Beniger 1986

Beniger, James. The Control Revolution: Technological and Economic Origins of the Information Society. Harvard UP, 1986.

Human society seems rather to evolve largely through changes so gradual as to be all but imperceptible, at least compared to the generational cycles of the individuals through whose lives they unfold. Second, contemporaries of major societal transformations are frequently distracted by events and trends more dramatic in immediate impact but less lasting in significance.

Page 2

…society is currently experiencing a revolu- tionary transformation on a global scale. Unlike most of the other writers, however, I do not conclude that the crest of change is either recent, current, or imminent. Instead, I trace the causes of change back to the middle and late nineteenth century, to a set of problems— in effect a crisis of control—generated by the industrial revolution in manufacturing and transportation. The response to this crisis, at least in technological innovation and restructuring of the economy, occurred most rapidly around the turn of the century and amounted to nothing less, I argue, than a revolution in societal control.

Page 6

Beginning most noticeably in the United States in the late nineteenth century, the Control Revolution was certainly a dramatic if not abrupt discontinuity in technological advance. Indeed, even the word revo- lution seems barely adequate to describe the development, within the span of a single lifetime, of virtually all of the basic communication technologies still in use a century later: photography and telegraphy (1830s), rotary power printing (1840s), the typewriter (1860s), trans- atlantic cable (1866), telephone (1876), motion pictures (1894), wireless telegraphy (1895), magnetic tape recording (1899), radio (1906), and television (1923).

Page 7

Along with these rapid changes in mass media and telecommuni- cations technologies, the Control Revolution also represented the be- ginning of a restoration—although with increasing centralization—of the economic and political control that was lost at more local levels of society during the Industrial Revolution. Before this time, control of government and markets had depended on personal relationships and face-to-face interactions; now control came to be reestablished by means of bureaucratic organization, the new infrastructures of trans- portation and telecommunications, and system-wide communication via the new mass media.

Page 7

Inseparable from the concept of control are the twin activities of information processing and reciprocal communication, complementary factors in any form of control. Information processing is essential to all purposive activity, which is by definition goal directed and must therefore involve the continual comparison of current states to future goals, a basic problem of information processing. So integral to control is this comparison of inputs to stored programs that the word control itself derives from the medieval Latin verb contrarotulare, to compare something “against the rolls,” the cylinders of paper that served as official records in ancient times.

Page 8

So central is communication to the process of control that the two have become the joint subject of the modern science of cybernetics, defined by one of its founders as “the entire field of control and communication theory, whether in the machine or in the…

Page 8

Because both the activities of information processing and commu- nication are inseparable components of the control function, a society’s ability to maintain control—at all levels from interpersonal to inter-

Page 8

Pen

Page 8

national relations—will be directly proportional to the development of its information technologies.

Page 9

‘Technology may therefore be considered as roughly equivalent to that which can be done, excluding only those capabilities that occur naturally in living systems.

Page 9

Like these earlier revolutions in matter and energy processing, the Control Revolution resulted from innovation at a most fundamental level of technology—that of infor- mation processing.

Page 9

Information processing may be more difficult to appreciate than matter or energy processing because information is epiphenomenal: it derives from the organization of the material world on which it is wholly dependent for its existence.

Page 9

…bureaucratic organization tends to appear wherever a collective activ- ity needs to be coordinated by several people toward explicit and impersonal goals, that is, to be controlled. Bureaucracy has served as the generalized means to control any large social system in most in- stitutional areas and in most cultures since the emergence of such systems by about 3000 B.c.

Page 13

Nevertheless, bureaucratic administration did not begin to achieve anything approximating its modern form until the late Industrial Revolution.

Page 14

Rationalization must therefore be seen, following Weber, as a complement to bureaucratization, one that served control in his day much as the preprocessing of information prior to its processing by computer serves control today.

Page 15

Tne reason why people can be governed more readily qua things is that the amount of information about them that needs to be processed is thereby greatly reduced and hence the degree of control—for any constant capacity to process information—is greatly enhanced. By means of rationalization, therefore, it is possible to main- tain large-scale, complex social systems that would be overwhelmed by arising tide of information they could not process were it necessary to govern by the particularistic considerations of family and kin that characterize preindustrial societies.

Page 15

In short, rationalization might be defined as the destruction or ig- noring of information in order to facilitate its processing.

Page 15

The rapid development of rationalization and bureaucracy in the mid- dle and late nineteenth century led to a succession of dramatic new information-processing and communication technologies. These inno- vations served to contain the control crisis of industrial society in what can be treated as three distinct areas of economic activity: production, distribution, and consumption of goods and services.

Page 16

Machinery itself came in- creasingly to be controlled by two new information-processing tech-

Page 16

nologies: closed-loop feedback devices like James Watt’s steam governor (1788) and preprogrammed open-loop controllers like those of the Jac- quard loom (1801). By 1890 Herman Hollerith had extended Jacquard’s punch cards to tabulation of U.S. census data. This information- processing technology survives to this day—if just barely—owing largely to the corporation to which Hollerith’s innovation gave life, Interna- tional Business Machines (IBM). F…

Page 17

The resulting flood of mass-produced goods demanded comparable innovation in control of a second area of the economy: distribution. Growing infrastructures of transportation, including rail networks, steamship lines, and urban traction systems, depended for control on a corresponding infrastructure of information processing and telecom- munications.

Page 17

This coevolution of the railroad and telegraph systems fostered the development of another communication infrastructure for control of mass distribution and consumption: the postal system.

Page 17

The mechanism for communicating information to a national audience of consumers developed with the first truly mass medium: power- driven, multiple-rotary printing and mass mailing by rail.

Page 18

Although most of the new information technologies originated in the private sector, where they were used to control production, distri- bution, and consumption of goods and services, their potential for controlling systems at the national and world level was not overlocked by government. Since at least the Roman Empire, where an extensive road system proved equally suited for moving either commerce or troops, communications infrastructures have served to control both economy and polity. As corporate bureaucracy came to control in-…

Page 20

One major result of the Control Revolution had been the emergence of the so-called Information Society. The concept dates from the late 1950s and the pioneering work of an economist, Fritz Machlup, who first measured that sector of the U.S. economy associated with what he called “the production and distribution of knowledge” (…

Page 21

The same report introduced the neologism telematics for this most recent stage of the Information Society, although similar words had been suggested earlier—for ex- ample, compunications (for “computing + communications”) by An-

Page 25

Crucial to telematics, compunications, or whatever word comes to be used for this convergence of information-processing and commu- nications technologies is increasing digitalization: coding into discon- tinuous values—usually two-valued or binary—of what even a few years ago would have been an analog signal varying continuously in time, whether a telephone conversation, a radio broadcast, or a television picture. Because most modern computers process digital information, the progressive digitalization of mass media and tele- communications content begins to blur earlier distinctions between the communication of information and its processing (as implied by the term compunications), as well as between people and machines. Dig- italization makes communication from persons to machines, between machines, and even from machines to persons as easy as it is between persons. Also blurred are the distinctions among information types: numbers, words, pictures, and sounds, and eventually tastes, odors, and possibly even sensations, all might one day be stored, processed, and communicated in the same digital form.

Page 25

In this way digitalization promises to transform currently diverse forms of information into a generalized medium for processing and exchange by the social system, much as, centuries ago, the institution

Page 25

of common currencies and exchange rates began to transform local markets into a single world economy. We might therefore expect the implications of digitalization to be as profound for macrosociology as the institution of money was for macroeconomics.

Page 26

…of control and the resulting Control Revo- lution is that particular attention to the material aspects of information processing, communication, and control makes possible the synthesis of a large proportion of the literature on contemporary social change.

Page 26

For example, could a control rev- olution have come before an industrial one? (The answer, as we shall see, is clearly no.)

000032.pdg

Pen

000032.pdg

That these circumstances have shifted from land and capital to information-that information has emerged as the material base of modern economies-challenges the social theory we have inherited from the nineteenth century, much as the Industrial Revolution challenged Marx and other thinkers of that era to recon- sider preindustrial theories.

000032.pdg

One such model, suggested in the previous chapter, is that of society as a processing system, one that sustains itself by extracting matter and energy from the environment and distributing them among its members.

000032.pdg

What we recognize in the end-directedness or purpose of organi- zation is the essential property of control, already defined as purposive influence toward a predetermined goal. Control accounts for the dif- ference between even the most complex inorganic crystal and simple organisms like the amoeba: the amoeba controls both itself and its environment; the crystal does not. As noted in the previous chapter, everything living processes information to effect control; nothing that is not alive can do so-nothing, that is, except certain artifacts of our own invention, artifacts that proliferated with the Control Revo- lution.

000035.pdg

Purposive organization and control, in other words, define the tan- gible discontinuity that distinguishes life from the inorganic universe. On one side, the exclusive province of the physical sciences, we find only matter, energy, and their ordering in the epiphenomenon we call information. On the other side, our own side in that we ourselves are living systems, we find structures purposively organized (in von Neu- mann's sense) for information processing, communication, and control, the special subject matter of the behavioral and life sciences.

000035.pdg

Here, then, is the most fundamental reason why the Control Rev- olution has been so profound in its impact on human society: it trans- formed no less than the essential life function itself. Rapid technological expansion of what Darwin called life's "marvellous structure and prop- erties" and what we now see to include organization, information pro- cessing, and communication to effect control constitute a change unprecedented in recorded history. We would have to go back at least to the emergence of the vertebrate brain if not to the first replicating molecule-marking the origin of life on earth-to find a leap in the capability to process information comparable to that of the Control Revolution.

000036.pdg

Because societies must also be concrete open systems if they are to sustain their organization against the progressive degrading of their collective energy, the view of organisms as concrete open processing systems applies equally to their socialaggregates. The essence of human society, in other words, is its continuous processing of physical throughputs, from their input to the concrete social system to their final consumption and output as waste.

000037.pdg

One idea directly inspired by the Control Revolution in technology is the concept of a program, a word that first appeared in the seventeenth century for public notices but which in the past 150 years has spread through other organizational and informational technologies: plans for formal proceedings (1837), political platforms (1895), broadcast pre- sentations (1923),electronic signals (1935),computer instructions (1945), educational procedures (1950), and training (1963). In general, pro- gram has come to mean any prean-anged information that guides sub- sequent behavior.

000039.pdg

All control is thus programmed: it depends on physically encoded information, which must include both the goals toward which a process is to be •influenced and the procedures for processing additional information toward that end.

000040.pdg

The difference reflects the progress of the Control Revolution which, as already noted, resulted in a fundamental change in human thought between the 1870s and 1930s.

000043.pdg

Born just three years after Maxwell's seminal paper on control theory, in the same decade that brought technological inno- vations like the telephone, phonograph, and microphone, the demon was not retired until the understanding of information had reached a high level of quantitative sophistication-…

000048.pdg

The basic lessons of Maxwell's demon-that control involves pro- gramming, that programs require inputs of information, that infor- mation does not exist independent of matter and energy and the ref ore must incur costs in terms of increased entropy-all seem commonplace today.

000048.pdg

It is hardly surprising that, with the growing understanding of pro- grammed control by computers after World War II, scientists would begin to decipher the ancient language of DNA and to exploit its programming in new technologies.

000056.pdg

Pen

000056.pdg

As happened so often since the advent of the Control Revolution, concepts from information and communicationtechnology-here Morse's binary telegraph code-helped scientists to reconceptualize traditional subjects like cellular biology.

000057.pdg

Thus the world commercial system foundered for centuries in a vicious cycle in which poor communications and the resulting lack of information prevented the increased specialization and control that would have made specialization itself less necessary.

000124.pdg

How, then, do we account for the obsession of the American colonial merchant with kin and personal relationships in his business or for the persistence of the family partnership as the dominant form of com- mercial organization well into the nineteenth century? The answer, as we shall see, lies in the need to control widely dispersed transactions without adequate telecommunications or effective legal sanctions. If lack of sufficient information-processing, communication, and control technology caused the retention of traditionalist values in commerce, it seems reasonable to expect the converse: that rationalization of these values will follow improvements in the same technology.

000128.pdg

DESPITE the modest technological and economic innovations in control under commercial capitalism, the world djstributional system, even in the early nineteenth century, • sim'ply did not require anything ap- proaching: the degree of control that would become ne~essary under raP,iilindustrializatio:p. Because· capital remained so mobile under the centuries-old commercial order, it served as the major medium-in the form of money and commercial paper-for communication and control of the world Sys\em.

000169.pdg

If profit provided the incentive to process matter faster under in- dustrial capitalism, steam power provided the means. The difference, in a word, was speed.

000169.pdg

To the extent that market conditions change faster than buyers and sellers are able to respond, using the infrastructure of information processing and communication available to them, or that the costs of information about prices remain high, markets will deviate from the classical economic model, which assumes free and perfect information and instantaneous transactions.

000172.pdg

The parameter that this book finds to be most central to the Control Revolution, namely speed, drew the most in- novations in its automatic control (six)-all after the mid-1780s. In short, Table 5.1, insofar as it establishes at least a temporal correlation between early industrialization and what Mayr calls a "veritable break- through" in feedback technology, bolsters a central argument of this book: the Industrial Revolution and the harnessing of inanimate sources of energy to material processes more generally led inevitably to an increased need for control.

000175.pdg

By developing this infrastructure, essential for the processing and distribution of matter and energy, the Commercial Revolution helped to establish necessary preconditions for the Industrial Revolution, in effect the application of inanimate energy to the material processing system.

000177.pdg

What made the Commercial Revolution truly revolutionary was that, for the first time, distributional and control systems including infor- mation processing, programming, and telecommunications could be sustained indefinitely on a global basis. Industrialization became rev- olutionary when the energy harnessed vastly exceeded that of any naturally occurring or animate source; the resulting throughput and processing speeds greatly exceeded the capability of unaided humans to control. What made the Control Revolution in fact revolutionary was the development of technologies far beyond the capability of any individual, whether in the form of the massive bureaucracies of the late nineteenth century or of the microprocessors of the late twentieth century. In all cases it was not the novelty of the commodities pro- cessed (whether matter, energy, or information) that proved decisive, contrary to Bell, but rather the transcendence of the information- processing capabilities of the individual organism by a much greater technological system.

000185.pdg

As we have already seen, the colonial merchant, a generalist who embraced all types of products and embodied all basic commercial functions, differed little from his counterpart in fifteenth-century Ven- ice. Within two generations, however, these general merchants had been largely replaced by more specialized workers: shipowners, fin- anciers, jobbers, transporters, insurers, brokers, auctioneers, retail- ers-a growing network of middlemen to process and move material goods. What merchants remained came increasingly to specialize in only one or two lines of goods, and to concentrate on a single commercial function: importing, wholesaling, retailing, or exporting.

000185.pdg

This squares with the systems view that tertiary services constitute a necessary precondition of industrialization-and not only because they facilitate the movement and processing of matter and energy by the system. Tertiary services also enable businessmen to specialize in only a few lines of goods and even to concentrate on a single commercial function…

000189.pdg

To power the extraction, processing, and distribution of material throughputs to industrial production, the system turned for its energy from sources animate (human and draft animal) to sources inanimate (steam-powered machinery) and from natural kinetic sources (wind and water) to sources chemical (coal and electric batteries). For this reason anthracite and iron mining gained in importance in the primary sector relative to agriculture, while textile and metals production- the most effective early applications of steam power-came to domi- nate the secondary sector. Steam-powered transportation, especially railroads and steamship lines, speeded processing and distribution; even faster electrical communicationvia a national telegraph grid helped to control the new systems of transportation and commerce.

000202.pdg

With the development of a year-round, dependable, and predictable transportation system that could move throughputs at the speed of steam, after 1840, the Industrial Revolution-grounded in similarly fast, steam-powered factory production-could at last take hold in the United States, nearly a century after its beginning in Great Britain.

000214.pdg

As domestic anthracite provided the energy, pig iron the matter, and steam-driven machinery the material processing for industrialization, the rapidly expanding rail network provided the infrastructure to move throughputs on an interregional, national, and finally continental scale.

000214.pdg

Dramatic problems of control first appeared, as might be expected, on the railroads, the first part of the material processing system to harness the speed of steam power on a large scale. Because early railroads operated for most of their length on only a single track at unprece- dented speeds of up to thirty miles per hour, they faced the problem of especially dangerous and costly head-on collisions. Lacking modern communication and control technology, most railroads adopted one of two solutions. On longer, lightly traveled roads, all trains ran one way one day and the other way the next. This solution did not prove eco- nomical or convenient enough for shorter, busier routes, however, where the first of two trains scheduled to meet running in opposite directions would wait at a midpoint station or siding until the other had passed. Without the technologies of centralized bureaucratic con- trol, telegraphic communication, and formalized operating procedures along the line, however, and lacking even standardized signals, time- tables, and synchronized watches aboard each train, many accidents did occur (Fig. 6.1).

000221.pdg

Control of each train became centralized in its conductor, who had standardized detailed programs for responding to delays, breakdowns, and other contingencies, who carried a watch synchronized with all others on the line, and who moved his train according to precise time- tables.

000225.pdg

To describe the conductors on the reorganized Western line as "pro- grammed" might at first seem anachronistic, a needless intrusion of contemporary jargon into the early nineteenth century. The fact re- mains, however, that in their control of trains the Western conductors might have been replaced in many of their functions by on-board micro- computers or, given modern telecommunications, by a more centralized means of computer control. Seen in this way, the Western conductors take on new significance: they are possibly the first persons in history to be used as programmable, distributed decision makers in the control of fast-moving flows through a system whose scale and speeds pre- cluded control by more centralized structures. This use of human beings, not for their strength or agility, nor for their knowledge or intelligence, but for the more objective capacity of their brains to store and process information, would become over the next century a dominant feature of employment in the Information Society.

000225.pdg

Pen

000225.pdg

With the rapid diffusion of the telegraph, after Morse's successful demonstration in 1844, and the adoption and refinement of the West- em's organizationalinnovations, the danger of collisionsno longer ranked as the railroad's major control problem by the 1850s. In the first half of that decade, which brought the first four trunk lines-the Erie (1851),Baltimore and Ohio (1852),New York Central (1853),and Penn- sylvania (1854)-connecting East and West, the control crisis of the railroads shifted from safety to efficiency in keeping track ·of trains, cars, and personnel in increasingly large, complex, and busy systems.

000226.pdg

The entire system of transportation and distribution derived pro- gressive coordination and integration-as Table 6.1 suggests-from a succession of new communication systems involving both infrastruc- ture or common carriers and generalized media of exchange. The com- mon caniers, which developed increasingly apart from the develop- ment of the railroads, included the telegraph, postal, and telephone systems-all point-to-point networks well suited to control transpor- tation and distribution. New generalized media of exchange-and therefore of communicationand control-included postage stamps (1852), the through bill of lading (1853), federal paper currency (1862), and postal money orders (1864), among many others.

000236.pdg

Just as the early pioneers of genetic biology were information specialists from physics and mathematics, as we saw in Chapter 2, and the pioneers of computer control came from mathematics and engi- neering, so too did the early leaders of organizational control-as typ- ified by Charles Babbage, mathematician, and David McCallum, engineer-represent abstract and analytic rather than practical ex- perience with information processing and decision. In times of true crisis, it would seem, experience with the old technologies provides little help in devising revolutionary new ones-more theoretical and general disciplines better fill that need.

000237.pdg

Pen

000237.pdg

Coordination and control of production therefore became no less important than the innovations in equipment or more intensive applications of energy that made pos- sible the continual increases in volume and speed. Producers with the best control technologies could maintain the greatest speeds, produce at the lowest costs, and thereby enjoy an important edge on compet- itors.

000238.pdg

Ironically enough, two technologies that would be central to much wider control of material economies in the twentieth century-namely, Hollerith punch cards and operations research-emerged early in the Industrial Revolution in England and France for the control of steam-driven mechanical production…

000245.pdg

The key innovation in social technology was the commodity exchange, based on the telegraph and later on telephone exchanges, which permitted crops to be sold in transit and even before harvest and allowed the exploitation of even minute-by-minute changes in prices.

000249.pdg

Commodity exchanges accompanied diffusion of the telegraph, which was launched in 1844 and in eight years comprised a continental tele- communications network of some twenty-three thousand miles (Fig.

000249.pdg

Successful coordination of through freight traffic, combined with the growing network of elevators, warehouses, and other storage facilities increasingly accessible by telegraphic communication, meant that the flow of agricultural commodities, despite its unprecedented velocity, could be regulated with precision. Deliveries could be scheduled for times when manufacturers would be ready to process or when retail inventories would likely be depleted, and trade in agricultural com- modities could be carried on throughout the year.

000252.pdg

As with the control of production (Table 6.2), early control of distribution came mostly through innovations in pre- processing, in this case to faciliate market transactions, the processing equivalent of throughputs to production. These marketing innovations included-within a fifty-year period-standardized methods of sort- ing, grading, weighing, and inspecting, packaging in containers of fixed sizes and weights, fixed prices, standardized sizes, and periodic pre- sentation to consumers via catalog.

000261.pdg

Not until steam power had sufficiently increased the speed and volume of material processing and the resulting increase in outputs could be distributed widely with the precision made possible by coevolving networks of railroad and telegraph did bureaucratic control become more efficient and more profitable than coordination by the market.

000262.pdg

Power-driven printing distributed by rail, the major mass medium before broadcasting, improved rapidly-parallel to the developing cri- sis in control of consumption-though a spate of innovations: the first electric press (1839) and rotary printing (1846), wood pulp and rag paper and the curved stereotype plate (1854), paper-folding machines (1856), the mechanical typesetter (1857), high-speed printing and fold- ing press (1875), and linotype (1886).

000271.pdg

The nineteenth-century revolution in information technology was pred- icated on if not directly caused by social changes associated with earlier innovations. Just as the Industrial Revolution presupposed a com- mercial system for capital allocations and the distribution of goods, as we saw in Chapters 4 and 5, the Control Revolution developed in re- sponse to problems arising out of advanced industrialization:a mounting crisis of control at the most aggregate level of national and international systems, levels that had had little practical relevance before the mass production, distribution, and consumption of factory goods.

000278.pdg

…the rapid development of officetechnology during early industrialization. In 1780 a modern American office might contain printed business forms and file cabinets, communicate via mail, parcel post and courier, subscribe to various types of news publications, and hire financial and other professional information services. In general, informational goods and services-as well as media and content-were still sharply separated. Bureaucracy, where it could be said to exist at all, lacked structural

000279.pdg

differentiation and specializationof function.

000280.pdg

A century later, in the midst of the revolution in generalized infor- mation processing and control, the modern American office had added a dozen major information technologies and services: telegraph and telephone, international record carriers and other local delivery ser- vices, newsletter, loose-leaf and directory subscriptions, news and ad- vertising services, and differentiated security systems (includingdistrict telegraphs that could summon police or the fire brigade with the turn of a crank). By the 1890stypewriters, phonographs, and cash registers had also come into common use in American business (Fig. 6.10). The new office technologies and services, added since the advent of indus- trialization and the resulting need for increased control, reflected a trend toward integration of informational goods and services, media and content, that has continued unabated to…

000280.pdg

…a spate of basic inventions to improve the generation of infor- mation within bureaucracy. These innovations included blotting paper (1856), the pencil with eraser and steel (Esterbrook) pen (1858), carbon paper (1872), and modern keyboard typewriter (1873);all can be found in many offices to this day. The new tools for creating information could also be applied-by the 1870s-to the preprocessing of new informational inputs to the modern office: not only the loose-leaf and directory services already mentioned but also the stock ticker (1870), messenger news service (1882), and press clipping service (1884). Pro- cessing of numerical data came to be facilitated by two inventions- the keyboard calculator (1887)and punch-card tabulator (1889)-whose social implications would be felt well into the twentieth century…

000283.pdg

Recording or information storage capabilities increased with the systematization of shorthand, including the first professional shorthand journal (1848), the systematization of office record keeping (early 1870s), and the dictating machine (1885).

000283.pdg

Office communica- tion improved with a continuous stream of innovations, from manu- factured envelopes (1839) to the desk telephone (1886) (Fig.

000284.pdg

Mass media were not sufficient to effect true control, however, with- out a means of feedback from potential consumers to advertisers, a mechanism that would restore to the emerging national and world markets an essential relationship of the earlier segmental markets: communication from consumer to producer…

000287.pdg

The smooth transition from control crisis to Control Revolution in the 1880s and 1890scan be attributed to three primary dynamics, each of which has sustained the steady development of information societies through the twentieth century. First, control technologies have co- evolved with energy utilization and processing speeds in a positive spi- ral, advances in any one factor causing or at least enabling improvements in the others.

000291.pdg

Second, increased control brought increased reliability and hence predicwbility of processes and flows, which in turn meant increasing

000291.pdg

economic returns on the application of information-processing tech- nology.

000292.pdg

Contrary to prevailing views, which locate the origins of the Information Society in World War II (Wiener 1948, 1950)or in the commercial development of tele- vision (McLuhan 1964)or computers (Berkeley 1962; Martin and Nor- man 1970;Tomeski 1970;Hawkes 1971)during the 1950s, or computer- based telecommunications in the 1960s and early 1970s (Brzezinski 1970; Oettinger 1971; Hiltz and Turoff 1978; Martin 1978, 1981; Nora and Mine 1978),or microprocessing technology in the late 1970s (Evans 1979;Forester 1980; Laurie 1981), we shall see from this analysis that the basic societal transformation from Industrial to Information So- ciety had been essentially completed by the late 1930s.

000293.pdg

Pen

000293.pdg

Only in the late 1870s and 1880s did advances in actual information processing-as opposed to information reduction or pre- processing-come to industrial production. These innovations gave the

000293.pdg

late nineteenth-century factory the components basic to all information processors: internal communication and process control, as in the shop- order systems based on routing slips (mid-1870s); hierarchical differ- entiation and specialization for control, as in the rate-fixing depart- ments (early 1880s); programmed control, as in the cost control of factories (1885); and data collection and storage, as with the new au- tomatic recording devices (late 1880s).

000294.pdg

In essence, scientific management aimed to preprocess the activities of individual workers qua processors, much as earlier efforts at pre- processing in industrial production-interchangeable parts, standard- ization of sizes, integration of flows-had focused on the entire factory as a continuous processor.

000294.pdg

Essential to the assembly line is the standardization and inter- changeability of parts, ideas "in the air" almost from the beginnings of industrialization in the late eighteenth century (Giedion 1948, pp. 47-50) and the basis of the American System of manufacturing by the 1840s.

000298.pdg

Certainly American contributions to automatic control came independently and brought increasing numbers of patents by the 1880s, when Elmer Sperry began work on a regulator for dynamo-electric machines that exploited au- tomatic control.. Of Sperry's nineteen patent applications between 1883 and 1887, eleven included some form of automatic control, more than half involving closed-loopfeedback (Hughes 1971,pp. 45-46). Although Sperry's inventions were hardly unique (the U.S. Patent Officegranted protection to twenty-two generator regulators in 1884alone), his early career does provide further evidence that information engineering, cybernetics, and even computer science trace their origins to the…

000304.pdg

and 1890s-the beginning of the Control Revolution-and not to World War II or to subsequent developments.

000306.pdg

Pen

000306.pdg

Increasingly the Control Revolution meant that industrial proc- esses could be run with little or no human intervention: a bread plant (1910), an electric substation (1914), a photographic film-developing studio (1926), the doors of commercial establishmen…

000307.pdg

Also crucial to control of rapid and long-distance transportation was radio, a wholly new medium of telecommunications that developed almost parallel to the Control Revolution followingpublication of Max- well's theory of electromagnetic radiation in 1873. Within the decade and for the remainder of the century, the problem of exploiting radio waves for telecommunications occupied leading scientists throughout the world…

000318.pdg

Computer scientists whom I have asked informally to compare these six information-processing problems rate the achievements of Sperry and Ford above those of Bush, Torres, and Fischer and Harris, ten- tative evidence that the control of transportation-which necessarily involves control of complex movements, processes, and speed-pre- sents a greater challenge for computing than number-crunching per se.

000323.pdg

Transportation has always been a field with countless and continuing problems of control and hence information processing-a field which historians of computing have not yet begun to explore.

000323.pdg

Pen

000323.pdg

The coevolvingnetworks of transportation and communicationserved to move still more generalized media of exchange, including travelers' checks (1891), precanceled stamps (1917), and facsimile bank checks (1926). In 1918 the first electric funds transfer system, known today as "Fedwire," eliminated the medium of paper in moving money be- tween the Federal Reserve and member banks; two years later Pitney Bowes eliminated the need for postage stamps when it secured federal approval of …

000329.pdg

Stereotyping, the casting of metal in pulp-paper molds to produce printing plates, allowed for print advertisements of more than a single column's width without distracting vertical rules. Despite the great expense of stereotyping machinery, forty-five such systems operated in the United States by 1880.

000356.pdg

Half-tone illustrated mass printing brought a new medium of mass advertising: the mass-circulation magazine. Pulp "mail-order maga- zines," so called because mail-order advertisements engulfed what lit- tle popular fictionthey contained, had reached circulationsof five hundred thousand by the early 1870s, with several million subscribers-mostly rural and poorly educated-in the late 1880s…

000357.pdg

With the rise of the mass-circulation magazine, a refinement of the essential idea of the mail-order magazine, printing had at last become- some four centuries after Gutenberg-not merely a means of mass production but also a mass medium, a new channel for advertising and hence the stimulation and control of mass production itself.

000358.pdg

As late as 1894, only 30 percent of advertisements in four selected publications con- tained illustrations; the proportion increased steadily to nearly 90 per- cent by 1919 (Kitson 1921, p.

000359.pdg

Thus did the power of visual communication to stimulate and control consumer demand come to be realized-nearly a half-century before commercial television-in the mass-circulation magazine.

000359.pdg

New technologies made possible the modern mass-circulation daily newspaper at about the same time. Except for the linotype, developed after 1886, the technological revolution in power mass printing had been· essentially completed in 1883, when Joseph Pulitzer took over the New York World and transformed it into what most newspaper historians consider America's first modern newspaper. The following half-century, the transitional period of the Control Revolution, brought several printing innovations: the sextuple press (1891), which could print and fold 90,000 four-page newspapers in an hour; the web-fed four-color rotary (1892) and color rotogravure (1904) presses; the au- tomatic plate-casting and finishing machine (1900), which greatly in- creased the speed of stereotype printing; and the teletypesetter (1932), a paper tape punch and drive for linotype.

000359.pdg

…the comic book (1904), originally compilations of colored cartoons previously published in newspapers; the crossword puzzle (1913), special rotogravure section (1914), illustrated daily tab- loid (1919),and compositephotographic layout or "composograph"(1925).

000360.pdg

The growth and decline of the daily newspaper as a means of mass communication in the United States perfectly parallels the transition to the Control Revolution in mass consumption.

000361.pdg

Despite continuing innovations in mass publishing, and even though newspa- pers have always attracted a greater share of advertising than have radio and television combined, after the 1920s bureaucratic control of consumption based on national advertising came increasingly to depend on radio and television broadcasting.

000362.pdg

In contrast to the twenty magazines that had attained circulations of 100,000by 1905when two millionread Pulitzer's World, radio broad- casts in the mid-1920s reached fifty million listeners; by the late 1930s the programming of a single advertising agency received one million fan letters per week (Fox 1984, pp. 152, 160).

000367.pdg

Chapter 6 de- scribed a crisis of control in office technology and bureaucracy in the 1880s, as the growing scope, complexity, and speed of information processing-including inventory, billing, and sales analysis-began to strain the manual handling systems of large business enterprises. This crisis had begun to ease by the 1890s, owing to innovations not only in the processor itself (formal bureaucratic structure) but also in its information creation or gathering (inputs), in its recording or storage (memory), in its formal rules and procedures (programming), and in its processing and communication (both internal and as outputs to its environment).

000390.pdg

Characteristic of the transition to the Control Revolution in bu- reaucracy is the history of the Library Bureau of Boston. Founded in 1876 as an offshoot of the American Library Association, the Bureau had discovered an increasingly good business in providing specialized library supplies and equipment not available elsewhere. By 1894, ac- cording to the Bureau's Classified Illustrated Catalog of that year, it had made an even more startling discovery: "There is hardly a library article on our list that is not also used in offices."

000394.pdg

Bolstered by the growing generalization and convergence of infor- mation-processing technology across all offices, the Library Bureau began a separate department of Improved Business Methods, joining the ranks of a growing number of "systematizers," what today would be called management consultants, an early application of scientific management to bureaucracy. What had less than twenty years earlier been a professional association for librarians, the first information sci- entists, now boasted that "among life and fire insurance companies, banks, railways, large manufacturing establishments, and to repre- sentative houses in almost every line, it has not only suggested and installed better methods and improved machinery, but it has also ef- fected great savings in expense." Two years later, in March 1896, the Library Bureau became the exclusive agent for Hollerith data- processing equipment in England, France, Germany, and Italy and soon contracted with Travelers' Insurance to compile a year's records using the new system…

000394.pdg

Desk-top calculators, which owe their intellectual origins to the seventeenth-century machinery of Wil- helm Schickard, Blaise Pascal, and Gottfried Leibniz, among others (Flad 1963), had been. commercially mass-produced in Europe since about 1820 (Turck 1…

000399.pdg

Punched paper cards, developed by Joseph-Marie Jacquard in 1801 to program the patterns woven in power looms (Fig. 6.3) and adopted by Babbage in 1833 for data inputs and programming for his Analytical Engine, had by 1884 been perfected by Herman Hollerith as a medium for electro- mechanical information processing and tabulation. In 1889 Hollerith received U.S. patent 395,781, entitled "Art of Compiling Statistics," for his electric punch-card tabulator…

000399.pdg

World War II, the period often cited as the origin of modern computing, the American pioneers of generalized information-processing hardware had already built a half-dozen impressive computing machines: the differential analyzer of MIT engineering professor Vannevar Bush (1930), the first automatic computer general enough to solve a wide variety of problems (Fig. 9.1); the "mechanical programmer" of Columbia Uni- versity astronomer Wallace Eckert (1933), which linked various IBM punch-card accounting machines to permit generalized and complex computation; Bush's electrical analog computer (1935), more general than his differential analyzer with punched-tape programming; an elec- tronic analog computer (1938) devised at the Foxboro Company; a working prototype of an electronic calculator (1939), under develop- ment by John Atanasoff at Iowa State University; and the Bell Lab- oratories Model I (1939), built by George Stibitz,…

000401.pdg

The concepts of information processing, program- ming, decision, and control and the intellectual stimulation of the re- lationships among them seemed in the air among European and American engineers, mathematicians, and philosophers by the mid- 1930s.

000403.pdg

In 1934an engineering student in Berlin, Konrad Zuse, began to design a universal calculating device that anticipated modern computers in several ways, including binary rather than decimal numbers and float- ing decimal point calculation (the first such applications to a machine), the programming rules of Boolean logic (unknown to Zuse), and the distinctive structure of a concrete open processor of information:punched tape (discarded 35mm movie film) input, a central processing unit, memory, programming, an internal controller, and an output device to display results (the similar structure of Babbage's Analytical Engine was also unkno…

000403.pdg

Beginning work on a concrete machine in 1936, Zuse had a mechan- ical prototype in 1938, an electromechanical relay machine in 1939, and by December 1941 the world's first general-purpose, program-con- trolled calculator in regular operation. Soon two specialized versions that analyzed the wing flutter of Nazi flying bombs displaced a com- putational office of thirty women at the Henschel Aircraft Company in Berlin.

000403.pdg

Little more than a giant electromechanical decimal calculator, the Mark I could nevertheless take two 23-digit numbers from paper tape input and within three seconds output their product onto punched cards.

000405.pdg

Even before Aiken had received IBM's backing for his proposal, Claude Shannon, then an MIT graduate student employed part-time to tend Vannevar Bush's differential analyzer, had published his mas- ter's thesis, "Symbolic Analysis of Relay and Switching Circuits" (Shannon 1938), which applied the propositional calculus of Whitehead and Russell's Principia Mathematica (1910-1913)to the design of elec- trical circuitry. Perhaps the most influential master's thesis ever writ- ten, in the words of Augarten (1984, pp. 100-101) it "not only helped transform circuit design from an art into a science, but its underlying

000405.pdg

message-that information can be treated like any other quantity and be subjected to the manipulation of a machine-had a profound effect on the first generation of computer pioneers." Shannon's paper also established that programming an electronic digital computer would be a problem not of arithmetic but of logic…

000406.pdg

Not as sophisticated as Zuse's relay machine, the Model I was per- manently wired to solve equations with complex numbers and could not be further programmed. It lacked a general-purpose central-proc- essing unit, a memory, and any clearly defined control unit. Because it input and output via teletype, however, it could be used from any- where in the telephone system. At the annual meeting of the American Mathematical Society at Dartmouth College in Hanover, New Hamp- shire, in September 1940, Stibitz installed a few teletypes to demon- strate the Model I in Manhattan-the first use of remote computing via telephone that would come to characterize the "telematic society" (~ora and Mine 1978) thirty years later. B…

000406.pdg

Pen

000406.pdg

Judging from this cumulative effort of the late 1930s,then, we might conclude that World War II interrupted work on generalized infor- mation-processing and computing technology as much as stimulated it.

000406.pdg

In either case, the idea that information-processing and computing hardware might be used to enhance bureaucratic control appears to have emerged only gradually during the transition phase of the Control Revolution.

000408.pdg

The turning point for the United States occurred in 1879 near the height of the crisis in control of the material economy, when General Francis Amasa Walker, then pro- fessor of political economy at Yale University, agreed to direct the 1880 census.

000408.pdg

Walker's encouragement of inv~mtionproduced two innovations in data-processing hardware: the Seaton tabulator and the Lanston add- ing machine.

000409.pdg

…the adding machine, developed by a lawyer friend of Seaton, Tolbert Lanston, who would patent a successful monotype machine for casting type in 1887. Lanston's adder allowed entry of numbers as they had been written-left to right-on the Seaton tab- ulator …

000410.pdg

In 1889, facing the prospect of a census that might be superseded by the next one before it could be completely tabulated, the Secretary of the Interior organized a committee of foremost statisticians to inves- tigate faster means of processing data. The committee decided to test three new systems: (1) hand transcription of data, using a different color ink for each characteristic, onto slips of paper that could be sorted by color and counted by hand; (2) transcription onto color-coded cards or "chips," also sorted and counted by hand; and (3) Herman Hollerith's method, which used a keyboard or "pantograph" punch to make holes in predetermined positions in standardized cards, counted individually by means of hand insertion into an electrical circuit-closing press, which had a pin contact for each possible hole location (Hollerith 1889).

000411.pdg

As his biographer, Geoffrey Austrian, con- cludes of Hollerith, "his schooling in the pervasive railroad technology of the day had a far greater influence on his development of tabulating machines than the more obvious example of other calculating and add- ing mechanisms" (1982; p. 36).

000412.pdg

Pen

000413.pdg

…the concrete open system for processing railroad cars and trains served Hollerith as the model for processing the in- formation-also as discrete material objects-that would soon serve in the control of such systems. This fact alone, quite apart from con- siderations of living systems more generally, suggests that information processing and communication cannot be understood independently of the matter, energy, and material processing systems that they control.

000413.pdg

Unlike the 1880count, the 1890census, even though it contained twenty more items (a total of 236), could be analyzed in every possible com- bination of variables, the most complicated tables produced at no more expense than the simplest ones. Boorstin (1973, p. 172) summarizes the impact of Hollerith's system: "Now it was as easy to tabulate the number of married carpenters 40 to 45 years of age as to tabulate the total number of persons 40 to 45 years of age."

000416.pdg

As we have already seen in the major beneficiaries of the Library Bureau's "systematizing and the Soviet Union's Gosplan, the first applications of the new information-processing technologies came in financial institutions, public utilities, railroads, and large manufac- turing operations. Life insurance companies ranked among the first companies to see profit in data processing: New York Life, which by the turn of the century had contracted to have its data punched onto Hollerith cards (Austrian 1982, p. 134), adopted about 1903the nation's first numerical insurance rating system, with values assigned to var- ious factors affecting the insurability of applicants. Telephone com- panies and other utilities shared a common data-processing problem: the continual recording and billing of large numbers of small amounts.

000422.pdg

Material cul- ture has also been crucial throughout human history, and yet capital did not begin to displace land as an economic base until the Industrial Revolution. To what comparable technological and economic "revolu- tion" might we attribute the similar displacement of the industrial capi- tal base by information and information-processinggoods and services, or the overshadowing of the Industrial by the Information Society? The answer, as we have seen, is the Control Revolution, a complex

000426.pdg

of rapid changes in the technological and economic arrangements by which information is collected, stored, processed, and communicated and through which formal or programmed decisions can effect societal control. From its origins in the last decades of the nineteenth century the Control Revolution has continued unabated to this day and in fact has accelerated recently with the development of microprocessing tech- nologies. In terms of the magnitude and pervasiveness of its impact upon society, intellectual and cultural no less than material, the Con- trol Revolution appears to be as important to the history of this century as the Industrial Revolution was to the last. Just as the Industrial Revolution marked an historical discontinuity in the ability to harness energy, the Control Revolution marks a similarly dramatic leap in our ability to exploit information.

000427.pdg

Once energy consumption, processing and transportation speeds, and the information requirements for control are seen to be interre- lated, the Industrial Revolution takes on new meaning. By far its greatest impact from this perspective was to speed up society's entire material processing system, thereby precipitating a crisis of control, a period in which innovations in information-processing and commu- nication technologies lagged behind those of energy and its application to manufacturing and transportation.

000427.pdg

As we have seen, what began as a crisis of safety on the railroads in the early 1840s hit distribution in the 1850s, production in the late 1860s, and marketing and the control of con- sumption in the early 1880s. As the crisis of control spread through the material economy, it inspired a continuing stream of innovations in control technology. These innovations, effected by transporters, producers, distributors, and marketers alike, reached something of a climax by the 1880s. With the rapid increase in bureaucratic control and a spate of innovations in industrial organization, telecommunications, and the mass _media,the technological and economic response to the crisis-the Control Rev- olution-had begun to remake societies throughout the world by the beginning of this c…

000429.pdg

Despite such rapid changes in mass media and telecommunications technologies, the Control Revolution also represented a restoration- although with increasing centralization-of the· economic and political control lost at more local levels during the Industrial Revolution. Be- fore this time, control of government and markets had depended on personal relationships and face-to-face interactions; by the 1890s, as we saw in Part III, control began to be reestablished by means of bureaucratic organization, the new infrastructures of transportation and telecommunications, and system-wide communication via the new mass media.

000433.pdg

Perhaps most pervasive of all rationalization is the increasing tendency to regulate interpersonal relationships in terms of a formal set of impersonal, quantifiable, and objective criteria, changes that greatly facilitate control by both government and business. The com- plex social systems that arose with the growth of capitalism and im- proved transportation and communication would have overwhelmed any information-processing system that operated on a case-by-case basis or by the particularistic considerations of family and kin that characterized preindustrial societies.

000435.pdg

The Information Society has not resulted from recent changes, as we have seen, but rather from increases in the speed of material processing and of flows through the material economy that began more than a century ago. Similarly, microprocessing and computing technology, contrary to currently fashionable opinion, do not represent a new force only recently unleashed on an unprepared society but merely the most recent installment in the continuing development of the Control Rev- olution. This explains why so many of the components of computer control have been anticipated, both by visionaries like Charles Babbage and by practical innovators like Daniel McCallum, since the first signs of a control crisis in the early nineteenth century.

000435.pdg