A more complete version of this interview is available at http://www.clbooks.com/nbb/strassmann.html
Paul A. Strassmann has jokingly referred to himself as "a cast iron, Establishment, corporate-bureaucrat kind of a business computer creature."... There's no doubting his credentials for that claim; during the past 35 years he has served as Chief Information Officer for General Foods, Kraft Foods, Xerox, even the Pentagon (the latter during the Bush Administration, controlling a $10 billion dollar budget!).In fact, Mr. Strassmann is anything but an MIS technocrat of narrow horizons; he is also an MIT-trained engineer, entrepreneur, document collector, consultant, West Point professor, and the author of four books. His intellectual interests encompass political theory, history, economics, and related topics. This is Part I of an interview done with Dan Doernberg in December 1994.
CLB: When did you first work with computers?
Strassmann: The first time I saw a computer was in a course I took at Harvard in 1953. I started working with computers at MIT in 1954, so this was as early as you can get; in fact, my 1955 Masters dissertation at MIT's Sloan Business School was the first in the nation on a commercial application of computers. I took punched cards from the New Jersey Turnpike and devised a method for scheduling toll collection personnel.
CLB: Your latest book, _The Politics of Information Management_ (1995), touches on many themes that you've written about before; how does it fit into the body of work you've produced over the years?
Strassmann: I've always been preoccupied with the maintenance of civilized society, as I've seen with my own eyes how even an advanced one can disintegrate. I have witnessed, in the Nazi and Soviet cases, how bureaucracies can be organized to execute socially destructive missions, and how they can run amok if they seize enough power to pursue criminality without any restraints whatsoever.
CLB: You've always had a very ecological view of business; thoroughly analyze the customers and competitors, identify the relevant assets, strategize, and then act.
Strassmann: My intellectual heritage is cybernetics. You may recall that in the 1940's Norbert Wiener started examining how self-adaptive capabilities function in biological organisms as well as in engineering servomechanisms. It encompasses the entire cycle of how an organism learns, governs, and adjusts so that it may survive and maybe even prosper. Wiener's followers tried to comprehend how socially constructed entities, as well as biological ones, manage to adapt to changing conditions, and how they evolve by linking their internal control mechanisms to the external realities. I started applying feedback theories in work that I did for General Foods in the 1960s. In those days you could find logistical anomalies, e.g. our Atlanta warehouse had a 40 year supply of 3 oz. lime Jell-O, but Jell-O has a shelf life of only 2 years! I started a detailed examination of how a presumably rational, profit-driven organization such as General Foods could continue accumulating useless inventory.
CLB: In general, has there been much cross-fertilization between economics and business and cybernetics?
Strassmann: Yes, to some extent... but much of the writing was just theoretical, devoid of any experimental testing. I'm too much of an engineer to tolerate theories that cannot be tested and applied in practice. My writings and research always end with lists of things to do. The intellectual heritage goes back to economists like Friedrich Hayek and, in fact, goes back more than 200 years ago, when the debates began whether and how mankind should plan its future, rather than leave it to some uncontrollable causes. This intellectual arrogance of presuming that planning can substitute for governance goes back to the Enlightenment philosophers Diderot and Voltaire in pre-Revolution France. Impatience with mankind's progress was transformed into a view that a self-appointed intellectual elite was sufficiently knowledgeable about what was needed to create a better future through planning. The basic premise here was that a good society could be planned and engineered. From these origins sprung forth two centuries of Socialist movements, in a variety of forms, including pathological "red" (Communist) and "blackshirt" (Fascist) varieties... What really fascinated me was the recognition that both Communism and Nazism were cybernetic run-away designs, because they abrogated the essence of all cybernetics, which is self-adaptive feedback controls. They self-destructed...
CLB: You place a tremendous emphasis on the maintenance and increase of "human intellectual capital" in organizations.
Strassmann: You would not have a microchip, perhaps the most complex thing mankind has ever put together, without a vast amassing of information capital. The manufacturing of microchips takes an enormous accumulation of skills and experiences about the relationships between matter, electrical characteristics and logical properties. Such knowledge is not reinvented for each new microchip. Innovation takes place through integration of added information capital on top of what has been already gathered. Any new microchip incorporates knowledge that it "inherited" from chips that were made 15 to 20 years ago.
CLB: Which is why you're also a proponent of object-oriented software techniques.
Strassmann: That's right. What is just incredible, and only gradually emerging into human consciousness, is that for the first time the underlying chemical limitations of the brain to accumulate knowledge will be overcome by manufacturing experience surrogates--software. The truly big takeoff in the evolution of mankind's learning capacity is still ahead of us. In terms of evolution, the tail-end of the 20th century will be seen as a totally unprecedented turning point in the relationship between humanity and Nature. Right now we do not have even a remote clue what that may lead to.
CLB: Right off the biological scale.
Strassmann: It goes right off the scale. The only comparable development is the dramatic jump that happened at the end of the 19th century with the advent of locomotion and manufacturing machines. Although a wide variety of mechanical artifacts have been around for at least two thousand years (very advanced machines were developed in Roman times), the capacity of society to integrate them into productive life remained dormant; the social structures of the various societies could not take advantage of them. In medieval times, the guilds successfully opposed machines, as machines threatened their monopolies based on control over the sources of skilled labor. The French developed very sophisticated machines, but they were mostly used to produce amusing toys for the king. In general, the ability to start accumulating knowledge about machines did not exist. In fact, the design of machines were always considered a proprietary trade secret.
What I'm saying is that information technology, or any given technology, comes into being only when there is a confluence in evolution and a certain freedom to move to the next plateau -- when a society is able to use innovation to accumulate knowledge with a distinct economic advantage.
CLB: The Business Value of Computers (1990), advanced your measure R-O-M (TM), Return-on-Management, as the ideal measure of business performance. What are your current feelings about this measure?
Strassmann: _The Business Value of Computers_ is an important book, perhaps my most valuable book in terms of theory. Its focus is on metrics, how to evaluate the productivity of businesses where knowledge, not capital, was the source of competitive advantage... In a few months I plan to publish in _Computerworld_ approximations of the R-O-M (TM) values (Information Productivity (TM) ratios) for the FORTUNE 1000 companies.
CLB: In Information Payoff (1985), you were ahead of your time in insisting that satisfying customers be the sole focus of a business, with everything else falling into place around that.
Strassmann: I find that I write about theoretical propositions fairly early, before they become fashionable. Regarding customer service in particular, that's derived again from servomechanism theory, what is called end-loop control... Bureaucracies are dysfunctional because they respond to their internal contentions rather than customer needs; the Purchasing Department works against the Engineering Department, which works against the Production Department...
After the demise of the Soviet Union, the Pentagon has become the world largest planned economy. I'm now extending the notion of customer service to the military. When you look today at Bosnia or Somalia, those are warlord engagements. Local warlords engage in tactical decisions at the point of the enemy (the "customer"); they improvise at the point of the engagement, in the field rather than at a headquarters. The principles of information design and information feedback are the same, regardless of where and how they are applied.
CLB: We were talking before about the historical desires of elites to exercise control; can you say a few words about your very public disagreement with Michael Hammer about reengineering?
Strassmann: Reengineering, like many lofty ideas... has to be examined by first finding out how it works instead of what it says. One of the fundamental precepts of reengineering is the idea that you start with a "blank sheet of paper" to design how an organization is supposed to function. The reengineers are revolutionaries, who consistently assert that the past should be discarded and the new must be "reinvented" from ideas that are not burdened by anything that has come before. Well, it is demonstrable that it is impossible to restructure an organization without any trace of what has come before.
All human beings in organizations bring with them their prejudices, their problems, their inhibitions, etc. After subjecting a group of workers to violent reengineering, they also have the trauma of being forcefully dislodged from what they were once doing. You end up with a disoriented team of people who are not accustomed to work together as the reengineers prescribe. Even if an organization "annihilates," "nukes," or "obliterates" all its people (terms that the reengineers have used with abandon), you still have to assemble human beings and find ways for them to willingly cooperate. The idea that one can completely design or reengineer organizations by first destroying them is of utopian and socialist origins, and for the last two hundred years has been nurtured by radicals seeking a theoretical legitimacy for seizing power...
Historically, under the cover of assertions that the "reengineers" are the avant-garde, the contemporary versions of Plato's philosopher kings, you find calls for the application of organizational violence as a way of speeding up inevitable progress. In this respect Michael Hammer is especially prolific in using metaphors of violence (e.g. turning guns on your own recalcitrant workers). Such calls to violence have always resulted in violence, the waging of civil war against one's own people for their own good. This is something that I have a very special allergy against.
Reengineering consultants also claim legitimacy for what they propose by saying "I have here a list of excellent business practices. I'm going to tell you what the excellent practices are, and if you follow the list, you'll be okay." The presumption that someone can compile a universal list of what makes, for instance, for an excellent information system is misguided. Every firm is different, every organization has different needs, and therefore every information system implementation must be unique. This is why awarding the Baldridge Prize for excellence based on a master checklist is unjustified. It's based on the idea that a body of experts can guarantee superior results if you follow their list of excellent characteristics. That ain't so. So far, 20% of the Baldridge winners have delivered dismal economic performance.
We are learning now that survival in ecological niches is feasible through a variety of unique adaptations. There are all kinds of fish that survive and there are all kinds of worms that prosper. It does not follow that one species will dominate because they are designed according to some preordained order or excellence... It does not follow at all that an organization that connects all employees to the Internet, uses client/servers, applies object-oriented software, and incurs heavy expenditures for information technologies will necessarily be superior to one that conducts modest operations on an old mainframe computer.
CLB: Instead of "reengineering", you advocate "business process improvement." How is that different from reengineering, or from Total Quality Management?
Strassmann: It's very close to Total Quality Management. This approach, especially as advocated by Juran and Deming, always starts organically from within the working levels of the organization, bottom up. You have to harness the imagination of the workers. The person who works on the assembly line tightening a screw, or the person in the bookstore talking to the customer, is the crux of all quality.
Copyright 1995, Computer Literacy Bookshops Inc. All rights reserved.