Paul Penfield, Jr., Time Constants, The MIT Faculty Newsletter, vol. X, no. 4, pp. 14-15; January/February 1998.

Time Constants

Paul Penfield, Jr.

The following article originally appeared in The Interface, IEEE Education Society and ASEE Electrical and Computer Engineering Division; Number 1, April 1997.

Some ideas came to a focus for me recently when we held a memorial service here at MIT for Gordon Brown. Many readers of The Interface will know instantly who Gordon Brown was, but for the others, let me explain.

Gordon served as Head of the Department of Electrical Engineering at MIT from 1952 to 1959, and then as Dean of Engineering for the next nine years. He retired in 1973, and lived his later years in Tucson, AZ where for some time he acted as an energetic citizen-champion to promote the use of system dynamics in the public schools. He was almost 89 when he died last August.

It is not too much to say that Gordon Brown had more impact on engineering education during the past 50 years than any other person. As department head and later as dean, he pushed through his vision of an engineering education based on fundamental science -- not the same science that was of interest to scientists, but rather "engineering science," those aspects that supported the practice of engineering. Although this concept seems perfectly natural today, it was radical in the 1950s.

In preparing my remarks for his memorial service, I came to appreciate better why such revolutionary ideas were necessary at that time, and also what other, equally radical ideas might be appropriate today.

It is all a matter of time constants. Yes, time constants. This is a concept that Gordon Brown, the expert in servomechanisms, knew very well. Just as electrical systems are characterized by natural time constants, so are natural phenomena and even social systems. We electrical engineers have a relative advantage in understanding the dynamics of such systems because of our familiarity with the dynamics of engineered systems.

We know that when dealing with rapidly changing, or dynamic, things, we can approximate slowly varying things as constant, or static. This approach is no less valid for natural or social systems than for engineered systems. For example, we think of the locations of continents as fixed on human time scales, even though they have moved on geological time scales. It is all a matter of time constants.

In engineering education, the most important time constant is 40 years, the length of one engineer's career. When designing an educational program, things that change slowly may be considered constant over a 40-year period, whereas things that change more rapidly must be considered as variable, in the sense that they may change during a single person's career.

Before 1900, advances in engineering (and in other fields) occurred at what seems today an incredibly slow rate. Fifty years passed between the inventions of the electric motor and the electric light. It took a hundred years for Coulomb's Law to turn into Maxwell's Equations. A practicing engineer could base an entire career on engineering techniques learned in school. The underlying science was changing so slowly that it could be considered as static, as far as an individual career was concerned. It is all a matter of time constants.

In the first half of the 20th century, science, especially physics, began to change more rapidly. Atomic theory, quantum mechanics, and relativity were introduced. But engineering education was not changed; the presumption continued to be that the science that was important to the practice of engineering was static, or unchanging.

The Second World War exposed the flaw in that reasoning dramatically. The atomic bomb was developed by physicists and chemists, who understood atomic theory. Radar was developed by physicists, who understood electromagnetism better. Engineers played a distinctly secondary role.

Gordon Brown recognized the problem. The presumption of a static science was obsolete. What was needed was a new model, in which the underlying science could change. Not only could specific branches of science advance rapidly, but other branches of science that had previously been of no engineering importance could suddenly become relevant.

As department head at MIT, Gordon led the revisions of the undergraduate electrical engineering curriculum to incorporate engineering science. He included an ample amount of science, to make it possible for our graduates to learn areas of engineering based on diverse sciences. He also exposed students to many different sciences, so that they would feel comfortable learning still other branches of science later in their careers.

Then, as dean, he extended this idea to other fields of engineering. Graduates of these programs, and similar programs elsewhere, went out to populate the faculty of many other universities, and the result was that today almost all engineering education has a heavy reliance on engineering science.

A question of critical importance to the readers of The Interface is whether this model of engineering grounded in a dynamic, changing base of science will serve us as well in the future as it has in the past. I personally believe it will. There is no indication that the rate of scientific advance is slowing, or that new sciences will be any less necessary. The only question might be which areas of science to incorporate. Many universities (including my own) are betting that biology will be as important in the future as physics is today. We now require a semester of biology for all students.

Another question that arises is whether, if Gordon Brown were alive today, he might have his sights set on a more important change than merely the selection of which sciences to include in what amounts. I believe he would. Let me explain. Again, it is all a matter of time constants.

We continue to educate our students as though the context in which engineering is done is static, or at worst slowly varying. By the word "context" I mean the cultural, political, industrial, social, and work environment in which an engineer practices. By failing to prepare our students to deal with a dynamic, or changing context, we are assuming that such changes have a time constant longer than 40 years. In other words, we recognize that science is dynamic, but we still think context is static.

There is ample evidence that this assumption is now obsolete. During the past 50 years we have seen several changes in context. Today society values the environment and disapproves of pollution in a way that was unknown in 1950. Today, almost half of the engineering undergraduates are women. Today, new countries are being formed every year, and new cultures are asserting their importance all the time. Today, American industry competes globally. Today, a successful engineer must be nimble, to cope with shorter design cycles, changing styles of activity, more teamwork, constantly improving design tools, and more effective global communications.

As these examples show, the context today is radically different from what it was 50 years ago or even ten years ago. That is, context is changing too rapidly to be considered constant during a 40-year career. Once again, it is all a matter of time constants.

What should we do to equip our students to deal with a rapidly changing context? That is probably the most important question facing electrical engineering education today. Different people have different ideas.

One idea is to ensure that students have a greater classroom exposure to various contexts. The traditional way of doing this is a liberal arts education. However, engineers also need their science base, and in addition they need the engineering approach to problem solving, which liberal arts programs do not usually supply. Perhaps what is needed is a new form of liberal arts education, with a much heavier dose of many sciences, along with some engineering experience. Or perhaps what is needed is a system where engineering is a professional, graduate program open only to graduates of a liberal arts program.

Another idea might be to strongly increase the creative, design portion of the curriculum so that a variety of contexts for design problems can be experienced. Another idea might be to emphasize exposure to multiple cultures through international exchange programs.

In judging any of these ideas, ask whether the graduates will be able to cope with profound changes in their context during their career. In other words, ask whether they will continue to learn after leaving school, both in technical and nontechnical areas.

One wall in our department headquarters at MIT is adorned with pictures of past department heads. Gordon Brown's picture is prominent among them. I see it every day as I come to work, and sometimes wonder how he would approach things if he were around today. The need for a change is quite clear -- it is all a matter of time constants. The way of satisfying that need seems much more elusive; it represents what is perhaps today's greatest and noblest challenge for us engineering educators.

[Paul Penfield, Jr. can be reached at penfield@mit.edu]


URL of this page: http://www-mtl.mit.edu/users/penfield/pubs/time_constants.html
Created: Feb 5, 1998  |  Modified: Dec 31, 1998
Related page: Penfield publication list
Site map  |  To Paul Penfield's home page  |  Your comments are welcome.