Home Page


WWII Era ('40s)
Korean Era ('50s)
Vietnam Era ('60s)
General Officers


OCS Association
OCS Notices
OCS Newsletter
Army News
Class Coordinators
Reunion Info
Other Links


Chief Locator
Web Submissions


Veterans' Salutes
Freedom Park
Brief Histories
Scrap Book
Chat Rooms
Charity Efforts



The Signal Corps During The Vietnam War

-  -  -  -  -  -  -

- This is Part III in a Three Part Series -

This article originally published on our Home Page in February 2012

Signal Corps During The Vietnam War

[Please Note: Any opinions expressed in this article are those of the author and do not necessarily represent the U.S. Army Signal Corps OCS Association]

This article is the third in a series regarding the Signal Corps and its path of evolution from WWII through Korea to Vietnam. The first two, The Signal Corps During World War II, and The Signal Corps During The Korean War, can be found on our Brief Histories page. There is a quick link to them at the bottom of this page.

Civil War Signal Corps balloonIn each article we have tried to take a 50,000 foot view of what the Signal Corps did during these wars, and determine from such observation how it affected the Corps’ evolution from the simple mission it had when it first came to prominence during the Civil War, to the complex mission structure it holds today. If one looks back on those early days and considers that at the time of the Civil War the Signal Corps’ primary task was simply to observe the enemy (usually from a hot air balloon), report on its activities, and deliver messages via pigeons or signal flags, one can see that today its mission to do everything from manage strategic DOA and DOD assets to being responsible for…

• Automation, communication, electronics and network planning, design, engineering, evaluation, management, installation, operation, logistical support, and maintenance of signal equipment and systems; to

Advising commanders, directors, and staff on command and control signal requirements, capabilities, and operations, including com­puter systems, data management, signals intelligence, signals monitoring, and network operation; to

Developing requirements for the design and implementation of local, regional and global data, mobile, and fixed communications systems and networks; as well as

Establishing, preparing, coordinating and directing programs, projects and activities engaged in unit level supply, logistics, maintenance, and life-cycle management of worldwide signal materiel; to

Integrating tactical, strategic and sustaining base communication, information processing, and management systems into a seamless global information network able to support knowledge dominance for the Army as well as joint and coalition operations; to

Civil War Signal Corps towerDirecting and controlling of the units and activities involved with the application of electrical, electronics, and systems engineering and management principles in the design, test acceptance, installation, operation, and maintenance of signal systems, equipment, databases, networks, and facilities; to more esoteric activities such as

Operating photo and video service undertakings that run the gamut from documenting combat activities to archiving the same, performing radio, data and other signal intelligence functions, to

Developing and implementing radio and radar countermeasures, establishing airway communications systems; and of course

Participating in all manner of combat activities from support of joint-assault signal operations through to the most simple but critical defense of individual signal sites…

we can see that much has changed in the Signal Corps.

The question we have been trying to answer through these three articles has been how did these changes come about and why. The answer we found is that the real time pressures of war, followed (in most cases) by government mismanagement of military budgets between wars, caused these changes.

In great part, the bulk of the changes in the Signal Corps’s approach to its duties came about during WWII, Korea and Vietnam... and the times between them. It’s because of this that our focus over the past two articles has been on the Signal Corps during the first two of these wars. In this article we finish our series by looking at how the Vietnam War forced further change upon the Signal Corps.

Civil War Signal Corps telegraphLooking back over the prior two pieces, we can see that one of the key lessons we learned in looking at the Signal Corps during WWII and Korea is that unlike most branches of service where the task is singular, comprising little more than one of giving combat in a manner that contributes to winning a war, the Signal Corps has evolved during these periods to fulfill two roles. In military speak, it could be said that while other branches of service focus narrowly and almost exclusively on their task at the operational level of war, the Signal Corps found that in order to meet its ever evolving mission, it needed to expand its operational concept to take in not only the application of military art and science to areas within the operational level of war, but also external to it.

In this regard, the first role the Signal Corps carries out obviously relates to being a partner war fighter, along with all of the other branches of the U.S. military. Considering that over the past 60 years the Signal Corps has been first a part of the combat arms, then not, and then later included again, being a partner war fighter has not always been easy. Whether formally a partner war fighter or not, in this role the Signal Corps, like its sister branches, puts its men on the line—engaging the enemy where and when needed, as it goes about its task of providing any and all support required to deliver the communication capabilities essential to the other branches sharing the combat field with it.

The second, as the reader can intuit from the list above, relates to providing the kind, type, and quantity of communication and signal capabilities necessitated by the nature and characteristics of the war, situation, or conflict underway. And while the glory in what the Signal Corps does may rest within the former role of a war fighter, it is the work done within this latter category that earns the Signal Corps its stripes.

Two simultaneous missions: that of a war fighter, and that of the provider of any and all manner of communication—or as we know it today, Information Technology (IT), Information and Communication Technology (ICT), Information Systems (IS), Information Management (IM), Knowledge Management (KM), Technical Science Management and Application (TSMA), and Data Management (DM)—as may be required by the nature and characteristics of the conflict in question.

Without these two tasks being successfully performed by the Signal Corps, combatants from the other branches would find themselves existing and fighting within a vacuum... a vacuum void of information about the enemy, his position, intentions, status, and pattern of methodological behavior. For if the truth be told, these latter five elements form the determinants of war in the modern age, and it is the Signal Corps that is first and foremost responsible for assisting in their identification, documentation, and communication to the rest of the military.

One can see then that as both the foundation and the glue that makes possible an effective response to the numerous war activities the U.S. military gets involved in, by enumerating to its sister branches an enemy’s position, intention, status, and pattern of methodological behavior, the Signal Corps is the enabler that allows the sister services to act with both precision and objective intent. In other words, because of the information communicated by the Signal Corps, its sister services, such as the Infantry, are able to use this knowledge to their collective advantage. In today’s modern world, it would be said that the Signal Corps enables the Infantry to turn its troops into knowledge workers.[1] And yet while this seems self obvious to us today, the reader should recognize that the calling to serve this purpose is not only a far cry from the job the Signal Corps originally set out to do when it was founded, it is as equally far a cry from that which it did during the second world war and Korea. The Signal Corps has evolved. As pundits would say today, it’s not your father’s Signal Corps anymore.Today's Knowledge Worker 

As we saw in the previous two articles, WWII challenged the Signal Corps to develop several capabilities that it did not previously have, while the Korean War helped the Signal Corps to figure out how to better deliver these capabilities. One of the more important of these capabilities involved expanding the role of the Signal Corps to support a method of war-fighting originally developed by Gen. Ulysses S. Grant, but not fully applied since then until WWII. A basic tenet of how America fights wars even today, Grant’s doctrine revolved around the emphasized, overwhelming, and continued application of military force directly against the enemy army, as well as indirectly against the enemy's civilian population (read: the civilian-industrial sector), to prevent the civilian sector from acquiring the resources (including the availability of civilian manpower itself) needed to support the military. With no insult meant, Colin Powell’s famous Powell Doctrine of the application of overwhelming force in time of war is simply a restatement of Grant’s original war strategy, and a not very original one at that.[2]

In applying force against an enemy’s Army, clearly, a key part of this is knowing where the enemy is and what his intentions are. In depriving the military and its supporting civilian population of their ability to provide resources to the enemy, the important part is identifying both what resources the military needs, as well as who is providing them to the military.

In both of these instances, a quick reflection will show that it is the Signal Corps that has, time and again, stepped forward to help solve the riddle of how to identify the who, what, where, when, and why embedded in matters of war, and communicate this information to the troops in the field. In the first instance, the Signal Corps’ development of RADAR serves to make this point. Whether it was the kind of RADAR that first detected Japanese Zeros approaching Hawaii on December 7, 1941, or the kind of X Band RADAR that was first used to locate mortars, the objective was the same: identify the enemy’s intentions, locate them, and distribute this information to those in the field.

In the second case it was the Signal Corps’ development of signals intelligence that led the way towards identifying the tie in between military resources and the civilian counterparts that provided them. Signals Intelligence, combined with the Signal Corps’ development of encrypted as well as burst radio communication, allowed U.S. saboteurs to step in and deny these resources to the enemy’s military.

World War II then served as a crucible in helping mold from the tailings of the old-school military of World War I a modern Signal Corps able to apply newer and rapidly evolving forms of technology to the purpose at hand. What it failed to do however was help the Signal Corps develop an organizational structure able to anticipate and respond to the sort of quickly changing battlefield conditions that were looming (as WWII was brought to a close) just over the horizon.

Korea did that.

As we saw in the last article, the Korean War brought home to roost the necessity for the Signal Corps to wrench itself from its stayed approach to handling operational situations, instead creating a means to transform itself on the fly… applying in each and every case that it was presented with the kind of American leadership, creativity, and problem solving skills that are required if one is to succeed in a fluid situation. As the Signal Corps learned then, key to doing this was being able to distinguish where rapid transformational abilities were needed and should be allowed, versus those situations where transformational pressures should be resisted and things forced to continue to be done “by the book.”

Remarkably, the Signal Corps succeeded in this vetting conundrum. It succeeded by unknowingly becoming the first military institution to define and apply process management to its mission. A term that came into vogue only in the early 1980s, the Signal Corps during the Korean War was one of the first to define this approach to task management, becoming its own internal proponent of the use of what is today known by the terms TQM, Six Sigma, QMS, process management, and a dozen others. With focus and purpose but unmindful that it was charting new territory, the Signal Corps wrote process management dictums into its SOPs even as the Korean War unfolded.

As to why this was necessary, battlefield conditions of the Korean War presented the Signal Corps with the need to integrate in real time its ability to find, trap and analyze exocentric knowledge of the enemy's intentions and activities... from all available sources and services... in order to build ever quicker, faster, cheaper, better endocentric means of analyzing and sharing this information with the combat arms most in need of it. Process management, when used as a means of solving real time war problems, is ideal for this purpose as it helps strain out nonstandard data points in the collection and analysis effort. For the Signal Corps then, the Korean War proved to be another important period of transition, in both how it selected, trained, organized, and managed its personnel, as well as how it managed itself in performing its core tasks of analyzing and communicating. In all of this, developing multiple technological means to address each communication need that appeared inadvertently led to what was likely the first ever effective application of process management techniques in a hot war environment.

Obama's military strategyBy the end of the Korean War the Signal Corps had found itself in a new place in military society. By 1960 the Signal Corps was the Army's third largest branch, comprising about seven percent of its strength. In 1961 the Army redesignated the Signal Corps as a combat arm again, a privilege it lost at the end of the second world war, while at the same time keeping its designation as a technical service arm.

Unfortunately, as with the end of WWII and every war that preceded it, with the suspension of combat operations in Korea America’s federal government set about the task of looting the military one more time, in a mad rush to reorganize it, ostensibly to “learn from the lessons of Korea.” Why the U.S. government continues with this charade of cutting military expenditures once a war has ended (under the pretense of making the military more efficient or effective), one can only imagine. Yet, like clockwork, as soon as a war has ended, government leaders set about wielding their ax to the military… as though the U.S. will never again fight another war. 

One can see it happening today. With President Obama announcing in January 2012 his plans to reorganize the post-Iraq, post-Afghanistan U.S. military, his efforts are at best feckless and at worst one of the most irresponsible things a president can do. Cutting the size of the military under the guise of making it more suitable to large scale naval engagements between nuclear powers, during a time of increasing tension between the U.S. and any number of countries, is in total contravention of the very reason for a people to have a government in the first place.[3] In this Editor's view, Obama’s actions today are no less imprudent and irresponsible than those of the leaders who, after World War II, gutted America’s military to the point that it was unable to fight in Korea without mounting a draft and scavenging the whole of Japan for every piece of armament that could be found. Decorated with disingenuous statements about how his new changes will make the U.S. Army better structured to fight the future wars that Leon Panetta says are coming, President Obama should learn from what happened in Iraq when Rumsfeld’s lofty goal of developing a new, more nimble, smaller footprint military had to be shelved because… gosh, what a surprise… wars require overwhelming force to win.[4]

This digressive rant aside, the military at the conclusion of the Korean War, the Signal Corps included, was gutted one more time, in another round of post-combat capability reductions… of a type that had an impact on the upcoming war in Is Iran next?Vietnam. Fortunately, unlike when the U.S. military’s global communication network ACAN (Army Command and Administration Network) was rent asunder between the end of WWII and the beginning of the Korean War, post-Korea the newly named and established Defense Communications Agency decided to maintain the global communications network then in place, and even expand this worldwide, long-haul system to provide still greater, secure communications. Thus, for the first time in the nation’s history the president, the secretary of defense, the Joint Chiefs of Staff, government agencies, and the military services, not to mention America’s diplomats abroad, had access to a fully integrated, secure, global communication network at the time the Vietnam War got underway.

As to how this network got built, under orders from the DCA the Signal Corps set about integrating what was left of the former ACAN system with the new network elements put up during the Korean War, and any other odd long haul network pieces that could be found laying around, bringing them all together into an expanded global network. As a new network it was titled the Strategic Army Communications Network (STARCOM). Yet while this effort proceeded smoothly and the network came on line as required, a new wrinkle in how the Signal Corps did its job was sneaking slowly into the process… a wrinkle that would have a profound effect in a few years when the Vietnam War got underway… and an even more profound effect by the time this precedent percolated its way down to the War in Iraq.

What was that wrinkle? The answer was that it was the partial contracting of the task of building STARCOM to America’s industrial sector: the defense industry. While at the time and on the surface this new approach of contracting military work to civilian companies seemed risk averse and a smart way of getting around the effects of the downsizing of the military at the end of the Korean War, behind the scenes and underneath it all a dangerous precedent was being set that would live with the Signal Corps forever. Worse, this new precedent and approach would inexorably expand and extend itself across all of the branches of the military, down to today.

Civilian militar partnershipThus today, not only are many of the Signal Corps’ communication systems designed by civilian contractors, they are built and run by them too. Similarly, and perhaps with far greater consequence when it comes to protecting civilians, NGOs, and U.S. government agency members in war zones, the Infantry itself has been co-opted into this program, finding many of its traditional roles in combat zones supplanted by firms like Blackwater, Greystone, and the other shadow armies that operate within what is now known as the Privatized Military Industry. It boggles the mind: America’s Army fighting side by side with shadow armies hired by the government so that the size of the military can be kept small.

Beginning with the government’s decision to downsize the U.S. military after the Korean War, a cruel joke was played on both the military and America’s citizens. Under the rubric of saving money, downsizing, and realigning the Army to fight smaller more mobile wars (a claim, as we stated above, that is still raised today whenever Congress or the president sets about cutting the military’s budget), in the late 1950s to early 1960s the government set about transferring much of the Signal Corps’ role to civilian industry players.

Thank you Dwight David Eisenhower. Your fear of the military–industrial complex was heard well. Unfortunately, the solution to the problem you and the presidents who followed you put in place only served to turn it from being a problem of budget matters being driven by the military–industrial complex into one of budget matters being driven by the industrial–military complex. The same bedfellows, they just swapped places in bed.[5]

For the Signal Corps, this new partnership with industry proved a double edged sword. On the one hand, because of the long standing relationships that existed between the Army’s research facilities at Ft. Monmouth and civilian industry, the Signal Corps had a cordial, synergistic working relationship with the civilian guys, one of the benefits of which was nearly immediate access for development purposes to the very latest in cutting edge technology and products. On the other, the inroads civilian industry made into the actual running of Signal Corps facilities put an enormous strain on the type, quality and amount of manpower available to the Signal Corps itself. After all, if civilians could do the work, what was the purpose of recruiting soldiers and Officers into the Signal Corps? What was the purpose of having training schools to turn out troops qualified to hold the numerous MOSs (reduced to just 17 as of today) that had been defined? Of what need were Officers if the complement of enlisted men was being downsized? Strangely, no one seemed to stop and think of how this would all play out if more armed hostilities broke out. Would the civilians who were running Signal Corps facilities be expected to ship out and take up residence in a war zone if war broke out, to build, run and maintain the Signal Corps facilities needed there? Nah, surely not.

Dien Bien Phu - now it's America's turnMeanwhile, while the Signal Corps was evolving once again… this time learning to embrace a new working relationship with civilian contractors who were snaking their way ever deeper into the Signal Corps’ operations and management structure... on the other side of the world life in South East Asia was beginning to go belly up. As we all know today, the French suffered a humiliating defeat at Điện Biên Phủ, after which they promptly withdrew from Indochina and left the U.S. to deal with the mess they created.

The U.S., seemingly ever solicitous of the French, decided to keep the "advisory group" (already in Vietnam) in place when the French left, allegedly to help guide the South Vietnamese Army now that the French were no longer available to do the job. In this act the Signal Corps clearly enmeshed itself in the evolving drama; not just taking a role in a side show foreign engagement, but inadvertently helping to move the show over the next few years from the wings of the theater to center stage. The reason the Signal Corps found itself going along for the ride, with one hand on the steering wheel, was simple: the Vietnamese Army contained a Signal Corps, and therein existed a ready-made excuse for the U.S. government to maintain listening posts in Vietnam as well as send more advisors along.[6]

In the end then, at the onset of the Vietnam War the Signal corps found itself ostensibly teaching operational and logistical signal matters to the Vietnamese Signal Corps, while in reality it was using its presence in-country to listen in on regional communications. As modern day historians, what matters to us is not what the Signal Corps was doing, but recognition of the fact that the U.S. Signal Corps was one of the first elements, if not the very first of the U.S. military, to take up an active role in the Vietnam War. In particular, in a series of steps between 1954 and 1965 the Signal Corps brought in more and more advisers, to the extent that they were assigned even down to the divisional level and to each of the Vietnamese Army's military regions.

By 1963 (at the time of Kennedy’s assassination) the U.S. had more than 16,000 U.S. military advisors in South Vietnam, the bulk of which were either assigned to or an intrinsic part of the Signal Corps. Yet among all of them there were no staff level Signal Officers. Instead, country oversight was handled by the Signal staff at the Pacific Command in Hawaii. Interestingly, despite this management from afar approach, things got done. One of those things included a very serious effort on the part of in-country Signal Corps staff to send their South Vietnamese counterparts to signals training at Forts Monmouth. Taking a leaf out of the Signal Corps’ training book, a similar effort was undertaken in 1961 when the U.S. sent 400 Special Operations Forces (Green Beret) to South Vietnam to begin training local ARVN troops in how to conduct what was, for the first time, called a counterinsurgency war against the Communist guerrillas in South Vietnam. Historians should mark this point as the time from which the term counterinsurgency entered the American lexicon. Today one could almost say that Afghanistan, or perhaps Pakistan, is a synonym for counterinsurgency.

Vietnam memorialLooking back now on how the Vietnam War got started, and the role the Signal Corps played, it seems strange to admit that the war itself had no beginning. Is that possible? Can America really get itself into a war that cost it 58,272 KIA, 303,644 WIA, 1,687 MIA, and 866 POWs, but for which Congress and repeated presidents didn’t have the time or consideration for its military to sit down and declare war on the enemy? No formal beginning. No formal end. Is that really possible? One almost begs to ask: what has our country come to when our elected leaders spend so much effort dissembling the truth about the foreign policy they are setting... so that it is palatable to the country at large... that they don’t have the time to declare as a war an undertaking that America's youth die in by the bucketful?

Back in those early days, for those on the ground in Vietnam, with or without a formal beginning, things moved inexorably towards a hot war. Incrementally, in a series of steps between 1950 and 1965, America found itself engaging in combat with the North Vietnamese.

- - -

One would have thought that this slow march to war would have given the U.S. military plenty of time to prepare. But that wasn’t the case, although not through the military’s fault alone. To prepare for war requires three key things: knowledge that it will happen, plans as to how the war should be prosecuted, and the money to secure the resources to prosecute the war as planned. Without a formal declaration of war from Congress, the funds needed to expand the military to undertake and win this oncoming war simply did not exist. For the Signal Corps, this meant that as the French left the country and took along the American supplied signal equipment that had been given to them, there were no longer any systems in place with which to tie the country together… nor any money to acquire what was needed. So again, as in the run up to the Korean war, the Signal Corps found itself cannibalizing signal facilities around the world in an effort to build a communication network that could support armed conflict. And since Japan had already been stripped to support Korea, that left only Europe as a warehouse from which to pilfer signal equipment to build what was needed in South Vietnam.

For the South Vietnamese government in power, the precarious situation it was in only became more obvious—even if it could rein in the crony capitalism, elitism, and corruption that was endemic in the country, without a telecom and radio infrastructure with which to reach the populace, it was going to prove near impossible to rally the South Vietnamese people to a cause of war with the north. The fact was, the commercial communication networks built by the French lay in disrepair and ruin after years of inattentiveness, and the South Vietnamese military, having no communication network of its own to underwrite its own war effort, was certainly in no position to help the civilian government tie the country together. No civilian communication infrastructure, no military communication network to back it up, no military communication network to use for its own, and no training in the form of combined arms tactics required to make effective use of a battlefield network in real time combat, all meant that South Vietnam found itself in a real mess as armed conflict escalated in the early 60s.

The reader can understand then that with this scenario presenting itself it was only natural that the “wrinkle” discussed earlier would raise its ugly head again—this time as the only viable solution to the problem at hand.

And thus it happened; to make available and stand up an operational communication system to serve the civilian, military, and government needs of South Vietnam, the Signal Corps turned to and hired contractors to construct a regional in-country network. In simple English, the military downsizing that Congress and the president mandated on the U.S. Army at the end of the Korean War forced the Signal Corps to enter this new war with civilian contractors doing the better part of its job for them. And in short order this first step was followed with a similar outsourcing of the Signal Corp’s mission to contractors that designed, built, and in many cases operated parallel networks in Cambodia, Laos, and Thailand.

The first of these efforts occurred in May 1960, when Page Communications undertook to build what was called the Pacific Scatter System,[7] a telecom and data network designed to serve the Army by linking the Philippines with Hawaii. Following a course of 7,800 miles, it leapt along a chain of islands and countries stretching from the Philippines to Guam, Midway, and on to Hawaii. Linking South Vietnam to the Philippines was accomplished via a portion of the military’s Strategic Army Communications (STARCOM) network, which was terminated at Phu Lam (Phu Lam translates as Rich Forest), outside of Tan Son Nhut. To complete the track, Hawaii was linked to the U.S. at Davis, California, via circuits that were originally part of the old ACAN network… yet another ironic example of how a once important piece of the U.S.’s global communication network was taken out of service because Washington dictated that the military be cut back at the end of one war, only to find a few years later that the systems taken down had to be hastily rebuilt to serve again when the next war popped up.[8]

BACKPORCH - Original network design, 1965

- - -

With the Pacific Scatter System in place, in 1962 Page Communications was awarded another contract (by the Secretary of the Air Force), this time to install a network called BACKPORCH. BACKPORCH was eventually to be a 72 channel, AN/MRC-85 Tropospheric Scatter system, with terminal sites at Da Nang, Nha Trang, Phu Lam, Pleiku, Qui Nhon, and Ubon, Thailand. In its early days however it started out with AN/TRC-90 tropo euipment. Why did the Air Force contract for the system instead of the Signal Corps? Again, the primary reason was that the Signal Corps' budget had been cut back so much that it didn't have the funds to build the networks needed to handle combat operations on the ground. Instead, equipment had to be scavenged from other places and other services.

Cost wise, for the Air Force, it was a good decision, as the BACKPORCH contract was for only $12 million and that included Page operating and maintaining the system for a year. For those Signal Officers who served in Vietnam but may not remember BACKPORCH, while you may not remember it well, you surely saw it, as BACKPORCH was the beast that caused all of those huge 60 foot “billboard” antennas to be placed all over the cities listed above, the mountain tops around them, and the rest of the country too. Perhaps the most important part of Vietnam’s overall battlefield, combat, area, and base camp communication net, BACKPORCH made it possible for standard Army short range multichannel radios to be connected literally from any active field combat area into the entire country wide network, covering all of the I, II, III and IV Corps Tactical Areas.

Unfortunately, while all of this was going on the Philippines-to-Vietnam portion of the STARCOM network set up by Page was proving to be a problem. STARCOM's radio circuits suffered from constantly fading signals, proving unreliable in the steamy, stifling environment of South Vietnam’s tropics. Needing something more dependable, the Signal Corps requested approval to supplement STARCOM with an underwater military cable. In 1962 the Joint Chiefs finally approved this request, with the construction of the new WETWASH cable system… again contracted to Page Communications, since the DOD had long since stripped the Signal Corps of its cable laying ships... being started shortly thereafter.

WETWASH linked South Vietnam to Clark Airbase, at Luzon, in the Philippines. However, as it would take time to complete WETWASH, an alternative 60 channel tropo scatter bridging link was requested by the Signal Corps, to try and connect South Vietnam to the rest of the world via still another route. This bridging link was designed and built by Philco, and linked Vietnam at Phu Lam to Bang Ping (near Bangkok), Thailand. At the Thailand end it was integrated with additional radio links to take it from Bang Ping to both Pakistan and Okinawa, and then on to the rest of the world. BACKPORCH Network

Sadly, the Philco tropospheric scatter bridge proved to be as unreliable as Page’s STARCOM link, to the point that the Signal Corps finally decided to take things into its own hands by replacing Philco’s work with a new link designed and installed by the 1st Signal Brigade, instead of yet another outside contractor. This link reconfigured the Philco approach by relocating the Bangkok terminal to Green Hill in Thailand, and the Saigon terminal to Vung Tau Hill. Not too strangely, it worked, perhaps because it was designed by junior level Signal Corps Officers with far less training and far less compensation than the well paid engineers at Page. Either way, when the Signal Corps took matters into its own hands, the work got done, and what was done worked. From these new locations the circuits were then brought into both Bangkok and Saigon via microwave links.[9]

In the end, while it took time for the lesson to be learned, the U.S. military got the message that while outsourcing to civilian contractors might be expeditious, if you wanted to get the job done to the point that the communication links actually worked under rigorous combat and environmental conditions, what the Signal Corps had to do was do it itself. Unfortunately, and here again we are beating a dead horse, while the Signal Corps seemed to have learned this lesson, Congress and the president seemed not to, as shortly after the Vietnam War was over, budgets cuts again tore into the Signal Corps’ ability to maintain and man the global network it had so rigorously built.

By 1965 all of the kinks had been worked out of the Vietnam-war-zone-to-the-rest-of-the-world communication system, with the troops in the field finally being able to depend on multichannel radio relay equipment to complete what were in fact intricate interconnections, tying together all manner of VHF, UHF, microwave, tropospheric, ionospheric, satellite, and undersea cables. Unlike in previous wars, the network put in place allowed combat commanders on the ground to have instant access from the field through standard field radios to anyone they might wish to talk to, from FACs in the field, to Arty at fire support bases, to Westmoreland’s HQ, any pilot sitting on any aircraft carrier in the entire Navy, all the way up to the president himself. With an effective combat communication network now in place, unit mobility greatly improved, allowing commanders to finally let loose with their Hueys, using them to full advantage in taking the fight to the Viet Cong and NVA. Perhaps best of all, having moved far beyond the days of field commanders having to depend on strung wire to communicate with, commanders could now move and shoot at any time they wanted, without losing communications for a minute, even while they were en route to their new positions.

Interestingly, a key part of the mobile capabilities the Signal Corps delivered came about not because of the fancy long haul links running over the tropo and other networks set up by the outside contractors, but by a simple airborne FM relay system set up by Signal people in units such as the 13th Signal Battalion. For example, in trying to meet the needs of the 1st Cavalry Division, their client, the Signal guys in the 13th Signal Battalion came up with the idea of mounting radios in fixed-wing aircraft and then circling those aircraft at 10,000 feet over the 1st Cav’s daily battle area. By doing this the 13th was able to set up a method for retransmitting messages between widely dispersed combat units on the ground. Through this simple expediency the LOS limits and electro-magnetic absorption effects of the triple canopy jungles on PRC-25s  could be overcome, effectively extending the PRC-25’s range from around 5 miles to over 60. When the word got out as to what the 13th Signal Battalion had done, Signal units throughout Vietnam found themselves beseeched with similar requests by battalion and brigade commanders to help them set up their own helicopter borne command centers, equipped with radio consoles and no-nonsense solutions that would make a ham operator cry with envy.

Overall then, as 1965 unfolded commanders found that the number of problems affecting the Signal Corps’ ability to stand up a solid communication network had been dramatically reduced, if not completely overcome. Sure, there were still problems with not enough circuits, but this was more a matter of a lack of available resources stemming from the downsizing the Signal Corps took after the Korean War than it was due to technical problems. Overall, traffic was flowing smoothly, albeit a backlog was beginning to develop.

IWCS - Phase IIITo make sure the backlog did not get out of hand, plans were made to design and build a base theater network. The network would involve a wide array of routing and transmission methods that would reinforce the simple approaches used in cases like the retransmitting aircraft mentioned above, with a more modern and well integrated battlefield communication network. Known as the Integrated Wideband Communications System (IWCS), the design was to blend automatic telephone, teletype, and data systems with coastal undersea cables, and integrate all of these with the BACKPORCH and WETWASH systems. The IWCS, designed to serve the Vietnam battle space as its first priority, would then be integrated into the global Defense Communication System.

With IWCS in place, things for the Signal Corps became much less dramatic as the war moved forward. By the time of Tet, everyone knew their place, signal links were humming, and work progressed almost without concern as Signal Corps troops went about their daily jobs. Even Tet turned out to be merely a speed bump to the Signal Corps’ daily activities, for while 10 of the IWCS signal sites were hit during Tet the damage they suffered barely affected the “up time” of the network.

Yes, problems with personnel did exist… such as a shortage of trained operators to run tropospheric scatter terminals. However, in most cases solutions could be found. Of interest again is that even in these cases where problems did exist—one more time everyone—they existed because of the downsizing of the Signal Corps at the end of the Korean War.

For example, while it was understandable that signal schools could not produce qualified graduates fast enough, what wasn’t understandable was why the Signal Corps could not reassign already qualified personnel sprinkled around the world, from where they were to where they were needed in Vietnam. The reason was that during the post-Korean War cuts, the section of the Signal Corps that tracked personnel assignments in relation to their MOS qualifications had been cut. Thus, since the records of previously trained personnel no longer existed in a format where they could be cross linked to current assignments, it was impossible to either recall those people who had left the military to active duty, or find and reassign them if they were still on active duty.

Adding to this difficulty, regulations at the time prohibited the involuntary reassignment of military personnel overseas for two years. And while this was eventually reduced to 9 months for specific skills, the only viable solution to the problem was to make it worthwhile for an enlisted man to “re-up” when his tour of duty was over. Thus, many a Signal Corps EM found himself with a little extra cash in his pocket as the DOD offered ever increasing pay and reenlistment bonuses to both recruit and retain the skilled, combat hardened soldiers the Signal Corps needed.

- - -

In early 1966 General Westmoreland created what was called the I and II Field Forces. These corps-sized headquarters were assigned the task of overseeing operations in the II and III Corps Tactical Zones [Editor’s Note: why the numbers don’t match, we don’t know… but they don’t]. At the time, II and III Corps Tactical Zones were seeing the heaviest fighting and were in the greatest need of additional oversight.

Each of these Field Forces was assigned a Signal Officer, along with a Signal Battalion. To make certain that these two Signal Battalions shared information among themselves and with the rest of the Signal Corps’ combat area commanders, as well as coordinate and improve their own command and control of signal operations based on shared knowledge of what was happening in other tactical zones, in the spring of 1966 the Signal Corps created the 1st Signal Brigade.

Once created, the 1st Signal Brigade grew like Topsy.

1st Signal BrigadeThis new command was the first TOE brigade in the Signal Corps’ history, holding within its arms all of the signal units in Vietnam except those that were intrinsic and organic to tactical units. As a unit, the 1st Signal Brigade consolidated all Signal units above the Field Force level into one command, essentially merging both tactical and strategic communication functions throughout the entire Vietnam combat area.

To make sure that each of the Tactical Zones were properly managed, the 2nd Signal Group was made subordinate to the 1st Signal Brigade and given command over Signal operations in the III and IV Corps Tactical Zones. Offsetting this, the 21st Signal Group was given authority over Signal operations in the I and II Corps Tactical Zones. Finally, in early 1967 the 160th Signal Group was added as a commanding element with responsibilities for Signal operations in the Saigon and Long Binh areas, with the 29th Signal Group in Thailand also being added to the 1st Signal Brigade, with responsibility for the flow of communication between the two countries.

By the end of 1967 the United States had committed nearly 500,000 troops to the Vietnam War. The Army provided about two-thirds of the total, including seven divisions and two separate brigades. The 1st Signal Brigade itself was comprised of twenty-one battalions, organized into five groups. Its strength peaked at 23,000 men in 1968, the majority of which served in one or another of the roughly 200 key signal sites spread throughout South Vietnam. Yet while they were posted to Signal Sites, that didn’t mean they weren’t in the thick of the fighting. Signal Sites, while strategically located for the purpose of providing communication, were little more than forward bases stuck in the middle of enemy territory. By the summer of 1968 enemy attacks on signal positions numbered an average of eighty per month.

In addition to American forces, Australia, New Zealand, South Korea and Thailand all contributed units, as of course did South Vietnam, bringing the total manpower engaged to well over a million. Unlike the situation during the earlier Korean War, however, the U.S. commander had no command authority over any of these friendly troops. In researching this article we attempted to see if a study had ever been done on whether the American commander’s lack of command authority over the entire complement of troops available in Vietnam had any impact on the outcome of the war. We were unable to find any such study… but one wonders what the result would be today if such had been the case.

Most readers know how the Vietnam War ended, so we won’t pursue here either how it was conducted nor the interplay of politics in its ending. If justice is to be done to these topics, space far larger than that available on this website would be needed. Instead, we will keep our focus on the Signal Corps.

Looking at just the Signal Corp side of things, as the war wound down the size of the 1st Signal Brigade decreased in lock step with the political mandates dictating the rate and type of withdrawal. By 1972 the 1st Signal Brigade’s strength stood at less than 2,500 men. On 7 November 1972 the brigade headquarters left Vietnam and transferred its colors to Korea. The 39th Signal Battalion, the first Signal unit to arrive in Vietnam, became the last to leave. Fittingly, as its final wartime mission the battalion supported the international peacekeeping force that monitored the troop withdrawal and prisoner exchange. The unit departed Vietnam on 15 March 1973, almost eleven years to the day after its first elements had arrived.[10]

By the summer of 1973 the United States had completed the withdrawal of its combat troops.

Lessons Learned

In terms of lessons learned and changes suffered, clearly the Signal Corps that fought the Vietnam War was unlike the Signal Corps that fought the Korean War, or certainly the war before that. The only thing all of these had in common was that they all shared the same name. Except for the most minimalist stating of its mission, everything about it, including its mission, had changed. As an example, the Chief Signal Officer had disappeared from the organizational chart and been replaced by a Chief of Communications-Electronics. While a nice modern title befitting of a civilian executive, the position held absolutely no operational responsibilities.

Communication In WarLater, as war activities increased, this move to strip the Signal Corps of operational command over its own people and assets proved to be a disaster. In simple English, the abolition of the Chief Signal Officer's position (in 1964) left the Signal Corps chain of command in near total disarray. The lack of coordination that ensued forced General Westmore-land (in July 1965) to disband the U.S. Army Support Command, Vietnam, (former-ly the U.S. Army Support Group, Vietnam), and create the U.S. Army, Vietnam (USARV).

USARV’s purpose was to try and undo the damage that had been done by McNamara and his boys working with Congress to gut the military, in order to save money. It wasn’t the money part that worried Westmoreland, it was the lack of command control over the people deciding the tactics and fighting the war.

Westmoreland, by creating USARV, was able to once again take under the military’s wing operational control over all military people in Vietnam (except for the advisers), and their assets. The reader should note however that while Westy’s little trick helped things in Vietnam, it did nothing to change how the military worked across the rest of the world. Even so, from the Signal Corp’s viewpoint, thanks to Westmoreland, the Signal Officer on the USARV staff was once again able to take responsibility for and command over the Army's tactical signal operations and personnel, with long-haul communications coming under the purview of the Strategic Communications Command.

In addition to fighting to gain control over its own tactics and people, the Signal Corps had to relearn in Vietnam what it was like to fight a war that did not conform with what Army planners had thought the next war would be like when the Korean War ended. Strangely, even though the use of nuclear weapons was approached and retreated from time and again in Korea, with nuclear weapons never being used, at the end of the war Army planners were convinced that in the post-Korean War period the next conflict would be a nuclear one. This of course meant one would be fighting across a nuclear battlefield. And this of course meant that the Army needed to be reorganized yet again, to accommodate what was thought would be highly fluid, front-line centric, combat conditions in the midst of nuclear fallout.

Accordingly, concepts for combat operations, like the "Pentomic Division", ROCID, and ROAD were developed and put in place. These forms of tactics were designed to support a fluid, aggressive, conforming type of combat. To support this, the Army equipped itself with tech-nology that matched the tactics. Thus, as the Vietnam War got underway the Army found itself with equipment like the Davy Crockett rocket, an ingenious bit of portable armament sporting an atomic warheadclearly something useful on the plains outside of Moscow, but totally useless in the jungles outside of Dalat. Instead, it became clear that since the troops in Vietnam faced guerrilla warfare in jungles and rice paddies, what was needed were weapons suitable to this environment, as well as tactics that matched both the environment and the enemy. In particular, since the enemy proved slippery, what was needed was a full reorganization to allow the Army to mount expeditions from fixed bases to both engage and fix the enemy.

For the Signal Corps this meant quickly developing a doctrine that supported rapid responses by the combat arms, via the use of communication equipment that, in the field, was small, lightweight, portable, and reliable, but back on base was supported by fixed-base communication via multiple forms of transmission, usually depending on large antennas and heavy equipment. Flexibility in deployment and mission support became the first of many lessons that the Signal Corps took from Vietnam.

Adding to this, since a typical divisional signal battalion in Vietnam ended up covering areas of 3,000 to 5,000 square miles, compared to the 200 to 300 miles that was expected in a ROAD type of conventional-cum-nuclear war, Signal units found themselves jostling to come up with the equipment needed. In this case, the problem was not lack of funds or poor planning, it was that the TOE allocation a Signal unit had was based on a design intended to provide for a much smaller footprint containing far fewer combat troops in need of support. When one looks back today on Rumsfeld's plan to slim down the Army and make it much more mobile and responsive to asymmetric warfare, or that of the Obama Administration today, this problem is the first one that comes to mind. Slimming down the complement of personnel is one thing, but are you then going to slim down the amount and type of equipment available too? If so, what will you do when you find yourself dealing with a 5,000 square mile combat area that, while it does not need a lot of manpower to keep it operational from a technical perspective, needs tons of equipment widely dispersed and aggressively defended by lots of people in order to keep it up and running?

Army budget cutsAn example of this can be seen in the 518th Signal Company, a unit this author was assigned to as X.O. towards the end of his tour of duty in Vietnam. The 518th, a company formed to provide tropo and microwave communication throughout the entire III and IV Corps Tactical Zones, was supporting some 14 microwave sites, 4 tropo sites, and a ton of smaller VHF and UHF sites when I was there. Manpower wise, it grew far beyond the normal complement of 80–225 odd people that a typical Signal Company might normally house. At the time of my service the 518th had more than 400 troops assigned to it.

Lesson wise then, Vietnam taught the Signal Corps that not only did it have to be flexible when it came to the type, design, and purpose of the communication equipment it filled its coffers with, but it also had to learn how to command a troop complement far larger and more greatly dispersed than anything encountered in any previous war. In Vietnam, if a Company Commander wanted to check on the status of his troops, as in the 518th where my men were spread over an area the size of Connecticut, it involved a lot more than merely walking out of my Nha Trang office and sauntering through the barracks or mess hall before heading to the Duy Tan bar for the night. Instead, it involved up to two months of travel to visit all of the signal sites. Often times this forced me to allocate less than a day at each, sitting and talking with no more than a handful of men for an hour or so until the chopper pilot impatiently signaled me that he had to move on to his next stop. Surely these men, sitting at a remote signal site experiencing combat every 3 to 5 days, deserved far more than a visit from their commander every  6 - 8 months, one that gave them just an hour or two of face time at that.

That's what happens when Congress takes a dull axe to the Army's budget, and Pentagon planners then make TOE decisions based not on the wars that we fight, but the ones the civilian appointee heading the DOD thinks may happen next. Better to over compensate in terms of types and quantities of military communication and armament systems than to try and outfit your military with some constantly changing concept of what the next war will be like and who it will be against. Strategic planning is an unscientific science. It doesn't always work. Rumsfeld himself, who was not a bad Secretary of Defense in his own right, said that the problem with making decisions based on strategic planning is that the process of strategic planning is far from an exact science. He said that the very first rule for strategic planning is to precisely define one's goals. In his latest book Known and Unknown, A Memoir, he said "Setting clear goals may sound obvious, but it is remarkable how rarely governments..." do it. Instead they spend their time thinking of "options or courses of action."  

Rumsfeld vs HusseinHe goes on say that if you want any chance of success in using strategic planning as a base from which to make  policy decisions, you need to prioritize your goals. He makes the point that without knowing "which goals are the most important, one ends up with little more than a wish list...". Looking at the government's latest plans to cut the military's budget while at the same time reorienting the military to fight Naval battles with China and Iran, one has to wonder if this is not the very kind of wish list thinking that Rumsfeld says comes out of poor strategic planning. After all, what is the most important goal here? Is it to cut spending in the military, or win the next war we get into. One could be forgiven for thinking that these are two mutually exclusive options.

In the end, the Vietnam War proved to be a study in contrasts, teaching the U.S. government one thing, the military another, and the Signal Corps still another. As for the enemy, the lessons they learned have been recorded and taught to every tin pot dictator and despot ruler on earth… giving each a way to poke its finger in our eye at any time they want, without fear of a military defeat. From Iran to Venezuela, North Korea, Pakistan, Syria, and even Turkey, Lebanon, Jordan and others, many nations have decided that while America’s military might be strong, a) its superior firepower can be matched in the field with rudimentary arms backed by fanatical fighters, b) America’s national debt and deficit will not allow it to fight long enough to win a prolonged war, c) our politicians will cut and run at the first sign that our populace has lost interest in the war, and d) you can count on our populace to lose interest and cry for an end to any war in, oh, about 2 – 3 years.

As for our own government, especially as regards how it treats its military, the U.S. government seems not to have learned any of the important lessons stemming from either WWII, Korea, or Vietnam. In this author’s view, they continue to make the same mistakes in trying to micromanage the military as they have in each of these wars, especially once a war ends, cost cutting battles begin, and Congressmen try to hive off ever larger pieces of the military’s budget to support bridge construction in their home district.

Anti America: Swedish protestorsPolitically, in relation to how our government addresses those foreign countries that wish us ill, they seem again not to have learned much. Compare if you will the current president’s comments in 2008 on Iran with those of Kissinger on Vietnam in 1972. In 2008 President Obama stated in a speech in Portland, Oregon, that Iran doesn’t “pose a serious threat to us” because, by his reckoning, “tiny countries” with small defense budgets can’t do us harm. Kissinger matched this idiocy when in 1972 he stated on his return from his famous Paris peace talks “peace is at hand.” I suppose that if you consider having the U.S. Navy taunted by speedboats in the Strait of Hormuz from a soon to be nuclear power not a serious threat, or the death of 58,272 American soldiers in a war that our government turned its back on for a last minute “peace” so that politics could go on as usual before the next presidential race, then both of these people must be right.

For the Signal Corps, there were lessons to be learned from Vietnam. And for the most part, at least from this distance of retirement, it appears that the Signal Corps has done its best to learn and apply these lessons, in spite of the rearguard action it has had to fight all these years, just to hold its own.  

While we have talked of politics and process as areas of lesson learning, another important lesson learned from Vietnam has to do with the impact of technology on a modern Army's ability to fight against a regressive society. Take the issue of the level, type and quality of communication available to both sides.

In Korea America experienced for the first time the impact of fighting against an enemy on horseback and mules, communicating via flags and whistles. Yet strangely, horses and mules aside, the difference in communication capabilities had little impact on how the war was fought or its outcome. In Vietnam the same disparity in communication capabilities existed, but the impact was far greater. Why?

Part of the answer lies in the fact that in Korea it was a man to man fight, while in Vietnam the fight was via a proxy: the local villagers. If you are fighting man to man all you have to do is beat the other guy and the battle is over. If however the fight is really about the hearts and minds of the local villagers, and the other guy is rallying the local villagers every night after your pull in your pickets and lock your front gates, you had better find a way to "communicate" that returns data not about what the enemy is saying, but what the villagers are thinking. The lesson learned then is that while the communication available on each side of a battle line might show a massive disparity, this doesn't mean that the enemy is at a disadvantage. All the enemy has to do to make up for any weaknesses it suffers in lack of technology or communication capacity is to simply change the form of battle it engages in. Battling via proxy fighters is the quickest and easiest way to do this, as is fighting a guerilla war. And if one thinks this lesson hasn’t been learned by North Korea today… or even Iran, then one is sorely mistaken.

Devolution of our means to communicateIn primitive societies such as those of Vietnam, North Korea, or Iran, enjoying effective means of secure combat area communication is virtually unknown. On our side, while we may enjoy the most sophisticated signaling systems ever seen on the battlefield, their utility is of little value if they do not support a better means to gather information about the enemy, his position, intentions, status, and pattern of methodological behavior, and transfer that information in the form intelligence to the troops on the ground. Advanced systems such as satellites, tropospheric scatter, FM radios, and fiber optics are of little value if this goal is not achieved in its entirety.

One of the lessons of Vietnam for the Signal Corps then should be that it needs to both broaden the number and type of forms of communication technology available to it, as well as expand its role in the war game itself, to vet the data gathered and deliver it in the form of actionable intelligence to the war fighters, in sub-real time responses. One can see the need for this latter point because, strangely, the need to be able to do these things in ever shortening degrees of real time activities is in direct proportion to the enemy’s increasingly sparing use of any form of communication. That is, the more the enemy goes quiet, the more imperative it is that Signal Corps systems and processes are able to work at a faster speed. Harking back to the Air Force’s lessons from the Korean War, one could say that the overwhelming technological superiority the Signal Corps holds becomes of little value if it cannot close its OODA loop faster than the enemy retreats from the use of technology. In other words, decision making in the 21st century will take place under conditions of ambiguity and hyper-speed in information: in a word, complexity. The Signal Corps must adapt its capabilities to support this new form of communication.

In closing, while as the reader can see from some of the comments in this article, a modicum of ill feeling and bitterness still remains in those who fought in Vietnam… at least with regard to how the Vietnam War was brought to a close. Nevertheless, there is little argument that the U.S. Army Signal Corps performed its mission admirably. It got the message through.

As stated  about the Vietnam War in Getting the Message Through, A Branch History of the U.S. Army, “in performing their mission, Signal Corps communicators sustained relatively heavy casualties, especially among radiotelephone operators accompanying combat operations. Their vital mission coupled with their high visibility, [and] the telltale antennas protruding from the radio sets, made them prime targets.” And while no amount of rationalization can negate the price these boys paid, it must be said that in support of their efforts their brother signalmen did their damndest to put in place and deliver efficient and rapid communications, if only to help reduce the battle fatalities of our fellow signalmen by speeding up the medical evacuation process.

Among the list of Signal Corps Officers we should pause to think of for their gallantry in Vietnam is Capt. Joseph Maxwell ("Max") Cleland, who received the Silver Star. Among those Signal Officers assigned to closely held signal companies embedded within the Infantry and other units, several soldiers serving as communicators earned recognition. One of them, “Capt. Euripides Rubio, Jr., communications officer for the 1st Battalion, 28th Infantry, posthumously won the award for his gallantry during Operation ATTLEBORO in Tay Ninh Province in November 1966. During an attack on 8 November, Rubio left the relative safety of his position to help distribute ammunition and aid the wounded. When the commander of a rifle company had to be evacuated, Rubio, already wounded himself, took over. Continuing to risk his life to protect his troops, he was eventually felled by hostile gunfire after tossing a misdirected smoke grenade into enemy lines.”

- - -

With Vietnam behind it, the Signal Corps moved on. In the troubling times that followed Vietnam, the Signal Corps underwent yet another significant transformation. This time however the change was due, with most of the changes being made in great measure because of the lessons learned from both Korea and Vietnam. Among the changes that took place are these:

– When Congress discontinued the draft in 1972, ushering in an all-volunteer organization, it was only natural that the Signal Corps would use the best of this concept to its advantage. As part of reorganizing to embrace the new Army, women were given an expanded role in the Signal Corps, with more career opportunities being made available to them. By 1976 over 7,000 enlisted women were distributed among all but a few of the then sixty-one communication MOS specialties.

– The ever present budget tightening forced the post-Vietnam Army to adopt a rationalized force structure based on sixteen Regular Army divisions, which were said to be strong enough to defend U.S. interests in Europe but lean enough to reduce the strain on the taxpayers’ pocketbooks. Fortunately, this idea never had to be tested. Equally fortunately, under the new “Total Army” concept, the Army Reserve and National Guard were made available to assume a greater role in the nation's defense, thus helping to round out under-strength units.

– 1973 saw the Army place its branch schools under the newly created Training and Doctrine Command (TRADOC). The next year the Signal Corps, fearing it might lose control over the excellent training schools it had spent decades building, began consolidating its own signal training at Fort Gordon, Georgia. So quickly did this effort move forward that by the summer of 1974 the Southeastern Signal School was re-designated as the U.S. Army Signal School, while the signal school at Fort Monmouth became the U.S. Army Communications–Electronics School. Shortly thereafter, on 1 October 1974, Fort Gordon became the U.S. Army Signal Center, with the fort being re-designated as the new “home of the Signal Corps.”

– During the same period the Strategic Communications Command (now located at Fort Huachuca) dropped the word strategic from its name and became simply the U.S. Army Communications Command (ACC). It was thought that the new title better described the broad range of mission objectives the command had, from providing communications within Army posts, camps, and stations to signaling around the world via satellites.

– In various forms Army planners undertook a number of revisions to tactical doctrines over the intervening years, hoping to encapsulate combat lessons from Vietnam with others learned from studying conflicts such as the Arab-Israeli war in 1973. For the most part, the new concepts that were rewritten as doctrines failed, with one in particular crashing and burning within 5 years of its publication. This one, summed up in the then new Field Manual 100-5, Operations, was tossed aside in 1982 after being continually and consistently ridiculed for being based on a “victory in the first battle is imperative” solution to each and every war to come.

– On the negative side, because of the heavy commitment the military made to advanced high-tech items such as the M1 tank, the Patriot air defense missile, the Bradley fighting vehicle, and the Apache attack helicopter “doctrine creep” began to set in. With a never ending need to justify the existence (and purchase) of more and more of these systems, combat doctrine found itself  being rewritten to depend on these weapons systems. In simple English, doctrine development became driven by the systems selected for development years earlier… towards the end of the Vietnam War. In this regard, a lesson that could and should have been learned from Vietnam and Korea was not. Fortunately, the Soviet Union imploded before America’s equipment driven approach to combat doctrine development could be tested.WIN-T Communications Equipment

– On the positive side again, the Signal Corps kept its focus throughout these years, working diligently to bring military communication firmly into the twenty-first century. In part it did this by working with its sister services to develop fully interoperable telecommunications systems managed under the auspice of the Joint Tactical Communications Program (TRI-TAC). During the same period, the Signal Corps promoted to the Army a new tactical communications architecture known as Mobile Subscriber Equipment, or MSE. Wondrously, to save time and money in implementing MSE, the Signal Corps took a page from its old Page and Philco days of Vietnam by endorsing the specs of a system that had already been developed for civilian use by GTE, rather than design a new one. Further, this time the Signal Corps served only to distribute the equipment and provide service support for it, leaving it up to the user to both cover its costs within its own budget, as well as operated the system. If one stops and thinks for a minute, one can see in this approach the beginnings of how the Signal Corps will move forward in the 21st Century. Simply put, if one wants to specify a strategic plan of evolution for the Signal Corps, this is probably it.

– To extend even better communication to the battlefield, especially at battalion level and below, the Signal Corps introduced new VHF-FM combat net radios. Known as the Single Channel Ground and Airborne Radio System (SINCGARS), it replaced the VRC-12 family of radios developed during the late 1950s. Designed as a family of radios, units were made available in man-packable, vehicular, and airborne versions. Of equal importance, SINCGARS was designed to be smaller, lighter, and able to provide more channels than its predecessor.

– Working to support its sister arms, the Signal Corps developed new data systems as part of an effort to modernize the capabilities of the Joint Tactical Information Distribution System (JTIDS) of the Air Defense Artillery section. The solutions implemented provided better missile fire control via an Enhanced Position Location Reporting System (EPLRS) that used radios to provide real-time position location, identification, and navigational information on the battlefield. Together the JTIDS and EPLRS were renamed the Army Data Distribution System (ADDS).

– In 1981 Army Chief of Staff General Edward C. Meyer approved the implementation of the United States Army Regimental System (USARS) to improve unit cohesion and esprit. An act that clearly resulted from the lessons of Vietnam where men rarely stayed in one unit long enough to develop binding feelings for it, under this new approach soldiers were assigned to regiments and, as originally conceived, would remain affiliated with them throughout their military careers. Within the Signal Corps and other combat support/combat service support branches, where a large portion of the soldiers served in units outside their assigned branch, the system was implemented on a "whole branch" basis. In other words, the entire Signal Corps was considered to be the Signal Corps regiment, and any soldier with a Signal MOS was automatically affiliated with the regiment upon graduation from the branch school. On 1 June 1986 the Signal Corps regiment was established as a component of the USARS with Fort Gordon as the regimental home base. Accordingly, on 3 June 1986 the commander/commandant of the Signal Center and Fort Gordon also became known as the Chief of Signal. Maj. Gen. Thurman D. Rodgers became the first to carry the new title.

More changes have followed these few listed above even until today, from the introduction of the Tactical Fire Direction System (TACFIRE), to fiber optics based electromagnetic pulse (EMP) reduction systems, to the assumption by the Signal Corps of the responsibility to develop, introduce, and manage a new paperless records management system for the Army as a whole. These and other efforts were, for the most part, rolled out as part of a doctrine that revolved around creation of an Information Mission Area (IMA). And while this concept has taken its hits and been repeatedly revised, it nevertheless has proven the case that as a result of the needs developed during the Vietnam War, the Signal Corps has adapted itself to be able to constantly move to meet and overcome the new challenges ever evolving modern warfare presents.

Firmly understanding that it must meet two simultaneous missions: that of a war fighter and that of the provider of any and all manner of communication and data systems and management, there is little doubt that the Signal Corps will continue to adapt to changing conditions in the information management environment. With a task of meeting the military’s needs for knowledge, information, and intelligence about the enemy, his position, intentions, status, and pattern of methodological behavior, the Signal Corps will continue to be an indispensible part of America’s Army.


 The first two articles in this series are available here:

  Go to Part I:   The Signal Corps During The Cold War       

Go to Part II: The Signal Corps During The Korean War  




[1] In business a knowledge worker is someone who is empowered, because of their access to real time, detailed information about an event, to make policy changing decisions as to how a company should respond to the event. They key element in this definition being access to real time, detailed information as an enabling force to empower an employee to make a decision that would either form new company policy regarding the issue in question, or go against company policy completely. In a combat environment, a knowledge worker would thus be someone who, again because of their knowledge of real time, detailed information of the event in progress, makes real time leadership and tactical decisions based on that information. The Signal Corps, in making available the delivery of such information to field combat personnel through its fully integrated communication networks, enables the empowerment of soldiers to act as knowledge workers, rather than simply forcing them to follow orders that, while they may have been proper for the occasion when first issued, are no longer relevant because of changes to the circumstances on the ground. The purpose of a knowledge worker's existence then is to gain real time access to the information needed (both audio, visual, and data) such that they are able to make a decision as to how an event should be addressed, while that event is in progress and in real time contact with the knowledge worker.– To return to your place in the text, click here: Return to Text

[2] The "Powell Doctrine" is a journalist-created term that begins with a long list of questions that should be answered before war is begun, and ends by asserting that when a “nation is engaging in war, every resource and tool should be used to achieve decisive force against the enemy, minimizing US casualties and ending the conflict quickly by forcing the weaker force to capitulate.” – To return to your place in the text, click here: Return to Text

[3] Re. the purpose of government: Known as the “social contract”, governments exist i) to protect the rights of the individual citizen, and ii) to preserve the property and homeland of those citizens. See for example: John Locke’s (1632-1704) Treatise on Government. Ibidem Bertrand Russell, A History of Western Philosophy, A Touchstone Edition by Simon & Schuster, May 2007, page 629. Reducing a nation’s ability to wage the kind and number of wars needed to protect its people flies in the face of the purpose of its existence in the first place. To return to your place in the text, click here: Return to text

[4] In 2003 Secretary of Defense Donald Rumsfeld, supported by General Pete Schoomaker, laid out a new program that, while long on explanation as to how it would streamline the military to be better able to fight asymmetric wars, was by many people’s reckoning little more than a disguised attempt to save money by cuts in the Army. The program set in motion a plan to convert the Army from a force of 10 active Divisions (of 15-20 thousand troops each) into a force of 40 Brigade Combat Teams (of 3 - 5 thousand troops each). Additional Combat Brigades would be set up in the National Guard via similar measures. In essence, the intent was to move from a Division centric deployment approach to one based on Brigades. The approach put in place has had mixed results. For example, some credit it with the ability to support sustained deployment of ground forces in Afghanistan, while others say that the same could have been just as easily accomplished within the old structure.  As to its effectiveness in asymmetric warfare, that may never be known, as the 2011 changes proposed by President Obama will change the military yet again. This time moving towards a structure "optimized" for fighting naval engagements against countries like Iran and China. To return to your place in the text, click here: Return to text

[5] The problem that Eisenhower created was compounded in 1961 when President John F. Kennedy and his secretary of defense, Robert S. McNamara, decided to “reorganize and strengthen” the armed forces to allow for a more flexible response to international crises. At that time McNamara initiated far-reaching managerial changes within the Defense Department, shifting power from the military services to the civilian bureaucracy. In terms of when the trend towards civilian management of military operations and assets began, and who was responsible for it, the answer is a) 1961, and b) Robert S. McNamara. Effectively, McNamara directed a gutting of the Army Staff, at its highest level. In support of this, on 16 January 1962 President Kennedy submitted a plan to Congress that abolished the technical services, with the exception of the Medical Department. Congress raised no objections, and the reorganization became effective on 17 February. In the process the positions of Chief Chemical Officer, Chief of Ordnance, and Quartermaster General were done away with. The positions of Chief Signal Officer (who would now report to the deputy chief of staff for military operations [DCSOPS]) and Chief of Transportation were allowed to continue to exist, albeit as special staff officers rather than as Chiefs of Services. The Chief of Engineers lost his military responsibilities, but was kindly allowed to retain his civil functions. In one fell swoop, by eliminating the technical services as independent agencies, McNamara effectively handed over military technical asset design and development, implementation, operations and management to the civilian industrial sector. To return to your place in the text, click here: Return to text

[6] Almost as soon as the announcement was made by Japan on August 15, 1945, that it was tossing in the towel (VJ Day), the U.S. dispatched a twelve-man team to Hanoi to arrange for the release of American prisoners. That team included four Signal Corps men, including two Signal Officers. These four established the first non-clandestine American communications station in Vietnam. The communication equipment was set up at the Hotel Metropole in Hanoi. From that point until the end of the Vietnam War, in one way or another, the Signal Corps maintained both people and active communication links from Vietnam back to the U.S. To return to your place in the text, click here: Return to text

[7] The Pacific Scatter System used tropospheric scatter and ionospheric scatter signal propagation. At inception the system was limited to two voice channels, one of which could be multiplexed into sixteen teletype circuits. To return to your place in the text, click here: Return to text

[8] So under staffed was the Signal Corps that by 1967 it was contracting with Page to build steel drum revetments around the signal buildings on the signal sites that dotted the country. Source: United States Army in Vietnam; Military Communications A Test for Technology, John D. Bergen, page 334. To return to your place in the text, click here: Return to text

[9] To be fair, much of the problem with signal fading came about not because of poor design by Page and Philco, but because of a combination of solar flares and temperature inversion problems. What actually transpired was that when Page and Philco proved unable to determine why they were losing signal strength, a group of experts from the Defense Communication Agency (DCA), headed by the Signal Corps, was brought in to analyze the problem. They determined that rarely, but on occasion, solar flares affected the tropo hops, while more frequently the problem happened because of the formation of local temperature inversions, a phenomenon that occurs when the upper layers of the atmosphere are uncharacteristically warmer than the lower layers.  – To return to your place in the text, click here: Return to text

[10] During its long tour of duty the 39th had participated in all 17 campaigns and earned 5 Meritorious Unit Commendations. – To return to your place in the text, click here: Return to text


Sources used in the writing of this article include:

United States Army in Vietnam; Military Communications A Test for Technology; by John D. Bergen

1951 - 1963: From Rice Paddy to STARCOM Station, Early American Strategic Communications In Vietnam; Josef W. Rokus

Getting the Message Through; A Branch History of the U.S. Army Signal Corps; Rebecca Robbins Raines

Johnathan M. House, Toward Combined Arms Warfare: A Survey of 20th-Century Tactics, Doctrine, and Organization; 1984. United States Army Combat Studies Institute. US Army Command and General Staff College. Fort Leavenworth, Kansas. p. 155.

Michelle Malkin, Barack Obama: Gaffe machine; May 21, 2008;

"TRI-TAC and You!," Army Communicator 1 (Spring 1976)

"Commander's Comments," Army Communicator 14 (Fall 1989), Annual Historical Review, Headquarters, U.S. Army Signal Center and Fort Gordon, Georgia, in re. how Information Systems Command became the proponent for the Army's data processing units, formerly Adjutant General Corps assets.

Nancy S. Dumas, "Fielding SINCGARS," Army Communicator 13 (Winter 1988).

The Journal of Military Electronics and Computing; Tech Refresh Strategies Bolster New Battlefield Compute Workloads; August 2011.


This page originally posted 1 February 2012, formatting updated March 2017. 

Top of Page


Original Site Design and Construction By John Hart, Class 07-66. Ongoing site design and maintenance by WebSpecks Incorporated, courtesy Class 09-67.
Content and design Copyright 1998 - 2012, by WebSpecks, Incorporated.