Engineering as a Social Enterprise (1991)

THOMAS P. HUGHES

During the past few decades, the scholarly study of the history of technology has made impressive strides toward placing technological change in a historical perspective and presenting the fruits of historical scholarship in a manner consonant with the experience of engineers, industrial scientists, and managers. Today we see technology as interacting with society; we now study technology as sociotechnical systems. The new history of technology shows that those who wish to preside over technological change need to master social, political, and economic factors as well as technical ones. This new history shows that the engineer's field of action is the sociotechnical system.

One purpose of this essay is to examine progress made in the field of the history of technology and to invite engineers, industrial scientists, and managers to draw on, and contribute to, this available body of knowledge. Another objective is to encourage students in engineering schools to broaden and enrich their understanding of technological change through the study of the new history of technology.

A quarter century ago, historians of technology were usually presenting earnestly detailed narratives of the development of machines, devices, and processes. Although scholarly and celebratory, the resulting articles and books did not analyze technological change. These histories were positivistic and reductionist in character, virtually ignoring nontechnical and nonscientific factors. Historians of technology now label this mode of presentation and interpretation

the internalist approach. It was implicit or explicit in its technological determinism and reductionist in its portrayal of modern technology as primarily applied science and economics.

Technological determinism embodies the widely held belief that hardware and software technology are the ultimate cause of social change. This assumption underlies much that we receive through our news media. We often hear that computers will change our work patterns, that western television will determine attitudes toward capitalism in Eastern Europe, and that nuclear weapons will maintain a lasting peace. Since World War II, it has been forecast that there would be numerous technological revolutions—among them atomic energy, the computer, and the information revolutions—followed by dramatic social changes. Even our scholarly histories still pretend to show how the Industrial Revolution in Britain ushered in modern times. One of the tenets of vulgar Marxism is a dogmatic insistence that technological change brings social change. Only rarely do we read or hear that values or social changes shape technology. When I was in engineering school, I often heard my professors dismiss politics as irrational and irrelevant, and identify technology as the root cause of all social improvement.

A History of Technology, published by Oxford University Press in the mid-1950s, clearly exemplifies the internalist genre in the field of history (Singer et al., 1954 –1958). Massively informative, painstakingly organized, and copiously illustrated, the five volumes covering the history until 1990 survey technological and applied science developments chronologically. The editors define technology as “how things are commonly done or made ” and “what things are done or made” (Singer et al., 1954, p. vii). Having virtually ignored the social relations of technology, they belatedly conclude their five volumes with an afterthought essay on “technology and its social consequences. ” In tone the work is internalist and unabashedly technologically deterministic.

The authors contributed chapters defined by engineering and industrial categories. For instance, chapters on electrical technology in the nineteenth century deal with “the generation of electricity” and “the distribution and utilization of electricity.” A cataloglike description of the basic discoveries of Alessandro Volta, Hans Christian Oersted, Michael Faraday, and others precede a chronicle describing progressively and more increasingly complex machines and processes, each assigned its basic technical characteristics and associated with the names of preeminent inventors and engineers. Magnetoelectric generators, for instance, give way to self-excited dynamos, and those supplying simple direct current to ones generating polyphase current. Such an approach leads the reader to conclude that technological change is essentially the progressive application of science to solve technical problems, which in turn results in an increasing variety of technical devices and processes of ever-increasing efficiency. Technological development thus takes place in a hermetically sealed world of invention, engineering, and science until the fruit of thought and

labor is loosed on the world to have its “social impact.” Such history is now termed Whiggish because it, like nineteenth century British liberal gentlemen, presupposed an uncritical belief in progress. One might dismiss this history as well-intentioned and harmless, but this is to overlook the probability that young engineers who are persuaded by the information and interpretation of A History of Technology are not likely to consider political and social factors as they design technology or aspire to preside over it.

Another salient characteristic of the internalist approach assumes that modern technology is mostly applied science. This interpretation is congruent with the argument often advanced in the past by science policy advisers that technological application will emerge willy-nilly from our support of pure science. For this reason, internalist history was and is usable history for those seeking support for pure science. History used in this way, however, leads to a misunderstanding of the nature of technological change, and perhaps, in the long run, to ineffective science policy. Today, the new history of technology offers a far more complex interpretation of the relation of science and technology and one far more in accord with the experience of engineers and industrial scientists.

As we have noted, internalist history of technology reinforces disciplinary and industrial category boundaries, taxonomies used often in engineering. For instance, chapters in the A History of Technology cover bridges and tunnels, the internal combustion engine, petroleum, machine tools, rubber, and mechanical road transport, to name only a few of the categories. Such organization facilitates the writing of a simple, clear narrative of technical developments but at the same time frustrates the presentation of the interconnections that transcend specialist categories. Using technological categories, for instance, precludes showing the interconnections among the internal combustion engine, petroleum engineering, and mechanical road transport and how such interactions like these helped bring about the modern, or second, industrial revolution. The internalist mode assigns plows to the category of food production and ignores social institutions like the medieval manor, so major sociotechnical changes such as the agricultural revolution of the early Middle Ages are overlooked. Our effort to understand technological change and to convey this understanding to others will continue to be severely handicapped if we employ only the internalist mode with its emphasis on individual artifacts evolving outside functional relationships to other artifacts and to social institutions.

When Lynn White, Jr., published his eloquent, erudite, and brilliant analysis of Medieval Technology and Social Change in 1962, he provided a new model of historical research and writing. He offered a viable alternative to an essential aspect of the internalist approach and new insights into the nature of technological change. He writes in a determinist mode, but his identification

FIGURE 1 Technology as systems.

of technology as systems rather than as isolated artifacts radically departs from the internalist approach. As a result, his history portrays complex and encompassing social change caused by evolving, interacting technology components such as plows, ox teams, and harnesses (Figure 1).

White argues that during an era when nine-tenths of the population of medieval Europe was involved in tillage, changes in the mode of plowing modified population, wealth, political relationships, leisure, and cultural expression (White, 1962, p. 39) and brought about one of the most prevailing and characteristic social institutions of the Middle Ages, the medieval manor. He also

persuasively portrays the introduction of the stirrup as dramatically altering the nature of medieval warfare in Western Europe. “Few inventions, ” he writes, “have been so simple as the stirrup, but few have had so catalytic an influence on history.” This advance ushered in the social system of feudalism dominated by an aristocracy of warriors endowed with land that sustained a new and highly specialized way of fighting. “Antiquity imaged the Centaur; the early Middle Ages made him the master of Europe” (White, 1962, p. 38).

We sample White's approach in more detail in his portrayal of the agricultural revolution of the Middle Ages, where he poses a problem familiar to engineers —that of transferring technology from one material and cultural environment to another. In late Roman and early medieval times, as people moved northward from the Italian peninsula, the peasant left the dry sandy earth and relatively dry climate to encounter the heavy alluvial soil of the river valleys and wet weather of northern Europe. In the south had evolved a technological system involving the scratch-plow with two oxen, shallow cross-plowing in easily pulverized, moisture-retaining solid, and square fields. After encountering the heavy, moisture-laden soil of the north, the tiller of the soil adopted a different system. The interacting components invented and developed included the wheeled heavy plow, with a coulter to cut vertically into the sod; a plowshare to slash the earth horizontally at the grass roots; and a mouldboard to turn the slice of earth to right or left. This heavier plow required eight yoked oxen to pull it. Because no cross-plowing was needed, the fields became long and narrow rather than square. This system substituted efficiently applied animal power for human energy and time and illustrates the sequential substitution of new components in a system of agriculture. ¹ Since few peasants owned eight oxen, they had to pool their animals, combine their land holdings into large open fields, and coordinate their planting and plowing. Communal decisions were made by a powerful village council of peasants. Thus did the characteristic manorial economy, a sociotechnical system, emerge. ²

Engineers, industrial scientists, and managers who preside over technological change today will also find congenial White's presentation of the dynamics of systems evolving. The introduction of the heavy plow, he reasoned, was only the first critical technological event of an extended sequence of innovations during the agricultural revolution of the early Middle Ages. Take for example the gradual substitution of the horse for the ox as draft animal. The introduction of a new harness and a nailed shoe made the horse an economic as well as military asset. A harnessed and shod horse moved so much more rapidly than the ox that, according to modern experiments, “he produces 50 percent more foot-pounds per second” and is also able to work one or two hours longer each day (White, 1962, p. 62). His circle of reasoning closes as he asserts that the new three-field system of crop rotation, stimulated by the introduction of the heavy plow, brought the cultivation of oats for the horse, and legumes for humans. With the cost of horse feed thus made lower than that of ox feed, the

horse repeatedly displaced the ox. White summarizes his account of systems evolving and revolutions transpiring as follows:

It is not merely the new quantity of food produced by improved agricultural methods, but the new type of food supply which goes far towards explaining, for northern Europe at least, the startling expansion of population, the growth and multiplication of cities, the rise in industrial production, the outreach of commerce, and the new exuberance of spirits which enlivened that age. In the full sense of the vernacular, the Middle Ages, from the tenth century onward, were full of beans (White, 1962, p. 76).

In history, as in science and technology, articulating patterns of change helps us see our world in new ways. White showed us systems of technological relationships, and other historians soon after explicitly identified technological systems in the historical events and structures they studied. Relationships and connections that had been previously overlooked in the source materials now emerged as historians pushed in new directions the idea of technology as evolving systems.

When I began writing a biography of Elmer Sperry, the biographical material published about his 1910 invention of the gyrocompass presented it as an isolated event in a series of seemingly unrelated inventive acts spread over his lifetime (Hughes, 1973). Internalist accounts of the invention only described the complex device and noted the substantial contribution made to navigation. If we consider the gyrocompass as a component and the ship as a system, however, then exciting insights into Sperry's invention of the compass follow. Since the mid-nineteenth century, steam engines displaced sails, wooden hulls gave way to iron ones, and electric motors and lights replaced steam-powered and petroleum-illuminated devices. An unintended consequence of the alteration of these components was the effect on other ship components. Changes cascaded through the system. For example, the ship was now filled with magnetic flux from the newly introduced iron hull, and electromagnetic fields generated by its electric motors, which affected the magnetic compass used to guide the wooden ship. Now the magnetic compass responded to these fields in addition to the magnetic field of the earth. Because of the improvement in gunnery and gunpowders, this malfunction became especially troublesome. With longer firing ranges possible, gunfire errors from flawed compass readings were magnified. Sperry and other inventors learned of this reverse salient in the evolving ship system and concentrated their creative talents on the solution of the problem (Figure 2). Research and development funds available because of an intensifying armaments race supported their inventive activities. By the eve of World War I, several inventors, including Sperry, had introduced the gyrocompass, a device unaffected by the irregularities of magnetic fields, but

FIGURE 2 Solving reverse salients.

dependent on the rotation of the earth. In this case, the inventorsresponded to a system undergoing dynamic change.

Based on my research on the development of the gyrocompass as a complex of systematic relationships, I then developed a model of technological systems evolving in which purposeful changes in some components in a technological system often lead to unintended malfunctions in others. Thus, alert inventors who monitor these omnipresent modern technological systems as they expand can then concentrate on inventions that solve the problems of the malfunctioning components. Once the system is in equilibrium, the opportunities for invention disappear, but as long as so-called improvements are being made, the cascading effect keeps inventors, industrial researchers, and others busy.

This Sperry episode, however, presumes too narrow a model of technological systems. Sperry also extended his horizons to encompass a sociotechnical system with nontechnical as well as technical components. Early in his career as an inventor, he found that new technology, including the gyrocompass, was rarely developed by existing institutions. Only infrequently did he locate an established manufacturing firm willing to abandon a line of products in which it had heavily invested skill, knowledge, and capital in order to develop a new innovation unrelated to its investment. So, on numerous occasions he, the inventor-entrepreneur, had to invent not only a device but an institution for manufacturing and marketing as well. Perusing his papers, one realizes that he drew no distinctions among the technical, the economic, and the institutional. The proposition that creativity is holistic can be grounded in empirical evidence. As he invented, Sperry wove a seamless web of technology and institutional change.

It was Alfred D. Chandler who further enlarged our historical horizons with his Pulitzer prize-winning volume, The Visible Hand (1977), in which he stresses the interaction of technology and institutions, especially the means of production and the business corporations that managed these. Not unlike White, Chandler tends to technological determinism. He argues that modern management practices make it possible for us to enter the doors that technology opens. He is especially adept at showing how modern management used the infrastructure of communication and transportation to coordinate and control the interacting means of production and distribution. Using examples from the period of about 1850 to 1950, when the market for goods in the United States was expanding, Chandler shows how the visible hand of management ensures the smooth flow of materials and energy through the stages of production and distribution, which continuously evolve because of technological innovation.

From Chandler, we learn to understand technological change better if we take into consideration the role of the manager and the firm. White's insistence that the multioxen team and deep plow brought about the institution of the manor, which in turn presided over the new technology, finds a counterpart in Chandler's argument that modern large-scale, capital-intensive technology paves the way for the multidivisional corporation which then rationalizes production through management. Chandler demonstrates that the implications of the technology revolution that transpired from about 1880 to 1930—the Second Industrial Revolution —would not have been realized without managerial, or organizational, innovations. In other words, the introduction of modern multidivisional business corporations contributed as much to increased production and productivity during the Second Industrial Revolution as did, say, the electric power system. Historians of technology attempting to explain the Second

Industrial Revolution now realize that, in addition to lectures on machine tools, steel production, and petroleum refining, they must include the rise of modern scientific management among their lecture topics.

In his new book, Scale and Scope, Chandler gives an example of how modern corporations realize the implications of technological change. After 1882, in order to maximize the flow of materials from well to consumer, managers of the Standard Oil Trust integrated new modes of petroleum refining and highly developed means of transportation and communication. Muckraking historians have characterized the formation of such trusts as efforts to establish profit-gouging enterprises. Chandler, by contrast, explains how the Standard Oil Trust provided a legal instrument to exploit the potential of modern technology. The management of the trust, through stock interchanges and other financial devices, controlled a loose federation of 40 companies, which owned oil fields, refineries, and transportation networks. Afterward, rationalization concentrated about a quarter of the world's kerosene production into three large and technically efficient refineries. As a result, economies of scale drastically lowered the cost of producing kerosene (Chandler, 1990, p. 24–25). Chandlerian managers fulfilled the goal that Thorstein Veblen defined early in this century when he called for production engineers to transform the nation's industry into “a system of interlocking mechanical processes ” (Veblen, 1921, pp. 72–74).

Although more sophisticated and subtle in their interpretations of technological change than the internalists, White and Chandler are, as we have noted, technological determinists. It was left to other historians to successfully attack the citadel of technological determinism and provide fresh understanding of the nature of technological change. Several decades before White and Chandler, Louis Hunter, an American economic historian, wrote vividly and concretely about the way in which geographic forces shape technology. Considering the persuasiveness of Hunter's monograph, one is surprised that internalist, technological-determinist history continued to prevail. Perhaps this is because Hunter was an economic historian, considered outside the mainstream of the historians of technology in his time.

Hunter's seminal work bears the memorable title Steamboats on the Western Rivers (1949). It is a marvelous example of comparative history on why steamboats navigating eastern coastal waters and rivers differed in technical detail from those on the Mississippi and it tributaries. After reading Hunter, it is difficult to argue that there is one best way to do technology, regardless of place. Hunter takes an ecological approach by placing technology in a geographic setting and showing how the technical characteristics mirror geographical features.

Robert Fulton, a steamboat inventor, made the mistake of assuming that

technology could be transferred from one geographical region to another without substantial modification. After successfully introducing steamboat transportation along the Hudson river, he obtained a monopoly to provide steamboat transportation along the Mississippi. He persuaded Nicholas Roosevelt, a New Jersey steam-engine builder, and workmen from New York to set up a factory in Pittsburgh to build steamboats much like those used on the Hudson and eastern coastal waters. The engines were efficient, low-pressure, and condensing, and the hulls were long and narrow, sitting deep in the water.

Within a few years, however, by 1818, Fulton abandoned his monopoly because his type of steamboat was superseded by broad-beamed, shallow-draft vessels with relatively inefficient, noncondensing, high-pressure steam engines. This new design, of indigenous origin, resulted from the empirical instincts and craft skills of engine designers and shipbuilders familiar with the geography of the Mississippi and its tributaries (Figure 3). Their high-pressure engine without condenser and air pump was lighter than the condensing engine, so their vessel rode higher in the shallows of the Mississippi, which was plagued with sandbars and snags. Their high-pressure engine offered more reserve power to negotiate extreme variations of current and depth; the noncondensing engine was not befouled by the muddy western waters; the less efficient engine used low-cost wood fuel in the still unsettled Mississippi Valley; and the broad-beamed hull had less draft than the eastern boats customarily used on the bays and tributary rivers. From Hunter 's monograph historians and students learn about the realities of technology transfer and the absurdity of arguing that there is one best engineering solution.

Hunter shows us that the concept of technological determinism alone fails to encompass the complexity of technological change. He demonstrates that geography often shapes or determines the development of technology. Other historians have recently shown how factors other than geography can also shape technology. If we consider the history of the comparative developments of electric light and power systems in the United States, Great Britain, and Germany early in this century, we find technological determinism once again flawed. In this case political factors shaped the technology. The story begins by comparing the backward electric supply system of London with that of Chicago. The Chicago supply had been consolidated into a single system with a high load factor and low cost, supplied by several large turbine-driven power stations, while greater London had 65 electrical utilities, 70 generating stations averaging a little over 5,000 kilowatts in capacity, and 49 different types of supply systems, 10 different frequencies, 32 voltage levels for transmission and 24 for distribution. A Londoner might toast bread in the morning, light the office during dark days, walk home under street lights, and read by lamps at

FIGURE 3 Geographical determinism.

home—all with different kinds of electricity. The overall load factor was low, costs were high, and usage was small by comparison with other metropolitan centers, especially those in the United States and Germany (Hughes, 1983, p. 227). Such was the British electrical industry lag, a cause for concern in an era when electric supply was a symbol of progress and power, political as well as technological.

An internalist historian would seek the cause of Britain's technological malaise in generators, motors, transformers, and engineers less efficient and science less well developed than those in Germany or the United States. Resort to the original sources does not sustain such a thesis. British science, especially physics, enjoyed great prestige, with achievement to match; the red brick universities offered electrical engineering courses comparable in content to those in the United States and Germany, and the many and varied power

stations were small but recognized as exceedingly well designed by painstaking engineers who used electrical equipment manufactured not only in Britain, but imported from, or designed in, the United States as well as Germany. The question of the lag could only baffle the internalist.

If, however, the historian looks beyond the confines of the world of technical and scientific affairs, beyond the reach of the internalist and the determinist, clues about the causes of the British electrical industry lag can be found. Original sources reveal that experienced British engineers and managers did look beyond the dynamos, transformers, and motors, all the way to the Houses of Parliament, where since 1882 a series of legislative enactments regulating electrical supply had emerged. About this legislation the engineers of the venerable Institution of Electrical Engineers declared

We hold that electrical enterprises should have their limits and boundaries set by economical considerations only, and that arbitrary boundaries, mostly of medieval ecclesiastical origin, should not limit the distribution or the growth of electrical systems (Hughes, 1983, pp. 233–234).

This text expressed the fundamental modern tension between a technocratic vision and a political ideology. The boundaries of medieval origin are those of the traditional local governments in Britain, the vestries, parishes, and counties. The British had long cherished these institutions and they realized that if technological systems, such as the electrical, transcended them geographically, they would likewise transcend their local authority. Local governments could not regulate or acquire electric light and power systems that spread over and beyond their political jurisdiction. Transformers in one authority and power plants in another would frustrate regulation then and now.

This episode provides a striking revelation of legislation shaping technology. One can probe even deeper and observe political power and values shaping the legislation and the technology. Those 75 power stations in London clearly state the message: “we are modern artifacts but we are shaped by values originating in medieval England.” This is not technological determinism; this is value construction of technology. In this instance, technology-shaping values were institutionalized into the legislation.

By 1980, historians of technology had abandoned the deterministic simplicities of the internalist approach. Technological systems for many had become the unit of study, and many saw geographical and political, as well as economic, factors as shaping technology. Some sociologists interested in technology were soon insisting that social factors must be there as well. In 1984, historians of like mind joined the sociologists at a conference in the Netherlands and articulated the social construction approach. From this conference emerged a volume of papers providing generalized argument and case histories. It was

appropriately titled, The Social Construction of Technological Systems (Bijker et al., 1987). In their essays, the authors identified professional status, education and practical experience, economic and institutional interests, and values among those social forces shaping technology.

In essence, social construction holds that technological systems bear the imprint of the social context in which they arise (Pfaffenberger, 1990, p. 15). Because many design alternatives exist, social interests, as well as other nontechnical factors, can play key roles in shaping the artifacts produced or the style of a system constructed from these artifacts. Pinch and Bijker (1987, pp. 17–50) show that during the late nineteenth century, different social groups such as sports cyclists, tourist cyclists, elderly men, and woman cyclists posed different sets of problems to which different producers responded with various solutions called bicycles. It was not at all clear to contemporaries which variants would “die” and which would “survive.” The power and strength of the various social groups contending to define a bicycle ultimately determined its technical characteristics, including the meaning attached to it. There was no one best solution, but a number of solutions, the surviving one or ones emerging from the context of social interest groups. Few would dispute the argument presented by Pinch and Bijker, but how many of us see, as they do, technology as congealed social interests?

We find another example of the social construction of technology in the history of the development of a cooling system for atomic reactors built during World War II. From the early days of the Manhattan Project, engineers and scientists differed about the best method of cooling plutonium-generating reactors. Some advocated gas and others liquid cooling. There was also disagreement over which gas and which liquid. One might analyze the episode in an internalist way and simply suggest that the search continued until the one best solution was found. Reflection and a social constructivist mode of interpretation result instead in a more complex and realistic analysis. The “best way” for engineers turned out to be different from the “best way” for scientists. Generally the engineers, especially those of the DuPont Company, following the dictates of experience, conservatively opted for gaseous cooling, either helium or air. Physicist Leo Szilard wanted to use the exotic coolant liquid bismuth. (With Albert Einstein he had patented in the 1930s a refrigerator using a liquid metal as a coolant.) Drawing on theoretical analysis, Eugene Wigner, also a physicist, recommended water cooling, the mode eventually adopted (Hughes, 1990, pp. 391–394). Professional background and experience clearly shaped the positions taken. There are a number of cases in the early history of nuclear energy showing how engineers differed from physicists on the best solution for technical problems.

Donald MacKenzie (1987) offers a variation of the social construction approach that introduces interaction between technology and social context. MacKenzie shows how interest groups with different technical capabilities and

strategic and tactical objectives influenced the accuracy of missile guidance systems and how, in turn, the development of missile accuracy shaped strategic and tactical objectives. “Technological developments, ” he argues, “cannot be treated in isolation from organizational, political, and economic matters.” This argument is not a contribution to the old debate on the social responsibility of the engineer: it contributes to our understanding of the nature of engineering design and technological change. From his interviews with engineers in the missile guidance field MacKenzie concludes that there is a rough correlation between how articulate they are on issues involving interactions of technical and nontechnical factors, and their worldly success.

He shows how successful system builders such as Charles Stark Draper engineered technical, economic, and political matters simultaneously. The drive of Draper and his laboratory for maximum missile accuracy was determined not only by technical considerations, but by the choice of problems that they, with their particular interests and capabilities, could solve. They believed that the historical experiences of their laboratory had given it a bias toward solving problems of accuracy at the expense at times of other problems such as reliability and cost. Other laboratories and manufacturers working in concert with government agencies decided on missile guidance technology that emphasized reliability and cost (MacKenzie, 1987, pp. 195–222). Clearly there was more than one best solution, a view rarely encountered in journalistic discussions of technology or in the curricula of our engineering schools.

As we have noted, the internalist approach to technology was wedded not only to technological determinism but also to the proposition that technology is applied science. This internalist persuasion was reinforced after World War II by influential technology policymakers, such as Vannevar Bush, a wartime mobilizer of science and technology and postwar author of Endless Horizons (1946), who wrote:

Basic research leads to new knowledge. It provides scientific capital. It creates the fund from which the practical applications of knowledge must be drawn. New products and new processes do not appear full-grown. They are founded on new principles and new conceptions, which in turn are painstakingly developed by research in the purest realms of science (Bush, 1946, pp. 52–53).

No clearer argument that technology is applied science could be made. This view was echoed repeatedly both by those seeking funding for pure science and those seeking status for engineering by associating it with prestigious postwar physics.

Research by historians, however, did not support the Bush proposition. In still-frequently cited articles written in the early 1970s, Edwin T. Layton, Jr.,

FIGURE 4 Science of technology.

counters both Bush's argument and the similar one found in the internalist A History of Technology. Layton believes that knowledge generated by science is not sufficient for engineers involved in designing artifacts and technological systems (Figure 4). “Science may indeed influence technology. . . . But this does not provide an adequate explanation of most technological change” (Layton, 1974, p. 39). Technology, he maintains, is an independent system of thought different from—and derived from sources not limited to —science. He refers to the mirror-image characteristics of the technological and scientific communities. The differences inhere in the ends of the two communities, for scientists seek to know, and technologists seek to design means to fulfill societal ends (Layton, 1971, pp. 576–578). ³

In a series of four articles in Technology and Culture, Walter Vincenti, an aeronautical engineer and historian, also provides a thoughtful and informative reading of the technology-science relationship. Simple concepts of technology as applied science did not survive his scrutiny. He explores in depth various aspects of engineering knowledge in the contexts of experimental research, theoretical analysis, and production and design. He is especially interested in discovering whether or not peculiarly technological methodologies exist. To formulate and test his hypotheses, he examines several case histories, including the air-propeller tests of W. F. Durand and E. P. Lesley, the development of control-volume analysis from 1910 to 1925, the innovation of flush riveting in American airplanes, and the problem of airfoil design in connection with the Davis wing (Vincenti 1979; 1982; 1984; 1986).

Vincenti's studies lead him to conclude that technological methods of experimentation differ both in form and object from those in the physical sciences. Technological methods permit the engineer to obtain information needed for design when no usable theoretical knowledge is available. Vincenti also finds that an engineering science exists and that it differs from other science in that the engineering science is acquired in order to serve the needs

of engineers who design artifacts such as fluid flow devices. Moreover, in the case of airfoil technology, Vincenti decides that over a period of years design engineers moved from a methodology that was essentially cut-and-try to one laden with theory derived from the cumulative experience of design and use. As a result, they were able to greatly reduce uncertainty. These and his other conclusions leave no doubt that engineering science and experimentation have developed alongside methodologies in natural sciences. From Vincenti we have more evidence that Vannevar Bush and others were far off the mark when they insisted that engineering, or technology, was essentially the application of natural science.

Over the past few decades the historian's understanding of the nature of technological change has developed dramatically. Today most historians of technology consider a descriptive narrative that ignores the interacting technical, economic, political, and social components of technological change reductionist and distorting. New modes of research and presentation take the technological, or the sociotechnical, system rather than individual artifacts as the unit of study. Deterministic dynamos have given way to seamless-web systems.

In my writing on the history of electrical technology, I present it as an evolving system of interaction among such components as turbines, dynamos, high-voltage grids, electrical utility management structure, electrical engineering departments in engineering schools, investment banks, regulatory agencies, and the weight of public opinion (Figure 5). Presiding over sociotechnical systems are system builders, including inventors, engineers, managers, and financiers who have prevailed during successive stages of system growth.

Epistemological developments of the magnitude of those taking place in the study of the history of technology during recent decades would surely be considered highly significant in a field of engineering or science. I suggest that engineers and scientists consider the impact that the new history of technology should have on their understanding of technological change.

The understanding derived from the new history can be used to encourage engineers and industrial scientists to present a more complex and less reductionist view when they explain technological change to laymen and policymakers. A reductionist view often leads to the application of unworkable technical fixes for problems that are in essence complex, multifaceted systems. Furthermore, the presentation of this more complex view of change will help laymen and policymakers appreciate that technological change, like political change, is complicated and deserves the attention of our most capable and imaginative leaders.

This new understanding can also be used to reform engineering and science education. At this time and for decades past, U.S. engineering and science

FIGURE 5 Sociotechnical system.

schools have left the impression with their graduates—in part due to the intellectual ambience of the engineering school that inculcates an unreflective technological determinism—that technological change results simply from the solution of technical and scientific problems. Students leave school with the

mistaken impression that political and social factors are extraneous. If the arguments offered in this essay are valid, then these young professionals are being ill-prepared to preside over either technological change or the development of sociotechnical systems. Experience may in time relieve them of this indoctrination, but why the prolongation of innocence in engineering education? Those of us who teach history to those who will become lawyers, businessmen, politicians, and other makers of the modern world do not dare shield them from the realities of change.

Courses introducing the complex realities of technological change should not be relegated to elective status for professionally oriented students. Doing so implies that the social and political components of change are peripheral. As I have shown in this essay, the political or social components and the problems they present are often the core constituent of the problem complex to which the engineers and scientists must address themselves. I can provide countless examples of experienced engineers and scientists who focused their energies on political and social matters in order to bring about innovation. If we do not prepare engineers and scientists for this imaginative flexibility, then we must relegate responsibility for long-range technological change to other professions.

1. White notes that historians and archaeologists have many exceptions to the rule that the heavy plow brought a system of cultivation, but he believes that this was the typical development.

2. This follows White's argument as he presents it in Medieval Technology and Social Change, pp. 39–57.

3. For an informed and thoughtful essay on the science-technology relationship, see John M. Staudenmaier (1985, pp. 83–120).

Bijker, W., T. Hughes, T. Pinch, eds. 1987. The Social Construction of Technological Systems: New Directions in the Sociology and History of Technology. Cambridge, Mass. : MIT Press.

Bush, V. 1946. Endless Horizons. Washington, D.C. : Public Affairs Press. Reprint, 1975. New York : Arno Press.

Chandler, A. 1990. Scale & Scope: The Dynamics of Induslrial Capitalism. Boston : Harvard University Press.

Chandler, A. 1977. The Visible Hand: The Managerial Revolution in AmericanBusiness. Cambridge, Mass. : Belknap Press.

Hughes, T. 1973. Elmer Sperry: Inventor and Engineering. Baltimore : Johns Hopkins University Press.

Hughes, T. 1983. Networks of Power: Electrification in Western Society, 1880–1930. Baltimore : Johns Hopkins University Press.

Hughes, T. 1990. American Genesis: A Century of Invention and Technological Enthusiasm, 1870–1970. New York : Penguin.

Hunter, L. 1949 Steamboats on the Western Rivers, An Economic and Technological History Cambridge, Mass. : Harvard University Press

Layton, E. 1971 Mirror-image twins: The communities of science and technology in 19th century America Technology and Culture 12 : 562–80

Layton, E., 1974 Technology as knowledge. Technology and Culture 15 :31–41

MacKenzie, D. 1987 Missile accuracy: A case study in the social processes of technological change Pp. 195–222 in The Social Construction of Technological Systems: New Directions in the Sociology and History of Technology, W. Bijker, T. Hughes, and T. Pinch, eds. 1987 Cambridge, Mass. : MIT Press

Pfaffenberger, B. 1990 Democratizing Information: Online Databases and the Rise of End-user Searching Boston : G. K. Hall & Company

Pinch, T., and W. Bijker 1987 The social construction of facts and artifacts: Or how the sociology of science and the sociology of technology might benefit each other Pp. 17–50 in The Social Construction of Technological Systems: New Directions in the Sociology and History of Technology, W. Bijker, T. Hughes, and T. Pinch, eds. 1987 Cambridge, Mass. : MIT Press

Singer, C., E. J. Holmyard, A. Hall, and T. Williams, eds. 1954–58 A History of Technology New York : Oxford University Press

Staudenmaier, J. M. 1985 Teehnology's Storytellers: Reweaving the Human Fabric Cambridge, Mass. : MIT Press

Veblen, T. 1921 The Engineers and the Price System Second Edition 1963 New York : Harcourt, Brace and World

Vincenti, W. 1979 The air-propeller tests of W. F Durand and E. P. Lesley: A case study in technology methodology Technology and Culture 20 : 712–751

Vincenti, W. 1982 Control-volume analysis: A difference in thinking between engineering and physics Technology and Culture 23 : 145–174

Vincenti, W. 1984 Technological knowledge without science: The innovation of flush riveting in American airplanes, ca. 1930–ca. 1950 Technology and Culture 25 : 540–576

Vincenti, W. 1986 The Davis wing and the problem of airfoil design: Uncertainty and growth in engineering knowledge Technology and Culture 27 : 717–758

White, L., Jr. 1962 Medieval Technology and Social Change. Oxford: Clarendon Press