Trying to predict the impact of tomorrow's inventions

WILL HAND-HELD satellite receivers, with a capacity for locating your position to within a few yards on the Earth's surface, make it impossible to get lost? Will today's paper money be rendered obsolete, not by plastic credit cards, but by the settlement of bills somewhere along the electronic superhighway?

Will international conferences in the future be conducted by participants sitting at their computer terminals, thousands of miles apart from each other? Will the mapping of the human genome lead to medical breakthroughs resulting in average life expectancies of a century or more?

Many executives, investors, and government officials hope they have answers to these questions, and they are spending billions of dollars in accordance with what they see in their crystal balls. They may want to stand back a bit, though, and keep their options open. A review of many important innovations, from the steam engine to the laser, shows an unsettling pattern--it seldom is possible to predict the full technological, economic, and social impact of inventions, even long after their commercial introduction.

Consider the laser, one of the century's most powerful advances since its invention three decades ago. Its versatility is breathtaking. Lasers are used for navigation, precision measurement, and chemical research. In surgery, detached retinas, a frequent cause of blindness, now are repaired with lasers on an outpatient basis. Lasers are gynecologists' instrument of choice in many surgical procedures. In the textile industry, they are used to cut material, speedily and accurately. Lasers are employed in millions of households for the high-quality reproduction of music recorded on compact discs.

The laser's most profound impact so far has been in telecommunications, where, with fiber optics, it is revolutionizing transmission. In 1966, the best transatlantic telephone cable could carry 138 conversations simultaneously between Europe and North America. The first fiber optic cable, installed in 1988, could carry 40,000; today's cables, nearly 1,500,000. Yet, lawyers at Bell Laboratories, which invented the laser, initially hesitated to apply for a patent, on the grounds that the laser had no possible relevance to telephones.

The experience of the laser is not unique. Guglielmo Marconi invented the radio, but thought it would be used primarily where communication by wire was impossible. (To this day, the British call the radio the wireless.") The radio in its early days was thought to be of potential use only for private communication--i.e., point-to-point communication, rather like the telephone--and not at all for communicating to a large audience of listeners. Surprising as it may seem today, the inventor of the radio did not think of it as an instrument for broadcasting. He visualized the users of his invention as steamship companies, newspapers, and navies, and these users needed directional, point-to-point communication--i.e., "narrowcasting," rather than broadcasting.

Such lack of foresight was by no means unique. According to James Martin, a communications authority, "When broadcasting was first proposed ... a man who was later to become one of the most distinguished leaders of the industry announced that it was very difficult to see uses for public broadcasting. About the only regular use he could think of was the broadcasting of Sunday sermons, because that is the only occasion when one man regularly addresses a mass public."

Many years later, in 1949, Thomas Watson, Sr., president of IBM, saw no large market for the computer. World demand, the company thought, could be satisfied by just 10 or 15 computers. When the invention of the transistor was made public in 1947, some saw the device as helpful for producing better hearing aids, but little more. This invention, one of the greatest of the century, was not announced on the front page of The New York Times, as one might have expected. On the contrary, it was a small item buried deep inside the newspaper, in a regular weekly column titled "News of Radio."

This recitation of failures to anticipate future uses and larger markets for new technologies could be expanded almost without limit. People ought not to be too complacent, though, for the factors that blinded earlier generations to the impact of innovations are likely to persist. The uncertainties connected with the uses of new technologies are not likely to decline drastically. A more useful issue to explore is what incentives, institutions, and policies are more likely to lead to a swifter resolution of these uncertainties.

Much of the difficulty is connected to the fact that new technologies typically come into the world in a very primitive condition. Their eventual uses, therefore, turn upon an extended improvement process that vastly expands their practical applications. Watson was not necessarily far off the mark when he concluded that the future market for the computer was extremely limited, if one thinks of the computer in the form in which it existed immediately after World War II. The first electronic digital computer, the ENIAC, contained no less than 18,000 vacuum tubes and filled a room more than 100 feet long. Any device that has to rely on the simultaneous working of that many vacuum tubes is bound to be notoriously unreliable. The failure in prediction was, at bottom, a failure to anticipate the demand for computers after they had been made far smaller, efficient, reliable--and much cheaper.

There are other factors. Major technological innovations often constitute entirely new technological systems, but it is difficult in the extreme to conceptualize an entirely new system. Thus, thinking about new technologies is likely to be handicapped severely by the tendency to regard them in terms of the old technologies they eventually will replace. Time and again, contemporaries of a new technology are found to have thought about it as a mere supplement that would offset certain inherent limitations of an existing one.

In the 1830s and 1840s, railroads were thought of merely as feeders into the existing canal system, to be constructed in places where the terrain made canals impractical. The telephone originally was conceptualized as primarily a business instrument, like the telegraph, to be used to exchange very specific messages, such as the terms of a prospective contractual agreement. This may explain why Alexander Graham Bell's original telephone patent was titled "Improvements in Telegraphy." One wonders what he would have thought of today's push-button telephone, with voice-mail capability, which transmits faxes and e-mail instantaneously around the world. Could he have anticipated the commercial importance of a seemingly simple addition to the telephone--the 800 number? One dare not ask what the strait-laced Bell might have thought of the introduction of the 900 number!

It is characteristic of a system that improvements in performance in one part are of only limited significance without simultaneous improvements in other parts. In this sense, technological systems may be thought of as comprising clusters of complementary inventions. Improvements in power generation can have only a limited impact on the delivered cost of electricity until improvements are made in the transmission network and the cost of transporting electricity over long distances. This need for further innovations in complementary activities is an important reason why even apparently spectacular breakthroughs usually have only a slowly rising productivity curve flowing from them. Within technological systems, therefore, major improvements in productivity seldom flow from single technological innovations, no matter how significant they may appear to be. At the same time, the cumulative effects of large numbers of improvements within a technological system eventually may be immense.

There is an additional reason why, historically, it often has proven so difficult to foresee the uses of a new technology. Many major inventions had their origins in an attempt to solve very specific, and often very narrowly defined, problems. However, once a solution has been found, it frequently turns out to have significant applications in totally unanticipated contexts. Much of the impact of new technologies is realized through intersectoral flows. Inventions have very serendipitous life histories.

The steam engine for example, was invented in the 18th century specifically as a device for pumping water out of flooded mines. For a long time, it was regarded exclusively as a pump. A succession of improvements later rendered the steam engine a feasible source of power for textile factories, iron mills, and an expanding array of industrial establishments. In the course of the early 19th century, it became a generalizable source of power and had major applications in transportation--railroads, steamships, and steamboats. Before the Civil War, the main use of the steam engine in the U.S. was not in manufacturing, but in transportation.

Later in the 19th century, the steam engine was used for a time to produce a new and even more generalizable source of power--electricity--that, in turn, satisfied innumerable final uses to which steam power itself was not directly applicable. Finally, the steam turbine displaced the steam engine in the generation of electric power. The special features of electricity--its ease of transmission over long distances, capacity for making power available in "fractionalized" units, and far greater flexibility of electrically powered equipment--sounded the eventual death knell of the steam engine itself.

The life history of the steam engine, then, was shaped by forces that hardly could have been foreseen by British inventors working on ways of removing water from increasingly flooded coal mines during George Washington's lifetime. Nevertheless, the very existence of the steam engine, once its operating principles had been understood thoroughly, served as a powerful stimulus to other inventions.

Inventions often arise as solutions to highly specific problems in a particular industry. Their subsequent inter-industry flow is bound to be highly uncertain. This is because the uses of a new technology in a quite different industrial context are especially difficult to anticipate. Yet, in some cases, a new technological capability may have not one, but multiple points of

Consider the impact of the computer upon air transportation. Its changing performance has been influenced at least as much by the application of the computer to new uses as by the research and development spending that has taken place within the industry itself.

* Supercomputers now perform a good deal of fundamental aerodynamic research, including much--but not all--that formerly was done in wind tunnels.

* Computers have been an important source of cost reduction in the design of specific components of aircraft, such as the wing. They played an important role in the wing designs of the Boeing 747, 757, and 767, as well as the Airbus 310.

* Computers now are responsible for much of the activity that takes place in the cockpit, including the automatic pilot.

* Together with weather satellites (a complementary technology that was developed many years later), which routinely determine the shifting location of high-altitude jet streams, computers widely are used in determining optimal flight paths. The annual fuel savings for the world commercial airline industry probably is well in excess of $1,000,000,000.

* Computers and computer networks are at the heart of the present worldwide ticketing and seat reservation system.

* Along with radar, the computer has become absolutely central to the operation of the air traffic control system.

* Computer simulation now is the preferred method of instruction in teaching neophytes how to fly.

There are many other forces that make technological prediction difficult, and perhaps none is greater than the elusiveness of imagination. Marconi had no sense of how the radio might enlarge human experience, but David Sarnoff, an uneducated Russian immigrant who headed RCA after World War I, had a lively vision of how it might transmit news, music, and other forms of entertainment into every household.

Similarly, Sony's Walkman takes existing technology (batteries, magnetic tape, earphones) and provides entertainment where it could not be delivered before--indeed, where no one had even thought of delivering it before.

Although the initial conception of the VCR had been of a capital good to be used by television stations, some American as well as Japanese participants were aware of the much larger home market possibilities. The crucial difference seems to have been the Japanese confidence, based upon their manufacturing experience, that they could come up with the necessary cost reductions and performance improvements. The rapid transformation of the VCR into one of Japan's largest export products was an achievement of both imagination and justified confidence in their engineering capabilities. The limited view once held of the potential for the VCR bears some parallels to the disdain of the mainframe computer makers toward the personal computer as it began to emerge about 15 years ago. It was then fashionable to dismiss the PC as a mere "hacker's toy," with no real prospects in the business world, and therefore no serious threat to the economic future of mainframes.

What can be learned from all this? First of all, there is humility. Innovation forecasting is extremely complex. Only the optimistic and naive would think that any single intellectual paradigm could provide a reliable guide to the future.

Second, and closely connected, is the importance of maintaining a critical attitude toward the notion of "relevance." The research community is being exhorted with increasing force to unfurl the flag of relevance to social and economic needs. The burden of much of what has been said here is that people frequently simply do not know what new findings may turn out to be relevant, or to what particular realm of human activity that relevance may eventually apply. A pervasive uncertainty not only characterizes basic research, where it generally is acknowledged, but the realm of product design and development as well. Consequently, early precommitment to any specific, large-scale technology project, as opposed to a more limited and sequential decision-making approach, is likely to be hazardous--i.e., wasteful. Evidence for this assertion abounds in such government-sponsored projects as weapons procurement, the space program, research on the development of an artificial heart, and synthetic fuels.

Given this inherent uncertainty, government ordinarily should resist championing any particular technology such as nuclear power and, instead, manage a deliberately diversified research portfolio. A fair criticism of the Federal government's postwar energy policy is not that it made a major commitment to nuclear power, but that it neglected alternative energy technologies that might have been invoked when nuclear power turned out to be problematic.

It is important not to make the same mistake with the information superhighway or other emerging technologies. Rather, government ought to open as many windows as possible, so that the private sector can explore a technological landscape that only can be discerned faintly from those windows.

Dr. Rosenberg, professor of economics, Stanford (Calif.) University, is director, Technology and Economic Growth Program, Stanford's Center for Economic Policy Research.

COPYRIGHT 1995 Society for the Advancement of Education
COPYRIGHT 2004 Gale Group