difficulties with the host countries. The period after the Vietnam War was one of exceptional instability in the Third World, and cryptologic sites, long held hostage to foreign aid by' host governments, were battered about quite unmercifully. If they survived at all, it was usually in an altered, and less favorable, condition.
(U) Turkey

I

E.O. 13526, section 1.4(c)(d)

I

Withheld from public release Pub. L. 86-36

-

(ug
II
Withheld from nublic release Pub. L. 86-36
~

27

. T6P SECRET UMBRA

E.O. 13526, section l.4(c)(d)

Withheld from public release Pub. L. 86-36

\0 0\ \0 M N 10
HANDa; ViX JALl:141 itE1JI6te eSMIfJlteefl'PR9LSY6Rilf6 ••em'R Y

Q H

U

28

o Q

OOClO: 523696

REF lO:A523696
lOP SECRET!:IMBM

I

E.O. 13526, section 1.4(c)(d)

I

Withheld from public release Pub. L. 86-36

IIANBb5

VIA

'fAr:.em

K5YUeb5

eaMIN'!' eati'fR9=

&¥S'l'EMS ,J9ImL¥

29

lOP S&CR&T l:IMBRA

DOClD: 523696
1'6P SE(RET UMBftA

REF lD:A523696

Withheld from public release Pub. L. 86-36

I

E.O. 13526, section 1.4(c)(d)

I

Withheld from public release Pub. L. 86-36

Withheld from public release Pub. L. 86-36

I

E.O.13526, section 1.4(c)(d)

I

Withheld from public release Pub. L. 86-36

fl1diBt.fJ ViA fAbBtff

ItS¥i19bS e9MUff 09NTROL 6'ISl'EMS cJQIN'l'J.Y

.

TOP SECRET l:JMBnA:

30

DOClD: 523696

REF lD:A523696
'FOP SECRETUMBM

E.O. 13526, section 1.4(c)(d)

Withheld from public release Pub. L. 86-36

CU) Ethiopia

E.O. 13526, section 1.4(c)(d)

Withheld from public release Pub. L. 86-36
(U) Initially threatened by budget cutters, Asmara ultimately fell to a different foe Third World instability. The Ethiopian regime of Haile Selassie, widely admired for its courageous stand against Mussolini in the 1930s, had been enlightened and progressive, especially by the standards of the area. But as the emperor grew old, his attention wandered from the business of government.· Long-suppressed tribal rivalries became more important. In Eritrea, the Eritrean Liberation Front became one of the strongest of the regime's opponents, and warfare broke out. This was compounded by tribal unrest in other parts of the country and by leftist movement within an increasingly fractured armed ~ forces. In 1973 a devastating famine in Wollo Province killed thousands of people and brought unrest into the streets of Addis Ababa. The students were eventually joined by the rebellious factions in the army, led by a five-foot-three-inch martinet named Mengistu Haile Miriam, known ominously the "Black Stalin of Africa."

a

.

as

31

fOP SECRETUMBM

DOCID: 523696
'fOp SECRETl:JM8M

REF ID:A523696

(U)

Ethiopia

HJalJaperiaJ ~esiy HaiLeSelusie I and Hon. Edw8l'd M. Kerry, U.s.· IUIIbaasador tolEthlopia, January 1967
(U)

HMmbl!: VI••• '4':M.~Ff KE'iIIGbS GGMINTGGN'FR9h 8¥SHMS d91N'FJ,¥

TOPSECitET UMIlItA

32

·1 -

DOClD: 523696

REF lD:A523696
fep SEERn l:JMBRA

(U) The revolution was initially bloodless. Key members of the armed forces, parliament, and the courts were rounded up and taken away. In September of 1974 the ruling Dergue (Amharic for "committee") arrested the emperor himself. After that, Mengistu abandoned all pretense ofbenevolenc:e. The capital became a bloodbath, and the provinces were roiled in unrest, famine, and fighting.zo
(U) Even without revolution, Asmara had been under seige. When ASA departed Asmara, base support facilities devolved to the Navy. The Navy stayed for only two years, and when they left, the base lacked a school, a medical facility, PX, commissary, post office, and other necessary logistics. Limited support would continue under. a contract with Collins International, but that too would dry up in fiscal year 1976, after which time the base would be unsupportable. (U) What it did have was a mission, so the people stayed on, improvising as they could. Harris Corporation, one of the STONEHOUSE contractors, accepted a contract add-on to ,provide a doctor, while the Americans left stranded in Asmara organized a school with support froth the consulate. The school was located on Kagnew Station."

I E.O. 13526, section 1.4(c)

Withheld from public release Pub. L. 86-36

(U) When the decision was made to close, the NSA contingent at STONEHOUSE was down to about 200 people, including dependents, Everyone lived on the economy, but gathered frequently for social events at the base officers club. That was the case the evening of January 31, 1975, when fighting broke out. Trigger-happy Ethiopian Army troops began fU'ing, apparently at rebel forces, and shots ricocheted through the walls of the club, while panicked Americans crawled under tables get out of the line of fire, They waited through the long night on the floor ofthe club, the party at an abrupt end,

to

(U) The next day the site chief, David Williams, and his deputy, Lewis Walls,closed the mission forever. With NSA's blessings, Williams began inauspiciously moving _ American dependents out of the country on commercial flights. Through February the effort picked up speed, and by mid-February only sixteen Americans were left at STONEHOUSE. 'They were engaged in packing all mission equipment for shipment on Ethiopian Airlines to Addis Ababa for repacking and shipping out of the country. They burned all the classified documents, and tried in vain to destroy the KG-13 crypto equipment with .incendiaries, (Incendiaries were notoriously unreliable, and Williams and his men wound up hacking them apart with fire axes.) 2S

HANBLE 'ilcA: T:A:LEN'f KEYU6Ul

C6f1UN'f CON'i'R6L 8\'S'fEMSJ61N'l'IX

33 '

tpp SECRIiT UMBRA

DOCID: 523696
1'6" SeeRE'f tlMBItA

REF ID:A523696

(U) Back in Washington, a behind-the-scenes struggle raged. P~ilip Habib at the State Department, with strong support from Secretary of State Henry Kissinger, opposed moving the Americans out of Asmara. Although Kissinger had support within 000, he did not have the support of'NSA's director, Lew Allen. In an angry letter to Kissinger on February 19, Allen said:
I consider that there is no longer &Dy operational need for StonehoWie commenaurate with the risk to my penonnel I have directed ChiefStonehouae to further reduce his workforee from

16 people to 8 people If local Asmara conditions further deteriorate. and in any cue when the packing and crating of my equipment is completed. it is my intention to further reduce my peI'llOnnelin Asmara below the eirht DOted above ...• The safety of my people i:S paramount. The safety of the equipment is secondary.

The State Department authorized the closure of Kagnew Station only two days after Allen's strong letter. After the last piece of equipment was out, David Williams flew to Addis Ababa to supervise the shipment from Ethiopia. He himself' departed in April of 1975, the last NSA official out of the country.24
(U) Thaliand

(U) During the years of war in Southeast Asia, NSA had used Thailand as a principal base of'cryptologic operations. The original ceiling of 1,000 cryptologists, while being a nice round number, soon ceased having any relationship to reality, ~nd over the years NSA had brought more SIGINTersinto Thailand, taking care of' the increases with postfacto authorizations by the Thai government. After the 1973 Vietnam cease-fire, a large slug of displaced SIGllffers entered the country, to be officially authorized by the power less Thais.~ (U) With the fall of Saigon in April of' 1975, the end of the American presence in Southeast Asia was only a matter of time. U.S. forces began leaving the country soon after, and the formidable base structure that had come into being during wartime quickly imploded. So where did that leave the cryptologists? \
'(& The cryptologic presence in Thailand was only partly related to Vietnam.
,-., -e ,-.,

'-' ~ ,...;

CJ

'" ...a
M

= .S CJ ~
Irl

-- - .. . .c ._ ...:l
•• ~ I

= e

Moreover, there was still a requirement to monitor the new communist regimes in Vietnam, Cambodia and Laos.
L..I -'

<lJ I,C <Ii ~ <lJ1,C

I

I

•••••

"O~oo
<lJ CJ

~ ,...; ~

.c . •• .=;..=;.

.- = =
~ Q.~

0

IlANDLS VIA TALENT K:I5'Yn9b5 C9M1N'f CONTROl;.SYST!lMB cJOlmbY TOP SECRR tlMBR:A

34

OOClO: 523696
E.O. 13526, section 1.4(c)(d)

REF lO:A523696
TO' SECRETUMIIlt.
,------------,

Withheld from public release Pub. L. 86-36

(U) Negotiations with the Thais consumed the whole of 1975, but with no resolution. The Royal Thai Government would clearly have been relieved to see the last of American forces, which by late in the year was made up of the cryptologists and virtually no one else . .The American embassy was on the side of the Thais, since the loss of the last American military forces would remove a thorn in the side of American-Thai relations.

(U) But in the end it wasn't enough .. The Thai government was getting fierce diplomatic pressure from the PRe, with whom they were negotiating an improved relationship. Moreover, the Thai military-run government was being squeezed by an internal communist insurgency in the bush and an urban leftist student movement emanating from the universities. With the communists victorious all across Southeast

Withheld from public release Pub. L. 86-36

I
35

E.O. 14526, section 1.4(c)(d)

MA~mbE '1M 'FM.8Nf KJ!lYlI9bB 69MIN'I' 69N'PR9b SYSHMS J9JNl'DY

. 'tOP SitRR UMBRA

DOCID: 523696
lQP SEC:RETUM8RA

REF ID:A523696

Asia, everyone, it seemed, wanted to be on the winning side. America did not appear to be the winning side. (U) Udorn, the nearest large town to Ramasun Station, had a university, and it was full of restive students. In 1975 they got a cause, the infamous Leuchai incident. Leuchai, who managed the officer's club accounts, got into trouble with the base commander over the disposition of some monies and was summarily fired. But Leuchai had friends, and they brought out the students from the university. The base commander at Bamasun was confronted with daily demonstrations at the main gate. One day the military police, apparently thinking that the base area was sovereign American ~rritory, arrested Leuchai, and the demonstrations got larger. In the end, Leuchai was released, the American ambassador was upset, and the Thai government, with newly stiffened spine, was ready to order the Americans out of Ramesun, 28 (U) The order to leave did not come until March 20, 1976, but in the intervening months the diplomatic game went back and forth several times. Operations at Ramasun became chaotic, as stop orders were followed by start orders. So when the order finally came to get out in four months, NSA and ASA were ready for a scorched-earth evacuation. The operation was shut down that very day, and the first transports began arriving at Ramasun within eighteen hours of the order. Operators took up wrenches, and the entire operation was torn down, to the last nineteen-inch rack. Everything that could be carried otrwasloaded aboard C-141 transports which were arriving in waves from Clark Air Base. ~Within days, 33,000 pounds of equipment had been airlifted to Clark. The FLR-9 was rendered useless, and the station was turned over to Division Six as a gutted shell. The only things salvaged for Division Six were ninety-nine R-390 receivers. Although AFSC officially accepted the station, the idea of using it ior SIGINT operations was ludicrous. The bill to run the diesel generators for a month was higher than the entire Division Six annual budget.29 ~S eeO} The SIGlm redeployment plan specified that the mission of U.SM·7would be reconstlt~ted at Clark Air Base, home of USA·57, and that is where the people and equipment went. Unfortunately, no one thought to tell the Ameri~an ambassador, William Sullivan. When he found out, all hell broke loose in Manila, because the evacuation from Thailand had caused the cryptologic ceiling in the Philippines to go through the roof, so to speak. But Sullivan needn't have worried .. There wasn't room for the Ramasun equipment on the operations floor at Clark, nor were there logistics facilities to handle the flood of people. Just as germane, the Ramasun mission could not, by and large, be heard from Clark because of the vagaries ofHF propagation. (This had been known for many years by operators.) So the equipment wound up at Vint Hill, Virginia, and the people scattered to various SIGINT sites around the globe.. Clark Air Base picked up only fragments of the Ramasun mission. The FLR-9 electronics were never used again.so

HkNflLe 'ilk .'fALEN'f KEYilOLE eOMfN'f eOtf'fROL SYfN5M5 cJOm'fb Y
(

)

lap SECRElI;JMBRA

36

DOClD: 523696
E.O. 13526, section 1.4(c)(d)
(U) Closures and Consolidations 1S='eE3G+ In Germany

REF lD:A523696
lOP SEEREl UMBRA

all resulted from budget cuts. The I Iconsolidation plan had actually originated from a study in 19~7 which showed the economies that could be achieved by closing the ASA sites at Reno, Rothwesten, and Bad Aibling and moving the people and mission to a single location. ASA organized the original I cadre in 1968, and the station was officially up and running in January 1972. Two years later the Security Service site at Darmstadt was closed, and the people and mission joined the triserviceoperatiorr

I

I base closures

Withheld from public release Pub. L. 86-36

I

I

Withheld from public release Pub. L 86-36
(U) The Airborne Communications Reconnaissance Program (ACRP) also slimmed down. In the 1960s it had consisted of a welter of strategic and tactical programs under various jurisdictions and controls. An Air Staff study in 1971 showed clearly that the program could be more economically managed if it were consolidated as a single program under a single manager. The outcome of the study was the RIVET JOINl' program. Under it, the worldwide ACRP programs were consolidated into a single airframe, the RC-135. Twelve airframes were modified for both COMINT and EUNT collection by E-Systems in Greenville, Texas. The Air Staff recommended that the new Airborne SIGINT Collection Program - ASRP - be jointly managed by SAC and USAFSS. Moreover, the new program operated under the Air Force's MOB-FOB concept. That is, there would be a main operating 'base - in this case Offutt in Omaha, SAC headquarters- and forward operating bases in each theater. The crews and airframes would be based at Offutt and would deploy to the forward bases on TDY for missions: The new RIVET JOINT marked the first successful attempt to rationalize and centralize a large number of programs that had grown like weeds duting the Cold war.P".

i '

IIANBJ:J! VIA TALENT IUlYUOLE eOMn,'l' eOU'PROL SYS'f'EMS d9Uf'fb¥

37

TOP SECRET UMBRA

DOClD: 523696
TOP SECRET tlMBItA

REF lD:A523696

(U)

Tactical Systems

(U) The war in Vietnam had displayed the inadequacies of the tactical SIGINTsystems that had rusted away during the era of nuclear dominance. Vietnam prpduced a spate of· development programs to fix the problem . .(Gr The Army came up with several .entries. CEFIRMLEADER was an airborne communications intercept, DF, and jamming syste~ aboard RU-21 dual-engine ~ircraf't. that had proved SO useful to the ARDF program. CEFIRM LEADERsupported tactical commanders at brigade, division, and corps levels. A second program, CEFLY LANCER,was a modernized version of the ARDF program, designed for deployment to Germany. The Army, being decentralized, fragmented its SlGINTeffort.~ . «*1'he Air Force, being farthest behind the curve, had to develop a system from scratch. Their entry was COMPASS EARS,a complete tactical SlGlNT support system based in mobile shelters. The collection system, called COMFY LEVI, was mostly airborne - two mobile shelters stuffed into a slightly modified C-130. Processing and reporting were done in tents and shelters located well back of the combat zone. As with Air Force doctrine generally, this system was highly centralized. There would be only one per tbeater." . .

J£f The Navy was least affected by the commotion in Vietnam. What was needed was simply an updating of shipboard SIGlNTsupport that had existed since World War II. The new program was called CLASSICOUTBOARD,an automated system designed to work against mobile naval emitters,l I
.$Cf Even NSA came up with a "tactical" system. Thel program, an ELINT innovation, permitted NSA to deploy ELINTintercept equipment I I I I Thls highly successful effort was one of what would become a large number of quick reaction systems to work against specific technical problems. sa

I

(U) REMOTING (U)

Tennis

'(s'CCaTThe origins of cryptologic remoting were in 1962 and stemmed from an idea attributed to Joseph Horn, an NSA engineer. The first communicationssatellite, Telstar, had just been launched and, with it, a new era in communications. Horn, in a paper entitled "A Proposal for Utilization of Satellite Relays to Provide an Early Warning and Extended SIGtNTCapability within the ZI," proposed that NSA look into the possibility of remoting signals intercepted in one location to another. The technology, he felt, could be developed to send large chunks of the RF spectrum from an overseas location to a location in the United States. Horn justified the effort that would be required on the basis of improved timeliness, reduction ofSIGINT people overseas, and cost-cutting."

lOP SECRET l:JM81tA

38

DOCID: 523696
I E.O. 13526, section L4(c) I

REF ID:A523696
Ip·L. 86-36

I
I

'fe' SECRET UMBRA ,--------Withheld from public release Pub. L. 86-36

J$f The proposal generated interest, and in 1964 NSA conducted experiments to see if what Horn proposed was really possible.

I

\'------

It worked, and everyone was ecstatic. aut for several years, that was it. The idea languished, awaiting sponsorship. 40

~ Horn's idea was revived in 1967 when K Group (which at .that time dealt with collection and signals analysis) established a study group headed by Alfred W. Andrews. Andrews named his project "TENNIS," a name evoking a signal, as if a tennis ball, bouncing back and forth between communications satellites. Within a year Andrews had produced a preliminary concept for remoting] I I Iback to a location at NSA. (U)Jo8epb Horn I I sites were small, and the Andrews group simply discarded them from the study because the expense of installing the operational and communications equipment for such a small site would not be feasible. The group took it as a given that the, technology was there - what was needed was practical appllcation." ~ The TENNIS idea did not have many sponsors in the early days. In particular, Dr. Albert Hall, assistant secretary of defense for intelligence, was known to oppose it as too expensive and technologically risky. But within NSA Dr. Robert Hermann adopted it as his own, and he set out to get sponsors, He created an "Industrial Advisory Board" to study the issue and enlisted important people from private industry to help him. His first ally outside of NSA was William Perry of ESD, w.howould later become secretary of defense. Within NSA, he had the support of Oliver Kirby, the assistant director for production. With this level of support, Hermann embarked on a major feasibility study. '2 published in 1969, proposed to remote I I to collection centers in the United States. Candidate locations wer~ \ Petaluma in California, \ I The follow-on system development plan produced the following year planned for an initial system, called PILOT TENNIS, in which I . . I in the U.S. The presumed success of the pilot would result in a wave of support, and by 1975 some thirteen sites would be part of the TENNIS system. NSA would close seven European and Mideast locations and six in the Far East. A residual force of about 20 percent of the total would remain in theater for tactical support. The savings would be staggering. Overall CCP I E.O. 13'526, section 1.4(c)

I

(8=000) The original

I

Istudy,

I

Withheld from public release Pub. L. 86-36

}MNl)bE 1Jh\ 1\'d.5N'I' KE¥HOLE COMIN'f eON'i'R6L SYS'fEMSJOIN"fLY

39

TOP SlieR&; UMBRA

OOClO: 523696
TOP SE(RE'l' tlMBItA

REF lO:A523696

I E.O. 13526,section 1.4(c)

Withheld from public release Pub. L. 86-36

economies would range] could be eliminated. SOme cryptologists overseas would come back. But the .up-front costs were equally hug~ [for the system through 1978 and 0 I Ito acquire dedicated communications satellites that were presumed to be required.f .

I

I

I

"(S) TENNIS

,

produced arguments galore. The biggest dispute was over thel ,Iapproaches. Horn had originally envisioned remoting large portions of lto the States,l and Petaluma.

I

I

~ The competing technology came to be called the long screwdriver approach. In this method, the ooerator sittin~ in the U.S. would remotely tune a receiVer in an overseas location.

I

I
..£S..COO) TENNIS also produced arguments over management. Theoretically. every intercepted signal in the world could be collected into a single facility, unot a single room. .Where would such a facility be? Was there enough room at Fort Meade? How would it be managed? What would the relationship be between collection and processing? Would operators accept being jerked out of their overseas bases and dumped in the high-cost Washington area? What kind of morale problems would result? Many elements of the Production organization lobbied for a TENNIS simulation facility to test out all these problems - a fly-before-buy approach. The engineering side naturally focused on the technical hurdles and ignored the management implications. A TENNIS simulation center was planned, but was never implemented. NSA bought the technology without testing the management problems rust." eel Ultimately, NSA succumbed to cost considerations and went for the long . screwdriver technology. Even under thel [program, however, communications requirements were stupendous. For instance, remoting the I This was why NSA became the largest single user of DoDcommunications satellite capacity."

I

I

(s CCot Dr. Hall continued to hold onto monies that NSA wanted for TENNIS .. Hermann's approach was radical - rather than scale back on the program to reduce the threat, he sent Hall a new proposal expanding TENNIS to include sixteen overseas sites, virtually wiping out the SCE component of the cryptologic system. All CONUS operator billets could be civilianized, less a 25 percent residual for tactical support. Financial

Withheld from public release Pub. L. 86-36

I E.O. 13526,section 1.4(c) I
I1AtiB1J!l "'bit 'I'*LflN'f K£lYliaLEl eaMIN'

G9Pf'fR9L&¥S'HlMBcJ911'R'bY

lOP S&CRE:rUMBRA

40

OOClO: 523696

REF lO:A523696
Tep SECRET l:JMBRA

savings from pulling people out of overseas locations and putting them in a single collection facility would be huge, both in direct operational costs and in logistics and overhead. Hermann's forceful approach finally got a tentative go-ahead from Hall.4i1
I

j..e) Table

447 -

E8tima~d TENNIS Communications Requirement by Site

E.O. 13526, section 1.4(c)(d)

Withheld from public release Pub. L. 86-36

~ When the Clements cuts hit NSA in 1973, the TENNIS concept seemed a heavensent solution to the budget crisis. Lew Allen became the director in AUgust 1973, and he barely had time to put his hat down before confronting the issue. Remoting seemed to be the answer, and he promptly convened a panel to consider it. He called it the DRA WSTRlNG TaskForce.
(u) Allen came from the high-tech side. of the Air Force, and he was well connected with private industry, which he considered an essential partner in solving big problems. The task force was composed of only four NSA people, plus representatives from fourteen companies, including such industry giants as Lockheed, Hughes, and IBM. Lew Allen

Ifh~t91oEYM 'f>I<IoSl1'l' KE'fl19lJil

C9MIN'P C9N'1'ReL SYSTSMS cl9Uffb¥

41

Tep SECRET UMBRA

E.O. 13526, section 1.4(c)(d)

Withheld from public release Pub. L. 86-36

H

c

MANN

1[""

TeT XVTYF¥WQ'

F'OM~'1TCOll;:ag"i~ln;JQnPI'L·r 42

U

o o

Tet SEEllET IIMBM

DOClD: 523696 I
E.O. 13526, section l.4(c)
understood .help.4I

REF lD:A523696
Withheld from public release Pub. L. 86-36
TOPSECRETlJMBRA

that the cryptologic community could not work its way out of this jam without

...(CCC~ He instru~ted the group to conside~ only

modernize lor use remoting. (Standing pat was not an optiqn.) The objective was clear - they were to devise a SIGINT system that was much less costly than the one that existed . .(S eee) The task force cast aside casual tinkering and recommended radical surgery. Although they did consider modernizing the overseas sites, they ended up recommending that the whole lot be remoted. I ITask Force recommended that every site remainingl Ibe remoted to Fort Meade.

I

I

~ They

had t~o options:

I

r
J!J'f Savings under the modernization

option would be significant, ,but using the remoting concept they would far exceed the 3 percent cut mandated by Clements (see Table 5). Of course, DoD would have to wait a few years for the return. The entire remoting scheme would cost I to be spread over a period of years from fiscal year 1976 to fiscal year 1981. Although each year's personnel savings would be significant, the procurement costs would not be completely amortized until fiscal year 1983 - fully ten years down the road .

I,

.(e') Full remoting

would require

that back to Fort Meade;
L..-

ate. would pass
--'

rldata,that ?=====;- .••••. o:--=.re=m=o:£te::-:s:::u::-chL1:h:::u:-::g=-e-:v=oT:lu=m=e=-=s:-:;of the panel recommended

NSA

purchase its own satellites rather than rent from the Defense Communications Satellite System (DCSS). Purchase would be more expensive, of course, but the amortization difference would only amount to less than a year. ~

is.eG~ Table 551
. Thel

IPlan Costs
Remoting modemization

Current Number positions

D

00
f--

Personnel AnnualCCP cost Estimated cost ofremoting

~ ~

,{S:.CGeTThe organization at Fort Meade would be a riightmare. Here, the panel only hinted at solutions, but did originate the concept of the "problem center." which was to

Withheld from public release Pub. L. 86-36

I

E.O. 13526, section 1.4(c)
.n•••••••• v

~ _. 43

19P5~CRETl:JM8RA .

DOCID: 523696
TOP SECRET UMBRA

REF ID:A523696

E.O. 13526, section 1.4(c)(d)

I
I

Withheld from public release . Pub~.L. 86-36

have a long life. At the Fort, signals would be shunted to areas that worked certain problems - for instance, all would go to one area. This would permit customized processing operations and would reduce duplication. For . instance, the problem center I I would not require a timely reporting mechanism, while the problem center (or "PC~) would not need equipmentl ~for transfer to the computer complex in the base~ent. 52

I

I

I

'(Q Consolidation at NSA would permit the introduction of many efficiencies that ~ight be unafTordable in a dispersed system. The panel foresaw the automation of search through the employment of automated sean systems/

'-------~.

I

I

Withheld from pubIic release Pub. L. 86-36

~ What emerged from,the private sector's blue-sky planning was an implementation plan, I , It represented what the cryptologic community could get . cranked into the CCP, and it was much different from thel system. Under it, NSA scaled the system back to a far more realistic plan, more in line with the original TENNIS planning (see Table 6).

I

"

I

~ Out of the c=JbilIets at theC] 8.ffectedsites,Dwould'remain overseas to do tactical support, Peacetime Aerial Reconnaissance Program, and 0Ter o~tions that would be difficult (ifnot impossible) to perform from Fort Meade. Some ople would

Withheld from public release Pu b. L. 86-36

I E.O. 13526, section 1.4(c)(d) I·
IlAUBLI!l 'lilt. 'l'ld:J!lN'f KFlYil6L1!l e6M1N'f eeN'fROL 9'IS'fEMs .JelN'lL Y

Tap SECRET UMBRA

44

DOClD: 523696

REF lD:A523696

fe' SECRET UMBRA
be moved back tothe collection operation center at NSA, and the billet savings would be onlyc=J The plan allowed for some medernizaticn at the residual overseas sites, but .offered specifics in only one case - the Navy site which would stay largely untouched by remotingl I At Fort Meade, the "problem center" organizational scheme was adopted from the I ~ plan.

I

I

J.G1 While the] Iplan remained through the end of the decade, harsh . realities soon intruded. Remoting would incur very high initial costs, and the ever-present Dr. Hall was willing to proceed initially with only one site./

I
~ ~ ~
l.o ~
I

•••.• '~
l.o

\C

'"O~QO

].::! ~
~ c.. ~

(U) Not even I Isurvived intact. Pieces of it were eventually implemented, but they resulted from pressures and events not even anticipated when the plan was written. The name survived, but the eventual system could not have been. recognized by the original planners.
11'-_

~::O~ .- = = (U\l

16+ The first remoted site had nothing to do with the grandiose plans originating from .thel . Iplanningefforts,1 Instead, thel Ibecame the guinea pig for the whole system.

I

mrmLE

VI,* 'f'IcLEN? KEYJfaLE CaMIN? CaNfft6f:; S)"S'fEMS40JN'fbY

45

lGP SECRET YMBRA

DOCID: 523696
TOP SECRET UMBRA

REF ID:A523696

Withheld from public release Pub. L. 86-36

I

E.O. 13526, section 1.4(c)(d)

I

Withheld from public release Pub. L. 86-36

Withheld from public release Pub. L. 86-36

19P SECRET tJMBItA

46

DOClD: 523696

REF ID:A523696
Tap SECRET tJMBItA

Withheld from public release Pub. L. 86-36

I E.O.

13526, section 1.4(c)

I

Withheld from public release Pub. L. 86-36

Withheld from public release Pub. L. 86-36

WAI>1'DY; ¥I'" 1"'1:01'>1>11 KE¥H9I:oE 09MI!fl' 09!fl'a9hS¥STEMScf9INTh¥

47

DOCID:523696
'fOP SECRE'f UMBRA

REF ID:A523696

Withheld from public release Pub. L. 86-36

E.O. 13526, section 1.4(c)(d)

Withheld from public release Pub. L. 86-36

Withheld from public release Pub. L. 86-36

HId~BtiE

-iIA T/d"flNT KE"t 1I6t1f.l eaMIN'!'

06NTR8b-

S¥SifiMS

cJ91N'FC '[

TOP SECRET l:tMlItA

48

DOCID: 523696

REF ID:A523696
. fOP SECRET UMBM

I E.O.

13526, section 1.4(c)

I

Withheld from public release Pub. L. 86-36

(U)AROF

I E.O.

13526, section 1.4(c)

I

Withheld from public release Pub. L. 86-36

Ib\NBbI!i VIA 'I'1tbEN'f KE'iH6LE e6MfN'f e6NTR6L SYSfilMS J6IN'fLY

49

Tap SECRET l:JMBRA

DOCID: 523696
T9P SECRET I:IMBRA

) REF ID :A523696

I E.O.

13526, section 1.4(c)

I

Withheld from public release Pub. L. 86-36

(U) The technology was different. though. ReA had custom-ciesigned the collection consoles. The Hazeltine receivers had an autostepping feature which eliminated handcranking a tuning knob in the time-encrusted method used since the invention of the radio. The time delay inherent in AROF remoting was almost a second. so for the operators everything seemed to be stepping in slow motion. The IATS system which still dominated the field was not in evidence in AROF. Instead. each position was equipped with a minicomputer to digitize the collection for later proeessing.f.9
(U)BROF Withheld from public release Pub. L. 86-36

I E.O.

13526, section 1.4(c) Withheld from public release Pub. L. 86-36

I

rtXNDL! ~ 1;1\ TitLENY KE'f HOLE COMINT eON'fROL S'lS'fBMS cfOfN'fb¥ T9P SECRET UMBRA

50

j

OOCIO: 523696
E.O. 13526, sect ion 1.4( c)( d)

REF IO"A523696 "

I

lOP SECRET UMBRA

Withheld from public release Pub. L. 86-36

(U) Remoting the Sinall Sites
~ TENNIS was never intended for the smalJ sites I I It had become a truism early in the project that the cost of earth terminals and ancillary equipment would make such a proposition uneconomical. TENNIS, DRAWSTRlNG,C] lall presupposed thatl Iwould become candidates for remoting .

I

..(Ceca) The implementation or remoting stood this assumption on ~ts head. As it turned out, the big payoff was in small-site remoting. Part oftms resulted {romthe deeline in earth terminal costs, but mostly it related to the importance or the mission. The small sites, with their land highly selective focus became the high value items in the system. .

I

I

I

(TSCCO TK) The fll'st step was data linking, in which operators at overseas sites intercepted signals and plugged the receiver outputs into communications channels.

With~eld from public release Pub. L. 86-36

I

E.O.13526,-s-ec-ti-o-n-l-.4-(c-)-(d";"')-

UA:NBL6 Yb\ 'ftd;Elfl' KEYUeLE eeMlN'f eeN'i'R6L S'lSftlMS d91N'£L¥

51

TOP SEeR!T tJMBRA !

DOClD: 523696
'FeP SECRET tJMBRA

REF ID:A523696

Withheld from public release Pub. L. 86-36

E.O. 13526, section 1.4(c)(d)

I

Withheld from public release Pub. L. 86-36

UA-NBI:B ViA TAI:BNT KeYI let:l!l eeMllff

e6foftIt6L SYStEMS J6lfftLY

TOP SECRET \:JMBRA

52

DOCID:'523696 I E.O.
13526, section 1.4(c)

REF ID:A523696 I
T()p SECRET l:JMBRA

Withheld from public release, Pub. L. 86-36

\'l's-ec~The advantage of data linking was speed - critical signals could be intercepted, forwarded and processed in something approaching near real time. It did not ' remove the operator at the distant end, nor did it reduce the number of people in the system. The operational payoff could be significantl Ibut these operations did not help withoverseas visibility, international balance of payments, " or CCP reductions.
'"$) The next system was a true remoting op~ration.1 an Army-sponsored project, sprang from the dismal budget-cutting days of the late 1960s, when ASA was .strapped for cash and looking for,a way to'reduce expenses. Th~ sites, although top producers, had been a financial drain for years. They were expensive to keep operating,\

I

I

/ tTS-CC9-) NSA recognized immediately that the potential payoff fori rem'oting w~s far greater than ASA realized.'

I
lTS-C~'In a lengthy memo in late 1971, MajorJGeneral John Morrison, NSA's ADP (assistant director fo~production, i.e., DDO), laid out the prospects. \ Icollection had to be data linked back to NSA. ASA's I Iwas a good idea, but it got the material only part of the way home. NSA needed a data link to get I I to FortMeade.7i ' ,----------:-I E.O. 13526, section 1.4(c)

I

I

Withheld from public release Pub. L. 86-36

IlJ.NBh£ WA fAbSl\''f' JfBYUebS C9MINT SeN'FR9b SY-STSMS49lNTbY

53

l6' SECRET l:JMBR:A

OOClO: 523696
"FOP SECRET l:JMBRA -

REF lD:A523696
~

I E.O.

13526, section 1.4(c) :LJI

..

Withheld from public release Pub. L. 86-36

.~ NSA's engineers became involved I from its inception, and in October of 1970 the ASA project manager, Colonel Vernon Robbins, formally invited NSA into the development process: ASA resources were strapped, and onlylNSA could provide the expertise to steer such a large project. NSA's.Richard Bernard was named the deputy project manager." .

r-

is-GCO) The combined ASAINSA project planning committee selected Radiation (later
called Harris) Corporation as the prime contractor and let a contract for $25 million. The committee had to scale back an early proposal I I

I

rAIillough NSA and Harris became ensnared in the almost inevitable cost overrun disputes, the system succeeded technically and operationally.lIl ~ For NSA, the payoff was the d~ta link. I

I
Withheld from public release Pub. L. 86-36
HANDLE VIA TALENT KEYHOLE COMINT CONTROL SY:

I E.O. E.O: 13526, section

1.4(c) 1.4(

I

TOP SECRET UMBRA

54

DOClD: 523696

REF lD:A523696
lGP SECRIiT UMBRA

U Guardrail E.O. 13526,section 1.4(c)

Withheld from public release Pub. L. 86-36

$) Once remoting was available, everybody wanted it. The earhes ie app ica ons were in Southeast Asia. where NSA began remoting signals from isolated mountaIntops during the later stages of the war in Vietnam. Called EXPLORER, this program got people out of danger zones and back into defensible base areas, while leavini the equipment {antennas. receivers, and communications) in exposed locations. The aptly named Black Widow Mountain along the Cambodian. border was the most famous of the remoting operations.

~) Remotingwas next employed to fIX serious SlGINT support problems I I The problems I Iarose from the disparity between tactical systems available to field commanders and strategic systems tailored for national-level support. By the early 19705, strategic SIGlNT had far outrun what was available tactically. In September of 1970, I Icomplained to Admiral Gayler (then DIRNSA) that his SlGINT support assets were not what they should be. I I I I His mobile collection equipment was antiquated I I Moreover. the intercept vans. I I 1 were too slow to get out of the way in case of attack I I I Communications were clearly inadequate] . I

I

I

-tG)1 knew about the systems that had been devised for Southeast Asia. and he wanted them j I He wanted airborne systems that did not have to retreat over roads that were vulnerable to interdiction. He wanted communications to get the intercept back to safe areas where they could be processed. And most of all, he wanted ARDF.83 I.
ts) At NSA. Gayler instigated a planning whirlwind. He sent an NSA team I I to look at the situation. The team devised a radical solution - an airborne remoting operation similar in concept to the! Iin Southeast Asia. When the matter came to a head in a JCS meeting in January of the following year, NSA was ready with the solution. The Agency called it GUARDRAIL. 114
'tCl GUARDRAIL would\

I

I

1'S=€GQ)

The first test was only partly succes·sful.

I
I E.O. 13526,section 1.4(c)

~======~--------------------~--~[
Withheld from public release Pub. L. 86-36
U. NIl' Ii' 1l[' TAoI I~

UYUObi

CQunn Cg~ROJ. S¥iTims JOl~Y
i

5.5

. tOP SECREt UMIIR.l

DOCID: 523696
TO' SEeRET tJ MBRA

REF ID:A523696

I E.O.

13526, section 1.4(c)

(U) Guardrail

aircraft

Withheld from public release Pub. L. 86-36

ffi.-OOO)

GUARDRAIL

II

was a s

ctacular success.

~ Early GUARDRAIL was an Army-specific asset. Despite the fact that air-related intelligence dominated the collection "take," the Air Force participated reluctantly, and then only after considerable prodding at the JCS level. One Air Force problem was survivability. The U-21 was a propeller-driven utility aircraft! I I The U-2 would be a far better platform. 88 It may also have been

I

Withheld from public release Pub. L. 86-36
TOP SECRET t:JMBRA 56

I E.O.

13526, section 1.4(c)

lIA1U)LE • I* TALENT KE i'HOLE eOMmT eOH'fROL 3'fBTEM5JOUfl'L ••..

DOCID:

523696
section 1.4(c)

RE'F ID :A523696
lOP SECRET UMBRA

I E.O. 13526,

I

that the Air Force feared Army dominance and wanted to use Air Force money to fund its '----------' own systems. «»-CUARDRAIL II became the final system. Even prior to its deployment, the Army, '--_---.II and NSA had all agreed that it would be left behind to provide tactical support. , There were no plans to fund a production system. at tested

Withheld from public release Pub. L. 86-36

';::1

This changed radically in 1972. Major General John Morrison proposed an lto do the same work 1 Ithat GUARDRAILwas doingD =='I-:A:-:t-a-s-:-tr-o-:-k-e-, -=-NSAwould be satisfying the constant demands of American ine=Jto improve SIGINTsupport and add a DF capability.
91

~

.

commanders

~CCO) The final system, called GUARDRAlLIV,looked a lot like GUARDRAIL I but it did not solve the strate 'c-tactical interface roblem. It used U-21s,

I'

It remained an integral part of the strategic SIGINT system. ' Once again, the Air Force entered the system reluctantly. Its concerns probably related to afear that GUARDRAIL IV threatened the continued viability of the RIVET JOINT fleet, rather than to any criticism of the way the program operated tech~ically or conceptually. 92

(U) REORGANIZATION (U) The war in Vietnam produced wide dissatisfaction with the performance of intelligence. This was in some ways unwarranted.. It had performed better than in Korea, and the problems that beset intelligence early in the war were on the way toward solution' by the time Richard Nixon became president in 1969. But the perceptions persisted and led to demands for change .. (U) The Fitzhugh Panel, (U) When Nixon assumed office, he called for a reexainination of the total Defense etTort, appointing a blue ribbon defense panel to recommend changes. The panel conducted the broadest review of the Defense Department since the Hoover Commission of the mid1950s. Part of that effort was a Panel on Command Control and Defense Intelligence

Withheld from public release Pub. L: 86-36

I E.O. 13526,
H.Y'lDJ:.S'/M'fM.EN'fJtS'fI19l:J1lCaMINfCON'fReLSYSTEMSJOfNTLY

section

1.4(c)

I '

57

+oP S&CRET UMBRA

DOCID: 523696
TOP S!CR!T tJMBRA

REF ID:A523696

chaired by Gilbert W. Fitzhugh. This committee consisted primarily of industry figures . . and lawyers and was clearly intended to represent a totally dispassionate view of Defense intelligence. lIS . ,
.

(U) The committee discovered that management was fragmented (not the fll'St time someone had discovered that salient fact), uncoordinated, and not well focused. There appeared to be no effective control of intelligence requirements, a great deal more information was collected than was required, and consumers were overwhelmed by a welter of disjointed reports from all corners of the intelligence structure. DoD had never developed a substantial corps of intelligence professionals. (The only exception appeared to be NSA, which had obtained special legislation.) (U) Fitzhugh recommended that the Office of the Secretary of Defense focus intelligence management under a single deputy, called the assistant secretary of defense for intelligence. (At the time, intelligence was loaded onto the assistant secretary of defense for administration as an additional duty.) Under him there would be a Defense Security Command (consciously modeled after the NSA structure), which would enjoy broad authority to supervise DIA, NSA, and all other Defense intelligence." Such changes might have been logical but politically fell very wide of the mark, The Fitzhugh Panel had little ultimate influence over the course of actual events.
(U) The Schlesinger Study·

(U) The Fitzhugh Panel had no .sooner submitted .its report than the president commissioned another study. But there were differences. This new study, chaired by James Schlesinger, head of OMB, dealt exclusively with intelligence, while Fitzhugh had also looked at command and control. More important, Schlesinger examined all of intelligence, while Fitzhugh had looked only at the Defense Department. 9~ (U) Not surprisingly (considering what job he held), Schlesinger concluded that intelligence centralization could best be effected by giving the DCI broader budget authority. Nixon invested then-Del Richard Helms with .a broad grant of authority to review all governmental intal l ige nee activities in order to rationalize programs and priorities within

.(U) Jame. Scbl •• lD,er

IIAffDLE •.•• IA 'fAUlN'f IES'fIf6tEC6MIN'f

CONTROL S¥SiEMS cJ91NTL¥

fOP SECRET UMBRA .

58..

DOClD: 523696

REF lD:A523696
lOP SECRET tJMBJtA

. the budgetary structure. But Nixon and Helms did not get on, and the president never followed this up with specific authorities for his DCI. Helms was left to study, to coordinate, to cajole, but he was no closer to reigning in the disparate parts ofintelligence, particularly those in Defense. He never did get what the Schlesinger study promised . him." . . (U) Helms did accomplish one thing, however, that had long-range effects. He created a small staff, composed of a cross-section of the intelligence community, to look at the budgets of the respective (and disrespectful) agencies. This staff still existed at Langley in 1973 when Schlesinger became DCI. The new intelligence chiefs intentions went awry as he struggled to contain the damage CromWatergate by reorganizing CIA, but he definitely • intended to grant that staff more power. William Colby, his successor in the job, pushed the status and authority of Schlesinger's small stafr, which had become known as the IC (Intelligence Community) Staff. At the time, President Ford issued a new executive order putting teeth in the IC Staff's authority to control the budgets of the warring intelligence agencies, and in 1978 President Carter issued 'the executive order which gave the DCI "full and exclusive authority for approval of the National Foreign Intelligence Program budget." By then the IC Staff had moved into its own quarters in downtown Washington, and thus attained its own facility, with its own identity. 'n

(U)CSS
(U) The cryptologic reorganization that occurred in the early 1970s was the culmination of two decades of conflict between NSA and the JCS over control of cryptologic assets and operations. As NSA gained more authority and as the cryptologic system became more centralized, Pentagon officials became less and less pleased. A decade of war in Vietnam had produced, among other things, an internal war over cryptology. NSA's attempts in the 1960s to further centralize the business were bitterly opposed within the JCS, which had embarked on efforts to fragment SIGINT by shaving off small areas that· they could call by different names (electronic warfare - EW, electronic support measures ESM, etc.) and rid itself oC the codewords that controlled dissemination. By the'time James Schlesinger looked at the organization of intelligence, thedeepfissures between NSA and the armed services had become almost unbridgeable. (U) Schlesinger intended to solve the problem for all time, in NSA's favor. Clearly driven by budgetary concerns, he proposed to stamp out any JCS control over, and even involvement in, the SlGINT business. The dispute over the control of cryptology that had continued since the end onVorld War II would come to an abrupt end.. (U) The "end of the war" came on November 5,1971, when Richard Nixon announced the conclusions of the Schlesinger Study. Buried in the text of this "Nixon Jetter" was the announcement that, by the first day of the' following year, there would be a "unified National Cryptologic Command" under the director, NSA, for the conduct oCUnited States government communications intelligence and electronic intelligence activities. 98

I+AN91:d5 11ff< 'fttbENf

KFJ"1l0LFJ COMlm'

CONTROL

SYI9'fEMS JOIN'fLY Tep SECRET lJMBAA

59

OOClO:

523696
TOP SECRET UMBRA

REF IO :A523696,

(U) And then controversy erupted. What was a National Cryptologic Command (NCC)? What did the president intend it to do, and what were its authorities? Was this really the end of SeA independence? What would the new organization, control? What was meant by "command"? Many, both within NSA and without, felt that it meant the death ofthe SCAs, and a new organization chart was even prepared showing all service collection activities directly under DlRNSA. One view was that the chief of the NCC would also serve as DlRNSA. In one role he would control the national cryptologic system as before; in the other, he would command the SCAs through the JCS chain of command. Most agreed that the SCA th~ater headquarters would expire arid that their functions w~uld be effectively assumed by existing NSA theater organizations. The opinion of Admiral Gayler counted the most, and Gayler viewed his role as akiri to that of a Unified & Specified (U&S) commander, with total control over assets within his purview.
(U)
,,-, <:J

= .9 .•..
~ '" vi
<:J
If)

-

'-' ...,.

(8=CCO) In the Pentagon, near panic ensued. Theoretically, the NCC would control all

~

N

~

.. . -= ..: -> .
~

0

.- c..~ ==

o '" ..= ~ ~ ~~ ~oo "0 ~ .::! . ...:l .•... ..c..c
I

5 ~~

SIGINTcollection.This could include the Navy's VQ squadrons, the Air Force's EC-47, and the Army's U-21 ARDF capability, the overhead mission ground stations, tactical EUNT(including the Third Party programs that the Air Force had guarded for so many years) ~ IUnder its NeC hat, NSA might begin managing Army and Air Force tactical SIGINT programs rendering support to field commanders. At the very least, the struggle to control EW and ESM programs would be resolved in NSA's favor.

I

I

(U) DIA predicted that NSA would swing hard toward satisfying national requirements and would cease paying any attention to the satisfaction of the SIGlNT requirements of tactical commanders. The independence of the SCAs would end, and, . worst of all, tactical ELlNTunits would find themselves answering to NSA through the NCC.1l9 (C) Within NSA a certain smugness settled in. The war was over, the battle was won, and to the victor belonged the spoils. The spoils consisted of those SIGlNT assets that had formerly been controlled by rival factions: primarily the. armed services and CIA. As November faded into December, plans were being laid to assume control of the outlying assets that NSA·had never owned. This was a big win - a major revolution in the way cryptology was handled. (U) But things began to go awry even before the end of the year. On December 23, Secretary of Defense Melvin Laird informed G~yler that 'the new organization would not be a command - it would be called the Central Security Service. Implicit in the new name was a diminished world view. "Services," after all, could not exactly "command." Laird instructed Gayler to come up with an organizational plan and to create the new organization by February 1, 1972, a slippage of one month from Nixon's or igina.l deadline. 100

IIAND~

VIA 'fALEN'P ICSYlIeLEl eeMfN'f eePffROL S'IS'feMSJOflffL¥

lOP SECRET UMBIbtc

60

DOCID: 523696

REF ID:A523696
lOP SECRET l;IMBRA

(U) Concurrently, a new NSCID 6 was being written. Issued in February of 1972, it gave NSA significant new powers - and failed to give it others that, in the heady days of November 1971, folks at Fort Meade assumed they would get.
)e) The directive officially established CSS, which would be collection oriented, and would "include SlGlNT functions previously performed by various Military Departments and other United States governmental elements engaged in SiGINT activities." It did not define these functions, nor did it refer to CIA, which by omission managed to hang onto its SIGlNT system. The mobile SIGINT system remained under military control, thus answering one of the biggest questions which had arisen from the Nixon Letter. But in NSA's favor, NSCID 6 resolved the EW issue by placing it under NSA control. And on the administrative front, NSCID 6 gave the director authority over tasking,logistics, research and development, security, and career management of personnel.'?'

(U) Following Laird's decision on December 23, Gayler created a ~ries of internal . panels to flesh out the CSS plan. Progress was uneven because no one seemed to agree what it should be or how it should function. Gayler gave the task of managing the disputatious committees to Paul Neff, a World War II cryptologic veteran who had held. key positions in NSA's policy councils for many years. Neffs most vital assistants were M~or General John Morrison for operations and Frank Austin for training. Much of the action fell into their bailiwicks.l~ . (U) Under severe time constraints (the plan was due to Laird by February I), the committees solved the easy problems and left the tough ones for later. The new cryptologic system would be unitary, with centralized control and decentralized execution (hardly a .new or controversial concept). It would be composed of NSA and the SCAs as they then existed, thus putting' off the question of the system acquiring assets then controlled by the JCS and CIA. The SCAs would provide men, equipment, and facilities ., CSS would operate the system. (U) CSS would be headed by DIRNSA in a dual-hat role, and it would be assisted by a staff of its own. Composed of some 205 billets (75 from operations), it looked just like the NSA staff (see Table 7). All the staff heads were dual-hatted with their respective NSA jobs - thus John Morrison was both head of NSA production:and chief of CSS operations, while Frank Austin headed NSA's training school and CSS's training organization. to! (U) The CSS plan produced serious fissures between Gayler and the SCA commanders, who viewed the new organization as the the death knell ofthe independent SCAs. So they fought back, and the struggle spilled over into almost every aspect of cryptologic crganizatlon. They fought the training plan because the role of training and equipping servicemen forcryptologic duty had always been central to their being. They fought NSA's encroachment into R&D and logistics in direct proportion to the size of their respective staffs in those functions.1M

/

H/.~T~LI51,<1••• 'MUSH'!' KEYUebS DeMIH'F DemReh

SYsrEMBJeINThY

61

fOP SECRET YMBRA

DOCID:

523696

REF ID:A523696

I
"

OIRtCTOR.

f":UIl:r. css

~

W$It

I

Withheld from public release Pub. L. 86-36.

r ~~'Rl

I

I
T
_""ie_UCW\s seteeee "~chnt\IXJV

D£,CH

en

I •.• ~ll e"',••••
"1

I h'.OM.'

I
tn ••• uedClnsl
lA4~~\C..

. M"'.'I!_ftt

I,.·....... 11$··.••. c_-II
I
SlCINT ProduCtion

I : \ r.,;oRIMI

I
t (
a,. •.•• Ion. t

,

IAql1uca

IH

l
Plel'U\tn,

I

en

...
I
De ••••1DPfMftl

/ .....«. '"lle .......n"••'••• ] [
SeeWII't

,

Ij
.

l
C,yptol0911: _ SchOol.

H.l~'"

I
., AAU

I
atP$

,

.

U.$.A'MY A".ftCy

$.0·"'11'

I
U•.i.N.".J
Sec"IUy Gtoyp

I

1r=~:'1

Co...anct

!r
[

l

U.S.41r S~wrUy .s.twlce

rwo·l

I· 1

1

I

1

tJ£U)
ACTIYlncs

\

DOClD: 523696

REF lD:A523696
lOP $iCR.ET UMBM .

(FOUO) A struggle ensued over cryptologic organization in the theaters. Gayler wanted SCA theater offices to collocate with the senior NSA/CSS headquarters, but eventually agreed that they could collocate instead with the component command headquarters. The senior SCA commander would be responsible for the SCA and CSS functions, and most of his people would do the same. Gayler also wanted component command level esGs to be NSA elements, and went toe to toe with Major General Carl Stapleton of USAFSS over this issue. Stapleton won, and all component command CSGs became part of their parent SCA. The chief was the senior SCA representative in the theater.105

(U)~r

General Carl Stapleton

(U) They enlisted U&S commanders to defend their interests. Admiral McCain, CINCPAC (which would soon become Admiral Gayler's own command), predicted the beginning of the end of responsive SIGINT support:
In summary, tne proposed plan is viewed as placing in concrete the sterile, innerently unresponsive centralization philosophy to which field commanders have so long been opposed. The centraliution of SIGINT haa not been tested in a major conflict. The concentration of analytical functions at !.he national level will soon C8UM a decline in the ability of !.he uniformed cryptoJogic activities to function responsibly in a support role in combat operations especially when access to !l national database is denied and integration with other intelligenCe data is vital. The proposal is a lorig step backward in the Armed Services quest for more responsive intelligence .... 108 .

...(€1 The most contentious issues related to resources, and it was here that NSA had eyes bigger than its stomach. In the first heady days ofCSS planning, many in the Agency envisioned swallowing every SIGINT collection asset worldwide, the theater ELINT centers, and even scientifie and technical centers like the ,Air Force's Foreign Technology Division. ffiTln April of 1972, Admiral Gayler convened a panel (which he himself chaired) to survey the field. The most cursory study revealed a very wide field indeed. For instance, NSA discovered that at least forty-three submarines had EUNT collection gear, as did all Navy surface combatants. The list of CIA sites was very long, and the theater ELINT centers were very well-entrenched tactical assets . .~When the smoke cleared from the battlefield, NSA had won operational control over some of the assets under contention, most notably Air Force SIGINT platforms doing national jobs. But theater EUNT centers remained under theater control; programs designed for purely tactical jobs stayed with their parent services; the Navy held onto its
i

HANDLE VIA TALENT KEYHOLE COMINT CONTROL SYSTEMS JOINTL Y

63

TOP SECRET UMBRA

DOClD: 523696
TOP 5&CRIiT UM8RA

REF lD:~523696

entire fleet of airborne SIGlNTreconnaissance aircraft; and the Army kept its electronic warfare companies. CIA assets were not even filtered into the mix, and NSA's relationship with Langley remained on hold.!" When confronted with determined service opposition, Gayler had elected to smooth the waters. (U) One of the key aspects of the CSS reorganization was to collocate the headquarters at Fort Meade, and a new DIRNSA, General Samuel Phillips, began looking at this in the fall of 1972. The move was superficially attractive because of the money that could be saved, and it would certainly permit further dual-hattingofSCA and NSA staffs. The idea did not begin to burn itself out until a study group quantified the amount of space needed: 550,000 square feet, to be exact.at a cost of $30 million. NSA, chronically short of space, was busy expanding into the Baltimore suburbs and could offer no space to the SCEs. It might be possible to get some office space on Fort Meade from 1st Army, but it was still inadequate, even if it could have been converted into cryptologic work space (a very doubtful proposition indeed). So the idea was virtually dead anyway when Major General Stapleton confronted Phillips with the most determined opposition that any aspect of CSS .had faced. It was obvious that the Air Force would never agree, and the plan was dropped.lo~ As Phillips later said, rather laconically, in a message to the theater cryptologic chiefs, n••• there is specific and determined opposition by the SCA chiefs to such collocation. It is the expressed view of the SCA chiefs that proximity to their service headquarters is more important than collocation with NSA/CSS." 109 . It was the understatement of the year. (U) At the Defense Department, Dr. Albert Hall told his chief of resources management,. Lieutenant General Phillip Davidson, to keep watch over the implementation of CSS. By January of 1973, Davidson's watchdog, Robert E. "Red" Morrison, was ready to throw in the towel. Morrison wrote to Hall that the CSS staff concept had not worked. Agency employees had not accepted the dual-hat idea and were . not ready to relinquish their carefully garnered authority. According to Morrison, " ... the 'dual-hat' concept has served mainly as a way to keep the status quo." NSA had never transferred authority over tactical SIGlNT assets to CSS, and field commanders had reciprocated with suspicion and mistrust of the CSS mechanism. CSS had cost NSA over 200 biJ lets and had produced nothing in return. (U) At NSA, Sam Phillips had seen enough. Lacking any semblance of DoD support, and unwilling to make the drastic changes in CSS authority that would be necessary to keep the concept functioning, Phillips killed it. The date of death was listed as April 16, 1973. On that date, Phillips eliminated the CSS staff, transferring authority instead to a new deputy director for field management and evaluation (DDF), who also became deputy chief, CSS. He dropped the idea of dual-hatting and instead transferred authority for CSS activities to existing NSA positions, elevating them at the same time to deputy director status. Thus assistant director for production became deputy director for operations, communications security be~ameru]edby a deputy director, and Phillips created the post of deputy director for research and engineering, with authority over boVt NSA and SCA

.

HAf4BLfi! 'f fA 'f)\Lfi!N"f ItEYII6LI!l e6MIN'f eerffft6L

S'[SftJMS rJ 91H1'LY

lOP SECRET \;JMBR:A

64

DOCID: 523696

REF ID:A523696
TOP SECRET I:IMBRA

research efforts. Other staff chiefs were elevated to assistant directors; all had additional responsibilities for CSS management.?" . . (U) In 1976, when a new director, Lew Allen, went looking for CSS, he found only a paper organization. Associated with CSS, his resource people could find only General Allen himself (he was named on paper as chiefofCSS); the DDF incumbent, who served as the deputy CSS; and a military stafToffewer than ten people.111 (U) The CSS exercise benefited the cryptologic system by further centralizing such functions as research and development, personnel administration, and certain aspects of logistics. In these areas, NSA's staff authority expanded into areas that were of common concern to NSA and the services. The biggest changes were in training, where Frank Austin, the dynamic leader of the National Cryptologic School, presided over a long-term centralization of training functions, and a rationalization of the system to the point where the individual SCAs served as executive agents to separate aspects of a:now-joint training system. And, though the meetings were often stormy, the SCA chiefs were brought into closer contact with Gayler and his staff. Gayler institutionalized this ,into Wednesday morning breakfasts with his SCA chiefs, and thus brought a more direct and personal atmosphere into what had been a remote and long-distance relationship.i'" (U) So in certain respects, the addition of "CSS" to the NSA logo marked a permanent change in the way business was done. But the larger changes that had been so keenly anticipated in the fall of 1971 would have required steamroller tactics worthy of Brownell at his best. The JCS had been bested by Brownell in 1952 because he had the backing of the president. Twenty years later the president was not engaged, and the JCS won.1l3
(U) The Murphy Commission

(U) The period following the Vietnam War was extraordinarily fruitful with reorganization studies. Those which touched cryptology bent the process in a new . direction. One such was the Murphy Commission. (U) The Murphy Commission was set up.by Congress rather than by the president. Its main purpose was to examine the process by which American foreign policy was set. The chairman, former ambassador Robert D. Murphy (then chairman of Corning Glass), was to report back to Congress.by June 1975. Murphy was looking at foreign policy at a time when Henry Kissinger occupied positions as both secretary of state and national security advisor, and perhaps this was the reason that Murphy concentrated on national security and intelligence issues.' Of the four subcommittees, the one on national security and intelligence, chaired by Murphy himself. dealt with NSA. (U) It was hardly surprising that Murphy should echo the climate of the times. Following Schlesinger (and a host of others before him), he recommended splitting the job of DCI into two people - the political advisor to the president should work.downtown, while the administrator of CIA. who would be his deputy, would manage the agency itself. He advocated giving the DCI further=control over the intelligence budget (meaning, in
, HANBL! VIA T1lu!J'(T KE I lieu

CeMfN'f Ceff'f'ftt)LSYS'ftlMS JeUft'b Y

65

ToP SECRET tJMBRA

'-

DOCID:

523696
TOP SECRET l:JMBRA

REF ID :A52·3696

essence, authority over the Defense component thereof). And he predictably proclaimed that the secretary of state and national security advisor roles should never again reside in the same person. (U) As for NSA, Murphy remarked rather quizzically that NSA was the only national cryptologic agency in the West that reported through the defense rather than the foreign affairs institution. This tended to bias the satisfaction of requirements in favor of military needs. But, having examined the pros and cons of that arrangement, Murphy opted to leave cryptology within Defense. He recommended, however, that the Agency report to an executive committee composed of the DCI and the assistant secretary of defense for intelligence; to broaden its responsiveness. Moreover, he favored changing the rule by which the director be strictly a military officer. The rule, he felt, .should be the same as at CIA - civilian or military did not matter as long as the director and his or her deputy were not both military officers. (U) The key thrust of the Murphy report, however, was in the direction of further centralization of the process. The SCAs should be abolished, and NSA should take on the job of cryptology unhindered and unassisted. This would at once simplify the process and eliminate the bickering that had characterized NSA-SCA relationships since the day NSA was established ,114
(U) The Hermann Study

Withheld from public release Pub. L. 86-36

(U) In the long run, the most influential study was one that was not even completed, let alone published and promulgated. In 1975 Dr. Robert Hermann askedLew Allen for the opportunity to study SIGlNT support to military commanders. Hermann formed a committee ofjust three people: himself,l land William Black. Together, they formulated an elegant and timeless statement of the problem that confronted cryptologic organiza tion. (U) To Hermann, the central dilemma emanated from the abortive establishment of CSS. NSA had been given theoretical control of the complete cryptologic process by which military commanders obtained cryptelogic support, but the enforcement. mechanism had never been implemented.
The most recent NSCID-6 ... provided for very broad NSA responBibilities and authorities well beyond present practices .•. , the 1971 Presidential Memorandum from which the directive was written specifically includes 'tactical intelligence' within the scope of the national level responsibility. However. the Pre,idclllial memorolldum and NSCID-6 are not btinB enforud alld are probably not enforuable. • .. The political forces which generated NSCID-6 did not develop the near term enforcement means necessary to persuade an unwilling management structure. , .• This has been a major cause of stagnation in the development of adequate SlGINT support to military operations aBwell &B inhibiting the general develOpment of SIGINT support for other purposes ..... (Emphasis added] us

(FOUO) Hermann pointed to a, cascade of changes to the SIGINT system which had irreversibly altered the way business was done. He referred to an "electronic explosion" in

HAN'f)LE VIA TAU!N'f Kf'itf6U! ceMINT CeNTRaL S'lSft)MSd6IN'fLY

lOP SECRET UMBRA

66

DOCID: 523696

REF ID:A523696
TOP S!CRET l:JM8RA

the signals environment which tactical commanders were increasingly occupied with and were exploiting to their own advantage. Electronic warfare, electronic support measures, and other terms were. being applied to signals in order to get them out from behind the codewords that restrained their dissemination and exploitation. According to Hermann, "The notion that all 'SIGINT'activity is naturally a part of a coherent SIGINTsystem organized separately to support all national interests and organizations at every echelon is probably unsound. SIGINT is clearly not the most natural primary management dimension for an increasing number of activities." While NSA held to the rigid codeword protection mechanisms that had been built up since 1952, these barriers were becoming increasingly anachronistic. The SCAs, confronted with a two-way tug on their loyalties, increasingly opted for allegiance to their own services. They no longer. hungered to expand the large field site system, no longer viewed their future as lying within a national cryptologic structure. According to the.study, "... the traditional role oftheSCA as the field collection arm of the national SIGINT system is eroding and is even now, not a viable mission." (FOUO) To solve the dilemma, Hermann recommended a revolutionary strategy. The SCAs should cease being cryptologic agencies and should become what he called Service Signal Wanare Agencies (SSWAs). They should be integrated with the commands they supported, and their main job would be to provide signal warfare functions such as ECM, ECCM, tactical SIGINT/electronic support measures, MIJI (meaconing, intrusion, jamming, or interference), and radar surveillance. Except in unusual cases, they would no longer. staff large fixed sites . ..(8J The existing classification system should be completely scrapped. According to Hermann, "... we now provide SI, TK, or EARPOP protection for sources that we no longer hold to be sufficiently sensitive to require these caveats. The reason for protection is , historical not deliberate." Cryptologists had cast aside the fine gradations which had , evolved during World War II to permit wider dissemination of less-sensitive SIGlNTand more restrictive handling of the products of cryptanalysis. In effect, everything was handled at a minimum Category II level, and the advantages of the World War II Y Service system had been lost. He pointed. to the handling of clear text speech intercept (then normally protected as Category II material) as an example of how not to protect information. Other sources, I I were scarcely more sensitive. Signals externals should not be held in COMINT channels unless clear justification was provided. I ~ I.C ~ ~ ~ ~ ~ I.C QO ~ "0 l. .j ~ ..c ..c
0
I

s

.~ ::c..c
Q.~

~

= =

-(STEven more radical was his proposal for the handling ofTK information. According to the study, "There very little justification today for providing SI access without TK. There is no justification for providing TK SIClNTaccess without Byeman access." (The Byeman compartment was created to protect technical and contractual details of overhead systems.) The study proposed that overhead SIGINT should be completely removed from the TK compartment and should be handled as ordinary SIGINT information and that Byeman .should be eliminated except as it related to the relationship with contractors.

is

1S=CC6t Hermann recommended new initiatives for SIGINT support to NATO, long a cryptologic planning backwater. I I

nANB~

VIA 'fAbSft'f 1{l!1YIlOb6 00 MUff OON'fML

S1S'fl!1MSotOIN'FLY

67

'fl)P SEetEf

l:JM8R:A

DOClD

. .

523696
TOP S!CR!'f UMBRA

REF lD:A 523696

E.O. 13526, section 1.4(c)

I

;

Withheld from public release Pub. L. 86-36

Withheld from public release Pub.L. 86-36

II:Mu)LE ';fA 'f1d;EN'f JEEYIIOLE COMlK'f COU'fR01:; SYS'fEMS 49IN'fI:;Y

TOP SECRET UMISItA

68

DOCID: 523696

REF ID:A523696
'FeP SECRET UMBRA

(U) The Ursano Study

(U) Robert Hermann's thinking dovetailed nicely with the direction that the Army was moving. That direction came out in very stark terms in 1975 as a result of the Intelligence Organization and Stationing Study (lOSS). (U) lOSS resulted from a memo from the secretary of the army, Howard Callaway, to Army chief of stair Frederick Weyand in late 1974. Commenting about Army intelligence, Calla way said, "We maintain considerable information which is of questionable value and .seldom used," a fact that "really makes me wonder about how much money we are wasting and raises serious questions as to the cost-effectiveness of our intelligence system." What was on Callaway'S mind was apparently money. The Army was continuing to take monstrous post-Vietnam cuts, and Callaway was looking at intelligence as a place to save money. 117 (U) The man Weyand appointed to study the issue, Major General James J. Ursano, was unencumbered by any experience with; or knowledge of, the intelligence function. At· the time, he was Weyand's director for management. His study group was not very high powered, nor did it contain much expertise in the discipline.l18 It was a completely outsider's look.

(U) Mlijor General James J. Ursano

UANBbB 'IIA '£ALEUT KEYHOLE eOMHff eO?ffROb S"lS'ftlMS ofOlN'ftl"

69

TOP SECRET tsMISRA

DOClD: 523696
TOP SECRET UMBRA

REF lD:A523696

(U) It did not take long for the Ursa no group to find out how fragmented and overlapping Army intelligence really was. Intelligence production was being carried out by a vast welter of rival organizations with competing agendas. The Army expended much I . effort toward HUMINTand comparatively little on SIGlNT, which was found to be isolated and neglected. ASA came under severe criticism. Since the creation of CSS, ASA amounted only to another bureaucratic layer. The elimination of its field headquarters in both the Pacific and Europe gave it an unmanageable span of control. It devoted too much of its, effort to field station operations, too little to tactical support. It had monopolized electronic warfare and held everything under a cloak of secrecy which inhibited real tactical support. In the field, the Army G2 had to manage two separate intelligence systems, SIGINT and everything else, and staff.to integrate the two sides was in short supply. us (U) Ursano looked at the vertical cryptologic command line which had been instituted following World War II and which had been reinforced with every subsequent study of Army intelligence. For once, someone took the opposite tack. Verticality must end, and ASA must rejoin the Army.·2O (U) Ursano'scentral and most important recommendation was to dismantle ASA. A new organization would be created, called INSCOM (Intelligence and Security Command), which would integrate all Army intelligence functions. Combining SIGlNT and HUMINT, Ursano recommended the amalgamation of USAINTA (U.S. Army Intelligence Agency) with strategic SIGINT. INSCOM would continue to manage eight field stations, to supply billets to NSA and other centralized cryptologic activities, and to provide S1GINT support to echelons above corps. Tactical assets (corps and be low) would join the supported command echelon. (8 CeOt INSCOM would be an interesting mix of SIGINT, HUMINT. and counterintelligence organizations. Joining the new command would be the military intelligence groups and to this were added groups in CONUS (CONUS.MI Group) I I TAREX, which had existed as a SIGINT-related effort since the waning days of World War II, would join the intelligence groups. There would be a unified Intelligence and Threat Analysis Center (ITAC) for all-source analysis. But, in sum, the new organization would be considerably smaller than ASA had been, primarily because of the loss of the tactical units. Training functions would be absorbed by other commands, and the training school at Fort Devens would belong to the Army Intelligence Center and School at Fort Huachuca, Arizona.121

I

I

(U) To virtually no one's surprise, Major General George Godding, the incumbent ASA commander, opposed the dissolution of his agency. Godding's reasoning, however, should have sounded bells somewhere in the Army staff. ASA should be retained because of the unique cryptologic expertise which had been developed and nurtured over a period of many years. Ursano's solution ignored -that aspect of the problem.l22 .

IIldffiLB 'RA 'fAUlN'f fCI'lYllOLrJ eOMIN'f eOU'fROL SYS'fI'lMSJOn~'fLY

TOP SECRET tJMBRA

DOCID: 523696

REF ID:A523696
TOP SECRET !:IMBRA

(U) Major General George Godding

(U) The proposals caught NSA seemingly by surprise. When routed for comments, the Ursano proposals elicited little reaction. Each staff element viewed the problem from its own very narrow perspective, and each concluded that the matter was an Army problem, not one which should interest NSA. At the Directorate level, Norman Boardman of the director's policy staff understood the implications: "It is our general feeling that the . I loading of all army intelligence, security, and EW functions onto ASA, with a new name, and the stripping of specialized support functions ... can do nothing but downgrade the quality and timeliness of SIGINT support to the army and army tactical commanders .... " 123 But NSA did not take a hard line, and its response to the Ursanoproposals was less than warlike. And 80 INSCOMofficially came into existence on January 1,1977, without NSA having taken a strong stand one way or the other. (SaCCO) When Vice Admiral Bobby Inman became director in July of 1977, he hit the roof. Noting that the CSS concept assumed central control of cryptologic assets, and that ASA was the organization that was to control the Army's component to that structure, he pointed out acerbically that divestiture of cryptologic assets at corps and below abrogated that agreement and fragmented the system. Moreover, cryptologic training, considered an

IIANBbIS VIA 'fAbEN'i' KEYliObB OOMUrr OQNTReb S¥STEM8~OU>fTL¥

.

71

TOP SECRET !:IMBRA

.L __

DOClD: 523696
lap SECRET UMBRA

REF lD:A523696,

essential aspect of ,maintaining a skilled cryptologic work force, had been removed from INSCOM's authority. TAREX, formerly an exclusive cryp~logic preserve, now appeared to be a SIGINT-HuMINT amalgam. "Throughout the plan SIGINT operational relationships and functions are described that impact directly on NSAlCSS. ,These relationships and functions have not been coordinated with this Agency." 124 In fact, they had been coordinated - hut with Lew Allen, not with Inman. And that train was much too far down the track for one angry admiral to turn it around. (U) The central problem of the INSCOM decision was one of expertise. The Army no longer had a unique cryptologic organization. It had been diluted by other disciplines and other interests. The cryptologic focus was lost and was replaced by a picture gone all dim and mushy. To participate in cryptology, the Army would have had to increase its , emphasis on technical specialization. It chose to go the other direction.
(U) The Creation of ESC

.retIn its own way, the Air Force chose the same path, but at a slower rate. The Air . Force Security Service had begun to lose its SIGINT focus in the late 19605. When the Air Foree Special Communications Center (AFSCC) S[GINT mission was moved to NSA in 1968, the organization survived by acquiring a new role. The mission, straight out of Vietnam, was to do electronic warfare analysis of tactical combat. Such analysis involved a variety of analytic skills, of which SIGINT was the largest component and was thus a natural fo!,'USAFSS. AFSCC could employ all the SIGINT and COMSEC skills of a seasoned work force in a new role of direct concern to Air Force commanders. \
(U) As the command shrank in size during the 1970s, the electronic warfare analysis being done in AFSCC grew proportionately larger. Like ASA, USAFSS slowly eased out of the business of providing manpower to large fixed sites. .Security Service sites which ' survived became smaller, and the command began shedding its management of air bases , Q.j ~ '" M around the world. In 1978.,USAFSS gave away its l~t remaining bases to other Air Force eo= commands: Goodfellow AFB went to Air Training Command.] I IrakIion, ~ (.J.j and Chicksands were turned over to USAFE, and PACAF began managing I I With its intermediate headquarters in Germany and Hawaii closed,the command ended the decade with just under 12,000 people, down from a peak size of over 28,000.125
I
,

E e t!::
~ ~

'* ~

-= .of :E..c
~ 6. d:

...(.Gf General Lew Allen, who had become Air Force chief of staff, was intensely unhappy with the Air Foree approach to, and use of, electronic warfare. His experience as DIRNSA had taught him how SIGINT could affect the modern battlefield. He had an especially keen appreciation for TEABALL, the command and control facility that had operated so effectively in Southeast Asia based on SIGINT support, and he wanted the new organization to create other such mechanisms. So he formed a high-level steering group to look at the problem.P"
(U) In April of 1978 the Air Force announced that it would disestablish Security Service and consolidate intelligence functions within a new intelligence center at.Kelly Air Force Base. This would involve USAFSS, the Foreign Technology Division at Wright-

ItM\'Bhtl 'ItA 'fAbEllff KEY! !6b:e C6MIN'f C6!ffR6b

S'lB't'ElMe 461N'fb Y

TOPSECItI!T tJMBItA

72

DOCID: 523696

REF ID:A523696
. TOP SECRET UMBRA

Patterson Air Force Base in Ohio, AFTAC(which monitored nuclear testing around the world), and Air Force Intelligence Service. The concept was clear, but the details were fuzzy; the affected organizations spent the summer thrashing out the implementation.P" (U) The grand Air Force Intelligence Center study became subsumed under two other high priority Air Force concerns: how to organize electronic warfare and what to do with a growing respcnsibil itycalled C3CM (command, control, and communications countermeasures). All three functions were closely related, and Allen wanted an . organization that combined all three. As it happened, USAFSS had the majority role in intelligence and C3CM and was a major player in electronic warfare. So whatever happened would surely center on the USAFSS complex at Kelly AFB. (U) In January of 1979 a .general officers board recommended to Allen that, not surprisingly, a new electronic warfare command be created, and that it be composed of all three . USAFSS missions. Like ASA,. USAFSS would continue as a major command. Unlike ASA, however, it would not swallow the other intelligence disciplines, at least not yet. USAFSS reopened its doors in August of 1979 under a new name, Electronic Security Command. Its commander, Major General Doyle Larson, was known to be a Lew Alien confidant. When he appointed Larson, Allen told hi I INSCOMr but (U) MlUor General Doyle Larson im not to emu ate to in/ure that all elements of electronic combat were integrated into a single structure .. Together, they were moving the Air Force away from a major role in cryptology, toward a closer tie with Air Force tactical combat. 121

tb\oNf)LE VIA T*LENT

KE'iH6LE

OOMJoNTeON'f'IteL

S'tS'fI!lMS d6IN'fL¥

73

TOl' SECRET \:IMBRA

DOCID: 523696
lOP SECRET UMBRA

REF ID ::A523696

Notes
1. (U) Deputy Director eDDIR), NSA, corrupondenc;e files, NSA retired recorda, 96026, box I, part 2, OVerview

of Soviet Cryptology.
2. 3 (U) NSA,Q_,*rly eU) RobertJ. MOfl46errulll Report (QMR),FY 1980, 2nd Quarter.

Hanyok, -Scaling Down NSA,· NSA New,utur,JanUArY

1995.

4.
5.

(U)QMR,9312,5.
(U) QMR, SOIL (U)CCHSeriea (U) Interview, VII.H.61.2. General (USAF. Ret.) lAw Allen, 17 June 1996. by Tom Jolmaon, OH 7,9·98, NSA.

6.
7. 8. 9. 10. 11. Withheld from public release Pub. L. 86-36

(V) CCH Seri88 XJJ.H.19.
(U) Interview William T. Kvetku, 10117July, 17 Aut: 1996. by TolX'Johnaon .• OH 25:96. NSA ..

(U) CCH SeriesXU.H.19. (U) eCH Series XU.H.lt; NSA An:hiveaacc
111'

27210, CBOK 68.

12·1'--13. (U) NSA retired reeorda, 44699, 84-228. 14. (U)lnterview. NSA.·

_
Colonel Cecil B. Fulford, 23 November 1987. by Robert D. Farley and Tom Johnson, . OH 30·87.

15. (V) NSA retired recorda 44669, 84.228; NsA Archives ace Dr 27263~ CaUB 11. 16. (V) Fullford interview; NSA rewed 17. (U) Ibici. r~rda 44669, 84-228.

r--------------I
19. 20. 21. IV) NSA retired recorda 10017.83.473; (U) Mary Anne Weaver. "B~the
(u) NSA retired

44669, 84-228."--_~·..:.A:.:H=iatorical Martyrs,·
Nelli

Overview,"
r:-::WCC'i t-:-h:-he-:-Id:-r:-ro-m-

YorUr (January

1993).

r8c:0I'ds, 44670,77.397.

public release Pub. L. 86·36

22. (U) Ibid.

~
24. (U) NSA retired recorda, 44760. 74-296; Allen interview;
<,

~r---------~
Williams interview. records, 28515. 8'-245. Security Filet, Gerald R. Ford. Pretidential Library, AnD Arbor, Miebig&Jl, in CCH Series

25. (V) NSA'retired 26. (V) National

r--------------, Withheld from public release Pub. L. 86-36

XVI.H., ·Southeut 2'1. 28.

AsialVietnamICambodia"; CCH

Sene. VI.HH.26.15. ---'
1992, by Charlee Baker and Tom Johoaoll., OH 8·92. NSA.

LI
(U)[nterviewl

'--------,--I

l23 December

TOP SECRET UMBRA

74

DOClD: 523696
Withheld from public release . Pub. L. 86-36

REF lD:A523696
lOP SECRET UMBRA

[=!~------------------------------------------------------------,
31.
(U) CCH Series

29.

(Uc==Jinterview.

vn.Q.1.l5:! Events in the History of Electronic Security Command, 1948·1988.

I-A command history

Chronology. of Significant available 32. 33. 34. (U)c:=Jetal (U) -A History

in CCH Series X.J.

.• "A Chronology."

of the USAFSS Airborne

SIGJN'l'

Reconnaissance

Program

(ASRP).

1950.1977;

20 ..

September

1977. available

in CCH Series X.J; Interview, at Air Intelligence

Gordon W. Sommers. byl Agency. San Antonio. Texas;c=J

Iand Jmes
et al., "A Chronology."

E. Pierson, JanWlry 1990. available
35. 36. 37. (U) NSAArc:hives.acc: (UHbid. (U)Ibid.

nr 33631. H01·0108·3.

~--------------~ Withheld from public release Pub. L. 86-36

38. (U) Ibid.
39. (U) Nolte. William M.• Project Te1U1i6: A Hi6to.., of SatelliU RemDUng.196B.1911 (Fort Meade: NSA,

1977). 40.. (V) Nolte. Proj"t 41. 42.
(U) Nolte,Project

Ten"i6. CCH Series VI.BB.l.6. Ttnnu. Hermann, 2 September 1994. by. Tom Johnson, OH 45·94, NSA.

(U) Interview.

Dr. RobertJ.

43. (V) Nolte. Project Te",u". '4. (U)Ibid ..

45". (U) NSA Archives. aee or 31614. HO·1·0308·5.
~. (U) Nolte.

Project Tennu.

4.,. 48. 49. 50. 51. 52. 53. 54. 55. 56. 57.

(U) NSA Archives, ace Dr 31614. HOl·0308·5. (V) Drawstring Task Force Report, 10 December 1973: in CCH Series VI.BB.1.4

(U) NSA Arcbives ace nr 32545. H01·0101·2. (U)lbid. (U) Drawstring (U)lbid. (Ullbid. (U) NSA retired recorda, 44959. 80-302. (U)(bid. (V) Nolte. Project Ttnni6. (U)Ibid. Task Force RepOrt.

58. (U)lbid.

\

IIkNa~

'ItA TaLENTIt! J HOLECOMIN;r CONtROL sVSl'EMSJOINTLY

75

TOP SE(:RET UMBRA

~
""':

-----

CJ

DOClD: 523696
TO' SECRET UMBRA
SQ.l\O

REF lD:A523696

.9
ell

c

(:j

Q.l

~~~

~

'O~Q()

0 r.;.l
-

I£l t'"l

M

.- = =
~ c. Q.,

:2 = . -,Q,Q

_ ....r---Q.lCJ...J

O"'t'"l

" ?J

59. ~

Nolte,Projeet TtllnU; Int.rviewl

1

I
60. (U~

1 interview.

Withheld from public release Pub. L. 86-36

p:J
63'~1 64. 65.
(U)

I

\

62. (U) Nolte,ProjecC TeJ1Ilis;NSA Archivesace Dr 18141. HOI-0505-5;NSA retired records. 44959. 80·302. The 40 Commi~t.e ruled on all covert int.lligence progralll8 for the White House. ~ Nolte. PrqjecCT8nni..

,-----------,

06. 67. (U) NSA Arehivea. ace IIr 12163. G12·0601.3.

'----------'

Withheld from public release Pub. L. 86-36

Mr.c::::::J
69.

68. (V) Interview~'--

...J119February i997, by Tom Johnson; Files in possesSion of
I .

ML..

...JI interview.

70. (u) NSA retired recorda. 44969.80·302. 71. (U)lbid. 72.1 73.

L..-

Withheld from 74. (U) NSA Archives. public release . P_U_b_._L_._8_6_-3_6---"

(Ue=J

I E.O.
8t al., "A Chronology"; NSA"retired records. 44959. 8O·30Z.

13526, section 1.4(c)

Withheld from public release Pub. L. 86-36

C
77. 80.

ace nr 39074. H02·0l03·5.

1 --------------~

76. (U) NSA retired records. 44969. 80-302; NSA"ArehiVes,3WOZ. HOoI·OI09·6; llIterview.I Tom~ohnaon,19February 1997.0H 4·97.
(u) NSA

1 by

retired records 44959. 80·302. 1996and

78. (u)NSA Archives ace nr 18802,CBnI 77; Interview, RichardL. Bernard, byTomJohnaon,4June 13 January 1997. OM 15·94. NSA. 79. (U)lbid.

(m NSA Archives, ace Dr 4200, CBUO 78; Bernard int.rvi8w.
---11 NSA Archives ace nr 18802, CBTM 71; NSA retired records

Withheld from public release Pub. L. 86-36

81.

(U) 1'44959, 80·302.

82. (U) NSA Archives 22965, HI8-0104·3. 83. (U) NSA Archives ace nr 16514. CBRG 3S. 84. (U~L.. Sprilli 1975, 15-18.
...•• "Guardrail:

I

A Joint TaCtical SIGINT Support System," Crypfl>lollic Spectru.m,

~. (ID~

~------ __-----------------....•

"1l1dfflbETffJ\ 'fld:I!lN"f KEYIf6Le e6MfUT ee!ffReb S'E5'!:'flM6 i6INTL¥

TOP SECRET UMBRA

76

DOClD: 52369~------~

Withheld from public release Pub. L. 86-36

REF lD:A523696
Tep SECRET t;JMBRA

87. 88. 89.

(U) Fact sneet, undated, (U) NSAArchives,ace:

NSA Archives. nr 16524, CBRG36;ace: ~r 2737,CBUC 73.

(U) CCH Series XI.R;NSA Archives ace: nr 16524, CBRG 37; ace: nr 2737, CBUC 73.

90.
91.

nn NSAArcnivesace:
(U) NSA Archives, ace

nr 16512, CBM35;
Dr

ace: nr 16512, CBRG61;acc:nr2897,

CBUC 52.

2737, CBUC 73.

92. (U) NSA Archives, acc nr 167524, CBRG 36; ace: nr 18970, C.BTB 48.
93. (U) RReport of the BlUe Ribbon Defense Panel [the Fitzhugh VI.C.l.31. 94. 95. (U) Ibid.
(u) Thomas

Panel report],"

1 July 1970, ill CCH Series

Powers,

Knopf, 1979),207·08. PresB,1986),59-61.

S. D. Breckenridge,

The Man Who Kept the Seeret.: Ricluzrd Helm. and the CIA (NJw York: T/r.e CIA and the U.s.lraeelligence System (Boulder, CO:

Alfred A. Westview

Withheld from public release Pub. L. 86-36

96. 97. 61.

(u) Breckenridge,

59-61.
1....1

(U) Interview,

Donald M. Showers (&ADM, USN, Ret.), by

JL 5 May

1992; Breckenridge,

98. (U) The Creation 1.5. 99. (U) Ibid.

of Central Security Service - Background

papers and memo files, in CCH Series Vl.QQ.l.1-

100. (U) Ibid. 101. (U) Ibid. 102. (U) Interview, 87,NSA. John R. Harney, by Robert D. Farley and TOlD Johnson, 17 December 1987, Oral History 32-

103. (U) The Creation ofCeiltral Security Service •.•.
104. (U) Interview, USAFSS interview Carl W. Stapleton available (Maj Gen, USAF, Ret.), by Lt Col William H. Buchholtz, unit date.in 1986,

at AlA, Kelly AFB, San Antonio. interview.

105_ (U) The Craation 106. (U) The Creation

of Central Security Service .... ; Stapleton of Central Security Service ...•

107. (U) Ibid.

Withheld from. public release Pub. L. 86-36

108. (U) CCH Series VIII.H.<I8; The Creation of Central Security Service •••• ; Interview,

GordO.riW. Sommers,

by

I

land James E. Pierson, Jsnuary

1990, available

at AlA, Kelly AFB, Texaa.

109. (U) DlRNSA msg 10 Nov 1972 ill CCH Series XII.H.48. 110. (U) CCH Series XII.H.48. 111. (U)[bid. 112. (U) Har~ey interview; CCH Series XII.H.48; Stapleton interview. .

H,".NI>Y 'RA T.\taENT KEYI[9b!ij C9MINT C9N'FR9L SY-5T5MSd9INTLY '77

TOP SECRET t;JMBRA

DOClD: 523696
TO' SECRET UMBRA

REF lD:A523696

113. {U}The DlRNSA. VADM Noel Gayler. was also hoping for a fourth star, and was loathe to jeopardiz~ advancement by seriously tangling with his potential benefactors. according to Major General Carl Stapleton; See Stapleton interview. 114. (U) Murphy Commission Report, Committee II -Intelligence. in CCH Series V1.C.1.23; Ray S. Cline. TIu!
CTA U1Id6r ReagAn, BlU4 and CQ.8ey (Washington. D.C.: Acropolis Books,1981). 261·62.

11S. (U) "SIGINT Support to Military Operations" [the Hermann Study], 28 April 1975. in NSA records center
28792. SO·O'l9:

116. (U) Hermann study; Interview. RobertJ. Hermann.by Tom Johnson and Charles Baker. 2 September 1994. OH 45-94. NSA; Tencap information is from Angelo. CodevilIa, lrzforming Sta~craft: In~lligcnce for a New CentlU)' (New York: The Free Press. 1992}.118·19. 117. (u) CCH Series XII.H.S7.2; Callaway quote is from draft .chapter 10 of a forthcoming history of Army intelligence. a joint Center for Military History-INSCOM project. 118. (U) CCH Series XII.H.57.2. 119. (U) Jack Finnegan. "lOSS And After" unpublished manuscript in HQ INSCOM files. 1987. 120. (U) Orai'\.Army hiatory. 121. (U) Finnegan. "IOSSAnd After." 122. (U) CCH Series XIl.H.57.2 123. (U) Ibid. 124. (U>Ibid. 125. CU) "History of the Electronic Security Command, I January - 31 December 1979." available at AlA. KeUy AFB. Texas. Holu.b,et al., "A Chronology of Significant Events .... " 126. (U) "History ofthe Electronic Security Command, i January·31 December 197.9." 127. (U) Sommers interview.. "History of the Electronic Security Command, 1 January-31 December 1979." Interview. Doyle Larson (Maj Gen, USAF. Ret). by TomJobnson. OH IS·97,NSA.

/

TOP SECRET U~BRA

78

-----

-.~

-

-

----

DOCID: 523696

REF ID:A523696
lOP SECRET UMBRA

(U)Chapter 16 Cryptology and the Watergate Era
; >

(U) BACKGROUND

TO SCANDAL

(U) The greatest political scandal in American history originated with an obscure note in the Metro section of the Washington Post on Sunday, June 18, 1972. In it, two Metro> section reporters, Bob Woodward and Carl Bernstein, covered what appeared to be an amateurish break-in at the Democratic National Headquarters in the Watergate Hotel in downtown Washington. (U) The Nixon administration managed to cover over the political effects of the break-in until after the elections in November. But when Congress returned in January, it was ready to investigate. In February 1973, the Senate voted to establish a Select Committee, commonly referred to as the Ervin Committee after Senator Sam Ervin, Democratic senator from North Carolina, to hold hearings. At the time, no one associated with the committee knew where they would get > information, since the administration was keeping a tight lip, and> the > Watergate burglarsweren't-talking. But on March 23, one of the burglars, James McCord, turned state's evidence; The federal judg~, John Siriea, had been pressuring the defendants by threatening lengthy prison terms if they did not cooperate. Now McCord was cooperating, and the entire thing began to unravel. The president, concerned with getting on with his second term, tried to shush the whole thing.

(U)

President NiKonand his inner circie, 1973>

(U) The scandal, of course, would not shush. Instead, it mushroomed, swallowing first Nixon's White House staff, then much of his cabinet, and finally the president himself. On August 8,1974, Nixon resigned and Gerald Ford moved into the White House. >(U) In a real sense, Watergate resulted from Vietnam. President Nixon was obsessed with the disorder and demonstrations that hurled the Johnson administration down and

!fAN fit!: 'Ii fA TALENT KE i MOt!: eOMmr eo14T1t6L S't'S'fEMSJ6lf(\'i;y

79

lOP SrCREllJMBRA

DOClD: 523696
TOP SECRET tJM11tA

REF lD:A523696
CRYPTOLOGIC QUARTERLY

played a large role in the defeat of Hubert Humphrey in 1968. One of the central incidents ofthe disorderly 19605 was Daniel Ellsberg's decision to publish a collection of the Johnson administration's papers on the war, which came to be known as the i'Pentagon Papers. Nixon ordered an investigation of Ellsberg, and two of his White House confidants, Egil "Bud" Krogh and David Young, put together a clandestine unit, which they called the "Plumbers" because the objective was to plugleaks. The group obtained the assistance of White House Special Counsel Charles Colson, who brought in some experts in clandestine surveillance formerly from CIA and FBI, among them Howard Hunt and G. Gordon Liddy. The Plumbersbroke into the office of ElIsberg's psychiatrist, Lewis Fielding. The unit itself was eventually disbanded, but the individuals were retained by the Committee to Re-Elect the President (CREEP), and they eventually bugged the office of Lawrence O'Brien, chairman of the Democratic National Committee, in the Waterga te complex. 1 ~For a time, cryptology was a bystander in this turmoil, but the antiwar demonstrations eventually touched NSA's business. In 1966, Stanford University students picketed Stanford Electronic Laboratories, where Lockheed Missile and Space Corporation (LMSC) was designing the P-ll SIGINT satellite payloads. When students occupied the building, James DeBroekert of LMSC smuggled one of the payloads out of the building, through Moffett Naval Air Station and over to Building 190 where the rest of the Lockheed SIGINT satellite effort resided. This very close call for the cryptologic payload had a happy ending only because the students never really knew what they were . picketing. 2 -reJNext year disorder hit the Princeton University campus. The radical group Students for a Democratic Society (SDS) discovered the existence en campus of the Communications Research Division of the Institutes for Defense Analyses (IDA/CROl, which had been set up in the late 1950s to help NSA with difficult cryptanalytic problems. Unclassified CRD publications appeared to link the organization with the Defense Department, andSDS set out to force a campus eviction. After several months of sporadic demonstrations, on May 4, 1970, students broke through police lines and vandalized the inside of the building. A few days later a student was arrested as he attempted to set the .building on fire. CRD built an eight-foot-high fence around the building and occupied it in a permanent siege mode. But the students had already achieved their objective. The atmosphere was no longer good for defense eontractors,.and Princeton asked CRD to move. CRD found other quarters offcampus and moved out in 1975.3 (U) In June 1971, amid the hysteria over the American invasion of Cambodia, the New York Times began publishing a series of documents relating to the war effort .: The papers had originally been given to journalist Neil Sheehan of the Times by one Daniel Ellsberg, a former defense analyst during the Johnson administration. Two days later a federal judge issued a restraining order, but that did not stop the presses. Ellsberg sent copies to seventeen more newspapers, and the revelations continued. On June 30, the court lifted its restraining order, and the Times published the rest of the batch. Journalists quickly labeled them the Pentagon Papers.

HANDLE VIA TALENT KEYHOLE COMINT CONTROL SYSTEMS JOINTLY

TOP SECRET UMBRA

80

DOCID: 523696

REF ID:A523696
TOP SECRET tJMBItA

(U) Ells~erg had been hired into the Pentagon as one of Robert McNamara's "whiz kids." In 1967 Ellsberg was assigned to a project under Lawrence Gelb to undertake a study of U.S. involvement in Vietnam. Brilliant and dogmatic, Ellsberg turned against the war. He felt that the documents could be damaging to the war effort, so when he left the Pentagon to take a job with the Rand Corporation, he reproduced a copy and carried it with him. (U) It was a very large document indeed - over 7,000 pages - and Ellsberg spent thousands of dollars making copies. For several years he tried to use the papers to convince policy makers (Henry .Kissinger and William Fullbright, among others) to change U.S. policy in Southeast Asia, but in vain. As a last resort, then, in 1971 he turned the documents over to the newspapers.4

(U) Daniel Ells berg

(U) Ellsberg claimed that the Pentagon Papers; although officially classified, were actually unclassified. In fact, the last four (of forty-seven) volumes contained COMnirr relating to, diplomatic negotiations with North Vietnam, and it was this information that the government was trying to protect when it applied for a restraining order. Newspapers' did not release the information in 1971, but journalist Jack Anderson got-the last four volumes and released them in 1972. Among the revelations was one concerning the intercept and exploitation of Soviet premier Kosygin's telephone calls while he was in London in February 1967. The intercept apparently came from the British, so from a technical point of view this incident revealed no American cryptologic informaticn," ..(S ceer NSA examined the four volumes and found five instances in which COMINT was undoubtedly the source of the information. Ambiguity prevailed in each case, and NSA's policy people bent over backwards to avoid having to charge Ellsworth or Anderson with violation of Section 798 of Title 18. But the director was concerned enough that he sent an emissary, Milton Zaslow (then deputy director for production), on a secret mission to try to convince the New York Times not to publish on the basis of national security. The Times editors viewed NSA as a stalking horse for the Nixon administration and published anyway. "You could," Zaslow said later, "cut the suspicion with a knife." 8 (U) The Pentagon Papers and subsequent Anderson columns began a trend. The trend was to tell all. It started small, but became a tidal-wave of revelations. That same year, for instance, Anderson revealed that NsA was reading the communications of the South Vietnamese embassy in Washington, through the" ingenious device of providing the·

IMllDLB VIA 'fl'tbSN''f KSYHOLECOMfN'f COti'fROL S'lS'fEMS iOIN'f'LY .

81

Tep SECRET l:IMBRA

OOClO: 523696
feP 5EeRE'f UMIItA

REF lO:A523696
CRYPTOLOGIC QUARTERLY

ciphers which controlled the Vietnamese equipment. Soon after, the Manchester GU4rdian. published an article about CIA COMINT operations in Laos." Then in the fall of 1971, in one of his more sensational columns, Anderson stated that the United States had an intercept operation in the American embassy in Moscow thatndt only intercepted Soviet communications, but was collecting and exploiting the private ear phone communications of Politburo leaders, a (U) Anderson, NSA later discovered, had acquired a boxof top secret CIA National Intelligence Digests (NIDs), the unwitting courtesy of an NSC staffer who had been in the habit of taking them home for a little bedtime reading; After a marital falling out, his wife took the accumulated NIDs to Anderson, who kept them in his office and used them in his columns over a period of years. 9

E.O. 13526, section 1.4(c)(d)

Withheld from public release Pub. L. 86-36

Withheld from public release Pub. L. 86-36

The previous insider-tells-all account, Herbert Yardley's The A~erican Black Chamber, had been written in a fit of greed (Yardley needed money). People like Fellwock could apparently be bought by id~logy. It echoed the climate of the 19305, when the Soviets got their spies for free (or at the very least, for expense money). (U) Ideology-based public revelations became fashionable with the publication in 1975 of ex-CIA agent Phillip Agee's Inside the Company - A CIA Diary. Although Agee's aim was CIA's covert operations organization, he knew much about SIGINT, and he revealed what he' knew. He claimed, for instance, that NSA had used close-in techniques to intercept plain text from the UAR embassy in Montevideo, Uruguay. He also claimed that Swiss-built Hagelin machines had vulnerabilities which NSA exploited to obtain plain text. 11

IfAN5LE •••• IA: ~UlN'f

KE'fH6LE

C6MfN''f' 60N'fHOL SYSTEMBJOINT.bY

lOP SEERET l:JMBRA

82

DOCID: 523696

REF ID:A523696
lOP SECRiT"MIRA

(U) Using the indefatigable Fellwock as a key source, the Canadian Broadcasting . Corporation did a 1974.series entitled '''rhe Fifth Estate - the Espionage Establishment," which made a wide-ranging exposure of Intelligence organizations in the United States. and Canada. This series laid out in sharp detail the overall cryptologic cooperative system encompassed within the UKUSA agreements. It was followed up by tag-on magazine articles, including several by British journalist Chapman Pincher regarding SIGINT at GCHQ.:Journalists exposed the role ofthe British intercept site.in Cyprus during the coup in 1974, and GCHQ's efforts to keep the station running during the fighting. That same year a Marvin Kalb biography of Henry Kissinger discussed NSA's exploitation of . Egyptian communications during the Yom Kippur War the previous year. I!

(U) NSA AND CLANDESTINE ACTIVITIES (U) Over the years, cryptologists had participated in two activities whose legality was eventually called into question. One, codenamedShamrock, was a way to intercept messages without setting up intercept sites. The other, Minaret, became enmeshed with an illegal use of information fordomestic law enforcement.
I

(U) Shamrock

(U) The easiest way to get access to telegrams was to get them from the cable companies which transmitted them. This method actually dated back to World War I, when the federal government, using the implied war powers of the president, set up cable and postal censorship offices. A copy of every cable arriving and departing from the United States was routinely sent to MI-8, which thus had a steady flow of traffic to analyze. After the war, the Army closed all intercept stations. Yardley's Black Chamber continued to use messages provided by the obliging cable companies unti11927, when the Radio Act of 1927 appeared to make this illegal, and the Communications Act of 1934 reinforced this. Lack of traffic forced Friedman's SIS to set up intercept stations in the 1930s.13 (U) In 1938, the Army's chief signal officer, General Joseph Mauborgne, approached David Sarnofl', president of RCA, with a request from the secretary of war to renew the arrangement whereby the Army received drop copies of cable traffic. Sarnoff was willing, and during the war the major cable companies (RCA, AT&T, and Western Union) once again provided cables to the cryptologists. Signal Intelligence Service set up Radio \ Intelligence Companies to collect cables through censors installed at the cable company offices. Following the surrender of Japan, military officials approached the companies to request their continued cooperation, as they had after World War I. . This time, however, they met considerable resistance. Cable company officials argued that the Federal Communications Act of 1934 appeared to make this illegal In peacetime. They wanted legislation.

HANDLE y h\ YALE Iff ItE'f116LE eeMUi'i' e9tiTRelo. S¥sti:MIii JQll>l"'tI Y

83

TOP SECRET !IMBRA

DOCID: 523696
(

REF ID:A523696
CRYPTOLOGIC QUARTERLY

TQP SECRET UMBRA
.

(U) What they got was a promise from the attorney general. Tom Clark, that they would be protected from lawsuits while the Justice Department sought authorizing legislation. (Opinions differ as to whether or not President Truman put this in writing.) But the legislation was not forthcoming, and in 1947 the company executives contacted Secretary of Defense James Forrestal, who had to renew Tom Clark's assurance that they would not be prosecuted, and that the operations would not be exposed. Two years later, sti1llacking legislation, they approached the new secretary of defense, Louis Johnson. He advised them again that Clark and Truman had been consulted, and had 9nce again approved the practice. Somewhat mollified, they finally dropped the subject." (U) At NSA the cable drop operation was treated as a compartmented matter, andonly a few employees knew where the traffic came from. Couriers carried cabled messages to NSA, but there was no direct contact with the cable companies themselves. NSA selected about 150,000 cables p~r month for further analysis - the rest were destroyed. Although not technically illegal;Lew Allen, who was director in the mid-1970s, said it did not pass the "smell test" very well. Stopping it was not a difficult decision for him."
(U) Minaret

.

(U) There is no stark line between "foreign intelligence" and domestic law enforcement. The phrases, which appear to be watertight, actually leak into each other at many points. But this never became an issue until the Watergate period: (U) In the collection of foreign intelligence, cryptologists often came across unrelated communications, which were routinely destroyed because of their irrelevance. But when items of importance to the FBI came available, they were normally passed on. This was done without much thought given to the boundaries between foreign intelligence and law enforcement, which were by law tobe kept separate. The practice began in the 1930s and· continued through the war years and into the 1950s.18 (U) In 1962, following the Cuban Missile Crisis, the Wl,ite House wanted to know who was traveling to Cuba (which had been made illegal but for exceptional cases). This involved passing on American names and violated customary SIGINT rules by which information on American citizens was to be ignored. It was clearly related to law enforeement, however, and it was the origins of the so-called "Watch List" which became known as the Minaret program. 17 . ..(SCCotThe idea proved to be irresistible. In 1965, as a result of the conclusions of the Warren Commission, the Secret Service asked NSA to be on the lookout for certain people who might be a threat to the president. The (11'stlist was composed almost entirely of Americans, but NSA complied because of the obvious implications of not providing such important information. In 1973 the Agency asked that the Americans bereinoved from .the list and hung onto that position despite anguished protests from the Secret Service." .. (U) The Watch List expanded in the 1960s to include people suspected of narcotics .trafficking. and at one point most of the names on the list were individuals suspected of

.

HIdJDLEl ViA 'fAf:;EHft' IEEYUOI:J5 eOMIN'P CONTROl. SYSTEMS JGINn,y

.TOP SECRET UMBIbtt

84

DOClD: 523696

REF lD:A523696
TOP SECRET UMBRA

narcotics-related activity. 'I'he list was formally documented by USIB in 1971.l~ But by far" the most controversial expansion of the list occurred in 1967, and it involved domestic terrorism . ..{S coer In 1967 the country appeared to be going up in flames. Vietnam War protests were becoming common, and "ghetto riots'; in America's urban centers had virtually destroyed sections of Detroit and Los Angeles. President Johnson wanted to know if the domestic antiwar movement was receiving help from abroad, and he commissioned Richard Helms at CIA to find out. CIA came up with very little, but in the process of mobilizing the intelligence 'community, the Army was tasked with monitoring communications for the purpose of answering Johnson's question. On October 20, Major , General William P . Yarborough, the Army chief of staff for intelligence, informed NSA of the effort, in which ASA was involved, and asked for help.2O (8 CeO} With FBI as the prime source of names, NSA began expanding the watch list to include domestic terrorist and foreign radical suspects. The watch list eventually contained over 1,600 names and included such personages as columnist Art Buchwald, journalist Tom Wicker, civil rights leaders Martin Luther King and Whitrley Young, the boxer Muhammed Ali, and even politicians such as Frank Church and Howard Baker. Virtually all the names were provided by other government organizations. However, NSA did add thirteen names, all but two of them Agency employees who were acknowledged spies, such as Martin and Mitchell. One of them was the aforementioned Percy Fellwock.21 ..(S cee; The project, which became known officially as Minaret in 1969, employed unusual procedures. NSA distributed reports without the usual serialization. They were designed to look like HUMINT reports rather than SIGINT, and readers could find no originating agency. Years later the NSA lawyer who first looked at the procedural aspects .stated that the people involved seemed to understand that the operation was disreputable' if not outright illegal. 22 .' (U) ASA's monitoring of domestic radical communications was almost certainly illegal,according to the legal opinions of two different groups of government lawyers. Even worse, it had come to public notice in 1970 when NBC aired a program alleging that ASA had monitored civilian radios during the Democratic Convention of 1968. ASA quickly closed it down and went out of the civil disturbance monitoring business.f'

(s ceo)- Minaret was quite another matter, and it did not depend on ASA for its existence. Lew Allen had been director for less than two weeks 'when his chieflawyer, Roy Banner, informed him of Minaret - it was the first the new director had known of the program. Banner noted a recent court decision on wiretaps that might atfect the Watch List. A federal judge had ruled in a case involving leading Weathermen (SDS radical wing) that all federal agencies, including NSA, must disclose any illegal wiretaps of the defendants. NSA's communications monitoring, although not technically a wiretap, could be construed as such by recent court decisions. Although the Weathermen in question might not be on the Watch List, the time was not far off when a court case would expose the list.

II A NPU VIA 'tA!.ENT KEYHgbE OgMll'lT OmlTlWL 6¥S!FEM6 clOmn. y

85

lOP SECRET UMBRA

DOClD: 523696
'FOP SECAETl:JMIM

.REF ID:A523696
. CRYPTOLOGIC QUARTERLY

(5-000) This operation did not pass the "smell test" either.
appeared Attorney retention to be a possible violation of constitutional guarantees. General Elliot Richardson to request that Richardson ofall individuals by name on the list. u' .

According to Allen, it He pro~pt1y wrote to himself authorize the

(U) This was in September 1973. The Watergate hearings in Congress had just wrapped up, and the special prosecutor, Archibald Cox, had subpoenaed the presidential tapes. The executive department was in chaos. Richardson's predecessor, Richard Kleindeinst, had been foreed out under pressure, and his predecessor, John Mitchell, was a'lmost sure to go to jail. In that atmosphere, the attorney general was not going to permit the continuation of an operation of such doubtful legality. He requested that NSA stop the operation until he had had a chance to review it. With that, Minaret came to a well- . deserved end. =s (U) Oandest/ne Methods
(U) If you can't break a code, the time-honored method is to steal it. Two of NSA's most cherished secrets, the black bag job and the wiretap, became public knowledge during the Watergate period.

(U) Black bag jobs referred to the art of breaking, entering, and theft of codes and cipher equipment. The OfTlce of Naval Intelligence (ONn, an unlikely leader in the field, became the first practitioner. In 1922 ONI picked the lock of the safe in the Japanese consulate in New York and filched a Japanese naval code. This theft led to the establishment of the fU'st permanent American naval cryptologic effort, OP-20-G, in 1924.26 ofthe art. Prior to World War II the Navy pilfered a diplomatic code .which was used at embassies which lacked a Purple machine. Joseph Mauborgne, the head of the Army Signal Corps, hit the overhead when he found out. Mauborgne reasoned that if the Japanese ever discovered the loss, they might change all their systems, including Purple, and extracted from the Navy an agreement that all such break-ins in the future would be coordinated with the Signal COrps.27

(U) ONI continued to be the main practitioner

I E.O.

13526, section 1.4(c)

Withheld from public release Pub. L. 86-36

l'()P SECRET l:JMBRA

86

DOClD: 523696

REF lD:A523696
"F9P SECRET UMBRA

I E.O.

13526, section 1.4(c)

I

Withheld from public release Pub. L. 86-36

(U)~. Edgar Hoover (U)The Huston Plan

(U) Richard Nixon had been president just over a year when he initiated a string of. actions which ultimately brought down his presidency. The White House-ordered invasion of Cambodia, a militarily ineffective foray, unleashed a wave of domestic protests, culminating in the shootings at Kent State in May of 1970.. Stung by the reaction, the president called the heads of the intelligence agencies, and on June 5 he told Richard Helms of CIA, J. Edgar Hoover of the FBI. Lieutenant General Donald Bennett of DIA, and Admiral Noel Gayler of NSA that he wanted to know what steps they and their agencies could take to get a better handle on domestic radicalism. According to journalist Theodore White, who later reconstructed the meeting:
He was dissatisfied with them all ... they were oveZ'8taffed,they weren't. getting the story, they were spending too much money, there was no production, they had to get together. In sum, he waated a thorough coordination ofall American intelligence agencies; he wanted to know what. . the links were between foreign groups
7"

al·Patah; the Arab terrorists; the Algerian subsidy

center - and domestic street turbu1eace. They would form a committee. J. Edgar Hoover would be the chairman, Tom Huston ofthe White House would be the sta1fman.
31

(U) Thomas Charles Huston, the evident object of the president's displeasure, was a young right-Wing lawyer who had been hired as an assistant to White House speech writer Patrick Buchanan. His only qualifications were political - he had been president of the Young Americans for Freedom, a conservative campus organization nationwide. And Huston wasn't even the key player. Hoover was named chair of the committee, in order to place him in a position in which the FBI would finally be forced to cpnfront domestic radicaHsm.S2 . .

87

lOP SECRE+U""8RA

DOClD: 523696
"FGP SECRET UMBRA

REF lD:A523696
CRYPTOLOGIC QUARTERLY

(U) The committee report confronted the issue, all right, and it laid out a number of "further steps," many of which were illegal. The report recommended increasing .wiretapping and microphone surveillance of radicals; relaxing restrictions on mail covers and mail. intercepts; carrying out selective break-ins against domestic radicals and organizations; lifting age restrictions on FBI campus informants; and broadening NSNs intercepts of the international communications of American citizens. But Hoover knew the score, and he attached footnotes to each of the techniques which he did not want the. FBI involved in. When it went to the president, it was carefully qualified by the FBI, the one organizations that would be the most involved." (U) The president sent word back to Huston, through Haldeman, of-his approval, but did not initiate any paperwork. So when the committee was tasked to implement the . recommendations,· it was tasked by Tom Charles Huston, not the president. Hoover informed John Mitchell, the attorney general, that he would not participate without a written order from Mitchell. Mitchell discussed this with Nixon, and both agreed that it would be too dangerous. Ultimately, the president voided the plan, but not before NSA had become directly involved in the seamier side of life.34 (8 eeo~ NSA was ambivalent. On the one hand, Gayler and his committee representative, Benson Bufiham, viewed it as a way to get Hoover to relax his damaging restrictions on break-ins and wiretaps. Gayler had personally pleaded with Hoover, to no avail; now the committee mechanism might force the stubborn director into a comer. But that was a legal matter for the FBI to sort out. When asked about intercepting the communications of Americans involved in domestic radicalism, Gayler and Buffham became more pensive. Th~y informed the committee that "NSA,currently interprets its jurisdictional mandate as precluding the production and dissemination of intelligence from communications between U.S. citizens, and as precluding specific targeting against communications of U.S. nationals;" Of course American names occasionally appeared in intercepted traffic, but use of even this incidental intercept needed to be regularized by a change to NSCIO 6.M As with the FBI, NSA wanted a legal leg to stand on. .JS.G€01 What stand did NSA take? Gayler genuinely wanted to be helpful, especially when the president so insisted on getting help; In meetings he seemed ready to turn NSA's legendary collection capability to the services of the Huston mandate. But his lawyers advised caution, and, according to Huston himself, NSA was more nervous than any of the other intelligence agencies. Gayler clearly wanted a legal mandate."
(U) The

White House Tapes

General Lew Allen, General Phillips's successor , came to the job with a strong admonition from his boss, Secretary of Defense James Schlesinger: stay as far away from Watergate as possible. He was aghast, then, when he learned on a Friday in January 1974 that a virtual army oflawyers was on its way to Fort Meade with the White House tapes. Howard Rosenblum, the director of research and engineering, had made it known that NSA might be able to analyze the infamous White House tapes which had been

(s-eeo)

RAN DLE V IA lAtENT It! I HOLE eOM1MT eONTlt6L SYSTEMS ofOlN'I'L'l
TOP SECRET UM81tA

88

DOClD: 523696

REFlD:A523696
T9P SECRET UMBRA

subpoenaed by the special prosecutor. They all arrived in staff cars on a Friday with boxes of tapes. NSA's experts went through the tapes for hours, then gave them back to the lawyers. They had found an eighteen-minute gap on one of the tapes. It appeared to be a deliberate erasure, as the tape had been gone over multiple times in a manner that did not support the president's contention that the erasure had been accidental."

(U) THE ALLEN ERA AT NSA· (U) Occasionally a person's impact on events demands that the period be named after him or her. General Lew Allen was such a man. But the "Allen Era" did not actually begin with Allen. (U) In July 1972 Noel Gayler departed the Agency. He got a fourth star and became CINCPAC. Gayler, an upwardly mobile officer with high ambitions, was the first director to move up. NSA had always been a dead end, where mavericks could end their careers at an agency where mavericks were appreciated, even required. He was not to be the last rather, Noel Gayler was the first offour officers in succession who gained their fourth star and moved on. The second was his successor, Air Force li~utenant general Sam Phillips . .,rerPhillips came from a highly technical background. A fighter pilot in World War n, .he came to NSA from the Apollo program, where he had been the director. The visibility of the program, and the accolades that had been heaped on his management of it, indicated that he was destined for bigger things. According to one source, .he knew before he arri ved that he would stay only one year, and would move on to command the Air Force Systems Command as a four-star. general. However, his successor, Lew Allen, believed that Phillips became aware of NSA's vulnerability to the Watergate mess once he was ensconced and that this influenced his determination to move on.3I (U) Lew Allen came from the same sort of background; but more so. He had a doctorate in nuclear physics,had worked at Lawrence Livermore Laboratories; worked in the satellite collection business for the Air Force, and when nominated to be DlRNSA, was de facto director of the Intelligence Community (IC)Staff. (U) He had become a prot~ge of James Schlesinger, who had brought him onto the Ie Staff. But owing to a temporary feud between Schlesinger and Congress over whether the job should be civilianor military, Allen had not been confirmed. So when Schlesinger became secretary of defense, he asked Allen to become DIRNSA, a position that did not require congressional confirmation." (U) Lew Allen was easy to·like. His quick mind was covered over by a kindly demeanor and a slowness to anger. Even Stansfield Turner, who feuded endlessly with Allen's successor, Bobby Inman, wrote that Allen "particularly impressed me with a ·firm statement that the NSA took its direction on what information to collect from the Director of Central Intelligence. All I needed, he said, was to tell him what I wanted." 40 .

fJA NDU: trIA TAbE~.fT KEYI EaL! GaMIN! CONTROL SYSTEMS JOINTLY

89

TOP SECRET YMBRA

DOClD: 523696
Tep SECRET I;IMBRA

REF lD:A523696
CRYPTOLOGIC QUARTERLY

(U) Lieutenant General Sam Pbillips

(U) Lieutenant General Lew Allen

E!F5 eeO-'FK) Lew Allen once described candidly the baggage that he brought with him to NSA. Schlesinger was convinced that NSA was too large and too expensive, and he told Allen to look into the charge. (He found it to be unsubstantiated.) He had always been impressed with the technical competence resident at NSA, but he felt that "NSA, like many large bureaucracies, had a lot of turf .... " Having come from the NRO side of the satellite business. he knew firsthand of NSA's desire to control SlGINT satellites and ground stations, and he felt that NSA harbored "ambitions for responsibilities that somewhat exceeded the grasp." He had heard that NSA had enormous warehouses of undecipherable tapes. (This too he found to be exaggerated.I"

-«»--His foeua on the technical side of life was perfect for NSA, a technical agency. Allen had no patience with bureaucratic turf battles, and he did not think that constant reorganizations were a good use of time. But he did .bring over from the Air Force a penchant for systems design, and for that, one needed a designer. So one of his first acts was to appoint an architectural planning staff to design the various components of the cryptologie system. He had an architect for everything: covert collection, Third Party, overhead, support to military operations, high-frequency systems, line-of-sight systems, signals search, and so on. One of Lew Allen's most important legacies was ,to institute a planning mentality where one had not existed.· . (FOUO) In 1977. in the last year ofliis tenure, he confronted a congressional proposal to pull NSA out of the Defense Department. To a man as firmly grounded in the military

UANBb8 Vb\ 'ffd:J!iN'f I{EYlJeLE eSMIN'f eeN'l'ReL SYS'fEM811eUfflll'

TOP SECRET I:IMBIbtt

90

DOCID: 523696

REF ID:A523696
TOP S!eRET UMBM

as Allen, this was a nonstarter. Pointing out that 75 to 80 percent of NSA's material supported the military, he came down firmly on the side of staying in the Defense Department. As to the concurrent proposal to civilianize the director's job, the continued credibility with military commanders was too important a qualification to lose.42

(U) THE CHURCH COMMITTEE (U) When John Dean, the president's legal counsel, began unburdening himself to the Ervin Committee in the spring of 1973, the testimony implicated the 'CIA in aspects of the Watergate scandal. So William Colby, the deputy for operations, decided to do a survey. 43 (U) The "Family Jewels" was a 693· .page report of possibly illegal CIA activities through the years. Colby, who had become DCI by the time the report was finished, informed the four chairmen of the House and Senate committees which had oversight of the CIA and succeeded in convincing all of them that the matter was over with-and that CIA would clean up its own house. But by then so many people within the CIA knew about the report that its eventual exposure became almost inevitable.

(U) WiUiam Colby

(U) On December 22, 1974, journalist seymour Hersh published a story in the New York Times based on the "Family Jewels," charging that the CIA had been involved in Chaos, an operation to monitor domestic radical groups during the Nixon administration." The next day, President Ford detailed Henry Kissinger to look into Hersh's allegations. (Although informing Congress, Colby had never told the White House apout the report.) Colby confirmed the general outlines of the story to Kissinger, and the president knew that he would have to investigate." So on January 4, Ford appointed a President's Commission on CIA Activities within the United States. It was headed by Vice President Rockefeller, and the press promptly dubbed it the Rockefeller Commission." (U) While the commission was deliberating, the president himself revealed. on January 16, that some of the allegations of wrongdoing included plots to assassinate foreign heads of state. As if enough controversy did not already surround the commission, this new charge served to scuttle its effectiveness. In the end it issued a very reasonable and workmanlike report which recommended certain structural reforms to guard against

fl:hNfll:d!l VIA TALENT !Hl't'H6l:iE C6MfN'f e6N1'ft6t 5YSTEMSJ61NTLl

91

TOP SIiC;RET "'MIRA

Doem:

523696
TOP SECRET l:IMBRA

REF ID:A523696
CRYPTOLOGIC QUARTERLY

(U) Nelson Rockefeller

future transgressions, and it set forth specific prohibitions of certain activities like illegal wiretaps and participation in domestic intelligence operations. (It declined to rule on assassinations, pleading lack oftime to get to the bottom of these allegations.) But by then no one was Iistening. 47 (U) Senators were clamoring for an investigation, and on January 27 the Senate established the Senate Select Committee on Intelligence. Philip Hart of Michigan was originally approached to chair the committee, but he was gravely ill with cancer, and so the job was offered to Frank Church of Idaho. Unlike Hart, Church harbored presidential ambitions, and some feared that he would use the committee as a pulpit to advance his ambitions. Like the Rockefeller Commission before it, this investigative body came to be known after its chair and has gone down in history as the Church Committee. (U) Some, like Church himself, were suspicious of the intelligence community and sought to expose as much as possible. Into this camp fell Democrats Gary Hart of Colorado and Walter Mondaleof Minnesota, along with Republicans Charles McMathias of Maryland and Richard Schweicker of Pennsylvania. Many were moderates (Warren Huddleston of Kentucky .and Howard Baker of Tennessee being examples) while two senators, Barry Goldwater of Arizona and John Tower of Texas, did not believe in exposing intelligence secrets no matter what the provccation." . (8 eeO) To begin with, NSA was net even on the target list. But in the course of preliminary investigation, two Senate staffers discovered in the National Archives files

ti1tlff)LE

YIA TALeNTK~'iHOL! eOMINTemmeL

8'18T!M8 JOIN'lL i

lOP SECRET UMBR-A

92

DOClD: 523696

REF lD:A523696
rap SECRET
UMBRA

(U) Frank Cburch

some Defense paperwork relating to domestic wiretaps which referred to NSA as the source of the request. The committee was not inclined to make use of this material, but the two staffers leaked the documents to Representative Bella Abzug of New York, who was starting her own investigation. Church terminated the two staffers, but the damage had been done, and the committee somewhat reluctantly broadened its investigation to include the National Security Agency.48
(8 ceer What the committee had found was the new Shamrock operation. It had become easier to use wiretaps than to get traffic fromcable companies, and NSA was using this technique with increasing frequency. But the Church staffers quickly uncovered the older Shamrock operation, and this became the focus of its early investigation of NSA. Knowing the ramifications, Allen terminated the portion of Shamrock that dealt with the cable companies on May 15, inthe middle of the preliminary hearings, so I

(FOUO) NSA's official relationship with the Church Committee began on May 20 with a visit from the committee staff; five days later Church himself came to Fort Meade for briefings and tours. This began a close association which extended over the entire summer and through October 1975. In the beginning it was a rough road, with committee staffers trying to dig deep: while NSA officials tried to protect. But with a few choice words from Allen, NSA's responsiveness improved and, with it, the cooperation of the committee. By the time it was all over it had become a model of how an intelligence agency should relate

1~AN9U; VIA TM.SN'l' KEYI leLB eeMIN' qeN'fReL SYS'fEMSJ~HN1'LY

93

TOP SECRET l:JMBRA

OOCIO: 523696
TOP SECRET tJMBItA

REF ID~A523696
CRYPl'OLOGIC QUARTERLY

to Congress, and it enhanced NSA's reputation on Capitol Hill: But it hadbeen tough 'slogging.~l (U) In September, the committee decided to request open testimony by Allen. They discussed two operations, Shamrock and Minaret, and in the end decided to question him about only Minaret. The committee discussions on the question were among the most rancorous of all, and Goldwater and Tower openly dissented from the proposition of requiring anyone at NSA to testify on any subject. But they were out.voted, and Allen was subpoenaed, despite a phone call from President Ford to Frank Church. ~t (8-000) Never had NSA been forced into such a position, and Lew Allen was very nervous. In a preliminary letter to Church he stated: .
As we prepare for open bearings. I am struck even more forcibly by the risks involved in this method of reporting to the American people•.. , Despite the honest and painstaking efforts of your pommittee and Staft'to work with
WI

to limit damage. I remain concerned that the open

hearing presents significant and unneco888ryrisks. 53

Allen pleaded that the cost of exposure of Minaret could be very high. The Watch List was a byproduct of NSA's operation to monitor ILC (int.ernational commercial) communications~ Withheld from public release I E.O. 13526, section 1.4(c) I Pub. L. 86-36
I

(U) The Chureh Committee conducted its open hearing on NSA on October 29, after two days of meticulous closed-door rehearsals. .The director began with a prepared statement describing NSA's mission in very generai terms and used historical examples (the Battle of Midway and the decryption of the Japanese Purple machine being two) to depict the value of such operations. He detailed the Agency's legal authorities and defined what NSA thought was meant by "foreign intelligence" and "foreign communication. Conceding the murky nature of the definitions, he then launched into a discussion of the Watch List, placing it in historical context and discussing how NSA interpreted the tasking and executed the support to requesting agencies. He stated that he himself had closed down Minaret two yearsbefote. ~
II

(FOUO) Lew Allen's performance was a triumph. Future vice president Walter Mondale noted to the director that "the performance of your stafl' and yourself before the committee is perhaps the most impressive presentation that we have had. And I consider your agency and your work to be possibly the single most important source of intelligence for this nation." Despite the accolades, however, when the committee in closed session discussed how much to tell about NSA, the mejority voted to include Shamrock, which Allen had oppOsed because of the embarrassment to the cable companies. Goldwater, Tower, and Howard Baker were set· in bitter opposition, but Church contended that legislation would be necessary to insure that abuses would not be repeated, and both Shamrock and Minaret constituted important material to back up the request for

IWf9b8

VIA 'J:hUS!f'i' KB¥U61:dS oeMUlT

CO!RReL S¥61'tiMS40INTLY

TOP SECRET UMBRA

94

DOCID: 523696

REF ID:A523696
lO' SECRiT UMBRA

legislation. When asked, Secretary of Defense James Schlesinger and the DCI, William .Colby, viewed the release of these two projects to be affordable.~ (U) When the Church Committee issued its final report in Febrbary 1976, the discussion of NSA was brief. FOCusingon what NSA could potentially do, rather than what it was doing, Church concluded:
The capabilities awesome. that. NSA now po88esa[es1 to intercept in technology and analyze communications are As

Future breakthro\1gbs barriers

will undoubtedly

in~..ease that capability.

the technological

to the interception

of all forms of communication safeguards

are being eroded,

there mWlt be a strengthening

of the legal and operational

thAt protect Arnericans .

. NSA's existence should be based on a congressional statute which established the limitations, rather than on an executive order then twenty-three years old. And so ended the discussion of NSA, just seven pages in a report comprising seven volumes of hearings. &7

(U) THE PIKE COMMITTEE (U) The backwash of Hersh's Family Jewels article. also infected the House of Representatives and produced the .predictable clamor to investigate. So the House held its own investigation, under Representative Otis Pike of New York. Not surprisingly, it became known as the Pike Committee. (U) But it did not begin that way. The first chairman was to be Lucien Nedzi, who chaired the Intelligence Subcommittee of the Armed Services committee. But this effort dissol ved in controversy when Democrats on the committee 'discovered that Colby had taken Nedzi into his confidence over the original Family Jewels report and had convinced him not to investigate. Fatally compromised, Nedzi resigned, and the task fell to Pike.~

(U)OUaPlke

{U) While the Church Committee focused on CIA, the Pike Committee had a much broader charter. It was to review the entire intelligence apparatus and to focus on operational effectiveness, coordination procedures, the protection of individual liberties, possible need for more congressional oversight, and on planning, programming, and budgeting. Pike promised to evaluate the performance of the Intelligence community

IllrlfBLE "IiA 'fALI!ltff ItEYiJ9L1!l e9MIN'f eetffR6L

S-ySftlMfl tJ9fNl'b Y

95

TOP SECRET liMBRA

DOClD: 523696
Tap SECRET \;IMBRIe

REF lD:A523696
CRYPTOLOGic QUARTERLY

against its budget. But the membership was liberal (somewhat more so than that of the Church Committee) and the staff intrusive. The focus quickly swung to the topic of abuses of individual liberties, and stayed there. 59 (FOUO) NSA had already had one experience with Pike, when he had chaired a subcommittee investigating the Pueblo capture of 1968. It had not been a happy encounter. The committee had leaked in camera testimony of the director, Lieutenant General Carter, to the press, and Carter was furious. Once burned, the NSA staff was wary (see American Cryptology during tbe Cold War, 1945·1989, Book [I: Centralization . Wins, 1960.1972, p. 449). (FOUD) The House charter gave the committee the power to determine its own rules concerning classification, handling, and release of executive department documents. Burned during the Pueblo investigation, NSA lawyers were anxious to nail down an agreed-upon set of procedures, but preliminary meetings yield~d no agreement on the procedures for handling SIGINT documents. Lew Allen, who later characterized the Pike Committee staffers as "Irresponsible," issued instructions to «limit our discussions with. the full House committee and staff to administrative, fiscal and management matters." 60 (8 GGO) Relationships quickly deteriorated. NSA officials described the committee staff as "hostile," the procedures for handling classified material as questionable, their willingness to learn about NSA as nonexistent. One NSAofficial noted that only one Pike staffer ever visited NSA, in contrast to the Church Committee, whose entire membership and staff visited Fort Meade in May 1975. Pike staffers objected to having NSA officials in the room when NSA employees were being questioned, and the staff interrogation of degenerated into a shoving match."

I

E.O. 13526, section 1.4(c)

I

I

.\

Withheld from public release Pub. L. 86-36

(FOUO) In August, the committee called Lew Allen to testify. The letter requesting. his presence stated that the budget policies and procedures would be the topic, but questioning soon turned to supposed monitoring of Americans. Allen objected to covering this ground in open session, and after a long committee wrangle and Allen's adamant refusa.l to go further, the committee voted to go into executive session. Summari:z.ing NSA's objections, he said: "I know of no way to preserve secrecy for an agency such as NSA other than to be as anonymous as possible, and to abide by the statutory restrictions which the Congress instructed us to, and those are that we do not discuss our operations; we do not discuss our organization; we do not discuss our budget in public." 6~ Thr;oughout Allen's appearance, Pike and Congressman Ron Dellums of California seemed suspicious and disbelieving.' At one point Pike interrupted the interrogation to say:
Now why don't you just tell any other member
WI

and be forthcoming.

without my having to drag it out of you, or of American

havin& to drag it out of you, what sort of communications how you Ilre intercepting

citizens you are intercepting,

tllam, wllat you are doing with them,

and why you feel it is necessary U; keep on doing it. lIS

The presumption of guilt was palpable.

\

RANDLE

y 1A 'fA:LEtfT tffiYIIOLE

eeMHff 60N'fROL &YstBMSdOINTL¥

TOP SECRET tJMlRA

96

DOClD: 523696

REF lD:A523696
TOP SECRET UMBRA

(8 eCOTOn September 8. the committee requested that NSA supply it with pertinent· intelligence products relating to the Yom Kippur War. The documents arrived on the 10th, and by the next day they were in the press, The Ford administration cut ofTall contact with the committee at that point. citing the leak of NSA materials, The passage . that resulted in the cut-off was a CIA summary which read:
Egypt - The (deleted) large·seale mobilization exercise may be an effort to soothe internal problel1lllas much aa to improve military capabilities. Mobilization of eome personnel. increaaing readiness of isolated united. Clfld greater cornmunicatio,.. sccurity are aU assessed sa part of the exercise routine .... (Italics added.) M

The phrase "and greater communications security" tipped off the COMINT origins of the information, and became known around NSA as the "four little words." it caused a crisis in executive-congressional relations because of the assertions by Pike that Congress could declassify on its own information classified by the executive department.· The matter was resolved, afte~ several weeks, by an agreement that the Ford administration did, indeed, control executive classified material, and in return agreed to relax its total ban on providing classified documents to the committee. NSA was soon forwarding material to the committee again .
.(s GCe) The final report criticized NSA's reporting policy. which amounted to firehosing the intelligence community. "NSA intercepts of Egyptian-Syrian war preparations in this period (Yom Kippur War 1 were so voluminous .•.an average of over 200·reports each week - that few analysts had time to' digest more than a small portion of them." It noted that NSA frequently- had the riglit answers, but that customers probably did not fully , understand what NSA was really saying. The Agency was also criticized for participating in the general intelligence failure during the 1968 Warsaw Pact invasion of Czechoslovakia. Like Church, Pike recommended that NSA's existence be authorized through congressional legislation and that "further. it is recommended that such legislation specifically define the role of NSA with reference to the monitoring of communications of Americans." 65

(U) The Pike Committee ended awash in controversy. On January 19, the committee distributed its final report. The Ford administration protested that it contained classified information, including several sections with codewordmaterial. The corrimittee voted, 8-4, not to delete the classified sections. and it sent the 340~pagereport to the House. Faced with anguished protests from the Ford administration, the House Rules Committee on January 29 voted 9-7 to reverse the Pike Committee decision, (Pike condemned this as "the biggest eoverup since Watergate.") 66 But it was already too late. On January 22 the .New York Times reported that it had knowledge of details of the report. On January 25, CBS correspondent Daniel Schor~ stated triumphantly on national television, ~I have the Pike Report." Four days later the House secured all copies of the report except the one in Schorr's possession. Fearing a Ford administration backlash and possible prosecution, . CBS refused to publish. Schorr then contracted with the Village Voice, and the report

HANDLE VIA TALENT KEYHOLE COMINTCONTROL SYSTEMS JOINTLY

97

TOP SECRET UMBRA

DOCID: 523696
TQp SECRET l:JMBAA

REF ID:A523696
CRYPTOLOGIC QUARTERLY

appeared in entirety in that publication in February, an event which led CBS to terminate his employment," (U) Despite protestations by Pike that the executive department was doing all the leaking, his own committee allpears to have been the source. 'I'he-draft report was distributed to committee members the morning of January 19, and by four o'clock that afternoon a New York Times reporter was already on the phone with the staff director asking questions based on the report. Versions of the report would appea~ in the press, the committee would make wording changes, and the next day the new wording would be in the newspapers." (U) Pike apparently began the investigation determined to produce a fair and balanced evaluation of American intelligence. He focused at first on job performance measured against funds expended. But the committee was top-heavy with liberal Democrats, and things quickly got out of hand ideologically. The committee and its staff refused to agree to commonly accepted rules for handling classified material, and when the executive department thwarted its desire to release classified material, it leaked like a sieve. The dispute with the administration over the release of NSA material produced an impasse, and diverted the committee from its original task, The House committee that was appointed to investigate the investigators turned up a shabby performance by the Pike Committee. In the end, it did Pike and Congress more damage than it did the Ford administration. All in all, it was a poor start for congressional oversight.

(U) THE ABZUG COMMITTEE (U) Serious (if ideologically polarized) inquiry descended into opera boufie with the charter of yet a third ·investigation. The leader was Bella Abzug, who had been elected to Congress in 1972 from a liberal district in New York City amid the early voter reactions to Watergate. (s eeO) Abzug chaired the Government Information and Individual Rights Subcommittee of the Committee on Government Operations. In mid-1975, with the Church Committee holding preliminary investigations in executive session, Abzug got hold of some of the more sensational information relating to Shamrock and Mw.aret. (The information was apparently leaked by Church Committee staffers.) 65 The climate for a
(U) Bella Abzu.

1:1 A lIlDLi \lI A TO, I.Em KEYHOU GOJflNT GO~lTROLSYS'RiMS J9IPfU?1

'TOP SECRET UMBRA

98

DOCID: 523696

REF ID:A523696
TOP SECRET I:JMaRA

full investigation of NSA was right. The press had picked up some of the themes resonating in the Church and Pike. hearings; An article in the September 8 edition of Newsweek described the "vacuum cleaner" approach to ILC collection and referred to NSA as "Orwellian." This was counterbalanced by a statement that "the NSA intends nothing like tyranny - it is probably the most apolitical agency in Washingt;Qn," But the fourth estate had clearly discovered the technological advances that permitted NSA to cast a very broad net. and characterized it as a potential threat to individual liberty .70 (S-CCO) NSA relationships with the Abzug Committee staff were poisonous. At their very first session. Abzug staffers refused to sign the normal indoctrination oath. and further discussions proceeded at the noncodeword level. Despite the refusal to accept executive department rules on clearances. the committee subpoenaed huge amounts of material. One subpoena. for instance, demanded every record, including tape recordings, of every scrap of information pertaining to the Agency's CO MINT mission since 1947. (Tape recordings alone comprised in excess of a million reels.) 11 Fearful of leaks that might dwarf those of the Pike Committee, the Ford administration decided to deny these requests. -rertn October, Abzug began maneuvering to get Lew Allen to testify in open session. The sparring sessions (Allen had no intention of complying) ended on October 29 when . , Allen appeared before the considerably less hostile Church Committee. Preempted, Abzug pressed for lower level NSA officials, and subpoenas began arriving at NSA. With the climate of mutual suspicion that existed, NSA resisted. Allen went to Jack Brooks, chairman of the full committee, to protest, and extracted a promise that Abzug could subpoena, but Brooks would refuse to enforce the subpoenas. In the end, Abzug got her hands on one unfortunate NSA official, Joseph Tomba, who appeared in open session and refused, at the request of DoD lawyers, to answer most .questions put to him. The committee held Tomba in contempt, but Jack Brooks was good to his promise, and the citation was not enforced. n ~ In the process of dealing with Abiug, Lew Allen and his staff were subjected to fearful browbeating, but they held fast, defended by not only the full executive department, but by Congressman Jack Brooks himself. Hearings dragged on into 1976, making Abzug the longest running of the investigative committees. Then, in September of 1976 they began to fade, as Abzug became involved in a campaign for the Senate. and hearings ceased. (She ultimately lost.) The committee eventually issued a draft report (February 1977) which predictably concluded that there were still loopholes which would allow NSA to intercept U.S. communications for foreign intelligence purposes and that these loopholes should be closed. But the importance was secondary. Church had already exposed the loopholes and had made the same recommendations, Moreover, by then President Ford had issued his new executive order, 11905, which forbade many of the "abuses" that Abzug had in mind. The committee faded into irrelevance." (U) With that, the investigative process had run its course. It had been a pretty thorough public housecleaning for all intelligence agencies. For CIA (and to a lesser extent FBI) it had been traumatic and damaging. For NSA, the trauma had been much

"

MANDL! \' f* Ti\L!NT K! t HOL! eOMlNT eONTKOL S't'SfEMS JOINTLY

99

lOP SECRET liMBRA

DOClD: 523696
TOP SECRET UMBRA

REF ID:A523696
CRYPTOLOGIC QUARTERLY

less. The principal reason was the director. Lew Allen - kindly, thoughtful, intellectual, and forthright - was just the right person at just the right time. He disarmed most of NSA's more reasoned critics with the way he directed his staff to respond to Congress. He headed off controversy before it got well started .. Most of all, his five-star performance before the Church Committee convinced many that NSA had not gone seriously off track and that it should be preserved at all cost. A glimpse under the cryptologic curtain convinced most senators and congressmen that NSA was the true gem of the intelligence world.

(U) THE BACKWASH (U) The Watergate era changed crypto10gy. The tell-an atmosphere resulted in a flood ofreveiations unprecedented then and now. It also resulted in new executive department restrictions on cryptologic operations and ushered in a new era of congressional oversight.
(U) The Revelations

(U) T~e investigations were conducted amid an absolute fury of press revelations, many apparently stemming from the committee staffs. The Washington Post termed NSA "America's Huge Vacuum Cleaner" and highlighted the reading of South Vietnamese diplomatic communications during the peace negotiations of 1972. Post articles in May 1975 revealed the atrocities of Pol Pot's government in Cambodia and indicated that . COMINT was the source. (This was probably a Ford administration leak.) The New York Times and Daily Telegram both exposed an alleged navy underwater SIGINT collection program called Holystone (which, if true, would have held the program at serious risk) .. The Times published articles about the extensive American support for a new SIGINT program for the shah of Iran. Penthouse published a lengthy expo~ of the nature and scope of NSA's operations, adding tidbits about a Third Party relationship with Israel, capability to track Soviet submarines, and the supposed monitoring of domestic communieations.P
I

(U) More serious still were articles on American cryptologic relationships with Second Parties. In November 1975 the Sunday Los Angeles Times revealed the location and function of three American SIGINT sites in Australia, including one at Pine Gap in central Australia. In New Zealand, members of Parliament demanded that the government confirm or deny the nation's membership in UKUSA.7s (U) Revelations continued the following year. In February the Far East Economic Review shone the spotlight on Ramasun Station, and the press coverage continued through the spring, thus increasing the chance that Thailand would close the station (which it did). Rolling Stone chimed in with an article by an ex-operator named Chet Lippe, who evidently wanted to follow in the footsteps of Wins10w Peck. David Kahn, the noted authority on cryptologic history, published a series of articles revealing cryptologic operations and sounding an alarm about potential violations of civil liberties. One article,
. (

1b\N9LE

VI!. TkLENT KI5YUObFl eOMINT GON'FROb SYs:FI5MS cJOIN'i'LY

fepS!CRE'f tlM11tA

100

DOClD: 523696

REF lD:A523696
TOP SECRET UMBRA

"Big Ear or Big Brother," depicted the theme of Orwellian intrusion. (Kahn had become exercised over the DES CData Encryption Standard) controversy which was then roiling academia; see p. 231). British and Australian journalists continued their revelations about the close UKUSA relationship - this trend- ended in the exposure of every UKUSA monitoring site in both countries: William Beecher, the investigative journalist who had been so proficient in digging out intelligence operations in the past, published revelations about an American collection operation in ,the U.S. embassy in Moscow and about Soviet attempts to interfere with it by bombarding the embassy with microwaves. 76
'. (U) Glomar Explorer

JGt One of the most intriguing exposes related to a CIA operation called Azorian. In 1968 a Soviet Golf-class nuclear submarine on patrol in the Pacific mysteriously went to the bottom with all hands. The Soviets could not locate the wreck, but the U.S. Navy could, and the U.S. began to study the feasibility of capturing it. Once it was concluded that it would be feasible, the job was .ven to DCI Richard Helms.

I E.O.

13526, section

1.4(c)

I

Withheld from public release Pub. L. 86-36

kS}'Ultimately the Azorian task force came up with a special ship, which could lower a "capture ship" to the Soviet sub, which rested in 1,700 feet water about 750 miles northwest of Hawaii. The capture ship had huge claws which would be capable of grabbing the submarine and bringing it to the surface as it was hoisted to the mother ship. Hughes Corporation became the prime contractor, and Sun Shipbuilding of Chester, Pennsylvania, -was selected to build the vessel. CIA devised a cover story that the ship was designed for mineral prospecting on the ocean floor.

of

I-E-.O-.-1-3-S-2-6-, -se-c-ti-o-n-l-.-4(-c-)-I-ffit In August 1974, with CIAI Ipeople aboard, the Hughes vessel, named , - Glomar Explorer, 'sent its capture vessel to the bottom .. Everything went fine until the
Withheld from public release Pub. L. 86-36 crew be an lifting the submarine from the ocean floor. The submarine hull snapped, and of it sank back down to the bottom.. The portion that CIA retrieved had I I They would have to go back . ..kSt Despite the fact that a Soviet seagoing salvage ship observed the operation from a safe distance. CIA planned to return to the site and risk exposure I I I I But then the press intruded. The original leak resulted from a burglary at Summa Corporation, a subcontractor for the operation. CIA feared that a Hughes' Corporation memo regarding Azorian might have been in some papers that disappeared from the office, and they decided to brief a few of the police investigators

HANDLE VIA TALENT KEYHOLE COMINT CONTROL SYSTEMS JOINTL Y

101

TOP SECRET UMBRA

oocrn. .523696
TGP SECAE=rUMBRA

REF ID:A523696
CRYPTOLOGIC QUARTERLY

(U) GlolIIaI' Ezploru

involved with the case. It was a potentially sensational story and, sure enough, it was leaked to Lo« Angeles Time. reporters covering the break-in. In March 1975. before the second salvage mission couid be mounted, Jack Anderson went public with it. and CIA decided to cancel all further attempts.71
(U)KOI'Ngate

I E.O.

13526, section 1.4(c)

Withheld from public release Pub. L. 86-36

Ib\NfJLI!l Vb\ 'f*LENT 1tE· •l.f6LE e6MfN'!' e6N'!K6L8'YS'fEMS J6IN'fLY

TOP

seem UMBRA

102

DOCID: 523696

REF ID:A523696
ref SECRET I:IMBRA

I E.O.

13526, section 1.4(c)

Withheld from public release Pub. L. 86-36

(U) Newspapers were, of course, following the Fraser investigation, and rumors began appearing that the indictment was based on NSA information. On Sept~mber 4, 1977, the New York Times published an article alleging that Henry Kissinger, Melvin Laird, and other top officials had been aware of the South Korean bribery ring at least as early as 1972. In discussing the source of this information, the Times said: "While the, investigators did not identify the documents precisely, other sources said that the

HANDL! 'iIA !ALEt~'f KeynOLe eOMINT eON'fROb~SftjM5 cf9UITl.¥

103

lap SECRETYMIRA

DOClD: 523696
TOP seCRET YMBRA

REF lD:A523696
CRYPl'OLOGIC QUARTERLY

/

documents came from the Central Intelligence Agency, whieh was earlier reported to have agents in the presidential executive mansion in Seoul, and from the National Security Agency, which has been reported to have intercepted South Korean cable traffic between Seoul and Washington."

Withheld from public release Pub. L. 86-36
(U) On September 6, two days after the Times story, a federal grand jury indicted Tong-Sun Park on thirty-six felony counts of bribery, conspiracy, mail fraud, illegal campaign contributions, and other charges. A California congressman and several former Korean intelligence officials were listedaa "unindicted co-conspirators." This placed the issue in the realm ofthe courts."
(u) But the Koreagate affair was hardly dead. In October 1977, the New York Times reported the bizarre case of Sohn Young Ho. Sohn, the top KCIA agent in New York City, was in the process of asking the United States for political asylum when Edward J. Derwinski, a member of the Fraser Committee, allegedly tipped off the KCIA, which went looking for Sohn, possibly intending to mailbag him back to Seoul for safekeeping. Fortunately, the FBI got to him first, but the source of the information about the Derwinski leak, according to the Times, was NSA.84

I E.O.

13526, section 1.4(c)

Withheld from public release Pub. L. 86-36

Congressional L-o-v-er-s"""ig'h"t'-w-as-;;Ofi-n-e-a-s-:-lo-n-g"i:7"t-w-as--'k-e-p-:-t -w-;-it:;-h-:-in-a-n-ar-ro-w-ra-n-g:-e-an--'d;-s-u"b-:-~ec::-:-;:-te:jd to the greatest restrictions. As a test of providing SlGINT support to law enforcement, however, ithad a much shorter influence. The Reagan administration began reversing that course in 1981,

IfANBbE VIA tALENt It! 'lUeU

CeMiNf CeN'FReJ:; S\'9'fBMS 4911't'H.Y
I

TOP SEEItEl UMBItf(

104

DOCID: 52369'6

REF ID:A523696
TOP SECRET ijMBItA

insisting that SIGINT be expanded to provide more, rather than less, support to domestic law enforcement.
(U) Executive Order 11905

(U) If the president did not act to restrict the intelligence community, it was clear that . Congress would. So during the fall of 1975. with the Church hearings in full throttle, President Ford appointed an Intelligence Coordinating Group, chaired. by White House counselor Jack Marsh, to draft a comprehensive order, at once organizing the intelligence community and placing checks on it. ae The result was Executive Order 11905. (U) Organizationally, the president gave the DCI more authority to supervise the intelligence community, including the critical budget review "club" .that Nixon had tentatively proferred to Richard Helms in 1971. The DCI became chairman of a new Council on Foreign .Intelligenee, which included the assistant secretary of defense for intelligence (a newly created position which would supervise NSA's director). Ford abolished the 40 Committee, which had ruled on all covert operations (including SIGINT peripheral reconnaissance missions) and replaced it with an Operations Advisory Group. , He continued the President's Foreign Intelligence Advisory Board and directed that three of its members constitute a special Intelligence Oversight Board to keep track of possibly illegal activities by intelligence organizations. The executive order attempted to draw a clear line between "foreign intelligence" and "domestic law enforcement." 87 (U) The organizational aspects were of less concern to NSA than were the specific prohibitions. The order prohibited, the intercept of communications made from, or . intended by the sender to be received in. the United States, or directed against U.S. persons abroad. except "under lawful electronic surveillance under procedures approved by the Attorney General." ae
(8-0eO) The 'new executive order resulted in the termination of many NSA activities

in support of law enforcement.

I E.O.

13526, section 1.4(c)

Withheld from public release Pub. L. 86-36

The crisp wording of the order obscured the resident subtleties. How did an analyst know if a person was an American citizen, a resident alien, or just a person with an .:American-sounding name? How would NSA segregate. within its database those

is-ccer

fiA1\(DLE "JA TALENT K! tHOLE eOMIN'! eaN'fR6J:; S'l-5'l'EMS c19INTbV

105

TOP SECRET UMlRA

DOCID: 523696
TOP SECRET UMBRA

REF ID:A523696
CRYPTOLOGIC QUARTERLY

individuals against whom collection was legal, from those against whom collection was authorized only in specific instances? In fast moving crises such as the Ma]agu.ez affair, how could NSA determine if collection was authorized? If it was not, but lives were in danger, who would rule on permissibility? And ho~ much easier it was to Monday moming quarterback the situation than to operate during crisis in the dim, floating world of possible prosecutability. In mid-1976 the NSA ODD, Robert Drake, noted to the Ie staff that "To the. question of whether or not day-to-day SIGlNT production can continue under the provisions of the Executive Order, the answer is yes. In other words, although the guidance is annoying, at times conflicting, and necessarily subject to interpretations at the desk level, I can cope with it. . .. On Monday morning, of course, we all can judge that that incident [Ma]aguez) was reportable but in cases such as this Monday may be too late." Despite such uncertainty, NSA drafted the general wording of the executive order into a new regulation, USSID 18, which stood the test of time for many years. As with the executive order, it was an attempt to preempt more restrictive congressional legislation; Lew Allen considered the matter to be extremely important and got White House approval.P" (U) One result of the Watergate period was to complicate NSA.'s life in the area of domestic wiretapping. The matter of wiretapping for law enforcement had been contentious since the first Supreme Court decision in 1927, which gave the federal government broad latitude to do electronic surveillance. Courts gradually narrowed this. .down, and by the 1970s the new climate of concern for individual liberties had basically made warrantless electronic surveillance inadmissible as evidence. But wiretaps for foreign intelligence did not.fall within this rule, and in the early 19705federal courts ruled that foreign intelligence wiretaps were legal.91 JS..Ge8) The "New Shamrock" operations involved wiretapping foreign embassies in the United States. Begun in the 1950s, those wiretaps had continued for years despite periodic resistance by J. Edgar Hoover. Through the decade of the 19605, the number of such wiretaps fluctuated in the sixty seventy range. But in December 1974 Attorney General Levi instituted new and eumbersome approval procedures which both lengthened the time needed for approval and broadened the exposure of specific operations from just a few people to a number spread around the intelligence and national security community. At the top of the heap, the attorney general maintained personal control and began disapproving requests that sported justifications that he regarded as weak. Lew Allen tried to divest Levi of control of domestic foreign intelligence. wiretaps, but was unsuccessful. But, though EO 11905 specificaHy stated that taps for foreign intelligence would be treated differently from taps for domestic law enforcement, successive attorneys general continued to control foreign intelligence taps through the Carter administration. To NSA, it was a cost of doing business that had not existed before Watergate.92

to

(U) The last act in the play occurred in 1978 when Congress passed, and the president signed, the Foreign Intelligence Surveillance Act (FISA). This added another approval layer, consisting of a special court of seven judges which would rule on requests from the attorney general for warrantless taps. Although this lengthened further the process of

lIANB~

VIA '!lALEn, KIH'neLE eeMIH'f eeU'PReL SYS'l'FlMS ;ieURLY

Ta, SECRETUMBRA

106

DOClD: 523696

REF lD:A523696
TOP SECRET UMBRA

Withheld from public release Pub. L. 86-36

instituting the taps, it had no effect on their approval.

(U) Congressional Oversight . .

(U) Congressional oversight of the intelligence community sprang frdm the Watergate period. Prior to the Church and Pike committees, oversight was more or less nominal and was confined to just four committees: the Armed Services and' Appropriations committees . in both houses of Congress. Had Congress no budget to approve, oversight probably would have been even more sketchy than it actually was.
(U) Each of the four committees set up special intelligence subcommittees, comprising the full committee chairman and three or four trusted members from both sides of the aisle. Their examination of funding requests was cursory, and they never asked embarrassing questions about operations. The president controlled the requests, and if someone's intelligence budget were to be shaved down, the executive department would have to do the shaving - congressmen did not get into those details. Thus, inclusion in the president's budget was tantamount to approval.

(U) In the Senate, one man dominated oversight - Richard Russell of Georgia. Serving from 1933 to 1971, Russell chaired both the Armed Services Committee and the Intelligence Subcommittee of the Appropriations Committee. In the House, a succession of chairmen, almost all from conservative southern states with strong national defense leanings, dominated the proceedings. Mendel Rivers, Carl Vinson, and F. Edward Hebern strongly supported intelligence projects and insured that the information was held as tightly as possible in Congress. Lawrence Houston, the CIA general counsel, once' said that "Security was impeccable. We never had the slightest breach." 94 Summing up the dealings with Congress, Clark Clifford said, "Congress chose not to 'be involved and preferred to be uninformed." a~ This situation lasted as long as bipartisan consensus continued. (U) Special intelligence clearances remained mysterious and obscure. In 1968, at the' time of the Tonkin Gulf hearings in the Senate Foreign Relations Committee, no committee members, not even the chairman, William Fulbright, had even heard of clearances above top secret. This problem tied the committee in knots during the testimony of Robert McNamara relating to the August 4,1964, attack (see Book II, p. 518) :

Senator Gore: Mr. Chairman, could we !mow what particular classification that is? I had not heard of this particular cluaif'ic:ation. Senator Fulbright: The IItaft'.Mr. Marcy. and Mr. Hold are cleare,dfor top secret information. This is aomething I never beard of before either. It is IOmething special witb regard to intelligence information. However. Mr. Bader was deared. for that.

Hl\NHL£ '(IA Tl\LENT a fHOLE eo MINT eONTK6L SYSTEMSJOIN!'L'"

107

TOP SECRET ijMBRA:

DOClD: 523696
TOP SECRET !:IMBRA

REF lD:A523696
CRYPTOLOGIC QUARTERLY

Secretary McNamara: If the staff would wish to. requast clearance, I am sure the Government would do it. Mr. Marcy: All of the membefs who are here submitted rel!ewal requests fOftop secret clearance
recently and,lIO rar as 1 know, all oftb088 requesta bave been granted.

Se~retary McNamara: But that is not the issue. Clea~ance is ab'ove top secret fOfthe particular information involved in this aituation:ee

(U) By the time the congressional hearings had ended in 1975, the culture had completely changed. Church had termed CIA a "rogue elephant," and Closercongressional scrutiny was inevitable. The first thought of Congress was to set up a joint House-Senate committee, but the House fell behind and, unwilling to wait, the Senate established the Senate Select Committee on Intelligence (SSC!) on May 19,.1976. The tardy House, consumed with procedural wrangling over the release of the Pike Report,. delayed until July 17,1977, more than a year later, when it established the House Permanent Select Committee on Intelligence (HPSCI). 91 (FOUO) Ultimately, all members of Congress were to be presumed cleared, and all staff members from the two oversight committees had Sland other security clearances to allow them to do their job. Clearances were also granted to select staff members of certain other committees (like Appropriations) to permit them to do their jobs. Though there were some rough spots at flrSt, NSA-congressional liaison came to be a more or less routine function bedeviled only occasionally by security problems. Certainly there were no repeats of the maverick Pike Committee performance. NSA senior Walter Deeley summed up the matter ten years later: "... I think one of the best things that ever happened to this country is the fact of the establishment of the House Committee on Intelligence and the Senate Committee on Intelligence, and they have total, absolute total, Scr:utinyover what NSA does." 81
(U) The Enabling Legislation

(U) The same Congress that decreed congressional oversight also wanted enabling legislation for the intelligence agencies that had not been established by law, as well as specific limiting legislation for CIA (which had already been established by the National Security Act of 1947).. NSA was the most visible of the agencies that had come into being by executive order, and the Agency was one of the main targets of the draft legislation. All the drafts took the same basic form. NSA would have the same authorities as under the Truman Memorandum and would remain within the Department of Defense. The director and deputy director would be appointed by the president and confirmed by the Senate. As with the CIA, the director could be either civilian or military, but if military, the deputy must be a career civilian. What distinguished these drafts from the Truman .Memorandum was ~ heavy emphasis on civil liberties, to be guaranteed through an overlay of oversight bodies - checkers and people to cheek the checkers. The driving force

I1ANBLB \'1" 'flW!lN'f KEYIJOLE GOMIN'!' CONTRaL S"t'S'fEMS JeINTLY

TOP SEERET I:lMBItA

108

DOClD: 523696

REF 1D :A523696
TOP SECRET I:JMBRA

behind the legislation seemed to be the final report of the Church Committee, in which the committee promised to end the abuses of the past.!$ .(e)1nitially the enabling legislation was pushed along by the strong breeze of reform dominating the Carter White House. But as the president settled into the business of governing, he found this focus on supposed abuses of previous administrations to be increasingly irrelevant. Moreover, the intelligence agencies, and especially NSA, yielded a cornucopia of information. He became less and less interested in pushing legislation that would remove NSA from his total control and give part of that control to Congress. The Carter White House allowed the breezes of reform to blow themselves out, and NSA remained firmly tied to the president's authorities. The Truman Memorandum stood. UlO
(U) The Enigma Revelations

(U) In England, far away from Watergate's tumultuous effects on government, a storm was brewing that was to help NSA, even as it stripped away the gauze of anonymity that remained. It became known as the Enigma revelations. (U) The story of cryptology's role in World War II had been kept secret since 1945. Only the Americans, who had publicly investigated the surprise attack on Pearl Harbor, had uncapped that bottle, and even they had managed to confine the story to 1940 and 1941, and to limit the disclosures to the breaking ofJapanese diplomatic codes and ciphers. The other 95 percent had remained hidden.
I

(U) The story began to trickle out in 1972, with the publication of John Masterman's book The Double Cross System, which covered the capture and, turning of German human agents in Britain during the war. How they were captured was another story and went to the heart of the Enigma story, but Masterman kept that part a. secret. 101 (U) The first break. to the Enigma story itself occurred in France in 1973, when Gustave Bertrand, the head of French intelligence before the war, published his memoirs revealing the Polish break into Enigma and the conference in 1939, just before the .German Blitzkrieg swept over the country, Bertrand detailed his key role in obtaining information on Enigma for the Poles, and he desc~ibed France's attack against Enigma in the final months preceding the German invasion of 1940. He also described what the British knew about the system.102 (U) For a time the British remained silent. But within the ranks of World War II veterans there was a movement to tell their own story, largely to set right what they felt were distortions in the Bertrand account. Leading this effort was Frederick Winterbotham,a former RAF lieutenant colonel who had devised the system for protecting SlGlNT during World War II. Winterbotham began working on his own book, published in 1974 as The Ultra Secret. He did not speak with a grant of authority from his government and had in fact been warned not to publish. But since the publication of Bertrand's book a year earlier, references to the British attack on Enigma had appeared in nooks and crevices of 'articles and book reviews, many of them authored by people who had
i

109

TOP SECRET I:JMBAA

DOCID: 523696
1'6' SECRET UM8~

REF ID:A523696
CRYPTOLOGIC QUARTERLY

participated in the operation during the war. Winterbotham knew that it was only a matter of time, and he determined to beat the rush. His book laid out the entire story of 103 Bletchley Park, albeit with certain inaccuracies which came with the fading of memory. . . (U) Following Winterbotham, many participants told their stories. For some, like Peter Calvocorresi, editor-in-chief of Penguin Books, revelation became eloquent literature. For others,like Gordon Welchman, it became a detailed technical description that caused the government to blanch (and NSA to pull his aecessest.!" (U) But none exceeded in scope and detail Harry Hinsley's book on British intelligence during World War II, which was largely a detailed history of Bletchley and the Enigma project. Alone among the writers and historians, Hinsley was given access to the stillclassified documents, so that a well-documented story would emerge from among the welter of revelations and memoirs. Hinsley was given permission to use classified documents largely to correct misimpressions stemming from the memory-based accounts of Winter both am, Calvocoressi, and others.los (U) The story of American codebreaking successes was later in coming. Ronald Clark's T~ Man who Broke Purple, a somewhat breathless (and not entirely accurate) biography of William Friedman, came out in 1977, and was followed by less memorable personal accounts by two Navy men, Edward Van Der Rhoer's Deadly Magic in 1978 and Jasper Holmes's Double-Edged Secrets in 1979. These could not compete in drama and readability with the stories churning out of the British press, and it took an Englishman, Ronald Lewin; to begin to ten the American story in his book The American Magic.1oe The British story captured the moment, while accounts of similarly significant American CO MINT successes bobbed unhappily in their wake. (U) Memoirs, biographies, and selective leaks of information would not, of course satisfy either the public or the historians. The only realistic alternative was to begin declassifying and releasing documents. Here, national security came to loggerheads with the public's right to know, and the issue was resolved only during the post-Watergate sorting out. The declassification effort resulted from two post-Watergate initiatives, FOIA and EO. (U) Congress passed a new Freedom of Information Act (FOIA) in 1974. In it the. congressmen took an old law relating to government documents, which required .the requester to prove the need for the documents, and reversed it, instead requiring the government to prove the need to maintain secrecy. 107 Under this new law each government agency set up special arrangements to process FOIA requests .. For several years NSA's ForA team routinely denied every request based on national security. This worked under President Ford, but the new Carter administration in 1977 took the side of the plaintiffs on FOIA. Releasing significant numbers of documents became only a matter of time. (FOUO) Executive Order 11652, issued in 1972, dealt with openness in government, and decreed that government documents be automatically declassified and released to the

'Fep SECRET tlMBItA

110

DOClD: 523696

REF ID:A523696
lO'UCRiTU.48RA

(0) World Warn SIGINT histories

National Archives after thirty years. loa The order actually preceded FOIA, but it did not have a major effect on NSA until after the Church and Pike hearings. By then, Lew Allen had become director, and Winterbotham had begun the Enigma revelations. Seeing that it was only a matter of time, Allen's staff began negotiating with GCHQ for a coordinated bilateral policy on release. They agreed to concentrate on World War II records (those most in demand) and to restrict their declassification initially to the COMINT efTort against German, Japanese, and Italian armed forces. In Britain. declassified records would go to the Public Records Office - in the United States, to the National Archives in Washington. NSA would also look at selected Korean War and Vietnam era records, but the British declined, citing a rule against proceeding into the postwar period.loo
I

(U) NSA began the Herculean task of reviewing millions of pages of World War II (and prior) records in 1976, with four reemployed annuitants hired on a temporary. sixty-day basis. The program expanded as more and more files were discovered. Admiral Inman decided to set up a classified NSA archives to hold the records which had been saved but were not yet ready for declassification, and the new "Cryptologic Archival Holding Area" was set up in SAB-2. which had been built in the early 1970s as a warehouse to hold material being transported to a records destruction facility. (At the time NSA did not have its own faeility.)110

flA~LP! Vb\: !'ALENT KFlYIfOLE eOMIN'f eON'fROb SYSTEMS40INTL¥

111

TOP SEa.ET ~MBRA

DOClD: 523696
TOP SECRET tJMBRA

REF ID:A523696
CRYPl'OLOGIC QUARTERLY

. (FOUO) FOIA ran parallel to the systematic declassification effort, and the two threads became frequently intertwined .. In 1978 a researcher named Earnest Bell, who ,. had worked in the Army's wartime COMINT office in London, submitted a FOIA request for all German and Japanese COMINT material for the entire war. NSA's legal counsel, Roy Banner, advised Inman that NSA would likely lose a lawsuit, and the Bell FOIArequest greatly expanded the volume of material that the reemployed annuitants had to review. Ultimately twenty-one REAs were hired under Inman to plow through the enormous pile of ra w COMINT reports to satisfy Bell's request. 111

'-

(U)THE IMPACT OF WATERGATE (U) The Watergate period resulted in a massive change in the way the cryptologic system related to the American public. Congressional oversight, which sprang from the Church and Pike Committees, fundamentally altered the way NSA related to the legislative branch of government. In a real sense, NSA had to answer to two masters, and the relatively simple life of prior decades became more complex. The new arrangements took some getting used to, but in many ways accountability worked to the advantage of an agency thatworked within the law, and within a decade few could imagine going back to the old way of doing business. (U) If congressional oversight ultimately worked to NSA's benefit, the public exposures accompanying the Watergate period did not. Too many sensitive operations were exposed; too many exposes were splashed across the newspapers. 'The deleterious effects of the Watergate period stayed with the cryptologic community for many years to come.

Notes
1. (U) Theodore H. White. Breach of Fllitl&: .TIM FIIU of Rich4rd Nisolt, 1st ed. (New York: Atheneum Publlsbera,1975)j Karnow, Vietn4rn,634.

2.
3.
4.

(u)

Bradburn, et aL, Th6 SIGINT RlCOnMi.,anu Sctclli~6, 5·37.

<U)CCH Sene. vt.A.l.6.2.
(U) Seymour

Hersh, TIw Prie« of Power: Kiuingtr in th6 Ni:wn White House (New York: Summit Books,

1983),321·32.

5.

(V) CCH Series VI.H.H.19.6; 19.16.

6. (U) CCH Series vt.H.H.19.6; Harrison Salisbury, Without Fear or Fauor (New York: Times Books, 1980>, 315-18; Interview, Milton Zaslow, by Tom Johnson, 19 June 1998. 7. 8. 9. (V) Hersh, Kiuinger; CCH VI.I.I.l.2. (V)lbid. (U) Interview, MeyerJ.Levin, by Robert D. Fatley Bnd Tom Johnaon, 14January 1987, OH 2·87, NSA.

lw,abfl 'If A 'ftdlflN'f !tE'ffIeb!

eeMINT 6ONTROL SYSTEMS JOINTLY 112

TOP SECRET tlMBR:A

DOClD: 523696

REF lD:A523696
TOP SECRET UMBRA

10. 50. U. 12.

(U) Winslow Peek (Perey Fellwockl, "U.s.Electronie

Espionage

- a Memoir," Ramparts, August

1972, 36-

(U) CCH Series VI.I .1.1.2.; DDIR files, 96026, box 13, "1974..· (U) CCH Series VI. U.l.2; (Knoxville, Tenn: Frank Smist, Jr .• C0rl8rn. Ouersle, Univerlli.ty of Tennelse~

the Uniud

SIllUt

Intelligence Community.
recorda,

1947-1989 96026 . . 13.

Press, 1990),184-86>; DDIR mea, NSA retired

(U) "Swnmary

of Statute

I

Which Relate

Specifically

to NSA and the Cryptologic

Activities .

of the

Government." 14.

undated file in CCH administrative

mea collection.

(U) Ath ••n Theoharia, by David Alvar~,

Spying on Americ~n.a:
Chapter

Politi.cGl SurveillAnce

from Hoouer

to

the Huston

Pl~n

(Philadelphia: manuscript 15. 16. 17. 18. 19.

Temple University

Press. 1978), 120; NSA Archives.

aec nr 18238, CBTF 36; Unpublished

1, in CCH filel.

(U) AUen interview. (U) CCHSerieaVn.B.4;F.19, Voll.

(Ul David Kahn. "Big Ear or Big Brother?." New York Times Magazine, May 16,1976, 13; 62·72. (U) CCH Series XII.H.57 .4.
(U) Church

Committee

hearings,

Vol V,10, in NSA rec:ords center, 28791·2, 80·079.

20. (U) Theoharia, Spyir18 on Americana, 121·22; james Bamford, The Pwak Palau: A R,por1 on America', Mo,t Secrd Agency (Boston: Houghton Miftlin Co., 1982), 24.8; Church Committee correspondence .. 21. (U) NSA Archives, ace nr 18238, CBTF 36; Church Committee 13.62. Tom Charlel Huston, by Gerald Haines L.I 20,121. correapondenc:e. Vol V, 12. Kahn, "Big Ear

or Big Brother?,· 22. 23. 24.. 25. ~6. 27. 28. 29.

(U) Interview,

JL 31 Janu.,ry

1986. OH 6·86, NSA.

(U) Theoh&ria,Sp,yillB

onA_rica~.

Withheld from public release Pub. L. 86-36

(U) Allen interview. (U)Theoharia,Spyirl8onAm'ric~na,122·23. (U) National (U) Ibid. (U) DDIR fLIes, 96026, box 8, "CIA Sellllit.ive Items, ~ (U) When Nixon became president, Carter tried to brief him on a eurrent with Carter, wiretap program, but Haldeman, 3·6 <Xtober Archives, Record Group 467, SRH·OO1.'

whom Carter c:alled "a first class son of a bitch," insisted on being briefed first. Carter refused: and he and Hoover agreed to cancel the operation. 1988,OH 30. 31. 32. 33. 34.. 35. 15·88, NSA. Spyi"8 on Americanr. DDIRfiles,96026, box 13. "1974." See NSA's oral histOry interview by Robert D. Farley,

(U) Theoharis.

(Ul White. Brecu:f& of Ftlitf&.133; Tbeoharis. Spyi.ng on AmericaM, 22·26. (U) Tbeoharis,
(U) Ibid.

Spyi"B

ollAmeriocaM,16.

22·31.

(Ul Ibid., 32·33. (U) File on the HWiton eO!llJllittee in CCH Series XII.D.; Churc:h Committee correspondence.

IIldfBt.S VfA 'i)\LBN'f IEE'lf feLE cellm.'l' eeNTReb S-YS'fSMS 401mb .•

113

TOP SECRET UMBRA

DOCID: 523696
lOP SECRET UMBRA

REF ID:A523696
CRYPTOLOGIC QUARTERLY

36.

(U) HllIiton interview; Kahn, "Big Ear or Big Brother?; 13, 62; Theo~ris, Spying ortArneMearts, 27.

Withheld from public release Pub. L. 86-36

C""J by CbarlesBaker,

(U) Allen interview; InterviewJ Tom Johnson, Michael Peterson 25 February 1993, OH 4-93, NSA; Interview, Paul Brady, b~ ~25 July 1995, OH 22·95, NSA; Interview, Howard Rosenblum, by Robert Farley and Charles Baker, 19 September 1991, OH 3-91, NSA. The existence of a taping system in the Oval Offiee bad been revealed to the Ervin Committee by a Nixon aide, Alexander Butterfield, in July of 1973. Individual tapes were under subpoena, and the most controversial of them contained a gap of eighteen minute&at • crucial point in the Waterpte coverup, It was the tapes under subpoena, and especially the tape containing' that eighteen-minute gaP. that were taken to NSA f~r examination.

37.

I

~

I

38. 39. 40. 24. 41. 42. . 43. 44.

(U) (U)

Allen interview. Ibid.
TM CIA in TrtlllBition (Boston: Houghton Mifflin, 1985),

(U) Stansfield Turner, Secrecy artd Democracy:

(U) Allen interview.

un CCH Series XII.D; DOIRfiles,
(U) Smiat., Congrus

96026, box 10, "Directorate Correspondence, Nov. 75·Dec 76."

,

(U) Powers, Mart WIIoKept tlw Secree., 288.. Oller,ee, tIw UniUd Sto.teslnUlli8ence
Community,

9-10,149; Powers, Man. WIIo Kept

the Secret., 288-89.

45.

(U) Theoharis, Spying on A~ricaM,

.9-11. of Gerald. Ford (Lawrence. Kansas: University of Kansas Press,

46. (U)John Robert Greene, TM Presidenq 1995),106. '47. <U)Greene, The Pr,.~M.y 48. 49. ..50. 51. 52.

of Gerald Ford, 106: NSA Archives ace nr 45146N, H07-0201·6. Stat.t.lnUUigerzee Community,

(U) Smiet, Congrc81 Ouersus tIw U~d

30.

(U) Huston interview; Smist., Con.gre" Ouersee. tlw UniUd State, InUUigence Commurtity, 63 • (U)Huston interview; Church Committee correspondence. (U)Ibid.
(u)

Christopher Andrew. For tIw Prui4erU',

Eye. Only:

Secret In.tclligerzee and 1M AmericGn. Pl'tsiderzey

from Wa,lai~gtorl to.Bula (New York: HarperCollins, 1994),415 ..

53, 5., 55. 56.

(U) Church'Committee correspondence; DIRNSA letter of7 October 1975. (U)lbid. (U) CCH Series VI.D.2.18; Smist, C;ongre48 Ouersees 1M United Stal.eslnleUigerzee Community. (U) Kahn, "Big Ear or Big Brother?," 65; Smist, Congrtlll
Over,ec& the United ' Community, 10; Kahn, "Big 73.

SllIk. InteUigence

COMrnunity,73.

57. CU)Smist, Congrell Owr,ee. tlw United StaI.e. IrtWligenu Brother? ".72; Church Committee correspondence.

Ear or Big

58. (U) Smist, Congre •• Oller,eel the UniUd Slllk~ lnUlligence Community, 135: Gerald K. Haines, "The Pike Committee Inveltigations and the CIA, ••Stwdin inlnteUigence(l997>, 41:3, 54. 59. (U) Church Conuclttee correspondence;. House Committee on Intelligence - correspondence files. 1975, in NSA retired recorda 28792, 80·079. Haines, "The Pike Committee," 56.

IIMtBLE ViA ,*LeU'f

KEYH6LE C6Mflff eON'fROL S"j'STEMS JOIU'i'LY

TOP SeCRET UMBRA

114

DOCID: 523696

REF ID:A523696
TOP SECRET UMBRA

60. (U) Allen interview: House Committee on Intelligenc. - correspond.Dce me; Smist, CongrcruO_a U nitCIdStlJtell riUUigUlU CO"lItJUtlity, 175. 61. 62. 63. (U) HoWIeCommittee on InteUigenee - correspondence file. (U) Ibid. (UJlbid.

the

64. (U) Smist, Congru, Ouersees tke Utlited St4tu Intel~e1lCe CommUllily, 185; The actual damage to aational security occasioned by the "four little worda" was surely less than Wallc~imed by the administration and, anyway, Henry Kiaainger had already leaked the wording of the report to a journalist. Thus the Pike Committee was not the ·firstto leak. See Haines, ~he Pike Committee Investigations." 65. 66. 67. 68. (U) House Committee on Intelligence - correspondence file. (UJIbid. (UJlbid.
(U) Ibid.

69. <l]}NSAArchives, 28795, 80-079.
70. (U) Ibid.

71.

(U) Gerald R. Fo~ Library, NSF, in CCH Series XVI.H., AbJug Committee."
ft

72. <Ul Huston interview; NSA retired raeords, 28795, 80-079. 73.
74. 75.
(u) NSA retired

recorda 28795, 80·079.

(U}CCH Series VU.I.1.2. <U}Ibid.

,

.

76.

<U) CCH Seriea 1..1.I .2; DDlR mes, 96026, 001 10, "Director's correspondence. Nov 75-Dec 76."

Withheld from public release Pub. L. 86-36

77. (Ull

i"Projeet Azorian - Th.eStory oCtile Hugh.es Glomar Expiorer:

StudUs in Tlltellig,fICCI, FaUI985; Interview, David Boak, by TomJohriaoD,11 May 1998.

78.
79.

(U)NeUl,/Mc.,

October31,1977: WcuMngtonPOBt,JlllIe6, 1977.

(U) Interview, Michael A. Smith, by Tom Johnson.andlL...-

---J1 September

8,1997, OH 14-97,

NSA.
. SO. (U) Nelli York Time" September 4,1977; Memo for the Record. Lt. Gen. Low A1len,tO March 1977, in CCH

Seri •• XII.D. 81.
(U) Smith

interview: Carter Library, NSF, in CCH Series XVI.I., "Koreagate."

82. (U)FIlCUon Fik, WeeJdy World NewlJDi.rest,June 11.1977.441,688; Carter Library, NSF, undated 1978 memo to Brzezinski (signature illegible). in CCH SerieaXVI, "Koreagate";New York Time6,September 4.1977. 83.. 84.
(U)

Facts on

ru« 44t', 688.

(U)Nelll

Yori Tim", October 20.1977.

85. (U) [Sipature iI1egibl~1 Memoto Brr.ezinski, unknown date in 1978. in Carter Library NSF. in CCH Series XVI.I., "Koreagate." 86. (U) Andrew, For
tIN

PruUknl', Eyu Only, 416.

87. <U) Ray S. Cline, The CTA Utlder Reagan, Buah and Cauy(WashiDgton, D.C.: Acropolia Books, 1981).

HAliBbE 'RA 'fIrLlilN'i' KEYIIOUlOOMUiT OOmReb SY-ErFSMS .JOINTLV

115

tOP SiCRE:r l:IMBRA

OOClO: 523696
TQP SECRET I:JMBR:A

REF IO:A523696
CRYPTOLOGlC QUARTERLY

Ip·L. 86-36 I
correspondence; CCH Series interview.

88. 89.

(U) Chureh Committee <U) Ford Library,

correspondenc:e. Cburch Committee

NSF, in CCH Series XVI.H .• "Legal";

Xl1.H.57.4.
~. (u) CCH Series XII.D., ~EO 11095";

Xn.H.S7.4,

Drake memo to IC staff, 9 August 1976; Allen

Ford Library, 91.

NSF, in CCH Series XVl.H .• "Legal," committee hearings. Vol. 5.81. in NSA retired records 28791·2, 80·079; House Committee on

(U) Church

Intelligence 92.

- correspondence

file. 1975, in NSA retired recorda 28792. 80·079.
bOl[

(U) DDlR file •• 96026.

13, "1974."

Ford Library,

NSF. "Wiretap/domestic

collection."

in CCH Series

XVI.
93. (U) Edwin Moise, Tonkin Gulf aild eM Escalation of the Vietnclm War (Chapel Press, 1996), 17.18; Carter Library,l the Inte/ligfnce HUl: .University of North Withheld from

Carolina 94. 95. 96.

Init interview.

June 15, 1994.

CU) Smist, Congre" Ouerue, (U)lbid.,5. .(U) Tbe Gulf of Tonkin.

Community. 4.

public release Pub. L. 86-36

the 1974 Incidents:

Hearing

Before the Committee

on Foreign

Relations.

90th

Cong., 2nd sesa., Feb. 20,1968. 97. 98.

CU) Cline. The CTAUnder Reagan;279.
(U) NSA Archives. ace nr 36740, CBPJ 47, Deeley testimony
(V) NSA Archives,

before Congress, September

27,1985.

99.

ace nr 42764, HOa·OSOl·4. Oversight." Yale University

100. (U) Carter 101. (U)John Press, 1972).

Library. NSF, in CCH Series XVI.I., "Intelligence

C. Masterman,

TIuDouble Cross System in the Warofl939 to 1945 (New Haven:
.

102. (U) Gustave .103. (U) Frederick

Bertrand,

Enigma: Tiu GreatelltEnignw oftlu War, 1939-1945 (Paris: PIon, 1913) . Tiu UltrtJ Secret (New York: Harper and Row, 1974). TM Hut

Winterbotham,

104. (U) Peter CalvQCOressi, Top Secret Ultnz (NeW York: Pantheon Books, 1980); Gordon Welchman, Siz SfDry: Brtaki1l8 the Enigm4 Coda (New York: McGraw-Hill. 1982).

105.

(U) Francis

H. Hinsley and CAG. Simkins, Brituh1ntellige~ Press, 1m).

in the Second Wor14War,3 vols (New York:
\ Farrar,

Cambridge

University

106. (U) Ronald Lewin, TM American Magic: Codes. Ciphers, a1Id the Defeat of Ja~n. (New York:

Straus and GiroUJ:, 1982>; See also Edward Van Der &hoar, Deadly Magic: A PereoMi Account ofCornmunication

Intelligence in World War IT in the Pacific (New York: Scribner, 1978) and Wilfred J. Holmes, Double·Edged Secretl: U.S. Ncuy TntelligeMe Operations in tlu Pacific During World War II (Annapolis: Navallnatitute Press,
1979). 107 . (U) Brady interview. 108. (U)NSA Archives,ace nr31218, CBOE 67. An~uitant (REA) Program in NSA: An Evaluation oCthe Agency ~papers. Boardman, by Robert D. Farley,

Withheld from public release Pub. L. 86-36

109. (U~
Archives, 110. (U~ 17 January
R

I ~e
me

Reemployed

undated paper in CCH records collection; DDIR files. 96026, box 2J Reemployed Annuitant Program"; Interview, Norman

1986,OH3·S6.NSA. interview;

111. (U> Boardman

tt1doff)U;

\i'fA 'f1\tEN'f

KE't'1l6LE COMIN'f CeN'fROL 116

S¥8TSMS

d91NTLY

Tap SECRET l:lMBRA

DOClD: 523696

REF lD:A523696
TOP SECRET (;IMIM

(U) Chapter 17 The New Targets and Techniques
(S-CCO) The demise of the Southeast Asia problem caused a revolution in SIGINT targeting. In many ways, though, it was no revolution at all, because the new focus was simply an old problem - the Soviet Union. In 1970, wheri Vietnamization was young, the Soviet Union occupied only 44 percent of NSA's attention. Five years later it had climbed back up to almost 60 percent and stayed the~e through the decade. Of the non-Soviet targets, only ILC increased in strength, from 5 percent to 10 percent. All the rest stayed stationary or declined.'

(U) STRATEGIC ARMS LIMITATION (U) History. shows that many presidents who have been given credit for starting something actually did not. This was the case with the negotiation of strategic arms limitations with the Soviets. President Lyndon Johnson, rather than Richard Nixon, initiated negotiations in 1967. At the time, Secretary of State Dean Rusk predicted that it would become "history's longest permanent floating crap game." ~ He was very nearly right. . (U) The Warsaw Pact invasion of Czechoslovakia in 1968 brought the abortive Johnson negotiations to an early and abrupt end. But Richard Nixon, hoping for some real departures in the foreign affairs field, got them started again. His new foreign policy ombudsman, National Security Advisor Henry Kissinger, contacted the Soviet ambassador to Washington, Anatoly Dobrynin, and they agreed to meetings in Helsinki. The "crap game" then floated to Vienna' and finally to Geneva, where it settled for the duration of the Cold War. Negotiations survived the bombing of Hanoi, the Watergate erisis, and the invasion of Afghanistan in 1979.3 (U) In May 1972 the protracted negotiations produced the first Strategic Arms Limitation Treaty, called SALT I. The treaty had two parts. a. Part 1 was defensive. The two sides agreed to limit their antiballistic missile forces to two locations. Each side was permitted to defend its capital city with defensive missiles, plus one other site, which would be a single cluster of silo-based launchers. This part oft-he treaty was of unlimited duration, to be reviewed every five years. b. Part 2 was offensive. It froze the silo-based missiles and submarine-launched ballistic missiles at their current (1972) level for five years (until October 1977). Since the Soviets would not admit what total number they possessed, the treaty did not express any numerical figures. American intelligence estimated that they posses