Technology Definition and History PDF

1 Can We Define “Technology”? One way to define “technology” is in terms of evolution. An animal may briefly use a natural object, such as a branch or a stone, for a purpose, but it was long thought that only human beings intentionally made objects, such as a rake or a hammer, for certain func...

1 Can We Define “Technology”? One way to define “technology” is in terms of evolution. An animal may briefly use a natural object, such as a branch or a stone, for a purpose, but it was long thought that only human beings intentionally made objects, such as a rake or a hammer, for certain functions. Benjamin Franklin and many others thought that tool use separated humans from all other creatures. Recent fieldwork complicates the picture. Jane Goodall watched a chim- panzee in its own habitat. It found a twig of a certain size, peeled off its bark, looked for a termite hill, thrust in the peeled twig, pulled it out covered with termites, and ate them. This chim- panzee not only made a tool, it did so with forethought. In 2004, scientists announced discovery of the bones of a previously unknown species in an Indonesian cave. Standing only three feet high, this dwarf species lived and used tools as recently as 12,000 years ago.1 Yet if Franklin’s idea needs modification, it seems that only intelligent apes and human species are toolmakers, while the vast majority of animals are not. Birds construct nests. Beavers cut down trees and build dams. Ants and bees build complex commu- nities that include a division of labor and food storage. But only a few species have made tools. Notable is a hand axe widely used by Homo erectus 1.6 million years ago. 2 Chapter 1 Homo sapiens have used tools for at least 400,000 years, and seem to have done so from their first emergence. Technologies are not foreign to “human nature” but inseparable from it. Our ancestors evolved an opposition between thumb and fingers that made it easier to grasp and control objects than it is for other species. Indeed, prehensile hands may even have evolved simultaneously with the enlarging human cortex. Learning to use tools was a cru- cial step in the species’ development, both because it increased adaptability and because it led to a more complex social life. Using tools, the relatively weak Homo sapiens were able to capture and domesticate animals, create and control fire, fashion artifacts, build shelters, and kill large animals. Deadly tools also facilitated murder and warfare. Tools emerged with the higher apes, and one might argue that humanity fashioned itself with tools.2 The central purpose of technologies has not been to provide necessities, such as food and shelter, for humans had achieved these goals very early in their existence. Rather, technologies have been used for social evolution. “Technology,” José Ortega y Gasset argued, “is the production of superfluities—today as in the Paleolithic age. That is why animals are atechnical; they are content with the simple act of living.”3 Humans, in contrast, con- tinually redefine their necessities to include more. Necessity is often not the mother of invention. In many cases, it surely has been just the opposite, and invention has been the mother of necessity. When humans possess a tool, they excel at finding new uses for it. The tool often exists before the problem to be solved. Latent in every tool are unforeseen transformations. Defining technology as inseparable from human evolution sug- gests that tools and machines are far more than objects whose meaning is revealed simply by their purposes. As the great stone circle at Stonehenge reminds us, they are part of systems of mean-  Can We Define “Technology?” 3 ing, and they express larger sequences of actions and ideas. Ulti- mately, the meaning of a tool is inseparable from the stories that surround it. Consider the similarity between what is involved in creating and using a tool and the sequence of a narrative. Even the chimpanzee picking up and peeling a twig to “fish” for termites requires the mental projection of a sequence, including an ini- tial desire, several actions, and successful feeding. The sequence becomes more complex when more tools are involved, or when the same tool is used in several ways. Composing a narrative and using a tool are not identical processes, but they have affinities. Each requires the imagination of altered circumstances, and in each case beings must see themselves to be living in time. Making a tool immediately implies a succession of events in which one exercises some control over outcomes. Either to tell a story or to make a tool is to adopt an imaginary position outside immediate sensory experience. In each case, one imagines how present cir- cumstances might be made different. When faced with an inadvertently locked automobile with the keys inside, for example, one has a problem with several possible solutions—in effect, a story with several potential endings. One could call a locksmith, or one could use a rock to break one of the car’s windows. Neither is as elegant a solution as passing a twisted coat hanger through a slightly open window and lifting the door handle from the inside. To improvise with tools or to tell stories requires the ability to imagine not just one outcome but several. To link technology and narrative does not yoke two disparate sub- jects; rather, it recalls an ancient relationship. Tools are older than written language (perhaps, as the chimpan- zee’s “fishing stick” suggests, even older than spoken language) and cannot merely be considered passive objects, or “signifieds.” Tools are known through the body at least as much as they are 4 Chapter 1 understood through the mind. The proper use of kitchen utensils and other tools is handed down primarily through direct observa- tion and imitation of others using them. Technologies are not just objects but also the skills needed to use them. Daily life is saturated with tacit knowledge of tools and machines. Coat hangers, water wheels, and baseball bats are solid and tangible, and we know them through physical experiences of texture, pressure, sight, smell, and sound during use more than through verbal descrip- tion. The slightly bent form of an American axe handle, when grasped, becomes an extension of the arms. To know such a tool it is not enough merely to look at it: one must sense its balance, swing it, and feel its blade sink into a log. Anyone who has used an axe retains a sense of its heft, the arc of its swing, and its sound. As with a baseball bat or an axe, every tool is known through the body. We develop a feel for it. In contrast, when one is only looking at an axe, it becomes a text that can be analyzed and placed in a cultural context. It can be a basis for verifiable statements about its size, shape, and uses, including its incorporation into literature and art. Based on such observations, one can construct a chronol- ogy of when it was invented, manufactured, and marketed, and of how people incorporated it into a particular time and place. But “reading” the axe yields a different kind of knowledge than using it. Telling stories and using tools are hardly identical, but there are similarities. Each involves the organization of sequences, either in words or in mental images. For another investigation it might be crucial to establish whether tools or narratives came first, but for my argument it matters only that they emerged many millennia ago. I do not propose to develop a grand theory of how human consciousness evolved in relation to tools. But the larger temporal framework is a necessary reminder that tools existed long before  Can We Define “Technology?” 5 written texts and that tools have always embodied latent narra- tives. My definition of technology does not depend on fixing pre- cisely when humans began to use tools, although it is pertinent that they did so thousands of years before anyone developed tools for writing. Cultures always emerge before texts. Long before the advent of writing, every culture had a system of artifacts that evolved together with spoken language. Objects do not define words, or vice-versa; both are needed to construct a cultural world. Only quite late in human development did anyone develop an alphabet, a stylus to mark clay tablets, or a quill adapted for writ- ing on paper. Storytelling was oral for most of human history. A tool always implies at least one small story. There is a situa- tion; something needs doing. Someone obtains or invents a tool in order to do it—a twisted coat hanger, for example. And after- wards, when the car door is opened, there is a new situation. Admittedly, this is not much of a narrative, taken in the abstract, but to conceive of a tool is to think in time and to imagine change. The existence of a tool also immediately implies that a cultural group has reached a point where it can remember past actions and reproduce them in memory. Tools require the ability to recollect what one has done and to see actions as a sequence in time. To explain what a tool is and how to use it seems to demand narrative. Which came first? This may be a misleading question. It seems more likely that storytelling and toolmaking evolved symbioti- cally, analogous to the way that oral performances are inseparable from gestures and mimicry. It is easy to imagine human beings as pre-literate, but it is diffi- cult to imagine them as pre-technological. Most Native American peoples, for example, did not write, but they did develop a wide range of tools, including snowshoes, traps, tents, drums, hatchets, bows, pottery, ovens, bricks, canals, and irrigation systems. All 6 Chapter 1 social groups use tools to provide music, shelter, protection, and food, and these devices are inseparable from verbal, visual, and kinetic systems of meaning. Each society both invents tools and selects devices from other cultures to establish its particular tech- nological repertoire of devices. In Herman Melville’s Moby Dick, Queequeg, a South Sea har- pooner visiting Nantucket, was offered a wheelbarrow to move his belongings from an inn to the dock. But he did not understand how it worked, and so, after putting all his gear into the wheel- barrow he lifted it onto his shoulders. Most travelers have done something that looked equally silly to the natives, for we are all unfamiliar with some local technologies. This is another way of saying that we do not know the many routines and small narra- tives that underlie everyday life in other societies. As the evolutionary perspective shows, technology is not some- thing new; it is more ancient than the stone circles at Stonehenge. Great stone blocks, the largest weighing up to 50 tons, rise out of the Salisbury Plain, put precisely into place in roughly 2000 B.C. The stones were not quarried nearby, but transported 20 miles from Marlborough Down. The builders contrived to situate them in a pattern of alignment that still registers the summer solstice and some astronomical events. The builders acquired many tech- nologies before they could construct such a site. Most obvi- ously, they learned to cut, hoist, and transport the stones, which required ropes, levers, rollers, wedges, hammers, and much more. Just as impressive, they observed the heavens, somehow recorded their observations, and designed a monument that embodied their knowledge. They did not leave written records, but Stone- henge stands as an impressive text from their culture, one that we are still learning to read. Transporting and placing the mas-  Can We Define “Technology?” 7 sive stones can only be considered a technological feat. Yet every arrowhead and potshard makes a similar point: that human beings mastered technologies thousands of years ago. Stonehenge suggests the truth of Walter Benjamin’s observation that “tech- nology is not the mastery of nature but of the relations between nature and man.”4 Technologies have been part of human society from as far back as archaeology can take us into the past, but “technology” is not an old word in English. The ancient Greeks had the word “techne,” which had to do with skill in the arts. Plato and Plotinus laid out a hierarchy of knowledge that stretched in an ascending scale from the crafts to the sciences, moving from the physical to the intellectual. The technical arts could at best occupy a middle position in this scheme. Aristotle had a “more neutral, simpler and far less value-laden concept of the productive arts.”5 He dis- cussed “techne” in the Nicomachean Ethics 6 (book 6, chapters 3 and 4). Using architecture as his example, he defined art as “a rational faculty exercised in making something... a productive quality exercised in combination with true reason.” “The business of every art,” he asserted, “is to bring something into existence.” A product of art, in contrast to a product of nature, “has its efficient cause in the maker and not in itself.”7 Such a definition includes such actions as making pottery, building a bridge, and carving a statue. Just as important, Aristotle related the crafts to the sci- ences, notably through mathematics. In Greek thought as a whole, however, work with the hands was decidedly inferior to philosophical speculation, and “techne” was a more restricted term than the capacious modern term “technology.” Perhaps because the term was more focused, classical thinkers realized, Leo Strauss wrote, “that one cannot be distrustful of political or social change without being distrustful of technological change.”8 As 8 Chapter 1 Strauss concluded, they “demanded the strict moral-political supervision of inventions; the good and wise city will determine which inventions are to be made use of and which are to be sup- pressed.”9 The Romans valued what we now call technology more highly than the Greeks. In De Natura Deorum Cicero praised the human ability to transform the environment and create a “second nature.” Other Roman poets praised the construction of roads and the pleasures of a well-built villa. Statius devoted an entire poem to praising technological progress, and Pliny authored prose works with a similar theme.10 Saint Augustine synthesized Plato and Aristotle with Cicero’s appreciation of skilled labor: “... there have been discovered and perfected, by the natural genius of man, innumerable arts and skills which minister not only to the necessities of life but also to human enjoyment. And even in those arts where the purposes may seem superfluous, per- ilous and pernicious, there is exercised an acuteness of intelli- gence of so high an order that it reveals how richly endowed our human nature is.”11 In contrast, Thomas Aquinas characterized the mechanical arts as merely servile.12 Some medieval thinkers, notably Albertus Magnus, appreciated iron smelting, the con- struction of drainage ditches, and the new plowing techniques that minimized erosion. A few drew upon Arabic thought, which presented the crafts as practical science and applied mathemat- ics. Roger Bacon, in his Communia Mathematica, imagined flying machines, self-propelled vehicles, submarines, and other con- quests of nature. Bacon put so much emphasis on the practical advantages of experiment and construction of useful objects that he “came close to reversing the usual hierarchy of the speculative and useful in medieval thought.”13 The full expression of a modern attitude toward technology appeared only centuries later, during the Renaissance, notably in  Can We Define “Technology?” 9 Francis Bacon’s New Atlantis (1627). Bacon imagined a perfect society whose king was advised by scientists and engineers organ- ized into research groups at an institution called Saloman’s House. They could predict the weather, and they had invented refrigera- tion, submarines, flying machines, loudspeakers, and dazzling medical procedures. Their domination of nature, which had no sinister side effects, satisfied material needs, abolished poverty, and eliminated injustice. This vision helped to inspire others to found the Royal Society.14 Established in London in 1662, this society institutionalized the belief that science and invention were the engines of progress. The Royal Society proved to be a per- manent body, in contrast to earlier, temporary groups that could also be seen as originators of modern research, such as those gath- ered in Tycho Brahe’s astronomical observatory on an island near Copenhagen, or Emperor Rudolf’s group of technicians and sci- entists in Prague. Today, a large bookstore typically devotes a section to the his- tory of science but scatters books on technological history through many departments, including sociology, cultural studies, women’s studies, history, media, anthropology, transportation, and do-it-yourself. The fundamental misconception remains that practical discoveries emerge from pure science and that technol- ogy is merely a working out or an application of scientific princi- ples. In fact, for most of human history technology came first; theory came along later and tried to make sense of practical results. A metallurgist at MIT, Cyril Stanley Smith, who helped design the first atomic bombs at Los Alamos, declared: “Tech- nology is more closely related to art than to science—not only materially, because art must somehow involve the selection and manipulation of matter, but conceptually as well, because the technologist, like the artist, must work with unanalyzable 10 Chapter 1 complexities.”15 Smith did not mean that these complexities are forever unanalyzable; he meant that at the moment of making something a technologist works within constraints of time, knowledge, funding, and the materials available. It is striking that he advances this argument when discussing the construc- tion of the first atomic bomb, which might seem to be the perfect example of an object whose possibility was deduced from pure science alone. However, Smith is correct to emphasize that the actual design of a bomb required far more than abstract thinking, particularly an ability to work with tools and materials. In fact, one sociologist of science has concluded that, although we cannot turn back the clock and “unlearn” the science that lies behind nuclear weapons, it is conceivable that we will manage to lose or forget the practical skills needed to make them.16 As Smith further pointed out, technology’s connection to science is generally misunderstood: “Nearly everyone believes, falsely, that technology is applied science. It is becoming so, and rapidly, but through most of history science has arisen from problems posed for intellectual solution by the technician’s more intimate experience of the behavior of matter and mecha- nisms.”17 Often the use of tools and machines has preceded a sci- entific explanation for how they work or why they fail. Thomas Newcomen, who made the first practical steam engines in Britain, worked as an artist in Aristotle’s sense of the term “techne.” He conceivably might have heard that a French scientist, Denis Papin, was studying steam and vacuum pumps. However, Newcomen had little formal education and could not have read Papin’s account of his experiments, published in Latin (1690) or in French (1695), though he conceivably could have seen a short summary published in English (1697). He never saw Papin’s small laboratory apparatus—and even had he seen it, it would not have  Can We Define “Technology?” 11 been a model for his much larger engine. Newcomen’s steam engine emerged from the trial and error of practical experiments. Papin’s scientific publications were less a basis for inventing a workable steam engine than a theoretical explanation for how a steam engine worked. However, further improvements in the steam engine did call for more scientific knowledge on the part of James Watt and later inventors. Likewise, Thomas Edison built his electrical system without the help of mathematical equations to explain the behavior of electricity. Later, Charles Steinmetz and others developed the theoretical knowledge that was necessary to explain the system mathematically and refine it, but this was after Edison’s laboratory group had invented and marketed all the components of the electrical system, including generators, bulbs, sockets, and a wiring system. Science has played a similar role in the refinement of many technologies, including the wind- mill, the water wheel, the locomotive, the automobile, and the airplane.18 The Wright Brothers were well-read and gifted bicycle mechanics, and they tested their designs in a wind tunnel of their own invention, but they were not scientists.19 If one bears these examples in mind, the emergence of the term “technology” into English from modern Latin in the seventeenth century makes considerable sense. At first, the term was almost exclusively employed to describe a systematic study of one of the arts. A book might be called a “technology” of glassmaking, for example. By the early eighteenth century, a characteristic defini- tion was “a description of the arts, especially the mechanical.” The word was seldom used in the United States before 1829, when Jacob Bigelow, a Harvard University professor, published a book titled Elements of Technology.20 As late as the 1840s, almost the only American use of the word was in reference to Bigelow’s book.21 In 1859, the year before he was elected president, Abraham Lincoln 12 Chapter 1 gave several versions of a lecture on discoveries and inventions without once using the word.22 Before 1855, even Scientific Ameri- can scarcely used “technology,” which only gradually came into circulation. Instead, people spoke of “the mechanic arts” or the “useful arts” or “invention” or “science” in contexts where they would use “technology” today. A search of prominent American periodicals shows that between 1860 and 1870 “technology” appeared only 149 times, while “invention” occurred 24,957 times. During the nineteenth century the term became embed- ded in the names of prominent educational institutions such as the Massachusetts Institute of Technology, but it had not yet become common in the discussion of industrialization.23 “At the time of the Industrial Revolution, and through most of the nine- teenth century,” Leo Marx writes, “the word technology primarily referred to a kind of book; except for a few lexical pioneers, it was not until the turn of [the twentieth] century that sophisticated writers like Thorstein Veblen began to use the word to mean the mechanic arts collectively. But that sense of the word did not gain wide currency until after World War I.”24 This broader definition owed much to German, which had two terms: “teknologie” and the broader “technik.” In the early twen- tieth century, “technik” was translated into English as “technics.”25 From roughly 1775 until the 1840s, “teknologie” referred to sys- tems of classification for the practical arts, but it was gradually abandoned. During the later nineteenth century, German engi- neers made “technik” central to their professional self-definition, elaborating a discourse that related the term to philosophy, eco- nomics, and high culture. “Technik” meant the totality of tools, machines, systems and processes used in the practical arts and engineering.26 Both Werner Sombart and Max Weber used the term extensively, influencing Thorstein Veblen and others writ-  Can We Define “Technology?” 13 ing in English. As late as 1934, Lewis Mumford’s landmark work Technics and Civilization echoed this German usage. However, Mumford also used the term “technology” not in the narrow Germanic sense but in reference to the sum total of systems of machines and techniques that underlie a civilization. In subse- quent decades the term “technics” died out in English usage and its capacious meanings were poured into “technology.”27 Mumford had these larger meanings and the German tradition in mind when he argued that three fundamentally different social and economic systems had succeeded one another in an evolu- tionary pattern. Each had its own “technological complex.” He called these “eotechnic” (before c. 1750), “paleotechnic” (1750– 1890), and “neotechnic” (1890 on). Mumford conceived these as overlapping and interpenetrating phases in history, so that their dates were approximate and varied from one nation to another. Each phase relied on a distinctive set of machines, processes, and materials. “Speaking in terms of power and characteristic materi- als,” Mumford wrote, “the eotechnic phase is a water-and-wood complex, the paleotechnic phase is a coal-and-iron complex, and the neotechnic phase is an electricity-and-alloy complex.”28 Although historians no longer use either Mumford’s terms or his chronology, the sense that history can be conceived as a sequence of technical systems has become common. Along with this sense of a larger sequence came the realization that machines cannot be understood in isolation. As Mumford put it: “The machine cannot be divorced from its larger social pattern; for it is this pattern that gives it meaning and purpose.”29 One important part of this pattern that Mumford missed, how- ever, was how thoroughly “technology” was shaped by gender. For example, legal records from the thirteenth and fourteenth centuries show that in rural England women were entirely 14 Chapter 1 responsible for producing ale, the most common drink of the peasantry. Men took control of alemaking only when it was com- mercialized.30 Similarly, some scholars argue that in the early medieval era European women worked in many trades, but that in early modern times women were gradually displaced by men.31 Ruth Oldenziel has persuasively extended such arguments into the twentieth century, showing that Western society only rela- tively recently defined the word “technology” as masculine. Between 1820 and 1910, as the word acquired its present meaning, it acquired male connotations. Before then, “the useful arts” included weaving, potterymaking, sewing, and any other activity that transformed matter for human use. The increasing adoption of the word “technology,” therefore, is not simply a measure of the rise of industrialization. It also measures the marginalization of women.32 In the United States, women were excluded from tech- nical education at the new university-level institutes, such as the Rensselaer Polytechnic Institute (established in 1824) and the Massachusetts Institute of Technology (founded in 1861). Never- theless, because one could become an engineer on the basis of job experience, there were several thousand female engineers in the United States during the nineteenth century. Likewise, despite many obstacles, there were female inventors. The women’s build- ings of the great world’s fairs in Philadelphia (1876), Chicago (1893), Buffalo (1901), and St. Louis (1904) highlighted women’s inventions and their contributions to the useful arts. Further- more, even though women had been almost entirely excluded from formal engineering education, many worked as technical assistants in laboratories, hospitals, and factories. Engineering was culturally defined as purely masculine, pushing women to the margins or to subordinate positions. Only in recent years have scholars begun to see technology in gendered terms, however, and this realization is not yet widely shared.  Can We Define “Technology?” 15 Indeed, the meaning of “technology” remained unstable in the second half of the twentieth century, when it evolved into an annoyingly vague abstraction. In a single author’s writing, the term could serve as both cause and effect, or as both object and process. The word’s meaning was further complicated in the 1990s, when newspapers, stock traders, and bookstores made “technology” a synonym for computers, telephones, and ancil- lary devices. “Technology” remains an unusually slippery term. It became a part of everyday English little more than 100 years ago. For several hundred years before then, it meant a technical description. Then it gradually became a more abstract term that referred to all the skills, machines, and systems one might study at a technical university. By the middle of the twentieth century, technology had emerged as a comprehensive term for complex systems of machines and techniques. Indeed, some thinkers began to argue that these systems had a life and a purpose of their own, and no sooner was “technology” in general use than some began to argue for “technological deter- minism.” A single scene in Stanley Kubrick’s film 2001 captures the essence of this idea. A primitive ancestor of modern man picks up a bone, uses it as a weapon, then throws it into the air, where it spins, rises, and metamorphoses into a space station. The implica- tions of this scene were obvious: a direct line of inevitable techno- logical development led from the first tools to the conquest of the stars. Should we accept such determinism? 2 Does Technology Control Us? Are technologies deterministic?1 Many people talk as though they are. Students have often told me that the spread of television or the Internet was “inevitable.” Likewise, most people find the idea of a modern world without automobiles unimaginable. However, history provides some interesting counterexamples to apparently inevitable technologies. The gun would appear to be the classic case of a weapon that no society could reject once it had been introduced. Yet the Japanese did just that. They adopted guns from Portuguese traders in 1543, learned how to make them, and gradually gave up the bow and the sword. As early as 1575 guns proved decisive in a major battle (Nagoshino), but then the Japan- ese abandoned them, for what can only be considered cultural reasons. The guns they produced worked well, but they had little symbolic value to warriors, who preferred traditional weapons.2 The government restricted gun production, but this alone would not be enough to explain Japan’s reversion to swords and arrows. Other governments have attempted to restrict gun ownership and use, often with little success. But the Japanese samurai class rejected the new weapon, and the gun disappeared. It re-entered society only after 1853, when Commodore Perry sailed his war- ships into Japanese waters and forced the country to open itself to the West. 18 Chapter 2 Japan’s long, successful rejection of guns is revealing. A society or a group that is able to act without outside interference can abol- ish a powerful technology. In the United States, the Mennonites and the Amish do not permit any device to be used before they have carefully evaluated its potential impact on the community. For example, they generally resist home telephones and prefer face-to-face communication, although they permit limited use of phones to deal with the outside world. They reject both automo- biles and gasoline tractors. Instead, they breed horses and build their own buggies and farm machinery. These choices make the community far more self-sufficient than it would be if each farmer annually spent thousands of dollars on farm machinery, gasoline, and artificial fertilizer, all of which would necessarily come from outside the community. Their leaders decide such matters, rather than leaving each individual to choose in the market. Such prac- tices might seem merely quaint, but they provide a buffer against such things as genetically modified foods or chemical pesticides, and they help to preserve the community. Indeed, the Amish are growing and flourishing. Both the Japanese rejection of the gun and the Amish selective acceptance of modern farming equip- ment show that communities can make self-conscious technolog- ical choices and can resist even very powerful technologies. Furthermore, these two examples suggest that the belief in determinism paradoxically seems to require a “free market.” The belief in technological determinism is widely accepted in individ- ualistic societies that embrace laissez-faire economics. What many people have in mind when they say that television or the Internet was “inevitable” boils down to an assumption that these technologies are so appealing that most consumers, given the chance, will buy them. Historians of technology often reject this view because they are concerned not only with consumers but  Does Technology Control Us? 19 also with inventors, entrepreneurs, and marketers. They see each new technology not simply as a product to be purchased, but as a part of a larger system. Few historians argue that machines deter- mine history. Instead, they contend that new technologies are shaped by social conditions, prices, traditions, popular attitudes, interest groups, class differences, and government policy.3 A surprising number of people, however, including many scholars, speak and write about technologies as though they were deterministic. According to one widely read book, television has “helped change the deferential Negro into the proud Black,” has “given women an outside view of their incarceration in the home,” and has “weakened visible authorities by destroying the distance and mystery that once enhanced their aura and pres- tige.”4 These examples suggest that technology has an inexorable logic, that it forces change. But is this the inexorable effect of introducing television into China or the Arab world? In some cases, one might argue, television is strengthening fundamental- ism. It simply will not do to assume that the peculiar structure of the American television market is natural. In the United States, television is secular, not religious; private, not public; funded by advertising, not taxation; and a conduit primarily of entertain- ment, not education. These are cultural choices. Many have made a similar mistake in writing about the Inter- net. Nicholas Negroponte declared, in a best-selling book, that “digital technology can be a natural force drawing people into greater world harmony.”5 This is nonsense. No technology is, has been, or will be a “natural force.” Nor will any technology by itself break down cultural barriers and bring world peace. Consider the wheel, an invention that most people think of as essential to civi- lization. Surely the wheel must be an irresistible force, even if the gun and the automobile are not! Much of North Africa, however, 20 Chapter 2 let the wheel fall into disuse after the third century A.D., prefer- ring to transport goods by camel. This was a sensible choice. Main- taining roads for wheeled carts and supplying watering sites for horses and oxen was far more expensive, given the terrain and the climate, than opting for the camel, which “can carry more, move faster, and travel further, on less food and water, than an ox,” needs “neither roads nor bridges,” and is able to “traverse rough ground and ford rivers and streams.”6 In short, societies that have used the wheel may turn away from it. Other civilizations, not- ably the Mayans and the Aztecs, knew of the wheel but never developed it for practical purposes. They put wheels on toys and ceremonial objects, yet apparently they did not use wheels in construction or transportation. In short, awareness of particular tools or machines does not automatically force a society to adopt them or to keep them. In Capitalism and Material Life, Fernand Braudel rejected tech- nological determinism. Reflecting on how slowly some societies adopt new methods and techniques, he declared: “Technology is only an instrument and man does not always know how to use it.”7 Like Braudel, most specialists in the history of technology do not see new machines as coercive agents dictating social change, and most remain unpersuaded by determinism, though they readily agree that people are often reluctant to give up conven- iences. For millennia people lived without electric light or central heating, but during the last 150 years many societies have adopted these technologies and made them part of their building codes. It is now illegal in many places to build or live in a house without indoor plumbing, heating, and electric lighting. In other words, people become enmeshed in a web of technical choices made for them by their ancestors. This is not determinism, though it does  Does Technology Control Us? 21 suggest why people may come to feel trapped by choices others have made. Often, adopting a new technology has unintended conse- quences. Governments build highways to relieve traffic conges- tion, but better roads may attract more traffic and reduce the use of mass transit as an alternative. Edward Tenner, in his book Why Things Bite Back, examines “the revenge of unintended conse- quences.”8 Among many examples, he notes that computers are expected to improve office efficiency, but in practice people spend enormous amounts of time adjusting to updated software and they suffer eyestrain, back problems, tendonitis, and cumulative trauma disorder.9 Furthermore, to the extent that computers replace secretaries, white-collar professionals often find them- selves doing routine tasks, such as copying and filing docu- ments and stuffing envelopes. Thus, despite many claims made for greater efficiency through computerization, a study by the American Manufacturing Association found that reducing staff raised profits for only 43 percent of the firms that tried it, and 24 percent actually suffered losses, despite the savings on wages. In some cases computerization reduced the time that highly skilled employees had available to perform skilled work. “Their jobs became more diverse in a negative way, including things like printing out letters that their secretaries once did.”10 For some white-collar workers, the computer had the unintended conse- quence of diminishing their specialization. In short, rather than assuming that technologies are deter- ministic, it appears more reasonable to assume that cultural choices shape their uses. While salesmen and promoters like to claim that a new machine is inevitable and urge us to buy it now or risk falling behind competitors, historical experience strongly 22 Chapter 2 suggests that the actual usefulness of a new technology is unpre- dictable. The idea that mechanical systems are deterministic remains so persistent, however, that a brief review of this tradition is neces- sary. In the middle of the nineteenth century, most European and American observers saw machines as the motor of change that pushed society toward the future. The phrase “industrial revo- lution,” which gradually came into use after c. 1875, likewise expressed the notion that new technologies were breaking deci- sively with the past. Early socialists and free-market capitalists agreed on little else, but both saw industrialization as an unfold- ing of rationality. Even harsh early critics tended to assume that the machine itself was neutral, and focused their attacks on people who misused it. Not until the twentieth century did many argue that technologies might be out of control or inherently dan- gerous. Technological determinism, which in the nineteenth century often seemed beneficent, appeared more threatening thereafter. Some Victorians worried that machinery seemed to proliferate more rapidly than the political means to govern it. Without any need of the word “technology,” Thomas Carlyle issued a full-scale indictment of industrialization that contained many of the neg- ative meanings that later would be poured into the term. His con- temporary Karl Marx saw the mechanization of society as part of an iron law of inevitable historical development.11 In The Critique of Political Economy, Marx argued that “the mode of production of material life determines the general character of the social, polit- ical, and spiritual process of life.”12 (Marx did not use the word “technology” in the first edition of Das Kapital,13 though it did appear in later editions. His collaborator, Engels, took up the term  Does Technology Control Us? 23 “technics” late in life.14) Marx argued that industrialization’s immediate results were largely negative for the working class. The skilled artisan who once had the satisfaction of making a finished product was subjected to the subdivision of labor. The worker, who once had decided when to work and when to take breaks, lost control of such choices in the new factories. Capital’s increasing control of the means of production went along with de-skilling of work and lowering of wages. Industrialization broke the bonds of communities and widened the gaps between social classes. Marx argued that capitalism would collapse not only because it was unjust and immoral, and not only because poverty and inequal- ity would goad the workers to revolt, but also because it would cre- ate economic crises of increasing intensity. These crises were not caused by greed or oppression, and they would occur no matter how well meaning capitalists themselves might be. For Marx, the logic of capitalism led to continual investment in better machines and factories, which tied up resources in “fixed capi- tal,” leaving less money available for wages (“variable capital”). As investments shifted from labor power to machinery, the amount available for wages and the number of workers employed had to decrease; otherwise the capitalist could not make a profit. This made sense for each individual capitalist, but the overall effect on society when many factories cut total wages and substituted machines for men was a decrease in demand. At the very time when a capitalist had more goods to sell (because he had a new and better production system), fewer people had money to pur- chase those goods. Thus, Marx argued, efficiency in production flooded the market with goods, but simultaneously the substitu- tion of machines for laborers undermined demand. A crisis was unavoidable. If a capitalist halted production until he had sold off surpluses, he reduced demand still further. If he raised wages 24 Chapter 2 to stimulate demand, profits fell. If he sought still greater effi- ciencies through mergers with rivals, he threw even more work- ers on the dole, and the imbalance between excessive supply and weak demand became more severe. Marx’s analysis posited the inevitable end of capitalism. As greater mechanization produced greater surpluses, it impoverished more workers, causing increas- ingly severe economic crises because supplies outran demand. Mechanization under capitalism apparently led unavoidably to worker exploitation, social inequality, class warfare, social col- lapse, and finally revolution. Marx did not reject technology itself. After the collapse of capi- talism, he expected, a succeeding socialist regime would appropri- ate the means of production and build an egalitarian life of plenty for all. If Marxism made a powerful critique of industrialization that included such concepts as class struggle, worker alienation, de-skilling of artisans, false consciousness, and reification, ulti- mately it was not hostile to the machine as such. Rather, both Marx and Engels expected that industrialization would provide the basis for a better world. Similarly, Lenin hoped that after the Russian Revolution the technical elite would rationally direct fur- ther industrialization and redistribute the wealth it produced. Lenin argued that revolutionary change “should not be confused with the question of the scientifically trained staff of engineers, agronomists and so on.” “These gentlemen,” he continued, “are working today in obedience to the wishes of the capitalists, and will work even better tomorrow in obedience to the wishes of the armed workers.”15 After the Revolution, the Soviet Union empha- sized electrification and mass production. Lenin famously declared that only when the Soviet Union had been completely electrified could it attain full socialism. He vigorously pursued a ten-year plan of building generating plants and incorporated them into a national grid, with the goal of extending electrical  Does Technology Control Us? 25 service to every home.16 As this example suggests, Marxists criti- cized how capitalists used technical systems but not industrializa- tion itself. The left generally assumed that a society’s technologies defined its economic system and social organization. Thus the primitive mill produced feudalism, while the steam engine produced capi- talism. They equated mechanization and industrialization with the rational unfolding of history. Evolutionary socialists agreed that technological systems ultimately would become the basis of a utopia, without, however, expecting that violent class conflict and revolution were necessary to attain it. They believed that new technologies would lead to the inevitable decline of capital- ism and the emergence of a better economic system. For example, German-born Charles Steinmetz, the leading scientist at General Electric in its first decades, expected socialism to emerge along with a national electrical grid, because it was an inherently inter- dependent basis for economic reorganization. Electricity could not be stored efficiently and had to be consumed through large distribution systems as soon as it was produced. “The relation between the steam engine as a source of power and the electric motor is thus about the same as the relation between the individ- ualist [capitalist] and the socialist.... The one is independent of everything else, is self-contained, the other, the electric motor, is dependent on every other user in the system.... The electric power is probably today the most powerful force tending towards co-ordination, that is cooperation [socialism].”17 Both Marxists and evolutionary socialists embraced not only the machine but also a sense of inevitable historical development based on techno- logical change. In contrast, Werner Sombart rejected such determinism in Tech- nik und Kultur, where he argued that cultures often shaped events more than technologies did. For example, Sombart thought that 26 Chapter 2 the failure of cultural and political institutions, and not tech- nological change, accounted for the decline of ancient Rome. Sombart accorded technology an important role in history, par- ticularly in modern times, but he also recognized the importance of culture and institutions. The Chicago School of sociology developed Sombart’s ideas in the United States. For example, when William Ogburn wrote about “the influence of invention and discovery,” he denied that “mechanical invention is the source of all change” and pointed to “social inventions” such as “the city manager form of government... which have had great effects upon social customs. While many social inventions are only remotely connected with mechanical inventions, others appear to be precipitated by” them, such as “the trade union and the tourist camp.” Influence could flow in either direction. Social inventions could stimulate technical invention.18 Ogburn admit- ted that mechanization had a powerful effect on society, yet he emphasized that “a social change is seldom the result of a single invention.” Women’s suffrage, for example, was the outcome of a great number of converging forces and influences, including mass production, urbanization, birth control, the adoption of the type- writer, improved education, and the theory of natural rights. Most historical changes were attributable to such a “piling up process.” Making the distinction between social invention and technical invention also suggested to Ogburn the notion of a cultural lag. “There is often a delay or lag in the adaptive culture after the material culture has changed, and sometimes these lags are very costly, as was the case with workmen’s compensation for indus- trial accidents.”19 “The more one studies the relationship between mechanical and social inventions,” Ogburn concluded, “the more interrelated they seem. Civilization is a complex of inter- connections between social institutions and customs on the one hand, and technology and science on the other.”20 Because “the  Does Technology Control Us? 27 whole interconnected mass is in motion,”21 it was difficult to estab- lish causation. The idea that technologies developed more rapidly than society remained attractive to some later theorists. During the 1960s, Marshall McLuhan won a large following as he argued that every major form of communication had reshaped the way people saw their world, causing changes in both public behavior and political institutions. For McLuhan, innovations in communications, notably the printing press, radio, and television, had automatic effects on society. Unlike Ogburn, McLuhan paid little attention to reciprocal effects or social inventions. For McLuhan, not only did the media extend the human sense organs; each new form of a medium disrupted the relationship between the senses. McLuhan argued that the phonetic alphabet intensified the visual function and that literate cultures devalued the other senses—a process that moveable type intensified. Furthermore, McLuhan thought electronic media extended the central nervous system and linked humanity together in a global network. Alvin Toffler reworked such deterministic ideas into Future Shock, a best-seller that argued that technological change had accelerated to the point that people scarcely could cope with it. Later, in The Third Wave, Toffler argued that a new industrial revolution was being driven by electronics, computers, and the space program.22 In such studies, the word “impact” suggests that machines inexorably impress change on society. Although the details of their analyses varied, both McLuhan’s arguments and Toffler’s were externalist, treating new technol- ogies as autonomous forces that compel society to change. The public has an appetite for proclamations that new technologies have beneficent “natural” effects with little government inter- vention or public planning. Externalist arguments attribute to a technology a dominant place within society, without focusing 28 Chapter 2 much on invention or technical details. Externalist studies of “technology transfer” often say little about machines and pro- cesses, such as firearms or textile factories, but a great deal about their “impact” on other countries.23 Externalists usually adopt the point of view of a third-person narrator who stands outside tech- nical processes. They seldom dwell on the (often protracted) dif- ficulties in defining the technological object at the time of its invention and early diffusion. Close analysis—common in the internalist approach to be described in chapter 4—tends to under- mine determinism, because it reveals the importance of particular individuals, accidents, chance, and local circumstances. Determinism is not limited to optimists. Between 1945 and 1970, many of the most pessimistic critics of technology were also determinists. Jacques Ellul paid little attention to the origins of individual inventions, but argued instead that an abstract “Tech- nique” had permeated all aspects of society and had become the new “milieu” that Western societies substituted for Nature. Readers of Ellul’s book The Technological Society24 were told that Technique was an autonomous and unrelenting substitution of means for ends. Modern society’s vast ensemble of techniques had become self-engendering and had accelerated out of human- ity’s control: “Technical progress tends to act, not according to an arithmetic, but according to a geometric progression.”25 Writers on the left found technology equally threatening, and many thought the only possible antidote to be a dramatic shift in consciousness. In One-Dimensional Man (1964) and other works, Herbert Marcuse, a Marxist sociologist whose work emerged from the Frankfurt School, attacked the technocratic state in both its capitalist and its socialist formations. He called for “revolutionary consciousness-raising” in preparation for a wholesale rejection of the managed system that everywhere was reducing people to  Does Technology Control Us? 29 unimaginative cogs in the machine of the state. Marcuse, who became popular with the student movements of the late 1960s, hoped that the “New Left” would spearhead the rejection of the technocratic regime. In The Making of a Counter Culture (1969), Theodore Roszak was equally critical but less confrontational, arguing that reform of the technocratic state was impossible. His first chapter, “Technocracy’s Children,” attacked the mystifica- tion of all decision making as it became clothed in the apparently irrefutable statistics and the terminology of technocrats. Western society had become a “technocracy,” defined by Roszak as “that society in which those who govern justify themselves by appeal to technical experts who, in turn, justify themselves by appeal to sci- entific forms of knowledge.”26 Such a technical ideology seemed “ideologically invisible” because its assumed ideals—rationality and efficiency—were accepted without discussion both in the communist East and the capitalist West. To resist technocracy, a de-technologized consciousness was needed, which Roszak sought through a combination of Zen Buddhism, post-Freudian psychology, and the construction of alternative grassroots institu- tions, such as those in the emerging hippie movement.27 As student radicalism faded during the 1970s, social revolution seemed less probable than technological domination, notably as analyzed in the work of Michel Foucault. He treated technology as the material expression of an overarching discourse that struc- tured individual consciousness and shaped institutions, notably hospitals, asylums, and prisons.28 In contrast to Marx, Foucault’s theory did not conceive of an economic or a technical “base” that drove changes in the social “superstructure.” Rather, Foucault saw history as the exfoliation of patterns of ideas and structures (“epis- temes”), which were expressed in art, in architecture, in classifica- tion systems, in social relations, and in all other aspects of the 30 Chapter 2 cultural discourse at a given historical moment. The epistemes did not evolve from one discursive system to the next but rather were separated by ruptures, or breaks in continuity. When a new dis- course emerged, it did not build upon previous systems. Rather, as a sympathetic critic summarized, “a new knowledge begins, it is unrelated to previous knowledge.”29 Foucault conceived history as a series of internally coherent epistemological systems, each built upon different premises. The individual author, inventor, or citi- zen was not the master of his or her fate but rather was penetrated and defined by discourses. Each was caught within, scarcely aware of, and ultimately articulated by structures of knowledge and power that were deployed and naturalized throughout society. In the modern episteme, Foucault was concerned with how power became anonymous and embedded in bureaucracies, making hierarchical surveillance a social norm. His determinism was far more comprehensive than that of most previous thinkers. Foucault, and later the postmodernist Francois Lyotard, authored academic best-sellers of the 1970s and the 1980s, but their grand deterministic theories found little favor among histo- rians of technology, whose research showed considerable evi- dence of human agency in the creation, dissemination, and use of new technologies. Leo Marx declared that postmodern theo- rists in effect ratify “the idea of the domination of life by large technological systems” and promote a “shrunken sense of human agency.”30 The most sweeping rejection of technological determinism came from Marx’s student Langdon Winner in Autonomous Technology, a book Winner said he had written in a spirit of “epistemological Luddism.”31 In dismantling determinis- tic ideologies, Winner made it easier to think of technologies as socially shaped, or constructed. Winner also emphasized Karl Marx’s more flexible views of technology in his earlier works. In  Does Technology Control Us? 31 The German Ideology (1846), Winner comments, “human beings do not stand at the mercy of a great deterministic punch press that cranks out precisely tailored persons at a certain rate during a given historical period. Instead, the situation Marx describes is one in which individuals are actively involved in the daily cre- ation and recreation, production and reproduction of the world in which they live.”32 While Marx’s labor theory of value might seem to suggest rigid determinism, Winner argues that his work as a whole does not support such a view. Technological determinism lacks a coherent philosophical tradi- tion, although it remains popular. A variety of thinkers on both the right and the left have put forward theories of technological determinism, but the majority of historians of technology have not found them useful. As the following two chapters will show, deterministic conceptions of technology seem misguided when one looks closely at the invention, the development, and the mar- keting of individual devices. 3 Is Technology Predictable? If technologies are not deterministic, then neither their emer- gence nor their social effects should be predictable. To consider this proposition in detail, we can divide technological prognosti- cation into three parts: prediction, forecasting, and projection. We predict the unknown, forecast possibilities, and project prob- abilities. These three terms correspond to the division common in business studies of innovation, between what James Utterbeck terms “invention (ideas or concepts for new products and pro- cesses), innovation (reduction of an idea to the first use or sale) and diffusion of technologies (their widespread use in the market).”1 Prediction concerns inventions that are fundamentally new devices. This is a more restrictive definition than the US Patent Office’s sense of “invention,” for that also includes “innovation,” treated here as a separate category. What is the distinction? The incandescent electric light was an invention; new kinds of fila- ments were innovations. The telephone was an invention, but the successive improvements in its operation were innovations. Inventions are fundamental breakthroughs, and there have been relatively few. In communications, they would include the tele- phone, the electric light, radio, television, the mainframe com- puter, the personal computer, and the Internet. While prediction 34 Chapter 3 Table 3.1 Form of Persons prognostication typically involved Their focus Their time frame Prediction Inventors, Breakthrough Long term utopian writers inventions Forecasting Engineers, Innovations Less than entrepreneurs 10 years Projection Designers, New models Less than marketers 3 years concerns such inventions, forecasting concerns innovations, which are far more numerous. Innovations are improvements and accessories to systems that emerged from inventions. The third term, “projection,” which I will discuss only briefly, con- cerns the future sales, profits, market share, and so forth of new models of established technologies. Prediction, forecasting, and projection typically involve differ- ent professionals working within different time frames. (See table 3.1.) These distinctions are not merely a matter of semantic con- venience. If one looks at the time frames involved, prediction deals with the long term or even indefinite periods, whereas fore- casting focuses on immediate choices about getting a new device perfected and into production. Those making projections must work within the shortest time frame, because they deal with new (often annual) models of devices that compete in the market. Who is centrally involved in prognostication depends on which category one is dealing with. Inventors, futurologists, and some academics predict or debunk the possibility of fundamental breakthroughs. Once a workable device exists, however, venture capitalists, engineers, and consultants busy themselves with fore- casting its possibilities. If a device is widely accepted, designers and marketers take a central role in projecting and extrapolating  Is Technology Predictable? 35 what new styles and models consumers will buy. In view of the differences in actors and in time frames, there are considerable differences in the aesthetics of invention, innovation, and prod- uct development, emphasizing, respectively, technical elegance, functionalism, and beauty.2 On television one mostly hears forecasting and projection, not prediction. For example, in 1998 a “technology guru” on the Cable News Network announced that voice recognition would be the “next big thing” in computers because keyboards could then be done away with, and small computers capable of respond- ing to verbal commands would be embedded in useful objects everywhere.3 Machine speech recognition was already used by telephone companies by that time; its possible extension and development to replace computer keyboards was forecast. Eight years later, voice recognition seems to have spread more slowly than that “guru” expected. All technological predictions and forecasts are in essence little narratives about the future. They are not full-scale narratives of utopia, but they are usually presented as stories about a better world to come. The most successful present an innovation as not just desirable but inevitable. Public-relations people are well aware that such stories can become self-fulfilling when investors and consumers believe them. As the consultant and critic John Perry Barlow once put it, “the best way to invent the future is to pre- dict it—if you can get enough people to believe your prediction, that is.”4 Selling stories of the wonders to come has been popular at least since the Chicago World’s Columbian Exposition of 1893,5 and they have become the stock in trade of investment newsletters, some technical magazines, and certain educational television programs. To put this another way, inventors and corporate 36 Chapter 3 research departments create not only products but also com- pelling narratives about how these new devices will fit into every- day life. They need to do this to get venture capital, and companies need to market such scenarios to get a return on investment. Yet accurate prediction is difficult, even for experts. George Wise, a historian who worked for years at the General Electric research labs in Schenectady, wrote his doctoral thesis on how well scientists, inventors, and sociologists predicted the future between 1890 and 1940. Examining 1,500 published predictions, he found that only one-third proved correct, while one-third were wrong and another one-third were still unproved. They used many methods, including intuition, analogy, extrapola- tion, studying leading indicators, and deduction, but all were of roughly equal accuracy.6 The technical experts, he found, per- formed only slightly better than others. In short, technological predictions, whoever made them and whatever method was employed, proved no more accurate than flipping a coin. If prediction has proved extremely difficult, what about fore- casting? That ought to be easier, because it deals with already invented technologies and builds on existing trends. Anyone interested in computers has heard of Moore’s Law, formulated in 1965, which predicted, quite accurately, that computer memory would double roughly every 18 to 24 months.7 (Note, however, that this may have been a self-fulfilling prophecy, because it estab- lished a benchmark for development in the computer industry.) Yet for every such success there are famous failures of forecasting. No demographer saw the United States’ post-World War II baby boom coming. American birth rates had fallen steadily for more than 100 years, and demographers were surprised when the decline did not continue. In the 1960s a great many sociologists projected that automation would reduce the average American’s  Is Technology Predictable? 37 work week to less than 25 hours by the century’s end. Instead, the average American today is working more hours than in 1968.8 Paul Ehrlich, in The Population Bomb, predicted in the early 1970s that it was already too late to save India from starvation.9 He did not foresee the tremendous increases in agricultural productivity. Social trends are difficult to anticipate. General forecasting is risky, failure common. Technological forecasting is no easier. In 1900, few investors forecast that the new automobiles would replace trolley cars.10 Trolley service had grown tremendously in the previous decade, and it was expanding into long-distance competition with the railroad. The automobile was still a rich man’s toy, and no one anticipated the emergence or the tremendous productivity of the automotive assembly line. In the 1930s, when only one in a hun- dred people had actually been up in an airplane, a majority of Americans mistakenly expected that soon every family would have one.11 In 1954, Chairman Lewis Strauss of the US Atomic Energy Commission told the National Association of Science Writers that their children would enjoy “electrical energy too cheap to meter.”12 IBM, thinking that mainframes would always be the core of the computer business, waited seven years before competing directly with Digital Computer’s minicomputers.13 Later, Apple mistakenly thought there was a market for its Newton, an early personal digital assistant that had good handwriting recognition but proved too large and too expensive for most con- sumers. The experts at Microsoft did not foresee the sudden emer- gence of the World Wide Web, and were slow to compete with Netscape when it appeared. These were all failures of forecasting. Projection might be expected to work reasonably well when the economy is stable. The total demand for most items will be stable, and extrapolations based on growth rates may prove accurate. But 38 Chapter 3 a stable market is full of competing products, and full of expand- ing and contracting firms. In the 1950s, Ford thought there was a market for the Edsel. Furthermore, business conditions are sel- dom stable for long. In the 1960s, American utility companies expected growth in the consumption of electricity to double every ten years, as it had done for decades. The utility companies did not foresee the energy crises of the 1970s, which would trigger a move toward conservation.14 The energy crisis likewise caught American automakers unprepared; they had projected continued demand for large cars, and they had few small, energy-efficient vehicles for sale. As these examples suggest, any trend that seems obvious, and any pattern that seems persistent, may be destabilized by changes in the economy, changes in technology, or some combination of social and technical factors. As the mathematician John Paulos put it, “futurists such as John Naisbitt and Alvin Toffler attempt to ‘add up’ the causes and effects of countless local stories in order to identify and project trends.” But “interactions among the various trends are commonly ignored, and unexpected developments, by definition, are not taken into account. As with weather forecast- ers, the farther ahead they predict, the less perspicacious they become.”15 It is not just futurists who stumble. Fundamental innovations almost always seem to come from outside the established market leaders, who suffer from “path dependency.” Established firms are usually too committed to a particular conception of what their product is. This commitment is embedded in its manufacturing process and endemic in the thinking of its managers. When a major innovation appears, a leading firm understands the tech- nology, but remains committed to its product and its production system. The case of IBM and the personal computer is a good example. At first IBM did not take the threat seriously enough, and  Is Technology Predictable? 39 competitors had the market for personal computers to themselves for at least four years before IBM entered the field. IBM then was clever enough to license others to manufacture its system, making it the standard, but it had to share the market with many other firms. In 2005, after 25 years, it withdrew from the market. In most cases, when an innovation such as the personal com- puter appears, established industries redouble their commitment to the traditional product that has made them the market leader. They make incremental improvements in manufacturing, and yet they lose market share to the invader. This occurs even in fast- changing electronic industries, where innovations come so fre- quently that there is little time for routines and habits to blind participants to the advantages of the next change. Utterback cites a comprehensive study of the manufacturers that supply semi- conductor firms with photolithographic alignment machines. During the invention and development of five distinct genera- tions of such machines, in no case did the market leader at one stage retain its top position at the next.16 A production system seems to gain such a powerful hold inside a firm that it seldom can move swiftly enough to adopt innovations.17 Another reason that forecasts and predictions are so hard to make is that consumers, not scientists, often discover what is “the next big thing.” Most new technologies are market-driven. Viagra was not developed as a sexual stimulant, but the college students who served as guinea pigs discovered what consumers would like about it. This general point can be put negatively: Just because something is technologically feasible, don’t expect the public to rush out and buy it. Consumers must want the product. There were many mistaken investments in machines that worked but which the public didn’t want. The classic case may be AT&T’s Picture Phone.18 It was technologically feasible, and it was promoted at the New York World’s Fair of 1964. But aiming imme- 40 Chapter 3 diately at the mass market, rather than starting more slowly with a niche market, proved a miscalculation. Few bought it, partly because they resisted its high price but also because they feared a visual invasion of their privacy and because they did not under- stand its potential as a data-display terminal. Though some appar- ently reasonable technologies fail to sell, people may nonetheless flock to “unreasonable” devices, such as Japanese electronic pets. Histories of new machines tend to focus on the process of inven- tion and to suggest that the market is driven by research and devel- opment. This is usually not so, even in the case of inventions that in retrospect clearly were fundamental to contemporary society: the telegraph, the telephone, the phonograph, the personal com- puter. When such things first appear, creating demand is more dif- ficult than creating supply. At first, Samuel Morse had trouble convincing anyone to invest in his telegraph. He spent five years “lecturing, lobbying, and negotiating” before he convinced the US Congress to pay for the construction of the first substantial telegraph line, which ran from Washington to Baltimore. Even after it was operating, he had difficulty finding customers inter- ested in using it.19 Likewise, Alexander Graham Bell could not find an investor to buy his patent on the telephone, and so he reluc- tantly decided to market it himself.20 Thomas Edison found few commercial applications for his phonograph, despite the sensa- tional publicity surrounding its discovery.21 He and his assistants had the following commercial ideas a month after the phono- graph was first shown to the world: to make a speaking doll and other toys, to manufacture speaking “clocks... to call the hour etc., for advertisements, for calling out directions automatically, delivering lectures, explaining the way,” and, almost as an after- thought at the end of the list, “as a musical instrument.”22 In the mid 1970s, a prototype personal computer, when first shown to a  Is Technology Predictable? 41 group of MIT professors, seemed rather uninteresting to them.23 They could think of few uses for it, and they suggested that per- haps it would be most useful to shut-ins. In short, the telegraph, the telephone, the phonograph, and the personal computer, surely four of the most important inven- tions in the history of communications, were initially understood as curiosities.24 Their commercial value was not immediately clear. It took both investors and the public time to discover what they could use them for. Eventually large corporations would manufacture each of these inventions, and each became the basis for an international form of communication. As people became familiar with these four technologies, they built them into daily life. Barlow argues that the public’s slow response time is genera- tional: “... it takes about thirty years for anything really new to arise from an invention, because that’s how long it takes for enough of the old and wary to die.”25 People need time to understand fundamental inventions, which is why they spread slowly; in contrast, innovations are eas- ier to understand and proliferate rapidly. The few fundamental inventions become the bases for entirely new systems, but most innovations plug into an existing system. Once the electrical grid, the telephone network, or the World Wide Web had been built, new application technologies or innovations proliferated. For example, as the electrical grid spread across the United States, small manufacturers rushed in with a stunning array of new products—electrified cigar lighters, model trains, Christmas tree lights, musical toilet-paper dispensers, and shaving cream warm- ers, as well as toasters, irons, refrigerators, and washing machines. As electric devices proliferated, the large manufacturers Westing- house and General Electric, like the computer hardware makers of today, soon found it impossible to compete in every area. Once 42 Chapter 3 several million PCs and Macs were in place, programmers created the software equivalent of the earlier appliances, with thousands of programs to compose music, calculate income tax, make archi- tectural drawings, encrypt messages, write novels, and so on. Ordinary consumers played a leading role by encouraging such innovations. They drove the rapid growth in sales of scanners, color printers, high-speed modems, external cartridge drives, and software that sends and receives snapshots and short videos.26 Selling the basic hardware for a communication system often ceases to be as profitable as selling software and services.27 People now spend far more money on things that use electricity than on the electricity itself, and this disproportion has been increasing since the 1920s.28 Something similar happened with the tele- phone. AT&T began with an absolute monopoly and expanded slowly during the period when no one could compete. During the 1890s, however, AT&T’s patent protection ran out, competitors appeared, the market doubled and redoubled in size, and the cost of telephone calls began to drop.29 The intensity of telephone use and the number of applications was still increasing 100 years later. Where once the telephone bill reflected a simple transaction between a customer and the phone company, now the technology of the telephone is the basis for a wide range of commercial rela- tions that includes toll-free calls to businesses, e-mail, faxes, and SMS messages. Telephones enable people not only to speak to one another, but also to send photographs, texts, news, and videos. As with the electrical system, the telephone provided the infrastruc- ture, or even the main platform, for many unanticipated busi- nesses. The recent proliferation of communication technologies interweaves and connects the electrical grid, the telephone, the television, the personal computer, and the Internet. The synergy of this mix of networked systems makes possible a particularly  Is Technology Predictable? 43 rich period of innovation. Many possibilities are latent or only partially developed, and that puts a premium on forecasting for the near future. In this dynamic market, the best design does not always win. Even if someone can accurately foresee the coming of a new tech- nology or an innovation, no one can be certain what design will prove most popular. Perhaps failure was obvious for the air- conditioned bed, the illuminated lawn sprinkler, and the electri- cally sterilized toilet seat, all marketed in the 1930s,30 but it was by no means obvious that Sony’s Betamax, the technically better machine, would lose out to VHS in the home video market. Mar- keting, not technological excellence, proved crucial. Sony decided not to share its system with others and expected to reap all the rewards. Its rival, JVC, allied itself with other manufacturers and licensed them to co-produce its VHS system. Consumers decided that more films were likely to be available in the VHS format because a consortium of companies stood behind it, and Betamax gradually lost out.31 Perhaps the most familiar recent example of a superior machine capturing only a small part of the market is that of Apple’s Macintosh computers. Here again a decision to “go it alone” appears to have been a decisive mistake.32 A somewhat different example is the case of FM radio, which is better for short- distance transmission than AM. It languished virtually unused for a generation because RCA discouraged its use while promoting its already well-established AM network.33 Consider an example from the electrical industry: district heat- ing vs. individual home heating. A hundred years ago, most power stations were near town centers, and they routinely marketed excess steam for the heating of apartment blocks, office buildings, and department stores. Since then district heating has failed to capture much of the American market,34 although in Scandinavia 44 Chapter 3 district heating is popular because it saves energy, lowers pollution levels, and reduces the cost of home heating. District heating was also widespread and apparently worked well in the former Soviet Union, but the plants are now often shut down for “cleaning,” especially during the summer, leaving apartments without hot water. American social values emphasize individualized technol- ogies. Every house has its own heating system, even though this is a wasteful and inefficient choice. If the market to some extent shapes technologies, the market in turn is inflected by cultural values. Even if one can predict which new technologies are possible and forecast which designs will thrive in the market, people may fail to foresee how they will be used. Edison invented the phonograph, but he thought his invention primarily would aid businessmen, who could use it to dictate letters, and he did not focus on music and entertainment even as late as 1890.35 As a result, competitors grabbed a considerable share of the market, and their system of a flat record on a turntable won out over his turning cylinders. Another example: Between 1900 and 1920 the new technology of radio was perceived by government and industry as an improved telegraph that needed no wires. They expected it to be used for point-to-point communications. When radio stations emerged after World War I as consumer-driven phenomena, the electrical corporations were caught off guard, but they quickly moved into the new market.36 The public used both the phonograph and the radio less for work than for fun.37 Likewise, many children use per- sonal computers less to write papers and pursue education than to play computer games and visit strange websites. These activ- ities may or may not be educational; my point is that they were unanticipated and consumer driven.  Is Technology Predictable? 45 Another example of unanticipated use is the higher-than- expected consumption of electricity by refrigerators, which so puzzled a California utility company that it hired anthropologists to find out what was going on.38 They discovered that families used the refrigerator for much more than food storage. It was also a place to hide money in fake cabbages, to protect photographic film, to give nylon stockings longer life, to allow pet snakes to hibernate, and to preserve drugs. At times people opened the refrigerator and gazed in without clear intentions, mentally forag- ing, trying to decide if they were hungry, often removing nothing before they closed the door again. The anthropologists concluded that the refrigerator, and by extension any tool, “enters into the determination of its own utilities, suggesting new ideas for its own definition... and... threatens to take on altogether new identi- ties....” The Internet offers a final, stunning example of this prin- ciple. Only military planners and scientists initially used this communication system. They developed a decentralized design so that messages could not easily be knocked out by power failures, downed computers, or a war. But this same feature made it diffi- cult to monitor and control the Internet. The developers did not imagine such things as Amazon.com, pornography on the net, downloading digitized music to a personal computer, or most of the other things people today use the Internet for. In short, when we review the history of the phonograph, the radio, the refrigera- tor, and the Internet, technologies conceived for one clearly defined use have acquired other, unexpected uses over time. Engi- neers and designers tend to think new devices will serve a narrow range of functions, while the public has a wide range of intentions and desires and usually brings far more imagination to new tech- nologies than those who first market them. 46 Chapter 3 Furthermore, a technology’s symbolic meanings may deter- mine its uses. Too often we think of technologies in purely func- tional terms. However, even so prosaic a device as the electric light bulb had powerful symbolic meanings and associations at its inception. Edison’s practical incandescent light of 1879 was pre- ceded by many forms of “impractical” electric lighting in theaters, where it was used for dramatic effects. A generation before Edi- son’s light bulb even began to reach most homes (after 1910), it was appropriated by the wealthy for conspicuous consumption, used to illuminate public monuments and skyscrapers, and put into electrical signs. As a result, by 1903 American cities were far more brightly lighted than their European counterparts: Chicago, New York, and Boston had three to five times as many electric lights per inhabitant as Paris, London, or Berlin.39 Intensive elec- tric lighting of American downtowns far exceeded the require- ments of safety. The Great White Way and its huge signs had become a national landmark by 1910, and postcards and photo- graphs of illuminated city skylines became common across the United States. In New York, during World War I when wartime energy saving darkened Times Square, the citizens complained that the city seemed “unnatural.” People demanded that the giant advertising signs be turned on again, and they soon were, with new slogans selling war bonds.40 This intensive use of lighting in the United States was in no sense a necessity, and the European preference for less electric advertising was not temporary or the expression of a “cultural lag.” Many European communities still resist electric signs and spectacular advertising displays. At the 1994 Winter Olympics in Norway, the city council of Lillehammer refused Coca-Cola and other sponsors the right to erect illuminated signs. On the city’s streets only wooden and metal signs were permitted. No neon or transparent plastic was allowed. Levels and methods of lighting  Is Technology Predictable? 47 vary from culture to culture, and what is considered normal or necessary in the United States may seem to be a violation of tradi- tion elsewhere.41 The preceding survey shows that, far from being deterministic, technologies are unpredictable. A fundamentally new invention often has no immediate impact; people need time to find out how they want to use it. Indeed, the best technologies at times fail to win acceptance. Furthermore, the meanings and uses people give to technologies are often unexpected and non-utilitarian. Eco- nomics does not always explain what is selected, how it is used, or what it means. From their inception, technologies have symbolic meanings and non-utilitarian attractions. A technology is not merely a system of machines with certain functions; rather, it is an expression of a social world. Electricity, the telephone, radio, television, the computer, and the Internet are not implacable forces moving through history, but social pro- cesses that vary from one time period to another and from one cul- ture to another. These technologies were not “things” that came from outside society and had an “impact”; rather, each was an internal development shaped by its social context. No technology exists in isolation. Each is an open-ended set of problems and pos- sibilities. Each technology is an extension of human lives: some- one makes it, someone owns it, some oppose it, many use it, and all interpret it. Because of the multiplicity of actors, the meanings of technology are diverse. This insight is useful for considering how historians understand technology (chapter 4) and for looking into the relationship between technology and cultural diversity (chapter 5). 4 How Do Historians Understand Technology? The previous two chapters suggest that one must reject techno- logical determinism and admit that the invention and the diffu- sion of technologies are not predictable. What is the alternative? Historians in the field give roughly equal weight to technical, social, economic, and political factors. Their case studies suggest that artifacts emerge as the expressions of social forces, personal needs, technical limits, markets, and political considerations. They often find that both the meanings and the design of an arti- fact are flexible, varying from one culture to another, and from one time period to another. Indeed, Henry Petroski, one of the most widely read experts on design, argues that there is no such thing as perfect form: “Designing anything, from a fence to a fac- tory, involves satisfying constraints, making choices, containing costs, and accepting compromises.”1 Technologies are social con- structions. Historians of technology have also generally agreed that after initial invention comes an equally important stage of “development.” Indeed, since the late 1960s the study of develop- ment has been at the center of much work in the field, for example in studies of Nicholas Otto’s internal-combustion engine or the development of the diesel engine.2 Nathan Rosenberg, a leading economic historian, emphasizes that for every new product or 50 Chapter 4 production technique “there is a long adjustment process during which the invention is improved, bugs ironed out, the technique modified to suit the specific needs of users, and the ‘tooling up’ and numerous adaptations made so that the new product (pro- cess) can not only be produced but can be produced at low cost.” Indeed, during this “shakedown period” of early production some feasible inventions are abandoned as unprofitable.3 As the study of development has increased, the heroic “lone inventor” has largely disappeared from scholarship. Anthony F. C. Wallace, a senior historian in the field, declared: “We shall view technology as a social product and shall not be over much interested in the pri- ority claims of individual inventors, for the actual course of work that leads to the conception and use of new technology always involves a group that has worked for a considerable period of time on the basic idea before success is achieved.”4 Variation in design continues during early stages of develop- ment, until one design meets with wide approval. Once a partic- ular design is widely accepted, however, variation in form gives way to innovation in production. Take the bicycle as an example. The earliest bicycles (high-wheelers) were handmade and cost an ordinary worker a year’s wages. Only the well-to-do could afford them. Most riders were young and male. The danger of toppling from a high-wheeler gave bicycling a macho aura. For more than a generation, low-wheel bikes were for women, clergymen, and old people. Tremendous experimentation took place as inventors changed the size of the wheels, made three-wheelers, moved the larger wheel from the front to the back, and tried out various materials, including wood and steel frames. They developed dif- ferent drive and braking systems, tried various shapes and posi- tions for the handlebars, and created accessories such as lights and panniers. Dunlop developed air-filled rubber tires that together  How Do Historians Understand Technology? 51 with a padded seat reduced discomfort from bumps and vibra- tions. At first, professional racers derided these “balloon” tires as being for sissies. However, bicycle design reached closure in the 1890s, when low-wheel models with front and back wheels of equal size and fitted with Dunlop tires proved to be the fastest on the racetrack. Once the low-wheel “safety bicycle” had become accepted as the standard, manufacturing changed. The leading producers of high-wheelers, such as the Pope Company, had prided themselves on durable construction by skilled artisans, who adjusted the wheel size of each cycle to match the length of a customer’ legs. In the 1880s such bicycles cost $300 or more, well beyond the reach of the average consumer. In the 1890s, however, mass pro- ducers such as Schwinn made bicycles with stamped and welded frames. They were of lesser quality, but priced as low as $50. By 1910 a used bicycle in working order could be had for $15, and ownership had spread to all segments of society. The bicycle had ceased to be a toy for the wealthy and had become a common form of transportation and recreation for millions. The military had adapted it to troop transport, delivery services had thousands of bicycle messenger boys, and bicycle racing had emerged as a professional sport. The social significance and use of the bicycle was not tech- nologically determined. For example, from the beginning some women adopted the bicycle, and during the era of the high- wheeler some joined the popular bicycle clubs. In 1888, eighteen women were members of a Philadelphia club, and one of them won the “Captain’s Cup,” awarded annually to the member who covered the most miles (in this case, 3,3041⁄4). However, women on wheels met opposition. Some physicians declared that the bicycle promoted immodesty in women and harmed their reproductive 52 Chapter 4 organs. Moralists thought women on bicycles were indecent because they wore shorter skirts to ride them, and worried that women would find straddling the seat sexually stimulating. The bicycle craze helped kill the bustle and the corset and encour- aged “common-sense dressing.” Many in the women’s suffrage movement adopted bicycles. In 1896, Susan B. Anthony declared: “Bicycling... has done more to emancipate women than any- thing else in the world. I stand and rejoice every time I see a woman ride by on a wheel. It gives women a feeling of freedom and self-reliance.”5 The women’s movement embraced the bicy- cle, and its democratization became part of their drive for social equality. Outside the United States, the bicycle persisted much longer as an important form of transportation. In the Netherlands and in Denmark, bicycles were more common than automobiles were until the early 1960s. In those countries, major roads have special lanes and special traffic signals for bicyclists, and government pro- grams encourage citizens to use bicycles instead of automobiles. But most Western societies have chosen the automobile as the primary mode of transportation instead, and even in Denmark and the Netherlands bicyclists are not as numerous as they were a generation ago. Despite the bicycle’s head start on the automo- bile, in most societies only the automobile seemed to achieve what Thomas Hughes calls “technological momentum.” Hughes argues that technical systems are not infinitely malle- able. If technologies such as the bicycle or the automobile are not independent forces shaping history, they can still exercise a “soft determinism” once they are in place. “Technological momentum” is a particularly useful concept for understanding large-scale sys- tems, such as the electric grid, the railway, or the automobile. In Networks of Power, Hughes examines five stages of system develop-  How Do Historians Understand Technology? 53 ment for the electrical grid, and these stages can apply to other inventions as well. In the case of electrification, the sequence began in the 1870s with invention and early development in a few locations (1875–1882). That was followed by technology transfer to other regions (1882–1890). With successful transfer came growth (1890–) and the development of subsidiary infrastructures of production, education, and consumption, leading to techno- logical momentum (after c. 1900) as electricity became a standard source of light, heat, and power. In the mature stage (after. c. 1910), the problems faced by management required financiers and consulting engineers.6 “Technological momentum” is not inherent in any technologi- cal system when first deployed. It arises as a consequence of early development and successful entrepreneurship, and it emerges at the culmination of a period of growth. The bicycle had such momentum in Denmark and the Netherlands from 1920 until the 1960s, with the result that a system of paved trails and cycling lanes were embedded in the infrastructure before the automo- bile achieved momentum. In the United States, the automobile became the center of a socio-technical system more quickly and achieved momentum a generation earlier. Only some systems achieve “technological momentum,” which Hughes has also applied to analysis of nitrogen fixation systems and atomic energy.7 The concept seems particularly useful for understanding large systems. These have some flexibility when being defined in their initial phases. But as technical specifications are established and widely adopted, and as a system comes to employ a bureau- cracy and thousands of workers, it becomes less responsive to out- side pressures. Hughes provided an example in American Genesis: “... the inertia of the system producing explosives for nuclear weapons arises from the involvement of numerous military, 54 Chapter 4 industrial, university, and other organizations, as well as from the commitment of hundreds of thousands of persons whose skills and employment are dependent on the system.”8 Similarly, at the end of the nineteenth century, once the width of railway tracks had been made uniform and several thousand miles were laid out, once bridges and grade crossings were designed with rail cars of certain dimensions in mind, it was expensive and impractical to reconfigure a railway system. Hughes makes clear when discussing “inertia” that the concept is not only technical but also cultural and institutional. A society may choose to adopt either direct current or alternating current, or to use 110 volts, or 220 volts, or some other voltage, but a gener- ation after these choices have been made it is costly and difficult to undo such a decision. Hundreds of appliance makers, thou- sands of electricians, and millions of homeowners have made a financial commitment to these technical standards. Further- more, people become accustomed to particular standards and soon begin to regard them as natural. Once built, an electrical grid is “less shaped by and more the shaper of its environment.”9 This may sound deterministic, but it is not entirely so, for people decided to build the grid and selected its specifications and com- ponents. To later generations, however, such technical systems seem to be deterministic.10 The US electrical system achieved technological momentum around 1900. By that time, it was “reinforced with a cultural con- text, and interacting in a systematic way with the elements of that context,” and “like high momentum matter [it] tended in time to resist changes in the direction of its development.”11 From 1900 on, growth was relentless and not easily deflected by con- tingencies. The electrical system was far more than machines themselves; it included utility companies, research laboratories,  How Do Historians Understand Technology? 55 regulatory agencies, and educational institutions, constituting what Hughes calls a “sociotechnical system.” It had high momen- tum, force, and direction because of its “institutionally structured nature, heavy capital investments, supportive legislation, and the commitment of know-how and experience.”12 Similarly, the auto- mobile achieved technological momentum not as an isolated machine, but as part of a system that included road building, driver education programs, gas stations, repair shops, manufac- turers of spare parts, and new forms of land use that spread out the population into suburbs that, practically speaking, were acces- sible only to cars and trucks. The concept of technological momentum provides a way to understand how large systems exercise a “soft determinism” once they are in place. Once a society chooses the automobile (rather than the bicycle supplemented by mass transit) as its preferred sys- tem of urban transportation, it is difficult to undo such a decision. The technological momentum of a system is not simply a matter of expense, although the cost of building highways, bicycle lanes, or railroad tracks is important. Ultimately, the momentum of a society’s transport system is embodied in the different kinds of cities and