Skip to content
Jun 18 / Barry Clemson

When Systems Thinking Ruled the Pentagon: A Personal Reflection

Author: Richard WrightVolume 1. Issue 1. June 18, 2012.
Abstract: This paper traces the rise of “systems” under McNamara at the Department of Defense and its resultant popularity within academia. McNamara’s shallow understanding of “systems” subsequently led to discrediting the entire movement. 

The 1960’s perhaps can best be described as a tumultuous period whose central events were the Civil Rights Movement, the Vietnam War, and the U.S. version of a cultural revolution. Lost in all the uproar was the formal introduction of the concepts of systems thinking for the first time to the U.S. Department of Defense (DOD). This paper discusses the ascendancy of “systems” under McNamara and the subsequent discrediting of systems thinking.

As a recent college graduate and a draftee, I was a peripheral witness to these events and a minor participant in the Vietnam War.  As a two-year draftee I was scheduled for the infantry until I was offered the opportunity to go regular by enlisting for two more years, in return for which I would receive a year’s training, and a soft job as an intelligence analyst. Of course I took the opportunity.  As a result, I never served in Vietnam, but worked in the surrounding countries in direct and indirect support for the 50,000 guys who actually had to do the fighting in that war (at its height there were 500,000 U.S. troops in Vietnam of whom only about 50,000 were actually fighting at any one time, with the rest in support roles. There were an additional 200,000 or so troops in the South Asia area also supporting the 50,000 “war fighters”, I was one of those troops).

All this is by way of saying that in the intelligence world we had a ringside seat on how the McNamara way of war actually worked and more generally how systems analysis, as applied by Secretary of Defense (SECDEF) McNamara (1961-1968), actually worked in war and peace.

The Rise of Systems Thinking (1947-1965)

Like many of his generation (1916-2009), Robert Strange McNamara was strongly influenced by his experiences in WWII.  During the war he was assigned to an Army Air Corps organization called ”Office Statistical Control”, which among other tasks used statistical date to determine the effectiveness and efficiency of the U.S. bombing campaigns in Europe and Asia.

After the war McNamara and nine of his colleagues were hired by the Ford Motor Company to rescue it from a badly eroding market share. Within a few years, thanks largely to McNamara and his “wiz kids” as they become known, Ford became a major competitor again. So what did they do to achieve this?

Applying lessons learned from the war, McNamara introduced the use of statistical analysis which later became widely known as “planning, programming and budget system” (PPBS) to guide everything from production to marketing. Although he never acknowledged it, McNamara was apparently influenced by the work of W. Edwards Deming in rebuilding the Toyota Car Company in post war Japan. He introduced the term “systems analysis” to describe his approach to automobile production. He saw auto production as a complete system, not a series of autonomous steps. While he was at Ford his PPBS approach proved highly successful. His most conspicuous triumph was the Ford Falcon, a moderate cost, economical small car that proved enormously popular in an age of “muscle cars”. (For more details see “Robert McNamara and the Evolution of Modern Management” Harvard Business Review Dec 2010)

When President John F. Kennedy was elected he and his advisors wanted somebody to bring the DOD under control since even then there was a tendency for DOD spending to get completely out of hand. With his WWII experience and success at Ford, McNamara apparently seemed an obvious choice to be Secretary of Defense.

McNamara and his wiz kids who came with him took the Pentagon by storm. Suddenly everybody in DOD was talking about the wonders of systems analysis and how it could be used to make DOD much more efficient and cost effective.  This was manifested in McNamara’s PPBS approach to defense spending. He introduced such innovations as analyzing total system costs rather than costing each individual component separately. Soon the civilian world was filled with talk of system analytic approaches to everything from sales and marketing to manufacturing. Academics began cranking out textbooks on systems analysis for courses in business schools across the country. The concept of systems thinking, i.e. thinking of everything as consisting of integrated components that function towards one or more goals, was suddenly the intellectual flavor of the day. So what happened? (Halberstam,  1993, Jordan 1999)

The Discrediting of Systems Thinking (1966-?)

As McNamara attempted to apply the systems analysis approach to DOD issues a serious flaw was discovered in his thinking; when he looked at a system he did not see or discounted any system component that could not be measured and graphed or otherwise presented as a numerical value. This led McNamara, e.g., to disregard service traditions and esprit de corps as irrelevant and to attempt to replace distinctive uniforms for the armed services with a single uniform. This idea was simply subverted by the service chiefs. More seriously his refusal to include intangibles in his systems analysis approach to the Vietnam War led him to disregard factors such as political will, courage, and ideological fanaticism. He never could get a handle on either the Viet Cong (South Vietnamese Communist Party) or the North Vietnamese. For example there is a whole sub-set of literature on the vain efforts of both military and civilian intelligence analysts to convince McNamara and the Military High Command that body counts were not providing accurate information. Because they were hard numbers McNamara could not believe they could be misleading. (Details on McNamara and the Vietnam War can be found in Hiam, 2006 and especially Allen, 2001)

This inability to understand intangibles was compounded by a misplaced arrogance that caused McNamara and his followers to disregard even hard data when it ran counter to their own ideas.  For example McNamara decided that joint services fighter was the most cost effective solution to the high price of fighter aircraft. As a result the TFX fighter-bomber project was born and ultimately produced the F-111 fighter for the Air Force alone, since the TFX proved unsuitable for carrier landings (he ignored the engineers who told him about the greater stress abrupt takeoffs and landings placed on air frames, and so on).  Although by 1972 a handful of F-111’s were operational, the whole TFX effort was a scandal in the late 1960’s. (See Art, 1968 for a full treatment of the TFX controversy).

It did not take long for systems analysis (and with it systems thinking) to be discredited as the plaything of fools. This was a tremendous setback for the world of systems analysis and systems thinking. That is why when the craft of systems engineering was developed much was made of the fact that the systems referred to were information systems (computers) and systems analysis was narrowly defined as focused on “operating systems” of one sort or another.  It is likely that there still is a residual feeling that the concept of systems thinking is somewhat disreputable (Halberstam, 1993).

Lessons Learned

Although he failed in the end, for a time Robert McNamara made systems analysis and systems thinking the talk of the town. He did so because he demonstrated at Ford that Systems Analysis (systems thinking) could be harnessed to achieve practical goals. He and his wiz kids successfully took the concept of “systems” from an academic concept understandable only by professors to the rough and tumble world of auto making. Although others including Edwards Deming were doing much the same thing, they didn’t have the publicity flair of McNamara.

McNamara’s failure to effectively apply systems thinking can be considered proof of the old cliché that “a little knowledge is a dangerous thing.”  He understood enough about the concepts of systems thinking to understand its utility, but not enough to effectively apply it to the complex systems that abound in DOD, then and now.

Still, in my opinion systems thinking and its close relatives systems analysis and operational research, are today still too often considered only as academic subjects with no practical applications. Too often people use systems thinking everyday, but are unaware of it because they believe it to be only an intellectual plaything. There is a moral somewhere in here.

Richard Wright was born in Ohio, raised in California, graduated from UCLA with a BA in Anthropology, and a Spanish minor, was drafted into the U.S. Army, went regular to become an intelligence analyst, worked as an intelligence analyst for twenty years before switching to the technical intelligence support learning network analysis and ultimately systems engineering along the way, left for the private sector and retired as a principal systems engineer. Now he is more or less blissfully retired with his wife and dog in the splendid isolation of the central New Mexico Mountains.

Staff who worked on this article: Editor: Gene Bellinger. Reviewers: Nicolas Stampf, Beth Robinson, Barry  Clemson, George McConnell

References

Art, Robert. (1968). The TFX Decision: McNamara and the Military, London, England: Little Brown and Co.

Allen, George W. (2001). None So Blind. Lanham, Maryland: Ivan R. Dee Publisher

Hiam, Michael C. (2006). Who the Hell Are We Fighting. Hanover, New Hampshire: Streeforth Press

Halberstam, Robert. (1993). The Best and the Brightest. New York, New York: Ballantine Books

Jordan, Amos, et al, (1999). American National Security.  Baltimore, Maryland: John Hopkins University

Citation details for this article

Wright, Richard. (2012). When Systems Thinking Ruled the pentagon: A Personal Reflection. Systems Thinking World Journal: Reflection in Action. [Online Journal]. Vol. 1 Issue 1. [Referred 2012-06-18]. Available:http://stwj.systemswiki.org . ISSN-L 2242-8577  ISSN 2242-8577

Print Friendly

2 Comments

leave a comment
  1. jbrieley / Jun 18 2012

    To follow-up with more evidence of what has been lost, in the late 1990’s, I had been asked to join a meeting in the Pentagon in which a consultancy that had been working to reduce the cost of weapons system by the U.S. Military. The consultancy had put forth a proposal that was designed to put in place a set of incentives that would reward procurement people’s efforts to buy the same systems but at the same time reduce the amount of money spent. Over a two year period, costs had escalated and (for some bizarre reason), the consultancy was brought back in to come up with a better plan. Sitting in this incredibly high-tech room filled with ‘decision-makers’ with mass-amounts of gold-braid on their uniforms, the consultancy’s presentation seemed to be just more of the same. During a break in the presentation, the other reviewer and I decided we would ‘loop-out’ the implications of what these new incentives might accomplish. The results were less than encouraging, so we took our findings (in the form of a causal-loop structure) to the Assistant Secretary of Defence who had invited us to sit in.

    The loops showed that whilst on the surface, the new incentive process looked like it made sense; but the loops also showed that the unintended consequence of this new procedure was that the procurement people were actually being dis-incentivised by the way that the procedure was going to be implemented. The ASD was intrigued and asked if we could spend another day to validate our findings through interviews with some of the procurement people.

    The next day we began talking to the people who would be most impacted by the new procedure (the new incentive scheme), and in one of the interviews, we heard from a procurement specialist that the previous scheme had not worked because of the system itself.

    The incentive scheme was basically a way to get the procurement people to ‘bargain’ with the producers of armaments and weapons systems. But what they encountered was solid resistance to any reduction in pricing, especially because the contracts were typically written to companies that had ‘exclusive’ (patented) rights to produce and sell their products to the Military. Clearly, there wasn’t any incentive for the producers to reduce their prices, so the procurement people simply crafted new agreements in which partial costs would be deferred into a different accounting category; i.e. instead of a fixed price for a bomb, the price might be listed as lower, but there would also be a contract written for ‘maintenance’ on the bomb. We were amazed at how creative the procurement people had been – they had found a way to receive their cash incentives by purchasing weapons systems at a lower cost on the accounting sheets, but in reality, the overall purchase price remained the same or increased – No real reduction in costs for the weapons, and new additional costs for the incentive payments. If that wasn’t enough, the person we were interviewing even said that what they were doing was simply ‘gaming the system.’

    We fed this information into our findings (still in the form of a causal loop structure) and met again with the ASD. His response was something to the effect of, ‘’now this all makes sense.’’ But he also added that whilst it was nice to know what was really going on – the gaming the system – there would be no effort to change the system, as it involved ‘politics.’

  2. Barry Clemson / Jun 18 2012

    The politics of weapons procurement are interesting. in recent decades, Presidents, Secretaries of Defense and the Joint Chiefs have all tried to stop major weapons programs. In most cases, Congress has over-ruled these efforts and has mandated that the programs continue.

    Major weapons systems involve jobs, lots and lots of jobs, and these jobs are spread all over the country. Every key congressman has bunches of constituents who work on these weapons systems. It is therefore usually politically impossible to stop these programs even when they make no sense militarily.

    Spending on weapons systems provides jobs, but very inefficiently. Building a tractor or machine tools, e.g., provides an economic multiplier effect. Building a tank or a sidewinder missile provides a short term economic boost (i.e., jobs), but the long term economic impact is a drag on the economy. We are in effect addicted to the short term boost of weapons systems spending but in the long term buying the wrong weapons systems is damaging to both our economy and our national security.
    In a real sense we are addicted to these programs

Leave a Comment